Ototoxicity

OTOTOXICITY

OTOTOXICITY Peter S. Roland, MD Professor and Chairman Department of Otolaryngology–Head and Neck Surgery The University of Texas Southwestern Medical Center at Dallas Dallas, Texas

John A. Rutka, MD, FRCSC Associate Professor Department of Otolaryngology University of Toronto Staff Neurotologist University Health Network Co-Director Multidisciplinary Neurotology Clinic and University Health Network Centre for Advanced Hearing and Balance Testing Toronto, Ontario

2004 BC Decker Inc Hamilton • London

EXIT

BC Decker Inc P.O. Box 620, L.C.D. 1 Hamilton, Ontario L8N 3K7 Tel: 905-522-7017; 800-568-7281 Fax: 905-522-7839; 888-311-4987 E-mail: [email protected] Website: www.bcdecker.com © 2004 BC Decker Inc All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior written permission from the publisher. 04 05 06 07 / FP / 9 8 7 6 5 4 3 2 1 ISBN 1-55009-263-4 Printed in Canada Sales and Distribution United States BC Decker Inc P.O. Box 785 Lewiston, NY 14092-0785 Tel: 905-522-7017 / 800-568-7281 Fax: 905-522-7839; 888-311-4987 E-mail: [email protected] Website: www.bcdecker.com Canada BC Decker Inc 20 Hughson Street South P.O. Box 620, L.C.D. 1 Hamilton, Ontario L8N 3K7 Tel: 905-522-7017 / 1-800-568-7281 Fax: 905-522-7839; 888-311-4987 E-mail: [email protected] Website: www.bcdecker.com Foreign Rights John Scott & Company International Publishers’ Agency P.O. Box 878 Kimberton, PA 19442 Tel: 610-827-1640 Fax: 610-827-1671 E-mail: [email protected] Japan Igaku-Shoin Ltd. Foreign Publications Department 3-24-17 Hongo Bunkyo-ku,Tokyo, Japan 113-8719 Tel: 3 3817 5680 Fax: 3 3815 6776 E-mail: [email protected]

UK, Europe, Scandinavia, Middle East Elsevier Science Customer Service Department Foots Cray High Street Sidcup, Kent DA14 5HP, UK Tel: 44 (0) 208 308 5760 Fax: 44 (0) 181 308 5702 E-mail: [email protected]

Mexico and Central America ETM SA de CV Calle de Tula 59 Colonia Condesa 06140 Mexico DF, Mexico Tel: 52-5-5553-6657 Fax: 52-5-5211-8468 E-mail: [email protected]

Singapore, Malaysia, Thailand, Philippines, Indonesia, Vietnam, Pacific Rim, Korea Elsevier Science Asia 583 Orchard Road #09/01, Forum Singapore 238884 Tel: 65-737-3593 Fax: 65-753-2145

Brazil Tecmedd Av. Maurílio Biagi, 2850 City Ribeirão Preto – SP – CEP: 14021-000 Tel: 0800 992236 Fax: (16) 3993-9000 E-mail: [email protected]

Australia, New Zealand Elsevier Science Australia Customer Service Department STM Division Locked Bag 16 St. Peters, New South Wales, 2044 Australia Tel: 61 02 9517-8999 Fax: 61 02 9517-2249 E-mail: [email protected] www.harcourt.com.au

India, Bangladesh, Pakistan, Sri Lanka Elsevier Health Sciences Division Customer Service Department 17A/1, Main Ring Road Lajpat Nagar IV New Delhi – 110024, India Tel: 91 11 2644 7160-64 Fax: 91 11 2644 7156 E-mail: [email protected]

Notice: The authors and publisher have made every effort to ensure that the patient care recommended herein, including choice of drugs and drug dosages, is in accord with the accepted standard and practice at the time of publication. However, since research and regulation constantly change clinical standards, the reader is urged to check the product information sheet included in the package of each drug, which includes recommended doses, warnings, and contraindications. This is particularly important with new or infrequently used drugs. Any treatment regimen, particularly one involving medication, involves inherent risk that must be weighed on a case-by-case basis against the benefits anticipated. The reader is cautioned that the purpose of this book is to inform and enlighten; the information contained herein is not intended as, and should not be employed as, a substitute for individual diagnosis and treatment.

This book is dedicated to our families: Marilena, Jake, Fiona Lauren, and Rory Melissa, Melinda, Evelyn, Jason, Britney, Kenzey and Thomas

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

10. Macrolides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Andrew R. Scott, BM, BS, MPhil, FRCS(ORL-HNS), and John A. Rutka, MD, FRCSC

Section I Anatomy and Physiology 1. Anatomy and Physiology of the Cochlea . . . . . . . . 1 Karen S. Pawlowski, PhD 2. Physiology of the Vestibular System . . . . . . . . . . . 20 John A. Rutka, MD, FRCSC Section II Systemic Toxicity 3. Salicylates, Nonsteroidal Anti-inflammatory Drugs, Quinine, and Heavy Metals. . . . . . . . . . . . 28 Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), and John A. Rutka, MD, FRCSC 4. Ototoxicity of Loop Diuretics . . . . . . . . . . . . . . . . 42 Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), Andrew R.Scott, BM, BS, MPhil, FRCS (ORL-HNS), and John A. Rutka, MD, FRCSC 5. Cinical Uses of Cisplatin . . . . . . . . . . . . . . . . . . . . 50 Jeremy Sturgeon, MD, FRCPC 6. Ototoxicity of Platinum Compounds . . . . . . . . . 60 Michael Anne Gratton, PhD, and Brendan J. Smyth, PhD, MD 7. Iron Chelating and Other Chemotherapeutic Agents: The Vinca Alkaloids . . . . . . . . . . . . . . . . . 76 Andrew R.Scott, BM, BS, MPhil, FRCS(ORL-HNS), Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), and John A. Rutka, MD, FRCSC 8. Clinical Aminoglycoside Ototoxicity . . . . . . . . . . 82 Coleman Rotstein, MD, FRCPC, and Lionel A. Mandell, MD, FRCPC, FRCP(Lond) 9. Mechanisms for Aminoglycoside Ototoxicity: Basic Science Research . . . . . . . . . . . . . . . . . . . . . . 93 Jochen Schacht, PhD

Section III Topical Toxicity 11. Middle Ear Effects of Ototopical Agents . . . . . . 107 Charles G. Wright, PhD, and Peter S. Roland, MD 12. Topical Aminoglycoside Cochlear Toxicity . . . . 114 Peter S. Roland, MD, and Charles G. Wright, PhD 13. Topical Aminoglycoside Vestibular Toxicity . . . 121 Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), Vitaly E. Kisilevsky, MD, and John A. Rutka, MD, FRCSC 14. Chloramphenicol, Colymycin, and Polymyxin . 128 Leonard P. Rybak, MD, PhD, and Srinivasan Krishna, MD, MPH 15. Topical Antifungals . . . . . . . . . . . . . . . . . . . . . . . 134 Lawrence W. C. Tom, MD, Lisa M. Elden, MS, MD, and Roger R. Marsh, PhD 16. Surgical Disinfectants and Antiseptics . . . . . . . . 140 Andrew R. Scott, BM, BS, MPhil, FRCS(ORL-HNS), Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), and John A. Rutka, MD, FRCSC Section IV Interventions 17. Genetic Factors in Aminoglycoside Ototoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 Nathan Fischel-Ghodsian, MD 18. Audiologic Monitoring for Ototoxicity . . . . . . . 153 Kathleen C. M. Campbell, PhD 19. Monitoring Vestibular Ototoxicity . . . . . . . . . . . 161 Vitaly E. Kisilevsky, MD, R. David Tomlinson, PhD, Paul Ranalli, MD, FRCPC, and Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL)

viii

Contents

20. Ototoxic Damage to Hearing: Otoprotective Therapies . . . . . . . . . . . . . . . . . . . 170 Thomas R. Van De Water, PhD, and Leonard P. Rybak, MD, PhD Section V Therapeutic Uses of Ototoxic Effects 21. Systemic Treatment of Bilateral Meniere’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 Sumit K. Agrawal, MD, and Lorne S. Parnes, MD, FRCSC 22. Intratympanic Gentamicin in the Treatment of Meniere’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . 191 Brian W. Blakley, MD, PhD, FRCSC

Section VI Medicolegal Concerns 23. Medicolegal Aspects of Ototoxicity . . . . . . . . . . 198 Peter E. Rhatican, JD, Sloan H. Mandel, LLB, and John A. Rutka, MD, FRCSC Appendix 1: Information Provided to Lawyers. . . . . 207 Appendix 2: Summary of the 2004 AAO-HNS Consensus Panel Recommendations . . . . . . . . . . . . . 213 Appendix 3: Major Groups of Agents Recognized to be Ototoxic in Humans . . . . . . . . . . . . . . . . . . . . . 214 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215

Preface

Ototoxicity can be broadly defined as the tendency of certain substances, either systemic or topical, to cause functional impairment and cellular damage to the tissues of the inner ear and especially to the end organs of the cochlear and vestibular divisions of the eight cranial nerves. Much less frequent would be the finding of impairment or injury at the level of the central nervous system. In relative terms, ototoxicity is not common. It nevertheless generates much interest amid controversy for those physicians who treat often life-threatening medical conditions with potentially ototoxic medications (primum non nocere) and for basic science researchers who try to determine at a genetic and cellular level the underlying pathophysiology of injury in the hopes of one day being able to prevent damage or perhaps even to regenerate damaged end-organ hair cells (fiat lux). However, we should never forget that when ototoxicity occurs it is not victimless and it places a truly considerable burden of disability on the affected individual. To lose one’s hearing profoundly limits one’s abilities to communicate and socialize in a hearing world; to lose one’s balance literally robs one of the ability to perform physical activities and ultimately limits independence. The book is divided into several sections according to broad topics. We approach ototoxicity with a basic review of the anatomy and physiology of the cochlea and the vestibular system. This is followed by chapters primarily concentrating on those agents that under certain circumstances have proven to be ototoxic, both historical agents (eg, acetylsalicylic acid, quinine, and the aminoglycoside class of antimicrobial therapy) and some recently introduced (eg, azithromycin, deferoxamine). Many of the book’s clinical chapters focus on the phenomenon of systemic aminoglycoside and platinum-based chemotherapeutic ototoxicity. In this regard there is no doubt that if we could eliminate all aminoglycosides, for example, we would certainly witness a significant decline in ototoxicity in most of the

industrialized world and in those populations (specifically in China) where certain genetic predispositions serve to make an individual extraordinarily sensitive to their ototoxic effects. Unfortunately, the above scenario is unlikely to occur in the foreseeable future, as clinical indications for aminoglycoside therapy (often lifethreatening sepsis), their convenience for use in outpatient treatment (single daily dosing), their overall lack of bacterial resistance, and the pharmacoeconomics (low per-unit cost) suggest they will be with us for some time. Until basic science research into the mechanisms of aminoglycoside and platinum-based ototoxicity might provide us with new clinically applicable treatment stratagems for its prevention (and even hair cell regeneration), clinicians still need to monitor patients appropriately for early changes indicative of cochleovestibular dysfunction and to act accordingly on those findings. One of the great controversies addressed in this book deals with the occurrence of topical aminoglycoside ototoxicity arising from the treatment of middle ear sepsis in the presence of a tympanic membrane defect. As recently as the 1980s most clinicians would have felt this to be a rare and mostly theoretical concern in humans, despite animal experimental data to the contrary. The efficacy of concentrated topical gentamicin in the treatment of incapacitating unilateral Meniere’s disease drew attention to this agent’s primary vestibulotoxic nature. This ultimately led to the recognition in certain patient cohorts and later its demonstration in a series of patients with Meniere’s disease that commonly used commercially available gentamicin ear drops could also be ototoxic in humans if used long enough in the presence of a dry middle ear. The development of nonototoxic fluoroquinolone antibiotic drops, for example, represents a major advance in the treatment of open middle ear sepsis and was in part expedited by major concerns regarding topical aminoglycoside ototoxicity. Recognizing that topical ototoxicity exists introduces the possibility that whatever gets into the middle ear

x

Preface

may eventually make its way into the inner ear—this should give us pause when using other ototopical agents and even the disinfectants and antiseptics routinely used in ear surgery. The final chapter looks at the medicolegal aspects of ototoxicity through the eyes of two distinguished trial lawyers, Sloan Mandel (Canada) and Peter Rhatican (United States). Although the laws between these two countries differ, the concepts behind them regarding the standard of care, harm, and what constitutes appropriate informed consent are remarkably similar. A learned discussion follows each presented case history involving ototoxicity, which the clinician is urged to read carefully (it might help prevent a medicolegal action from occurring in future). The inclusion of three appendices at the end brings our book to a close. Appendix 1 provides the reader with full disclosure and the expert information our trial lawyers would have used to formulate their opinions. Appendix 2 contains a summary of the 2004 American Academy of Otolaryngology–Head and Neck Surgery (AAO-HNS) Consensus Panel Recommendations regarding the use of potentially ototoxic topical antibiotics for middle ear use. This landmark report founded on evidence-based medicine (EBM) sets a new standard for patient care and should be carefully read by all physicians who use topical therapy

for the treatment of middle ear disease. In Appendix 3 we have provided a handy reference list for those agents currently recognized to be ototoxic in humans should there ever be any doubt. We are indebted to several individuals whose help made this book possible. We acknowledge the vision of Chuck Inman, who challenged us while providing us with the necessary encouragement to write this book. We thank our publisher and those individuals at BC Decker Inc, namely Rochelle Decker, Paula Presutti, and Monika Holden, for their help with the organization, editing, typesetting, and the usual things publishers do. Many of the illustrations in the book and on its cover were made possible by permission from Solvay Pharma; with thanks to Meena Bhogal for her assistance in this matter. We would be remiss to not specifically thank the many authors who wrote their chapters without complaint and in a timely fashion. To them we are eternally grateful. Finally, we would like to formally acknowledge the respective roles of our families, who understood our need to complete this work, which has constituted the better part of our academic life over the past decade, and provided us the latitude to do so. PETER ROLAND JOHN RUTKA July 2004

List of Contributors

Sumit K. Agrawal, MD Department of Otolaryngology University of Western Ontario London, Ontario

Roger R. Marsh, PhD Department of Otorhinolaryngology–Head and Neck Surgery University of Pennsylvania Philadelphia, Pennsylvania

Brian W. Blakley, MD, PhD, FRCSC Department of Otolarygology University of Manitoba Winnipeg, Manitoba

Lorne S. Parnes, MD, FRCSC Department of Otolaryngology and Clinical Neurological Sciences University of Western Ontario London, Ontario

Kathleen C. M. Campbell, PhD Department of Surgery Southern Illinois University School of Medicine Springfield, Illinois Lisa M. Elden, MS, MD Department of Otorhinolaryngology–Head and Neck Surgery University of Pennsylvania Philadelphia, Pennsylvania Nathan Fischel-Ghodsian, MD Department of Pediatrics Cedars-Sinai Medical Center Los Angeles, California Michael Anne Gratton, PhD Department of Otorhinolaryngology–Head and Neck Surgery University of Pennsylvania School of Medicine Philadelphia, Pennsylvania Vitaly E. Kisilevsky, MD University Health Network University of Toronto Toronto, Ontario Srinivasan Krishna, MD, MPH Department of Surgery Southern Illinois University School of Medicine Springfield, Illinois

Karen S. Pawlowski, PhD Department of Otolaryngology–Head and Neck Surgery University of Texas Southwestern Medical Center Dallas, Texas Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL) Department of Otorhinolaryngology University Health Network University of Toronto Toronto, Ontario Paul J. Ranalli, MD, FRCPC Department of Medicine University of Toronto Toronto, Ontario Peter E. Rhatican, JD Mendham, New Jersey Peter S. Roland, MD Department of Otolaryngology–Head and Neck Surgery The University of Texas Southwestern Medical Center at Dallas Dallas, Texas Coleman Rotstein, MD, FRCPC Department of Medicine (Infectious Diseases) McMaster University Hamilton, Ontario

Sloan H. Mandel, LLB Thompson Rogers Toronto, Ontario

John A. Rutka, MD, FRCSC Department of Otolaryngology University of Toronto Toronto, Ontario

Lionel A. Mandell, MD, FRCPC, FRCP(Lond) Department of Medicine (Infectious Diseases) McMaster University Hamilton, Ontario

Leonard P. Rybak, MD, PhD Department of Surgery Southern Illinois University School of Medicine Springfield, Illinois

xii

List of Contributors

Jochen Schacht, PhD Department of Otolaryngology University of Michigan Medical School Andrew R. Scott, BM, BS, MPhil, FRCS(ORL-HNS) Department of Otolaryngology University Health Network Toronto, Ontario

Lawrence W. C. Tom, MD Department of Otorhinolaryngology–Head and Neck Surgery University of Pennsylvania Philadelphia, Pennsylvania R. David Tomlinson, PhD Department of Otolaryngology University of Toronto Toronto, Ontario

Brendan J. Smyth, PhD, MD Department of Clinical Pharmacology Georgetown University School of Medicine Washington, District of Columbia

Thomas R. Van De Water, PhD Department of Otolaryngology University of Miami School of Medicine Miami, Florida

Jeremy Sturgeon, MB, BS, DIBIM(Med Oncol), FRCPC Department of Medicine University of Toronto Toronto, Ontario

Charles G. Wright, PhD Department of Otolaryngology–Head and Neck Surgery University of Texas Southwestern Medical Center Dallas, Texas

Anatomy and Physiology CHAPTER 1

Anatomy and Physiology of the Cochlea Karen S. Pawlowski, PhD

This overview of the anatomy and physiology of the various components of the cochlea includes information about (1) the position of the cochlea within the temporal bone; (2) the anatomy and physiology of the cochlear fluid spaces, the vascular system, the supporting tissues, and the neuroepithelia that are responsible for the sensorineural transduction of sound; and (3) the electrophysiology associated with sensorineural transduction.

A

C

TM

COCHLEAR ANATOMY The inner ear includes the osseous labyrinth, a series of interconnected chambers within the temporal bone, and the membranous labyrinth, which occupies the osseous labyrinth. The membranous labyrinth houses the neuroepithelia responsible for auditory and vestibular sensation. The osseous labyrinth lies within the petrous portion of the temporal bone and forms part of the skull base (Figure 1-1A). The auditory component of the inner ear is the cochlea. The cochlea of mammals has a coiled, snailshell shape. Its main axis runs anterior to posterior. The narrow, apical end lies anterior in the head, and the wide, basal end of the spiral lies posterior and slightly more medial and dorsal to the apex (Figure 1-1B). The number of turns in the cochlear spiral varies among species; humans have 2.5 to 3 turns.1 The sensory cells of the auditory system lie along the turns of the cochlea on a flexible membrane, called the basilar membrane (see Basilar Membrane). The two types of sensory cells that lie on the basilar membrane, inner hair cells (IHCs) and outer hair cells (OHCs), have different roles in hearing (see Organ of Corti). The IHCs are responsible for transferring the fluid motion at their surface into a neural signal that can be passed to the brain to be perceived as hearing. The OHCs lie on the flexible portion of the basilar membrane. They have the capacity to contract and elongate as the basilar membrane moves. Their role in hearing is to modify the incoming signal by altering the fluid motion at the top of the

EAC

B

IAC

V

ES

IAC V TM

CA

C EAC E Figure 1-1 Diagrams showing the position of the cochlea within the human head. A, Horizontal view of the skull base with a close-up of the left temporal bone. B, Coronal view of the head with a close-up of the external, middle, and inner ear. The inner ear is composed of the cochlea and the vestibular apparatus. The vestibular nerve, cochlear nerve, and a portion of the facial nerve all enter the brain cavity via the internal auditory canal. C = cochlea; CA = cochlear aqueduct; E = eustachian tube; EAC = external auditory canal; ES = endolymphatic sac; IAC = internal auditory canal; TM = tympanic membrane; V = vestibular apparatus.

Anatomy and Physiology

2

H

SV

SL St

RM SM

OS

SV

IS

SM B SG

ST

O of C ST

SG OSL

Figure 1-3 The constituents of the scala media (SM) showing the adjacent areas of the scala vestibuli (SV) and scala tympani (ST). The stria vascularis (St) is responsible for production of the endocochlear potential and the high potassium concentration in the endolymph. The organ of Corti (O of C) houses the sensory receptor cells, the inner and outer hair cells. B = Böttcher cells; BM = basilar membrane; IS = inner sulcus cells; OS = outer sulcus cells; OSL = osseous spiral lamina; RM = Reissner’s membrane; SG = spiral ganglion inside the OSL; SL = spiral ligament.

RWM

Figure 1-2 A right cochlea cut in half along the modiolar axis showing the three cochlear scalae. The scala vestibuli (SV) and scala tympani (ST) are continuous with one another at the helicotrema (H). Arrows in the scala tympani indicate direction of spiral of the scalae. RWM = round window membrane; SG = spiral ganglion within the modiolus; SM = scala media.

The scala media is sandwiched between the scala tympani and scala vestibuli as they spiral up the cochlea from base to apex. These three compartments spiral along the cochlear turns and around a nerve-filled bony central axis, the modiolus. The fluid that occupies the scala tympani and scala vestibuli is perilymph.2 The ion composition of perilymph is similar to that of cerebrospinal fluid (CSF) and many extracellular fluids (Table 1-1), with a low sodium (Na+) concentration and a high potassium (K+) concentration.3–8 The scala media lies between scala tympani and scala vestibuli and contains endolymph (Figure 1-3). The endolymph is high in K+ and low in Na+, in inverse

IHC. Details of the roles of these cells are covered in this chapter. Fluid Spaces The cochlea is subdivided into three fluid-filled compartments. The scala tympani and scala vestibuli compartments are continuous with one another at the helicotrema, in the apex of the cochlea (Figure 1-2). Table 1-1 Composition of Fluids (mMol) Component

Scala Media

Scala Vestibuli

Scala Tympani

Endolymphatic Sac

CSF

Plasma

Na+

1–1.3

141–145

138–148

103

149

145

+

157

6

4.2

15

3.1

5

Ca2+

0.023

0.6–2

1–2

0.5

1.2

2.6

131–132

121

119

129

106

HCO3–

31

18

21

19

18

pH

7.4

7.3

7.3

7.3

7.3

K

Cl–

Composition of the various fluids found within the cochlea. Other low-concentration components, such as proteoglycans and glycosaminoglycans, are not included. Sodium and calcium ion concentrations are low in the endolymphatic compartment (scala media) and high in the perilymphatic compartments (scala tympani and scala vestibuli), whereas potassium ion concentration is the reverse. These ions, with their relative concentrations within the compartments of the inner ear, are important in the activation of hair cells. Values derived from several sources.3,5–8 CSF = cerebrospinal fluid.

Anatomy and Physiology of the Cochlea

3

ES

SMA

ED Cr CMV

DR

LA

CD VCA

*

AICA RWM BA

SMV

Figure 1-4 The vasculature in a human cochlea. The darker vessels illustrate arterial flow. The lighter vessels illustrate venous flow. Oxygenated blood enters via the basilar artery (BA); the anterior inferior cerebellar artery (AICA) branches off the BA and extends to the labyrinthine artery (LA), which extends into the cochlea and vestibular apparatus. The spiral modiolar artery (SMA) extends from the cochlear portion of the LA to supply the modiolus, lateral wall, and organ of Corti. The venous drain starts from the spiral modiolar vein (SMV), which drains into the cochleomodiolar vein (CMV), which joins up with the venous system of the vestibular apparatus at the vein of the cochlear aqueduct (VCA). RWM = round window membrane. Adapted from Axelsson A, Ryan AF. Circulation of the inner ear. I. Comparative study of vascular anatomy in the mammalian cochlea. In: Jahn AF, Santos-Sacchi J, editors. Physiology of the ear. New York: Raven; 1988. p. 295–315.

ratio relative to the perilymph, and this difference is critical for hearing.2,3 The perilymph of the scala tympani communicates with CSF via the cochlear aqueduct (see Figure 1-1B). There is also some communication between perilymph and CSF via the internal auditory meatus.5,9 These communication routes are not completely open, as the cochlear aqueduct in humans is filled with loose connective tissue and the internal auditory meatus is filled with the auditory, vestibular, and facial nerves and their associated tissues. Scala tympani is separated from the middle ear space by the round window membrane (see Figure 1-2). The stapes footplate separates the vestibule from the middle ear space. Scala vestibuli is continuous with the vestibule at the basal end of the cochlea. Scala tympani and scala vestibuli are also in contact with the vascular system. Radiating arterioles extend out along the bony wall of scala vestibuli from the spiral modiolar artery. A network of collecting venules lines the bony wall of scala tympani before the venules connect with the spiral modiolar vein (Figure 1-4; see Blood Supply and Vasculature).10 Communication routes into the endolymphatic space of the scala media are limited to a vascular route, via the stria vascularis, or a perilymphatic route, via scala tympani and scala vestibuli. Scala media is part of the membranous labyrinth. The endolymphatic space

SM UM Figure 1-5 The membranous labyrinth. Endolymph is formed in the pars superior portion (dark shaded area) of the membranous labyrinth (utricle and semicircular canals) and in the pars inferior portion (light shaded area) of the membranous labyrinth (cochlear duct and saccule). Endolymph from the cochlear duct (CD) drains into the saccule via the ductus reuniens (DR); from there it enters the endolymphatic duct (ED). Endolymph drains from the pars superior via the utriculoendolymphatic valve of Bast (*) into the ED. From the ED, endolymph drains into the endolymphatic sac (ES), where it is broken down and resorbed into the vascular system. Cr = cristae ampularis of the superior, posterior, and lateral semicircular ducts; SM = sacular macula; UM = utricular macula.

of the scala media is a fluid-filled compartment that extends from the apex through the base of the cochlea, continuing into the saccule and the rest of the vestibular apparatus via the ductus reuniens (Figure 1-5). The endolymphatic space of the utricle and semicircular canals is somewhat isolated from the endolymphatic space of the cochlea and saccule by a small valve, called the utriculoendolymphatic valve of Bast. This valve lies at the opening of the utricle into the endolymphatic duct system, which connects the endolymph of the utricle and semicircular canals to the endolymph of the cochlea, saccule, and endolymphatic duct. The membranous labyrinth continues from the vestibular space through the endolymphatic duct to terminate in the endolymphatic sac. The endolymphatic sac lies within the dura on the medial aspect of the temporal bone (see Figure 1-1B). Both the perilymphatic fluid space and the membranous labyrinth can be affected by fluid pressure changes within the brain cavity due to fluid communication via the cochlear aqueduct, the endolymphatic duct and sac, and the internal auditory meatus. This relationship is important for fluid pressure regulation within the inner ear.9–13 The fluid pressure ratio between endolymphatic and perilymphatic compartments of the cochlea need to be stable in order to allow proper fluid motion at the tip of the sensory

4


cells. A shift of the endolymph–perilymph fluid ratio has been implicated in several conditions involving tinnitus and hearing loss.4 Blood Supply and Vasculature The cochlear blood supply is one route of entry for cochleotoxic agents. Blood is supplied to the inner ear via the labyrinthine artery, which branches from the anterior inferior cerebellar artery branch of the basilar artery (see Figure 1-4). From the labyrinthine artery, blood to the cochlea is circulated via the spiral modiolar artery and vein.11,15,16 Arterioles branch off the large spiral modiolar artery to form capillary beds in the tissues of the modiolus, osseous spiral lamina, and tympanic lip along the outside lip of the spiral lamina. Branches from the modiolus also radiate out toward the lateral wall along the bony wall of scala vestibuli. The stria vascularis, spiral ligament, and spiral prominence comprise the lateral wall of the scala media (see Figure 1-3). Arterioles branching from the modiolus form distinct capillary beds in the lateral wall: one each for the spiral ligament, stria vascularis, and spiral prominence. Separate capillary networks supply the various tissues in the lateral wall. The three capillary beds in the lateral wall serve distinct purposes, and their walls are constructed to reflect their purpose. The vessel walls of these lateral wall networks are specialized to filter substances from the blood supply to the cochlear tissues. 16,17 In general, the structure of the vessel endothelial cells prevents free diffusion of solutes from the blood into cochlear tissues. Under normal conditions, solutes must first cross the endothelial cells, via active transport, before they can enter the cochlear tissues. Tight junctions between adjacent endothelial cells prevent passive diffusion of substances from the blood to the perilymphatic tissues and the perilymph or to the stria vascularis and the endolymph. The tightness of this capillary system varies within the cochlea depending on location.17,18 In the spiral ligament, capillaries that lie near Reissner’s membrane (suprastrial region) have endothelial cells that contain stress fibers that may be able to regulate the flow of blood to the rest of the lateral wall vasculature.18 The portion of the spiral ligament that lies adjacent to the stria vascularis has vascular walls that are composed of continuous layers of endothelial cells connected by tight junctions, with a prominent basement membrane surrounded by pericytes. This arrangement creates a fairly tight blood–perilymph barrier. The vessels in the spiral prominence, near the basilar membrane, appear to be more permeable than those of the adjacent spiral ligament and stria vascularis because they have less basal lamina. This region is nearest the collecting venules of the lateral wall; therefore, these vessels may serve as the entry point for substances

coming from the endolymph and surrounding tissues to the vasculature. The densest capillary network of the lateral wall is in the stria vascularis. Strial capillaries are surrounded by a continuous layer of endothelial cells, a thickened basal lamina, and pericytes that are similar in appearance to those of the adjacent spiral ligament.15,16 However, the basal lamina of stria vascularis vessels is twice as thick as that of the vessels that lie in the adjacent spiral ligament. The basement membrane of the stria vascularis runs between closely packed marginal, intermediate, and basal cells, whereas the spiral ligament basal lamina runs through loosely packed connective tissue. The arrangement of the basal lamina within the stria vascularis is thought to efficiently supply nutrients to the strial cells and can serve as a pathway for other substances to the strial cells, once they pass from the vasculature to the intracellular space.17 Capillaries in the central portion of the modiolus are fenestrated, allowing quick passage of fluids and solutes. Capillaries that are found in the tympanic lip portion of the spiral limbus, near the basilar membrane, are thought to supply the organ of Corti with oxygen. These capillaries are not fenestrated but have endothelial cells connected by tight junctions, similar to the vessels of the lateral wall.11,19 Because of the blood–perilymph and blood–strial barriers, entry of solutes into the inner ear is mediated by their interaction with the vascular endothelial cells, vascular pericytes, and tissues along the pathway to the fluids. This is true not only of oxygen and nutrients, but also drugs that have accumulated in the vasculature. The lag time for substance accumulation or equilibration between endolymph, perilymph, and blood concentrations of an intravascularly administered substance indicates that there is a moderate barrier between blood and perilymph and a high barrier between blood and endolymph and between endolymph and perilymph.17,18 The cochlear vasculature is also important for the maintenance of the endocochlear potential (see Lateral Wall ). The active system that maintains endocochlear potential is strongly dependent on oxygen availability, which is supplied by the cochlear artery and the vasculature of the stria vascularis. Lateral Wall The lateral wall of scala media houses structures important in the production of the endocochlear potential. The endocochlear potential is a standing potential that exists between the endolymphatic space and the perilymphatic space.20,21 The lateral wall is also responsible for active accumulation of potassium ions (K+) in the endolymphatic space.20 The endocochlear potential and high endolymphatic K+ are important


for K+ entry and triggering of an action potential in hair cells when ion channels in these cells are opened.20 The spiral ligament consists of fibrous tissue, vasculature, epithelial cells, and extracellular matrix. The stria vascularis is situated between the spiral ligament and the endolymphatic space. It consists of three types of cells: marginal, intermediate, and basal. The strial vessels occupy the central portion of the stria vascularis adjacent to all three cell types (Figure 1-6). The strial cells are isolated from their surroundings by a series of tight junctions.22,23 The marginal cells are adjacent to the endolymphatic space and are joined to one another by tight junctional complexes near the endolymphatic surface. Marginal cells interdigitate with strial vessels, intermediate cells, and basal cells. Intermediate cells lay within the middle layer of the stria vascularis, adjacent to the strial vessels. They are in direct communication with basal cells via gap junctions.24 The basal cells are flat, overlapping cells on the lateral side of the stria vascularis. A series of tight junctional complexes between basal cells separates this side of the stria vascularis from the perilymphatic fluid spaces adjacent to it. The basal cells are in direct communication with some of the fibrocytes in the spiral ligament via gap junctions.25,26 Basilar Membrane The basilar membrane is a multilayered fibrous structure that extends from the osseous spiral lamina laterally to the spiral ligament. It separates scala media from scala tympani (see Figure 1-3). Epithelial cells cover the basilar membrane on the scala tympani side. The organ of Corti, the inner and outer sulcus cells, and Böttcher cells line the scala media side of the basilar membrane. Mechanical properties of the basilar membrane dictate the vibratory pattern of the membrane in response to a sound stimulus.27,28 The mechanics of the basilar membrane are complex and frequency specific. The basilar membrane is narrow and stiff at the basal end, gradually becoming wide and flexible, with an increased mass at the apical end. This causes the peak amplitude of vibration in response to high-frequency stimuli to occur at the basal end. The higher the frequency, the more basal the location of peak amplitude of vibration, and the lower the frequency, the more apical the peak vibration point. These mechanical properties, along with the action of the OHCs and the refractive properties of the individual nerve cells (see below), dictate the location at which the various frequency components of a sound stimulus are encoded along the basilar membrane.20 This description is an oversimplification, as the position of maximum vibration to a certain frequency is also dependent on the intensity of the sound carrying that frequency. An increase in intensity of a simple tone will cause the point of maximum vibration amplitude to shift toward the apex. Complex sounds also affect frequency coding

5

EL MC

IC

V IC

BC BC

F Figure 1-6 The stria vascularis. Medium dark cells adjacent to the endolymphatic space (EL) illustrate marginal cells (MC). Light cells in the middle of the tissue, adjacent to the vessel (V), are intermediate cells (IC); the darkest cells beneath the IC and above the fibrocytes (F) are basal cells (BC). Tight junctions at the endolymphatic surface of the marginal cells, between the basal cells and between the vascular endothelia cells, isolate the intercellular fluid of the stria vascularis from its surroundings. Gap junctions allow direct communication between BC and F of the spiral ligament as well as between BC and IC.

at the basilar membrane (for more information see Pickles 20 ). The coding of complex sounds is very involved and is not discussed in this chapter. Tectorial Membrane The tectorial membrane is a gel-like structure, consisting of highly hydrated layers of collagen, related proteins, and glycosaminoglycans. 29–31 It lies near the surface of the reticular lamina of the organ of Corti (Figure 1-7). In mammals the tectorial membrane has several distinct zones. Some function as structural elements for the tectorial membrane, whereas others serve as an interface between the tectorial membrane and the organ of Corti. The tectorial membrane is in contact with the OHCs. The stereocilia of the IHCs are close to the tectorial membrane, but they do not make direct contact with it.31,32 Although the tectorial membrane is not part of the organ of Corti, it plays a key role in organ of Corti function. Organ of Corti The organ of Corti houses the auditory sensory epithelia (see Figure 1-7). It is a long ribbon of cells that

6


HS TM

B

IB

HB

3

2

S

CP

Mi IP

1

SC I

HC OP

N DC

MSO

IPh HP

SG OSL

Figure 1-7 Cross section of the organ of Corti and its associ-

ation with the tectorial membrane (TM) and Hensen’s stripe (HS). 1, 2, 3 = first, second, third row outer hair cells, respectively; DC = Deiters’ cells; HC = Hensen’s cells; HP = habenula perforata; I = inner hair cell; IB = inner border cell; IP = inner pillar cells; IPh = inner phalangeal cell; MSO = efferent fiber from the medial superior olivary bundle; OP = outer pillar cells; OSL = osseous spiral lamina; SG = spiral ganglion.

spirals along the basilar membrane from base to apex.32–35 This tissue modifies and transduces the wave motion of the basilar membrane into the neurochemical signal. The cells of the organ of Corti include the sensory cells (IHCs and OHCs); their supporting cells (the inner border and inner phalangeal and Deiters’ cells, respectively); afferent (type I and type II spiral ganglion) and efferent nerve endings (the medial and lateral olivocochlear bundles); the inner and outer pillar cells; and Hensen’s cells.32,34 The configuration of the cells of the organ of Corti changes as it spirals apically. For example, OHCs at the apex of the cochlea in humans are longer than those at the base.35–38 Also, in humans, the Böttcher cells that line the basilar membrane just lateral to the Hensen’s cells at the base slowly disappear by midcochlea. These gradations in structure of the organ of Corti, along with associated structures, such as the basilar membrane and lateral wall, are important for tuning the frequencies to which an individual hair cell responds. Outer Hair Cells and Supporting Cells Of the two types of sensory cells in the cochlea, the OHCs are more abundant than IHCs (approximately 12,000 OHCs compared with 3,500 IHCs in humans35,36), yet the IHCs are responsible for coding of sound waves into neural signals. The OHCs are responsible for fine-tuning the wave action of the basilar membrane and reticular lamina.38 In mammals, the OHCs are arranged in three or four rows that lie more lateral to the modiolus than do

DC AF EF Figure 1-8 An outer hair cell, associated Deiters’ cell (DC),

and afferent (AF) and efferent (EF) nerve endings. All hair cells have an eccentric nucleus (N), numerous mitochondria (Mi), and a basal body (B), and their cuticular plates (CP) are connected to the adjacent cells via tight junction complexes. Stereocilia (S) of varying heights project from the surface of the cuticular plate. They are connected to each other via tip and side links, and they possess a dense core that runs down into or through the cuticular plate. The denseness of the subsurface cisternae (SC) and the Hensen’s body (HB) are unique to the outer hair cells, as is the fact that the tallest row of stereocilia is attached to the underside of the tectorial membrane.

the IHCs. The endolymphatic surfaces of the IHCs and OHCs consist of the cuticular plate with stereocilia. The cuticular plate is a hard cuticle-like component of the endolymphatic portion of the cell. Cytoplasmic gaps are interspersed within the cuticular plate near the complex cellular junctions between the hair cells and supporting cells.39–41 These gaps in the cuticle are thought to be important for intercellular communication between the endolymphatic surface of the cell and the rest of the cytoplasm. The OHC is connected to the surrounding supporting cells, Deiters’ cells, by tight junction complexes at the level of the cuticular plate (Figure 1-8).42 The stereocilia are long projections extending from the cuticular plate on the surface of the hair cells (Figures 1-8 to 1-10). Stereocilia of OHCs are shaped like a “W” and progress in height from a short row toward the modiolar side of the cell to a tall row that is embedded in the tectorial membrane on the lateral side of the cell (see Figure 1-8).32,33,41 All of these stereocilia are connected together by a network of side links. Their tips are connected with tip links that are part of a


7

S B

CP

SC IB RER

Figure 1-9 Scanning electron micrograph of a mouse organ

of Corti. A row of inner hair cells shows the flat U shape of the stereocilia. Rows of outer hair cells show the W shape of the stereocilia. The bar represents 10 microns. 1, 2, 3 = first, second, third row outer hair cells, respectively; I = inner hair cells. Courtesy of Dr. Charles G. Wright, UT Southwestern Medical Center.

spring-loaded gate system that opens a nonselective cation gate when the stereocilia are displaced toward the apex and laterally, toward the tallest row of stereocilia. The influx of K+ ions then sets off a cycle, which ends in changes in cell shape and the release of synaptic vesicles into the synaptic cleft at the base of the hair cell.44 The center of the stereocilia is mainly organized actin fibers, with a dense core, called the stereocilia rootlet, running from close to midway on the portion of the stereocilia that is above the cuticular plate to the bottom or beneath the cuticular plate (see Figures 1-8 and 1-10). This portion of the stereocilia is important in communicating the inflow of K+ at the tip of the stereocilia to the rest of the cell.39,40 The supporting cells that surround the OHCs are Deiters’ cells and pillar cells. Pillar cell projections contact the modiolar side of the first (most modiolar) and second row of OHCs.40 The first-row OHCs make contact with pillar cells on the modiolar and lateral side of the OHCs. Deiters’ cell processes only contact the firstrow OHCs on either side of the lateral pillar processes. The pillar cells contact the second row of OHCs only at the modiolar side. The remaining sides of the secondrow OHCs plus third-row OHCs are all contacted by Deiters’ cell processes. Junctional complexes that connect OHCs, Deiters’ cells, and outer pillar cells create a tight barrier between endolymph and perilymph. These junctions are arranged in a complex manner so that they do not open even during scar formation, suggesting that maintenance of this barrier is very important in the continued health of the organ of Corti surrounding the traumatized area. The outer and inner pillar cells form the tunnel of Corti, separating the IHC region of the organ of Corti from the OHC region (see Figures 1-7 and 1-9).45 The

Ve

N

Mi

IPh

AF

EF

Figure 1-10 An inner hair cell, with its associated supporting cells and nerve endings. Like the outer hair cells these cells have an eccentric nucleus (N), mitochondria (Mi), and a basal body (B), with a cuticular plate (CP) that is connected to the adjacent cells via tight junction complexes. The inner hair cell is goblet-shaped with numerous vesicles (Ve) and rough endoplasmic reticulum (RER) in the “neck” of the goblet. Like the outer hair cells, the stereocilia (S) of the inner hair cell vary in height, are linked together via tip and side links, and contain rootlets that project to and through the cuticular plate. The stereocilia, however, are not attached to the underside of the tectorial membrane, and there are fewer rows of them than on the outer hair cell. AF = afferent nerve fiber; EF = efferent nerve fiber; IB = inner border cell; IPh = inner phalangeal cell; SC = subsurface cisterna.

inner and outer pillar cells are major structural cells of the organ of Corti that are filled with microfilaments and microtubules for strength. The Deiters’ cell is a highly evolved epithelial cell whose purpose is to support the OHC and buffer its environment (see Figure 1-7). 45 The architectural arrangement of the OHC and Deiters’ cell is such that the Deiters’ cell acts as a cable to the rigid beam structure of the OHC, allowing contraction but limited expansion of the OHC.

8


Inner Hair Cells and Supporting Cells The IHCs are responsible for the signal that is sent to the brain via the primary afferent nerve for hearing, the auditory nerve.46 The IHCs are goblet-shaped cells that lie in a single row on the modiolar side of the organ of Corti and basilar membrane (see Figure 1-10). Situated at the lip of the osseous spiral limbus, on the modiolar side of the pillar cells and tunnel of Corti, beneath and slightly lateral to Hensen’s stripe, the stereocilia of the IHCs are positioned to receive the most refined fluid wave action during sound stimulation. Unlike the OHCs, the IHC is not in direct contact with the tectorial membrane.47–49 Therefore all of the input to the cell through the stereocilia occurs via fluid mechanics. The IHC stereocilia differ from OHC stereocilia in that they are arranged in more of a “U” or collapsed “W” shape on the surface of the cell, which correlates with a smaller, more ovoid shape to the cuticular plate within which they are embedded (see Figure 1-9).50 Like the OHC, the cuticular plate of the IHC has a fonticulus, occupied by a basal body, and cytoplasmic gaps that run through the cuticular plate (see Figure 1-10).39 The body of the IHC is shaped like a flask, with the apical neck portion filled with mitochondria, Golgi apparatus, endoplasmic reticulum, lysosomes, and multivesicular bodies.51 A cytoplasmic cistern lines the sides of the cell apically, to midcell region. The synaptic region basal to the cell nucleus is filled with synaptic specializations, as one IHC is in contact with an average of 20 type I afferent nerve fibers.52 Adult IHCs are rarely directly contacted by efferent nerve fibers. This contact when it occurs is more common in the apical region of the cochlea. Typically, efferent nerves associated with IHCs synapse on the afferent nerve endings. The cuticular plates of the IHCs and OHCs, along with their supporting cells, comprise the reticular lamina. A thick coat of proteoglycans covers the structures at the endolymphatic surface and protects the cell bodies of the organ of Corti from the otherwise toxic components of the endolymph.53 Neural Structure Cochlear innervation includes afferent and efferent innervation to the organ of Corti plus a sympathetic innervation system.54–57 Organ of Corti innervation is composed of four types: type I and type II afferent nerves and the efferent nerves, the medial and lateral olivocochlear bundles. The type I afferents predominantly innervate the IHCs. In humans, each IHC synapses with an average of 20 afferent nerve fibers.52 The exact numbers vary, but not in any base-to-apex gradient. Often the highest concentration of afferent fibers is in a region of the basilar membrane that represents the most sensitive frequency for a given species. 57 The type I spiral

ganglion cells are the most numerous, representing 90 to 95% of the total ganglion within the modiolus. The IHC afferent specializations are characterized by synaptic bars or ribbons surrounded by clear, irregularly shaped vesicles (see Figure 1-10).52 The presence of synaptic ribbons indicates that a high number of synaptic vesicles are released onto the nerve ending during activation of the hair cell. Type I spiral ganglion are bipolar cells with peripheral endings that radiate from the modiolus to the IHCs, which lie close to the nerve cell soma. The central endings of type I fibers bifurcate twice before entering three different areas of the ipsilateral cochlear nucleus in the brainstem.58 The type I fibers are myelinated from the central projection to the periphery up to the point where they leave the modiolus to cross the basilar membrane and enter the organ of Corti. The first node of Ranvier occurs near the organ of Corti. These nerve cells differ in their firing characteristics depending, in part, on their spontaneous activity.59,60 Each IHC is connected to nerve cells that represent all levels of spontaneous activity. The spontaneous activity of a nerve cell is dictated by the nerve’s refractory property (the time it takes for a nerve cell to “reset” itself after firing). Low, intermediate, or high spontaneous rate neurons lie within specific areas beneath the IHC. High spontaneous rate fibers tend to synapse on the lateral side of the IHC and low spontaneous rate fibers synapse on the medial side, closest to the habenulae perforatae. The major neurotransmitter released by the IHC is glutamate (Table 1-2).61,62 Two major types of non–Nmethyl-D-aspartate (non-NMDA) glutamate receptors have been found on the postsynaptic nerve ending, kainate and α-amino-3-hydroxy-5-methylisoxazole-4propionic acid (AMPA), AMPA being the more abundant of the two. Non-NMDA glutamate receptors are ion channels that are thought to be responsible for rapid afferent transmission. Several non-NMDA glutamate receptors have been found in the organ of Corti. They include glutamate receptor 3 (GR3), GR4, GR5, GR6, and kainate receptor KA1.63 Receptors for AMPA, quisqualate, and domoic acid have also been indicated via changes in neural output after application of these neurochemicals.64 NMDA receptors have also been found.65 The role of NMDA receptors is typically modulation of the ionic glutamate receptors. P2X2 receptors for adenosine triphosphate (ATP) have been found on postsynaptic nerve endings, indicating a potential modulator role for ATP in these cells.66 There is also some evidence that the type I afferents contain metabotropic glutamate receptors. These types of receptors initiate a metabolic change within a cell that ultimately changes the cell’s firing characteristics. Their purpose within the auditory system is not yet clear.

Type II SG/OHC Synapse Neurotransmitters

Glutamate

OHC Receptors

?

Type II Dendrite Receptors

GluR

Type I SG/IHC Synapse Neurotransmitters

Glutamate

IHC Receptors

GluR

Type I Dendrite Receptors

Non-NMDA glutamate receptors

?

?

MOC Efferent Receptors

GABAR

LOC Efferent Receptors

NPR

NPR

AChr (α-9 and α-10 nicotinic)

GABAR

AChr (α-7 nicotinic)

IHC Receptors

Dopamine

Dynorphins

Enkephalins

OHC Receptors

GABA

GABA Opioid peptides:

Acetylcholine CGRP

Acetylcholine CGRP

MOC/OHC Synapse Neurotransmitters

OHC

The complete list of neurotransmitters and receptors is still being compiled, but the main excitatory afferent neurotransmitter for both type I and type II spiral ganglion (SG) is most likely glutamate. The major neurotransmitters for the efferent system are acetylcholine, calcitonin gene-related peptide (CGRP), and γ-aminobutyric acid (GABA). AChr = acetylcholine receptor; AMPA = α-amino-3-hydroxy5-methylisoxazole-4-propianic acid; ATP = adenosine triphosphate; GluR = glutamate receptor; IHC = inner hair cell; LOC = lateral olivocochlear efferent; MOC = medial olivocochlear efferent; NMDA = N-methyl-D-aspartate; NPR = neuropeptide receptors; OHC = outer hair cell. Information for this table was derived from several sources.61,62

Metabotropic receptors

P2X2

ATP receptors

NMDA receptors

Domoic acid

Quisqualic acid

Kainic acid

AMPA

IHC

OHC

IHC LOC/IHC Synapse Neurotransmitters

Efferent Nerves

Afferent Nerves

Table 1-2 Neurotransmitters and Receptors within the Organ of Corti

Anatomy and Physiology of the Cochlea 9

10


Approximately 5% of the spiral ganglion population is composed of type II afferent fibers.55 These cells are bipolar or pseudobipolar in character with little or no myelination and are covered by a simple sheath of Schwann’s cells, surrounded by a basement membrane. Their peripheral targets are the OHCs. The type II spiral ganglion fibers, called outer spiral fibers, cross the floor of the tunnel of Corti to extend out under the three rows of OHCs. From there they make contact with the base of several of the OHCs as they extend up to 0.6 mm apically.67 The central target of these neurons is the shell region of the cochlear nucleus, which projects to the superior olivary complex.68 One function of the type II spiral ganglion is believed to be a feedback loop to the OHCs. The neurotransmitters at the type II synapse are still being elucidated, but glutamate is a probable candidate (see Table 1-2).63,69 The cochlear efferent system is composed of two separate fiber tracts, the medial olivocochlear bundle (MOC) and the lateral olivocochlear bundle (LOC) fibers.70 The MOC efferents originate predominantly from the contralateral medial superior olivary nucleus in the brainstem, with some ipsilateral contribution.71 The MOC fibers are myelinated fibers that become unmyelinated at the habenulae perforatae, where they cross the tunnel of Corti at midlevel (see Figure 1-7) and run parallel to the three rows of OHCs to form the outer spiral bundle.71,72 They synapse with a multitude of OHCs along the way. The uncrossed (ipsilateral) MOC efferents synapse almost equally with OHCs from the apex of the cochlea to the base, with a slight increase in synapse count in the midregion of the cochlea. The crossed (contralateral) MOC efferents preferentially synapse with the more basal, highfrequency components of the cochlea. In fact, the fibers typically enter the OHC area basally and branch out toward the apex.72 How the MOC efferents function in hearing is still being elucidated, but it has been shown that activation of the crossed MOC efferents can change the response to evoked otoacoustic emissions (OAEs) and can decrease the size of compound action potentials.73 The LOC efferents are unmyelinated and originate in the lateral superior olivary nucleus in the brainstem.70 From there they can run either ipsilateral or contralateral to the nucleus. All LOC efferent axons terminate, without much branching, on afferent nerve endings in the vicinity of the IHCs. The ipsilateral projections synapse in essentially equal numbers from base to apex. The crossed projections synapse predominantly in the apical, lower-frequency regions of the cochlea.72 The IHC synaptic ultrastructure for efferent and afferent nerve endings closely resembles that of standard configurations. This is not the case for OHC synaptic ultrastructure. Many large efferent endings

synapse directly onto the OHCs (see Figure 1-8). Also, evidence of reciprocal synapses has been found in human OHCs. 74 At nerve endings with reciprocal synapses, both afferent and efferent configurations can be found on the same nerve ending. The roles of the efferent and afferent innervations to the OHC are not completely clear. However, there is evidence that activation of the efferent innervation changes the activity of the OHCs, as recorded by distortion product OAEs and compound action potentials.73 Sympathetic nerve fibers in the cochlea are predominantly noradrenergic and enter the cochlea along the cochlear artery in the internal auditory meatus.56 Many of the fibers stay associated with the vasculature, but some branch out into the cochlear nerve, spiral ganglion, and Rosenthal’s canal. Some radiate out into the osseous spiral lamina as far as the habenulae perforatae. There is some disagreement whether fibers that are not associated with vessels reach into the organ of Corti itself, but the diffusion properties of norepinephrine allow it to influence nearby neural activity without direct contact. Therefore, the organ of Corti activity may be directly affected by sympathetic nerve activity.

COCHLEAR PHYSIOLOGY The anatomic arrangement of structures within the cochlea is essential for the transformation of acoustic signals into neurochemical signals that can be interpreted by the brain as sound. Equally important is the physiology of the system. Cochlear physiology is usually divided into two types: the cellular physiology of the cochlear tissues, involving biochemistry and molecular biology of the cells, and the electrophysiology of neural transduction, involving measurement of neural activity of the neuroepithelial cells. The cellular physiology, molecular biology, and biochemistry of tissues are directly affected by ototoxic agents, with consequences to the anatomy and electrophysiology of the system. These aspects vary by cell and by tissue. Therefore, ototoxic agents affect different tissues, depending on the agent’s effect on the biology of the cell and access to the vulnerable tissue. The cochlear duct houses the tissue structures that are responsible for sensorineural transduction. Two major structures of the cochlear duct are the organ of Corti and the lateral wall (see Figure 1-3). The lateral wall consists of the spiral ligament and the stria vascularis. The stria vascularis and spiral ligament are responsible for maintaining the high potassium levels in the endolymph as well as the endocochlear potential required for the sensorineural transduction.75 The spiral ligament contains four different types of fibrous tissue that are important for ion transport activity.76 The type II fibrocytes near the basilar membrane and the type I fibrocytes adjacent to the stria vascularis are important


for the recycling of K+ from the area of the organ of Corti back to the stria. Fibrocytes that are in the region of the spiral ligament that is adjacent to the basilar membrane contain stress fibers that can supply tension to the basilar membrane. These fibrocytes are thought to play an active role in basilar membrane mechanics.77 There are three major cell types in the stria vascularis: marginal, intermediate, and basal cells (see Figure 1-6). Strial marginal cells are polar cells with a flat hexagonal surface adjacent to the endolymphatic space and long interdigitating projections that extend deep into the strial tissue. Tight junctional complexes on the endolymphatic side of the cells isolate the strial intracellular space from the endolymphatic space. 17,23,78 Marginal cells contain a high concentration of Na+–K+ adenosine triphosphase (ATPase) and numerous mitochondria, which are required for pumping K+ into the endolymph. Strial basal cells line the spiral ligament side of the stria vascularis. They are in direct communication with spiral ligament cells and strial intermediate cells via gap junctions. Tight junctions between the basal cells isolate the strial intracellular space from the spiral ligament, which is key for endocochlear potential production. 78 Strial intermediate cells are neural crest–derived cells that are situated between the marginal and basal cells, nearest the vasculature. These cells function as a buffering system as they typically react first to insults, and they are essential for endocochlear potential generation.79,80 The stria vascularis is important in secreting K+ into the endolymph and in creating a large potential across the epithelia that lines the scala media (Figure 111). A standing current flows through the cochlear tissues, between the current source (the stria vascularis) and the two current sinks (the perilymphatic fluids of the scala vestibuli and the scala tympani). This current is required for sensorineural transduction by hair cells.78,81 The composition of the endolymphatic and perilymphatic fluids is crucial in this process as they provide the charge-carrying molecules and the ionic milieu for current to flow and transduction to occur. The endolymphatic fluid space is physically separated from the perilymphatic spaces of scala tympani and scala vestibuli by a series of tight junction complexes that rim the endolymphatic fluid space, along with the series of tight junction complexes that isolate the stria vascularis from its surroundings (see Figure 1-11, bold lines).23,46 The composition of the endolymph is poisonous to most cells’ external surfaces, and isolation of the endolymph from the rest of the tissues is needed to protect the tissue from endolymphatic poisoning.53 Isolation of endolymph from surrounding tissues is very important to the homeostasis of the cochlea. Therefore, physiologic mechanisms are in place to minimize intermixing of fluids even during scar formation in the organ of Corti.42

11

P

RM

EL

St

SL

O of C P

Figure 1-11 The generation of the standing current in

the cochlea. K+ is pumped into the endolymph (EL) via Na+–Ka+ ATPase in the marginal cells of the stria vascularis (St). A barrier exists between the EL and perilymphatic (P) spaces created by tight junctional complexes between the cells lining the endolymphatic space and stria vascularis (bold lines). Some K+ can leak across Reissner’s membrane (RM), and it crosses the organ of Corti (O of C) during hair cell transduction. K+ is then taken up by cells lining the perilymphatic space and travels intracellularly to fibrocytes of the spiral ligament (SL, long arrows). K+ then passes into the stria vascularis through gap junctions between the fibrocytes, the strial basal cells, and the strial intermediate cells to supply the K+ for uptake by the strial marginal cells (small arrow).

The osseous spiral lamina and organ of Corti house the auditory nerves and neuroepithelia (see Figure 1-3). The two types of neuroepithelial cells within the organ of Corti, the IHCs and OHCs, have their apices exposed to endolymph and their bases exposed to perilymph-like fluids (cortilymph).45 The OHC has a unique ultrastructure of cisterns along its sidewalls that are continuous with the Hensen’s body in the apical region and with the synaptic cistern at the base. This cistern is involved with mechanical cell changes that modify the wave action of the basilar membrane. The walls of these cells are also responsible for maintaining the high hydrostatic pressure within the cell. Deiters’ cells are the main supporting cells of the OHCs (see Figure 1-8). Deiters’ cells have been shown to undergo conformational changes after injection of current in vitro similar to what has been shown for OHCs. This indicates a strong interaction between the two cell types. Thus, OHCs are “cradled” by the Deiters’ cell in

12


the synaptic region of the cell, which indicates a buffering role for this type of cell. Even with this buffering system in place, OHCs tend to exhibit the first sign of degeneration after noise or other ototoxic insults to the organ of Corti.23 Similar to the OHC supporting cells, the cells surrounding the IHC are thought to buffer the IHC environment.46 They have also been suggested to be essential for maintenance of the afferent nerves. The loss of afferent nerve endings after an insult to the organ of Corti is often accompanied by the loss of these supporting cells, as well as the IHCs.57 As stated earlier, movement of OHC stereocilia results in conformational changes in the cell and, secondarily, release of neurotransmitters at the synapse. This differs from IHC activation, which results in a much larger release of neurotransmitters into the synaptic gap between the sensory cell and the surrounding nerves. Unlike OHC–type II afferent nerve interactions, much more is known about IHC–type I afferent nerve interactions. The predominant neurotransmitter at the IHC afferent synapse is glutamate. Glutaminergic neurons are subject to excitotoxicity, which occurs when too much glutamate is released into the synaptic gap at one time, causing swelling in the postsynaptic cell.58 Like glutaminergic central nervous system (CNS) neurons, overstimulation of the IHC can cause excitotoxicity in the postsynaptic nerve ending. Unlike CNS neurons, where excitotoxicity affects the soma, the portion of the type I neuron that is affected in cochlear excitotoxicity is the unmyelinated part of the fiber, peripheral to the habenula perforata and relatively distant from the cell soma. In many cases of cochlear excitotoxicity the peripheral process will recover as long as the IHC is intact. If the IHC and supporting cells are lost during the insult, recovery is lost. However, animal experiments have shown that it is possible to stimulate regrowth of the dendrites by applying neurotrophic factors, such as glial cell line–derived neurotrophic factor (GDNF) and brain-derived neurotrophic factor (BDNF), in animals.58 Although not considered part of the organ of Corti, the tectorial membrane plays an important role in the transduction that takes place there. Fluid dynamics at the IHC stereocilia are important in the sensorineural transduction of an acoustic stimulus.46 The movement of the reticular lamina in relation to the tectorial membrane dictates fluid dynamics. The movement of the basilar membrane and OHC activity alters the association of the reticular lamina with the tectorial membrane. The molecular components of the tectorial membrane are key to this association. The type II and type IX collagen fibers contribute to its tensile strength and resist stretch.82 They are aligned in the membrane in an apical–radial direction, consistent with the movement of the membrane in response to

sound stimulation. Because the glycosaminoglycans and proteoglycans in the membrane have a high negative-charge density, they create a highly hydrated, loose gel matrix that resists the forces of compression and radial expansion that are endured by the membrane during sound stimulation.46 Alterations in tectorial membrane hydration or proteoglycan content can alter its shape and mechanical properties, which could alter the membranes relationship with the reticular lamina during sound stimulation, thus altering sound transduction.47

ELECTROPHYSIOLOGY Sensation in the auditory system, as well as every other sensory system, relies on neural conduction of chemoelectric potentials, resulting from a flow of ions through channels in the membranes of either neurons or the receptor cells.83,84 These electrical events create electric fields with varying electric potentials that can be tracked through the use of electrodes, placed either locally (connected to the cell in patch clamp procedures or next to the cell in near field recordings) or remotely (placed on the surface of the head). Only the more clinically relevant remote fields will be discussed here. Electric events occur in the auditory system on an ongoing basis. The electrical potentials can be divided into two types: spontaneous responses (activity that occurs in absence of an external stimulus) and stimulus-evoked responses (which occur in response to acoustic stimulation). Here we focus on stimulusevoked responses that occur in the cochlea and primary auditory nerve (the eighth nerve). This neural activity is measured using electrodes sensitive to changes in electrical currents that occur when a hair cell or neuron is activated. Electrophysiology can be used to study cochlear responses to sound stimulation. Cochlear Potentials There are several steps involved in the auditory signal transduction that generates electrochemical potentials within the organ of Corti. These potentials can be recorded and used to observe various components of electrophysiology within the cochlea. The potentials included in this chapter are cochlear microphonics (CM), the summating potential (SP), the compound action potential (CAP), and the auditory brainstem response (ABR). Cochlear Microphonics CM is an evoked response that follows the spectral components of the stimulus, similar to how a microphone transduces sound into analog changes in electric current. This potential is alternating current (AC) in nature and follows the frequency and amplitude fluctuations of the acoustic stimulation of moderate amplitude. Weaver and Brey, who originally recorded CM,


Summating Potential SP also arises from activity in the organ of Corti.86 SP is recorded as a direct current (DC), unidirectional shift in potential in response to a stimulus that lasts as long as the stimulus. This shift can be either positive or negative, depending on the parameters of the stimulus and recording techniques. This potential is also a consequence of hair cell activation and is thought to arise from the hair cell transduction process. Studies have shown that both IHCs and OHCs contribute to this potential, but IHCs are the major contributors. 20 Abnormal SPs occur when there is loss of IHCs and OHCs or when there is a bias in the position of the basilar membrane, such as in endolymphatic hydrops.87 Compound Action Potential CAP is a measure of the synchrony of firing for fibers of the auditory nerve. Each nerve fiber is connected to an auditory nerve cell that “fires” or creates an action potential when the cell reaches the threshold of firing.88 These cells fire even in an unstimulated, or resting, state, but the activity is more or less random and asynchronous. They fire in synchrony when they are driven by an external stimulus. This pattern of activity is what the CAP measures. A normal CAP occurs at a specific time after presentation of a stimulus (latency) and with a large negative-going voltage peak (measured in microvolts), followed shortly by a second negative peak.89 The latency and the amplitude of the first peak vary with intensity of the stimulus. The minimum intensity sound to create the CAP is called the CAP threshold for that sound. The CAP threshold for puretone, short-duration sounds varies according to the frequency of the sound. The inner ear sensitivity to sound varies with frequency, and different species demonstrate different sensitivities. Loss of sensitivity of the CAP could be because of an attenuation of the sound stimulus reaching the inner ear, a problem with the hair

-

I III

V

V L 80dB

I

III

+ AI .25uV

+ .25uV

thought that the source of the current fluctuations was the auditory nerve.84 Later work done by Tasaki and colleagues using recording electrodes within the cochlea demonstrated that the source of CM is the hair cells, within the organ of Corti.81 As there are three times as many OHCs as IHCs, the OHCs are the major contributor to CM.85 CM represents the vector sum of activity from a large number of hair cells. CM amplitude varies with the intensity of the input stimulus and does not appear to adapt over the length of the stimulus signal, unlike some of the other potentials.21 CM is initiated by displacement of the basilar membrane, and therefore the spatial distribution of CM activation coincides with area of peak vibration of the membrane, which changes according to frequency and intensity of the input signal.

13

R 80dB

LATENCY 4.00 ms / div

-

Figure 1-12 A normal human auditory brainstem response

(ABR) recorded on a standard clinical evoked potential recorder at 80 dB nHL stimulus intensity. Results from the left ear are presented on the left and the right ear presented on the right. Positive potential (+) is up and measured in 0.25 microvolt units. Latencies are indicated at the abscissa in 4 millisecond divisions. I, III, and V denote the first, third, and fifth peak in the wave form. Courtesy of Judy DeMorest, AuD, University of Texas at Dallas Callier Center for Communication Disorders.

cells in the organ of Corti, or a problem with the nerve cells themselves. Brainstem Potentials Brainstem potentials can also be recorded using electrode placements similar to those used for recording remote cochlear potentials. The main protocol for studying brainstem potentials that are associated with cochlear function is the auditory-evoked brainstem potential, also called the ABR.90 Measurement of the ABR is used to observe the activity of the lateral and medial aspects of the auditory nerve and the auditory brainstem, up to about the inferior colliculus.90 The ABR is preferred clinically over measurement of the CAP because it is more robust and gives more information about nerve function. The graph of a normal human ABR displays both positive and negative voltage peaks occurring at different times after the stimulus (Figure 1-12). The first positive peak of the ABR is similar in latency and intensity to the first negative peak of the CAP. In humans, this first positive peak is thought to arise from the synchronous activity of the peripheral part of the auditory nerve, with the second peak arising from the central component of the auditory nerve as it synapses with the cochlear nucleus. These two peaks merge in smaller animals because of the proximity of the peripheral and central components of the nerve, which reduces the latencies between the two peaks and is harder to separate spatially on the smaller heads of the animals. The third peak of an ABR graph comes from the upper brainstem in humans, and the latency between the first and fifth peaks is used clinically to identify lower brainstem pathology.

OTOACOUSTIC EMISSIONS AND BASILAR MEMBRANE MECHANICS Basilar Membrane Mechanics Sensorineural transduction of acoustic waves into neural signals occurs in the cochlea. The structure of

14


the basilar membrane and its resonant properties are key to the physical separation of an acoustic wave into its component frequencies. In 1952, George von Békésy was the first to observe the wave motion of the basilar membrane in human and animal cadaver cochleae in response to an acoustic stimulus.21 The inner ear is essentially closed fluid-filled compartments with a piston-like stapes footplate associated with one fluid compartment (scala vestibuli) and an elastic membrane (round window membrane) stretched across the communication route between the fluids of another fluid compartment (scala tympani) and the middle ear space. At the stapes footplate, acoustic waves transmitted by the ossicles of the middle ear are transferred from a wave traveling in air to one traveling in fluid. The in-and-out piston action of the stapes (initiated by the transfer of an acoustic wave through the tympanic membrane and ossicular chain to the stapes) and the elasticity of the round window membrane act on the essentially incompressible fluids of the inner ear to create a wave of pressure fluctuations through the cochlea. As a result of the acoustic wave traveling through fluid, the basilar membrane is set in motion. This is the motion von Békésy observed.21 The organ of Corti lies on the basilar membrane, and the hair cells are situated so that they are stimulated when the basilar membrane is forced to vibrate as the wave arrives.91 The basilar membrane is shaped so that the low-frequency components of the acoustic wave set the more apical areas into motion, with progressively higher frequency components setting progressively more basal components of the basilar membrane into motion.21,92 The representation of frequencies is laid out along the basilar membrane in a continuous manner, similar to that of a piano keyboard. Factors affecting basilar membrane motion other than the properties of the basilar membrane itself are the inertia of the fluid mass within the cochlea, the frictional resistance created by the fluid motion and stiffness of the associated tissues (the round window membrane, the ligaments of the stapes, Reissner’s membrane, the organ of Corti, and the spiral ligament), and an active component associated with the action of the OHCs.92 The extents to which these properties affect the traveling wave motion are still under investigation. The motion that von Békésy observed was a broadtipped, wave-like motion with a peak activity corresponding to the frequency of the stimulus.21 The wave appeared to travel from the base of the cochlea toward the apex. The more basal portions of the basilar membrane are lighter and stiffer than the apex. However, the basal portions are set into motion with relative ease compared with the apex because of the increase in mass toward the apex.92,93 As the factor of mass is overcome,

the wave of vibration “travels” toward the apex. Highfrequency sounds do not travel far toward the apex, as their place of maximum vibration is at the base. Correspondingly, the lower the frequency of the stimulus, the farther toward the apex the wave travels, until it reaches a point of maximum vibration. The wave is quickly dampened apical to this point. The envelope of maximum vibration observed by von Békésy was extremely broad, which did not correspond with what had been learned by physiologic measures, that frequency tuning at the basilar membrane is sharp and the internal filters are narrow. Since von Békésy’s work, this discrepancy has been attributed to an active tuning system at the level of the basilar membrane that is responsible for sharpening the tuning.92 The OHCs have been shown to be the major component of this active tuning system, along with the efferent system that supplies the OHCs.94,95 The action of the OHCs alters the input to the IHCs, thus changing the output of the auditory nerve. This system is responsible not only for the sharpness of the tuning, but also for an increase in sensitivity of 30 to 40 dB sound pressure level over a damaged system that lacks OHC input.96,97 Otoacoustic Emissions Active events other than electric potentials occur within the cochlea and can be useful in examining the condition of the cochlea. OAEs are acoustic waves of various frequencies generated at the basilar membrane that can be measured with a microphone in the external auditory canal.98 The waves arise from the action of the basilar membrane as a consequence of an active, electromotile response occurring in the organ of Corti after OHC activation (active tuning system). The waves propagate through the middle ear in the opposite direction to that of sounds from the environment.99 OAEs are a reflection of the activity of the basilar membrane, basilar membrane mechanics, and the overall health of the organ of Corti. They can be measured with acoustic measurement equipment placed in the external auditory canal. OAEs can occur spontaneously, without intentional stimulation, or as a response to intentional stimulation (evoked).100 Non-audible spontaneous OAEs (SOAEs ) are typical in humans and are thought to arise from nonlinearities in the activity of the basilar membrane.101,102 SOAEs occur more often in right ears versus left and in women more often than men.102 Evoked forms of OAEs have proven useful clinically, as the nonlinearity of basilar membrane motion is linked to the active tuning process in the organ of Corti, thus the OHCs. Damage to the OHCs results in a loss or decreased amplitude of evoked OAEs.97,103 As OHCs are typically the most vulnerable cells in the organ of Corti to toxic insults, OAE tests can give

Anatomy and Physiology of the Cochlea Patient: JAD Birthdate: Comment:

On the basis of our present understanding of cochlear anatomy and physiology we have been able to demonstrate that

Ear: Left ID: Result: PASS

1. Ototoxic substances affect the anatomy and physiology of the auditory system. 2. The ability of ototoxic substances to affect specific tissues of the cochlea is dependent on the entry of the substance into the cochlear compartment that harbors that tissue and by its ability to interact with the physiology of that tissue. 3. Ototoxic substances enter the cochlear tissues via the vascular system, the CSF (via either the cochlear aqueduct or the internal auditory meatus), or via the middle ear, across the round window membrane. 4. Ototoxic substances can directly affect the tissues that comprise the neurosensory epithelia (organ of Corti). 5. Ototoxic substances can also affect the tissues that support the normal physiologic functioning of the organ of Corti (such as stria vascularis), thereby affecting the organ of Corti function in an indirect manner. 6. The most common anatomical change within the organ of Corti resulting from ototoxic damage occurs in the OHCs. The degree of change ranges from altered stereocilia anatomy or complete loss of the cell. 7. The preferred method for detection of OHC damage is the measurement of OAEs.

dB TE Signal NF Noise 0

-10

0

1

2

3

4

5

15

kHz

Figure 1-13 Transient evoked otoacoustic emissions

(TOAEs) recorded from the external ear canal from a normal adult human. abscissa = frequency of emission; NF = noise floor (black area), ordinate = level of emission in dB, relative to stimulus; TE = evoked emission (gray area). Courtesy of Judy DeMorest, AuD, University of Texas at Dallas Callier Center for Communication Disorders.

valuable, noninvasive information about the condition of the organ of Corti. There are two types of OAEs that are typically used clinically to test the health of the organ of Corti: transient evoked OAEs (TEOAEs) and distortion product OAEs (DPOAEs) (Figures 1-13 and 1-14). TEOAEs are typically produced using click stimuli, which represent a pan-frequency stimulus. This stimulus produces a robust response spectrum that represents the spectrum of the stimulus in a nonspecific manner.104 DPOAEs, on the other hand, are a product of a two-tone stimulus, which represent the basilar membrane activity a specific distance away from the two eliciting tones. The position along the basilar membrane that is represented is determined by the difference in frequency and amplitude of the two tones. DPOAEs allow the tester to examine the condition of the active tuning system (the OHCs) at specific points along the basilar membrane.104,105

CONCLUSION The better our understanding of the anatomy and general physiology of the cochlea, the more we can learn about the mechanisms involved in ototoxicity. The anatomy of the cochlea is fairly well understood. However, work in understanding the cellular and electrophysiology of the peripheral hearing organ is ongoing. Less yet is known about the complex physiology of the central auditory system. As more is understood about the normal functioning of these systems in the perception of simple and complex sounds, the better our understanding of the affects of ototoxic drugs will be.

Lastname Firstname Ear Date Result Left 30-Sep-03 Pass JAD Right 30-Sep-03 Pass JAD dB 60 40 20 0 -20 2

2

DP-Gram

DP-Gram

kHz F2 Frequency

Figure 1-14 Distortion product otoacoustic emissions (DPOAEs) recorded from the external canals from a normal adult human. Results from the left and right ears are on the left and the right side of the graph, respectively. The two top lines represent the input levels of the two test frequencies, the middle line represents the output level of the emission, and the bottom line represents the output level of the noise floor. Courtesy of Judy DeMorest, AuD, University of Texas at Dallas Callier Center for Communication Disorders.

16


8. The sensory receptor for the auditory nerve is the IHC, which synapses with type I spiral ganglion. Type I spiral ganglia are responsible for conveying the auditory signal to the CNS. 9. Loss of function of the type I spiral ganglion or the IHC results in loss of acoustic information from that portion on the basilar membrane. However, OHC loss will only reduce the sensitivity of that portion. The OHCs’ main function is to modify the acoustic information that reaches the IHCs. 10. Preferred methods of detection of auditory nerve activity are measurement of the CAP or measurement of the ABR.

SUMMARY • The effects of substances on the peripheral auditory system are caused either by a whole body reaction causing secondary affects on the inner ear or by the entry of substances into the inner ear where they affect specific cellular targets. Substances entering into the inner ear can cause toxicity by their interaction with the biochemistry of the tissues and physiology of individual cell types. In either case, the administration of an ototoxic agent ultimately results in observable, pathologic changes in the anatomy and electrophysiology of the auditory system. To understand the pathologic changes to peripheral auditory system, it is necessary to understand the systems’ normal anatomy and physiology. • The inner ear is a fluid-filled, neurosensory structure within the skull base that converts the mechanical, acoustic waves of sounds into chemoelectric signals that are transported to the brain via the auditory nerve. The health of this system can be directly observed in postmortem, histological preparations or by premortem observation of chemoelectric signals via electrophysiologic testing. • There are several fluid compartments in the cochlea: the endolymph of the scala media; the perilymph of the scala vestibuli, which is continuous with the perilymph of the scala tympani; the cortilymph; and the vascular system. Routes of entry into the cochlear tissues for ototoxic substances are through one or more of these fluids. Substances that enter the fluids either can affect the health of the cells that are directly involved in sensorineural transduction (the hair cells and nerve endings) or can affect other tissues within the cochlea that play supporting roles (such as stria vascularis). • The unique architecture of the inner ear structure allows for tuning of the sensory epithelium

(basilar membrane and organ of Corti) to the frequency and temporal components of sound. The specificity and sensitivity of tuning of the basilar membrane and organ of Corti in a normal ear involve an active mechanism, a major component of which is the mechanical action of the OHCs. The OHCs are especially susceptible to a variety of ototoxic insults, and the health of these cells can be monitored using otoacoustic testing or electrophysiologic testing.

REFERENCES 1. Schuknecht HF, Gulya AJ. Anatomy of the temporal bone with surgical implications. Philadephia (PA): Lea & Febiger; 1986. p. 130. 2. Smith CA, Lowry GH, Wu ML. The electrolytes of the labyrinthine fluids. Laryngoscope 1954;64: 141–53. 3. Anniko M, Wroblewski R. Ionic environment of cochlea hair cells. Hear Res 1986;22:279–93. 4. Feldman AM. Cochlear biochemistry. In: Brown RD, Daigneault EA, editors. Pharmacology of hearing. New York: Wiley; 1981. p. 51–80. 5. Waggeman P, Schacht J. Homeostatic mechanisms in the cochlea. In: Dallos P, Popper AN, Fay RR, editors. The cochlea. New York: Springer-Verlag; 1996. p. 151–3. 6. Mori N, Ninoyu O, Morgenstern C. Cation transport in the ampulla of the semicircular canal and endolymphatic sac. Arch Otorhinolaryngol 1987; 244:61–5. 7. Ninoyu O, Meyer zum Gottesberge AM. Ca++ activity in the endolymphatic space. Arch Otorhinolaryngol 1986;243:141–2 8. Hara A, Komeno M, Senarita M, et al. Effect of asphyxia on the composition of cationic elements in the perilymph. Hear Res 1995;90:228–31. 9. Anson BJ. The vestibular and cochlear aqueducts: their variational anatomy in the adult human ear. Laryngoscope 1965;75:1203–23. 10. Axelsson A. The vascular anatomy of the cochlea in the guinea pig and in man. Acta Otolaryngol Suppl 1968;243:3. 11. Anson BJ, Donaldson JA, Warpena RL, Winch TR. A critical appraisal of the anatomy of the perilymphatic system in man. Laryngoscope 1964;74:945. 12. Marchbanks RJ. Hydromechanical interactions of the intracranial and intralabyrinthine fluids. In: Ernst A, Marchbanks R, Samii M, editors. Intracranial labyrinthine fluids. Berlin: Springer Verlag; 1996. p. 51–61. 13. Carlborg B, Konradson KS, Farmer JC. Pressure relation between labyrinthine and intracranial fluids: experimental study in cats. In: Ernst A, Marchbanks R, Samii M, editors. Intracranial labyrinthine fluids. Berlin: Springer Verlag; 1996. p. 63–72.


14. Horner KC. Review: morphological changes associated with endolymphatic hydrops. Scanning Microscopy 1993;7:223–38. 15. Nomura Y, Hiraide F. Cochlear blood vessel. Arch Otolaryngol 1968;88:231–41. 16. Smith CA. Capillary areas of the membranous labyrinth. Ann Otol Rhinol Laryngol 1954; 63:435–47. 17. Duvall AJ, Quick CA, Sutherland CR. Horseradish peroxidase in the lateral cochlear wall: an electron microscopic study. Arch Otolaryngol 1971;93: 304–16. 18. Gorgas K, Jahnke K. The permeability of blood vessels in the guinea pig cochlea. II Vessels in the spiral ligament and stria vascularis. Anat Embryol 1974;146:33–42. 19. Smith CA. Vascular patterns of the membranous labyrinth. In: de Lorenzo AJD, editor. Vascular disorders and hearing defects. Baltimore: University Park Press; 1975. p. 1–18. 20. Pickles JO. An introduction to the physiology of hearing. London: Academic Press; 1988. p. 58. 21. von Békésy G. DC resting potentials inside the cochlear partition. J Acoust Soc Am 1952;24: 72–6. 22. Jahnke K. Die feinstruktur gefrigergaitzer Zellmambran-Haftstellen der Stria Vascularis. Anat Embryol 1975;147:189–201. 23. Duvall AJ, Klinkner A. Macromolecular tracers in the mammalian cochlea. Am J Otolaryngol 1983;4: 400–10. 24. Forge A. Gap junctions in the stria vascularis and the effects of ethacrynic acid. Hear Res 1984;13: 189–200. 25. Spicer SS, Schulte BA. Differentiation of inner ear fibrocytes according to their ion transport related activity. Hear Res 1991;56:53–64. 26. Spicer SS, Schulte BA. The fine structure of spiral ligament cells relates to ion return to the stria and varies with place-frequency. Hear Res 1986;100: 80–100. 27. Slepecky NB. Cochlear structure. In: Dallos P, Popper AN, Fay RR, editors. The cochlea. New York: Springer-Verlag; 1996. p. 66–70. 28. Johnstone BM, Patuzzi R, Yates GK. Basilar membrane measurements and the traveling wave. Hear Res 1986;22:147–53. 29. Sellick PM, Patuzzi R, Johnstone BM. Measurement of basilar membrane motion in the guinea pig using the Mossbauer technique. J Acoust Soc Am 1982;72:131–41. 30. Santi PA, Lease MK, Harrison RD,Wicker EM. Ultrastructure of proteoglycans in the tectorial membrane. J Electron Microsc Tech 1990;15:293–300. 31. Thalmann I, Thallinger G, Comegys TH, Thalmann R. Collagen II: the prodominate protein of

32.

33.

34.

35. 36.

37.

38. 39. 40.

41.

42.

43.

44.

45.

46.

47.

48.

17

the tectorial membrane. Otorhinolaryngology 1985;48:116–23. Thalmann I, Thallinger G, Comegys TH, et al. Composition and supramolecular organization of the tectorial membrane. Laryngoscope 1987;97: 357–67. Lim DJ, Lane WC. Cochlear sensory epithelium: a scanning electron microscopic study. Ann Otol Rhinol Laryngol 1969;78:827–41. Lim DJ. Fine morphology of the tectorial membrane: its relationship to the organ of Corti. Arch Otolaryngol 1972;96:199–215. Corti A. Recherches sur l’organe de l’ouie des mammifères. Z Wiss Zool 1851;3:109. Retzius G. Das gerhörorgan der wirbeltieve. II. Das Gerhörorgan der Reptilien, der Vögel, und der Säugetieve. Stockholm: Samson and Wallin; 1884. Wright A, Davis A, Bredberg G, et al. Hair cell distributions in the normal human cochlea. Acta Otolaryngol Suppl 1987;444:1–48. Smith CA. Ultrastructure of the organ of Corti. Adv Sci 1968;24:419–33. Lim DJ. Functional structure of the organ of Corti: a review. Hear Res 1986;22:117–46. Libermann MC. Chronic ultrastructural changes in acoustic trauma: serial section reconstruction of stereocilia and cuticular plates. Hear Res 1987; 26:45–64. Libermann MC, Dodds LW. Acute ultrastructural changes in acoustic trauma: serial section reconstruction of stereocilia and cuticular plates. Hear Res 1987;26:65–88. Raphael Y, Altschuler RA. Reorganization of cytoskeletal and junctional proteins during cochlear hair cell degeneration. Cell Motil Cytoskeleton 1991;18:215–27. Gulley RL. Intercellular junctions in the reticular lamina of the organ of Corti. J Neurocytol 1976;5:479–507. Kimura RS. Hairs of the cochlear sensory cells and their attachment to the tectorial membrane. Acta Otolaryngol 1966;61:55–72. Hudspeth AJ. Models of mechanoelectrical transduction by hair cells. Prog Clin Biol Res 1985; 176:193–205. Engström H, Wersäll J. Structure and innervation of the inner ear sensory epithelium. Int Rev Cytol 1958;7:535–85. Freeman DM, Cotanch DA, Ehsani F, Weiss TF. The osmotic response of the isolated tectorial membrane of the chick to isosmotic solutions: effect of Na+, K+, and Ca2+ concentration. Hear Res 1994;79:197–215. Takasaka T, Shinkawa H, Hashimoto H, et al. Highvoltage electron microscopic study of the inner ear.

18

49.

50.

51.

52.

53.

54.

55. 56. 57.

58.

59.

60. 61.

62.

63.

64.


Technique and preliminary results. Ann Otol Rhinol Laryngol Suppl 1983;101:1–12. Bohne BA, Harding GW, Nordmann AS, et al. Survival-fixation of the cochlea: a technique for following time-dependent degeneration and repair in noise-exposed chinchillas. Hear Res 1999;134: 163–78. Flock A, Kimura R, Lundquist PG, Wersäll J. Morphological basis of directional sensitivity of the outer hair cells of the organ of Corti. J Acoust Soc Am 1962;34:1351–5. Furness DN, Hackney CM. Comparative ultrastructure of subsurface cisternae in inner and outer hair cells of the guinea pig cochlea. Eur Arch Otorhinolaryngol 1990;247:12–5. Nadol JB Jr. Serial section reconstruction of the neural poles of hair cells in the human organ of Corti I: inner hair cells. Laryngoscope 1983;93: 599–614. Horner KC, Guilhaume A. Ultrastructural changes in the hydropic cochlea of the guinea pig. Eur J Neurosci 1995;7:1305–12. Spoendlin H, Lichlenstieger W. The adrenergic innervation of the labyrinth. Acta Otolaryngol 1966;16:423–34. Spoendlin H. Innervation densities of the cochlea. Acta Otolaryngol 1972;73:235–48. Spoendlin H. Autonomic innervation of the inner ear. Adv Otorhinolaryngol 1981;27:1–13. Spoendlin H, Schrott A. The spiral ganglion and the innervation of the human organ of Corti. Acta Otolaryngol 1988;105:403–10. Ryugo DK. The auditory nerve: peripheral innervation, cells body morphology, and central projections. In: Webster DB, Popper AN, Fay RR, editors. The mammalian auditory pathway: neuroanatomy. New York: Springer-Verlag; 1992. Libermann MC. Morphological differences among radial fibers in the cat cochlea. Hear Res 1980; 3:45–63. Libermann MC. Single-neuron labeling in the cat auditory nerve. Science 1982;216:1239–41. Raphael Y, Altschuler RA. Structure and innervation of the cochlea. Brain Res Bull 2003;60: 397–422. Sewell WF. Neurotransmitters and synaptic transmission. In: Dallos P, Popper AN, Fay RR, editors. The cochlea. New York: Springer-Verlag; 1996. p. 503–14. Ryan AF, Brumm D, Kraft M. Occurrence and distribution of non-NMDA glutamate receptor mRNAs in the cochlea. Neuroreport 1991;2:643–6. Kuriyama H, Jenkins O, Altschuler RA. Immunocytochemical localization of AMPA selective glutamate receptor subunits in the rat cochlea. Hear Res 1994;80:233–40.

65. Kuriyama H, Albin RL, Altschuler RA. Expression of NMDA-receptor mRNA in the rat cochlea. Hear Res 1993;69:215–20. 66. Housely GD, Janjhan R, Raybould NP, et al. Expression of the P2X(2) receptor subunit of the ATP-gated ion channel in the cochlea: implications for sound transduction and auditory neurotransmission. J Neurosci 1999;19:8377–88. 67. Schuknecht HF. Pathology of the ear. 2nd ed. Philadelphia (PA): Lea & Febiger; 1993. p. 66–70. 68. Morgan YV, Ryugo DK, Brown MC. Central trajectories of type II (thin) fibers of the auditory nerve in cats. Hear Res 1994;79:74–82. 69. Altschuler RA, Sheridan CE, Horn JW, Wenthold RJ. Immunocytochemical localization of glutamate immunoreactivity in guinea pig cochlea. Hear Res 1989;42:167–73. 70. Warr WB. Efferent components of the auditory system. Ann Otol Rhinol Laryngol Suppl 1980; 89(5 Pt 2):114–20. 71. Guinan JJ Jr, Warr WB, Norris BE. Topographic organization of the olivocochlear projections from the lateral and medial zones of the superior olivary complex. J Comp Neurol 1984;226:21–7. 72. Ishii D, Balogh K Jr. Distribution of efferent nerve endings in the organ of Corti: their graphic reconstruction in cochleae by localization of acetylcholinesterase activity. Acta Otolaryngol 1968; 66:282–8. 73. Guinan JJ Jr. Physiology of olivocochlear efferents. In: Dallas P, Popper AN, Fay RR, editors. The cochlea. New York: Springer-Verlag; 1996. p. 435–67. 74. Nadol JB Jr. Serial section reconstruction of the neural poles of hair cells in the human organ of Corti II: outer hair cells. Laryngoscope 1983; 93:780–91. 75. Zidanic M, Brownell WE. Fine structure of the intracochlear potential field. I. The silent current. Biophys J 1990;57:1253–68. 76. Spicer SS, Schulte BA. The fine structure of spiral ligament cells relates to ion return to the stria and varies with place-frequency. Hear Res 1996;100: 80–100. 77. Henson MM, Henson OW Jr, Jenkins DB. The attachment of the spiral ligament to the cochlear wall: anchoring cells and the creation of tension. Hear Res 1984;16:231–42. 78. Salt AN, Melichar I, Thalmann R. Mechanisms of endocochlear potential generation by stria vascularis. Laryngoscope 1987;97:984–91. 79. Hilding DA, Ginzberg RD. Pigmentation of the stria vascularis. The contribution of neural crest melanocytes. Acta Otolaryngol 1977;84:24–37. 80. Forge A. Observations on the stria vascularis of the guinea pig cochlea and the changes resulting from


81.

82.

83. 84.

85.

86.

87. 88.

89.

90.

91. 92.

93.

the administration of the diuretic furosemide. Clin Otolaryngol 1976;1:211–9. Tasaki I, Davis H, Eldredge DH. Exploration of cochlear potentials in guinea pig with a microelectrode. J Acoust Soc Am 1954;26:765–73. Slepecky NB, Cefaratti LK, Yoo TJ. Type II and type IX collagen form heterotypic fibers in the tectorial membrane of the inner ear. Matrix 1992;12: 80–96. Møller AR. Sensory systems. San Diego: Academic Press; 2003. p. 63–72. Wever EG, Bray CW. Action currents in the auditory nerve in response to acoustical stimulation. Proc Natl Acad Sci U S A 1930;16:344–50. Dallos P. Cochlear potentials and cochlear mechanics. In: Møller AR. Basic mechanisms in hearing. New York: Academic Press; 1973. p. 335–72. Dallos P, Schoeny ZG, Cheatham MA. Cochlear summating potentials: descriptive aspects. Acta Otolaryngol Suppl 1972;302:1–46. Møller AR. Hearing: its physiology and pathophysiology. New York: Academic Press; 2000. p. 99–100 Teas DC, Eldredge DH, Davis H. Cochlear responses to acoustic transients: an interpretation of whole-nerve action potentials. J Acoust Soc Am 1962;32:1438–59. Møller AR. Hearing: its physiology and pathophysiology. New York: Academic Press; 2000. p. 100–10. Møller AR. Hearing: its physiology and pathophysiology. New York: Academic Press; 2000. p. 299–322. de Boer E. Physical principles in hearing theory I. Phys Rep 1980;62:87–174. Pickles JO. An introduction to the physiology of hearing. 2nd ed. San Diego: Academic Press; 1992. p. 47–53. Lighthill J. Biomechanics of hearing sensitivity. J Sound Vib 1991;113:1–13.

19

94. Davis H. An active process in cochlear mechanics. Hear Res 1983;9:79–90. 95. Brownell WE, Bader CR, Bertrand D, de Ribaupierre Y. Evoked mechanical responses of isolated cochlear hair cells. Science 1985;227:194–6. 96. Kiang NY-S, Moxon EC, Levine RA. Auditorynerve activity in cats with normal and abnormal cochleas. In: Wolstenholme GEW, Knight J, editors. Sensorineural hearing loss. CIBA Foundation Symposium. London: Churchill; 1970. p. 241–68. 97. Dallos P. The active cochlea. J Neurosci 1992;12: 4575–85. 98. Pickles JO. An introduction to the physiology of hearing. 2nd ed. San Diego: Academic Press; 1992. p. 147–57. 99. Kemp DT. Stimulated acoustic emissions from within the human auditory system. J Acoust Soc Am 1978;64:1386–91. 100. Probst R. Otoacoustic emissions: an overview. Adv Otorhinolaryngol 1990;44:1–91. 101. Kemp DT. Evidence of mechanical nonlinearity and frequency selective wave amplification in the cochlea. Arch Otorhinolaryngol 1979;224:37–46. 102. Bright K. Spontaneous otoacoustic emissions. In: Robinette MS, Glattke TJ, editors. Otoacoustic emissions: clinical applications. New York: Thieme Medical Publishers, Inc; 1997. p. 46–62. 103. Probst R, Lonsbury-Martin BL, Coats A. Otoacoustic emissions in ears with hearing loss. Am J Otolaryngol 1987;8:73–81. 104. Robinette MS, Glatte TJ. Otoacoustic emissions. In: Roeser RJ, Valente M, Hosford-Dun H, editors. Audiology diagnosis. New York: Thieme Medical Publishers, Inc.; 2000. p. 503–9. 105. Lonsbury-Martin BL, Martin GK, Whitehead ML. Distortion product otoacoustic emissions. In: Robinette MS, Glatte TJ, editors. Otoacoustic emissions: clinical applications. New York: Thieme Medical Publishers, Inc.; 1997. p. 83–109.

CHAPTER 2

Physiology of the Vestibular System John A. Rutka, MD, FRCSC

Within the inner ear are specialized sensory receptors responsible for the perception of the forces associated with head movement and gravity. Control centers within the brainstem integrate this information along with other biologic signals derived from vision and proprioceptive sensors in the final determination of an individual’s orientation in three-dimensional space.1 Although anatomically developed and responsive at birth, the vestibular system matures along with other senses in the first 7 to 10 years of life.2 Recognition of the head’s movement relative to the body is provided by the linear (otolithic macula) and angular (semicircular canals) acceleration receptors of the inner ear. Electrical activity generated within the inner ear travels along the vestibular nerve (primary afferent neuronal pathway) to the central vestibular nuclei of the brainstem, forming second-order neuronal pathways that become the vestibulo-ocular reflex (VOR), the vestibulospinal tracts, and the vestibulocerebellar tracts. Pathways derived from vestibular information also travel to the brainstem emetic centers, which serves to explain vegetative symptoms such as nausea, vomiting, and perspiration that a patient typically experiences following an acute unilateral vestibular loss (Figure 2-1). Disruption of peripheral (inner ear and the vestibular nerve) or central vestibular pathways as a result of, for example , trauma, ototoxicity, or surgical deafferentation leads to the patient experiencing a distortion in orientation. The patient often uses the term “dizziness,” which is an all-encompassing yet relatively nonspecific term that can include symptoms such as giddiness, lightheadedness, and floating sensation. Clinically, the term “vertigo” is best suited to describe a precise type of dizziness—a hallucination of movement involving oneself (subjective vertigo) or the surrounding environment (objective vertigo) that is apt to occur when there is an acute interruption of vestibular pathways.1

Of all the human vestibular pathways, the VOR remains the most important and most studied. At its simplest level the VOR is required to maintain a stable retinal image with active head movement. When an active head movement is not accompanied by an equal but opposite conjugate movement of the eyes, retinal slip occurs. When the VOR is affected bilaterally (as could occur from systemic aminoglycoside poisoning) patients characteristically complain of visual blurring with head movement, better known as oscillopsia, in addition to having significant complaints of imbalance and ataxia.3 True vertigo is typically not a feature of a bilateral peripheral vestibular loss.

Oculomotor nuclei

Vestibulo-ocular Pathways Medial longitudinal fasciculus Reticular formation S

S

L

L D

Vestibulospinal Pathways

Utricle

D M

M Vestibular nucleus Medial longitudinal fasciculus

Other Pathways (not demonstrated)

• To Emetic Centers • To Vestibulocerebellum

Figure 2-1 Schematic representation of the vestibular system

and its pathways.

Physiology of the Vestibular System

21

Superior Posterior Vestibule

Semicircular Canals

Lateral

Superior Vestibular Nerve Inferior Vestibular Nerve Utricle Saccule Figure 2-2 Anatomic organization of the peripheral vestibular system (vestibular end-organs and the vestibular nerve).

Reproduced with permission from Solvay Pharma, Canada.

ANATOMY OF THE VESTIBULAR SYSTEM Peripheral Vestibular System The peripheral vestibular system includes the paired vestibular sensory end-organs of the semicircular canals (SCCs) and the otolithic organs. These receptors are found within the fluid-filled bony channels of the otic capsule (the dense endochondral-derived bone that surrounds the labyrinth and cochlea) and are responsible for perception of both the sense of position and motion. The vestibular nerve (both superior and inferior divisions of the VIIIth nerve) is the afferent connection to the brainstem nuclei for the peripheral vestibular system (Figure 2-2). Perception of angular accelerations is chiefly the responsibility of the three paired SCCs (superior, posterior, and lateral). Within the ampullated portion of the membranous labyrinth are the end-organs of the cristae, containing specialized hair cells that transduce mechanical shearing forces into neural impulses. Histologically the hair cells of the ampulla are located on its surface. Their cilia extend into a gelatinous matrix better known as the cupula, which acts like a hinged gate between the vestibule and the canal itself (Figure 2-3). The otolithic organs of the utricle and the saccule are found within the vestibule. Chiefly responsible for the perception of linear accelerations (eg, gravity, deceleration in a car), their end-organs consist of a flattened, hair cell–rich macular area whose cilia project into a similar gelatinous matrix. The matrix, however, differs from the matrix associated with the SCCs in its support of a blanket of calcium carbonate crystals better known as otoliths, which have a mean thickness of approximately 50 µm (Figure 2-4).4

Information from the vestibular end-organs is transmitted along the superior (which receives information from the superior, horizontal SCCs and utricle) and inferior (which receives information from the posterior SCC and saccule) divisions of the vestibular nerve. Although its role is primarily afferent in the transmission of electrical activity to the central vestibular nuclei of the brainstem, an efferent system does exist that probably serves to modify end-organ activity.5 Each vestibular nerve consists of approximately 25,000 bipolar neurons whose cell bodies are located in a structure known as Scarpa’s ganglion, which is typically found within the internal auditory canal (IAC).6 Type I neurons of the vestibular nerve derive information from corresponding type 1 hair cells, whereas type II neurons derive information from corresponding type 2 hair cells at its simplest.

Cupula

Type 1 and 2 Hair Cells Supporting Cells Ampullary Nerve Figure 2-3 Stylized representation of the crista: angular acceleration receptor. Reproduced with permission from Solvay Pharma, Canada.

22

Systemic Toxicity Blanket of Calcium Carbonate Crystals Gelatinous Matrix Type 1 and 2 Hair Cells

Macula

Afferent Neuron At Rest

Figure 2-4 Stylized representation of macular end-organ: linear acceleration receptor. Reproduced with permission from Solvay

Pharma, Canada.

Central Vestibular System Primary vestibular afferents enter the brainstem dividing into ascending and descending branches. Within the brainstem there appears to exist a nuclear region with four distinct anatomic types of second-order neurons that have been traditionally considered to constitute the vestibular nuclei. It appears, however, that not all these neurons receive input from the peripheral vestibular system.4,7 The main nuclei are generally recognized as the superior (Bechterew’s nucleus), lateral (Deiters’ nucleus), medial (Schwalbe’s nucleus ), and descending (spinal vestibular nucleus). Functionally, in primate models, the superior vestibular nucleus appears to be a major relay station for conjugate ocular reflexes mediated by the SCCs. The lateral vestibular nucleus appears to be important for control of ipsilateral vestibulospinal (the so-called “righting”) reflexes. The medial vestibular nucleus, because of its other connections with the medial longitudinal fasciculus, appears to be responsible for coordinating eye, head, and neck movements. The descending vestibular nucleus appears to have an integrative function with respect to signals from both vestibular nuclei, the cerebellum, and an amorphous area in the reticular formation postulated to be a region of neural integration. Commonly referred to as the “neural integrator” among neurophysiologists, it is responsible for the ultimate velocity and position command for the final common pathway for conjugate versional eye movements and position.8 The vestibular nerve in part also projects directly to the phylogenetically oldest parts of the cerebellum— namely, the flocculus, nodulus, ventral uvula, and the ventral paraflocculus—on its way directly through the vestibular nucleus. Better known as the vestibulocerebellum, this area also receives input from other

neuronal pathways in the central nervous system (CNS) responsible for conjugate eye movements, especially smooth-pursuit eye movements, which, in addition to the VOR, are responsible for holding the image of a moving target within a certain velocity range on the fovea of the retina. The Purkinje’s cells of the flocculus are the main recipients of this information, of which some appears to be directed back toward the ipsilateral vestibular nucleus for the purposes of modulating eye movements in relation to gaze (eye in space) velocity with the head still or during combined eye–head (vestibular signal-derived) tracking.9,10 Important for cancelling the effects of the VOR on eye movement when it is not in the best interest of the individual (think of twirling ballet dancers or figure skaters and how they can spin without getting dizzy), the vestibulocerebellum is also important in the compensation process for a unilateral vestibular loss.7,9,10

The Hair Cells The fundamental unit for vestibular activity on a microscopic basis inside the inner ear consists of broadly classified type 1 and 2 hair cells (Figure 2-5). Type 1 hair cells are flask-shaped and surrounded by the afferent nerve terminal at its base in a chalicelike fashion. One unique characteristic of the afferent nerve fibers that envelop type 1 hair cells is that they are among the largest in the nervous system (up to 20 µm in diameter). The high amount of both tonic (spontaneous) and dynamic (kinetic) electrical activity at any time arising from type 1 hair cells has probably necessitated this feature for the neurons that transfer this information to the CNS. Type 2 hair cells are more cylindrical and at their base are typically surrounded by multiple nerve terminals in contradistinction.11

Physiology of the Vestibular System Type 1

23

Type 2

KC

KC H

H Ct

Ct

M Nu

M

NC

Nu

NE 2

N

With Otolithic Displacement

NE 1 NE 2

Figure 2-6 Physiology of macular stimulation and inhibition

from otolithic shift and its shearing effect on stereocilia and kinocilium of the hair cells. Figure 2-5 Schematic representation of type 1 and type 2 hair cells. Ct = cuticular plate; H = hairs; KC = kinocillum; M = mitochondria; NC = nerve chalice; NE = nerve ending; Nu = nucleus.

Each hair cell contains on its top a bundle of 50 to 100 stereocilia and one long kinocilium that project into the gelatinous matrix of the cupula or macula. It is thought that the location of the kinocilium relative to the stereocilia gives each hair cell an intrinsic polarity that can be influenced by angular or linear accelerations. It is important to realize that an individual is born with a maximum number of type 1 and 2 hair cells that cannot be replaced or regenerated if lost as a result of the effects of pathology (eg, ototoxicity or surgical trauma) or aging (the postulated presbyvestibular dropout from cellular apoptosis). Presumably the same process holds for the type I and II neurons that comprise the vestibular nerve.

The SCCs largely appear to be responsible for the equal but opposite corresponding eye-to-head movements better known as the VOR. The otolithic organs are primarily responsible for ocular counter-rolling with tilts of the head and for vestibulospinal reflexes that help in the maintenance of body posture and muscle tone. In order to ultimately produce conjugate versional VOR-mediated movements of the eyes, each vestibular nucleus receives electrical information from both sides that is exchanged via the vestibular commissure in the brainstem. The organization is generally believed to be specific across the commissure. Neurons in the right vestibular nucleus, for example, that receive type I input from the right horizontal SCC project across the commissure to the neurons found in the left vestibular nucleus that are driven by the left horizontal SCC receiving contralateral type II input and vice versa.7 Displacement of Sensory Hairs

APPLIED PHYSIOLOGY At the microscopic level, movements of the head or changes in linear accelerations deflect the cupula or shift the gelatinous matrix of the otolithic organs with its load of otolithic crystals that will either stimulate (depolarize) or inhibit (hyperpolarize) electrical activity from type 1 and 2 hair cells. Displacement of the stereocilia either toward or away from the kinocilium influences calcium influx mechanisms at the apex of the cell that causes either the release or reduction of neurotransmitters from the cell to the surrounding afferent neurons (Figures 2-6 and 2-7 ).12 The electrical activity generated is then transferred along the vestibular nerve to the vestibular nuclei in the brainstem. Information above the tonic (spontaneous) firing rate of the type 1 hair cells transmitted along type I neurons is largely thought to have a stimulatory effect in contrast to a more inhibitory effect attributable to type 2 hair cells and type II neurons.

Resting Rate

Toward Kinocilium

Away from Kinocilium

Discharge Rate Vestibular Nerve Tonic Resting Activity

Stimulation (depolarization)

Inhibition (hyperpolarization)

Figure 2-7 The physiology of motion and position sense. Concept of hair cell signal (electrical activity generation) at rest (resting discharge rate) and with respect to effects of movement resulting in depolarization (stimulation) and hyperpolarization (inhibition).

24

Systemic Toxicity

KEY CONCEPTS According to Leigh and Zee’s seminal text, the key concepts of vestibular physiology can be best appreciated in the context that “the push–pull pairings of the canals, the resting vestibular tone and exchange of neural input through the vestibular commissure maximize vestibular sensitivity in health and provide a substrate for compensation and adaptation.”7 VOR Gain In order to maintain a stable retinal image during head movement, the eyes should move in an equal but opposite direction to head movement. Anything less than unity (corresponding eye movement/head movement) may result in the perception of visual blurring with head movement—oscillopsia being the classic symptomatic complaint of an individual with a bilateral peripheral vestibular loss as might result from gentamicin vestibulotoxicity. Nystagmus Defined as a rhythmic to-and-fro, back-and-forth movement of the eyes, nystagmus represents the cardinal sign of unilateral peripheral vestibular or central vestibular dysfunction. In an acute unilateral loss of peripheral vestibular activity that might occur from topical aminoglycoside drops or certain disinfectant surgical preparation solutions used in the presence of a tympanic membrane defect, injury to the end-organ causes a difference in neural activity between the left and right vestibular nuclei. Should the push–pull pairings of the canals be affected as a result of pathology, the eyes are typically driven with a slow movement toward the affected side only to be corrected by a fast corrective saccade generated within the CNS away from the side of the lesion in a repetitive fashion. Although somewhat misguided, the direction of the nystagmus by convention refers to the fast phase, typically away from the side of the lesion under circumstances of an acute unilateral peripheral vestibular loss. Habituation and Adaptation In humans the CNS may habituate (show a reduced response) the VOR depending on the environmental circumstances. This may happen in individuals who are blind or in those exposed to constant velocity rotations or continuous low-frequency oscillations (such as on a ship). The mechanisms for adaptation or the adaptive plasticity of the VOR are usually visually driven and have been experimentally studied by subjects wearing reversing prisms.13 This phenomenon is frequently experienced by those wearing new prescriptive glasses with the explanation that “they take some time to get used to.” Eventually one adapts to the new lenses as the

gain of the VOR changes accordingly. The same holds true to some extent for those with a unilateral peripheral vestibular loss, where the gain can be somewhat influenced, though not perfectly. Compensation Clinical improvement following acute unilateral peripheral vestibular deafferentation requires the presence of intact central vestibular connections primarily at the level of the vestibulocerebellum.4,7 The loss of tonic or spontaneous vestibular activity from the endorgan is ultimately replaced by the development of spontaneous electrical activity arising within the vestibular nuclei of the affected side.14 At rest the asymmetries that would be expected from the push–pull effects from the canals are kept in check, and as a result there is the gradual resolution of the once-present spontaneous nystagmus. Quick head movements producing changes in the dynamic electrical activity, however, can never be completely compensated through this mechanism on the affected side, and a bilateral loss of inner ear function never does despite the insertion of midrotation corrective saccades. For a more detailed explanation of the phenomenon of compensation and why it often fails in the setting of a bilateral vestibular loss see Chapter 19, “Monitoring Vestibular Toxicity.”

CLINICAL MANIFESTATIONS OF VESTIBULAR DYSFUNCTION Loss of vestibular function is associated with several signs and symptoms. Unilateral Peripheral Vestibular Loss With a loss of unilateral vestibular function the patient acutely experiences the sensation of true vertigo from interruptions of VOR pathways and tends to lie perfectly still, as any movement aggravates vegetative symptoms such as nausea and vomiting that arise from the emetic centers. Nystagmus beating away from the side of lesion is the cardinal physical sign that obeys Alexander’s law (the quick phase of the nystagmus induced by the imbalance in activity at the level of the vestibular nuclei is greatest in amplitude and frequency when the eyes are turned away from the side of the lesion).15 Interruption in vestibulospinal tract pathways causes the patient to fall or list toward the affected side. Findings of ipsilateral hemispheric cerebellar dysfunction presenting with behaviors such as past-pointing, an inability to perform rapid alternating movements (dysdiadochokinesis), and gait ataxia reflect acute vestibulocerebellar tract involvement. Features distinguishing peripheral from central mediated nystagmus can be found in Table 2-1. With compensation (implying the existence of a normal functioning CNS and contralateral peripheral vestibular system) there may be minimal symptoma-


25

Table 2-1 Characteristics of Nystagmus/Oculomotor Abnormalities in Peripheral Vestibular vs Central Pathology Feature

Acute Unilateral Peripheral Loss

Bilateral Peripheral Loss

Central

Direction of nystagmus

Mixed horizontal torsional (arching)

None expected

Mixed or pure torsional or vertical

Fixation/suppression

Yes

Yes

No

Slow phase of nystagmus Constant

No nystagmus expected

Constant or increasing/ decreasing exponentially

Smooth pursuit

Normal

Normal

Usually saccadic

Saccades

Normal

Normal

Often dysmetric

Caloric tests

Unilateral loss

Bilateral loss

Intact/direction of nystagmus often perverted (reverse direction)

CNS symptoms

Absent

Absent

Often present

Symptoms

Severe motion aggravated vertigo/vegetative symptoms

Oscillopsia/imbalance/gait Vertigo not as severe as in ataxia, vertigo not a complaint acute unilateral loss

CNS = central nervous system.

tology that is only brought out by very rapid head movements. The spontaneous nystagmus disappears, vegetative symptoms resolve, gait improves, and in the case of a chronic condition the patient may experience only a slight imbalance when turning quickly.

the patient requires assistive devices for ambulation or is relegated to a wheelchair. Compensation is generally unlikely to occur despite the best efforts of vestibular rehabilitation therapy and a greater reliance on information from visual and proprioceptive receptors.

Bilateral Peripheral Vestibular Loss Vertigo is not a feature of a bilateral vestibular loss even when it occurs in an acute fashion. Injury to the endorgans as might occur in systemic aminoglycoside vestibulotoxicity causes a bilateral loss of function that tends to be electrically symmetric at the level of the vestibular nuclei in the brainstem. Instead the patient tends to complain of oscillopsia and imbalance. The gait is typically broad-based and ataxic, especially with eyes closed. Falls are not infrequent and in many instances

SPECIAL CLINICAL TESTS OF VESTIBULAR FUNCTION The following clinical tests are used at the bedside in the assessment of vestibular function and are specific for the VOR. A summary can be found in Table 2-2. High-Frequency Head Thrust or Halmagyi Maneuver Perhaps the most specific test for horizontal VOR function, the high-frequency head thrust, shares the

Table 2-2 Special Clinical Tests of Vestibular Function Clinical Vestibular Test

Purpose

High-frequency head thrust (Halmagyi maneuver) High-frequency test of VOR function usually performed in the horizontal plane. Presence of refixation saccades to stabilize eyes on a target following fast head movement suggests a defect in the horizontal VOR Head shake test for 15–20 seconds

High-frequency vestibular test. Presence of post-headshake nystagmus correlates well with increasing right/left excitability difference on caloric testing. Fast phase of nystagmus usually directed away from side of lesion.

Oscillopsia test

Visual loss of more than 5 lines with rapid horizontal head shaking while looking at a standard Snellen’s chart suggests a bilateral vestibular loss.

VOR suppression test

Inability to visually suppress nystagmus during head rotations suggests a defect at the level of the vestibulocerebellum. Pursuit eye movements are invariably saccadic

VOR = vestibulo-ocular reflex.

26

Systemic Toxicity

same physiologic basis as the “doll’s eye” maneuver in neurology. During a quick head movement to the normal side with the patient focusing on a target, there should be almost perfect compensatory conjugate versional movement of the eyes in an equal but opposite direction to head movement. When a defect occurs in the VOR, quick movements of the head are usually associated with incomplete compensatory conjugate eye movements often requiring refixation saccades to stabilize gaze.16,17 Not surprisingly patients with a unilateral vestibular loss often volunteer subjective blurring of vision when they move their head to the affected side. Although a positive head thrust maneuver is always indicative of pathology somewhere along the course of VOR, false-negative findings can arise. This can occur when an individual throws in corrective midrotation saccades that are imperceptible to the human eye. Identification with advanced eye-movement recording systems such as magnetic scleral coil search studies would typically be required.18 Head Shake Test Rapid horizontal head shaking for 15 to 20 seconds occasionally results in horizontal post-headshake nystagmus usually (but not always) directed away from the side of a unilateral vestibular loss.19 Frenzel’s glasses are generally worn to prevent ocular fixation and suppression of nystagmus by vision. Headshake nystagmus is generally thought to occur when asymmetries in resting vestibular tone are exaggerated at the level of the postulated central velocity storage mechanism in the brainstem.20 Dynamic Visual Acuity (Oscillopsia Testing) Complaints of oscillopsia are best assessed by asking the patient to read the lowest line possible on a standard Snellen’s chart. While repetitively shaking the head in the horizontal plane (at a frequency > 2 Hz) the patient is asked to read the lowest line possible for comparison purposes. A loss of more than five lines during active head-shaking is considered definitely clinically significant for a bilateral vestibular loss.21 The test can also be performed using an optotype such as the letter “E” (the so-called dynamic illegible “E,” or DIE, test) presented in random orientations and different sizes on a chart or monitor. Comparison between the static and dynamic optotypes correctly identified has been used to determine if a bilateral vestibular loss exists.22,23 VOR Suppression Testing (Fixation of VestibularInduced Nystagmus) This test assesses the ability of the CNS to suppress a vestibular signal. It can be assessed by asking patients to follow with the head in the same direction an object that rotates (eg, patients look at their outstretched

hands held together while seated in a chair that rotates). If the vestibulocerebellum is intact then the eyes should remain stable in the orbit from visual fixation and suppression of the VOR. In central vestibular pathology, oculomotor testing typically reveals pursuit eye movements that are saccadic associated with the presence of a breakthrough nystagmus during head rotations as fixation is incomplete.24 VOR suppression testing is somewhat analogous to the parallel phenomenon of failure of fixation suppression during caloric testing when visual fixation does not suppress caloric-induced nystagmus.

SUMMARY • The vestibular system is responsible for the perception of both the sense of position and motion. Angular accelerations are perceived by the SCCs, linear accelerations by the otolithic macula of the utricle and saccule. • Vestibular pathways include the VOR, vestibulospinal tracts, and vestibulocerebellar tracts. Overall the VOR remains the most important and most clinically studied vestibular pathways, being responsible for the maintenance of a stable retinal image with active head movement. Defects in these pathways arising from pathology demonstrate several well-recognized physical signs, the cardinal sign of a unilateral peripheral vestibular loss being nystagmus. • The concepts of VOR gain, nystagmus, habituation or adaptive plasticity, and compensation all have their substrates in the push–pull pairings of the canals, resting vestibular tone, and exchange of neural input through the vestibular commissure from peripheral and central vestibular pathways that maximize vestibular sensitivity in health and demonstrate pathologic features when diseased. (For detailed information regarding oculomotor and vestibular physiology, see Leigh and Zee7 and Baloh and Honrubia.4) • True vertigo is usually present in an acute unilateral peripheral vestibular loss. A bilateral peripheral vestibular loss typically results in oscillopsia (visual blurring with head movement) and imbalance that worsens in the absence of visual clues. True vertigo is rarely ever a feature of bilateral peripheral vestibular loss. • Clinical bedside tests apply the concepts of vestibular physiology and are important in recognizing disease pathology involving the vestibular system. The high-frequency head thrust (Halmagyi maneuver) is probably the most specific clinical test of vestibular function. Other tests include the headshake test, oscillopsia testing, and VOR suppression and fixation.


ACKNOWLEDGMENTS The author thanks Mr David Mazierski (medical illustrator), Ms Elise Walmsley (medical illustrator, TriFocal Communications), and Solvay Pharma Canada (illustrations and permissions).

REFERENCES 1. Rutka JA. Evaluation of vertigo. In: Blitzer A, Pillsbury HC, Jahn AF, Binder WJ, editors. Office based surgery in otolaryngology. New York: Thieme; 1998. p. 71–8. 2. Westcott SL, Lowes LP, Richardson PK. Evaluation of postural stability in children. Current therapies and assessment tools. Phys Ther 1997;77:629–45. 3. Brickner R. Oscillopsia: a new symptom commonly occurring in multiple sclerosis. Arch Neurol Psych 1936;36:586–90. 4. Baloh RW, Honrubia V. Contemporary neurology series: clinical neurophysiology of the vestibular system. Philadelphia (PA): FA Davis Company; 1979. 5. Goldberg JM, Fernandez C. Efferent vestibular system in the squirrel monkey: anatomical location and influence on afferent activity. J Neurophysiol 1980;43:986–1025. 6. Park JJ, Tang Y, Lopez I, Ishiyama A. Age related change in the number of neurons in the human vestibular ganglion. J Comp Neurol 2001;431: 437–43. 7. Leigh RJ, Zee DS. Contemporary neurology series: the neurology of eye movements. Philadelphia (PA): FA Davis Company; 1983. 8. Robinson DA. Ocular motor control signals. In: Lennerstrand G, Bach-y-Rita P, editors. Basic mechanisms of ocular motility and their clinical implications. Oxford: Pergammon Press; 1975. p. 337–74. 9. Zee D, Yamazaki A, Butler PH, et al. Effects of ablation of the flocculus and paraflocculus on eye movements in the primate. J Neurophysiol 1981;46:878–99. 10. Nedzelski JM. Cerebellopontine angle tumors: bilateral flocculus compression as cause of associated oculomotor abnormalities. Laryngoscope 1983;93:1251–60. 11. Ades HW, Engstrom H. Form and innervation of the vestibular epithelia. Symposium on the role of the vestibular organs in the exploration of space. US Naval School Aviat Med, Pensacola, Florida,

12.

13.

14. 15.

16.

17. 18.

19.

20.

21.

22.

23.

24.

27

1965. NASA SP-77. Washington (DC): National Aeronautics and Space Administration; 1965. Ersall J, Lundquist PG. In: Graybiel A, editor. Second symposium on the role of vestibular organs in space exploration. NASA SP-115. Washington (DC): National Aeronautics and Space Administration; 1966. Gonshor A, Melvill Jones G. Extreme vestibularocular adaptation induced by prolonged optical reversal of vision. J Physiol 1976;256:381–7. Precht W. Vestibular mechanisms. Annu Rev Neurosci 1979;2:265–78. Doslak MJ, Dell’Osso L, Daroff RB. A model for Alexander’s law of vestibular nystagmus. Biol Cybern 1979; 34:181–90. Halmagyi M, Curthoys IS, Ae ST, et al. The human vestibular-ocular reflex after unilateral vestibular deafferentation. The results of high frequency acceleration impulsive testing. In: Sharpe J, Barber HO, editors. The vestibulo-ocular reflex and vertigo. New York: Raven Press; 1993. p. 45–55. Halmagyi GM, Curthoys IS. A clinical sign of canal paresis. Arch Neurol 1988;45:737–9. Prepageran N, Kisilevsky V, Tomilinson D, et al. Symptomatic high frequency vestibular loss: consideration for a new clinical syndrome of vestibular dysfunction. Acta Otolaryngol 2004; 124:1–7. Asawavichianginda S, Fujimoto M, Mai M, et al. Significance of head shaking nystagmus in the evaluation of the dizzy patient. Acta Otolaryngol Suppl 1999;540:27–33. Hain TC, Sindler J. Head-shaking nystagmus. In: Sharpe J, Barber HO, editors. The vestibulo-ocular reflex and vertigo. New York: Raven Press; 1993. p. 217–28. Chambers BR, Mai M, Barber HO. Bilateral vestibular loss, oscillopsia and the cervico-ocular reflex. Otolaryngol Head Neck Surg 1985;93: 403–7. Longridge NS, Mallinson AI. The dynamic illegible E (DIE) test: a simple technique for assessing the ability of the vestibule-ocular reflex to overcome vestibular pathology. J Otolaryngol 1987;16: 97–103. Herdman SJ, Tusa RJ, Blatt P, et al. Computerized dynamic visual acuity test in the assessment of vestibular deficits. Am J Otol 1998;19:790–6. Barber HO. About teaching otoneurology. J Otolaryngol 1982;11:141–7.

Systemic Toxicity CHAPTER 3

Salicylates, Nonsteroidal Anti-inflammatory Drugs, Quinine, and Heavy Metals Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), and John A. Rutka, MD, FRCSC

Salicylates are one of the most widely used drugs. Acetylsalicylic acid (ASA), commonly known as Aspirin, is widely used for its anti-inflammatory, antipyretic, and analgesic properties. ASA inhibits platelet aggregation and is used in the treatment of patients with a history of transient ischemic attacks (TIAs), stroke, unstable angina, and myocardial infarcts. Up to 20,000 tons of salicylates are consumed annually in the United States.1 ASA is rapidly absorbed after oral administration and is hydrolyzed in the liver into its active form, salicylic acid. Therapeutic levels range from 25 to 50 µg/mL for analgesic and antipyretic effects to 150 to 300 µg/mL for treatment of acute rheumatic fever. 2 Nonsteroidal anti-inflammatory drugs (NSAIDs) constitute a heterogeneous group of compounds that have therapeutic actions and side effects similar to those of salicylates. Since these drugs can be obtained without prescription, they are potentially available for long-term use and abuse.3 Quinine is another compound with therapeutic actions and clinical side effects similar to those of salicylates. One of the side effects of salicylates and many NSAIDs is ototoxicity manifesting as mild to moderate usually reversible hearing loss and tinnitus. ASA ototoxicity has been reported to occur in 11 per 1,000 patients, with a much higher incidence reported for long-acting salicylates.3,4 Elderly patients have been reported to be at a significantly higher risk for salicylate toxicity, even at lower doses.5 Although the use of quinine as an antimalarial drug has been decreasing in favor of less toxic semisynthetic derivatives, it is commonly used as a treatment for nocturnal leg cramps. For this reason, quinine ototoxicity is still occasionally seen in clinical practice.1

SALICYLATE OTOTOXICITY History The use of naturally occurring salicylates, such as the bark of the willow tree, for their medicinal properties

dates to at least the fourth century BCE. Hippocrates recommended their use to relieve the pain of childbirth.2 The active ingredient of willow bark, called salicin, was first isolated by Leroux in 1829. The antipyretic properties of salicylate, in the form of sodium salicylate, were used in 1875 for the treatment of rheumatic fever.1 In 1899, Hoffman, a chemist with Bayer, prepared the ASA compound that was called Aspirin.6 Muller first identified the ototoxic effects of high doses of salicylate in 1877; reports of deafness from Aspirin surfaced in 1884.1,7,8 Pharmacokinetics of Salicylate ASA is rapidly absorbed in the upper small intestines and stomach. Appreciable serum levels of ASA have been noted within 30 minutes of oral ingestion, with peak levels occurring at 2 hours.1,6 The passive diffusion of salicylic acid through the mucosa is determined by various factors: tablet dissolution rate, mucosal pH value, and the gastric emptying rate. Subsequently, salicylate is widely distributed, undergoes biotransformation in the liver, and is excreted by the kidneys. The main metabolic products are salicyluric acid, phenolic glucuronide (the ether), and acyl glucuronide (the ester). Salicylate metabolites excreted by the kidney in the urine include free salicylic acid (10%), salicyluric acid (75%), salicylate phenolic (10%), acyl glucuronides (5%), and gentisic acid (< 1%). The plasma half-life of ASA is about 15 minutes, whereas the plasma half-life of salicylate is 2 to 3 hours for low doses and 12 hours or longer for higher doses.1,6 Ishii and colleagues in 1967 documented the rapid entry of salicylate into the cochlea after systemic administration of tritium-labeled salicylate. Autoradiography detected the tritium-labeled salicylate in the stria vascularis and spiral ligament. Within 1 hour, salicylate was detected around the outer hair cells (OHCs) and near the spiral ganglion cells. There was no accumulation in any specific structures.9 Evidently, salicylate undergoes rapid absorption and promptly reaches the

Salicylates, Nonsteroidal Anti-inflammatory Drugs, Quinine, and Heavy Metals

cochlea via the arteries, where it readily diffuses to all parts of the cochlea.1 Salicylic acid, transported in the undissociated or un-ionized form, is highly permeable across lipid layers10 and human red blood cell membranes.11 When salicylic acid enters a cell, dissociation to salicylate occurs. The final concentration of the dissociated form appears dependent on intracellular pH.12 Serum salicylate levels and perilymph levels have been extensively studied in animals and humans. Woodruff and colleagues reported serum salicylate levels of 51 to 78 mg/dL, with a median of 65.3 mg/dL, 3 hours after intramuscular injection of sodium salicylate (400 mg/kg) in chinchillas.1,13 Salicylate concentration has also been documented in serum, cerebrospinal fluid (CSF), and perilymph after intraperitoneal injection of ASA (300 mg/kg).14 Peak levels appeared within 0.5 to 1 hour in serum and 1 hour in CSF, whereas perilymph levels peaked at 2 hours. The concentration of salicylate in perilymph was the lowest (10 mg/dL), approximately 20% of the serum concentration (65 mg/dL), but only slightly lower than CSF levels (14 mg/dL).1 Silverstein and colleagues reported perilymph levels of 25 to 40 mg/dL and serum concentrations of 51 to 86 mg/dL, 5 to 7 hours after injecting salicylate (350 mg/kg) into cat peritoneum.1,15 Jastreboff and colleagues in 1986 documented a peak serum level of 70 mg/dL 3 to 4 hours after an intraperitoneal dose of sodium salicylate in guinea pigs and 60 to 70 mg/dL in rats. They also documented that the salicylate levels in perilymph were closely related to both drug levels in serum and CSF. Higher levels were systematically observed in perilymph than in CSF. The conclusion drawn was that salicylate is not passively transported into perilymph across CSF but, rather, appears to be transported from blood directly into perilymph.16 Day and colleagues studied the concentration response relationships for salicylate-induced ototoxicity in normal volunteers and reported that total and unbound plasma salicylate concentration increased disproportionately with increasing daily doses of ASA. The increase in unbound salicylate in plasma was relatively greater since the percentage of unbound salicylate increased over the dose range investigated from a mean of 3.9 to 10.4%.17 Studies have correlated serum salicylate levels to hearing loss. Myers and Bernstein18 and Bernstein and Weiss19 reported a correlation (r = .84) between hearing loss and increasing serum levels up to 40 mg/dL. Levels above this did not produce a significantly greater degree of hearing loss.1 Bonding and McFadden and colleagues21 found strong correlations (r = .73 and r = .6 respectively) between hearing loss and serum salicylate levels.1,20,21 Various salicylate serum concentrations (ranging from 19.9 to concentrations > 67 mg/dL) have been reported to correlate with documented ototoxicity.22 Levy reported that the time

29

courses of various pharmacologic effects of single doses of ASA are not coincident with the plasma concentration of either ASA or salicylic acid. There was reasonably good evidence, however, that the pharmacologic effects were related to the concentration of ASA, salicylic acid, or both.23 He concluded that the pharmacologic activity of salicylate is produced by free (unbound) drugs. The plasma protein that binds salicylic acid is concentration dependent and subject to pronounced individual variations, making it preferable to monitor unbound rather than total concentrations of salicylate in plasma.23 Halla and colleagues identified that most reports of salicylate toxicity have been based on total serum levels, whereas unbound serum salicylate concentrations appear to reflect more closely the risk of ototoxicity.22 Day and colleagues reported that in a study of volunteers hearing loss and tinnitus intensity increased progressively with ASA dosage and increasing concentration of both total and unbound plasma salicylate concentrations. Hearing loss and unbound salicylate concentration appear to have a linear relationship.17 Conversely, Halla and colleagues concluded that tinnitus and subjective hearing loss were too nonspecific and not sensitive enough to be used as a useful indicator of serum salicylate concentration.22,24 Mechanisms of Salicylate Ototoxicity The dominant ototoxic effects of salicylates are tinnitus and a reversible sensorineural hearing loss (SNHL). The hearing loss is typically mild to moderate and bilaterally symmetric. Recovery usually occurs 24 to 72 hours after cessation of the drug. As such, the mechanism of salicylate ototoxicity is probably more related in humans to reversible biochemical or metabolic changes in the cochlea rather than to morphologic abnormalities.1 Pathology The pathology of salicylate ototoxicity has been documented from human and animal temporal bone histopathology and ultrastructural examinations of the organ of Corti, especially the cochlear hair cells. In many instances, there have been conflicting and variable interspecies differences. For example, Kirchner in 1881 identified areas of hemorrhage in the organ of Corti and the cochlear labyrinth in a human temporal bone following salicylate ototoxicity.1 Similar findings were noted in guinea pigs in 1938.1 Bernstein and Weiss in 1967 reported histopathologic findings in temporal bones from two patients who had received large doses of ASA (more than 5 g/d) before death.19 They documented that the organ of Corti was essentially uninvolved. Most hair cells were normal, with the exception of the hair cells in the basal turn, which were consistent with change as a result of aging.1 DeMoura and

30

Systemic Toxicity

Hayden in 1968 also reported the organ of Corti to be normal with no significant hair cell loss in a patient who apparently had a 40 to 50 dB hearing loss from taking ASA of up to 5.2 g/d for 7 months.25 The patient recovered 20 dB of hearing 3 days after she stopped taking ASA.1 Histopathologic examinations of temporal bones have revealed minor and presumably reversible microscopic changes in the cochlea. In 1938, Covell noted evidence of mitochondrial change and dilatation of stria vascularis in the strial and OHCs of the guinea pigs.26 The spiral ganglia, however, appeared to be normal. Other investigators have noted different microscopic changes. Gotlib, in 1957, reported some changes in the spiral ganglion with normal strial tissues. 27 Although Falbe-Hansen (1941) noted some loss of hair cells, Myers and Bernstein (1965) did not find any significant abnormalities in the cochlea microstructure in subjects when compared with a control group.1,18 In an animal study using chinchilla exposed to 400 mg/kg of salicylate, Woodford and colleagues noted scattered OHC loss.1 Conversely, again using the chinchilla as the animal model, Deer and Hunter-Duvar found no significant abnormalities.1,28 There was no evidence of cell destruction in the stria vascularis or organ of Corti. The cochlea vessels and stereocilia of inner and outer hair cells were normal. Transmission electron microscopy showed no difference in the inner hair cells, supporting cells, or stria vascularis between the study and the control group.1,28 The OHCs of the experimental animals, however, revealed an accumulation of organelles, namely membrane bound vacuoles, dark staining spheres near the Golgi apparatus, swollen mitochondria, lysosomes, and whorls of fenestrated smooth cisternae surrounding degraded mitochondria.1 A similar ultrastructural study of longitudinal and transverse sections of guinea pig temporal bones after exposure to a high dose of sodium salicylate revealed no abnormalities. Falk found no abnormalities in the cell membrane, nuclear membrane, mitochondria, Golgi bodies, endoplasmic reticulum, myelin sheath of spiral ganglia, axoplasm, Schwann’s nucleus, or auditory nerve cytoplasm.1,29 Morphologic changes secondary to salicylate ototoxicity have been documented with the aid of electron microscopy in the last few decades. In 1983, Douek and colleagues reported cochlea morphologic findings in guinea pigs after administering 375 mg/kg of sodium salicylate daily for 1 week. Although no significant changes were noted on light microscopy in the stria vascularis and the number of hair cells between the study and control groups, electron microscopy revealed extensive vacuolization of the endoplasmic reticulum underlying the cell membrane of the OHCs in the study group. In addition, long-term salicylate administration caused bending of the OHC stereocilia.1,30 Dieler and

colleagues reported certain ultrastructural morphologic changes on the OHCs of guinea pigs after salicylate infusion. They noted subsurface cisternae dilatation and vesiculation. These changes appeared to be time and dose dependent and were reversible over 30 minutes.31 They speculated that intact, unfenestrated subsurface cisternae were required for optimal generation of electrically induced motility in mammalian OHCs. The OHC changes may therefore form the morphologic basis by which salicylate ototoxicity reduces spontaneous and evoked emissions.1 Physiology Physiologic studies of salicylate ototoxicity have predominantly focused on the evaluations of cochlear potentials, acoustic emissions, and cochlear blood flow.1 Effects on Cochlear and Auditory Potentials Early animal studies have identified some electrophysiologic changes after exposure to salicylate. In 1973, Mitchell and colleagues noted that a single subcutaneous dose of sodium salicylate reversibly interfered with the generation of cochlear nerve action potentials (APs) in guinea pigs.32 This process was more pronounced in the higher frequencies of hearing. The cochlear microphonics (CM) remained normal. Silverstein and colleagues, however, noted decreases in the amplitudes of both CM and APs in cats after intraperitoneal injections of salicylate.15 McPherson and Miller noted similar results in guinea pigs in 1974.33 Stypulkowski in 1990 reported his findings of increased CM and reduced summating potentials.34 No significant change was identified in endocochlear potentials. Other researchers have reported varying increases in the spontaneous activity of auditory neurons after salicylate administration.35 Evans and Borerwe noted an increase in the spontaneous activity of auditory neurons.36 This finding was further supported by the works of Kumagai in 199237 and Stypulkowski in 1990.34 Chen and Jastreboff in 1995 identified increased central activity in the inferior colliculus.38 A similar pattern was noted by Jastreboff and Sasaki 39 in 1986 and Manabe and colleagues40 in 1997. Ochi and Eggermont, however, identified increased spontaneous activity in the auditory cortex after salicylate administration.41 Most physiologic observations from animal studies were obtained after a single high dose of salicylate in an acute setting. This differs from most human studies where high doses of salicylate were repeatedly administered over days.35 Effects on Otoacoustic Emissions The discovery and application of otoacoustic emission (OAE) testing have helped to further our understanding of salicylate ototoxicity. OAEs are presumed to reflect an active mechanical process in the cochlea, especially at


the level of the OHCs.1 Johnson and Eberling reported decreased OAEs after the ingestion of 10 g of ASA.42 McFadden and Plattsmier measured spontaneous OAEs prior to, during, and following administration of ASA.43 They noted that all spontaneous OAEs gradually diminished and disappeared during the treatment protocol. Small emissions were noted to disappear early, within 14 to 20 hours of drug ingestion, whereas larger emissions took 40 to 70 hours to disappear completely. There was complete recovery once the drug administration ceased; the length of recovery lasted from 24 hours to a few days. Long and Tubis noted decreased spontaneous emissions, delayed OAEs, and synchronous emissions after ingestion of high-dose ASA.44 Reduction of OAE induced by salicylates suggested abnormalities in the OHCs.1 Brownell and Winston reported a decrease in the turgor of the OHCs, which reduced their ability to contract.45 Alterations in cell membrane permeability are believed to be the result of salicylate increasing the potassium ion conductance of the OHCs, which in turn has been postulated to cause the decrease in the OHC turgor.34,45 Decrease in OHC turgidity and interference with electromotility has been demonstrated by Shehata and colleagues to be dose dependent and reversible.46 More recently, Tunstall and colleagues. showed that salicylates reduce OHC-membrane capacitance.47 This was later confirmed by Kakehata and Santos-Sacchi in 1996.12 Effects on Cochlear Blood Flow A vascular basis for salicylate ototoxicity has also been recently highlighted. Reduction of blood flow to the cochlea has been suggested as a possible mechanism for salicylate toxicity.1,48 Hawkins noted that acute ASA ingestion produced vasoconstriction of the capillaries of the spiral ligament and stria vascularis.48 The localized drug accumulation and vasoconstriction in the auditory microvasculature may have been mediated by antiprostaglandin activity in these agents.49 Using a laser Doppler flowmeter, Didier and colleagues demonstrated decreased cochlear blood flow by 25% that was associated with significant elevated compound AP thresholds with high doses of ASA (300 mg/kg).50,51 Jung and colleagues found a similar significant reduction of cochlear blood flow both by systemic (300 mg/kg) and round window membrane (RWM) application of sodium salicylate.1 Biochemistry Transient biochemical abnormalities have long been believed to be responsible for the reversible nature of salicylate ototoxicity. Biochemical investigations of salicylate ototoxicity include abnormalities in inner ear enzymes, catecholamines, and cations such as zinc and arachidonic metabolism.1 Silverstein and colleagues in 1967 demonstrated reduced malic dehydrogenase activity in the perilymph and endolymph and decreased electrical activity in the

31

cochlea, possibly reflecting an inhibition in the stria vascularis and organ of Corti.15 Ishii and colleagues noted that salicylate interfered with cholinesterase activity in the organ of Corti. 9 Krzanowski and Matschinsky in 1971 reported that salicylate interferes with phosphate metabolism in phosphocreatine reserve at the basal turn of the organ of Corti and in the adenosine triphosphate (ATP) of Reissner’s membrane.52 Jung and colleagues in 1990 found elevated levels of catecholamines, in particular, norepinephrine in the perilymph of chinchillas after intraperitoneal injection of salicylate.53 Other researchers have discovered that antagonists of norepinephrine prevent salicylate ototoxicity.54,55 These findings further strengthen the possibility of abnormal catecholamine metabolism in salicylate ototoxicity. Gunther and colleagues in 1989 noted that simultaneous administration of zinc prevented salicylateinduced hearing loss. They proposed that a relative zinc and magnesium deficiency could conceivably play a role in salicylate ototoxicity.56,57 In 1971 Vane discovered that salicylates and NSAIDs inhibited cyclooxygenase pathways in the arachidonic acid cascade from producing prostaglandin (PG).58 This led to speculation that inhibition of PG might play a role in salicylate ototoxicity.59 In the mid1980s Jung and colleagues reported that perilymph contained relatively higher levels of 6-ketoprostaglandin F1 than did CSF and that indomethacin and salicylate markedly reduced these levels. 1,60,61 Decreased PG levels and synthesis in the guinea pig cochlea were noted after exposure to salicylate.60 Kawata and colleagues in 1988 reported in vitro PG synthesis in the cochlea.61 These findings further verified the possible role of PG metabolism in salicylate ototoxicity. Altered arachidonic acid metabolisms after salicylate administration has also been postulated to play an important role in ototoxicity. This was first noted by Jung and colleagues when decreased PG levels (cyclooxygenase products) and increased leukotriene (LT) or lipooxygenase products were noted in the perilymph of chinchillas after systemic or RWM application of salicylates.62,63 Applying exogenous LT on the RWM in animal studies caused hearing loss and changes in the eicosanoid levels similar to those found in salicylate ototoxicity.1,64 The role of PG and LT in salicylate ototoxicity was further validated when salicylate ototoxicity was prevented by the use of LT blockers and steroids before salicylate treatment.65–67 Pretreatment with an LT blocker prevented the reduction of cochlear blood flow.68 This series of studies underlines the probable importance of increased LT levels in the inner ear in the pathogenesis of salicylate ototoxicity.1 To summarize, it appears that the mechanisms for salicylate ototoxicity are both multifactorial and multilocational. Minimal morphologic abnormalities have

32

Systemic Toxicity

been reported. Acoustic emission studies suggest OHC abnormalities in salicylate ototoxicity. Decreased cochlear blood flow also plays a role. This may be mediated by catecholamines or arachidonic acid metabolites. Biochemical changes with increased levels of norepinephrine, decreased PGs, and increased LTs have also been demonstrated. Abnormal levels of arachidonic metabolites appear to reduce cochlear blood flow and abnormal permeability of the OHC.1

MANIFESTATIONS OF SALICYLATE OTOTOXICITY Manifestations of salicylate toxicity in humans are well documented. Nausea, vomiting, tinnitus, hearing loss, headache, mental confusion, quickened pulse, and increased respiration have all been reported.6 Hearing Loss The classic description of salicylate-induced SNHL is of a mild to moderate bilaterally symmetric loss, which may be flat or only in the higher frequencies. The hearing loss is typically reversible, and recovery usually occurs 24 to 72 hours after cessation of the drug.69 In the 1960s, Myers and Bernstein and Bernstein and Weiss reported salicylate ototoxicity in patients with rheumatoid arthritis after large daily doses of ASA, 6 to 8 g/d. The SNHL was noted to be flat bilaterally in the order of 20 to 40 dB. Hearing apparently recovered within hours after drug cessation.18,19 In 1965 McCabe and Dey noted temporary high-frequency hearing loss with tinnitus in five healthy volunteers after high doses of ASA. The hearing loss progressed from 4 dB after 1 day to 28 dB after 5 days (at 6 kHz). Increased duration and dosage caused greater hearing loss. All five volunteers had their hearing return to normal within 72 hours upon termination of treatment.69 McFadden and Plattsmeir in 1983 demonstrated that both dosage and duration of ASA ingestion affected the degree of hearing loss. They reported that 3.9 g/d of ASA for nearly 3 days resulted in a hearing loss of 2 to 19 dB at between 2.5 and 8 kHz. A dose of 4.9 g/d for 4 days resulted in threshold shifts of 18 dB at 4 kHz and 25 dB at 8 kHz. They also reported hearing loss of 8.4 to 21.6 dB at 3.5 kHz in three of five subjects receiving 3.9 g of ASA per day for 5 days.70 Some studies have demonstrated an additive effect of salicylate administration and noise exposure on the threshold shift. Most studies, however, demonstrate that salicylate does not aggravate noise-induced hearing loss or cause cochlear damage.1,71 The recovery periods for hearing loss after salicylate ototoxicity appear to range from a few days to 2 weeks and show large individual variability with no apparent relation to the duration of treatment, serum salicylate levels, or amount of hearing loss.18,35,44 Most cases of hearing loss after salicylate toxicity appear to be reversible, although a few cases of permanent hearing loss have been reported.72–74 One patient has also

been reported to have developed symptoms of cochlear hydrops with fluctuating hearing loss, unsteadiness, tinnitus, and aural fullness after taking 2 g/d of ASA. The patient suffered a 40 to 70 dB hearing loss, which recovered to 20 dB after 3 weeks.1,75 Ototoxicity has also been reported after topical application of salicylates for psoriasis.76 Salicylate ototoxicity appears to be associated with changes in suprathreshold characteristics of hearing, such as a decrease in temporal integration, poorer temporal resolution, and impaired frequency selectivity.1 Animal studies have documented changes in hearing threshold levels following administration of salicylates. The hearing loss in animals varies with the species of the animal tested. This variability may relate to the differences in physiology and serum and perilymph metabolism of salicylate. Myers and Bernstein measured the hearing threshold in monkeys following subcutaneous injection of 500 to 600 mg/kg of salicylate. Hearing loss was noted to be flat, ranging from 17 to 36 dB with an average of 30 dB.18 A similar study with monkeys after an intramuscular injection of 250 mg/kg of sodium salicylate revealed a 22 dB loss at 4 kHz within 1 hour of injection and complete recovery within hours. Higher doses of salicylates caused higher threshold shifts with complete recovery within 24 hours.1,77 Other studies involving guinea pigs, chinchillas, and cats with hearing losses measured by various methods (such as auditory brainstem response [ABR], evoked responses from electrodes permanently placed in inferior colliculus, and AP-threshold shifts) have revealed reversible hearing losses of varying degrees.13,62,78–81 Some studies have demonstrated an additive effect of salicylate administration and noise exposure on the threshold shift. Most studies, however, demonstrate that salicylate does not aggravate noise-induced hearing loss or cause cochlear damage.1,70 Overall, it appears that high-dose salicylate administration causes a reversible, bilaterally symmetric, mild to moderate hearing loss, which may be flat or present in the high frequencies. The threshold shifts typically recover after cessation of salicylate. The duration of recovery varies from individual to individual. Tinnitus Tinnitus also occurs with salicylate toxicity. Salicylateinduced tinnitus has been characterized as tonal, and pitch matching has identified the tinnitus frequencies to be usually around 7 to 9 kHz.70 Clinically the onset of tinnitus has been used as the earliest sign of salicylate toxicity. Several studies had been carried out to correlate serum salicylate levels with the onset of tinnitus. In 1973, Mongan and colleagues reported that tinnitus


may be a useful indicator of salicylate ototoxicity in patients with normal hearing.82 This was based on their study where patients with normal hearing and impaired hearing were given ASA in an increasing dosage as an analgesic. They noted that all patients with normal hearing eventually developed tinnitus, compared with 31% of patients with a preexisting hearing impairment. Most patients developed tinnitus with a minimum serum salicylate level of 19.6 mg/dL. Day and colleagues in 1989 reported a study involving volunteers whereby hearing loss and tinnitus intensity increased progressively with both the ASA dosage and increasing concentration of total and unbound plasma salicylate concentrations. There appeared to be a linear relationship between hearing loss and the unbound salicylate concentration.17 Ototoxic symptoms were noted at lower concentrations of total salicylate than previously observed. There appears to be no minimal serum salicylate concentration for threshold shift.1 Conversely, Halla and colleagues concluded that tinnitus or subjective hearing loss was too nonspecific and not sensitive to be used as a useful indicator of serum salicylate concentration.22,24 They noted that in patients with rheumatoid arthritis, mean salicylate levels were not higher in those with tinnitus than in those without. The presence of tinnitus correlated with serum salicylate levels in only 30% of patients. Clinical reports indicate that during long-term salicylate treatment in humans, tinnitus occurs after several days, becomes louder as treatment is continued, and is characterized as a high-pitched noise.17,70,83 The loudness of salicylate-induced tinnitus from long-term salicylate treatment has been matched in humans to pure-tone levels ranging from 20 to 60 dB sound pressure level.17 Studies on the spontaneous neural activity of the auditory system in animal models have contributed to our understanding of salicylate-induced tinnitus. Various degrees of increase in the spontaneous activity of auditory neurons after salicylate administration have been reported at the level of the auditory nerve,36,37 inferior colliculus,38–40 and auditory cortex.40 An ensemble measure of auditory nerve spontaneous firing rate can be obtained in animals from the average spectrum of recordings in silent conditions from a gross electrode on the nerve or on the RWM.83–85 This spectrum measure changed after a single dose of salicylate.85,86 Similar measures carried out in humans during eighth cranial nerve surgery indicated that in several patients with tinnitus, the average spectrum of activity of the auditory nerve was altered.35,86,87 Cazals and colleagues measured the average spectrum of electrophysiologic cochleoneural activity (ASECA) and auditory nerve compound action potentials (CAPs) in guinea pigs after salicylate administration. They noted ASECA alterations without changes in CAP. This find-

33

ing provides evidence that tinnitus is the first subjective sign of salicylate ototoxicity, before hearing loss.35 Although tinnitus appears to be the initial sign of impending ototoxicity, serum salicylate levels apparently do not correlate well with tinnitus in patients with salicylate ototoxicity.

OTOTOXICITY OF NONSTEROIDAL ANTI-INFLAMMATORY DRUGS The ototoxicity of NSAIDs is in general similar to that of salicylates. NSAIDs are a heterogeneous group of compounds that share similar therapeutic actions and side effects with salicylates.1,6 Most NSAIDs act as analgesics and anti-inflammatories by inhibiting cyclooxygenase pathways. Chapman in 1982 reported a series of five patients who sustained hearing loss with naproxen; two recovered their hearing after discontinuing the drug.88 Rarely does a significant and permanent loss of hearing occur shortly after the start of NSAIDs. Few animal studies have been carried out to assess the ototoxicity of NSAIDs. Koopman and colleagues found no alterations in the ABR of guinea pigs after long-term ibuprofen treatment.89 Morrison and Blakely noted no ultrastructural abnormalities in guinea pigs treated with indomethacin except for a questionable distention of Reissner’s membrane.1,90 Jung and colleagues in 1992 noted that RWM application of indomethacin in chinchillas resulted in hearing loss, decreased PGs, increased LTs, and reduced cochlear flow, similar to the effects of salicylate.1,63,91

QUININE OTOTOXICITY Research into the mechanisms of quinine and salicylate ototoxicity has historically been closely linked because of their similar clinical manifestations. Quinine is an effective drug in suppressing an acute attack of malaria. Intravenous quinine continues to be the initial treatment for severe Plasmodium falciparum malaria following the occurrence of widespread chloroquineresistant strains92 Although the use of quinine as an antimalarial agent has been decreasing in favor of less toxic semisynthetic derivatives, its use for nocturnal leg cramps continues to rise. Research into quinine ototoxicity has been comparatively minimal. As a result, the exact mechanism of quinine ototoxicity is still largely unknown, except that it has a distinctly different mechanism from salicylate.1 History Quinine has been used in the treatment of malaria since the early 1600s. Derived from the cinchona bark, indigenous to parts of South America, it was initially also used as an antipyretic. The tree apparently was named after the wife of a viceroy to Peru, Countess Anna del Chinchon, whose cure from its bark led to the

34

Systemic Toxicity

tree’s importation to Spain in 1639.1,93 It was popularly known as “Jesuit’s bark” as the Jesuit fathers were the main distributors in Europe at that time.1 The first side effect from the treatment of malaria with quinine was reported by Richard Morton in 1692. He described the treatment as rather safe: “I have never known anyone suffer a misfortune as a result of using Cinchona bark (quinine) other than a distressing type of hearing loss.” 92 Today, quinine is largely used in tonic beverages or converted to its stereoisomer, quinidine, which has many of quinine’s side effects, including, on rare occasion, ototoxicity.1 Pharmacokinetics of Quinine The metabolism of quinine begins with its absorption in the upper small intestine and is over 80% complete, with peak levels occurring 1 to 3 hours after a single oral dose. Plasma concentration falls with a half-life of 11 hours. Around 90% of plasma quinine is protein bound, with CSF levels ranging between 2 and 5% of that of plasma. It is metabolized extensively in the liver, with only 10% eliminated by the kidneys unaltered.1,93 Among the plasma proteins, α1-acid glycoprotein appears to be the main plasma-binding protein for quinine.92 Intravenous quinine is in many instances the mainstay of treatment for severe P. falciparum infection. For therapeutic effect, a plasma concentration of 10 mg/L is advised.94,95 However, plasma quinine concentrations above 5 mg/L in malaria patients can become ototoxic, selectively affecting high-frequency hearing.94,96,97 Reports of serious side effects of quinine are relatively rare, which could be because of the slow infusion rates and extensive binding of quinine to plasma proteins in the acute phase of falciparum malaria.94,98 The pharmacokinetic properties of quinine in malaria patients have been noted to be different from those in healthy patients.94,99,100

MECHANISMS OF QUININE OTOTOXICITY Morphology Large doses of quinine are known to produce reversible hearing loss and tinnitus, similar to salicylates. Cochlear OHCs seems to be the common site for the ototoxic effect of both drugs. Studies have shown that prolonged administration of high-dose quinine in guinea pigs resulted in loss of OHCs, which would explain the hearing loss.6,101 This finding probably represents the end point of prolonged exposure. Reversible hearing loss is unlikely to be associated with such permanent morphologic changes.1 Karlsson and Flock have reported ultrastructural changes in the OHCs in the cochlea of guinea pigs after exposure to quinine.92,102 Recently, Dieler and colleagues studied the morphologic changes of the isolated OHCs of guinea pigs exposed to large doses of quinine. Ototoxic doses of

quinine did not induce any morphologic changes. No changes in turgor, shape, or fine structure of OHCs were noted. As a result they concluded that the underlying mechanisms of quinine toxicity are considerably different from those of salicylates, although both substances lead to identical symptoms.103 Physiology High doses of quinine have demonstrated a selective effect on the CAP in response to both high- and lowfrequency responses.1,104 Intermediate frequencies were normal. CMs were suppressed in most animals. Researchers additionally have noted other physiologic changes. Dieler and colleagues reported that quinine exposure led to a hyperpolarization followed by a depolarization of the hair cells’ membrane potential. It also caused a diminution of evoked rapid motile responses that was more apparent in response to hyperpolarizing pulses than depolarizing pulses. Responses were largely dose dependent and reversible. 103 Exposure of quinine to OHCs and Hensen’s cells revealed alterations of the micromechanical tuning of the organ of Corti.105 Kenmochi and Eggermont recently measured local field potentials in cats’ primary cortex before and after application of salicylates and quinine. They found a significant decrease with both drugs, suggesting a central effect of both drugs in addition to a peripheral one.106 Reduction of cochlear blood flow has been postulated as another explanation for quinine ototoxicity. Gradenigo in 1893 and Wittmaack in 1903 suggested the possibility of vasoconstriction for ototoxicity.48 Vasoconstriction had been noted in guinea pigs following quinine in the capillaries of the suprastrial spiral ligament, stria vascularis, and basilar membrane.48 Smith and colleagues additionally noted a significant reduction in erythrocytes in all turns of the cochlea and localized vessel narrowing as a result of endothelial cell swelling.104 There is speculation that quinine induces osmotic changes in endothelial cells as it does in red blood cells and hepatocytes, by blocking calcium-dependent potassium channels. Motion photographic studies of cochlear microcirculations after quinine demonstrate a temporary cessation of blood flow through the terminal capillaries in the basilar membrane.1,107 Lee and colleagues noted a significant decrease in cochlear blood flow, measured by laser Doppler flowmeter, that corresponds to hearing loss.108 Quinine, in susceptible individuals, binds to plasma protein and acts as a hapten, triggering the complement cascade, leading to thrombocytopenic purpura, disseminated intravascular coagulation, and hemolytic anemia.109,110 This mechanism has been suggested by some researchers to be responsible for the microvascular changes in the cochlea after quinine ingestion.111


Biochemistry Reduction in endogenous PGs has been noted in rats after exposure to high doses of quinine. This has been postulated to interfere with the generation of APs in nerves and vascular smooth muscle. Others propose that quinine’s antiprostaglandin effects arise from the inhibition of the phospholipase A 2 enzyme. 1,112,113 Quinine has also been noted to block calcium-dependent potassium channels, which are found in a wide variety of cells.103,114

MANIFESTATIONS OF QUININE OTOTOXICITY Quinine ototoxicity, or cinchonism, presents as hearing loss, tinnitus, vertigo, headache, nausea, and visual loss. Hearing Loss Transient hearing loss appears to be the first manifestation of quinine ototoxicity. It occurs a few hours after initiating high-dose therapy (up to 2 g in the treatment of malaria). After prolonged daily dose courses of 200 to 300 mg, up to 20% of patients might have some degree of hearing loss. 1,115 The hearing loss is typically reversible and a bilateral symmetric SNHL that affects the higher frequencies initially (at 4, 6, and 8 kHz) with a characteristic 4 kHz notch. Discrimination scores have been noted to drop below 30%.1,75 Although the hearing loss after quinine administration is typically reversible, permanent hearing loss has been reported, affecting the conversational frequencies.115 In addition, to prevent irreversible hearing loss, ultrahigh-frequency audiometry (10 to 20 kHz) has been advocated for accurate monitoring of impending ototoxicity.97 Tange and colleagues administered high doses of quinine-dihydrochloride intravenously to 12 healthy volunteers and 10 patients with falciparum malaria. In healthy subjects, hearing loss was documented at 2 to 4 hours after quinine infusion at a mean maximal plasma quinine concentration of only 2 mg/L. Both high-frequency audiometry (HFA) at 10, 12, 14, and 16 kHz and conventional audiometry (CA) were performed. A unilateral hearing loss was initially noted in five healthy volunteers during infusion itself (four in HFA and one in CA). Maximal hearing loss was measured 2 to 4 hours after infusion unilaterally in nine subjects (seven in HFA and two in CA). Hearing loss did not exceed 25 dB, except in one ear, which showed 35 dB loss at 10 and 13 kHz with a persistent loss of 20 dB at 14 kHz after 14 months. All others recovered completely within 1 week.92 This study underscores the importance of HFA in the early detection of quinine ototoxicity. All patients with malaria experienced ototoxicity initially; 9 had hearing loss, 10 had tinnitus, 8 had aural pressure, and 4 felt giddy. The hearing loss

35

was maximal on the third day of infusion. Final audiograms were normal, indicating reversibility in hearing loss.92 Similar findings were noted by Claessen and colleagues in 1998.94 In cases of quinine self-poisoning, clinical hearing loss is common only at plasma concentrations over 10 mg/L94 Hearing loss secondary to quinine ototoxicity in patients with plasma levels above 10 mg/L has been found to be largely, if not completely, reversible.94,96,97 Clinical auditory toxicity from quinine has been reported sparingly in malaria patients, despite quinine concentrations almost invariably exceeding 10 mg/L. This is probably explained by the differences in protein binding, the free fraction of quinine being reduced by 25% in patients with uncomplicated malaria and up to 40% for severe malaria.94,100 Healthy volunteers were noted to have only one-third of the concentration of α1-acid glycoprotein (the main plasma binding protein for quinine) found in malaria patients.92 Tinnitus and Vertigo Tinnitus is a known symptom of quinine ototoxicity. Quinine-induced tinnitus has been reported to be similar to salicylate-induced tinnitus, which typically produces a high-pitched narrow-band tone.116,117 Some studies indicate that quinine-induced tinnitus can be blocked with nimodipine, a calcium channel blocker, in a dose-dependent manner.116 The calcium channel blocker verapamil, however, did not prevent hearing loss after quinine administration to guinea pigs.118 The vestibular effects of quinine are also well recognized. Minimal amounts of quinine are found in tonic beverages, which may lead to low-serum quinine concentrations adequate for producing clinically significant vestibular changes.1 Blood quinine levels of 0.2 mg/L have been found in pilots who died in aviation accidents, suggesting that quinine toxicity may have played a causative role.1,119 Transient positional abnormalities have also been noted during electronystagmography in volunteers who drank 1.6 L of tonic water (105 mg) daily for 2 weeks.111

HEAVY METALS (MERCURY AND LEAD) Mercury Mercury (Hg) is a toxic heavy metal; its different chemical forms account for various degrees of toxicity. In Asia, cinnabar (a naturally occurring mercuric sulfide [HgS] compound) has been used in combination with Chinese herbal medicine as a sedative for more than 200 years. Abuse of cinnabar as a sedative for infants has resulted in cases of Hg toxicity.120 HgS is almost insoluble in water. Orally administered HgS is absorbed and has been reported to accumulate in the kidney and liver.120 Impaired hearing and deafness have been reported to result from developmental and adult exposure to

36

Systemic Toxicity

methyl mercury, a chemical form of Hg.120,121 Mercury toxicity may be subclinical, and ABR has been reported to be a sensitive tool in detecting early subclinical Hg toxicity before clinical manifestations. Discalzi and colleagues in 1993 recorded ABRs from patients exposed to industrial Hg or lead, comparing them with age- and sex-matched control subjects who were never exposed to either heavy metal. They noted a significant prolongation of wave I–V time compared with control subjects.122 Counter and colleagues in 1997 found similar prolongation in mercury- and lead-exposed workers.123 Chang and colleagues in 1995 noted prolonged neural conduction times in auditory-evoked AP studies of patients exposed to Hg vapor.124 Animal studies have revealed similar results.120 Monkeys exposed to methyl mercury from birth displayed selective high-frequency deficits.125 Analysis of ABRs in mice treated with mercurial compounds revealed significant elevation of physiologic hearing threshold, as well as prolongation of interwave latency of I–V that increased with mean blood Hg level.120 Animal studies have aided our understanding of the pathophysiology of Hg toxicity. Inhibition of Na+–K+ adenosine triphosphatase activity and overproduction of nitric oxide in the brainstem subsequently alter neuronal activity in the auditory brainstem.120 This has been one of the postulated mechanisms for Hg ototoxicity. Lead Lead toxicity is a well-documented phenomenon with multisystemic toxic effects, particularly involving the peripheral and central nervous systems. A review of studies in humans suggests that lead toxicity affects the auditory system.126 Auditory thresholds of adults and children exposed to lead were noted to be elevated.127,128 ABRs were noted to be abnormal in lead-exposed children and adults.129,130 Discalzi and colleagues studied 49 lead-exposed workers (mean exposure was 7.4 years) and compared their ABRs with those of age- and sexmatched control subjects. They noted marked prolongation of interpeak differences in the study group.131 Animal studies suggest that the auditory pathways, especially the higher auditory centers, may be unusually sensitive to the toxic effects of lead.126 Yamamura and colleagues in 1989 reported that lead toxicity affected the VIIIth nerve CAP of adult guinea pigs but not the CM or the endocochlear potential.132 They noted that the cochlea was spared from lead toxicity. Lasky and colleagues conversely noted abnormal OAE in leadexposed monkeys, implicating cochlear pathology. They also noted similar ABR-prolonged latencies in experimental animals.126 These experimental results suggest that lead toxicity affects neural transmission primarily in auditory pathways.

SUMMARY • High-dose salicylate administration usually causes a reversible, bilaterally symmetric, mild to moderate hearing loss that may be flat or present in the higher frequencies. Threshold shifts typically recover after cessation of salicylate. • The duration of recovery varies from individual to individual. • High-pitched tinnitus is usually the first sign of impending salicylate ototoxicity and typically resolves when salicylates are discontinued. Serum salicylate levels do not correlate well with tinnitus in patients with salicylate ototoxicity. • Mechanisms for salicylate ototoxicity are multifactorial and multilocational. • Large doses of quinine have been associated with reversible hearing loss and tinnitus similar to those seen in salicylate ototoxicity. • Toxic levels of mercury and lead can result in auditory toxicity. Studies have suggested that the toxicity is primarily neural, affecting electrical transmission along the VIIIth nerve and brainstem.

REFERENCES 1. Jung TT, Rhee CK, Lee CS, et al. Ototoxicity of salicylates, nonsteroidal anti-inflammatory drugs and quinine. Otolaryngol Clin North Am 1993;26: 791–810. 2. Kalkanis J, Galtz F, Campbell KCM, Rybak LP. Inner ear, ototoxicity. Emedicine. Oct 14, 2002. Available at: http://www.emedicine.com/ent/topic699.htm (accessed March 21, 2002). 3. Scott PM, Griffiths MV. A clinical review of ototoxicology. Clin Otolaryngol 1994;19:3–8. 4. Boston Collaborative Drug Surveillance Program: drug induced deafness. JAMA 1973;224:515–6. 5. Grigor RR, Spitz PW, Frust DE. Salicylate toxicity in elderly patients with rheumatoid arthritis. J Rheumatol 1987;14:60–6. 6. Insel PA. Analgesic-antipyretics and anti-inflammatory agents; drugs employed in treatment of rheumatoid arthritis and gout. In: Gilman AG, Goodman LS, Rall TW, et al, editors. Goodman and Gilman’s the pharmacological basis of therapeutics. 8th ed. New York: Pergamon Press; 1990. p. 638–81. 7. Muller G. Beitrag zur Wirking der salicylasuren Natrons beim Diabetes Melleus. Berl Klin Wochenschr 1877;14:29–31. 8. Schwabach. Ueber Bleibende Storungen im Gehororgan nach Chinin und Salicylsauregebrauch. Dtsch Med Wochenschr 1884;10:163–6.


9. Ishii T, Bernstein J, Balogh K. Distribution of tritium-labelled salicylate in the cochlea. Ann Otol Rhinol Laryngol 1967;76:368–76. 10. Gutknecht J, Tosteson DC. Diffusion of weak acids through lipid bilayer membranes: effect of chemical reactions in the aqueous unstirred layer. Science 1973;182:1258–61. 11. Joy MM, Cutler DJ. On the mechanism of transport of salicylate and p-hydroxybenzoic acid across human cell membrane. J Pharm Pharmacol 1987;39:266–71. 12. Kakehata S, Santos-Sacchi J. Effects of salicylate and lanthanides on outer hair cell motility and associated gating charge. J Neurosci 1996;16:4881–9. 13. Woodford C, Henderson D, Hamernik R. Effects of combination of sodium salicylate and noise on the auditory threshold. Ann Otol Rhinol Laryngol 1978;87:117–27. 14. Juhn SK, Rybak LP, Jung TTK. Transport characteristics of the blood–labyrinth barrier. In: Drescher D, editor. Auditory biochemistry. Springfield (IL): Charles C Thomas; 1985. p. 488–99. 15. Silverstein H, Bernstein JM, Davies DG. Salicylate ototoxicity: a biochemical and electrophysiological study. Ann Otol Rhinol Laryngol 1967;76: 118–28. 16. Jastreboff PJ, Hansen R, Sasaki PG, Sasaki CT. Differential uptake of salicylate in serum, cerebrospinal fluid and perilymph. Arch Otolaryngol Head Neck Surg 1986;112:1050–3. 17. Day RO, Graham GG, Beiri D, et al. Concentration-response relationships for salicylate-induced ototoxicity in normal volunteers. Br J Clin Pharmacol 1989;28:695–702. 18. Myers EN, Bernstein JM. Salicylate ototoxicity. Arch Otolaryngol 1965;82:483–93. 19. Bernstein JM, Weiss AD. Further observations on salicylate ototoxicity. J Laryngol Otol 1967;89: 915–25. 20. Bonding P. Critical bandwidth in patients with a hearing loss induced by salicylates. Audiology 1979;18:133–44. 21. McFadden D, Plattsmier H, Pasanen E. Aspirininduced hearing loss as a model of sensorineural hearing loss. Hear Res 1984;16:251–60. 22. Halla JT, Atchinson SL, Hardin JG. Symptomatic salicylate ototoxicity: a useful indicator of serum salicylate concentration? Ann Rheum Dis 1991;50: 682–4. 23. Levy G. Clinical pharmacokinetics of salicylates: a re-assessment. Br J Clin Pharmacol 1980;10 Suppl 2:285S–90S. 24. Halla JT, Hardin JG. Salicylate ototoxicity in patients with rheumatoid arthritis: a controlled study. Ann Rheum Dis 1988;47:134–7.

37

25. DeMoura LF, Hayden RC. Salicylate ototoxicity. Arch Otolaryngol 1968;87:60–4. 26. Covell WP. A cytologic study on the effects of drugs on the cochlear. Arch Otolaryngol 1938;23:633–41. 27. Gotlib IL. Morphologic changes in the cells of the cochlear and vestibular analysers following ingestion of sodium salicylates. Vestn Otorinolaringol 1957;196:31–5. 28. Deer B, Hunter-Duvar I. Salicylate ototoxicity in the chinchilla: a behavioral and electromicroscope study. J Otolaryngol 1982;11:260–4. 29. Falk SA. Sodium salicylate. Arch Otolaryngol 1974;99:393. 30. Douek E, Dodson H, Bannister L. The effects of sodium salicylate on the cochlear of the guinea pig. J Laryngol Otol 1983;93:793–9. 31. Dieler R, Shehata-Dieler W, Brownell W. Concomitant salicylate-induced alterations of the outer hair cell subsurface cisternae and electromotility. J Neurocytol 1991;20:637–53. 32. Mitchell D, Brummett R, Himes D, et al. Electrophysiological study of the effect of sodium salicylate on the cochlea. Arch Otolaryngol 1973;98: 297–301. 33. McPherson D, Miller J. Choline salicylate: effects on cochlear function. Arch Otolaryngol 1974; 99:305–8. 34. Stypulkowski P. Mechanisms of salicylate ototoxicity. Hear Res 1990;49:113–46. 35. Cazals Y, Horner KC, Huang ZW. Alterations in average spectrum of cochleoneural activity by long-term salicylate treatment in the guinea pig: a plausible index of tinnitus. J Neurophysiol 1998; 80:2113–20. 36. Evans EF, Borerwe TA. Ototoxic effects of salicylates on the responses of the single cochlear nerve fibres and on cochlear potentials. Br J Audiol 1982;16:101–8. 37. Kumagai M. Effects of intravenous injection of aspirin on the cochlea. Hokkaido Igaku Zasshi 1992;67:216–33. 38. Chen G, Jastreboff PJ. Salicylate-induced abnormal activity in the inferior colliculus of rats. Hear Res 1995;82:158–78. 39. Jastreboff PJ, Sasaki CT. Salicylate-induced changes in spontaneous activity of single units in the inferior colliculus of the guinea pig. J Acoust Soc Am 1986;80:1384–91. 40. Manabe Y, Yoshida S, Saito H, Oka H. Effects of lidocaine on a salicylate-induced discharge of neurons in the inferior colliculus of the guinea pigs. Hear Res 1997;103:192–8. 41. Ochi K, Eggermont JJ. Effects of salicylate on neural activity in cat primary cortex. Hear Res 1996;95:63–76.

38

Systemic Toxicity

42. Johnson NJ, Eberling C. Evoked acoustic emissions from the human ear, I. Equipment and response parameters. Scand Audiol 1986;11:3–12. 43. McFadden D, Plattsmier H. Aspirin abolished spontaneous oto-acoustic emissions. J Acoust Soc Am 1984;76:443–8. 44. Long GR, Tubis A. Modification of spontaneous and evoked otoacoustic emissions and associated psychoacoustic microstructure by aspirin consumption. J Acoust Soc Am 1988;84:1343–53. 45. Brownell WE, Winston JB. Slow electrically evoked volume changes in guinea pig outer hair cells. In: Abstracts of the Twelfth Midwinter Meeting of the Association for Research in Otolarygology; 1989. p. 138. 46. Shehata WE, Brownell WE, Deiler R. Effects of salicylate on shape, electromotility and membrane characteristics of isolated outer hair cells from guinea pigs cochlear. Acta Otolaryngol 1991; 111:707–18. 47. Tunstall MJ, Gale LE, Ashmore JF. Action of salicylate on membrane capacitance of outer hair cells from the guinea pig cochlear. J Physiol 1995; 485:739–52. 48. Hawkins JE. Drug ototoxicity. In: Keidel WD, Neff WD, editors. Handbook of sensory physiology. Berlin: Springer-Verlag; 1976. p. 707–48. 49. Brien JA. Ototoxicity associated with salicylates. A brief review. Drug Saf 1993;9:143–8. 50. Didier A, Nuttall AL, Miller J. Sodium salicylate induced blood flow changes and hearing loss in guinea pig cochlea. In: Abstracts of the Thirteenth Midwinter Meeting of the Association for Research in Otolaryngology; 1990. p. 310. 51. Didier A, Miller JM, Nuttall AL. The vascular component of sodium salicylate ototoxicity in the guinea pig. Hear Res 1993;69:199–206. 52. Krzanowski J, Matschinsky F. A phosphocreatine gradient opposite to that of glycogen in the organ of Corti and the effect of salicylate on adenosine triphosphate and P-creatinine in cochlear structures. J Histochem 1971;19:321. 53. Jung T, Kim P, Kim D, et al. Effects of sodium salicylate on levels of catecholamine in the perilymph. In: Abstracts of the Thirteenth Midwinter Research Meeting of the Association for Research in Otolaryngology; 1990. p. 52. 54. Cazals Y, Li X, Aurousseau C, Didier A. Acute effects of noradrenalin related vasoactive agents on the ototoxicity of aspirin: an experimental study in the guinea pig. Hear Res 1988;36:89–96. 55. Fratianni T, Jung TTK, Miller SK, et al. Effect of adrenergic blockers on sodium salicylate induced ototoxicity. Otolaryngol Head Neck Surg 1990; 103:233.

56. Gunther T, Rebentisch E, Vormann J. Enhanced ototoxicity of salicylate by magnesium deficiency. Magnes Bull 1989;11:15–18. 57. Gunther T, Rebentisch E, Vormann J. Protection against salicylate ototoxicity by zinc. J Trace Elem Electrolytes Health Dis 1989;3:51–3. 58. Vane JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin like drugs. Nat New Biol 1971;231:232–5. 59. Brown RD, Feldman AM. Pharmacology of hearing and ototoxicity. Am Rev Pharmacol Toxicol 1978;18:233–52. 60. Jung T, Juhn S, Morizono T, et al. The effect of aspirin on the concentration of prostaglandins in the perilymph. In: Myers E, editor. New dimensions in otorhinolarygology-head neck surgery. Vol 2. Amsterdam: Elsevier Science; 1985. p. 134–5. 61. Kawata R, Urade Y, Tachibana E, et al. Prostaglandin synthesis by the cochlea. Prostaglandins 1988;35:173–84. 62. Jung T, Miller S, Park Y, et al. Effects of nonsteroidal anti-inflammatory drugs on levels of lipoxygenase products in perilymph and hearing. In: Abstracts of the Twelfth Midwinter Research Meeting of the Association for Research in Otolaryngology; 1989. p. 105. 63. Jung T, Miller S, Rozehnal S, et al. Effects of round window membrane application of salicylate and indomethacin on hearing and levels of arachidonic acid metabolites in perilymph. Acta Otolaryngol Suppl 1992;493:81–7. 64. Jung T, Park Y, Miller S, et al. Effects of exogenous arachidonic acid metabolites applied on round window membrane on hearing and their levels in perilymph. Acta Otolaryngol Suppl 1992;493:171–6. 65. Jung T, Miller S, Mowery G, et al. Effects of leukotriene inhibitor (L-663536) metabolites in salicylate ototoxicity. In: Abstracts of the Twelfth Midwinter Research Meeting of the Association for Research in Otolaryngology; 1989. p. 73. 66. Jung T, Park Y, Miller D. Effects of lipoxygenase inhibitor (SCH 37224) on hearing and perilymph levels of arachidonic acid metabolites in ototoxicity induced by sodium salicylate. In: Abstracts of the Thirteenth Midwinter Research Meeting of the Association for Research in Otolaryngology; 1990. p. 412. 67. Park YS, Jung TT, Choi DJ, Rhee CK. Effects of corticosteroid treatment on salicylate ototoxicity. Ann Otol Rhinol Laryngol 1994;103:896–900. 68. Jung T, Hwang A, Miller S. Effects of sodium salicylate and leukotriene blocker on cochlear blood flow. In: Abstracts of the Fourteenth Midwinter Research Meeting of the Association for Research in Otolaryngology; 1991. p. 117.


69. McCabe PA, Dey DL. The effect of aspirin upon auditory sensitivity. Ann Otol Rhinol Laryngol 1965;74:312–25. 70. McFadden D, Plattsmier H, Pasanen E. Aspirin can potentiate the temporary hearing loss induced by intense sounds. Hear Res 1983;9:295–316. 71. Carson S, Prazma J, Pulver S, et al. Combined effects of aspirin and noise in causing permanent hearing loss. Arch Otolaryngol Head Neck Surg 1989;115:1070–5. 72. Gignonx M, Martin H, Calgfinger H. Troubles cochleovestibulaire après tentative de suicide a l’aspirin. J Fr Oto-Rhinolaryngol 1966;15:631–5. 73. Jarvis JF. A case of unilateral permanent hearing deafness following acetylsalicylic acid. J Laryngol 1966;80:318–20. 74. Kapur YP. Ototoxicity of acetylsalicylic acid. Arch Otolaryngol 1965;81:134–8. 75. Koegel L Jr. Ototoxicity: a contemporary review of aminoglycosides, loop diuretics, acetylsalicylic acid, quinine, erythromycin and cisplatinum. Am J Otol 1985;6:190–9. 76. Perlman LV. Salicylate intoxication from skin application. N Engl J Med 1966;274:164–7. 77. Stebbins WC, Miller JM, Johnsson LG, Hawkins JE Jr. Ototoxic hearing loss and cochlear pathology in the monkey. Ann Otol Rhinol Laryngol 1969; 78:1007–25. 78. Gold A, Wilpizeski CR. Studies in auditory adaptation: the effects of sodium salicylate on evoked auditory potential in cats. Laryngoscope 1966; 76:674–85. 79. Eddy L, Morgan R, Carrey H. Hearing loss due to combined effects of noise and sodium salicylate. ISA Trans 1976;15:103–8. 80. Boettcher FA, Bancroft BR, Salvi RJ, et al. Effects of sodium salicylate on evoked response measures of hearing in the chinchilla. Hear Res 1989;42:129. 81. Jung T, Woo H, Baer W, et al. Effects of nonsteroidal anti-inflammatory drugs on the hearing and prostaglandin levels in the perilymph. Otolaryngol Head Neck Surg 1988;99:154. 82. Mongan E, Kelly P, Nies K, et al. Tinnitus as an indication of therapeutic serum salicylate levels. JAMA 1973;226:142–5. 83. Cazals Y, Huang ZW. Average spectrum of cochlear activity: a possible synchronized firing, its olivocochlear feedback and alterations under anaesthesia. Hear Res 1996;101:81–92. 84. Dolan DF, Nuttall AL, Avinash G. Asynchronous neural activity recorded from the round window. J Acoust Soc Am 1990;87:2621–7. 85. Schreiner CE, Synder RL. A physiological animal model of peripheral tinnitus. In: Feldmann H, editor. Proceedings of 3rd International Tinnitus

86.

87.

88. 89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

39

Seminar; 1987. Karlsruhe: Harsch Verlag; 1987. p. 100–6. Martin WH, Schwegler JW, Scheiblhoffer J, Ronis ML. Salicylate-induced changes in cat auditory nerve activity. Laryngoscope 1993;103:600–4. Feldmeier I, Lenarz T. An electrophysiological approach to the localization of tinnitus generators. In: Abstract book of 1996 Midwinter Meeting of Association for Research in Otolaryngology. Des Moines (IA): ARO Press; 1996. p. 208. Chapman P. Naproxen and sudden hearing loss. J Laryngol Otol 1982;96:163–6. Koopman CF Jr, Glattke TA, Caffrey JD. Effect of ibuprofen upon hearing in the guinea pig. Otolaryngol Head Neck Surg 1982;90:819–23. Morrison MD, Blakely BW. The effect of indomethacin on inner ear fluids and morphology. J Otolaryngol 1978;7:149–57. Malotte M, Jung TTK, Miller SK, et al. Effect of nonsteroidal anti-inflammatory drugs on cochlear blood flow. Otolaryngol Head Neck Surg 1990; 103:187. Tange RA, Dreschler WA, Claessen FAP, Perenboom RM. Ototoxic reactions of quinine in healthy persons and patients with Plasmodium falciparum infection. Auris Nasus Larynx 1997; 24:131–6. Webster LT. Drugs used in the chemotherapy of protozoal infections. In: Gilman AG, Goodman LS, Rall TW, et al, editors. Goodman and Gilman’s the pharmacological basis of therapeutics. 7th ed. New York: Macmillan; 1985. p. 1029–48. Claessen FA, Van Boxtel CJ, Perenboom RM, et al. Quinine pharmacokinetics: ototoxic and cardiotoxic effects in healthy Caucasian subjects and in patients with falciparum malaria. Trop Med Int Health 1998;3:482–9. Looareesuwan S, Charoenpan P, Ho M, et al. Fatal Plasmodium malaria after an inadequate response to quinine treatment. J Infect Dis 1990;161: 577–80. Roche RJ, Pukrittayakamee S, Looareesuwan S, et al. Quinine induced reversible high tone hearing loss. Br J Clin Pharmacol 1990;29:780–2. Nielsen-Abbring FW, Perenboom RM, Van Hulst JR. Quinine induced hearing loss. ORL J Otorhinolaryngol Relat Spec 1990;52:65–8. Mansor SM, Molyneux ME, Taylor TE, et al. Effect of Plasmodium falciparum malaria infection on the plasma concentration of alpha-acid glycoprotein and the binding of quinine in Malawian children. Br J Clin Pharmacol 1991;32:317–23. Trenholme GM, Williams RL, Rieckmann KH, et al. Quinine disposition during malaria and induced fever. Clin Pharmacol Ther 1976;19:459–67.

40

Systemic Toxicity

100. Krishna S, White NJ. Pharmacokinetics of quinine, chloroquine and amodiaquine. Clinical implications. Clin Pharmacokinet 1996;30:263–99. 101. Ruedi L, Furrer W, Luthy F, et al. Further observation concerning the toxic effects of streptomycin and quinine on the auditory organ of guinea pigs. Laryngoscope 1952;62:333–51. 102. Karlsson KK, Flock A. Quinine causes isolated outer hair cells to change length. Neurosci Lett 1990;116:101–5. 103. Dieler R, Davies C, Shehata-Dieler WE. The effects of quinine on active motile responses and fine structure of isolated outer hair cells from the guinea pig cochlea. Laryngorhinootologie 2002;81:196–203. 104. Smith DI, Lawrence M, Hawkins JE. Effects of noise and quinine on the vessels of stria vascularis. An image analysis study. Am J Otolaryngol 1985;6:280–9. 105. Karlsson KK, Ulfendahl M, Khanna SM, Flock A. The effects of quinine on the cochlear mechanics in the isolated temporal bone preparation. Hear Res 1991;53:95–100. 106. Kenmochi M, Eggermont JJ. Salicylate and quinine toxicity affect the central nervous system. Hear Res 1997:113:110–6. 107. Lawrence M. Circulation in the capillaries of the basilar membrane. Laryngoscope 1970;80: 1364–75. 108. Lee CS, Heinrich J, Jung TTK. Quinine-induced ototoxicity: alterations in cochlear blood flow. Otolaryngol Head Neck Surg 1992;107:233. 109. Horwitz CA. Autoimmune hemolytic anemia. II. Drug induced type. Postgrad Med 1979;66: 199–202. 110. Moss RA. Drug induced immune thrombocytopenia. Am J Hematol 1980;9:439–46. 111. Zajtchuk JT, Mihail R, Jewell JS, et al. Electronystagmographic findings in long term low-dose quinine ingestion. Arch Otolaryngol 1984;110: 788–91. 112. Manku MS, Horrobin DF. Chloroquine, quinine, procaine, quinidine, tricylic anti-depressants and methylxanthines as prostaglandin agonists and antagonists. Lancet 1976;ii:1115–7. 113. Markus HB, Ball EG. Inhibition of lipolytic processes in rat adipose tissue by antimalarial drugs. Biochim Biophys Acta 1969;187:486–91. 114. Eleno N, Botana L, Espinosa J. K+ channel blocking drugs induce histamine release and 45Ca uptake in isolated mast cells. Int Arch Allergy Appl Immunol 1990;92:162–7. 115. Miller JJ, editor. Antimalarial drugs. In: CRC handbook of ototoxicity. Boca Raton (FL): CRC Press; 1985. p. 9–15.

116. Jastreboff PJ, Brennan JF, Sasaki CT. Quinine induced tinnitus in rats. Arch Otolaryngol Head Neck Surg 1991;117:1162–6. 117. Quick CA. Chemical and drug effects on the inner ear. In: Paparella, Shumrick, editors. Otolaryngology. Vol. II. Philadelphia: WB Saunders; 1980. p. 1823. 118. Jager W, Idrizbegovic E, Karlsson KK, Alvan G. Quinine-induced hearing loss in the guinea pig is not affected by the Ca2+ channel antagonist verapamil. Acta Otolaryngol 1997;117:46–8. 119. Balfour AJ. The bite of Jesuits’ bark. Aviat Space Environ Med 1989;60(7 Pt 2):A4–5. 120. Chuu JJ, Hsu CJ, Lin-Shiau SY. Abnormal auditory brainstem responses for mice treated with mercurial compounds: involvement of excessive nitric oxide. Toxicology 2001;162(1):11–22. 121. Rice DC. Age related increase in auditory impairment in monkeys exposed in utero plus postnatally to methylmercury. Toxicol Sci 1998;44:191–6. 122. Discalzi G, Fabbro D, Meliga F, et al. Effects of occupational exposure to mercury and lead on brainstem auditory evoked potentials. Int J Psychophysiol 1993;14:21–5. 123. Counter SA, Buchanan LH, Laurell G, Ortega F. Blood mercury and auditory neuro-sensory responses in children and adults in Nambija gold mining area of Ecuador. Neurotoxicology 1998;19:185–96. 124. Chang YC, Yeh CY, Wang JD. Subclinical neurotoxicity of mercury vapor revealed by a multimodality evoked potential study of chloralkali workers. Am J Ind Med 1995;27:271–9. 125. Rice DC, Gilbert SG. Exposure to methyl mercury from birth to adulthood impairs high-frequency hearing in monkeys. Toxicol Appl Pharmacol 1992;115(1):6–10. 126. Lasky RE, Maeier MM, Snodgrass EB, et al. The effects of lead on otoacoustic emissions and auditory evoked potentials in monkeys. Neurotoxicol Teratol 1995;17:633–44. 127. Repko JD, Corum CR. Critical review and evaluation of the neurological and behavioral sequalae of inorganic lead absorption. CRC Crit Rev Toxicol 1979;6:135–87. 128. Schwartz J, Otto D. Blood lead, hearing thresholds and neuro-behavioral development in children and youth. Arch Environ Health 1987;42:153–60. 129. Holdstein Y, Pratt H, Goldsher M, et al. Auditory brainstem evoked potentials in asymptomatic leadexposed subjects. J Laryngol Otol 1986;100:1031–6. 130. Otto D, Robinson G, Bauman S, et al. Five year follow-up study of children with low to moderate lead absorption: electrophysiological evaluation. Environ Res 1985;38:168–86.


131. Discalzi GL, Capellaro F, Bottalo L, et al. Auditory brainstem evoked potential (BAEPs) in lead-exposed workers. Neurotoxicology 1992;13: 207–9.

41

132. Yamamura K, Terayama K, Yamamoto N, et al. Effects of acute lead acetate exposure on adult guinea pigs: electrophysiological study of the inner ear. Fundam Appl Toxicol 1989;13:509–15.

CHAPTER 4

Ototoxicity of Loop Diuretics Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), Andrew R. Scott, BM, BS, MPhil, FRCS(ORL-HNS), and John A. Rutka, MD, FRCSC

Loop diuretics are a class of diuretics widely used and occasionally implicated in ototoxicity. They are organic compounds that exert potent saliuretic effects by blocking the reabsorption of sodium and chloride from the epithelial cells in the loop of Henle and the proximal renal tubule.1,2 Loop diuretics comprise several groups of compounds with diverse chemical structures— sulfonamides, phenoxyacetic acid derivatives, and heterocyclic compounds.1 Furosemide, bumetanide, and ethacrynic acid are the most commonly used loop diuretics. Ethacrynic acid is usually reserved for those cases allergic or refractory to furosemide.1 Other members of this group include piretanide, azosemide, triflocin, and indapamide.2 Loop diuretics are generally used in the treatment of congestive cardiac failure, renal failure, cirrhosis, and hypertension. These diuretics, especially furosemide, are also widely used in the neonatal intensive care unit in the treatment of bronchopulmonary dysplasias.3 Patients at greatest risk for loop diuretic ototoxicity include those with renal impairment, premature infants, and possibly those receiving aminoglycoside antibiotics.2 As loop diuretics are eliminated by glomerular filtration, renal impairment will prolong their half-life. This increase in serum half-life may permit accumulation of these drugs in the inner ear, thereby increasing the risk for ototoxicity.2

ETHACRYNIC ACID Ethacrynic acid is a highly potent saliuretic diuretic. Chemically, it is composed of an unsaturated ketone derivative of an aryloxyacetic acid [2,3-dichloro-4-(2methylenebutyryl)-phenoxyacetic acid]. Ethacrynic acid is rarely used, except where subjects are allergic or refractory to furosemide. Ethacrynic acid inhibits active sodium reabsorption in the ascending limb of the loop of Henle.4 The recommended dose of intravenous ethacrynic acid for adults is about 1.0 mg/kg body weight. In patients with impaired renal function, serum levels rise quickly

because of impaired secretion of the drug, enhancing its toxic effects. Clinical Manifestations The ototoxicity of ethacrynic acid has been well documented by case reports of sensorineural hearing loss (SNHL) in humans and experimental animals. Ototoxicity in humans was first reported by Maher and Schreiner in 1965.5 They noted temporary hearing loss in patients treated with ethacrynic acid for refractory edema. The hearing recovered to its pretreatment level after 6 to 24 hours. This historical report was soon followed by additional case reports of temporary hearing loss noted by various authors.6–9 Subsequent publications have documented permanent hearing loss in several patients.10–13 However, most, if not all, of these reports involved patients with impaired renal function.14 David and Hitzig in 1971 reported a case of an oliguric patient who, soon after intravenous (IV) administration of 100 mg of ethacrynic acid, developed a severe hearing loss that persisted for 1 hour.15,16 Homer in 1971 reported a transient hearing loss of 4 hours in a patient who received 50 mg of IV ethacrynic acid for ascites.17 Meriwether and colleagues later documented seven cases of permanent SNHL related to ethacrynic acid, most of which occurred in patients with renal failure.1,5 Rybak, in 1988, reported a renal transplant recipient who sustained a permanent profound SNHL of middle–high frequency in both ears after receiving a total of 200 mg of ethacrynic acid intravenously.18 A cooperative study reported the incidence of hearing loss with ethacrynic acid therapy to be 7 patients per 1,000 treated.1,19 Reversible tinnitus and vertigo have also been reported after administration of loop diuretics.20 Mechanisms of Toxicity Experimental animal studies have aided our understanding of the mechanisms involved in the toxicity of

Ototoxicity of Loop Diuretics

loop diuretics. Studies looking at ethacrynic acid have suggested the following mechanisms for ototoxicity: Reduction in Endocochlear Potential Both ethacrynic acid and furosemide produce a doserelated, reversible reduction in endocochlear potential in animal experiments.20,21 Prazma and colleagues have demonstrated depression of the endocochlear potential following IV administration of ethacrynic acid in guinea pigs. They speculated that this was secondary to the inhibition of sodium–potassium (Na + –Ka + )activated adenosine triphosphatase (ATPase) of the stria vascularis, thought to be the origin of endocochlear potentials. 4,14 Silverstein and colleagues reported consistent and marked depressions in cochlear potentials after administration of ethacrynic acid with recovery of electrical potentials within 24 hours.16 They concluded that the cochlear potential changes were reversible and transient. Other studies have revealed reductions in cochlear microphonics (CM), summating potentials (SPs), and eighth nerve compound action potentials (APs) for variable periods following ethacrynic acid administration.22–25 Perilymph or Endolymph Electrolyte Alterations Several investigators have noted alterations in the Na+ and K+ and potassium concentrations of the labyrinthine fluids following the administration of ethacrynic acid. A marked increase in K+ in perilymph with a corresponding increase in the Na+ concentration, from 5.9 to 145 mEq/L, has been noted, as has a decrease in K+ concentration from 154 to 25 mEq/L in endolymph.26 It has also been suggested that ethacrynic acid inhibits active ion transport in the ductus cochlearis, reversing the K+ and Na+ concentration, thereby producing a reduction in cochlear potentials.26,27 This attractive theory, however, has been disputed by subsequent studies, where alterations of the cochlear potential were noted without intralabyrinthine electrolyte changes.28 Alterations in Hair Cell Glycogen Metabolism High concentrations of glycogen have been noted in the hair cells of the inner ear, and several researchers have reported that ethacrynic acid appears to have a direct inhibiting effect on glycolysis. 29 Duggan and Noll reported the ability of ethacrynic acid, similar to certain cardiac glycosides, to inhibit the Na+–K+-activated ATPase, an enzyme associated with active ion transport.30 Other researchers have found similar results.31,32 Morphologic and Histologic Changes Morphologic changes have been documented in ethacrynic acid ototoxicity both in animal models and human temporal bones. Degenerative changes have been noted in the stria vascularis in both guinea pigs

43

and cats after ethacrynic acid administration.28 Atrophy of the intermediate layer of the stria vascularis was particularly prominent.33 Characteristic edema and cystic changes in stria vascularis, especially in the intercellular spaces, have also been noted. Thickening of the stria and irregularities of the marginal cell borders facing the endolymph were also identified. The membranous labyrinth, hair cells, and spiral ganglia appeared normal. Examination of animal temporal bones 1 month after initial administration revealed a normal-looking organ of Corti and a marked decrease in intercellular strial edema. These morphologic changes were noted to recover more slowly when compared with inner ear electrical potentials.16 Freeze-fracture studies of guinea pigs’ cochlea after ethacrynic acid administration also revealed alterations in the gap junction morphology in the stria vascularis. These changes suggest that a physiologic uncoupling of stria cells could have occurred in response to ethacrynic acid.1,34 Martz and colleagues and Martz and Hinojosa reported histologic changes in humans after exposure to ethacrynic acid.10,35 They noted edema of the stria vascularis, loss of outer hair cells (OHCs) in the basal turn of the cochlea, and cystic degeneration of hair cells in the ampulla of the posterior semicircular canal and macula of saccule.

FUROSEMIDE Furosemide represents the most widely used loop diuretic, especially in the treatment of congestive cardiac failure, renal failure, cirrhosis, and hypertension. It is also widely used in bronchopulmonary dysplasia, a common sequela among premature infants with low birth weight treated for respiratory distress syndrome with oxygen and ventilator therapy. 3 Furosemide appears to improve pulmonary function by promoting the clearance of excess pulmonary interstitial edema. Adverse effects, such as volume loss and chloride depletion, nephrocalcinosis, and ototoxicity, can occur.36 Metabolic alkalosis, hypokalemia, and cholelithiasis appear related to the long-term administration of furosemide. 37 This is more pronounced in premature infants in whom reduced renal tubular secretion of furosemide into the urine causes an accumulation of furosemide in plasma to potentially ototoxic levels.38 Furosemide’s serum half-life is normally 47 to 53 minutes but has been found to be significantly higher in neonates and patients with renal failure.2 Clinical Manifestations Clinical experience with furosemide indicates that it may produce transient or permanent hearing loss in a fashion similar to ethacrynic acid following either oral or IV administration.2 Reports of ototoxicity surfaced soon after furosemide was introduced into clinical use.39 Schwartz and colleagues in 1970 reported temporary

44

Systemic Toxicity

hearing loss in five patients who received high doses of furosemide intravenously. Two of the patients also developed vertigo, suggesting possible vestibular toxicity as well.39 Similar reports surfaced in 1971 and 1973, documenting furosemide ototoxicity.40,41 Administrative Effects Associated with Ototoxicity Rate of Infusion The rate of infusion of furosemide has been noted to influence ototoxicity. Clinical studies have suggested that rapid infusion of furosemide increases the incidence of ototoxicity. Heiland and Wigand reported that the infusion of furosemide at a constant rate of 25 mg per minute caused noticeable hearing loss in two-thirds of patients. However, when the infusion rate was reduced to 15 mg per minute in patients with severe renal impairment, only minor hearing losses were noted. They suggested that furosemide not be given at a rate of more than 4 mg per minute.42 Bosher in 1977 reported patients who had received furosemide at a rate of 25 mg per minute developing symptoms of ototoxicity, whereas those whose dosage rate was 5.6 mg per minute or less developed no symptoms of ototoxicity.43 Bolus Dosing Large bolus dosings of furosemide appear in general to be more toxic to the inner ear. Venkateswaran reported transient severe hearing loss in a patient with chronic renal failure and edema who was given 500 mg of furosemide as an IV bolus injection over 3 minutes. The hearing loss, however, was recovered after 4 hours. A later infusion of 240 mg over 5 minutes apparently did not cause subjective hearing loss.44 Large bolus dosings have been reported to cause tinnitus without hearing loss. In one study, a large IV dose of 3,200 mg of furosemide given over less than 4 hours caused severe tinnitus that spontaneously disappeared after the drug was stopped or the rate of infusion was reduced. No subjective hearing loss was reported by the patients, and normal audiograms were obtained on testing.45 Route of Administration In a study of patients with renal failure, those receiving only oral furosemide demonstrated no hearing loss. One patient, however, who received furosemide 480 mg/d orally and 60 mg/d intravenously for 10 days, had an average hearing loss of 25 dB in the right ear and 30 dB in the left ear.46 Notwithstanding, a critical review of the world literature reveals that the incidence of furosemide ototoxicity is not entirely dependent on dose or route of administration. Freis and colleagues, for example, studied a group of patients with renal failure who received 500 to 1,000 mg of furosemide by infusion over 6 hours.

They found no evidence of hearing impairment in this study group. 47 Similarly, Rastogi and colleagues reported no hearing loss in patients receiving large oral doses of furosemide, up to 2 g/d.48 Conversely, other authors have reported permanent hearing loss from much smaller oral doses of furosemide.49–51 Permanent hearing loss with furosemide after IV administration has also been reported, although most cases of furosemide ototoxicity have been reversible.52–54 Other Considerations Audiometric studies in patients receiving high doses of furosemide have demonstrated some unexpected findings. Wigand and colleagues identified acute reversible hearing loss, greatest in the middle frequencies, after rapid infusion of 1,000 mg of furosemide over 40 minutes in patients with renal impairment.55 Tuzel compared the ototoxic effects of bumetanide (another loop diuretic) and furosemide. He reviewed audiometric changes in patients receiving either bumetanide or furosemide and found that only 1.1% of patients receiving bumetanide had a minimum 15 dB elevation of pure tone audiometry thresholds, compared with 6.4% of patients receiving furosemide.56 Several studies in neonates have also confirmed the ototoxicity of furosemide in this age group. Salamy and colleagues studied 224 low-birth-weight infants with repeated auditory brainstem response (ABR) testing, first in the newborn nursery, then at 6-month intervals for the first 2 years of follow-up, and annually until the age of 4 years. SNHL was statistically associated with greater amounts of furosemide administered for longer periods and in combination with aminoglycoside antibiotics. 3,57 Brown and colleagues studied 35 neonates with SNHL and in 70 age-matched control subjects with normal ABRs. Several factors, including seizures and exposure to anticonvulsant drugs, kanamycin, and furosemide, were associated with SNHL. After multivariate analysis, however, only the exposure to furosemide remained a consistent risk factor for hearing loss.3,58 Mechanisms of Toxicity The mechanisms of furosemide ototoxicity are probably similar to those of ethacrynic acid. Further experimental studies with furosemide in animal models, however, have revealed some new findings for consideration. Reductions in Endocochlear Potentials and Auditory Electrical Pathway Activity Chodynicki and Kostrzewska first demonstrated that systemic furosemide administrations reduced endocochlear potentials in guinea pigs.59 Subsequent studies have demonstrated a dose-related reduction of endocochlear potentials in animal experiments following


furosemide administration.60–64 In addition, furosemide has been reported to reduce CM potentials and reduce the amplitude of the cochlear-vestibular nerve APs in animal studies.25,65,66 Studies involving animal auditory nerves have revealed that furosemide causes a temporary loss of sharpness of tuning in the cochlear nerve fibers that have characteristic frequencies of 7 to 30 kHz.67 The spontaneous firing rate of the auditory nerve was noted to be reduced in proportion to the amount of reduction of the endocochlear potential.68 Furosemide has also been reported to depress evoked cortical potentials in cats and brainstem-evoked responses in guinea pigs.69,70 Relationship to Hypoalbuminemia Animal studies have revealed that rats deficient in albumin are more susceptible to furosemide ototoxicity than those with a normal serum albumin. Doseresponse curves of both endocochlear potentials and compound APs revealed that the dose of furosemide required to produce half-maximal changes in albumindeficient rats was about one-half of the dose required in normal rats. This supports the hypothesis that the ability of furosemide to reach its site of ototoxic action in the cochlea depends on the amount of unbound furosemide in the serum.71 Morphologic and Histologic Changes Morphologic changes following furosemide ototoxicity have been extensively reported. However, no permanent damage to the hair cells has been identified in temporal bone studies involving microscopic examination of the cochlea in both guinea pigs and dogs.72,73 Nevertheless, strial edema and degeneration of intermediate cells in guinea pigs treated with 200 mg/kg of IV furosemide have been noted.54 Strial edema correlated well with alterations in endocochlear potentials in this model.74 Other changes have been noted. The endolymph– perilymph barrier, for example, appears to be altered by the administration of furosemide. Rarey and Ross found a lower number of apical ridges and greater than normal depths of the strands in the tight junctions of the marginal cells of the stria vascularis. 75 Greater cross-linking of the more basal strands was also noted. Endolymphatic sac morphology appeared altered after the administration of furosemide. Increased cytoplasmic contents of endoplasmic reticulum and more prominent Golgi structures of the light cells have been noted immediately after administration. After 10 days of administration, the epithelial cytoarchitecture of the endoplasmic sac was altered, with pronounced veiling of the light cells by the dark cells in one study.76 OHCs and their stereocilia can be affected by furosemide. Histochemical studies in animals by Tamura and by Comis and colleagues have suggested

45

that the oxidative metabolism of OHCs is impaired.77,78 Splaying of OHCs, stretching or breakage of tip links, and erosion and fracture of cross-links between stereocilia of furosemide-treated guinea pigs have been documented.79,80 Human temporal bone studies of a patient with suspected ototoxicity from a loop diuretic revealed granular and densely staining hair cells in the vestibular neuroepithelium and organ of Corti, particularly in the basal turn. The endoplasmic reticulum of some spiral cells appeared dilated. The major cytologic changes, however, were found in the stria vascularis of the cochlea and dark cell areas of the vestibular neuroepithelium.20 Biochemical Changes Biochemical alterations have also been noted in experimental animal studies. In animal studies, the adenylate cyclase complex of the stria vascularis can be inhibited by both furosemide and ethacrynic acid. Further observation revealed that the G-protein complex that regulates adenylate cyclase is affected by furosemide. 81 Brusilow reported increased endolymph Na+ concentration and reduced K+ activity in endolymph with furosemide, in conjunction with decreased endocochlear potential.60 Interestingly, some compounds have been noted to reduce the ototoxic effect of furosemide on the endocochlear potential in animal studies. These include quinine, iodinated benzoic acid derivatives, and probenecid.82–84

BUMETANIDE Bumetanide, introduced in the 1970s, inhibits Na+ transport in the thick ascending limb of the loop of Henle. It is structurally and pharmacologically similar to furosemide, but it is approximately 40 times more potent by mass. After oral administration, it is rapidly absorbed, with peak serum concentrations attained at approximately 30 minutes. Its pharmacokinetic properties are similar to those of furosemide. Bumetanide has demonstrated efficacy in the management of edema associated with congestive cardiac failure, liver cirrhosis, and renal impairment. Its adverse effects profile is similar to that of furosemide, although the incidence of hypochloremia and hypokalemia appears to be greater. The incidence of hyperglycemia and ototoxicity, however, seems to be greater with furosemide. As a result, the principal indication for bumetanide may be in patients with an increased risk for ototoxicity. Unfortunately, cost considerations have relegated bumetanide to a secondary role for the treatment of Na+ and fluid retention in most clinical settings.85 Reports of ototoxicity first surfaced shortly after bumetanide’s introduction to the marketplace. Asano

46

Systemic Toxicity

and colleagues in 1974 reported hearing loss in a patient who received 1 mg of bumetanide for 2 weeks.1,86 Nevertheless, Bourke reported hearing loss in a patient after furosemide administration that was completely recovered when bumetanide was substitute for furosemide.1,87 Evans and colleagues carried out a prospective study of patients receiving bumetanide by performing high-frequency audiograms (8–20 kHz). They could not detect any significant changes in pure tone threshold of those frequencies.1,88 Brown compared IV bumetanide and furosemide in beagle dogs to study the effects of these agents on CM and primary auditory afferent activity. He noted that the acute ototoxic potency of bumetanide in beagle dogs was 6.5 times that of furosemide, whereas its diuretic potency was 40 to 60 times that of furosemide. When the clinical dosages of the two drugs were considered, the relative acute ototoxic potency of bumetanide in the dogs was approximately 0.11 to 0.16 times that of furosemide. A similar range has been obtained in cats. Histologic examination of the cochlea did not reveal any significant pathology.89 In summary, bumetanide is probably a more potent diuretic with less ototoxic potential than furosemide has. Relative cost appears to be the ratelimiting factor in its clinical use. Nevertheless, it would be appropriate to consider this medication as an alternative to furosemide, especially when patients exhibit symptoms suggestive of ototoxicity from furosemide.

OTHER LOOP DIURETICS Other class agents include torsemide, piretanide, azosemide, ozolinone, and indacrinone. They are rarely used clinically. Torsemide, for example, produces a more prolonged water and electrolyte excretion than equipotent doses of furosemide, but it does not cause kaliuresis to the same extent. Studies have confirmed the efficacy of low-dose torsemide (2.5–5 mg/d) in the treatment of hypertension and shown it to be effective when administered orally at a dosage of 5 to 20 mg/d in the management of congestive cardiac failure. Dosage of torsemide of 20 mg/d has been noted to increase plasma renin levels, angiotensin II activity, and urinary dopamine and prostaglandin E secretion. Adverse effects are usually mild and transient in nature. No evidence of ototoxicity has been demonstrated to date, and torsemide does not appear to affect blood glucose levels, serum uric concentrations, or serum potassium levels at dosages below 5 mg/d.90 Animal studies have shown no permanent hearing impairment, even at very high doses of torsemide.91 Piretanide, however, has been shown in animal studies to reduce endocochlear potentials and compound APs.1,92,93 Azosemide, ozolinone, and indacrinone have been shown to affect compound APs in

animal studies as well. 93,94 Both ozolinone and indacrinone have relatively less ototoxic potential than do other loop diuretics.

SUMMARY • Both reversible SNHL and permanent SNHL have been reported in patients receiving loop diuretics. • Patients at greatest risk for loop diureticinduced ototoxicity include those with renal impairment, premature infants, and those receiving aminoglycoside antibiotics. • Furosemide ototoxicity clinically has been primarily identified in those who have received furosemide intravenously or a large bolus dose over a short period, or with a rapid rate of infusion (typically > 5 mg/min). • Bumetanide appears more potent and less ototoxic when compared with furosemide and as such would represent a reasonable alternative in patients who exhibit symptoms suggestive of ototoxicity from furosemide.

REFERENCES 1. Rybak LP. Ototoxicity of loop diuretics. Otolaryngol Clin North Am 1993;26:829–44. 2. Koegel L Jr. Ototoxicity: a contemporary review of aminoglycosides, loop diuretics, acetylsalicylic acid, quinine, erythromycin and cisplatinum. Am J Otol 1985;6:190–9. 3. Henley CM, Rybak LP. Developmental ototoxicity. Otolaryngol Clin North Am 1993;26:857–71. 4. Prazma J, Thomas WG, Fischer ND, Prslar MJ. Ototoxicity of the ethacrynic acid. Arch Otolaryngol 1972;95:448–56. 5. Maher JF, Schreiner GE. Studies on ethacrynic acid in patients with refractory edema. Ann Intern Med 1965;62:15–29. 6. Schneider WJ, Becker L. Acute transient hearing loss after ethacrynic acid therapy. Arch Intern Med 1966;117:715–7. 7. Matz GJ, Nauton RF. Ototoxic drugs and poor renal function. JAMA 1968;206:2119. 8. Hanzelik E, Peppercorn M. Deafness after ethacrynic acid. Lancet 1969;i:416. 9. Meriwether WD, Mangi RJ, Serpick AA. Deafness following standard intravenous dose of ethacrynic acid. JAMA 1971;216:795–8. 10. Matz JM, Beal DD, Krames L. Ototoxicity of ethacrynic acid. Arch Otolaryngol 1969;90:152–5. 11. Ng PS, Conley CE, Ing TS. Deafness after ethacrynic acid. Lancet 1969;i:673–4. 12. Pillay VKG, Schwartz FD, Aimi K, et al. Transient and permanent deafness following treatment with ethacrynic acid in renal failure. Lancet 1969; i:77–9.


13. Kohonen A, Jauhiainen T, Tarkkanen J. Experimental deafness caused by ethacrynic acid. Acta Otolaryngol 1970;70:187–9. 14. McCurdy JA Jr, McCormick JG, Harrill JA. Ototoxicity of ethacrynic acid in the anuric guinea pig. Arch Otolaryngol 1974;100:143–7. 15. David DS, Hitzig P. Diuretics and ototoxicity. N Engl J Med 1971;284:1328. 16. Silverstein H, Begin R. Ethacrynic acid—its reversible ototoxicity. Laryngoscope 1974;84:976–89. 17. Homer M. Deafness after ethacrynic acid. N Engl J Med 1971;285:1151. 18. Rybak LP. Ototoxicity of ethacrynic acid (a persistent clinical problem). J Laryngol Otol 1988; 102:518–20. 19. Boston Collaborative Drug Surveillance Program. Drug induced deafness: a cooperative study. JAMA 1973;224:515–6. 20. Arnold W, Nadol JB Jr, Geidauer H. Ultrastructural histopathology in a case of human ototoxicity due to loop diuretics. Acta Otolaryngol 1981;91: 391–414. 21. Rybak LP. Pathophysiology of furosemide ototoxicity. Otolaryngology 1982;11:127–33. 22. Goldman WJ, Bielinski TC, Mattis PA. Cochlear microphonic potential response of the dog to diuretic compounds. Toxicol Appl Pharmacol 1973;25:259–66. 23. Komune A, Morimitsu T. Dissociation of the cochlear microphonics and endocochlear potential after ethacrynic acid injection. Arch Otorhinolaryngol 1982;241:149–56. 24. Syka J, Melichar I. Comparison of the effects of furosemide and ethacrynic acid upon the cochlear function in the guinea pig. Scand Audiol Suppl 1981;14 Suppl:63–9. 25. Brown RD, McElwee TW Jr. Effects of intraarterially and intravenously administered ethacrynic acid and furosemide on cochlear N1 in cats. Toxicol Appl Pharmacol 1972;48:157–69. 26. Wilson KS, Juhn SK. The effect of ethacrynic acid on perilymph Na and K. Pract Otorhinolaryngol 1970;32:279–87. 27. Cohn ES, Gordes EH, Brusilow SW. Ethacrynic acid effects on the composition of cochlear fluids. Science 1971;171:910–1. 28. Silverstein H, Yules RB. The effect of diuretics on cochlear potentials and inner ear fluids. Laryngoscope 1971;81:873–88. 29. Klahr S, Bourgoignie JJ, Yates J, et al. Inhibition of glycolysis by ethacrynic acid and furosemide. Fed Proc 1971;30:608. 30. Duggan DE, Noll RM. Effect of ethacrynic acid and cardiac glycosides upon a membrane adenosine triphosphatase of the renal cortex. Arch Biochem 1965;109:388–96.

47

31. Hoffman JF, Kregenov FM. The characterization of new energy-dependant cation transport processes in red blood cells. Ann N Y Acad Sci 1966; 137:566–76. 32. Banerjee SP, Khanna VK, Sen AK. Inhibition of sodium and potassium-dependent adenosine triphosphatase by ethacrynic acid: two modes of action. Mol Pharmacol 1970;6:680–90. 33. Quick CA, Duvall AJ. Early changes in the cochlear duct from ethacrynic acid: an electronmicroscope evaluation. Laryngoscope 1970;80:954–65. 34. Forge A. Gap junctions in the stria vascularis and effects of ethacrynic acid. Hear Res 1984;13: 189–200. 35. Matz GJ, Hinojosa R. Histopathology following use of ethacrynic acid. Surg Forum 1973;24:488. 36. Miceli JJ, Kramer PA, Chapron DJ, et al. Pharmacokinetics and dynamics of furosemide in the newborn piglet. J Pharmacol Exp Ther 1990;253: 1126–32. 37. Guignard JP, Dubourg L, Gouyon JB. Diuretics in the neonatal period. Rev Med Suisse Romande 1995;115(8):583–90. 38. Chemtob S, Papageorgiou A, DuSouich P, Aranda JV. Cumulative increase in serum furosemide concentration following repeated doses in the newborn. Am J Perinatol 1987;4:203–5. 39. Schwartz GN, David DS, Riggo RR, et al. Ototoxicity induced by furosemide. N Engl J Med 1970; 181:1413–4. 40. Morrison JC, Fort AJ, Fish SA. Diuretic induced ototoxicity in pre-eclampsia. J Tenn Med Assoc 1971;64:36–7. 41. Cooperman LG, Rubin IL. Toxicity of ethacrynic acid and furosemide. Am Heart J 1973;85:831–4. 42. Hiedland H, Wigand ME. The effect of furosemide at high doses on auditory sensitivity in patients with uremia. Klin Wochenschr 1970;48:1052–6. 43. Bosher SK. Ethacrynic ototoxicity as a general model in cochlear pathology. Adv Otorhinolaryngol 1977;22:81–9. 44. Venkateswaran PS. Transient deafness from high doses of furosemide. BMJ 1971;3:113–4. 45. Cantarovich F, Locatelli A, Fernandez JC, et al. Furosemide in high doses in the treatment of acute renal failure. Postgrad Med J 1971;47 Suppl: 13–7. 46. Muth RG. Furosemide in severe renal insufficiency. Postgrad Med J 1971;47 Suppl:21–5. 47. Fries D, Pozet N, Dubois N, et al. The use of large doses of furosemide in acute renal failure. Postgrad Med J 1971;47 Suppl:18–20. 48. Rastogi SP, Volans G, Elliott RW, et al. High dose furosemide in the treatment of hypertension in chronic renal insufficiency and of terminal renal failure. Postgrad Med J 1971;47 Suppl:45–53.

48

Systemic Toxicity

49. Gallagher KL, Jones JK. Furosemide-induced ototoxicity. Ann Intern Med 1979;91:744–5. 50. Keefe RE. Otoxicity from oral furosemide. Drug Intell Clin Pharm 1978;2:428. 51. Rifkin SI, DeQuesada AM, Pickering MJ, et al. Deafness associated with oral furosemide. South Med J 1978;71:86–8. 52. Brown CG, Ogg CS, Cameron JS, et al. High dose furosemide in acute reversible intrinsic renal failure. Scott Med J 1974;19 Suppl 1:35–8. 53. Lloyd-Mostyn RM, Lord IJ. Ototoxicity of intravenous furosemide. Lancet 1971;2:1156. 54. Quick CA, Hoppe W. Permanent deafness associated with furosemide administration. Ann Otol 1975;84:94–101. 55. Wigand ME, Heidland A. Ototoxic side effects of high doses of furosemide in patients with uremia. Postgrad Med J 1971;47 Suppl:54–6. 56. Tuzel IH. Comparison of adverse reactions to bumetanide and furosemide. J Clin Pharmacol 1981;21:615–9. 57. Salamy A, Eldredge L, Tooley WH. Neonatal status and hearing loss in high risk infants. J Pediatr 1989;114;847–52. 58. Brown DR, Watchko JF, Sabo D. Neonatal sensorineural hearing loss associated with furosemide: a case control study. Dev Pharmacol Ther 1989; 13:70–7. 59. Chodynicki S, Kostrzewska A. Effect of furosemide and ethacrynic acid on endolymph potential in guinea pigs. Otolaryngol Pol 1974;28:5–8. 60. Brusilow SW. Propanolol antagonism to the effect of furosemide on the composition of endolymph in guinea pig. Can J Physiol Pharmacol 1976; 54:42–8. 61. Green TP, Rybak LP, Mirkin BL, et al. Pharmacologic determinants of ototoxicity of furosemide in the chinchilla. J Pharmacol Exp Ther 1981;216: 537–42. 62. Kusakari J, Ise I, Comegys TH, et al. Effect of ethacrynic acid, furosemide and ouabain upon the endolymphatic potential and upon high energy phosphates of the stria vascularis. Laryngoscope 1978;88:12–37. 63. Ng PSY, Conley CE, Ing TS. Deafness after ethacrynic acid. Lancet 1969;1:673–4. 64. Rybak LP, Whitworth C. Some organic acids attenuate the effects of furosemide on the endocochlear potential. Hear Res 1987;26:89–93. 65. Comis SD, Pratt SR, Hayward TL. The effect of sulfamyl loop diuterics on the crossed olivo-cochlear bundle. Neuropharmacology 1979;18:739–41. 66. Jung W, Schon F. Effects of loop diuretics on the inner ear. A quantitative evaluation using computer technics. Arch Otorhinolaryngol 1979; 224:143–7.

67. Klinke R, Evans EF. The effects of drugs on the sharpness of the tuning of single cochlear nerve fibres. Pflugers Arch 1974;347 Suppl:R53. 68. Sewell WF. The relation between the endocochlear potential and spontaneous activity in the auditory nerve fibres in the cat. J Physiol 1984;347:685–96. 69. Jung W, Rosskopf K. Evoked response audiometry (ERA) am Meerschweinchen vor und nach LasixInduriertem Horstorz. Laryngol Rhinol Otol (Stuttg) 1975;54:411–8. 70. Mathog RH, Matz GJ. Ototoxic effects of ethacrynic acid. Ann Otol 1972;81:871–6. 71. Whitworth C, Morris C, Scott V, Rybak LP. Doseresponse relationships for furosemide ototoxicity in rat. Hear Res 1993;71:202–7. 72. Brown RD, Manno JE, Daigneault EA, et al. Comparative acute ototoxicity of intravenous bumetanide and furosemide in the pure bred beagle. Toxicol Appl Pharmacol 1979;48:157–69. 73. Federspil P, Mausen H. Experimentelle untersuchungen zur Ototoxicatat des Furosemids. Res Exp Med 1973;161:175–84. 74. Pike D, Bosher SK. The time course of the strial changes produced by intravenous furosemide. Hear Res 1980;3:79–89. 75. Rarey KE, Ross MD. A survey of the effects of loop diuretics on the zonulae occludentes of the perilymph-endolymph barrier by freeze fracture. Acta Otolaryngol 1982;94:307–16. 76. Erwall C, Friberg U, Bagger-Sjoback D, RaskAndersen H. Effects of ototoxic diuretics (loop diuretics) on the endolymphatic sac. ORL J Otorhinolaryngol Relat Spec 1988;50:42–53. 77. Tamura H. Histochemical study on pathogenesis of the inner ear damage caused by ethacrynic acid and furosemide. Audiology Jpn 1978;21:668–87. 78. Comis SD, Pratt SR, Hayward TL. The effect of furosemide, piretanide and bumetanide on cochlear succinic dehydrogenase. Neuropharmacology 1981;2:405–7. 79. Comis SD, Osborne MP, Jeffries D Jr. The effect of furosemide upon the morphology of hair bundles in the guinea pigs cochlear hair cells. Acta Otolaryngol 1990;109:49–56. 80. Forge A, Brown AM. Ultrastructural and electrophysiological studies of acute ototoxic effects of furosemide. Br J Audiol 1982;16:109–16. 81. Koch T, Gloddek B. Inhibition of adenylate cyclasecoupled G protein complex by ototoxic diuretics and cis-platinum in the inner ear of the guinea pig. Eur Arch Otorhinolaryngol 1991;248:459–64. 82. Rybak LP, Whitworth C. Quinine reduces noxious cochlear effects of furosemide and ethacrynic acid. Am J Otolaryngol 1988;9:238–43. 83. Rybak LP, Green TP, Juhn SK, et al, Probenecid reduces cochlear effects and perilymph penetra-


84.

85.

86.

87. 88.

89.

tion of furosemide in chinchilla. J Pharmacol Exp Ther 1984;230:706–9. Hirashima N. Blocking effect of radio-constrast media on cochlear depression. Ann Otol Rhinol Laryngol 1978;87:32–6. Halstenson CE, Matzke GR. Bumetanide a new loop diuretic (Bumex, Roche Laboratories). Drug Intell Clin Pharm 1983;17:786–97. Asano Y, Masazawa H, Tabata Y, et al. Clinical trial of bumetanide. Basic Pharmacol Ther 1974; 2:47–54. Bourke E. Furosemide, bumetanide and ototoxicity. Lancet 1976;1:917–8. Evans EA, Klinke R. The effects of intracochlear and systemic furosemide on the properties of single cochlear nerve fibres in the cat. J Physiol 1982; 331:409–28. Brown RD. Comparative acute cochlear toxicity of intravenous bumetanide and furosemide in the

49

pure bred beagle. J Clin Pharmacol 1981;21(11–12 Pt 2):620–7. 90. Dunn CJ, Fitton A, Brogden RN. Torasemide. An update of its pharmacological properties and therapeutic efficacy. Drugs 1995;49:121–42. 91. Klinke R, Mertens M. Quantitative assessment of torasemide ototoxicity. Arzneimittelforschung 1988;38:153–5. 92. Rybak LP, Whitworth C. Comparative ototoxicity of furosemide and piretanide. Acta Otolaryngol 1986;101:59–65. 93. Gottl KH, Roesch A, Klinke R. Quantitative evaluation of ototoxic side effects of furosemide, piretanide, bumetanide, azosemide and ozolinone in the cat—a new approach to the problem of ototoxicity. Naunyn Schmiedebergs Arch Pharmacol 1985;331:275–82. 94. Rybak LP, Whitworth C. Ototoxicity of indacrionone is stereospecific. Hear Res 1987;31:169–74.

CHAPTER 5

Clinical Uses of Cisplatin Jeremy Sturgeon, MD, FRCPC

Of all the drugs used in the modern chemotherapy of cancer, cisplatin occupies a unique role in causing significant ototoxicity in a large percentage of patients treated. In 1965, Rosenberg and his coworkers observed that an electric current delivered between platinum electrodes inhibited the proliferation of Escherichia coli.1,2 This inhibition was related to the formation of inorganic platinum-containing moieties, in the presence of ammonium and chloride ions. Cisplatin, or cis-diamminedichloroplatinum (II), was noted to be the most active platinum complex in experimental tumor systems, and it was soon introduced into clinical oncology in the early and mid-1970s. As an oncologic drug, it has been extensively tested against most human tumors.

RELEVANT CLINICAL PHARMACOLOGY Mechanism of Action and Resistance The toxicity of cisplatin is believed to be related to the development of platinum deoxyribonucleic acid (DNA) interstrand cross-links and also to the formation of intrastrand bifunctional N7 adducts at d(GpG) and d(ApG).3 Other effects of cisplatin may be related to the inhibition of sodium–potassium adenosine triphosphatase, transport of essential amino acids, calcium channel function, and mitochondrial function.4–6 The mechanisms of resistance are believed to include reduced cellular drug accumulation, cytosolic inactivation of drug, and the enhancement of DNA repair. Certain genes and proteins may be important in determining the sensitivity of cells to cisplatin (see Chapter 6, “Ototoxicity of Platinum Compounds”).7–9 Pharmacokinetics After an intravenous injection of cisplatin, peak plasma levels are reached almost immediately, and these decrease by over 50% within 2 hours. The clearance of cisplatin is triphasic, with a distribution half-life (t 1/2 α) of 13 minutes, an elimination half-life (t 1/2 β) of 43 minutes, and a terminal half-life (t 1/2 γ) of 5.4 days.10

About 25% of a dose of cisplatin is eliminated from the body during the first 24 hours, with renal clearance accounting for more than 90%. Toxicity The use of cisplatin is associated with severe nausea and vomiting in almost all patients.11 The introduction of 5-hydroxytryptamine (5-HT3) receptor antagonists into clinical use has dramatically reduced the amount and severity of nausea and vomiting with this drug.12 The effects on vomiting have been shown to be significantly better with this class of drugs than the effects on nausea, which remains a problem in patient management. The regular use of these agents permits treatment with cisplatin to be administered in an ambulatory care setting. Neurotoxicity (Including Ototoxicity) The neurotoxicity associated with the use of cisplatin includes peripheral sensory neuropathy, high-frequency hearing loss, and autonomic neuropathy, most commonly producing constipation. Much more rarely, seizures and encephalopathy have been reported. Neuropathy has been shown to be dose dependent and can occur in over 80% of patients who have a cumulative dose greater than 300 mg/m 2. In up to 50% of patients, the neuropathy is not reversible.13 High-frequency hearing loss is believed to be caused by damage to the outer hair cells in the organ of Corti and is permanent.14 Although several types of drugs are being evaluated to protect patients from neurotoxicity, these are largely experimental and include free oxygen radical scavengers and nucleophilic sulfur thiols.15 Today neurotoxicity has become the major dose-limiting toxicity of this drug (see Chapter 6,“Ototoxicity of Platinum Compounds”). Nephrotoxicity The nephrotoxicity of cisplatin can be severe but can be moderated with appropriate therapy. 16,17 Proximal tubular damage leads to a decreased reabsorption of

Clinical Uses of Cisplatin

sodium and water and, subsequently, impairment of distal tubular re-absorption, renal blood flow, and glomerular filtration lead to an enhanced excretion of enzymes, proteins, and other electrolytes, particularly potassium and magnesium. In early clinical trials, it was shown clearly that aggressive hydration with normal saline, together with infusion of hypertonic saline and with mannitol-induced diuresis, could significantly reduce the renal toxicity of cisplatin. It is important to note that in patients who have preexisting renal impairment, cisplatin should be used with caution and should not be used if the glomerular filtration rate (GFR) is less than 30 mL/min. Full doses are usually employed if the GFR is more than 50 mL/min.18 Drugs such as amifostine, a thiol-containing compound, have been shown to reduce renal toxicity from cisplatin.19 Amifostine is preferentially taken up and metabolized by normal cells. The active thiol moiety acts as a nucleophile, which can inactivate carbonium ions generated by alkylating agents and prevent DNA damage. Clinical trials have shown that treatment with amifostine can protect against cisplatin-induced nephrotoxicity without compromising the antitumor effects of this drug.

CLINICAL USES OF CISPLATIN When cisplatin was combined with other drugs, including bleomycin and vinblastine, it had significant curative potential in treating metastatic germ cell tumors, which were hitherto almost always fatal. When it was first tested in epithelial ovarian cancers, it produced significantly higher response rates than had previously been seen with alkylating agents and antimetabolites. Cisplatin in combination with cyclophosphamide, with or without the addition of doxorubicin, rapidly became the standard treatment for ovarian cancer. Cisplatin has been widely used in the treatment of lung cancer, head and neck cancer, and carcinoma of the cervix. In some of these diseases, it has been combined with external beam irradiation therapy and has significantly improved treatment outcomes. More recently, cisplatin has been employed in salvage regimens for patients with lymphoma whose disease has failed to respond or has progressed after initial chemotherapy and who are being considered for autologous stem cell transplant. Germ Cell Tumors of the Testis Testicular cancer represents the primary example of a disease in which cisplatin has been proven to cure large numbers of patients when combined with other effective agents. Further, in this disease site the substitution of carboplatin for cisplatin produces inferior results, and cisplatin-containing combination chemotherapy remains the standard of care after more than 25 years in clinical use.

51

Table 5-1 Chemotherapy Regimens in Testicular Cancer Treatment Regimen

Dose

Frequency

BEP* Bleomycin

30 U

IV weekly on days 2, 9, and 16

Etoposide

100 mg/m2

IV daily × 5

Cisplatin

20 mg/m2

IV daily × 5

Etoposide

100 mg/m2

IV daily × 5

Cisplatin

20 mg/m2

IV daily × 5

Etoposide

75 mg/m2

IV daily × 5

Ifosfamide

1.2 g/m2

IV daily × 5

EP †

VP-I-P ‡

Cisplatin

20 mg/m

2

IV daily × 5

IV = intravenous. * Cycles repeated every 21 days for 3 or 4 cycles. † Cycles repeated every 21 days. ‡ Cycles repeated every 21–28 days.

Currently, the most widely used regimen for managing nonseminomatous testicular cancer in North America and Europe is the BEP regimen, using for its name the initials of its components: bleomycin, etoposide, and platinum (cisplatin) (Table 5-1).20 These drugs are used for three or four cycles of treatment, depending on the risk category of the individual patient. Although short-term toxicity is moderately severe, patients with this condition are generally young and tolerate treatment well. Long-term studies have shown that, with the use of cisplatin, high-frequency hearing is permanently impaired and GFR is permanently reduced as demonstrated by serial measurements of creatinine clearance before and after treatment. Normal fertility is recovered in at least 50% of patients treated with this regimen. A randomized trial at Memorial Hospital compared four cycles of etoposide and cisplatin with four cycles of etoposide and carboplatin (EP) in 270 patients who were considered to have good-risk testicular cancer. Although the complete response rates were comparable in both arms, the event-free and relapse-free survivals were less favorable in the arm containing carboplatin, such that the authors recommended restricting carboplatin to the category of investigational status in patients with germ cell tumors.21 Because of the risk of serious lung toxicity from bleomycin, studies have been undertaken to see whether this drug can be safely removed from the treatment regimen. Comparisons of treatment with and without bleomycin have led to a good deal of controversy.22 At

52

Systemic Toxicity

Table 5-2 Randomized Trials of Chemotherapy with Cisplatin in Ovarian Cancer Study

Kaye et al 25

McGuire et al 26

Erlich et al 27

Conte et al 28

Number of Patients

159

485

56

133

Regimen

Platinum Dose (mg/m 2)

Median Survival (mo)

CP

50 q3 wk × 6

17

CP

100 q3 wk × 6

29

CP

50 q3 wk × 8

20

CP

100 q3 wk × 4

21

CAP

CEP

50 q3 wk × 6

23

100 q4 wk × 3

27

50 q4 wk × 6

24

100 q4 wk × 6

29

CAP = cyclophosphamide, doxorubicin, and cisplatin; CEP = cyclophosphamide, epidoxorubicin, and cisplatin; CP = cyclophosphamide, and cisplatin.

the present time, for good-risk patients, it is proposed that standard treatment include three cycles of BEP, equivalent to four cycles of EP. For patients with poorrisk testicular cancer, the standard of care at this time is the use of four cycles of BEP. Although the cure rate is high in this disease, approximately 30% of patients who have poor-risk testicular cancer will fail to achieve a complete remission with four courses of BEP or will undergo a relapse, which occurs usually within 1 to 2 years of completing initial therapy. Such patients receive salvage regimens that include vinblastine, ifosfamide, cisplatin, and paclitaxel. 23 At least 25% of patients treated with second-line chemotherapy will achieve stable complete remission. The use of high-dose chemotherapy and stem cell transplant for patients who relapse after primary therapy remains controversial, although it is clear that a subset of patients may be cured with this aggressive approach.24 Despite these drugs being used to treat testicular cancer for over 25 years, it has not been possible to decrease the doses of chemotherapy in a way that would reduce ototoxicity or nephrotoxicity. Because of the high curability of this disease, most patients are aggressively managed and most are willing to accept the predicted toxicities when they are offered the chance of cure of their disease. Although undesirable, most of these toxicities are not life threatening. For metastatic seminomas of the testis, the standard of care today usually includes the use of etoposide and cisplatin without bleomycin. Some studies in the literature suggest that for certain patients with metastatic seminoma, complete remission and long-term cure may be achieved with the use of cisplatin alone. Despite this, most patients are currently treated with the twodrug combination. Toxicity is moderately severe but of short duration, and the complete response and longterm survival rates are high.

Epithelial Ovarian Cancer When cisplatin was first introduced into clinical practice, it was rapidly found to be an active agent in the treatment of advanced epithelial ovarian cancer. Prior to this time, patients were treated with single alkylating agent chemotherapy, and clinical remissions were of short duration. When cisplatin was first used, it was rapidly introduced into combination chemotherapy. This happened in the 1970s, when superiority had been demonstrated for combination chemotherapy compared with single-agent treatment for the management of patients with advanced Hodgkin’s disease and also in the treatment of men with advanced testicular cancer. The assumption was therefore made that combination chemotherapy would probably prove superior in the management of other malignant diseases. However, 20 years later there is still uncertainty as to the superiority of single-agent versus combination chemotherapy in the management of epithelial ovarian cancer. Throughout the 1980s, the standard management of patients with advanced ovarian cancer reflected in most reviews and textbooks included either the combination of cyclophosphamide and cisplatin or the combination of cyclophosphamide, doxorubicin, and cisplatin.25–28 Unfortunately, although this treatment produced a high response rate, 70 to 75% in most trials of advanced ovarian cancer, the vast majority of patients relapsed from treatment and died from their disease despite subsequent chemotherapy. The median survival was 22 to 26 months (Table 5-2). Once this became clearly recognized, more attention was paid to the toxicity of treatment when weighed against the probable outcome of therapy. In order to answer some of the questions concerning the most appropriate therapy for advanced ovarian cancer, the Advanced Ovarian Cancer Trialists’ Group undertook a meta-analysis of all of the randomized trials in the literature. The most recent update from this


53

Table 5-3 Randomized Trials of Paclitaxel-Cisplatin* versus Paclitaxel-Carboplatin in Advanced Ovarian Cancer

Study

Number of Patients

Ozols et al33

798

Sandercock et al34

Neijt et al35

798

258

Median PFS (mo)

Median Overall Survival (mo)

p

Paclitaxel (135 mg/m 2/24 h) + cisplatin

19.4

49

NA

Paclitaxel (175/3 h) + carboplatin AUC 7.5

20.7

57


17.2

43.3

Paclitaxel (185 mg/m 2/3 h) + carboplatin AUC 6

19.1

44.1


16

30

Paclitaxel (175 mg/m 2/3 h) + carboplatin AUC 5

16

32

Arms

NS

NS

AUC = area under curve; NA = not available; NS = not significant; PFS = progression-free survival. *Dose of cisplatin 75 mg/m2.

group in 2002 suggested the following: (1) platinumcontaining combination regimens are superior to similar regimens without cisplatin; (2) platinum-combination chemotherapy is superior to single-agent platinum therapy; and (3) there is no difference in efficacy between cisplatin and carboplatin. 29 Many studies address the question of dose intensity of the drugs employed, and the conclusion that can be drawn is that intensifying the dose beyond the accepted standard leads to unacceptable toxicity. Until the late 1980s, therefore, standard treatment for advanced ovarian cancer was 500 mg/m 2 to 1,000 mg/m2 cyclophosphamide with 50 mg/m2 to 100 mg/m2 cisplatin. Controversy continues as to the efficacy of adding doxorubicin to this regimen.30 In the late 1980s, the Gynecologic Oncology Group (GOG) presented evidence from a randomized trial comparing cisplatin and cyclophosphamide with cisplatin plus a new taxane compound, paclitaxel. The results of this study showed an improvement in overall survival of more than 12 months in the patient arm treated with paclitaxel.31 A subsequent intergroup trial followed from the European Organization for Research and Treatment of Cancer (EORTC) and from the National Cancer Institute of Canada (NCIC) and the Nordic Gynecological Study Group (NOCOVA).32 This large trial confirmed a survival advantage to paclitaxel plus cisplatin as compared with the previous standard of cyclophosphamide plus cisplatin. These studies used different doses of each drug and different time schedules for the administration of paclitaxel. From these studies, it was widely concluded that the new standard of care for patients with advanced ovarian cancer was the use of paclitaxel plus cisplatin.

Because carboplatin had come into clinical use by this time and was recognized as producing less nephrotoxicity and ototoxicity, although with greater myelosuppression than with cisplatin, several clinical trial groups explored the use of carboplatin together with paclitaxel in ovarian cancer. The goal of these studies was to determine whether the use of carboplatin would lead to reduced toxicity without impairing the high response rate seen in patients treated with combination chemotherapy. Table 5-3 presents information from randomized trials comparing combination regimens with either cisplatin or carboplatin. The conclusion has been drawn that there is no disadvantage to the use of paclitaxel and carboplatin; therefore, in North America, this regimen is now widely accepted as the standard of care.33–35 One additional result from a clinical trial has added to the controversy over the most appropriate initial management of advanced ovarian cancer. A large European study, ICON3, compared the combination regimen of cyclophosphamide, doxorubicin, and cisplatin with single-agent carboplatin in over 1,500 patients with ovarian cancer who required chemotherapy. No survival difference was seen in either group.36 The median survival was 33 months in both groups. This result questioned the assumption that combination chemotherapy was superior to single-agent treatment, as well as the assumption that an anthracycline-containing treatment was better than a non–anthracycline-containing treatment. Therefore, either no regimen has demonstrated superiority to single-agent carboplatin or no regimen has demonstrated superiority to the combination of paclitaxel plus carboplatin.

54

Systemic Toxicity

In summary, most patients with advanced ovarian cancer are probably being treated with the combination of paclitaxel and carboplatin unless additional features such as advanced age and poor performance status suggest a more conservative approach, in which case single-agent carboplatin or, less often, cisplatin is employed. This treatment provides a high clinical complete remission rate of 75%; long-term survival is seen in only a small minority (< 5%) of patients. Intraperitoneal Chemotherapy in Ovarian Cancer For over 20 years, investigators have studied the use of intraperitoneal chemotherapy with cisplatin for patients with ovarian cancer. Such treatment has attracted interest because this disease is largely confined to the peritoneal cavity and because higher doses of treatment can be administered intraperitoneally with increased pharmacokinetic advantages. To date, no firm conclusions have been reached about the efficacy of intraperitoneal treatment or of the advantages of this route of treatment as compared with standard systemic intravenous chemotherapy. Patients with small amounts of residual disease in the peritoneal cavity would be most likely to respond to treatment by the intraperitoneal route. Cisplatin is one of the drugs shown to provide maximal penetration of tumors. Several clinical trials have been carried out on patients with modest amounts of residual tumor comparing intraperitoneal with intravenous chemotherapy. These trials suggest that progression-free survival improves by approximately 25% in the intraperitoneal therapy arm. Moreover, overall survival may be improved in patients treated by this route by approximately 8 months.37,38 Further clinical trials of large numbers of patients would be necessary before a major shift in current clinical practice occurs favoring the use of intravenous chemotherapy for all stages of epithelial ovarian cancer. Cancer of the Cervix When megavoltage irradiation was developed in the 1950s, it became the mainstay of treatment for patients with invasive cervical cancer. Approximately 50% of patients with this disease have locally advanced disease at the time of diagnosis. With pelvic irradiation, 5-year survival rates of 65% overall have been achieved for patients with carcinoma of the cervix. Recently, strategies use concurrent chemotherapy with radiation in the treatment of this disease. Cisplatin remains the single agent most likely to produce clinical responses in patients with metastatic cancer of the cervix, although such responses are mostly of short duration. It has been proposed that combining chemotherapy with cisplatin and radiation may increase the killing of tumor cells. Chemotherapy may act synergistically with radiation to inhibit the repair of radiation-induced damage, and it may also act independently to increase the rate of death

of tumor cells. Several recent trials comparing radiotherapy alone with radiation and concurrent cisplatin chemotherapy have shown a significant improvement in the control of pelvic disease and also in overall survival.39–41 Because of these studies, the National Cancer Institute issued a recommendation that strong consideration be given to incorporating concurrent cisplatinbased chemotherapy with radiation therapy in women who require radiation therapy for the treatment of cervical cancer. Many centers employ weekly doses of cisplatin of 40 mg/m2, to a maximum individual dose of 70 mg IV. Although the systemic toxicity from the cisplatin is not usually severe, local reactions may be more marked at the site of radiation, and patients need careful monitoring of their renal function and of their levels of magnesium and potassium, which may require supplementation. Squamous Cell Cancer of the Head and Neck Until recently, chemotherapy was largely employed only in the palliative management of patients with metastatic head and neck cancer, but more recently it has been added to the early treatment of this disease with surgery and radiation. A more aggressive approach has been developed in recognition of the fact that patients who present with locally advanced disease have expected cure rates of approximately 50% with the standard approach to treatment of surgery and radiation therapy. Treatment failure with these methods is usually ascribed to an inability to control local and regional disease because of inadequate margins at surgery and because of the presence of lymph node metastases within the neck. The addition of chemotherapy at the onset of treatment is based on the supposition that improved drug delivery to tumors with intact vascular beds may lead to an improvement in local control. Theoretically, systemic micrometastases can be treated at an early stage in their development. Most of the induction regimens presently used employ cisplatin or 5fluorouracil (5-FU). Complete clinical remission rates of 20 to 50% are seen in patients who have locally advanced disease, but as yet this has not been shown to lead to an improvement in patient survival.42,43 Two randomized trials have shown that in laryngeal cancer, laryngeal preservation could be achieved with induction chemotherapy followed by radiation therapy and that survival rates were similar to those achieved with surgical resection followed by postoperative radiation therapy for laryngeal cancer.44 At the present time, concurrent chemotherapy and radiotherapy are being increasingly explored. A metaanalysis of chemotherapy in the Head and Neck Cancer Collaborative Group review of 65 randomized trials conducted between 1965 and 1993 compared local treatment in locoregionally advanced disease with or without chemotherapy. A significant absolute survival


55

Table 5-4 Randomized Trials of Chemotherapy with Cisplatin in Metastatic Transitional Cell Carcinoma of the Bladder

Study

NBCCG

46

Number of Patients

Chemotherapy

109

Cisplatin CP

ECOG

47

SCSG 48

Intergroup 49

135

91

244

Response Rate (%)

20

Median Survival (mo)

Survival

Not stated

No difference

No difference

11.9

Cisplatin

17

6

CAP

33

7.3

Cisplatin

15

5.2

CAP

21

7.2

Cisplatin

12

8.2

MVAC

39

12.5

No difference

MVAC superior

CAP = cyclophosphamide, doxorubicin, and cisplatin; CP = cyclophosphamide and cisplatin; ECOG = Eastern Cooperative Oncology Group; MVAC = methotrexate, vincristine, doxorubicin, and cisplatin; NBCCG = National Bladder Cancer Collaborative Group; SCSG = Southeastern Cancer Study Group.

advantage of 4% at 5 years was found for patients receiving concurrent chemotherapy and radiation.45 Most studies have demonstrated that platinum-based chemotherapy and concurrent radiotherapy lead to better local control as well as to increased survival when they are compared with radiation alone. Several chemotherapeutic agents have activity when combined with radiation in the treatment of advanced head and neck cancer. These include cisplatin, 5-FU, carboplatin, methotrexate, and mitomycin-C. Of these, cisplatin has been studied most extensively because of its properties as a radiation sensitizer. It is not clear whether combination chemotherapy is superior to single-agent chemotherapy when given concurrently with radiation for head and neck cancer. Chemotherapy-induced mucositis remains a very severe problem in this treatment group. The optimum radiation fractionation schedules are not clear. The optional use of chemotherapy and radiation remains to be defined by further clinical trials in head and neck cancer. Bladder Cancer Transitional cell carcinomas of the bladder are sensitive to several chemotherapeutic drugs, including cisplatin, methotrexate, doxorubicin, ifosfamide, gemcitabine, and paclitaxel. The most active single agent is cisplatin, and response rates of between 15 and 30% have been reported from phase II and III trials in patients with metastatic disease. Whereas some studies comparing cisplatin alone with cisplatin-containing combinations have shown no difference in overall survival, the MVAC regimen (methotrexate, vincristine, doxorubicin, and cisplatin) has been shown to improve survival when compared with cisplatin alone (Table 5-4).46–49 With the MVAC regimen, long-term survival is about 4%, but

there is significant toxicity and treatment-related mortality for this population, which is generally of advanced age with concurrent medical problems. More recent studies have identified promising results from phase II studies of gemcitabine and cisplatin and one randomized phase III clinical trial comparing MVAC with gemcitabine and cisplatin, showed that the gemcitabine and cisplatin regimen led to similar response rates and survival with a lower treatment-related mortality. In carcinoma of the bladder, preexisting renal impairment may present difficulties with the administration of cisplatin-based chemotherapy. Where appropriate, obstruction can be managed with stents or nephrostomies, thereby permitting the use of this agent. Lung Cancer Small Cell Lung Cancer Small cell lung cancer tends to disseminate early in the course of its natural history and to display aggressive clinical behavior. Death is usually a result of disseminated systemic disease, and, as a result, much attention has been paid to the use of chemotherapy in the primary treatment of this disease. Small cell carcinoma is sensitive to a wide variety of chemotherapy agents. Among the more active agents are nitrogen mustard, doxorubicin, methotrexate, ifosfamide, etoposide, vincristine, vindesine, nitrosoureas, cisplatin, and carboplatin.50,51 Randomized trials conducted in the 1970s showed that combination chemotherapy produced superior survival compared with single-agent treatment. For the last 20 years, therefore, combination chemotherapy has been the primary method of managing this disease.52,53 The earliest combinations, from the late 1970s and early 1980s, tended to employ cyclophosphamide-based regimens, often including doxorubicin and vincristine.

56

Systemic Toxicity

Slightly later, regimens containing cisplatin and etoposide became widely used. When the combination of cyclophosphamide, doxorubicin, and vincristine was compared in trials with cisplatin and etoposide, there was no clear survival advantage for either regimen for patients with advanced-stage disease. The treatment choice is often based on a patient’s other medical problems. Patients with preexisting renal disease or preexisting neuropathy may not be offered a cisplatincontaining regimen. Similarly, patients with preexisting heart disease may not be appropriate candidates to receive cisplatin because of the requirement for aggressive intravenous hydration. With most of the active regimens, objective response rates in the range of 80 to 90% are seen, with complete remissions occurring in up to one-half of patients, depending on their clinical stage at presentation. The best results are seen in patients with good performance status and limited-stage disease. Such patients may enjoy median survivals up to 20 months.54 Non–Small Cell Lung Cancer Non–small cell lung cancer accounts for about 80% of patients with a diagnosis of lung cancer. The major histologic subtypes of non–small cell lung cancer are adenocarcinoma and squamous cell carcinoma. A smaller number of patients present with large cell carcinoma or bronchioalveolar carcinoma. In contradistinction to small cell lung cancer, which is highly responsive to chemotherapy, until recently, the role of chemotherapy in non–small cell lung cancer was less clear. Clinical data show that patients with advanced-stage non–small cell lung cancer who receive combination chemotherapy have an improved median survival of 3 to 4 months and also 1-year survival (35–40%). Studies have also shown that the quality of life of patients receiving such combination chemotherapy is improved when thoracic radiation is given as well.55 The regimens chosen to treat non–small cell lung cancer are many and varied, but most of them contain cisplatin or carboplatin.56 Paclitaxel with carboplatin or cisplatin is used frequently in the United States, whereas in Canada combined vinorelbine and cisplatin is usually used as the first line of treatment. A combination of gemcitabine with cisplatin is also considered to be effective. Many new classes of chemotherapy agents are undergoing clinical evaluation in non–small cell lung cancer. Patients with this disease should be entered into clinical trials of new agents whenever possible. In early-stage non–small cell lung cancer, adjuvant chemotherapy must be considered experimental. Lymphoma Cisplatin does not play a significant role in the primary treatment of Hodgkin’s disease or the non-Hodgkin’s lymphomas. For patients with Hodgkin’s disease and for

patients with intermediate-grade non-Hodgkin’s lymphomas, induction chemotherapy regimens have become established over the last 25 years, and they lead to a high proportion of patients achieving complete remission. When patients relapse, they may be considered for high-dose chemotherapy and autologous stem cell transplant. Before being accepted into such a program, such patients must be shown to have chemotherapy-sensitive disease. In this setting they are treated with a variety of regimens, some of which include cisplatin, such as the ESHAP 57 regimen (etoposide, methylprednisolone, cytarabine, and cisplatin) and the DHAP58 regimen (dexamethasone, cytarabine, and cisplatin). A more recent regimen includes the use of cisplatin with gemcitabine. With such regimens, more than 60% of patients will achieve remissions, and of these a proportion may go on to high-dose chemotherapy and receive an autologous stem cell transplant. There are occasions when cisplatin-containing protocols such as these are used in patients who will not be proceeding to transplant. For various reasons, such as age, the mortality risk from transplant may be considered excessive, but if such patients have a good performance status and no serious concurrent medical problems, a further attempt at intensive chemotherapy may be justified.

SUMMARY • The curative role of cisplatin in the treatment of germ cell tumors has been established for more than 20 years, and this drug has maintained its place in first-line therapy. • In other diseases such as ovarian cancer, despite the initial success with cisplatin in achieving remissions, it has gradually been supplanted by carboplatin because of a reduction in nephrotoxicity and neurotoxicity, without jeopardizing response rates or survival time. In other disease settings the development of other drugs has led to the replacement of cisplatin in many first-line chemotherapy regimens. Cisplatin, however, retains its place as one of the most important drugs used in combination therapy for the management of malignant disease. • The toxicity profile of cisplatin is well recognized. To this end, nephrotoxicity may be largely prevented by aggressive hydration and correction of electrolyte imbalance. The use of drugs to reduce the severity of peripheral neuropathy has been described. Unfortunately, there is no known way to prevent dose-related ototoxicity despite its predictability. Future developments in chemoprotective agents might help minimize and possibly prevent this complication. • The decision to use cisplatin in any clinical situation must always be based upon the balance between the expected outcome of treatment and


the toxicity, as well as on the identification of other medical problems that may increase the likelihood of developing side effects, such as preexisting renal disease or neuropathy.

REFERENCES 1. Rosenberg B, Vancamp L, Krigas T. Inhibition of cell division in Escherichia coli by electrolysis products from a platinum electrode. Nature 1965;205:698–9. 2. Rosenberg B, Van Camp L, Grimley EB, et al. The inhibition of growth or cell division in Escherichia coli by different ionic species of platinum (IV) complexes. J Biol Chem 1967;242:1347–52. 3. Pinto AL, Lippard SJ. Binding of the antitumor drug cis-diamminedichloroplatinum (II) (cisplatin) to DNA. Biochim Biophys Acta 1985; 780:167–80. 4. Guarino AM, Miller DS, Arnold ST, et al. Platinate toxicity: past, present, and prospects. Cancer Treat Rep 1979;63:1475–83. 5. Vassilev PM, Kanazirska MP, Charamella LJ, et al. Changes in calcium channel activity in membranes from cis-diammine-dichloroplatinum (II)-resistant and -sensitive L1210 cells. Cancer Res 1987; 47:519–22. 6. Gordon JA, Gattone VH. Mitochondrial alterations in cisplatin-induced acute renal failure. Am J Physiol 1986;250:F991–8. 7. Gately DP, Howell SB. Cellular accumulation of the anticancer agent cisplatin: a review. Br J Cancer 1993;67:1171–6. 8. Fujii R, Mutoh M, Niwa K, et al. Active efflux system for cisplatin in cisplatin-resistant human KB cells. Jpn J Cancer Res 1994;85:426–33. 9. Parker RJ, Eastman A, Bostick-Bruton F, et al. Acquired cisplatin resistance in human ovarian cancer cells is associated with enhanced repair of cisplatin-DNA lesions and reduced drug accumulation. J Clin Invest 1991;87:772–7. 10. Vermorken JB, van der Vijgh WJ, Klein I, et al. Pharmacokinetics of free and total platinum species after short-term infusion of cisplatin. Cancer Treat Rep 1984;68:505–13. 11. Gralla RJ, Itri LM, Pisko SE, et al. Antiemetic efficacy of high-dose metoclopramide: randomized trials with placebo and prochlorperazine in patients with chemotherapy-induced nausea and vomiting. N Engl J Med 1981;305:905–9. 12. Marty M, Pouillart P, Scholl S, et al. Comparison of the 5-hydroxytryptamine3 (serotonin) antagonist ondansetron (GR 38032F) with high-dose metoclopramide in the control of cisplatin-induced emesis. N Engl J Med 1990;322:816–21. 13. Cersosimo RJ. Cisplatin neurotoxicity. Cancer Treat Rev 1989;16:195–211.

57

14. Stadnicki SW, Fleischman RW, Schaeppi U, et al. Cisdichlorodiammineplatinum (II) (NSC-119875): hearing loss and other toxic effects in rhesus monkeys. Cancer Chemother Rep 1975;59:467–80. 15. Alberts DS, Noel JK. Cisplatin-associated neurotoxicity: can it be prevented? Anticancer Drugs 1995;6:369–83. 16. Ozols RF, Corden BJ, Jacob J, et al. High-dose cisplatin in hypertonic saline. Ann Intern Med 1984;100:19–24. 17. Al Sarraf M, Fletcher W, Oishi N, et al. Cisplatin hydration with and without mannitol diuresis in refractory disseminated malignant melanoma: a Southwest Oncology Group Study. Cancer Treat Rep 1982;66:31–5. 18. Kintzel PE, Dorr RT. Anticancer drug renal toxicity and elimination: dosing guidelines for altered renal function. Cancer Treat Rev 1995;21:33–64. 19. Kemp G, Rose P, Lurain J, et al. Amifostine pretreatment for protection against cyclophosphamide-induced and cisplatin-induced toxicities: results of a randomized control trial in patients with advanced ovarian cancer. J Clin Oncol 1996;14:2101–12. 20. Einhorn LH, Williams SD, Loehrer PJ, et al. Evaluation of optimal duration of chemotherapy in favorable-prognosis disseminated germ cell tumors: a Southeastern Cancer Study Group protocol. J Clin Oncol 1989;7:387–91. 21. Bajorin DF, Sarosdy MF, Pfister DG, et al. Randomized trial of etoposide and cisplatin versus etoposide and carboplatin in patients with goodrisk germ cell tumors: a multiinstitutional study. J Clin Oncol 1993;11:598–606. 22. Loehrer PJ Sr, Johnson D, Elson P, et al. Importance of bleomycin in favorable-prognosis disseminated germ cell tumors: an Eastern Cooperative Oncology Group trial. J Clin Oncol 1995;13:470–6. 23. Loehrer PJ Sr, Gonin R, Nichols CR, et al. Vinblastine plus ifosfamide plus cisplatin as initial salvage therapy in recurrent germ cell tumor. J Clin Oncol 1998;16:2500–4. 24. Bhatia S, Abonour R, Porcu P, et al. High-dose chemotherapy as initial salvage chemotherapy in patients with relapsed testicular cancer. J Clin Oncol 2000;18:3346–51. 25. Kaye SB, Paul J, Cassidy J, et al. Mature results of a randomized trial of two doses of cisplatin for the treatment of ovarian cancer. Scottish Gynecology Cancer Trials Group. J Clin Oncol 1996;14:2113–9. 26. McGuire WP, Hoskins WJ, Brady MF, et al. Assessment of dose-intensive therapy in suboptimally debulked ovarian cancer: a Gynecologic Oncology Group study. J Clin Oncol 1995;13:1589–99. 27. Ehrlich CE, Einhorn L, Stehman FB, et al. Treatment of advanced epithelial ovarian cancer using

58

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

Systemic Toxicity

cisplatin, Adriamycin and cytoxan—the Indiana University experience. Clin Obstet Gynaecol 1983;10:325–35. Conte PF, Bruzzone M, Carnino F, et al. High-dose versus low-dose cisplatin in combination with cyclophosphamide and epidoxorubicin in suboptimal ovarian cancer: a randomized study of the Gruppo Oncologico Nord-Ovest. J Clin Oncol 1996;14:351–6. Advanced Ovarian Cancer Trialists Group. Chemotherapy for advanced ovarian cancer. Cochrane Database of Systematic Reviews. 4th ed. Chichester (UK): John Wiley & Sons, Ltd; 2002. Cyclophosphamide plus cisplatin versus cyclophosphamide, doxorubicin, and cisplatin chemotherapy of ovarian carcinoma: a metaanalysis. The Ovarian Cancer Meta-Analysis Project. J Clin Oncol 1991;9:1668–74. McGuire WP, Hoskins WJ, Brady MF, et al. Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med 1996; 334:1–6. Piccart MJ, Bertelsen K, James K, et al. Randomized intergroup trial of cisplatin-paclitaxel versus cisplatin-cyclophosphamide in women with advanced epithelial ovarian cancer: three-year results. J Natl Cancer Inst 2000;92:699–708. Ozols RF, Bundy BN, Fowler J, et al. Randomized phase III study of cisplatin/paclitaxel versus carboplatin/paclitaxel in optimal stage III epithelial ovarian cancer: a Gynecologic Oncology Group Trial (GOG 158). Proc Am Soc Clin Oncol 1999; 18:356a. Sandercock J, Parmar MK, Torri V, et al. First-line treatment for advanced ovarian cancer: paclitaxel, platinum and the evidence. Br J Cancer 2002; 87:815–24. Neijt JP, Engelholm SA, Tuxen MK, et al. Exploratory phase III study of paclitaxel and cisplatin versus paclitaxel and carboplatin in advanced ovarian cancer. J Clin Oncol 2000;18:3084–92. International Collaborative Ovarian Neoplasm Group. Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide, doxorubicin, and cisplatin in women with ovarian cancer: the ICON3 randomised trial. Lancet 2002;360:505–15. Alberts DS, Liu PY, Hannigan EV, et al. Intraperitoneal cisplatin plus intravenous cyclophosphamide versus intravenous cisplatin plus intravenous cyclophosphamide for stage III ovarian cancer. N Engl J Med 1996;335:1950–5. Markman M, Bundy BN, Alberts DS, et al. Phase III trial of standard-dose intravenous cisplatin plus

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

paclitaxel versus moderately high-dose carboplatin followed by intravenous paclitaxel and intraperitoneal cisplatin in small-volume stage III ovarian carcinoma: an intergroup study of the Gynecologic Oncology Group, Southwestern Oncology Group, and Eastern Cooperative Oncology Group. J Clin Oncol 2001;19:1001–7. Keys HM, Bundy BN, Stehman FB, et al. Cisplatin, radiation, and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy for bulky stage IB cervical carcinoma. N Engl J Med 1999;340:1154–61. Rose PG, Bundy BN, Watkins EB, et al. Concurrent cisplatin-based radiotherapy and chemotherapy for locally advanced cervical cancer. N Engl J Med 1999;340:1144–53. Morris M, Eifel PJ, Lu J, et al. Pelvic radiation with concurrent chemotherapy compared with pelvic and para-aortic radiation for high-risk cervical cancer. N Engl J Med 1999;340:1137–43. Jeremic B, Shibamoto Y, Milicic B, et al. Hyperfractionated radiation therapy with or without concurrent low-dose daily cisplatin in locally advanced squamous cell carcinoma of the head and neck: a prospective randomized trial. J Clin Oncol 2000;18:1458–64. Adelstein DJ, Li Y, Adams GL, et al. An intergroup phase III comparison of standard radiation therapy and two schedules of concurrent chemoradiotherapy in patients with unresectable squamous cell head and neck cancer. J Clin Oncol 2003;21:92–8. Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer. The Department of Veterans Affairs Laryngeal Cancer Study Group. N Engl J Med 1991;324:1685–90. Pignon JP, Bourhis J, Domenge C, et al. Chemotherapy added to locoregional treatment for head and neck squamous-cell carcinoma: three metaanalyses of updated individual data. MACH-NC Collaborative Group. Meta-Analysis of Chemotherapy on Head and Neck Cancer. Lancet 2000; 355:949–55. Soloway MS, Einstein A, Corder MP, et al. A comparison of cisplatin and the combination of cisplatin and cyclophosphamide in advanced urothelial cancer. A National Bladder Cancer Collaborative Group A Study. Cancer 1983;52:767–72. Khandekar JD, Elson PJ, DeWys WD, et al. Comparative activity and toxicity of cis-diamminedichloroplatinum (DDP) and a combination of doxorubicin, cyclophosphamide, and DDP in disseminated transitional cell carcinomas of the urinary tract. J Clin Oncol 1985;3:539–45. Troner M, Birch R, Omura GA, et al. Phase III comparison of cisplatin alone versus cisplatin,


49.

50. 51.

52. 53.

doxorubicin and cyclophosphamide in the treatment of bladder (urothelial) cancer: a Southeastern Cancer Study Group trial. J Urol 1987; 137:660–2. Loehrer PJ Sr, Einhorn LH, Elson PJ, et al. A randomized comparison of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: a cooperative group study. J Clin Oncol 1992;10:1066–73. Johnson DH. New drugs in the management of small cell lung cancer. Lung Cancer 1989;5:221–3. Ettinger DS, Finkelstein DM, Ritch P. Randomized trial of single agents versus combination chemotherapy in extensive stage small cell lung cancer. Proc Am Soc Clin Oncol 1992;11:295. Ihde DC. Chemotherapy of lung cancer. N Engl J Med 1992;327:1434–41. Ihde D, Pass H, Glatstein EJ. Small cell lung cancer. In: DeVita VT, Hellman S, Rosenberg SA, editors. Cancer principles and practice of oncology. 4th ed. Philadelphia (PA): JB Lippincott Company; 1993. p. 723–58.

59

54. Johnson DH, Kim K, Turrisi AT, et al. Cisplatin and etoposide and concurrent thoracic radiotherapy administered once versus twice daily for limited stage small cell lung cancer: preliminary results of an intergroup trial. Proc Am Soc Clin Oncol 1994;12:333. 55. The Non-Small Cell Lung Cancer Collaborative Group. Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised controlled trials. BMJ 1995;311:899–909. 56. Schiller JH, Harrington A, Sandler A, et al. A randomized phase III trial of four chemotherapy regimens in advanced non-small cell lung cancer. Proc Am Soc Clin Oncol 2000;19:1a. 57. Rodriguez MA, Cabanillas FC, Velasquez W, et al. Results of a salvage treatment program for relapsing lymphoma: MINE consolidated with ESHAP. J Clin Oncol 1995;13:1734–41. 58. Velasquez WS, Cabanillas F, Salvador P, et al. Effective salvage therapy for lymphoma with cisplatin in combination with high-dose Ara-C and dexamethasone (DHAP). Blood 1988;71:117–22.

CHAPTER 6

Ototoxicity of Platinum Compounds Michael Anne Gratton, PhD, and Brendan J. Smyth, PhD, MD

Cisplatin may be likened to a medical pearl: following a serendipitous discovery, its simple inorganic structure fascinates, its fundamental mechanism of action captivates, and its marvelous chemotherapeutics across tumor types equates it to one of the most powerful anti-neoplastic agents ever introduced. Yet its toxicity continues to elude and evade, with the result that carboplatin, third-generation oxaliplatin, and now fourth-generation platinum agents are in the process of proving their chemotherapeutic “salt.” The impetus to review the ototoxicity of platinum compounds stems from the issue of quality of life for cancer survivors. The platinum compounds are widely used in gynecologic, testicular, lung, central nervous system, and head and neck cancers. They are non–cell-cycle-specific agents that insert into the deoxyribonucleic acid (DNA) helix, disrupting replication. The dose and efficacy of platinum chemotherapy is limited largely by adverse effects. Cisplatin is the most ototoxic of the platinum compounds even with the inclusion of hypertonic saline, prehydration, or mannitol diuresis in chemotherapeutic regimens.

HISTORY OF CHEMOTHERAPEUTIC DISCOVERY The chemotherapeutic potential of platinum complexes was first recognized by Rosenberg and Cavalieri in 1964 while they were investigating the effects of electric fields on Escherichia coli growth. 1 Clinical trials in 1971 demonstrated cisplatin’s potent antineoplastic activity.2 Cisplatin ototoxicity initially appeared inconsequential. Kovach and colleagues described the ototoxicity of cisplatin as resulting in a mild threshold elevation above 4 kHz in eight patients, whereas DeConti and colleagues reported only two cases of tinnitus persisting for several hours to 1 week.3,4 Cisplatin’s cumulative ototoxic potential was repeatedly obscured by inconsistencies in case reports.5 Elucidation of cisplatin’s ototoxic pathogenesis was delayed when promising chemotherapy phase I clinical trials were abandoned secondary to cisplatin nephrotoxicity. Following the

Table 6-1 Clinical Presentation of Cisplatin Ototoxicity

Tinnitus High-frequency sensorineural hearing loss Progression toward lower frequencies Bilateral and usually symmetrical hearing loss Permanent and usually irreversible hearing loss Total cumulative dose generally > 200 mg

discovery that forced saline diuresis is renoprotective, cisplatin was finally approved for use in human malignancies in 1978.6 Unfortunately, renoprotective protocols do not reduce the ototoxic potential of cisplatin.7 By the 1980s, cisplatin ototoxicity was characterized and risk factors established (Tables 6-1 and 6-2). During the past two decades, cisplatin became the foundation for the development of curative regimens against solid epithelial tumors. The drug remains unrivaled in efficacy against germ cell, ovarian, endometrial, cervical, urothelial, head and neck, lung, and brain cancers (see Chapter 5, “Clinical Uses of Cisplatin”).8 Unfortunately, cisplatin is systemically toxic and predictably ototoxic.7,9 It is the standard for the 5/5 rating of the Hesketh Classification of Potential Scale.10 These factors, plus a narrow tumor indication and the occurrence of tumor resistance, continue to spur development of newer generations of platinumbased agents.11,12 Table 6-2 Cisplatin Ototoxicity Predisposing Factors

Dose, duration, and mode of administration Age extremes Previous or concurrent cranial irradiation Previous history of hearing loss Renal disease Concomitant use of other ototoxic agents Volume status

Ototoxicity of Platinum Compounds

To date, more than 1,000 cisplatin analogues have been synthesized. More than 20 have entered clinical trials.12 Rosenberg’s team was first with the synthesis of carboplatin. This less toxic, second-generation platinum compound gained approval of the US Food and Drug Administration (FDA) in 1989. 13,14 Thirdgeneration oxaliplatin is currently in active clinical trials.15 Although third- and fourth-generation agents have the purported benefit of decreased ototoxic potential, their clinical utility and long-term postexposure toxic sequelae remain to be established. Thus, cisplatin is expected to remain in the chemotherapeutic armamentarium indefinitely, prompting development of effective strategies to protect or rescue the cancer patient from its ototoxicity.16–19

CISPLATIN Chemistry The platinum compound standard is cisplatin, a squareplanar inorganic platinum (Pt) molecule consisting of a divalent Pt(II) central atom accommodating ligands of cis positioned pairs of chlorine atoms or amine groups (Table 6-3).12 Cisplatin’s cis geometry is crucial for cytotoxic activity and optimal Pt(II) binding to sulfur-donating cytosolic “platinophiles” (such as glutathione and methionine) or DNA’s purine N7 atom.20 Cisplatin’s systemic toxicity may be related to the ease with which Pt-N or Pt-S reactions occur following nonenzymatic displacement of the chlorine atoms by hydrolysis.21,22 In contrast, the reduced systemic toxicity of the new Pt(IV) agents is attributed to ligand inertness to deactivation reactions.12 In addition, many Pt(IV) agents being developed are Pt(II) prodrugs in which chemical substitutions in the cisplatin amino groups result in altered lipophilia and tissue distribution.23,24 Mechanism of Action The initial intracellular action of cisplatin is similar to that of the alkylating agents. Both agents preferentially form covalent bonds with N7 of the purine guanine by hydrolytic displacement of chlorine atoms. The extremely reactive intrastrand and interstrand crosslinked Pt-DNA adducts formed by the covalent bonds inhibit template function and DNA replication.20 Binding to non-DNA targets follows, with subsequent induction of cell death through apoptosis, necrosis, or a combination of both mechanisms of cytolysis.25 Clinical Pharmacology Following intravenous administration, cisplatin has an initial plasma half-life (t1/2) of 20 to 30 minutes. The variability of the terminal t1/2 of 6 to 47 days is related to the extensive (> 90%) plasma protein binding displayed by cisplatin. Cisplatin undergoes incomplete urinary excretion (only 35–50% after 5 days) via renal tubule

61

secretion and glomerular filtration, with detectability in tissue samples for as long as 4 months post administration. Cisplatin preferentially concentrates in the liver, kidneys, and large and small intestines, with low penetration of the central nervous system.26 Indications and Clinical Use Cisplatin is employed as palliative therapy in established combination treatment regimens for metastatic testicular tumors in patients who have already received surgical, radiotherapeutic, or chemotherapeutic procedures. It is used as a secondary therapy for metastatic ovarian tumors that are refractory to standard chemotherapy. Other indications include advancedstage and refractory bladder carcinomas as well as in head and neck squamous cell carcinoma (see Table 6-3). A complete listing of cisplatin-based chemotherapy regimens, acronyms, and links to their clinical trial results can be found in a concept report on the Web site of the National Cancer Institute (NCI) at . Nonotological Manifestations of Cisplatin Toxicity The primary nonotological toxicities of cisplatin are listed in Table 6-4. Peripheral neuropathy is currently the main dose-limiting toxicity of cisplatin. Permanent neuropathy occurs in 30 to 50% of patients receiving a high cumulative cisplatin dose.27 Although the precise mechanism for the renoprotection is unknown, cisplatin-induced nephrotoxicity is minimized by saline diuresis protocols and volume repletion. 28 Nevertheless, high cumulative doses (> 120 mg/m2) of cisplatin produce renal failure despite diuresis.29 Unrelenting acute and delayed nausea with vomiting remains a patient’s most feared side effect of cisplatin.30 Emesis does not appear to be related to vestibular toxicity. Premedication with 5-hydroxytryptamine-3 (5-HT3) receptor antagonists and corticosteroids is the standard of care for the emetogenicity associated with cisplatin. Cisplatin-induced myelosuppression is doserelated and characterized as anemia, leukopenia, and thrombocytopenia. Severe thrombocytopenia or leukopenia occurs in about 5% of cases.31 Myelosuppression is generally reversible. Cisplatin Ototoxicity The incidence for irreversible ototoxicity varies, depending upon the dose regimen and the criteria used to define ototoxicity.32–35 Patients with cisplatin ototoxicity generally present with high-frequency hearing loss. The principal characteristics of the ototoxicity are summarized in Table 6-1.36,37 Cisplatin ototoxicity is considered to be exclusively confined to the cochlea. Risk factors (see Table 6-2) for ototoxicity include age

OCO

Pt

O

O

C

C

Pt

O

O

Oxaliplatin

NH2

NH2

Carboplatin

Pt

OCO

Cisplatin

Pt

O

O

CH3

Pt

CI

CI

O

Pt

O

O

CI

CI

2004 FDA fast track phase II–III clinical trials

2002 FDA fast track phase II–III clinical trials

1995 Phase I–II clinical trials

1996

—

—

Japan

Worldwide

Worldwide

Worldwide

1978

1985

Clinical Status

Year Approved

Hormone-resistant prostate cancer

Hormone-resistant prostate cancer

—

Colorectal cancer

Ovarian cancer

Bladder, testicular, and ovarian cancers

FDA-Approved Indications

Breast, cervical, colon, gastric, ovarian, NSCLC, SCLC

Bladder, breast, cervical, mesothelioma, ovarian

NSCLC, SCLC, bladder, cervical, esophageal, head and neck, ovarian, testicular

Breast, head and neck, mesothelioma, NHL, ovarian, pancreatic, testicular

ALL, AML, bladder, bone marrow ablation, breast, cervical, germ cell, head and neck, lung, malignant glioma, neuroblastoma, osteogenic sarcoma, soft tissue sarcoma, stem cell transplant preparation, testicular, Wilms’ tumor

Astrocytoma, breast, carcinoid, cervical, desmoid, esophageal, gastric head and neck, hepatocellular, malignant glioma, malignant melanoma, neuroblastoma, NSCLC, NHL, osteogenic sarcoma, penile

Cancer Indications (In Clinical Trial)

ALL = acute lymphocytic leukemia; AML = acute myelogenous leukemia; NHL = non-Hodgkin’s lymphoma; NSCLC = non–small cell lung cancer; SCLC = small cell lung cancer.

Satraplatin (JM-216)

NH2

H 3N

O

AMD-473 (ZD0473)

N

H 3N

O

Cl

NH3

Nedaplatin (254-S)

H 3N

H 3N

H 3N

H 3N

Cl

H 3N

Platinum Compound

Table 6-3 Platinum Compounds in Clinical Use or in Active Clinical Trials

62

Systemic Toxicity


63

Table 6-4 Toxicity Profiles of Platinum Compounds Toxicity

Cisplatin

Carboplatin

Myelosuppression

Oxaliplatin

XX

Nephrotoxicity

XX

Neurotoxicity

X

Ototoxicity

X

X

Nausea/vomiting

X

X

Nedaplatin

AMD-473

Satraplatin

XX

XX

XX

X

X

X XX X X

X

X = significant; XX = main dose-limiting toxicity.

extremes, renal insufficiency, intravenous bolus administration or high cumulative dosage of cisplatin, coadministration with aminoglycoside antibiotics or loop-inhibiting diuretics, and excessive noise exposure with concomitant cisplatin administration.38–41 Cisplatin-based regimens commonly contain other potentially ototoxic chemotherapeutic agents. Examples of these agents include cytarabine, bleomycin, nitrogen mustard, vincristine, and vinorelbine.40,42–47 It is imperative to assess auditory acuity at the onset of therapy and prior to each successive treatment given the variability of occurrence and individual susceptibility to cisplatin ototoxicity. Epidemiology of Cisplatin Ototoxicity Cisplatin has the highest ototoxic potential of all platinum compounds.48 Differences in chemotherapy regimens, in the definition of ototoxicity, and in methods used to assess auditory ability contribute to the variance in the epidemiology of cisplatin. Thus, Moroso and Blair, in a review of early clinical studies, found a 9 to 91% range in the incidence of ototoxicity, although more recent studies reported a 50 to 60% incidence of ototoxicity.7,34,49–53 Approximately 50% of head and neck cancer patients treated with cisplatin develop ototoxicity.41 Cisplatin ototoxicity is related to dose. In a large retrospective study covering the period of 1990 to 2001, de Jongh and colleagues found that 42% of 400 patients receiving high-dose cisplatin (70–85 mg/m2; median cumulative dose of 420 mg) incurred symptomatic ototoxicity (Common Toxicity Criteria [CTC] grade 3–4; see Table 6-5).39 In contrast, cisplatin ototoxicity was incurred by only 20% of patients receiving a low-dose

cisplatin therapy but 75 to 100% of patients receiving a very high-dose regimen.35,40,54 Although these studies seem to link cisplatin ototoxicity to dosage, the firstorder predictor of cisplatin ototoxicity is cumulative dose in both clinical and animal studies.55 Bokemeyer and colleagues, as well as others, have found that the incidence of ototoxicity increased dramatically when the total cumulative dose exceeds 400 mg, although McKeague states that the critical total dosage is 600 mg.40,56–60 The method used to determine the presence of ototoxicity influences epidemiological reports. The most commonly employed measure of auditory status is conventional audiometry (0.5–8 kHz). The use of high-frequency audiometry (9–20 kHz), however, is the optimal method for early detection of cisplatin sensitivity (see Chapter 18, “Audiologic Monitoring for Ototoxicity”).34,56,61–64 In a comparative study, increased thresholds were noted with high-frequency audiometry when the cumulative dose approximated 200 mg. However, with use of conventional audiometry, hearing loss was not detected until the cumulative dose approximated 400 mg.56 The greater cumulative dose associated with hearing loss in the frequency range of conventional audiometry reflects the progressive nature of cisplatin ototoxicity. A five-frequency sequence (8, 9, 10, 11.2, and 12.5 kHz) was recently proposed by Fausti and colleagues for rapid and sensitive detection of early-onset ototoxicity.64 The auditory brainstem response (ABR) is often employed to assess hearing status and determine the presence of cisplatin ototoxicity in pediatric cancer patients.65–67 Significant increases in wave V latency for click stimuli were found to precede threshold elevations

Table 6-5 Common Toxicity Criteria for Hearing Adverse Event

Inner ear/hearing

0

1

2

3

4

Normal

Hearing loss on audiometry only

Tinnitus or hearing loss, not requiring hearing aid or treatment

Tinnitus or hearing loss, correctable with hearing aid or treatment

Severe unilateral or bilateral hearing loss (deafness), not correctable

Adapted from Cancer Therapy Evaluation Program, Common Toxicity Criteria, Version 2 72 DCTD, NCI, NIH, DHHS, March 1998.

64

Systemic Toxicity

for conventional audiometry by several cycles of chemotherapy.67 Use of the ABR technique with high-frequency specific stimuli (ie, tone bursts) has been proposed as a screening device for ototoxicity.65 Evoked otoacoustic emissions (OAEs) can also be used as a screening tool for cisplatin ototoxicity, although controversy exists regarding whether transient OAEs or distortion-product OAEs are more sensitive in detection of cisplatin ototoxicity.68,69 Reduced amplitude of the distortion-product OAEs has been noted to precede the threshold changes noted by conventional audiometry.70,71 The final methods of evaluating cisplatin ototoxicity are the CTC (Table 6-5) and World Health Organization Grades of Hearing Impairment.72,73 These criteria can be used for subjective rather than audiometric assessment of auditory status.39,72,73 However, subjective assessment underestimates the incidence of ototoxicity since a slight increase in threshold or hearing loss confined to high frequencies will not be detected. Characteristics of Cisplatin Ototoxicity Cisplatin ototoxicity is characterized as a sensorineural hearing loss that is initially detected in the very high frequencies. The hearing loss is bilateral and usually symmetric. The high-frequency (> 2 kHz) character of the hearing loss causes difficulty in speech discrimination, especially in the presence of background noise. Higher cumulative dosing results in further increase in the severity of the high-frequency hearing loss as well as a progression to lower frequencies. However, a bolus administration of cisplatin to a susceptible patient may result in rapid-onset severe hearing loss. Only one case of complete recovery and several cases of partial recovery have been reported following cessation of cisplatin chemotherapy.34,51 Most investigators have reported virtually no recovery in the ototoxic hearing loss following cessation of cisplatin treatment. Thus, cisplatin ototoxicity is regarded as permanent. A larger threshold shift may exist for a few days immediately following the cisplatin administration compared with the hearing threshold 1 week after drug administration (Gratton, unpublished data, 1988). Pathophysiology of Cisplatin Ototoxicity The pathogenesis of cisplatin ototoxicity is multifactorial. Animal research indicates that reduced auditory acuity is partially mediated by free radical generation and antioxidant inhibition.74,75 The presence of superoxide radicals leads to changes in the stria vascularis, organ of Corti, and spiral ganglion cells. The permanent hearing loss associated with cisplatin is linked to the degeneration of cochlear outer hair cells. However, cisplatin is toxic to the stria vascularis as well as the organ of Corti.

Several animal studies report that following lowdose cisplatin, alterations to the stria vascularis precede changes to the organ of Corti. The strial damage occurs primarily in the marginal cells, with some changes noted in the intermediate cells.76–82 Decreases in cell size, which lead to overall thinning of the stria, occur following low-dose cisplatin.76 Histopathology from single high-dose injections or prolonged treatment of cisplatin show bulging or ruptured marginal cells as well as strial atrophy. Overall, in the same cochlear region, the strial changes are minor compared with the cisplatin-induced damage observed in the organ of Corti. In the organ of Corti, low-dose cisplatin causes damage primarily to the hair cell stereocilia.83,84 Wright and Schaefer, as well as Comis and colleagues, observed stereocilia abnormalities including a rough surface coat, disruption of the stereocilia cross-links, fusion, splaying or drooping of the stereocilia, and formation of giant stereocilia.85,86 Larger doses of cisplatin cause hair cell degeneration through sequelae that involve softening of the cuticular plate, aggregation of lysosomes in the supranuclear portion of the hair cell body, and cytoplasmic extrusion from the hair cell. The loss of outer hair cells follows a pattern similar to that seen with aminoglycoside antibiotics. Initially, outer hair cells in the most basal region of the cochlea degenerate.87 As the ototoxic reaction increases, outer hair cells located more apically degenerate.84 Hair cell destruction is most pronounced in the first row and least in the third row of outer hair cells. Inner hair cells show damage and degeneration only after all three rows of outer hair cells in the same region have degenerated.88,89 High doses of cisplatin cause focal damage to pillar cells and other supporting cells.87,90 The cisplatin-induced pathology noted in animal studies has been verified by examination of human temporal bones.83,85,91,92 In addition to noting the hallmark of outer hair cell degeneration, atrophy of the stria vascularis and degeneration of the fibers of the eighth nerve and spiral ganglion cells were observed. Moreover, Hoistad and colleagues reported that the otopathology from cisplatin could be aggravated by combination therapy. They noted that temporal bones removed from patients who had received cranial irradiation as well as cisplatin therapy included altered cochlear vasculature with serum effusion and fibrosis in addition to damage to the organ of Corti and stria vascularis histopathology detailed above.92 Assessment of cochlear function using the ABR and OAE techniques after cisplatin administration to animals agrees with human data showing variable, but reduced, threshold sensitivity and OAE amplitude. These data, in conjunction with decreased cochlear microphonics, are consistent with outer hair cell pathology.78,87,88,90 Pathology in the stria vascularis is associated with a lowered endocochlear potential. Interestingly,


endolymphatic ionic concentrations are not altered, nor is strial sodium, potassium adenosine triphosphatase (Na+, K+-ATPase) activity decreased.88 Other Manifestations of Cisplatin Ototoxicity Tinnitus as a sequela to cisplatin therapy occurs in as many as 60% of patients.40,93,94 However, tinnitus is not predictive of hearing loss.95 The occurrence of tinnitus can be transient, disappearing after the cessation of chemotherapy. Although cisplatin vestibulotoxicity has been the subject of several studies, no clear and consistent evidence of toxicity is reported.96–100 Vestibulotoxic effects have been found only in those cancer patients whose vestibular system was previously exposed to a damaging agent (eg, aging or aminoglycoside antibiotic therapy) unrelated to the cisplatin chemotherapy.101 Risk Factors There is substantial variability in susceptibility to cisplatin-induced ototoxicity. Factors affecting the severity of the ototoxic reaction to cisplatin are a high cumulative dose, age extremes, preexisting hearing loss, anemia, coadministration of other ototraumatic agents or high-dose vinca alkaloids, and prior cranial irradiation (see Table 6-2).32,37,39–41,86,102–112 Cisplatin-mediated ototoxicity is age dependent.50,51 Studies involving older adults or young children report a greater degree of hearing loss in these age groups. Helson and colleagues report that patients aged 8 to 20 years or older than 46 years of age who received a high cumulative dose of cisplatin (> 400 mg) incurred a more severe hearing loss than that incurred by 21- to 45-year-old patients receiving a high cumulative cisplatin dose.50 In summary, a greater sensitivity to cisplatin ototoxicity exists for both older and pediatric patients once a critical cumulative dose of cisplatin has been reached. However, in the study by Helson and colleagues, the average preexposure threshold at 4 to 8 kHz for patients 47 years of age or older ranged from 30 to 40 dB (American National Standards Institute [ANSI], 1969), whereas that of the younger patients was within the normal range (≤ 25 dB HL). The study concluded that cisplatin ototoxicity increased with age, but it is not possible to differentiate the effect of age from that of a preexisting hearing loss.50 Two clinical studies reported that the degree of the cisplatin ototoxicity in patients with a preexisting hearing loss was the same as that of patients with normal baseline hearing.93 In contrast, Aguilar-Markulis and colleagues reported that preexisting cochlear hearing loss “sensitized” the auditory system, resulting in greater hearing loss from the cisplatin than was incurred by those with normal baseline hearing ability.51 Fleming and colleagues agreed, finding that a “greater than average” baseline threshold

65

resulted in a greater degree of ototoxicity.113 In a study that compared different schedules and rates of cisplatin infusion, Vermorken and colleagues found that patients with baseline thresholds > 15 dB HL developed ototoxicity from a lower cumulative dose of cisplatin than did patients who had normal baseline hearing. However, since the degree of ototoxicity was the same regardless of baseline hearing status, Vermorken and colleagues concluded that the presence of a preexisting hearing loss did not increase susceptibility to cisplatin ototoxicity.7 In summary, the results of studies addressing the issue of preexisting hearing loss are equivocal in their results, requiring further study of this topic. Several clinical studies observed that concurrent or prior cranial irradiation increased susceptibility to the ototoxic potential of cisplatin.37,41,103,112 Brock and colleagues reported that 64% of their pediatric patients displayed severe cisplatin ototoxicity at a cumulative dose of only 200 mg, half that of the critical cumulative dose of cisplatin as a sole agent.73 Cranial irradiation as a risk factor for cisplatin ototoxicity was evaluated in an animal study. 114 The bulla, which received cranial radiation prior to administration of cisplatin, displayed greater ototoxicity than did the control ear exposed solely to cisplatin. Thus, animal research supports the clinical observation that cranial irradiation predisposes the chemotherapy patient to cisplatin ototoxicity. Interaction between Cisplatin and Other Ototraumatic Agents The interaction of cisplatin with other potentially ototoxic drugs, such as the aminoglycoside antibiotics and the loop-inhibiting diuretics, has been explored in animal studies, as has the relationship between noise and cisplatin.86,104–108 In each instance, the degree of ototoxicity from the cisplatin was enhanced by exposure to another ototraumatic agent. Based upon ABR and cochleogram data, Schweitzer and colleagues reported that administration of both cisplatin and kanamycin caused greater ototoxicity than occurred from either of the two control agents.84 The kanamycin was administered at a dosage that produced little or no ototoxicity. The cisplatin alone showed a 50% loss of basal turn outer hair cells. Administration of both agents produced a complete or near-complete loss of outer hair cells in the basal turn of the cochlea. The decrease in hearing status as assessed by ABR correlated with histologic data. However, the combinationtreated group sustained substantial nephrotoxicity that may have contributed to the loss of auditory sensitivity. Nevertheless, an additive interaction was noted in this study between cisplatin and the aminoglycoside antibiotic.86 A second study examining the potentiation of cisplatin ototoxicity by gentamicin varied the time during which cisplatin was administered during the 14-day

66

Systemic Toxicity

course of gentamicin injections in guinea pigs.104 A time-dependency for the potentiation of the ototoxicity was found. Exposure to cisplatin during the beginning of the course of gentamicin administration resulted in greater ototoxicity than did the presence of cisplatin late in the gentamicin treatment or with the administration of each agent alone. The possibility that hearing loss from cisplatin might be augmented by administration of a loopinhibiting diuretic was explored in two studies.107,108 Komune and Snow found that the combination of cisplatin and ethacrynic acid affected the amplitude of the cochlear microphonic and resulted in a greater loss of outer and inner hair cells than did administration of either agent alone.107 Brummett and colleagues measured the change in the stimulus intensity necessary to elicit the cochlear microphonic after administration of cisplatin or a loop-inhibiting diuretic (ethacrynic acid, furosemide, bumetanide, or piretanide).108 The results indicated that administration of cisplatin with a loopinhibiting diuretic caused a decrease in hearing sensitivity greater than that expected from administration of either agent alone. The severity of the ototoxicity was the greatest for the combination of ethacrynic acid and cisplatin. This is reasonable since ethacrynic acid is the most potent of the loop-inhibiting diuretics. However, the augmentation of cisplatin ototoxicity by the administration of a loop-inhibiting diuretic has clinical implications. The technique of forced diuresis to minimize cisplatin nephrotoxicity requires using a non–loop-inhibiting diuretic, such as mannitol. A history of noise exposure has been reported to increase the risk of cisplatin ototoxicity threefold.40 Several animal studies have investigated the potential for noise exposure to interact with cisplatin therapy.105,106 Laurell demonstrated that noise exposure that preceded cisplatin administration resulted in a potentiation of the toxic effects of both ototraumatic agents, whereas noise exposure that followed the cisplatin administration did not result in a potentiation of the cisplatin ototoxicity.105 The threshold of noise intensity was investigated by Gratton and colleagues in a study that concomitantly exposed cisplatin-treated chinchillas to noise. A synergistic interaction was found between low-dose cisplatin and noise of 85 dB SPL.106

CARBOPLATIN Chemistry Carboplatin is a second-generation analog of cisplatin consisting of a central platinum atom in the same plane as the two ammonia groups and chloride or 1,1-cyclobutanedicarboxylate ligands in the cis position. The ligand-leaving groups are present in a ring structure versus the two chloride arms of cisplatin. This difference confers stability to the carboplatin molecule with

a resulting decrease in nephrotoxicity, neurotoxicity, ototoxicity, and emetogenicity.27 Mechanism of Action The exact mechanism of carboplatin cytotoxicity is not known. Carboplatin, like cisplatin, induces platinum–DNA adducts, although requiring a 10-fold higher drug concentration and a 7.5-fold longer incubation time.115 Carboplatin becomes hydrolyzed, similar to cisplatin, but at a hundredfold slower rate.116 Carboplatin is a non–cell-cycle-specific agent as well as a radiation sensitizer.117,118 Clinical Pharmacology Following intravenous administration of carboplatin, the initial t1/2 = 24 hours. Carboplatin binds plasma proteins with greater than > 85% efficiency. A wide tissue distribution has been noted including kidney, liver, skin, and tumor tissue in addition to erythrocytes. As much as 70% of carboplatin is excreted unchanged in the urine within 24 hours of administration. Carboplatin is not secreted by the proximal renal tubules, thus permitting calculation of total glomerular filtration rate by creatinine clearance.119 Indications and Clinical Use Carboplatin is used primarily in combination with other chemotherapy agents (see Table 6-3). Carboplatin constitutes the first successful application of pharmacokinetic-directed therapy in oncology. The carboplatin–etoposide combination, with or without cyclophosphamide or ifosfamide, has become the backbone of high-dose chemotherapy for germ cell tumors with stem cell support.120 In addition, carboplatinbased high-dose chemotherapy combinations have been extensively employed for treatment of pediatric brain tumors.121,122 Moreover, carboplatin is equally effective as cisplatin for testicular (with bleomycin and vinblastine), bladder, head and neck (with bleomycin and fluorouracil), ovarian (with cyclophosphamide or doxorubicin), and advanced-stage small cell or non–small cell lung cancers (with etoposide). 27,123 Carboplatin-based chemotherapy regimens, acronyms, and links to clinical trial results can be found in a concept report on the Web site of the National Cancer Institute at . Nonotological Manifestations of Carboplatin Toxicity The primary nonotological toxicities of carboplatin are delineated in Table 6-4. Prior exposure to carboplatin substantially increases the risk and severity of toxicities such as myelosuppression, emetogenesis, peripheral neuropathy, and ototoxicity.124 Thrombocytopenia is the primary dose-limiting toxicity of carboplatin. The


incidence of severe thrombocytopenia and neutropenia is 25 and 18%, respectively. Myelotoxicity is minimized by use of the Calvert AUC formula, which estimates carboplatin dose from the area under the concentration (AUC) time curve based on creatinine clearance, plus a constant for nonrenal clearance in opposition to a fixed dosing based on the body surface area.124–126 Myelosuppression less commonly presents as leukopenia or anemia. Risk factors for myelotoxicity include advanced age, impaired renal function, and concurrent myelosuppressive therapy.124 Anemia is more common with increased carboplatin exposure and has been linked to ototoxicity.40 Blood transfusions with growth factors may be needed during prolonged (> 6 cycles) carboplatin therapy. Peripheral neuropathy occurs in 3% of patients.124 Nephrotoxicity occurs after high-dose carboplatin.119 Carboplatin, although less emetogenic than cisplatin, still requires aggressive antiemetic therapy. Carboplatin Ototoxicity Carboplatin ototoxicity following conventional dose regimens is generally reported in only 1% of patients. However, the incidence following high-dose or combined carboplatin chemotherapy is approximately 33%.127–129 A direct relationship between AUC and auditory toxicity exists with high-dose chemotherapy. 130 Parsons and colleagues reported significant hearing loss in 82% of children with advanced-stage neuroblastoma who received high-dose carboplatin as part of their conditioning regimen for autologous bone marrow transplant.131 Wandt and colleagues reported a dose-limiting CTC grade 2 to 3 ototoxicity (Table 6-5) during a phase I/II study in advanced ovarian cancer, which involved sequential cycles of high-dose chemotherapy with dose escalation of carboplatin with or without paclitaxel supported by granulocyte colonystimulating factor (G-CSF) mobilized peripheral blood progenitor cells.132 In this study, the maximum cumulative tolerated dose of sequential carboplatin that avoided ototoxicity was 1,600 mg. Pathophysiology of Carboplatin Ototoxicity The pathogenesis of carboplatin ototoxicity has not been elucidated. Like cisplatin, carboplatin is toxic to the organ of Corti. Surprisingly however, the preferred target of carboplatin is the inner hair cells rather the outer hair cells that are targeted by cisplatin. Additional damage to outer hair cells has been observed only at higher doses of carboplatin. The preference of carboplatin ototoxicity for the inner hair cells has proven an advantage to research investigating the specific role in audition of inner hair cells versus outer hair cells.

67

NEDAPLATIN Chemistry Nedaplatin (254-S, cis-diammineglycolatoplatinum) is a second-generation cisplatin analogue. The central Pt(II) atom accommodates two cis positioned ammonia groups and a glycolate ring. Mechanism of Action The exact mechanism(s) of action of nedaplatin is unknown. Nedaplatin induces the same platinum–DNA adducts as cisplatin and is cell cycle phase nonspecific. The fact that it displays the same cross-resistance to cisplatin tumor types as carboplatin suggests that these drugs are affected by the same resistance mechanisms. Clinical Pharmacology The dose range for nedaplatin is 60 to 100 mg/m2.133 Similar to other second-generation platinum compounds, nedaplatin can be given without hydration. Indications and Clinical Use Nedaplatin, which is similar to carboplatin in many ways, has been approved for clinical use in Japan since 1995. Nedaplatin has shown promising response rates in phase II trials for treatment of squamous cell carcinoma of the head and neck, lung, esophagus, and uterine cervix (see Table 6-3).134–137 A recently published phase I study combined nedaplatin with paclitaxel to successfully treat unresectable lung, thymic, and head and neck squamous cell carcinoma.133 In Japan, nedaplatin has also been approved for the treatment of ovarian cancer.138 Nonotological Nedaplatin Toxicity Similar to carboplatin, the dose-limiting toxicity of nedaplatin is myelosuppression in the form of thrombocytopenia (see Table 6-4).138 Nedaplatin Ototoxicity Horiuchi and colleagues evaluated the use of nedaplatin in phase II studies. 139 In addition to nedaplatin as a single agent, nedaplatin plus vindesine and cisplatin plus vindesine were included in the study. The incidence of hearing loss was 25.8% (16/62) for the nedaplatin group, 17.6% for the nedaplatin–vindesine group, and 20.0% for the cisplatin–vindesine group. Each combination group consisted of fewer than 20 patients. These results indicate that the ototoxicity of nedaplatin occurs more frequently than does carboplatin ototoxicity but is less prevalent than with cisplatin. Conversely, Nishida and colleagues reported that five or six patients incurred a “slight” hearing loss.140

68

Systemic Toxicity

OXALIPLATIN Chemistry Oxaliplatin is a third-generation analogue of cisplatin in which the amine and chlorine groups are replaced by a 1,2-diaminocyclohexane (DACH) carrier group and an oxalate-leaving group, respectively (see Table 63). The DACH group provides unique antitumor activity, whereas the oxalate group improves water solubility and greatly accelerates DNA adduct formation to 50 times that of cisplatin. The DACH group also affects cellular uptake beyond that of cisplatin.141 Aroplatin, a novel liposomal formulation of oxaliplatin, is currently under clinical trial.142 Mechanism of Action The cellular and molecular details of the mechanisms of action for oxaliplatin have yet to be elucidated. Similar to the other platinum compounds, oxaliplatin undergoes nonenzymatic hydrolysis in physiologic solutions to several active derivatives via displacement of its labile oxalate ligand. Oxaliplatin inhibits DNA replication and transcription through the formation of N7 intra- and interstrand DNA adducts, which are bulkier and more hydrophobic than are those of cisplatin or carboplatin. These properties contribute to the enhanced activity in DNA inhibition and synthesis as well as the lack of cross-resistance displayed by oxaliplatin. Like other platinum compounds, oxaliplatin is a non–cell-cycle-specific agent. Synergistic activity has been demonstrated between oxaliplatin and 5-fluorouracil (5-FU) and leucovorin (LV), all components of the FOLFOX4 regimen, in both in vitro and in vivo tumor models.141 Clinical Pharmacology Following a single 2-hour intravenous infusion of 85 mg/m2 of oxaliplatin, platinum levels are triphasic: initial t1/2 = 0.43 hours, second t1/2 =16.8 hours, and terminal t1/2 = 391 hours. Oxaliplatin displays an irreversible plasma protein binding greater than 90%. It is rapidly incorporated into a wide distribution of tissues. Renal clearance for oxaliplatin accounts for approximately 50% of the total plasma clearance. However, because tissue distribution is a major factor in oxaliplatin clearance, renal clearance alone is not a useful predictor of platinum exposure and toxicity after oxaliplatin administration.141 Indications and Clinical Use Oxaliplatin in combination with 5-FU and LV was approved as first-line treatment for metastatic colorectal cancer in Europe (1999) and recently in the United States by the FDA (January 2004). Clinical trials are ongoing for relapsed and refractory non-Hodgkin’s lymphoma (eg, DHAOx regimen, consisting of dexametha-

sone, high-dose cytarabine, and oxaliplatin).143 Other earlier clinical trials investigated oxaliplatin in the treatment of ovarian, gastric, pancreatic, head and neck, non–small cell lung, and breast cancers (see Table 6-3).144 Oxaliplatin clinical trial updates are available at . Oxaliplatin Toxicity Unlike cisplatin or carboplatin, oxaliplatin is not associated with significant nephrotoxicity or ototoxicity, whereas hematologic toxicity is usually mild.141 The main dose-limiting toxicity of oxaliplatin is an unusual peripheral sensory neuropathy that can occur in 85 to 95% of patients and is often aggravated by exposure to cold temperatures (see Table 6-4).141 It is this side effect, which has two forms—acute and chronic—that is of interest to the otolaryngologist. The toxicity can present as an acute dysesthesia affecting the extremities either during or shortly after the infusion and more rarely as a pharyngolaryngeal dysesthesia, which resolves in hours.147 The chronic form of peripheral neuropathy usually resolves a few months following cessation of treatment.141

NEW PLATINUM COMPOUNDS IN CLINICAL TRIAL The latest platinum compounds are designed for outpatient oral administration. They display efficient solid-tumor cytotoxicity, as well as reduced systemic toxicity and tumor resistance. AMD473 (M473 or ZD473) is a sterically hindered Pt(II) compound, which overcomes cisplatin resistance by forming different DNA adducts than previous platinum compounds.148 AMD473 is in active clinical trials for hormone-refractory prostate cancer (HRPC), ovarian cancer, small and non–small cell lung cancers, and mesothelioma. AMD473 has a toxicity profile similar to carboplatin for nephrotoxicity, neurotoxicity, and ototoxicity.149 Satraplatin (JM-216) is a Pt(IV) complex that is metabolized to active Pt(II) compounds. Satraplatin may hold promise against prostate, ovarian, and non–small cell lung cancers.150 It is currently undergoing phase III clinical trials to evaluate its activity as second-line treatment of docetaxel-refractory HRPC. Satraplatin is reported to lack cisplatin cross-resistance and has reduced nephrotoxicity and neurotoxicity.150 Ototoxicity has not yet been reported.

TUMOR RESISTANCE The leading cause of chemotherapeutic failure is the development of tumor multidrug resistance. No clearcut dominant mechanism of resistance to cisplatin has been identified. Nucleotide excision repair appears to be the most important pathway for cisplatin–DNA


damage. The critical gene appears to be excision repair cross-complementing 1 (ERCC1). High levels of the ERCC1-relative messenger ribonucleic acid (mRNA) are associated with response and survival after cisplatin treatment. 151 Another possible cisplatin resistance mechanism is a genetic abnormality affecting TP53, a gene associated with apoptosis. Mutations in TP53 are found in 60% of non–small cell lung cancer patients.151 Overall, studies have revealed that a combination of reduced platinum transport, increased cellular tolerance to Pt–DNA adducts, increased cytoplasmic detoxification via elevated glutathione or metallothionein levels, and enhanced DNA repair may underlie resistance to platinum compounds.152 In the development of new platinum compounds, alterations in rate of hydrolysis, transport to and within the cell, and DNA binding are being used to deter tumor resistance.12,153

OTOPROTECTION The future success of platinum chemotherapeutic regimens depends not only on their clinical efficacy but also on strategies that directly improve quality of life by reducing ototoxic risk. The principal cause for hair cell death implicates signaling in the apoptotic pathway via (1) reactive oxygen species and free radicals formed as by-products that elicit damage by reacting with cellular proteins, (2) caspase activation, and (3) calpain activation.154,155 Several laboratories have been investigating not only the mechanisms of ototoxicity but also the pharmacodynamics of potential chemoprotective agents in animal models. Strategies for prevention of free radical formation have included the temporal or anatomic separation of the platinum compounds with agents such as vitamin E (α-tocopherol), sodium thiosulfate, D-methionine, or N-acetylcysteine (see Chapter 20, “Ototoxic Damage to Hearing: Otoprotective Strategies”).8,17,74,156–159 Superoxide radicals can cause cell death via ironand calcium-dependent pathways.160 The combination of a calcium chelator (Quin 2-AM) and the iron chelator (2,2-DPD) provided protection from hair cell degeneration in cochlear explants, indicating the presence of both iron- and calcium-dependent components in cisplatin ototoxicity. These findings suggest that use of iron chelators may prevent cisplatin ototoxicity.75,160 The protection from platinum ototoxicity provided by salicylates may involve the scavenging of platinumgenerated reactive oxygen species (ROS), inhibition of NFκB translocation to the nucleus, or inhibition of tumor necrosis factor (TNF-α) via IκB stabilization. Regardless, the final action of the salicylates is to inhibit cisplatin activation of apoptotic pathways.161,162 Manipulation of the signal transduction pathways to increase survival and decrease apoptosis pathways may ameliorate cisplatin ototoxicity. For example, inorganic thiophosphates, such as WR2721 (amifostine),

69

may protect against nephrotoxicity and ototoxicity.129,163 This mechanism of protection may involve enhanced DNA repair and synthesis or the release of free thiols after dephosphorylation with alkaline phosphatase acting as free radical scavengers and metalbinding centers.164,165 Sodium thiosulfate is a well-studied cytoprotective agent. Use of this agent in a clinical trial of patients with brain tumors has been found to decrease the degree of ototoxicity when there is a delay of 4 hours between administration of carboplatin and the sodium thiosulfate.171 Anatomic separation with this agent using a two-compartment model has also been shown to have some success in protecting against ototoxicity.8 This model uses the relative impermeability of the blood-brain barrier to allow only the intra-arterially administered chemotherapeutic agent access to the tumor while the intravenously administered protective agent acts on other parts of the body. There is a need for additional clinical trials to evaluate the efficacy of the chemotherapy in relation to dose and timing of sodium thiosulfate. The other chemoprotective agents mentioned above have been evaluated only in animal or in vitro tumor models. Their effectiveness in reducing platinum ototoxicity while maintaining the needed antitumor effects of the platinum compounds has yet to be evaluated in clinical trials.

SUMMARY • Platinum based chemotherapy is used widely for the treatment of gynecologic, testicular, lung, central nervous system, and head and neck cancers. Platinum compounds are non–cell-cyclespecific agents that inhibit DNA replication. Binding to non-DNA targets induces cell death through mechanisms of apoptosis, necrosis, or a combination of both mechanisms. Cisplatin is the still the most widely used and ototoxic platinum based agent. Non-otologic toxicities of cisplatin include peripheral neuropathy (in 30–50%), nephrotoxicity (which can be largely minimized by saline diuresis protocols and volume repletion) and myelosuppression (dose dependent and severe in approximately 5%). • Cisplatin ototoxicity is related to dose and in particular total cumulative dosage (especially when > 400 mg). It is primarily cochleotoxic causing a permanent, usually symmetric sensorineural hearing loss that begins in the higher frequencies and progresses toward the lower frequencies. Other predisposing factors include age extremes (the very young or old), previous or concurrent cranial irradiation, renal disease, concomitant use of other ototoxic agents (ie aminoglycosides), and a previous history of

70

Systemic Toxicity

hearing loss. Tinnitus may occur in up to 60% of patients receiving cisplatin. • The pathogenesis of cisplatin ototoxicity is multifactorial with changes noted in the stria vascularis, spiral ganglion cells, and the outer hair cells (OHCs) especially. Carboplatin, a second generation analog conversely appears to be preferentially toxic for the inner hair cells (IHCs) and is also a radiation sensitizer. • Inhibition of cisplatin activated cellular apoptosis has formed the basis for current investigative otoprotective strategies. Strategies that prevent free radical formation have included the temporal and anatomic separation of platinum compounds with agents such as inorganic thiophosphates (WR2721), sodium thiosulfate, or iron chelating agents.

REFERENCES 1. Rosenberg BH, Cavalieri LF. Template deoxyribonucleic acid and the control of replication. Nature 1965;206:999–1001. 2. Hill J, Speer R. Organo-platinum complexes as antitumor agents. Anticancer Res 1982;2:173–82. 3. Kovach JS, Moertel CG, Schutt AJ, et al. Phase II study of cis-diamminedichloroplatinum (NSC119875) in advanced carcinoma of the large bowel. Cancer Chemother Rep 1973;57:357–9. 4. DeConti RC, Toftness BR, Lange RC, et al. Clinical and pharmacological studies with cis-diamminedichloroplatinum (II). Cancer Res 1973;33:1310–5. 5. Rossof A, Slayton R, Perlia C. Preliminary clinical experience with cis-diamminedichloroplatinum (II) (NSC 119875, CACP). Cancer 1972;30:1451–6. 6. Rosenberg B. Fundamental studies with cisplatin. Cancer 1984;55:2303–16. 7. Vermorken JB, Kapteijn TS, Hart AA, et al. Ototoxicity of cis-diamminedichloroplatinum (II): influence of dose, schedule and mode of administration. Eur J Cancer Clin Oncol 1983;19:53–8. 8. Blakley BW, Cohen JI, Doolittle ND, et al. Strategies for prevention of toxicity caused by platinumbased chemotherapy: review and summary. Laryngoscope 2002;112:1997–2001. 9. Oldfield EH, Clark WC, Dedrick RL, et al. Reduced systemic drug exposure by combining intraarterial cis-diamminedichloroplatinum(II) with hemodialysis of regional venous drainage. Cancer Res 1987;47:1962–7. 10. Hesketh PJ, Kris MG, Grunberg SM, et al. Proposal for classifying the acute emetogenicity of cancer chemotherapy. J Clin Oncol 1997;15:103–9. 11. Waterhouse DM, Reynolds RK, Natale RB. Combined carboplatin and cisplatin. Limited prospects for dose intensification. Cancer 1993;71:4060–6.

12. Fuertes MA, Castilla J, Alonso C, et al. Novel concepts in the development of platinum antitumor drugs. Curr Med Chem Anti-Cancer Agents 2002; 2:539–51. 13. Egorin MJ, Van Echo DA, Tipping SJ, et al. Pharmacokinetics and dosage reduction of cis-diammine (1,1-cyclobutanedicarboxylato) platinum in patients with impaired renal function. Cancer Res 1984; 44:5432–8. 14. Woloschuk DM, Pruemer JM, Cluxton RJ Jr. Carboplatin: a new cisplatin analog. Drug Intell Clin Pharm 1988;22:843–9. 15. Soulie P, Raymond E, Brienza S, et al. Oxaliplatin: the first DACH platinum in clinical practice. Bull Cancer 1997;84:665–73. 16. Yalcin S, Muftuoglu S, Cetin E, et al. Protection against cisplatin-induced nephrotoxicity by recombinant human erythropoietin. Med Oncol 2003;20:169–74. 17. Wimmer C, Mees K, Stumpf P, et al. Round window application of D -methionine, sodium thiosulfate, brain-derived neurotrophic factor, and fibroblast growth factor-2 in cisplatin-induced ototoxicity. Otol Neurotol 2004;25:33–40. 18. Tanaka F, Whitworth CA, Rybak LP. Round window pH manipulation alters the ototoxicity of systemic cisplatin. Hear Res 2004;187:44–50. 19. Leitao DJ, Blakley BW. Quantification of sodium thiosulphate protection on cisplatin-induced toxicities. J Otolaryngol 2003;32:146–50. 20. Roberts J, Pera M. DNA as a target for anticancer coordination compounds. In: Lippard S, editor. Platinum, gold, and the metal chemotherapeutic agents. Washington (DC): American Chemical Society; 1983. p. 3. 21. Calabresi P, Chabner B. Chemotherapy of neoplastic disease. In: Gilman A, Rall T, Nies A, et al, editors. Goodman and Gilman’s the pharmacological basis of therapeutics. 8th ed. Elmsford (NY): Pergamon Press; 1990. p. 1249–51. 22. Jamieson E, Lippard S. Structure, recognition and processing of cisplatin–DNA adducts. Chem Rev 1999;99:2467–98. 23. Raynaud FI, Mistry P, Donaghue A, et al. Biotransformation of the platinum drug JM216 following oral administration to cancer patients. Cancer Chemother Pharmacol 1996;38:155–62. 24. Reedijk J. New clues for platinum antitumor chemistry: kinetically controlled metal binding to DNA. Proc Natl Acad Sci U S A 2003;100:3611–6. 25. Gonzalez VM, Fuertes MA, Alonso C, et al. Is cisplatin-induced cell death always produced by apoptosis? Mol Pharmacol 2001;59:657–63. 26. Safirstein R, Daye M, Guttenplan J. Mutagenic activity and identification of excreted platinum in human and rat urine and rat plasma after


27.

28.

29. 30.

31.

32.

33.

34.

35.

36. 37.

38.

39.

40.

41.

42.

43.

administration of cisplatin. Cancer Lett 1983; 18:329. Go R, Adjei A. Review of the comparative pharmacology and clinical activity of cisplatin and carboplatin. J Clin Oncol 1999;17:409–22. Santoso JT, Lucci JA III, Coleman RL, et al. Saline, mannitol, and furosemide hydration in acute cisplatin nephrotoxicity: a randomized trial. Cancer Chemother Pharmacol 2003;52:13–8. Arany I, Safirstein R. Cisplatin nephrotoxicity. Semin Nephrol 2003;23:460–4. Perez E. Use of dexamethasone with 5-HT3-receptor antagonists for chemotherapy-induced nausea and vomiting. Cancer J Sci Am 1998;4:72–7. Von Hoff DD, Schilsky R, Reichert CM, et al. Toxic effects of cis-dichlorodiammineplatinum(II) in man. Cancer Treat Rep 1979;63:1527–31. Ilveskoski I, Saarinen UM, Wiklund T, et al. Ototoxicity in children with malignant brain tumors treated with the “8 in 1” chemotherapy protocol. Med Pediatr Oncol 1996;27:26–31. Simon T, Hero B, Dupuis W, et al. The incidence of hearing impairment after successful treatment of neuroblastoma. Klin Padiatr 2002;214:149–52. Fausti SA, Schechter MA, Rappaport BZ, et al. Early detection of cisplatin ototoxicity. Selected case reports. Cancer 1984;53:224–31. Pollera CF, Marolla P, Nardi M, et al. Very highdose cisplatin-induced ototoxicity: a preliminary report on early and long-term effects. Cancer Chemother Pharmacol 1988;21:61–4. Laurell G, Engstrom B, Hirsch A, et al. Ototoxicity of cisplatin. Int J Audiol 1987;10:359–62. Nagy JL, Adelstein DJ, Newman CW, et al. Cisplatin ototoxicity: the importance of baseline audiometry. Am J Clin Oncol 1999;22:305–8. Reddy AT, Witek K. Neurologic complications of chemotherapy for children with cancer. Curr Neurol Neurosci Rep 2003;3:137–42. de Jongh FE, van Veen RN, Veltman SJ, et al. Weekly high-dose cisplatin is a feasible treatment option: analysis on prognostic factors for toxicity in 400 patients. Br J Cancer 2003;88:1199–206. Bokemeyer C, Berger CC, Hartmann JT, et al. Analysis of risk factors for cisplatin-induced ototoxicity in patients with testicular cancer. Br J Cancer 1998;77:1355–62. Blakley BW, Gupta AK, Myers SF, et al. Risk factors for ototoxicity due to cisplatin. Arch Otolaryngol Head Neck Surg 1994;120:541–6. Corden BJ, Strauss LC, Killmond T, et al. Cisplatin, ara-C and etoposide (PAE) in the treatment of recurrent childhood brain tumors. J Neurooncol 1991;11:57–63. Anthoney D, McKean M, Roberts J, et al. Bleomycin, vincristine, cisplatin/bleomycin, etoposide, cisplatin

44.

45.

46.

47.

48. 49. 50.

51.

52.

53.

54.

55.

56.

57.

71

chemotherapy: an alternating, dose intense regimen producing promising results in untreated patients with intermediate or poor prognosis malignant germ-cell tumours. Br J Cancer 2004;90:601–6. Loeffler JS, Kretschmar CS, Sallan SE, et al. Preradiation chemotherapy for infants and poor prognosis children with medulloblastoma. Int J Radiat Oncol Biol Phys 1988;15:177–81. Schweitzer VG. Ototoxicity of chemotherapeutic agents. Otolaryngol Clin North Am 1993;26: 759–89. Moss PE, Hickman S, Harrison BR. Ototoxicity associated with vinblastine. Ann Pharmacother 1999;33:423–5. Gridelli C, Gallo C, Shepherd FA, et al. Gemcitabine plus vinorelbine compared with cisplatin plus vinorelbine or cisplatin plus gemcitabine for advanced non-small-cell lung cancer: a phase III trial of the Italian GEMVIN Investigators and the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2003;21:3025–34. Anniko M, Sobin A. Cisplatin: evaluation of its ototoxic potential. Am J Otolaryngol 1986;7:276–93. Moroso MJ, Blair RL. A review of cis-platinum ototoxicity. J Otolaryngol 1983;12:365–9. Helson L, Okonkwo E, Anton L, Cvitkovic E. cisPlatinum ototoxicity. Clin Toxicol 1978;13: 469–78. Aguilar-Markulis NV, Beckley S, Priore R, et al. Auditory toxicity effects of long-term cis-dichlorodiammineplatinum II therapy in genitourinary cancer patients. J Surg Oncol 1981;16:111–23. Reddel RR, Kefford RF, Grant JM, et al. Ototoxicity in patients receiving cisplatin: importance of dose and method of drug administration. Cancer Treat Rep 1982;66:19–23. Brown RL, Nuss RC, Patterson R, et al. Audiometric monitoring of cis-platinum ototoxicity. Gynecol Oncol 1983;16:254–62. Kopelman J, Budnick AS, Sessions RB, et al. Ototoxicity of high-dose cisplatin by bolus administration in patients with advanced cancers and normal hearing. Laryngoscope 1988;98:858–64. Klis SF, O’Leary SJ, Wijbenga J, et al. Partial recovery of cisplatin-induced hearing loss in the albino guinea pig in relation to cisplatin dose. Hear Res 2002;164:138–46. Park KR. The utility of acoustic reflex thresholds and other conventional audiologic tests for monitoring cisplatin ototoxicity in the pediatric population. Ear Hear 1996;17:107–15. McKeage M. Clinical toxicology of platinum-based cancer chemotherapeutic agents. In: Kelland L, Farrell N, editors. Platinum-based drugs in cancer therapy. Totowa (NJ): Humana Press; 2000. p. 251–75.

72

Systemic Toxicity

58. Schaefer SD, Post JD, Close LG, et al. Ototoxicity of low- and moderate-dose cisplatin. Cancer 1985;56: 1934–9. 59. Perin G, Dallorso S, Stura M, et al. High-dose cisplatin and etoposide in advanced malignancies of childhood. Pediatr Hematol Oncol 1987;4: 329–36. 60. Waters GS, Ahmad M, Katsarkas A, et al. Ototoxicity due to cis-diamminedichloroplatinum in the treatment of ovarian cancer: influence of dosage and schedule of administration. Ear Hear 1991; 12:91–102. 61. Dreschler WA, van der Hulst RJ, Tange RA, et al. The role of high-frequency audiometry in early detection of ototoxicity. Audiology 1985;24:387–95. 62. Dreschler WA, van der Hulst RJ, Tange RA, et al. Role of high-frequency audiometry in the early detection of ototoxicity. II. Clinical aspects. Audiology 1989;28:211–20. 63. Fausti SA, Larson VD, Noffsinger D, et al. Highfrequency audiometric monitoring strategies for early detection of ototoxicity. Ear Hear 1994; 15:232–9. 64. Fausti SA, Henry JA, Helt WJ, et al. An individualized, sensitive frequency range for early detection of ototoxicity. Ear Hear 1999;20:497–505. 65. Coupland SG, Ponton CW, Eggermont JJ, et al. Assessment of cisplatin-induced ototoxicity using derived-band ABRs. Int J Pediatr Otorhinolaryngol 1991;22:237–48. 66. Berg AL, Spitzer JB, Garvin JH, Jr. Ototoxic impact of cisplatin in pediatric oncology patients. Laryngoscope 1999;109:1806–14. 67. De Lauretis A, De Capua B, Barbieri MT, et al. ABR evaluation of ototoxicity in cancer patients receiving cisplatin or carboplatin. Scand Audiol 1999; 28:139–43. 68. Allen GC, Tiu C, Koike K, et al. Transient-evoked otoacoustic emissions in children after cisplatin chemotherapy. Otolaryngol Head Neck Surg 1998;118:584–8. 69. Toral-Martinon R, Shkurovich-Bialik P, ColladoCorona MA, et al. Distortion product otoacoustic emissions test is useful in children undergoing cisplatin treatment. Arch Med Res 2003;34:205–8. 70. Ozturan O, Jerger J, Lew H, et al. Monitoring of cisplatin ototoxicity by distortion-product otoacoustic emissions. Auris Nasus Larynx 1996; 23:147–51. 71. Stavroulaki P, Apostolopoulos N, Segas J, et al. Evoked otoacoustic emissions—an approach for monitoring cisplatin induced ototoxicity in children. Int J Pediatr Otorhinolaryngol 2001;59:47–57. 72. Cancer Therapy Evaluation Program. Common Toxicity Criteria, Version 2: DCTD, NCI, NIH, DHHS; 1998.

73. Brock PR, Bellman SC, Yeomans EC, et al. Cisplatin ototoxicity in children: a practical grading system. Med Pediatr Oncol 1991;19:295–300. 74. Evans P, Halliwell B. Free radicals and hearing. Cause, consequence, and criteria. Ann N Y Acad Sci 1999;884:19–40. 75. Rybak LP, Kelly T. Ototoxicity: bioprotective mechanisms. Curr Opin Otolaryngol Head Neck Surg 2003;11:328–33. 76. Nakai Y, Konishi K, Chang KC, et al. Ototoxicity of the anticancer drug cisplatin. An experimental study. Acta Otolaryngol 1982;93:227–32. 77. Tange RA, Vuzevski VD. Changes in the stria vascularis of the guinea pig due to cis-platinum. Arch Otorhinolaryngol 1984;239:41–7. 78. Kohn S, Fradis M, Podoshin L, et al. Toxic effects of cisplatin alone and in combination with gentamicin in stria vascularis of guinea pigs. Laryngoscope 1991;101:709–16. 79. Meech RP, Campbell KC, Hughes LP, et al. A semiquantitative analysis of the effects of cisplatin on the rat stria vascularis. Hear Res 1998;124:44–59. 80. Campbell KC, Meech RP, Rybak LP, et al. D Methionine protects against cisplatin damage to the stria vascularis. Hear Res 1999;138:13–28. 81. Hamers FP, Wijbenga J, Wolters FL, et al. Cisplatin ototoxicity involves organ of Corti, stria vascularis and spiral ganglion: modulation by alphaMSH and ORG 2766. Audiol Neurootol 2003;8: 305–15. 82. Sluyter S, Klis SF, de Groot JC, et al. Alterations in the stria vascularis in relation to cisplatin ototoxicity and recovery. Hear Res 2003;185:49–56. 83. Strauss M, Towfighi J, Lord S, et al. Cisplatinum ototoxicity: clinical experience and temporal bone histopathology. Laryngoscope 1983;93: 1408–13. 84. Schweitzer VG, Hawkins JE, Lilly DJ, et al. Ototoxic and nephrotoxic effects of combined treatment with cis-diamminedichloroplatinum and kanamycin in the guinea pig. Otolaryngol Head Neck Surg 1984;92:38–49. 85. Wright CG, Schaefer SD. Inner ear histopathology in patients treated with cis-platinum. Laryngoscope 1982;92:1408–13. 86. Comis S, Rhys-Evans P, Osborne M, et al. Early morphological and chemical changes induced by cisplatin in the guinea pig organ of Corti. J Laryngol Otol 1986;100:1375–83. 87. Fleischman RW, Stadnicki SW, Ethier MF, et al. Ototoxicity of cis-dichlorodiammine platinum (II) in the guinea pig. Toxicol Appl Pharmacol 1975;33:320–32. 88. Barron SE, Daigneault EA. Effect of cisplatin on hair cell morphology and lateral wall Na,K-ATPase activity. Hear Res 1987;26:131–7.


89. Marco-Algarra J, Basterra J, Marco J. Cis-diamminedichloroplatinum ototoxicity. An experimental study. Acta Otolaryngol 1985;99:343–7. 90. Estrem SA, Babin RW, Ryu JH, et al. Cis-diamminedichloroplatinum (II) ototoxicity in the guinea pig. Otolaryngol Head Neck Surg 1981; 89:638–45. 91. Hinojosa R, Riggs LC, Strauss M, et al. Temporal bone histopathology of cisplatin ototoxicity. Am J Otol 1995;16:731–40. 92. Hoistad DL, Ondrey FG, Mutlu C, et al. Histopathology of human temporal bone after cisplatinum, radiation, or both. Otolaryngol Head Neck Surg 1998;118:825–32. 93. Piel I, Meyer D, Perlia C, et al. Effects of cisdiamminedichloroplatinum (NSC-119875) on hearing function in man. Cancer Chemother Rep 1974;58:871–5. 94. Fossa SD, de Wit R, Roberts JT, et al. Quality of life in good prognosis patients with metastatic germ cell cancer: a prospective study of the European Organization for Research and Treatment of Cancer Genitourinary Group/Medical Research Council Testicular Cancer Study Group (30941/TE20). J Clin Oncol 2003;21:1107–18. 95. Domenech J, Carulla M, Traserra J. Tinnitus in the diagnosis and prognosis of ototoxicity. Acta Otorrinolaringol Esp 1990;41:7–9. 96. Kitsigianis GA, O’Leary DP, Davis LL. Vestibular autorotation testing of cisplatin chemotherapy patients. Adv Otorhinolaryngol 1988;42:250–3. 97. Kobayashi H, Ohashi N, Watanabe Y, et al. Clinical features of cisplatin vestibulotoxicity and hearing loss. ORL J Otorhinolaryngol Relat Spec 1987;49:67–72. 98. Black FO, Gianna-Poulin C, Pesznecker SC. Recovery from vestibular ototoxicity. Otol Neurotol 2001;22:662–71. 99. Black FO, Myers EN, Schramm VL, et al. Cisplatin vestibular ototoxicity: preliminary report. Laryngoscope 1982;92:1363–8. 100. Schweitzer VG, Rarey KE, Dolan DF, et al. Vestibular morphological analysis of the effects of cisplatin vs. platinum analogs, CBDCA (JM-8) and CHIP (JM-9). Laryngoscope 1986;96:959–74. 101. Schaefer SD, Wright CG, Post JD, et al. Cisplatinum vestibular toxicity. Cancer 1981;47: 857–9. 102. Highley M, Calvert A. Clinical experience with cisplatin and carboplatin. In: Kelland L, Farrell N, editors. Platinum-based drugs in cancer therapy. Totowa (NJ): Humana Press; 2000. p. 171–94. 103. Schell MJ, McHaney VA, Green AA, et al. Hearing loss in children and young adults receiving cisplatin with or without prior cranial irradiation. J Clin Oncol 1989;7:754–60.

73

104. Riggs LC, Brummett RE, Guitjens SK, et al. Ototoxicity resulting from combined administration of cisplatin and gentamicin. Laryngoscope 1996; 106:401–6. 105. Laurell GF. Combined effects of noise and cisplatin: short- and long-term follow-up. Ann Otol Rhinol Laryngol 1992;101:969–76. 106. Gratton MA, Salvi RJ, Kamen BA, et al. Interaction of cisplatin and noise on the peripheral auditory system. Hear Res 1990;50:211–23. 107. Komune S, Snow JB Jr. Potentiating effects of cisplatin and ethacrynic acid in ototoxicity. Arch Otolaryngol 1981;107:594–7. 108. Brummett RE. Ototoxicity resulting from the combined administration of potent diuretics and other agents. Scand Audiol Suppl 1981;14: 215–24. 109. Samson MK, Baker LH, Devos JM, et al. Phase I clinical trial of combined therapy with vinblastine (NSC-49842), bleomycin (NSC-125066), and cisdichlorodiammineplatinum(II)(NSC-119875). Cancer Treat Rep 1976;60:91–7. 110. Miettinen S, Laurikainen E, Johansson R, et al. Radiotherapy enhanced ototoxicity of cisplatin in children. Acta Otolaryngol Suppl 1997;529: 90–4. 111. Feun LG, Stewart DJ, Maor M, et al. A pilot study of cis-diamminedichloroplatinum and radiation therapy in patients with high grade astrocytomas. J Neurooncol 1983;1:109–13. 112. Granowetter L, Rosenstock JG, Packer RJ. Enhanced cis-platinum neurotoxicity in pediatric patients with brain tumors. J Neurooncol 1983; 1:293–7. 113. Fleming S, Peppard S, Ratanatharathorn V, et al. Ototoxicity from cis-platinum in patients with stages III and IV previously untreated squamous cell cancer of the head and neck. Am J Clin Oncol 1985;8:302–6. 114. Baranak CC, Wetmore RF, Packer RJ. Cis-platinum ototoxicity after radiation treatment: an animal model. J Neurooncol 1988;6:261–7. 115. Hongo A, Seki S, Akiyama K, et al. A comparison of in vitro platinum-DNA adduct formation between carboplatin and cisplatin. Int J Biochem 1994;26:1009–16. 116. Knox R, Friedlos F, Lydall D, et al. Mechanism of cytotoxicity of anticancer platinum drugs: evidence that cis-diamminedichloroplatinum (II) and cis-diammine- (1,1-cyclobutanedicarboxylato) platinum (II) differ only in the kinetics of their interaction with DNA. Cancer Res 1986; 46:1972–9. 117. Atagi S, Kawahara M, Ogawara M, et al. Phase II trial of daily low-dose carboplatin and thoracic radiotherapy in elderly patients with locally

74

118.

119.

120.

121.

122.

123.

124.

125.

126. 127. 128.

129.

130.

131.

Systemic Toxicity

advanced non-small cell lung cancer. Jpn J Clin Oncol 2000;30:59–64. Groen H, van der Leest A, de Vries E, et al. Continuous carboplatin infusion during 6 weeks’ radiotherapy in locally inoperable non-small-cell lung cancer: a phase I and pharmacokinetic study. Br J Cancer 1995;72:992–7. Reed E. Cisplatin and analogs. In: Chabner B, Longo D, editors. Cancer chemotherapy and biotherapy: principles and practice. Philadelphia (PA): Lippincott Williams and Wilkins; 2001. p. 447–65. Nieto Y, Vaughan WP. Pharmacokinetics of highdose chemotherapy. Bone Marrow Transplant 2004;33:259–69. Guruangan S, Dunkel I, Goldman S, et al. Myeloablative chemotherapy with autologous bone marrow rescue in young children with recurrent malignant brain tumors. J Clin Oncol 1998; 16:2486–93. Mason W, Grovas A, Halpern S, et al. Intensive chemotherapy and bone marrow rescue for young children with newly diagnosed malignant brain tumors. J Clin Oncol 1998;16:210–21. Hartmann JT, Lipp HP. Toxicity of platinum compounds. Expert Opin Pharmacother 2003; 4:889-901. Bristol-Myers Squibb. Paraplatin-aq product monograph. Montreal (QC): Bristol-Myers Squibb; 1994. Calvert A, Newell D, Gumbrell L, et al. Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J Clin Oncol 1989;7:1748–56. Calvert A. Dose optimisation of carboplatin in adults. Anticancer Res 1994;14:2273–8. Alberts DS. Clinical pharmacology of carboplatin. Semin Oncol 1990;17:6–8. Macdonald MR, Harrison RV, Wake M, et al. Ototoxicity of carboplatin: comparing animal and clinical models at the Hospital for Sick Children. J Otolaryngol 1994;23:151–9. Kanat O, Evrensel T, Baran I, et al. Protective effect of amifostine against toxicity of paclitaxel and carboplatin in non-small cell lung cancer: a single center randomized study. Med Oncol 2003; 20:237–45. Huitema AD, Spaander M, Mathjt RA, et al. Relationship between exposure and toxicity in highdose chemotherapy with cyclophosphamide, thiotepa and carboplatin. Ann Oncol 2002; 13:374–84. Parsons SK, Neault MW, Lehmann LE, et al. Severe ototoxicity following carboplatin-containing conditioning regimen for autologous marrow transplantation for neuroblastoma. Bone Marrow Transplant 1998;22:669–74.

132. Wandt H, Birkmann J, Denzel T, et al. Sequential cycles of high-dose chemotherapy with dose escalation of carboplatin with or without paclitaxel supported by G-CSF mobilized peripheral blood progenitor cells: a phase I/II study in advanced ovarian cancer. Bone Marrow Transplant 1999; 23:763–70. 133. Sekine I, Nokihara H, Horiike A, et al. Phase I study of cisplatin analogue nedaplatin (254-S) and paclitaxel in patients with unresectable squamous cell carcinoma. Br J Cancer 2004;90: 1125–8. 134. Inuyama Y, Miyake H, Horiuchi M, et al. A late phase II clinical study of cis-diammine glycolatoplatinum, 254-S, for head and neck cancers. Gan To Kagaku Ryoho 1992;19:871–7. 135. Yamamoto N, Tamura T, Kurata T. Phase I and pharmacokinetic (PK) study of (glycolate-0, 00)diammine platinum (II) (nedaplatin: 254-S) in elderly patients with non-small cell lung cancer (NSCLC). Proc Am Soc Clin Oncol 2000;19:203a. 136. Taguchi T, Wakui A, Nabeya K, et al. A phase II clinical study of cis-diammine glycolatoplatinum, 254-S, for gastrointestinal cancers. 254-S Gastrointestinal Cancer Study Group. Gan To Kagaku Ryoho 1992;19:483–8. 137. Noda K, Ikeda M, Yakushiji M, et al. A phase II clinical study of cis-diammine glycolatoplatinum, 254-S, for cervical cancer of the uterus. Gan To Kagaku Ryoho 1992;19:885–92. 138. Vermorken JB. The integration of paclitaxel and new platinum compounds in the treatment of advanced ovarian cancer. Int J Gynecol Cancer 2001;11 Suppl 1:21–30. 139. Horiuchi M, Miyake H, Ota K. Ototoxicity of cisdiammine glycolatoplatinum, 254-S. Gan To Kagaku Ryoho 1992;19:1327–32. 140. Nishida M, Satoh Y, Nishide K, et al. Phase I study of a combination chemotherapy of nedaplatin and cisplatin. Gan To Kagaku Ryoho 1999;26:2209–15. 141. Sanofi-Synthelabo. Oxaliplatin product monograph. New York: Sanofi-Synthelabo Inc; 2002. 142. Antigenics. Aroplatin product monograph. Lexington (MA): Antigenics; 2002. 143. Machover D, Delmas-Marsalet B, Misra S, et al. Dexamethasone, high-dose cytarabine, and oxaliplatin (DHAOx) as salvage treatment for patients with initially refractory or relapsed non-Hodgkin’s lymphoma. Ann Oncol 2001;12:1439–43. 144. Graham J, Muhsin M, Kirkpatrick P. Fresh from the pipeline: oxaliplatin. Nat Rev Drug Discov 2004;3:11–2. 145. Cavaletti G, Tredici G, Petruccioli MG, et al. Effects of different schedules of oxaliplatin treatment on the peripheral nervous system of the rat. Eur J Cancer 2001;37:2457–63.


146. Haller DG. Safety of oxaliplatin in the treatment of colorectal cancer. Oncology 2000;14:15–20. 147. Culy C, Clemett D, Wiseman L. Oxaliplatin. A review of its pharmacological properties and clinical efficacy in metastatic colorectal cancer and its potential in other malignancies. Drugs 2000; 60:895–924. 148. Michalska D, Wysokiski R. Molecular structure and bonding in platinum-picoline anticancer complex: density functional study. Collect Czech Chem Commun 2004;69:63–72. 149. AnorMed. Facts and figures: AMD473 and other platinum based anti-cancer agents. Langley (BC): AnoreMed; 2003. 150. Wosikowski K, Caligiuri M, Rattel B, et al. Efficacy of satraplatin (JM216) and its metabolites is maintained in taxane-resistant tumors. In: Abstracts of the First ISC International Conference on Cancer Therapeutics: Molecular Targets, Pharmacology and Clinical Applications. February 19–21, 2004, Florence, Italy. 151. Giaccone G. Clinical perspectives on platinum resistance. Drugs 2000;59:9–17. 152. Gordon M, Hollander S. Review of platinum anticancer compounds. J Med 1993;24:209–65. 153. Boulikas T, Vougiouka M. Cisplatin and platinum drugs at the molecular level. Oncol Rep 2003; 10:1663–82. 154. Scarpidis U, Madnani D, Shoemaker C, et al. Arrest of apoptosis in auditory neurons: implications for sensorineural preservation in cochlear implantation. Otol Neurotol 2003;24:409–17. 155. Lefebvre PP, Malgrange B, Lallemend F, et al. Mechanisms of cell death in the injured auditory system: otoprotective strategies. Audiol Neurootol 2002;7:165–70. 156. Teranishi M, Nakashima T, Wakabayashi T. Effects of alpha-tocopherol on cisplatin-induced ototoxicity in guinea pigs. Hear Res 2001;151:61–70. 157. Campbell KC, Rybak LP, Meech RP, et al. D Methionine provides excellent protection from cisplatin ototoxicity in the rat. Hear Res 1996; 102:90–8. 158. Ekborn A, Laurell G, Johnstrom P, et al. D Methionine and cisplatin ototoxicity in the guinea pig: D-methionine influences cisplatin pharmacokinetics. Hear Res 2002;165:53–61. 159. Treskes M, van der Vijgh WJ. WR2721 as a modulator of cisplatin- and carboplatin-induced side

160.

161.

162.

163.

164.

165.

166.

167.

168. 169.

170.

171.

75

effects in comparison with other chemoprotective agents: a molecular approach. Cancer Chemother Pharmacol 1993;33:93–106. Laurell G, Teixeira M, Sterkers O, et al. Local administration of antioxidants to the inner ear. Kinetics and distribution. Hear Res 2003;173:198–209. Ramesh G, Reeves WB. Salicylate reduces cisplatin nephrotoxicity by inhibition of tumor necrosis factor-alpha. Kidney Int 2004;65:490–9. Li G, Sha SH, Zotova E, et al. Salicylate protects hearing and kidney function from cisplatin toxicity without compromising its oncolytic action. Lab Invest 2002;82:585–96. Kaltenbach JA, Church MW, Blakley BW, et al. Comparison of five agents in protecting the cochlea against the ototoxic effects of cisplatin in the hamster. Otolaryngol Head Neck Surg 1997; 117:493–500. Weiss J. Pharmacologic approaches to protection against radiation-induced lethality and other damage. Environ Health Perspect 1997;105:1473–8. Santini V, Giles F. The potential of amifostine: from cytoprotectant to therapeutic agent. Haematologica 1999;84:1035–42. van der Hulst RJ, Dreschler WA, Urbanus NA. High frequency audiometry in prospective clinical research of ototoxicity due to platinum derivatives. Ann Otol Rhinol Laryngol 1988; 97:133–7. Walker DA, Pillow J, Waters KD, et al. Enhanced cis-platinum ototoxicity in children with brain tumours who have received simultaneous or prior cranial irradiation. Med Pediatr Oncol 1989; 17:48–52. McKeage MJ. Comparative adverse effect profiles of platinum drugs. Drug Saf 1995;13:228–44. Shulman A. The cochleovestibular system/ototoxicity/clinical issues. Ann N Y Acad Sci 1999; 884:433–6. Feliu J, Martin G, Madronal C, et al. Combination of low-dose cisplatin and gemcitabine for treatment of elderly patients with advanced nonsmall-cell lung cancer. Cancer Chemother Pharmacol 2003;52:247–52. Doolittle ND, Muldoon, LL, Brumett RE, et al. Delayed sodium thiosulfate as an otoprotectant against carboplatin-induced hearing loss in patients with malignant brain tumors. Clin Cancer Res 2001;7:493–500.

CHAPTER 7

Iron-Chelating and Other Chemotherapeutic Agents: The Vinca Alkaloids Andrew R. Scott, BM, BS, MPhil, FRCS(ORL-HNS), Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), and John A. Rutka, MD, FRCSC

IRON-CHELATING AGENTS Deferoxamine Deferoxamine (DFO), also known as desferrioxamine, is an iron-chelating agent available for intramuscular, subcutaneous, and intravenous administration. It obtained approval from the US Food and Drug Administration (FDA) for this use in 1968.1 It is indicated for the treatment of acute iron intoxication and of chronic iron overload secondary to multiple transfusions (as may occur in the treatment of chronic anemias, including thalassemia, thalassemia intermedia, and Diamond-Blackfan anemia).1,2 DFO also has antimalarial properties, antioxidant effects, antiproliferative effects, and the ability to chelate aluminum. These are areas of research and not licensed clinical indications for DFO use at this time.1 Reports of ototoxicity attributed to DFO emerged in the 1980s.2 Somewhat paradoxically, DFO has been used more recently in animal studies, for its ironchelating and antioxidant effects, to try to better understand the ototoxic mechanisms of other agents, including excessive noise, aminoglycosides, to cisplatin, and to try to protect the inner ear from these agents.3–14 De Virgillis and colleagues reported sensorineural hearing loss (SNHL) in thalassemia major patients receiving DFO in 1979.15 However, the authors did not initially consider DFO to be responsible.2 It should also be noted that thalassemic patients appeared more prone to conductive hearing loss, possibly related to adenoidal hypertrophy causing eustachian tube dysfunction, than were normal subjects.2 Marsh and colleagues described a case of reversible tinnitus associated with DFO in 1981.16 The patient, a woman with thalassemia intermedia, developed tinnitus while receiving DFO that resolved upon discontinuation of DFO. Of interest, the tinnitus returned on further DFO therapy, suggesting a drug effect. No hearing loss was detected on audiometry, however. A further case of suspected

DFO ototoxicity was reported by Guerin and colleagues in 1985.17 They described a 26-year-old woman on hemodialysis who had acquired iron overload secondary to multiple transfusions for severe anemia. After 7 months of treatment with DFO the patient noticed a hearing loss. Audiometry demonstrated a mid- to high-frequency SNHL. Hearing reportedly returned to normal within 5 weeks following cessation of DFO. The authors commented that the toxicity had occurred once the iron levels had fallen to normal and cautioned that DFO should be discontinued once signs of metal overload disappeared. Further isolated case reports have continued.18–20 Heightened awareness of the potential for ototoxicity from DFO has led to several published case series from the mid-1980s to the present day. These are summarized in Table 7-1 according to underlying pathology, number of patients affected, size of series, hearing loss described, and whether hearing loss was supported audiometrically.21–35 A wide range in incidence of SNHL in these populations has been identified (3.8–56%). This may be partly explained by the different populations examined, with some series focusing on a pediatric setting with varying pathologies. Reporting may also vary, with some authors including only those patients in whom they felt the SNHL was definitely attributable to DFO, whereas others included all SNHL in the populations studied. Nevertheless, only 3 of the 15 studies concluded that the risk of DFO ototoxicity was exaggerated or negligible when compared with the general population.24,27,34 The pattern of SNHL encountered was predominantly high frequency, although some series also reported other patterns, including notched hearing at 3 or 6 kHz or pan-frequency losses.22,25,33 Reversibility of the hearing loss is variable in the published reports. Although some patients recovered their hearing completely, others experienced no improvement, and a proportion required amplification with hearing aids.2 Although cause and effect remains

Iron-Chelating and Other Chemotherapeutic Agents

77

Table 7-1 Published Case Series of SNHL with DFO Study

Pathology

Patients (%)

Audiogram

SNHL Pattern

Reversibility

Olivieri et al, 1986 21

Hemoglobinopathy

22/89

(25)

Yes

High frequency

4 complete, 1 partial

Barratt and Toogood, 1987 22

Hemoglobinopathy

9/27

(33)

Yes

6 high frequency, 2 flat line, 1 midfrequency

NP

Albera et al, 1988 23

Hemoglobinopathy

58/153 (38)

Yes

High frequency

NP

Masala et al, 1988 24

Hemoglobinopathy

12/100 (12)

Yes

High frequency

NP

Wonke et al, 1989 25

Hemoglobinopathy

13/54

(26)

Yes

6 high-frequency notch, 1 complete, 2 low frequency, 4 partial 5 high frequency

Porter et al, 1989 26

Hemoglobinopathy

9/47

(19)

Yes

NP

2 partial

27

Hemoglobinopathy

2/52

(3.8) Yes

High frequency

1 partial

Cases et al, 1990 28

Hemodialysis

6/41

(15)

Yes

Mid- to high frequency

3 complete, 3 partial

Triantafyllou et al, 1992 29

Hemoglobinopathy

26/120 (22)

Yes

NP

NP

De Espana et al, 1992 30

Hemodialysis

3/20

(15)

NP

NP

NP

Styles and Vichinsky, 1996 31

Hemoglobinopathy

8/28

(29)

Yes

High frequency

8 complete

Kontzoglou et al, 1996 32

Hemoglobinopathy

24/88

(27)

Yes

High frequency

12 partial

Chiodo et al, 1997 33

Hemoglobinopathy

22/75

(29)

Yes

High frequency, High frequency notch

NP

Ambrosetti et al, 200034

Hemoglobinopathy

15/57

(26)

Yes

Mild to moderate highfrequency loss

No change

Karimi et al, 2002 35

Hemoglobinopathy

72/128 (56)

Yes

High frequency

NP

Cohen et al, 1990

Adapted and updated from Kanno et al.2 DFO = deferoxamine; NP = data not provided; SNHL = sensorineural hearing loss.

unproven in the case series studies, an association between DFO therapy and SNHL in these groups of patients seems likely at this time. Risk Factors and Minimizing DFO Ototoxicity In their earlier review of the subject, Kanno and colleagues identified five risk factors for DFO ototoxicity: (1) younger age, (2) higher dose of DFO, (3) lower serum ferritin level, (4) monthly DFO amount, and (5) better compliance with chelation therapy.2 The total cumulative dose of DFO, however, did not appear to be a risk factor. The observation that hearing loss often became apparent when serum ferritin levels had fallen to normal led Porter and colleagues to recommend the use of a therapeutic index to adjust the dosage of DFO to minimize the risk of ototoxicity.26,36,37 The therapeutic index is calculated by dividing the mean daily dose of

DFO by the serum ferritin level. If the therapeutic index is less than 0.025, then the risk of developing ototoxicity is said to be reduced. Porter also recommended the mean daily dose not exceed 40 mg/kg and that routine audiologic monitoring be performed.37 Treatment if toxicity develops, largely consists of discontinuing DFO (depending on the risk–benefit assessment for that individual patient). Pinna and colleagues described a case in which oral zinc supplemented the cessation of DFO. 38 The patient was a 25-year-old woman with thalassemia major who developed presumed DFO-induced optic neuropathy and SNHL (confirmed audiometrically). The visual disturbance and audiogram returned to normal after 2 days of withholding DFO and of treatment with oral zinc sulfate (100 mg twice daily). This was a much quicker recovery than would normally be anticipated by withholding DFO alone. They concluded that oral zinc sulfate might accelerate neurotoxicity reversal.

78

Systemic Toxicity

Mechanism of DFO Ototoxicity Neurologic disturbances with DFO treatment include peripheral sensory, motor, or mixed neuropathy. 1 Where DFO exerts its main effect on the auditory pathway is unclear. Positive recruitment tests and normal auditory brainstem responses (ABRs) measured in several of the case series have suggested a cochlear effect.15,23,25,26,28 Conversely, abnormal ABRs in another series would point to a retrocochlear process.30 Animal studies are also conflicting. Shirane and Harrison administered DFO to chinchillas using a chronic regimen of 100 mg/kg/d for 5 days per week for 3 months.39,40 Functionally, no significant difference in auditory thresholds compared with control animals was found. Morphologically, no evidence of cochlea damage was seen on scanning electron microscopy (SEM). Ryals and colleagues, however, did find morphological evidence of cochlea damage on light microscopy when they examined adult quails that had received 300 or 750 mg/kg DFO for 30 days.41 At the lower dosage, supporting cells only were damaged, whereas the higher dose also affected the sensory hair cells. Yamanobe and Kanno were also able to demonstrate DFO ototoxicity in an animal model.42 A regimen of 600 mg/kg DFO per day administered to guinea pigs produced an increase in thresholds of compound action potential (CAP) from the auditory nerve. SEM studies on these animals showed damage predominantly to the outer hair cells of the cochlea. Thus, DFO appears to have a direct effect on the cochlea, at least at an animal level. However, the possibility remains that higher pathways could also be involved by the neuropathic effects of DFO.

VINCA ALKALOIDS The vinca alkaloids are a class of antitumor drugs that include vincristine, vinblastine, and vinorelbine.1 They are structurally similar compounds that are made up of two multiringed units, vindoline and catharanthine. Vincristine and vinblastine are derived from an alkaloid obtained from the periwinkle plant (Vinca rosea, Linn.). Vinorelbine is semisynthetic. Vinca alkaloids are predominantly metabolized by the liver.1 Their precise mechanism of cytotoxicity continues to be under investigation. Vinorelbine and vincristine appear to mainly affect microtubule formation during mitosis.1 Vinblastine differs in that it seems to interfere with the metabolic pathways of amino acids leading from glutamic acid to the citric acid cycle and to urea.1 Although there are many reports of ototoxicity with combination chemotherapy, isolating the responsible agent is often not possible. This is particularly true of the vinca alkaloids, which are not infrequently used along with cisplatin, for example, in the treatment of testicular tumors. Cisplatin is known to be ototoxic, so

it is often impossible to separate the relative effects of the combination agents. Vincristine Vincristine obtained FDA approval in 1984, and its current indications include acute leukemia, selected lymphomas, rhabdomyosarcoma, and Wilms’ tumor.1 Mahajan and colleagues described one case of ototoxicity attributed to vincristine.43 They documented a 73-year-old woman with non-Hodgkin’s lymphoma who had two episodes of bilateral SNHL (60 dB averages) after the sixth and seventh 2 mg doses of vincristine. Her hearing gradually returned over 2 to 3 months following discontinuation of the vincristine. Yousif and colleagues described a further case, in a woman, also in her eighth decade of life, also with nonHodgkin’s lymphoma and running a similar course.44 Deafness after the first course of chemotherapy (vincristine, chlorambucil, and prednisolone) was reported as being attributed partly to a concomitant upper respiratory tract infection. The patient was described as becoming profoundly deaf in both ears after the second course of chemotherapy with tuning fork tests indicating a SNHL. No audiologic data were provided. The hearing reportedly returned to near pretreatment levels over a few weeks. Lugassy and Shapira also described a case of SNHL associated with vincristine treatment in a 64-year-old patient with multiple myeloma.45 They went on to conduct a prospective cohort study to further assess this possible association.46 Twenty-three patients were followed audiometrically before and during treatment. No adverse effects on hearing, including pure tone audiometry and speech audiometry, were reported at moderate doses of vincristine (6–16 mg total dose). However, in the patient who received high-dose vincristine (24 mg total dose), a mild SNHL was observed. In an earlier study, Engstrom and colleagues looked at the ototoxic effects of combination chemotherapy in 26 patients with small cell carcinoma of the lung.47 Each patient received combination treatment with doxorubicin, vincristine, cyclophosphamide, and methotrexate. Hearing was monitored with repeated audiometry during treatment. No significant impairment of hearing was observed. In addition, postmortem examination of the temporal bones was performed in seven patients, and no alterations in cochlear morphology were reported. As mentioned previously, identifying the injurious agent in combination therapy is often not possible. However, in a multivariant analysis of the risk factors for cisplatin ototoxicity in a group of 86 patients with testicular cancer, high-dose vincristine was identified as an independent risk factor, and greater vigilance was called for in this subset of patients.48 In summary, the clinical evidence suggests a potential for ototoxicity with vincristine, particularly at higher doses.


The mechanism for vincristine ototoxicity is unclear. Vincristine is known to have neuropathic effects on peripheral, autonomic, and cranial nerves.1,49,50 Some evidence from animal work suggests that it may also be directly toxic to the cochlea. Using light microscopy, Serafy and Hashash examined the cochleas of six rabbits exposed to vincristine via intraperitoneal injections compared with six animals that received control injections.49 One-half of the animals were sacrificed and examined after three injections and the rest after five. Atrophy of the organ of Corti was reported in the animals that received vincristine. There was also degeneration of the nerve fibers in the bony spiral lamina, spiral ganglion, and cochlear nerve. Moreover, the neural degeneration was greater in the animals that received the longer course of vincristine. The authors postulated that the changes seen in the organ of Corti were the direct effect of the drug or secondary to the degeneration of spiral ganglion cells. The direct effect hypothesis was favored given that damage to the organ of Corti appeared relatively greater than that to the neural structures in the animals that had received the shorter course of vincristine. Either explanation, however, remains possible. Vinblastine Vinblastine obtained FDA approval in 1964, and its current indications include breast carcinoma, embryonal cell carcinoma, testicular carcinoma, choriocarcinoma, histiocytosis X, selected lymphomas, mycosis fungoides, Kaposi’s sarcoma, and teratocarcinoma.1 Again, identifying the toxic agent in multidrug treatment regimens is difficult. A case report by Moss and colleagues, however, suggests that vinblastine may cause ototoxicity.51 They described a 29-year-old man with recurrent Hodgkin’s disease who received doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) chemotherapy, every 2 weeks for 12 cycles. The patient reported tinnitus after each treatment cycle, and when audiology was performed a mild high-frequency SNHL was noted. Vinblastine was implicated because of its temporal association with the auditory symptoms and its similarities with vincristine. Serafy and colleagues examined the ototoxicity of vinblastine in an animal model, similar to their work on vincristine.52 They noted progressive damage to the organ of Corti with increasing doses of vinblastine. The nerve cells of the spiral ganglion appeared preserved (contrary to the findings with vincristine), and therefore a direct toxic effect on the end organ was concluded. In summary, vinblastine appears to be potentially ototoxic, but clinical evidence remains sparse for the doses and regimens being used in current practice. Vinorelbine Vinorelbine is a relatively newer agent, obtaining FDA approval in 1994 for the treatment of lung carcinoma.1

79

A case of ototoxicity in a patient treated with vinorelbine and paclitaxel for metastatic breast carcinoma has been reported.53 Vinorelbine is known to cause peripheral neuropathy, and ototoxicity has been observed in combination with cisplatin previously.1 Again, it is not possible to determine which agent was responsible for the ototoxicity or indeed if both may have had an effect.

SUMMARY • Iron-chelating agents, namely deferoxamine (DFO), also known as desferrioxamine, have been associated with a predominantly highfrequency SNHL is some series. • Risk factors for DFO toxicity include younger age, higher dose of DFO, lower serum ferritin levels, monthly DFO amount, and, somewhat paradoxically, better compliance with DFO therapy. A therapeutic index for DFO (mean daily dose of DFO [mg/kg] divided by serum ferritin level) of less than 0.025 appears to be associated with a reduced risk for ototoxicity. • Vinca alkaloids (vincristine, vinblastine, and vinorelbine) have all been associated with risk for ototoxicity at higher doses. Determining whether a vinca alkaloid in itself is responsible for ototoxicity is often not possible as these agents are typically used in combination with other potentially ototoxic chemotherapeutic agents (eg, cisplatin).

REFERENCES 1. Mosby’s drug consult. Elsevier; 2003. Available at: www3.us.elsevierhealth.com/DrugConsult/ (accessed Aug 2003). 2. Kanno H, Yamanobe S, Rybak LP. The ototoxicity of deferoxamine mesylate. Am J Otolaryngol 1995;16: 148–52. 3. Clerici WJ, DiMartino DL, Prasad MR. Direct effects of reactive oxygen species on cochlear outer hair cell shape in vitro. Hear Res 1995;84:30–40. 4. Song BB, Schacht J. Variable efficacy of radical scavengers and iron chelators to attenuate gentamycin ototoxicity in guinea pig in vivo. Hear Res 1996;94:87–93. 5. Clerici WJ, Yang L. Direct effects of intraperilymphatic reactive oxygen species generation on cochlear function. Hear Res 1996;101:14–22. 6. Ryals B, Westbrook E, Schacht J. Morphological evidence of ototoxicity of the iron chelator deferoxamine. Hear Res 1997;112:44–8. 7. Conlon BJ, Perry BP, Smith DW. Attenuation of neomycin ototoxicity by iron chelation. Laryngoscope 1998;108:284–7. 8. Song BB, Sha SH, Schacht J. Iron chelators protect from aminoglycoside-induced cochleo- and

80

9.

10.

11.

12.

13.

14.

15.

16.

17. 18.

19.

20.

21.

22.

23.

Systemic Toxicity

vestibulo-toxicity. Free Radic Biol Med 1998; 25:189–95. Wantanabe H, Kanno H. Experimental studies of the protective effect of deferoxamine mesilate on cisplatin induced toxicity [Japanese]. J Otorhinolaryngol Soc Jpn 1998;101:967–78. Yamasoba T, Schacht J, Shoji F, Miller JM. Attenuation of cochlear damage from noise trauma by an iron chelator, a free radical scavenger and glial cell line-derived neurotrophic factor in vivo. Brain Res 1999;815:317–25. Forge A, Li L. Apoptotic death of hair cells in mammalian vestibular sensory epithelia. Hear Res 2000;139:97–115. Dehne N, Lautermann J, Petrat F, et al. Cisplatin ototoxicity: involvement of iron and enhanced formation of superoxide anion radicals. Toxicol Appl Pharmacol 2001;174:27–34. Tabuchi K, Okubo H, Fujihira K, et al. Protection of outer hair cells from reperfusion injury by an iron chelator and a nitric oxide synthase inhibitor in the guinea pig cochlea. Neurosci Lett 2001;307:29–32. Tabuchi K, Tsuji S, Asaka Y, et al. Ischemia-reperfusion injury of the cochlea: effects of an iron chelator and nitric oxide synthase inhibitor. Hear Res 2001;160:31–6. De Virgillis S, Argiolu F, Sanna G, et al. Auditory involvement in thalassemia major. Acta Haematol 1979;61:209–15. Marsh MN, Holbrook IB, Clark C, Schaffer JL. Tinnitus in a patient with beta-thalassaemia intermedia on long-term treatment with desferrioxamine. Postgrad Med J 1981;57:582–4. Guerin A, London G, Marchais S, et al. Acute deafness and desferrioxamine. Lancet 1985;ii:39–40. Orton RB, de Veber LL, Sulh HM. Ocular and auditory toxicity of long-term, high-dose subcutaneous deferoxamine therapy. Can J Ophthalmol 1985; 20:153–6. Cases A, Campistol JM, Sabater M, et al. Desferrioxamine-induced acute neurosensorial deafness. Nephron 1988;48:326. Cases A, Kelly J, Sabater J, et al. Acute visual and auditory neurotoxicity in patients with end-stage renal disease receiving desferrioxamine. Clin Nephrol 1988;29:176–8. Olivieri NF, Buncic JR, Chew E, et al. Visual and auditory neurotoxicity in patients receiving subcutaneous deferoxamine infusions. N Engl J Med 1986;314:869–73. Barratt PS, Toogood IRG. Hearing loss attributed to desferrioxamine in patients with beta-thalassaemia major. Med J Aust 1987;147:177–9. Albera R, Pia F, Morra B, et al. Hearing loss and desferrioxamine in homozygous beta-thallasaemia. Audiology 1988;27:207–14.

24. Masala W, Meloni F, Gallisai D, et al. Can deferoxamine be considered an ototoxic drug? Scand Audiol Suppl 1988;30:237–8. 25. Wonke B, Hoffbrand AV, Aldouri M, et al. Reversal of desferrioxamine induced auditory neurotoxicity during treatment with G-DTPA. Arch Dis Child 1989;64:77–82. 26. Porter JB, Jawson MS, Huehns ER, et al. Desferrioxamine ototoxicity: evaluation of risk factors in thalassaemic patients and guidelines for safe dosage. Br J Haematol 1989;73:403–9. 27. Cohen A, Martin M, Mizanin J, et al. Vision and hearing during deferoxamine therapy. J Pediatr 1990;117(2 Pt 1):326–30. 28. Cases A, Kelly J, Sabater F, et al. Ocular and auditory toxicity in haemodialyzed patients receiving desferrioxamine. Nephron 1990;56:19–23. 29. Triantafyllou N, Fisfis M, Sideris G, et al. Neurophysiological and neuto-otological study of homozygous beta-thalassemia under long-term desferrioxamine (DFO) treatment. Acta Neurol Scand 1991;83:306–8. 30. De Espana R, Biurrun O, Lorente J, et al. Ototoxicity of deferoxamine. An Otorrinolaringol Ibero Am 1992;19:341–7. 31. Styles LA, Vichinsky EP. Ototoxicity in hemoglobinopathy patients chelated with desferrioxamine. J Pediatr Hematol Oncol 1996;18: 41–5. 32. Kontzoglou G, Koussi A, Tsatra J, et al. Sensorineural hearing loss in children with thalassemia major in Northern Greece. Int J Pediatr Otorhinolaryngol 1996;35:223–30. 33. Chiodo AA, Alberti PW, Sher GD, et al. Desferrioxamine ototoxicity in an adult transfusiondependent population. J Otolaryngol 1997; 26:116–22. 34. Ambrosetti U, Donde E, Piatti G, Cappellini MD. Audiological evaluation in adult beta-thalassemia major patients under regular chelation treatment. Pharmacol Res 2000;42:485–7. 35. Karimi M, Asadi-Pooya AA, Khademi B, et al. Evaluation of the incidence of sensorineural hearing loss in beta-thalassemia major patients under regular chelation therapy with desferrioxamine. Acta Haematol 2002;108:79–83. 36. Porter JB, Faherty A, Stallibrass L, et al. A trial to investigate the relationship between DFO phasmacokinetics and metabolism and DFO-related toxicity. Ann N Y Acad Sci 1998;850:483–7. 37. Porter JB. A risk-benefit assessment of ironchelation therapy. Drug Saf 1997;17:407–21. 38. Antonio P, Corda L, Carta F. Rapid recovery with oral zinc sulphate in deferoxamine-induced presumed optic neuropathy and hearing loss. J Neuroophthalmol 2001;21:32–3.


39. Shirane M, Harrison RV. A study of the ototoxicity of deferoxamine in chinchilla. J Otolaryngol 1987; 16:334–9. 40. Shirane M, Harrison RV. The effects of deferoxamine mesylate and hypoxia on the cochlea. Acta Otolaryngol 1987;104:99–107. 41. Ryals B, Westbrook E, Schacht J. Morphological evidence of ototoxicity of the iron chelator deferoxamine. Hear Res 1997;112:44–8. 42. Yamanobe S, Kanno H. An experimental study of ototoxicity induced by deferoxamine mesilate. J Otorhinolaryngol Soc Jpn 1998;101:979–87. 43. Mahajan SL, Ikeda Y, Myers TJ, Baldini MG. Acute acoustic nerve palsy associated with vincristine therapy. Cancer 1981;47:2404–6. 44. Yousif H, Richardson SG, Saunders WA. Partially reversible nerve deafness due to vincristine. Postgrad Med J 1990;66:688–9. 45. Lugassy G, Shapira A. Sensorineural hearing loss associated with vincristine treatment. Blut 1990; 61:320–1. 46. Lugassy G, Shapira A. A prospective cohort study of the effect of vincristine on audition. Anticancer Drugs 1996;7:525–6.

81

47. Engstrom B, Hillerdal M, Hillerdal G, Nou E. Hearing and cochlear morphology in patients treated with a combination of cytostatics. Acta Otolaryngol Suppl 1990;470:119–23. 48. Bokemeyer C, Berger CC, Hartmann JT, et al. Analysis of risk factors for cisplatin-induced ototoxicity in patients with testicular cancer. Br J Cancer 1998;77:1355–62. 49. Serafy A, Hashash M. The effect of vincristine on the neurological elements of the rabbit cochlea. J Laryngol Otol 1981;95:49–54. 50. Schweitzer VG. Ototoxicity of chemotherapeutic agents. Otolaryngol Clin North Am 1993;26: 759–89. 51. Moss PE, Hickman S, Harrison BR. Ototoxicity associated with vinblastine. Ann Pharmacother 1999;33:423–5. 52. Serafy A, Hashash M. The effect of vinblastine on the neurological elements of the rabbit cochlea. J Laryngol Otol 1982;96:975–9. 53. Tibaldi C, Pazzagli I, Berrettini S, De Vito A. A case of ototoxicity in a patient with metastatic carcinoma of the breast treated with paclitaxel and vinorelbine. Eur J Cancer 1997;34:1133.

CHAPTER 8

Clinical Aminoglycoside Ototoxicity Coleman Rotstein, MD, FRCPC, and Lionel Mandell, MD, FRCPC, FRCP(Lond)

Aminoglycoside antibiotics have been an important and integral part of our antibacterial drug armamentarium since their discovery in the 1940s. They possess potent in vitro activity against Pseudomonas aeruginosa and most other aerobic gram-negative bacilli while also exhibiting some activity against Staphylococcus aureus.1 They also demonstrate a synergistic activity when used in combination with certain β-lactam antibiotics.2 Their toxicity, however, has led to comparatively restrained use in the past two decades, particularly since the broadspectrum cephalosporins, carbapenems, and fluoroquinolones were introduced. Indeed, it is fair to say that their narrow therapeutic window has limited their use.3 The major toxicities of note shared by all aminoglycosides are nephrotoxicity, which occurs in approximately 15% of patients receiving divided-dose regimens, and ototoxicity, which is manifested by hearing loss (5%) and vestibular toxicity (3%).4–10 Aminoglycosides are fermentation products or semisynthetic derivatives from soil actinomycetes. They are produced or derived from Streptomyces or Micromonospora species,1 and the suffix of the drug name indicates its origin. Thus, aminoglycosides ending in “mycin” are derived directly or indirectly from Streptomyces. Similarly, the aminoglycosides ending in “micin” are products of Micromonospora. Currently, eight aminoglycosides (gentamicin, tobramycin, amikacin, streptomycin, neomycin, kanamycin, paromomycin, and spectinomycin) are approved by the Food and Drug Administration and marketed in the United States. In Canada, only six products are presently available (gentamicin, tobramycin, amikacin, streptomycin, netilmicin, and paromomycin) (Table 8-1). Only certain aminoglycoside products are available for parenteral, oral, or topical use (otic or ophthalmologic solutions) (see Table 8-1).11,12 All aminoglycosides have a six-member ring (aminocyclitol) with amino-group substituents and glycosidic bonds between the aminocyclitol ring and two or more sugars, thereby accounting for the name

aminoglycoside. The compounds are highly soluble in water. Their insolubility in organic solvents limits their ability to cross lipid-containing cellular membranes.1 At an alkaline pH, aminoglycosides have a very high positive charge (ie, cationic), which enhances their antibacterial activity. In contrast, acidic environments (eg, the stomach) reduce their activity. Compared with other drugs, such as the fluoroquinolones, the aminoglycosides do not have well understood structure-activity relationships.

CLINICAL INDICATIONS FOR AMINOGLYCOSIDE THERAPY Aminoglycosides are considered first and foremost to be bactericidal antimicrobial agents. They act at various sites of the bacterial cell. As a first step, cationic aminoglycosides bind to the anionic outer membrane of the gram-negative organisms, thereby disrupting the integrity of the cell wall’s normal permeability function. Their uptake into the bacterial cell is increased in an alkaline environment producing greater intracellular concentrations, thus enhancing the antibacterial effect of the drug. Second and most important, aminoglycosides impair bacterial protein synthesis by binding to the 30S ribosomal subunit leading to misreading of the genetic code and inhibition of translocation.13,14 Elongation of the amino acid chain fails to occur, resulting in bacterial death. Two important pharmacodynamic properties related to the activity of aminoglycosides are their postantibiotic effect and concentration-dependent killing. The postantibiotic effect can be simply described as persistent suppression of bacterial growth that occurs after the drug has been removed in vitro or cleared by metabolism and eliminated in vivo.15 In addition, aminoglycosides kill bacteria more rapidly and efficiently when higher drug concentrations are achieved. High drug concentrations may be easily achieved with once-daily dosing compared with twice-

Clinical Aminoglycoside Ototoxicity

83

Table 8-1 Origin and Availability of Parenteral, Oral, and Topical (Otic/Ophthalmic Solutions) Aminoglycosides Parenteral Availability or Oral Preparation

Otic/Ophthalmic* Solution Availability

Streptomyces 1972 kanamyceticus

Canada, United States (IM, IV)

NA

Gentamicin

Micromonospora 1963 purpurea and M. echinospora

Canada, United States (IM, IV)

Canada, alone or in combination with betamethasone; United States

Kanamycin

S. kanamyceticus 1957

United States (IM, IV)

NA

Neomycin

S. fradiae

1949

United States (oral powder, solution, or tablets)

United States, in combination with polymyxin B and hydrocortisone; Canada, gramicidin, neomycin, and polymyxin B combination, framycetin (neomycin II), gramicidin and dexamethasone combination

Netilmicin

M. inyoensis

1975

Canada (IM, IV)

NA

Paromomycin S. fradiae

1959

United States, Canada (oral capsules only)

NA

Spectinomycin S. spectabilis

1962

United States (IM)

NA

Streptomycin

S. griseus

1944

United States, Canada (IM)

NA

Tobramycin

S. tenebrarius

1968

United States, Canada (IM, IV)

United States/Canada, tobramycin alone and with dexamethasone combination*

Drug

Origin

Amikacin

Year of Discovery

IM = intramuscular; IV = intravenous; NA = not available. *Ophthalmic solutions containing aminoglycosides are not infrequently used in an “off-label” fashion for otic disease.

or thrice-daily dosing while avoiding elevated trough concentrations that are linked to nephrotoxicity.16 In Vitro Activity The aminoglycosides exhibit in vitro activity against a wide range of aerobic gram-negative pathogens, including Enterobacteriaceae (Escherichia coli, Klebsiella spp, Enterobacter spp, Proteus spp, Serratia spp, and Citrobacter spp) and Pseudomonas spp. In vitro activity against Burkholderia cepacia, Stenotrophomonas maltophilia, the anaerobic bacteria such as Bacteroides spp and Clostridium spp, and Streptococcus pneumoniae is absent or poor. In addition, aminoglycosides have activity against methicillin-susceptible but not -resistant S. aureus. Streptomycin remains the most active aminoglycoside against Mycobacterium tuberculosis, whereas amikacin is more active against M. aviumintracellulare and other nontuberculous mycobacteria. Streptomycin is also considered the drug of choice for Yersinia pestis infection. Gentamicin, tobramycin, netilmicin, and amikacin possess virtually identical spectrums of activity, although gentamicin and tobramycin are consistently more active than the other drugs against Serratia and P. aeruginosa, respectively. 1 Spectinomycin is used

clinically for Neisseria gonorrhoeae, whereas paromomycin can by used for the treatment of Entamoeba histolytica. Therapeutic Indications Because of their spectrum of activity and bactericidal properties, the aminoglycosides are clinically useful agents for documented serious infections such as septicemia, complicated urinary tract infections, intraabdominal infections, respiratory tract infections, and osteomyelitis caused by susceptible aerobic gramnegative bacilli (Enterobacteriaceae). Aminoglycosides are also often employed in combination with other antibiotics (usually extended spectrum penicillins with or without a β-lactamase inhibitor, cephalosporins, or carbapenems) for serious infections resulting from Pseudomonas spp.1 Combination therapy with gentamicin is frequently used for the treatment of enterococcal infections, particularly endocarditis when the enterococci do not exhibit high-level aminoglycoside resistance. Gentamicin may also be used for 2 weeks initially in combination with antistaphylococcal penicillins or vancomycin for serious staphylococcal infections to enhance bacterial eradication. Gentamicin specifically

84

Systemic Toxicity

demonstrates synergistic activity with ampicillin for the treatment of Listeria monocytogenes infections. Aminoglycosides are also routinely used in combination with broad-spectrum β-lactam agents with activity against aerobic gram-negative bacilli as empiric antibacterial therapy for febrile neutropenic cancer patients. Aminoglycoside monotherapy, however, even for susceptible organisms, in neutropenic cancer patients is suboptimal and often results in unacceptable failure rates.17 Finally, gentamicin in combination with ampicillin or vancomycin remains a mainstay as prophylaxis to diminish the risk of developing infective endocarditis related to procedures involving the gastrointestinal or genitourinary tracts. As mentioned previously, aminoglycosides play a role in the treatment of mycobacterial diseases. Streptomycin is the most active of the aminoglycosides against M. tuberculosis, whereas amikacin is the one employed in nontuberculous mycobacterial infections.1 Compared with the cephalosporins and fluoroquinolones, the aminoglycosides have demonstrated relative stability against the development of resistance despite many years of use. Resistance is infrequently encountered during treatment of gram-negative bacillary infections. The main mechanisms of resistance to aminoglycosides are (1) the bacterial production of enzymes that modify the antibiotics and inactivate them and (2) the decreased uptake of the antibiotics.14 The latter mechanism is mainly seen in Pseudomonas spp and is likely caused by membrane impermeability.14 The former mechanism, however, is of greater importance. Aminoglycosides are inactivated by phosphorylation, adenylation, or acetylation by means of enzymes encoded in plasmids, but they are also associated with transposable elements.14 Alteration of the ribosomal binding site can occur from enzymatic activity or mutational modification, as seen in M. tuberculosis.1 Low-concentration aminoglycoside resistance is observed in enterococci because of their anaerobic metabolism. This may be overcome through synergism when a cell wall active agent such as penicillin is combined with the aminoglycoside. This synergism is eliminated, however, when high-level resistance to the aminoglycoside (minimum inhibitory concentrations > 2,000 mg/L) exists in enterococci.1

CLINICAL TOXICITIES ASSOCIATED WITH AMINOGLYCOSIDE THERAPY All aminoglycosides, except for spectinomycin, share the potential to be nephrotoxic, to damage the cochlea or vestibular apparatus, and to produce neuromuscular blockade. Nephrotoxicity The reported incidence of nephrotoxicity varies widely, from 0 to 34%, with most reports claiming between

5 and 25% incidence.1,4,18–21 This range of nephrotoxicity is a result of variations in study design, toxicity definitions, patient populations, frequency of measurements, and concomitant risk factors. In general, gentamicin and tobramycin appear to be equally nephrotoxic and only slightly more so than amikacin and netilmicin.18 The pathogenesis of the acute renal insufficiency resulting from aminoglycosides stems from the development of acute tubular necrosis. This adverse event is relatively common, with a rise in plasma creatinine of 0.5 to 1.0 mg/dL (44 to 88 µmol/L), occurring in 10 to 20% of patients.22,23 As the aminoglycosides are freely filtered in the kidney, almost all of these drugs are then excreted. Only a small portion of the drug is taken up by and stored in the renal tubular cells, particularly the lysosomes of proximal tubule cells. It is this subsequent concentration that damages the cells. As aminoglycosides accumulate within the proximal tubular epithelial cells in lysosomal phospholipid complexes, these complexes eventually rupture and initiate cell death. As a consequence, the local renin–angiotensin system is activated, producing local vasoconstriction and diminished glomerular filtration rate, thereby increasing plasma creatinine values.22 Acute renal failure may occur even if drug serum concentrations are closely monitored, but it tends to occur more frequently in patients with higher serum concentrations.23–25 The propensity to accumulate aminoglycosides in the tubular cells may be overcome by administering the drugs once a day rather than in divided doses because the uptake mechanism in the proximal tubule is saturable. A large single dose will therefore not increase drug uptake once the resorptive capacity is exceeded.26,27 It should be noted that tubular injury is typically reversible, and recovery of renal function may occur despite continued therapy if decreased cellular exposure occurs, as may be accomplished by prolonging the dosage interval.28 The molecular charge on the aminoglycoside molecule interestingly may be a determinant of nephrotoxicity. The number of cationic amino groups (NH3+) per module appears to be a contributing factor for the development of nephrotoxicity. Since neomycin possesses the largest number of amino groups (six groups), it has the greatest predilection for renal compromise, compared with streptomycin (three groups), which has the least. 23,29 Gentamicin, tobramycin, netilmicin, and amikacin (four or five groups) have intermediate potential for nephrotoxicity.24 Cationically charged aminoglycosides bind to megalin, an anionic protein in the luminal membrane, which facilitates the movement of the complex to the lysosome and thereafter fusion with lysosome membranes. Lysosome localization of aminoglycosides is observed about an hour after exposure. The megalin dissociates and returns to the cell membrane. The subsequent rupture


of lysosomal complexes culminates in cell death, local vasoconstriction, and decreased glomerular filtration (ie, renal insufficiency). Investigators have experimentally attempted to ameliorate these effects in animals to prevent aminoglycoside-induced nephrotoxicity.30 Ototoxicity Ototoxicity is an adverse effect of aminoglycoside use affecting both the auditory and vestibular functions of the ear. The pathophysiology involves the destruction of the sensory hair cells in the cochlea and vestibular labyrinth.8 Although ototoxicity is the second most common form of toxicity encountered with aminoglycosides, its precise incidence is controversial. Some investigators have reported auditory ototoxicity in up to 41% of subjects,31 whereas others have reported a much lower incidence of about 7%.32 Pooled data from meta-analyses have demonstrated an approximate 5% incidence of auditory toxicity (as measured by at least a 15 dB change in hearing by audiometry) with multiple daily dosing (MDD) of aminoglycosides, compared with a slightly higher 5.8% incidence when single daily dosing (SDD) of aminoglycosides was employed.33,34 A similar situation is reflected in the relatively sparse vestibular toxicity data. Vestibular toxicity has been reported in 0 to 7% of those patients receiving aminoglycosides.18,35–37 What defines a vestibulotoxic event, however, remains somewhat unclear. For example, a decrement in vestibular function as evidenced by a 5 to 8% reduction in caloric test scores, though clinically insignificant, may occur even in healthy volunteers receiving tobramycin.38 Discordant results in the incidence of auditory and vestibular aminoglycoside-induced toxicity have arisen because of the discrepancies between clinical observations demonstrating that very few patients receiving these drugs actually complain of diminished hearing or vertigo and the results of systematic objective testing for these toxicities. With respect to hearing loss, aminoglycoside toxicity primarily affects high-frequency hearing earlier than low-frequency hearing.8 Hearing loss at higher frequencies (> 4,000 Hz) is often detectable audiometrically well before it becomes severe enough to involve the speech frequency range and thereby be perceived by patients. The subliminal nature of a vestibular loss also makes vestibular toxicity difficult to detect at times. As aminoglycoside vestibular toxicity typically occurs without an accompanying injury to the auditory system,35 a vestibular injury can often be compensated by visual and proprioception cues, which probably helps mask the true incidence of this form of aminoglycoside toxicity. Unfortunately, when auditory or vestibular toxicity occurs, it is often irreversible, in contrast to nephrotoxicity. At one time only parentally administered aminoglycosides were thought to cause both forms of

85

ototoxicity. In recent years topical aminoglycoside eardrops have also been implicated as a cause for ototoxicity (see Chapter 12, “Topical Aminoglycoside Cochlear Toxicity” and Chapter 13, “Topical Aminoglycoside Vestibular Toxicity,”).39 Otic solutions containing gentamicin instilled into the ear canal in the presence of a tympanic membrane defect have produced both hearing loss and vestibular damage.39,40 The round window membrane (RWM) is thought to play a key role in the absorption of topical aminoglycosides into the inner ear. The degree of ototoxicity appears to be directly related to the duration and dose of the preparation in contact with the RWM. Toxic levels of aminoglycosides may therefore occur in the inner ear with repeated exposures to the eardrops.41,42 Aminoglycoside-induced ototoxicity is a complex process (see Chapter 9, “Mechanisms for Aminoglycoside Ototoxicity: Basic Science Research”) involving the effect of the drugs on the sensory hair cells in the cochlea and vestibular labyrinth as well as a superimposed genetic predisposition of these cells (see Chapter 17, “Genetic Factors of Aminoglycoside Ototoxicity”) for this adverse event.8,43 Following parenteral administration aminoglycosides enter the inner ear rapidly but do not apparently accumulate in the perilymph and endolymph as the concentrations never exceed those of serum.43,44 Once again, there appears to be an initial interaction between the cationic aminoglycosides and the anionic membranes of the hair cells, which leads to their rapid transport into the cells.45 The exact intracellular target is unknown. One postulate is that aminoglycosides, such as gentamicin, form complexes with iron.46 This activates molecular oxygen-producing superoxides that progress to free hydroxyl radicals, which in turn damage the cell. Another hypothesis of aminoglycoside ototoxicity has been based on the overactivation of glutamate receptors on cochlear synapses.47 The proposed mechanism is related to receptors for N-methyl-D-aspartate (NMDA), which are present at the synapse between cochlear hair cells and neural afferents. Aminoglycosides mimic the positive modulation of polyamines at these receptors, possibly producing excitotoxic damage. The administration of NMDA antagonists such as dizocilpine and ifenprodil can markedly attenuate hearing loss in animals.48 This effect may have been confounded by the fact that dimethyl sulfoxide (DMSO), a potent hydroxyl radical scavenger, was the vehicle for the administration of the antagonists and may have enhanced the protection provided by the antagonists.47 Genetic factors may also contribute to susceptibility for ototoxicity. There appears to be a genetic predisposition linked to two (possibly more) mutations in the mitochondrial chromosome. The first mutation discovered is in location 1555 of the mitochondrial ribosomal ribonucleic acid (RNA) where A-to-G substitution

86

Systemic Toxicity

Table 8-2 Risk Factors for Aminoglycoside Auditory Toxicity Risk Factor

Reference

Duration of therapy

Moore et al51

Bacteremia

Moore et al51

Renal dysfunction

Forge and Schacht,47 Moore et al,51 Neu and Beadush52

Peak temperature

Moore et al51

Liver dysfunction

Moore et al51

Advanced age

Forge and Schact,47 Gatell et al53

Concomitant ototoxic drugs

Govaerts et al54

Higher serum concentrations

Black et al55

Preexisting hearing disorders

Forge and Schacht47

occurred.49 This mutation has been identified in several Chinese and Japanese families and correlated with a maternally inherited hypersensitivity to aminoglycoside-induced hair cell death.43 It is noteworthy that patients with this mutation have exhibited only cochlear toxicity with no involvement of the vestibular system.47 A second point mutation in the small 12S ribosomal RNA gene has also been described.50 The mutant human RNA binds aminoglycosides with high affinity; in contrast, the wild-type human RNA does not bind aminoglycosides. The binding was also tighter for the clinically more toxic drugs such as neomycin than for others such as gentamicin and tobramycin.50 Several other risk factors have been implicated in the development of aminoglycoside-induced ototoxicity (Table 8-2). These studies have predominantly assessed the association of these factors with auditory ototoxicity. Auditory toxicity may be unilateral or bilateral.56 In contrast, vestibular toxicity is more likely to occur in patients with prior disorders of balance57 and renal dysfunction.47 One must also be cognizant of the varying potential for cochlear and vestibular ototoxicity among the aminoglycosides. It is somewhat uncommon to have both vestibular and cochlear symptoms in the same patient,1 but either form of ototoxicity may occur with any of the aminoglycosides. A hierarchy of auditory toxic potential has been established among the aminoglycosides: neomycin is the most toxic, followed in order of decreasing toxicity by gentamicin, tobramycin, amikacin, and netilmicin.1,35,45,54,58 Vestibular symptoms of dizziness, imbalance, nausea, and oscillopsia are comparable with gentamicin and tobramycin, less frequent with amikacin, and least frequent with netilmicin.45,57

An additional issue to be considered whenever discussing both auditory and vestibular toxicity resulting from aminoglycosides is the methods employed to detect these toxicities. Auditory toxicity is typically verified by a decrease in hearing acuity. The common standard used in the literature is a 15 dB reduction in the hearing threshold at two or more frequencies.59 Nevertheless, other definitions such as 10 dB or 20 dB reductions have also been used.32,60 Testing for vestibular toxicity is far less standardized. Diagnostic testing for vestibular toxicity covers a spectrum from clinical symptom documentation to caloric testing with electronystagmography (ENG) and other advanced tests of vestibular function. Variability in the reporting of vestibular toxicity in the literature might have been avoided by standardized testing. Nevertheless, standardization has usually been hampered in practice by the lack of availability of tests such as an ENG.61,62 Neuromuscular Blockade Neuromuscular blockade is a rarely reported adverse effect of aminoglycoside use, although it may be more common than previously believed.63 It is more likely to occur when aminoglycosides are given intravenously and coadministered with neuromuscular blocking drugs or anesthetic agents.63 It has been noted with all of the parenterally administered aminoglycosides. The risk of blockade is magnified by hypomagnesemia, hypocalcemia, and perhaps calcium channel-blocking agents.64 The blockade results from inhibition of the presynaptic release of acetylcholine and blockage of postsynaptic receptor sites of acetylcholine. Calcium internalization must occur before acetylcholine release, which aminoglycosides prevent. In addition, the aminoglycosides blunt the response of postsynaptic receptors to acetylcholine.65 Neomycin is more apt to inhibit presynaptic release, whereas streptomycin and netilmicin act at the postsynaptic site.1

AMINOGLYCOSIDE ADMINISTRATION: MULTIPLEDAILY VERSUS ONCE-DAILY DOSING All parenterally administered aminoglycosides possess similar pharmacokinetics. The drugs’ distribution and elimination occur in three phases.66 The first or distributive phase is the result of drug distribution from the vascular to the extravascular compartment. This phase typically occurs over 15 to 30 minutes. This is the reason peak drug concentrations are usually checked 30 minutes after the completion of a 15- to 30-minute infusion. The second or β phase of elimination encompasses excretion of the drug from the plasma and extravascular space. This phase is determined by the glomerular filtration rate. With poor renal function the elimination rate is more prolonged. In general, the half-life of aminoglycosides with normal renal function is 1.5 to


3.5 hours.1,67,68 The third or γ phase represents the slow elimination of drug that has accumulated in the kidney. As a result of these pharmacokinetic properties, aminoglycosides were traditionally administered in MDD regimens either every 8 or 12 hours commencing with a loading dose to rapidly achieve therapeutic plasma concentrations and followed by a maintenance dose to maintain the drug in a steady state of therapeutic peak concentrations. With this approach, the desired therapeutic concentrations would achieve clinical success. Similarly, by measuring peak and trough serum concentrations of the drug, one ensured that sufficient drug was being administered for clinical efficacy while obviating the development of tissue uptake that might prove to be toxic to the kidney or inner ear.69 In contrast, based on the above-mentioned pharmacodynamic properties of concentration-dependent killing and their postantibiotic effect, aminoglycosides may be administered once daily. As previously described, aminoglycosides kill microorganisms more efficiently and rapidly when higher drug concentrations are present.15 This fact, coupled with the concept that aminoglycoside uptake into the renal proximal tubule cell is saturable, led to the introduction of oncedaily dosing in the hopes that it might reduce the incidence of nephrotoxicity. Drug administration costs were also expected to be somewhat curtailed with this strategy. As a result, numerous trials have attempted to compare MDD with SDD treatment regimens. Published data on once-daily dosing have recently been reviewed by Turnidge.61 He reported that SDD was compared with MDD in 45 clinical trials involving over 6,500 patients where identical or similar total daily aminoglycoside doses were administered. Another 900 patients have also been studied in whom SDD was employed in an uncontrolled fashion or with unequal drug dosages compared with MDD. Data have been compiled in meta-analytic fashion.10,33,34,70–76 The metaanalyses demonstrate superiority for SDD or favor SDD for bacteriologic eradication and a decreased incidence of nephrotoxicity, although the results were not always consistently statistically significant. For ototoxicity predominantly assessed by auditory toxicity, no trend emerged. 10,33,34,70,71,73,75 Some reports favored SDD, 10,33,34,73 whereas others favored MDD. 70,71,75 Vestibular toxicity summary data were inconsistent and demonstrated no statistically significant differences between the SDD and MDD regimens.10,70,71,76 The inconsistencies in the meta-analyses described above with regard to ototoxicity are understandable. This variation is two-pronged. First, there has been tremendous variability in the clinical studies evaluating auditory toxicity as far as testing is concerned, both in performing the test at baseline and end of therapy and in adhering to a standardized definition of toxicity. The variability is even more pronounced regarding the data

87

on vestibular toxicity, where the interpretation and deficiencies in testing vestibular function have hampered meaningful comparisons. Second, the pathogenesis of ototoxicity as previously discussed does not provide a sound rationale for SDD. In contrast to the kidney, where the uptake mechanism of aminoglycosides into the proximal tubule cells is saturable and favors the SDD regimen, hair cell uptake of aminoglycosides in the inner ear does not emulate the aforementioned saturable process, making SDD a bane. Although SDD of aminoglycosides has emerged as the standard of care when employing these agents because of its efficacy, lower drug-induced nephrotoxicity, simplicity (especially in the outpatient setting), and cost reductions, some caveats remain as to its acceptance in all clinical situations. MDD of aminoglycosides, is still preferred for enterococcal endocarditis, severe renal compromise, cystic fibrosis, burns, pregnancy, and meningitis.1,77

RENAL AND CONCENTRATION SERUM MONITORING Rapid attainment of therapeutic concentrations of aminoglycosides has been correlated with clinically successful therapy. When using MDD of aminoglycosides in patients with normal renal function, gentamicin, tobramycin, and netilmicin are administered three times daily according to the manufacturers’ recommendations, whereas amikacin is administered twice daily. The treatment regimens are divided into an initial loading dose (2 mg/kg for gentamicin, tobramycin, and netilmicin but 7.5 mg/kg for amikacin) and subsequent maintenance doses (1.7 mg/kg every 8 hours for gentamicin and tobramycin, 2 mg/kg every 8 hours for netilmicin, and 7.5 mg/kg every 12 hours for amikacin).1 The loading and maintenance aminoglycoside doses are based on ideal body weight, with a modification for obese individuals.1,34,61 Appropriate dosage adjustments are required in the presence of renal insufficiency.61 The dosing regimen is somewhat different when administering aminoglycosides in the SDD format. With this approach, gentamicin and tobramycin are administered at a single daily dose of 5 mg/kg in patients with normal renal function. In patients with an anticipated increased volume of distribution, the dosage is 7 mg/kg/day.78 For netilmicin and amikacin the doses are 6 mg/kg and 15 mg/kg, respectively. The doses are based on ideal body weight, and there is an adjustment in the dosage interval should renal insufficiency be a factor.1,34,61 Monitoring of serum aminoglycoside concentrations is essential for both efficacy and the avoidance of toxicity during MDD therapy. Serum aminoglycoside peak and trough levels are measured after the first or second maintenance dose, and then the maintenance dose

88

Systemic Toxicity

is adjusted accordingly.1 Peak serum concentrations of 4 to 10 mg/L and trough concentrations < 2 mg/L are desired goals for gentamicin, tobramycin, and netilmicin. The corresponding values for amikacin are 15 to 30 mg/L and < 10 mg/L, respectively.1 An accurate record of aminoglycoside administration times and the time the serum concentrations were obtained is essential for interpreting the results. Moreover, serum creatinine should be monitored every 2 to 3 days. If the creatinine level is stable, it is usually not necessary to repeat the serum aminoglycoside concentrations. However, if renal function changes, the dosage must be adjusted, and thereafter the peak and trough serum concentrations require sampling again. SDD of aminoglycosides targets much higher peak concentrations and permits trough concentrations < 1 mg/L. Elevated peak serum concentrations for gentamicin and tobramycin of 20 mg/L, netilmicin of 26 mg/L, and amikacin of 60 mg/L are to be expected in patients with normal renal function.1 As with MDD, peak serum levels are measured to ensure efficacy, and trough levels are performed to monitor for the occurrence of toxicity. However, rather than measuring peak serum levels 30 minutes after the completion of a 30minute infusion and trough levels within 30 minutes of the ensuing infusion as is the routine for the MDD of an aminoglycoside, it may be preferable to obtain a level between 6 and 14 hours after the dose. This serum level is subsequently applied to a nomogram to determine the dosage interval.34,78,79 Alternatively, the more familiar practice of obtaining peak and trough serum concentrations may be employed. As previously mentioned serum creatinine determinations are monitored every 2 to 3 days to provide information about renal function. Dosage interval adjustments are predicated from renal function. The question remains whether serum concentration monitoring is at all necessary if in the final analysis dosage adjustments are carried out solely based on determinations of renal function. Development of elevated serum concentrations ensues secondary to the reduction in renal clearance of the aminoglycoside. Accordingly one may anticipate elevations in peak or trough serum concentrations as a consequence of elevations in serum creatinine and diminished renal clearance of the drug. Following this logic, serum aminoglycoside determinations may be deemed superfluous. On the other hand, a rational approach may be to determine peak and trough serum concentrations at the commencement of therapy to ensure that one is in the therapeutic range while precluding potential accumulation of the aminoglycoside in tissues and thereby reducing toxicity. Prevention of renal dysfunction may in turn reduce the possibility of ototoxicity. Monitoring of serum creatinine every 2 to 3 days may be sufficient while reducing the unnecessary costs of

measuring peak and trough serum aminoglycoside concentrations. To date there is no convincing evidence in the literature to substantiate a causal relationship between peak or trough serum concentrations for aminoglycosides administered in MDD and treatment outcome.19 Moore and colleagues reported on the outcome of 37 cases of gram-negative pneumonia treated with MDD of aminoglycosides.80 Successfully treated patients had higher mean and maximal peak serum concentrations, with a mean peak serum concentration of ≥ 6 mg/L correlating with an improvement in outcome. Nevertheless, these data are of little import in an era of SDD for aminoglycosides where peak serum levels approach 20 mg/L for gentamicin and tobramycin, 26 mg/L for netilmicin, and 60 mg/L for amikacin.1 Similarly, treatment outcomes are not enhanced via the use of individualized dosing of aminoglycosides with concomitant serum concentration monitoring.19 Although elevated serum concentrations of aminoglycosides have been demonstrated to be a risk factor for ototoxicity, particularly auditory toxicity,55,81 no specific threshold serum level has been identified as the precipitating factor for this toxicity.19 Although a higher incidence of nephrotoxicity has been observed with trough aminoglycoside levels > 2 mg/L,4,82 it is unclear whether this elevation actually preceded serum creatinine increases. Some clinicians have suggested that changes in trough serum concentrations occur more rapidly than do serum creatinine determinations. 19 As previously noted, the mechanism of renal insufficiency (ie, decreased creatinine clearance) prompts elevations in serum aminoglycoside concentrations. In summary, serum creatinine measurements remain the most accessible and economical means of monitoring renal function, thereby preventing the development of significant nephrotoxicity and debilitating ototoxicity. As previously recommended, creatinine levels should be determined every 2 or 3 days while on therapy.

RECOMMENDATIONS FOR AMINOGLYCOSIDE THERAPY Although aminoglycosides are potent and clinically effective antimicrobial agents for treating infections resulting from susceptible bacteria, their use has been largely supplanted with the introduction of fluoroquinolones and broad-spectrum cephalosporins that possess comparable spectrums of activity. Despite their inherent and perceived nephrotoxicity and ototoxicity, aminoglycosides remain indispensable agents for patients allergic to fluoroquinolones and cephalosporins and in combination therapy for staphylococcal, enterococcal, L. monocytogenes, and mycobacterial infections. They may also be considered to be salvage drugs in the face of higher antibiotic resistance rates to the fluoro-


quinolones and cephalosporins in aerobic gram-negative bacillary infections. They remain a mainstay of prophylactic therapy in combination with ampicillin or vancomycin to diminish the risk of endocarditis related to gastrointestinal or genitourinary procedures. When considering aminoglycoside therapy, risk factors that impair renal function, such as the concomitant use of nephrotoxins (vancomycin,1,4,83 amphotericin B,1 and furosemide1), should be noted. Efforts to avoid or minimize risk factors that predispose patients to ototoxicity, such as older age, concomitant ototoxic drugs, preexisting hearing or balance disorders, and renal dysfunction, should be undertaken. As the clinical indications for aminoglycoside use have diminished, the usual precautionary considerations are often not exercised as frequently. When aminoglycoside use is necessary in patients at risk, it is prudent to employ once-daily dosing unless this method of administration is inappropriate based on the clinical infection. The use of an aminoglycoside should be restricted to the shortest time interval possible. Serum concentrations should be obtained at the outset of therapy to ensure therapeutic efficacy. Renal function should be monitored every 2 to 3 days. For patients requiring a more protracted treatment course of aminoglycosides of more than 14 days, baseline and weekly auditory (see Chapter 18, “Audiologic Monitoring for Ototoxicity”) and vestibular function should be monitored to detect ototoxicity.19 Others have advocated daily bedside testing of vestibular function, such as the head thrust test, dynamic visual acuity evaluation, and Romberg test, and an assessment of gait while on therapy and after the completion of therapy if patients develop symptoms of vestibular hypofunction (see Chapter 19, “Monitoring Vestibular Ototoxicity).62 At the first sign of auditory or vestibular dysfunction, the aminoglycoside should be discontinued immediately to allow for reversibility of the condition. No outcome studies, however, have demonstrated whether monitoring for ototoxicity will actually reduce morbidity.

CONCLUSION • Aminoglycosides have potent antimicrobial activity against P. aeruginosa and most other gram-negative bacilli, as well as S. aureus. Their spectrum of activity and bactericidal properties have made them clinically useful agents for the treatment of septicemia; complicated urinary tract, respiratory tract, and intra-abdominal infections; and osteomyelitis caused by susceptible aerobic gram-negative bacilli. Their major toxicities include nephrotoxicity, ototoxicity, and neuromuscular blockade. Since the 1980s their use has waned and been supplanted by the broadspectrum cephalosporins and fluoroquinolones with impressive gram-negative coverage.

89

• Despite reservations concerning their toxicities, bacterial resistance to aminoglycosides continues to be low. They remain indispensable agents in patients allergic to fluoroquinolones and cephalosporins and in combination therapy for staphylococcal, enterococcal, L. monocytogenes, and mycobacterial infections. They may also be considered salvage drugs in the face of heightened antibiotic resistance rates and have remained a mainstay for prophylactic therapy in combination with other antibiotics for bacterial endocarditis prophylactic therapy. • Auditory and vestibular ototoxicity has been well documented with both parenteral (MDD and SDD therapy) and topical use. Unlike nephrotoxicity, the full extent of ototoxicity problem is not well appreciated because of limitations in testing for these toxicities as well as inadequate reporting of symptoms. • SDD for aminoglycosides has pharmacodynamic advantages over MDD based on heightened concentration-dependent killing and the postantibiotic effect. Pharmacoeconomic advantages for SDD include relative ease of delivery in the outpatient setting and reduced monitoring costs. Meta-analyses overall suggest that SDD delivery of aminoglycosides results in less nephrotoxicity. The same claim cannot be made with regard to ototoxicity as the mechanisms for toxicity, including their entry into the sensory hair cells of the inner ear, are different. Nephrotoxicity is typically reversible. The same cannot be said for ototoxicity. • There are no a priori surrogate markers available to clinicians designating which patients will develop ototoxicity. Serum creatinine determinations should be measured every 2 to 3 days. Peak aminoglycoside levels in SDD and MDD regimens are measured to ensure efficacy and trough levels measured to monitor toxicity. Prevention of renal dysfunction may reduce the possibility of ototoxicity but does not eliminate it completely. • Prudent medical practice dictates that clinicians must apprise their patients of the potential for ototoxicity and vigilantly monitor patients for this important adverse event. Clinicians should at all times carefully weigh the risk-to-benefit ratio when contemplating the use of aminoglycosides in their patients, especially in the case of prolonged treatment.

REFERENCES 1. Gilbert DN. Aminoglycosides. In: Mandell G, Bennet JE, Dolin R, editors. Principles and practice of infectious diseases. 5th ed. Philadelphia (PA): Churchill Livingstone; 2000. p. 307–36.

90

Systemic Toxicity

2. Barriere SL. Bacterial resistance to beta-lactams and its prevention with combination therapy. Pharmacotherapy 1992;12:397–402. 3. Miyagawa CI. Aminoglycosides in the intensive care unit: an old drug in a dynamic environment. New Horizons 1993;1:172–80. 4. Cimino MA, Rotstein C, Slaughter RL, Emrich LJ. Relationship of serum antibiotic concentrations to nephrotoxicity in cancer patients receiving concurrent aminoglycoside and vancomycin therapy. Am J Med 1987;83:1091–7. 5. Lietman PS, Smith CR. Aminoglycoside nephrotoxicity in humans. Rev Infect Dis 1983;5 Suppl 2:S284–92. 6. Bertino JS, Booker LA, Franck PA, et al. Incidence of and significant risk factors for aminoglycosideassociated nephrotoxicity in patients dosed by using individualized pharmacokinetic monitoring. J Infect Dis 1993;167:173–9. 7. Hottendorf GH, Williams PD. Aminoglycoside nephrotoxicity. Toxicol Pathol 1986;14:66–72. 8. Brummett RE, Fox KE. Aminoglycoside-induced hearing loss in humans. Antimicrob Agents Chemother 1989;33:797–800. 9. Lerner AM, Cone LA, Jansen W, et al. Randomized controlled trial of the comparative efficacy, auditory toxicity and nephrotoxicity of tobramycin and nephrotoxicity of tobramycin and netilmicin. Lancet 1983;i:1123–6. 10. Munckhof WJ, Grayson ML, Turnidge JD. A metaanalysis of studies on the safety and efficacy either once daily or as divided doses. J Antimicrob Chemother 1996;37:645–63. 11. American Society of Health-System Pharmacists. Aminoglycosides in AHFS drug information 2003. Bethesda (MD): American Society of HealthSystem Pharmacists Inc.; 2003. p. 61–76. 12. Canadian Pharmacists Association. Compendium of pharmaceuticals and specialties: the Canadian drug reference for health professionals. Ottawa (ON): Canadian Pharmacists Association; 2003. 13. Fourmy D, Recht MI, Blanchard SC, Puglisi JD. Structure of the A site of Escherichia coli 16S ribosomal RNA complexed with an aminoglycoside antibiotic. Science 1996;274:1367–71. 14. Mingeot-LeClercq MP, Glupczynski Y, Tulkens PM. Aminoglycosides: activity and resistance. Antimicrob Agents Chemother 1999;43:727–37. 15. Fantin B, Ebert S, Leggett J, et al. Factors affecting duration of in-vivo post antibiotic effect for aminoglycosides against gram-negative bacilli. J Antimicrob Chemother 1991;27:829–36. 16. McLean AJ, Joannides-Demos LL, Li SC, et al. Bactericidal effect of gentamicin peak concentration provides a rational for administration of bolus doses. J Antimicrob Chemother 1993;32:301–5.

17. Hughes WT, Armstrong D, Bodey GP, et al. 2002 Guidelines for the use of antimicrobial agents in neutropenic patients with cancer. Clin Infect Dis 2002;34:730–51. 18. Kahlmeter G, Dahlager JI. Aminoglycoside toxicity—a review of clinical studies published between 1975 and 1982. J Antimicrob Chemother 1984; 13 Suppl A:S9–22. 19. McCormack JP, Jewesson PJ. A critical reevaluation of the “therapeutic range” of aminoglycosides. Clin Infect Dis 1992;14:320–9. 20. Lietman PS, Smith CR. Aminoglycoside nephrotoxicity in man. Rev Infect Dis 1983;5:284–93. 21. Bertino JS Jr, Booker LA, Franck PA, et al. Incidence of and significant risk factors for aminoglycoside-associated nephrotoxicity in patients dosed by using individualized pharmacokinetic monitoring. J Infect Dis 1993;167:173–9. 22. Mingeot-LeClercq M, Tulkens PM. Aminoglycoside nephrotoxicity. Antimicrob Agents Chemother 1999;43:1003–12. 23. Humes HD. Aminoglycoside nephrotoxicity. Kidney Int 1998;33:900–11. 24. Moore RD, Smith CR, Lipsky JJ, et al. Risk factors for nephrotoxicity in patients treated with aminoglycosides. Ann Intern Med 1984;100:352–7. 25. Smith CR, Lipsuy JJ, Oetty G, et al. Double-blind comparison of the nephrotoxicity and auditory toxicity of gentamicin and tobramycin. N Engl J Med 1980;302:1106–9. 26. Mattie H, Craig WA, Pechere JC. Determinants of efficacy and toxicity of aminoglycosides. J Antimicrob Chemother 1989;24:281–93. 27. Gilbert DN. Once-daily aminoglycoside therapy. Antimicrob Agents Chemother 1999;35:399–405. 28. Trollfors B. Gentamicin-associated changes in renal function reversible during continued treatment. J Antimicrob Chemother 1983;12:285–7. 29. Bennett WM, Wood CA, Houghton DC, Gilbert DN. Modification of experimental aminoglycoside toxicity. Am J Kidney Dis 1986;8:292–6. 30. Ernst S. Model of gentamicin-induced nephrotoxicity and its amelioration by calcium and thyroxine. Med Hypotheses 1989;30:195–202. 31. Sataloff J, Wagner S, Menduke H. Kanamycin ototoxicity in healthy men. Arch Otolaryngol 1964; 19:331–7. 32. Moore RD, Lerner SA, Levine DP. Nephrotoxicity and ototoxicity of aztreonam versus aminoglycoside therapy in seriously ill non-neutropenic patients. J Infect Dis 1992;165:683–8. 33. Hatala R, Dinh T, Cook DJ. Once-daily aminoglycoside dosing in immunocompetent adults: a meta-analysis. Ann Intern Med 1996;124:717–25. 34. Bailey TC, Little JR, Littenberg B, et al. A metaanalysis of extended-interval dosing versus


35.

36.

37.

38.

39. 40.

41.

42.

43.

44.

45.

46.

47. 48.

multiple daily dosing of aminoglycosides. Clin Infect Dis 1997;24:786–95. Lerner AM, Cone LA, Jansen W, et al. Randomized controlled trial of the comparative efficacy, auditory toxicity and nephrotoxicity of tobramycin and netilmicin. Lancet 1983;i:1123–6. Lerner SA, Schmitt BA, Seligsohn R, et al. Comparative study of ototoxicity and nephrotoxicity in patients randomly assigned treatment with amikacin and gentamicin. Am J Med 1986; 80 Suppl 5B:S98–S104. The International Antimicrobial Therapy Cooperative Group of the European Organization for Research and Treatment of Cancer. Efficacy and toxicity of single daily doses of amikacin and ceftriaxone versus multiple daily doses of amikacin and ceftazidime for injection in patients with cancer and granulocytopenia. Ann Intern Med 1993;119:584–93. Proctor L, Petty B, Thakor R, et al. A study of potential vestibulotoxic effects of once daily versus thrice daily administration of tobramycin. Laryngoscope 1987;97:1443–9. Helal A. Aminoglycoside eardrops and ototoxicity. Can Med Assoc J 1997;156:1056. Bath AP, Walsh RM, Bance ML, Rutka JA. Ototoxicity of topical gentamicin preparations. Laryngoscope 1999;109:1088–93. Smith BM, Myers MG. The penetration of gentamicin and neomycin into perilymph across the round window membrane. Otolaryngol Head Neck Surg 1979;87:888–91. Harada T, Iwamori M, Nagai Y, et al. Ototoxicity of neomycin and its penetration through the round window membrane into the perilymph. Ann Otol Rhinolaryngol 1986;95:404–7. Hutchin T, Cortopassi G. Proposed molecular and cellular mechanism for aminoglycoside ototoxicity. Antimicrob Agents Chemother 1994;38:2517–20. Henley CM, Schacht J. Pharmacokinetics of aminoglycoside antibiotics in blood, inner ear fluids and tissues and their relationship to ototoxicity. Audiology 1988;27;137–46. Hakashima T, Teranishi M, Hibi T, Kobayashi M. Vestibular and cochlear toxicity of aminoglycosides—a review. Acta Otolaryngol 2000;120:904–11. Song B-B, Schacht J. Variable efficacy of radical scavengers and iron chelators to attenuate gentamicin ototoxicity in guinea pig in vivo. Hear Res 1996;94:987–93. Forge A, Schacht J. Aminoglycosidic antibiotics. Audiol Neurootol 2000;5:3–22. Basile AS, Huang J-M, Zie C, et al. N-Methyl-Daspartate antagonists limit aminoglycoside antibiotic-induced hearing loss. Nat Med 1996;2: 1338–43.

91

49. Fischel-Ghodsian N, Prezant TR, Chaltraw WE, et al. Mitochondrial gene mutation is a significant predisposing factor in aminoglycoside ototoxicity. Am J Otolaryngol 1997;18:173–8. 50. Hamasaki K, Rando RR. Specific binding of aminoglycoside to a human RNA construct based on a DNA polymophism which causes aminoglycoside-induced deafness. Biochemistry 1997;36: 12323–8. 51. Moore RD, Smith CR, Lietman PS. Risk factors for the development of auditory toxicity in patients receiving aminoglycosides. J Infect Dis 1984; 149:23–30. 52. Neu HC, Bendush CL. Ototoxicity of tobramycin: a clinical overview. J Infect Dis 1976;134: S206–18. 53. Gatell JM, Ferran F, Araujo V, et al. Univariate and multivariate analyses of risk factors predisposing to auditory toxicity in patients receiving aminoglycosides. Antimicrob Agents Chemother 1987; 31:1383–7. 54. Govaerts P, Van De Heyning PH, Jorens G, et al. Aminoglycoside induced ototoxicity. Toxicol Lett 1990;53:227–51. 55. Black RE, Lau WK, Weinstein RJ, et al. Ototoxicity of amikacin. Antimicrob Agents Chemother 1976;9:956–61. 56. Manian FA, Stone WJ, Alford R. Adverse antibiotic effects associated with renal insufficiency. Rev Infect Dis 1990;12:236–49. 57. Black FO, Pesznecker SC. Vestibular ototoxicity: clinical considerations. Otolaryngol Clin North Am 1993;26:713–36. 58. Tange RA. Ototoxicity. Adverse Drug React Toxicol Rev 1998;17:75–89. 59. Rybak MJ, Abate BJ, Kang SL, et al. Prospective evaluation of the effect of an aminoglycoside dosing regimen on rates of observed nephrotoxicity and ototoxicity. Antimicrob Agents Chemother 1999;43:1549–55. 60. Nordstrom L, Rinberg H, Cronberg S, et al. Does administration of an aminoglycoside in a single daily dose affect its efficacy and toxicity? J Antimicrob Chemother 1990;25:159–73. 61. Turnidge J. Pharmacodynamics and dosing of aminoglycosides. Infect Dis Clin North Am 2003;17:503–28. 62. Minor LB. Gentamicin-induced bilateral vestibular hypofunction. JAMA 1998;279:541–4. 63. Snavely SR, Hodges GR. The neurotoxicity of antibacterial agents. Ann Intern Med 1984;101: 92–104. 64. Del-Pozo E, Baezem JM. Effects of calcium channel blockers on neuromuscular blockade induced by aminoglycoside antibiotics. Eur J Pharmacol 1986;128:49–54.

92

Systemic Toxicity

65. Wright JM, Collier B. The effects of neomycin upon transmitter release and action. J Pharmacol Exp Ther 1977;200:576–87. 66. Laskin OL, Longstreth JA, Smith CR, et al. Netilmicin and gentamicin multidose kinetics in normal subjects. Clin Pharmacol Ther 1983;34: 644–50. 67. Barza M, Brown RB, Shen D, et al. Predictability of blood levels of gentamicin in man. J Infect Dis 1975;132:165–74. 68. Neu HC. Pharmacology of aminoglycosides. In: Whelton A, Neu HC, editors. The aminoglycosides: microbiology, clinical use and toxicology. New York: Marcel Dekker Inc.; 1982. p. 125–42. 69. Schentag JJ. Aminoglycoside pharmacokinetics as a guide to therapy and toxicology. In: Whelton A, Neu HC, editors. The aminoglycosides: microbiology, clinical use and toxicology. New York: Marcel Dekker Inc.; 1982. p. 143–67. 70. Ali MZ, Goetz MB. A meta-analysis of the relative efficacy and toxicity of single daily dosing versus multiple daily dosing of aminoglycosides. Clin Infect Dis 1997;24:796–809. 71. Barza M, Ioannidis JF, Capelleri JC, Lau J. Single or multiple daily doses of aminoglycosides. Br Med J 1996;312:338–45. 72. Hatala R, Dinh TT, Cook DJ. Single daily dosing of aminoglycosides in immunocompromised adults: a systematic review. Clin Infect Dis 1997;24: 810–5. 73. Ferriolos-Lisart R, Alos-Alminana M. Effectiveness and safety of once-daily aminoglycosides: a metaanalysis. Am J Health Syst Pharm 1996;53: 1141–50. 74. Freeman CD, Strayer AH. Mega-analysis of metaanalysis: an examination of meta-analysis with an

75.

76.

77. 78.

79.

80.

81.

82.

83.

emphasis on once-daily aminoglycoside comparative trials. Pharmacotherapy 1996;16:1093–102. Galloe AM, Gradal N, Christensen HR, Kampman JP. Aminoglycoside single or multiple dosing? A meta-analysis on efficacy and safety. Eur J Clin Pharmacol 1995;48:39–43. Kale-Pradhan PB, Habowski SR, Chase HC, Catranova FC. Once-daily aminoglycosides: a metaanalysis of non-neutropenic and neutropenic adults. J Pharm Technol 1998;14:22–9. Gerberding JL. Aminoglycoside dosing: timing is of the essence. Am J Med 1998;105:256–8. Freeman CD, Nicolau DP, Belliveau PP, et al. Oncedaily dosing of aminoglycosides: review and recommendations for clinical practice. J Antimicrob Chemother 1997;39:677–86. Nicolau DP, Freeman CD, Belliveau PP, et al. Experience with once-daily aminoglycoside program administered to 2184 adult patients. Antimicrob Agents Chemother 1995;39:650–5. Moore RD, Smith CR, Lietman PS. Association of aminoglycoside plasma levels with therapeutic outcome in gram-negative pneumonia. Am J Med 1984;77:657–62. Lerner SA, Schmitt BA, Seligsohn R, Matz GJ. Comparative study of ototoxicity and nephrotoxicity in patients randomly assigned to treatment with amikacin or gentamicin. Am J Med 1986; 80:98–104. Matzke GR, Lucarotti RL, Shapiro HS. Controlled comparison of gentamicin and tobramycin nephrotoxicity. Am J Nephrol 1983;3:11–7. Rybak MJ, Albrecht LM, Boike SC, et al. Nephrotoxicity of vancomycin alone and with an aminoglycoside. J Antimicrob Chemother 1990;25: 679–87.

CHAPTER 9

Mechanisms for Aminoglycoside Ototoxicity: Basic Science Research Jochen Schacht, PhD

The toxic side effects of aminoglycoside antibiotics against the kidney (nephrotoxicity) and the inner ear (ototoxicity) became evident in the first clinical trial,1 just a year after the discovery of the first compound in this class, streptomycin, by Selman Waksman.2 It then took almost 50 years for the mechanisms of this toxicity to be unraveled. The last decade has brought significant advances in our understanding of aminoglycoside action, to the point that we can now suggest a rational hypothesis explaining the death of hair cells induced by this class of antibiotics. Based on the basic research into these mechanisms, successful protective therapies against the ototoxic side effects have already been formulated. This review covers pertinent studies elucidating the biochemical and molecular steps leading to cell death. Although protection from drug-induced hearing loss is covered later (see Chapter 20, “Ototoxic Damage to Hearing: Otoprotective Therapies”), some of the same literature is drawn upon here because much of what is known about the mechanisms of cell death comes from attempts to prevent it.

STRUCTURE AND ACTIVITY OF AMINOGLYCOSIDES Aminoglycoside antibiotics are valued for their bactericidal properties, which are the result of a complex series of events starting at the cell membrane and culminating in the inhibition of bacterial protein synthesis. They are produced by different strains of soil actinomycetes: Streptomyces, which produce aminoglycosides classified by the suffix “-mycin,” and by Micromonospora, which produce aminoglycosides identified by the suffix “-micin.” Various other drugs carry the same suffixes, although they are structurally and pharmacologically unrelated to aminoglycoside antibiotics, for example, the antineoplastic drug bleomycin or the macrolide antibiotic erythromycin. Within the aminoglycoside group, however, the individual drugs, from the first-discovered streptomycin to dihydrostrepto-

mycin, neomycin, tobramycin, kanamycin, gentamicin, and their semisynthetic derivatives, such as netilmicin and amikacin, all share common properties. They have a molecular weight of approximately 300 to 600 daltons and are composed of ring structures, primarily sixmembered saturated cyclitol rings and five- or sixmembered sugar rings that are linked via glycosidic bonds. All aminoglycosides carry hydroxyl and amino groups, giving them their basic character and water solubility. Because of the similarity in their chemical structures and behaviors, results obtained for one aminoglycoside can often be extrapolated to the entire class of these drugs. Although there are exceptions to this rule, evidence suggests that the mechanisms of ototoxicity discussed here are shared by all aminoglycoside antibiotics.

DEFINITION OF OTOTOXICITY When considering molecular mechanisms of ototoxicity, bear in mind that aminoglycoside antibiotics have a multitude of effects on cells of the auditory (and vestibular) system as well as on cells in other tissues of the body. Some of these actions lead to only a temporary impairment of auditory function (at least in experimental animals) and therefore should not be strictly considered “ototoxic.” In particular, the propensity of aminoglycoside antibiotics to displace calcium from its binding sites and to block calcium channels renders any physiological event relying on calcium susceptible to these drugs. Examples are depression of the microphonic output of the lateral line by streptomycin3 or depression of the cochlear microphonics,4 both of which are transient and easily reversed by applying solutions of high K+ or high Ca++ concentrations. Likewise, aminoglycosides can block transduction channels of stereocilia, although this action does not directly lead to hair cell death.5 An acute phase of aminoglycoside intoxication is also seen in the dendrites beneath the inner hair cells (IHCs), which swell in response to aminoglycoside treatment.6 The strict definition of

94

Systemic Toxicity

ototoxicity is that of a permanent auditory threshold shift as a result of the irrevocable loss of outer hair cells (OHCs) and, to some degree, IHCs as well. Regardless of the specific aminoglycoside used or the species of animal studied, ototoxicity progresses from the high frequencies to the low frequencies, corresponding to a loss of hair cells from the base of the cochlea to the apex. Within this base-to-apex gradient, the OHCs seem to be most vulnerable to an aminoglycoside attack, whereas IHCs succumb much later in treatment or at higher doses of the drug.

PHARMACOKINETICS OF AMINOGLYCOSIDE ANTIBIOTICS An obvious prerequisite to death from aminoglycoside toxicity is the uptake of the drug into the cells. Aminoglycoside antibiotics enter the inner ear rapidly following parenteral administration. The drugs can be found in the inner ear within minutes and may reach a plateau as soon as 30 minutes to 3 hours following administration.7 Much has been said about an accumulation of aminoglycosides in inner ear fluids being responsible for the selective death of auditory structures. This, however, is not the case since aminoglycoside antibiotics are present in the fluids (perilymph and endolymph) and the tissues of the inner ear at significantly lower levels than are reached in the serum. Aminoglycosides, however, persist in inner ear tissues long after the drug has essentially been cleared from the bloodstream. Although the half-life of most aminoglycosides in serum is around 3 to 5 hours, the drugs may persist in inner ear fluids for months after treatment, which may account for the delayed hair cell death that can occur after cessation of treatment.7,8 Following systemic application, aminoglycosides can be detected mostly in hair cells and to a lesser extent in supporting cells of the basilar membrane and the tissues of the lateral wall.9–12 In hair cells, the aminoglycosides appear to be initially transported into lysosomal structures near the apical surface, compatible with a receptor-mediated endocytotic internalization.13,14 The precise nature of the aminoglycoside transporter in the inner ear is unresolved. In the kidney, the glycoprotein and mediator of endocytotic uptake, megalin, is strongly supported as an aminoglycoside transporter in proximal tubules. 15 Megalin is also present in the epithelia of the cochlear duct, including lateral wall tissues, but has been claimed to be absent from OHCs, the primary targets of aminoglycoside toxicity.16,17 An alternative component of the cell membrane that may be involved in aminoglycoside uptake is the unconventional myosin-VII-A. Explants from the inner ear of mutant mice lacking this myosin do not take up aminoglycosides and are protected from their toxicity.18 A polyamine-like uptake mechanism has also been discussed.19

More precise information on uptake mechanisms would be valuable in designing therapeutic protection, assuming that exclusion from the cell interior would prevent the toxic actions. However, considering the cellular mechanisms of cell death, uptake is a necessary, but not sufficient, factor for aminoglycoside ototoxicity. Several lines of evidence argue against the correlation of uptake (presence in cells) and toxicity. First, the uptake is widespread in the cochlea, whereas the destruction of hair cells by aminoglycoside antibiotics is rather selective. Second, the concentrations reached in the inner ear by different aminoglycosides do not correlate with the magnitude of their ototoxic potential.20 Third, primarily vestibulotoxic or cochleotoxic drugs do not show a preferential uptake into the structures that they are targeting for destruction.21 Interestingly, even in OHCs, the presence of the drug alone will not lead to an immediate functional impairment or to morphological changes. Although the drugs reach the inner ear within minutes of their systemic application, there is a profound time lag of several days before the death of hair cells.14,22 This time lag points to a complex series of intracellular reactions that slowly reach and surpass a toxic threshold.

CELL DEATH Apoptosis and necrosis are two classically accepted pathways of cell death. More recent literature, however, includes subtle distinctions from these paradigms, such as apoptosis-like cell death and necrosis-like cell death, which can be distinguished by morphological features and the associated biochemical reactions. 23 These subtleties have not yet been addressed in aminoglycoside ototoxicity. Apoptosis is most often mentioned as the prevailing (or even only) form of cell death. It is the irreversible end stage of programmed cell death that removes damaged cells without producing inflammation. Consistent with such a way of death, the affected hair cells are removed from the organ of Corti and replaced by expansion of their supporting cells, which form a scar-like closure of the reticula lamina.24 However, some caution is necessary in interpreting these results. Most of the studies on mechanisms of cell death have been carried out in model systems such as vestibular or organ of Corti explants. It remains an open question whether the acute exposure of cells in tissue or organ culture to the drugs is indeed a precise reflection of the slow cell death that occurs in chronic systemic application of aminoglycosides. Apoptosis indeed seems to be a major pathway of cell death in tissue culture, reported for both cochlear and vestibular hair cells and based on morphological features of apoptosis, such as condensed and marginated chromatin.25–27 Necrotic events are mentioned not nearly as much in the context of aminoglycoside toxicity. However, in vivo studies that report apoptosis also concede an

Mechanisms for Aminoglycoside Ototoxicity: Basic Science Research

undetermined amount of necrotic events taking place in the cochlea.28,29 It seems most likely that different conditions of intoxication (eg, in vitro vs in vivo, low vs high dose of drugs) will have a different balance of apoptotic and necrotic pathways. Since apoptosis is a controlled event of cell disintegration, it proceeds via defined pathways. These pathways are triggered by an initial noxious stimulus that activates signaling cascades of sequential activation of protein kinases, which ultimately activate transcription factors that then will translocate into the nucleus. When attached to their binding regions on deoxyribonucleic acid (DNA), these transcription factors will initiate the expression of specific genes. In apoptotic pathways, the resulting proteins and enzymes will contribute to the demise of the cell. Conversely, signaling pathways exist for restitution of homeostasis and rescue of cells from such noxious stimuli. A well-characterized pathway of apoptosis proceeds via the activation of c-Jun N-terminal protein kinase (JNK), a member of the mitogen-activated protein kinase family that is involved in the phosphorylation of precursors to transcription factors. The activation of JNK can be strongly inferred from studies that show that inhibitors of this pathway will rescue hair cells from the toxic effects of aminoglycoside antibiotics.30 Although apoptotic pathways are well defined, there are complex overlapping and interconnected pathways that eventually determine cell fate. One of the committing steps in another sequence to cell death is the activation of caspases, specific proteases. Caspases are involved in aminoglycoside-induced cell death since inhibitors of caspase activation protect from the toxic effects of the drugs.31 Although events such as the activation of JNK or caspases are essential to apoptotic cell death, these events are in turn regulated by upstream pathways that respond to the initial noxious stimulus (Figure 9-1). So-called small guanosine triphosphatases (GTPases) are common activators of JNK, and it appears that the subfamily of Rho-GTPases plays a role in aminoglycoside toxicity signaling. When inhibitors of Rho-family GTPases are added to tissue cultures treated with gentamicin, these cultures are protected from ototoxicity, implying this family of GTPases as an upstream activator consistent with the existence of JNK signaling pathways.32 Likewise, permeability changes in mitochondria precede the activation of caspases, and such permeability transitions have been shown in response to aminoglycosides in cultured cells.33 Consideration of signaling pathways leads to the question: what are the initial noxious events that trigger these apoptotic pathways? Compelling evidence leads us to believe that free radicals—reactive oxygen species (ROS) and possibly reactive nitrogen species (RNS)—are the ultimate triggers of aminoglycosideinduced cell death.

95

ROS

Change in redox status GSH depletion

Activation of signaling pathways

Mitochondria

JNK

Caspases

Gene transcription

APOPTOSIS [ NECROSIS ] Figure 9-1 Putative cell death pathways in aminoglycoside

ototoxicity. The noxious formation of reactive oxygen species (ROS; free radicals) exhausts the cellular antioxidant defenses and changes the redox status of the cell, most notably the glutathione (GSH) balance. This triggers the activation of signaling pathways (eg, Rho guanosine triphosphatases) that set an apoptotic machinery in motion. Represented are the pathways involving mitochondria and caspase activation and c-Jun N-terminal protein kinase (JNK) leading to gene transcription. The stippled arrow is a reminder that additional pathways (for example, to necrosis) exist and that pathways frequently cross-talk and influence each other’s activity.

FREE RADICAL FORMATION AS THE CAUSE OF CELL DEATH Free radicals such as superoxide, hydroxyl, and nitric oxide are naturally produced compounds that distinguish themselves from other cell metabolites by the presence of an unpaired electron in their structure. This unpaired electron renders them chemically reactive, and in the worst-case scenario, free radicals can attack cell constituents ranging from cell membranes to proteins to DNA, thereby causing irreparable damage and, ultimately, cell death. Conversely, ROS and RNS serve distinct and important roles in the physiologic control of cell function.34 They function as intermediates in synthetic reactions (eg, for prostaglandins), signaling molecules in homeostatic reactions to balance cellular metabolism, or as second messengers like nitric oxide. Their toxic effects are held in check by naturally occurring antioxidants in the cell, notably by glutathione, and by enzymes that inactivate free radicals (or their proradicals) such as superoxide dismutase, catalase, and glutathione peroxidase. If enhanced production of free radicals by cellular overstimulation or by drugs exceeds the capacity of the antioxidant system,

96

Systemic Toxicity

free radical damage can ensue. Oxidative stress as a result of an excess of free radicals has been associated with a variety of pathologic conditions, including artheriosclerosis, inflammation, ultraviolet ray damage, certain cancers, and even aging. The suggestion that aminoglycoside antibiotics produce free radicals was made relatively early based on studies in renal mitochondria35 and in the inner ear.36 However, subsequent failure to link free radicals to nephrotoxicity 37or to ototoxicity 38 laid this hypothesis initially to rest. Only in recent years has evidence accumulated both from in vivo and in vitro experiments to clearly establish a causal relationship between free radicals and ototoxicity. Discussions have centered on ROS derived from superoxide and RNS derived from nitric oxide.

NITRIC OXIDE IN AMINOGLYCOSIDE OTOTOXICITY Nitric oxide is a physiologically important second messenger and neuromodulator but also the potential precursor of the highly destructive peroxynitrite radical. The enzyme responsible for its synthesis, nitric oxide synthase, can be activated by calcium influx through N-methyl-D-aspartate (NMDA) receptors. Such receptors are associated with the afferent system in the cochlea, as they respond to glutamate, the neurotransmitter at the IHC synapse. NMDA receptors, specifically those of the NR1A/NR2B type, are sensitive to modulation and stimulation by polyamines such as spermine. Aminoglycosides share some basic features with polyamines, and polyamine–aminoglycoside interactions are well documented in many systems, including the inner ear, where polyamines interfere with aminoglycoside uptake and inhibit cochlear ornithine decarboxylase. 19,39 Overstimulation of NMDA receptors and the resulting “excitotoxicity” has been postulated as a mechanism of aminoglycoside ototoxicity.40 Although this is an interesting concept, the original study is difficult to interpret because of confounding factors. The hypothesis of a polyaminelike action responsible for aminoglycoside ototoxicity was supported by the observation that NMDA receptor antagonists prevented aminoglycoside ototoxicity in vivo. The design of the study, however, did not control for the vehicle used for the drugs, dimethyl sulfoxide, which itself is a radical scavenger. The vehicle alone subsequently was shown to convey protection against aminoglycoside-induced auditory damage.41 Another difficulty in interpreting this hypothesis is the fact that excitotoxicity manifests itself as a transient swelling of afferent nerve endings but does not affect OHCs.42 Swollen afferent dendrites beneath the IHCs were indeed observed following administration of the aminoglycoside amikacin,6 but the authors considered this pathology to be an acute phase of insult whereas

the chronic phase resulted in complete loss of OHCs. The NMDA receptor antagonists MK-801 gave only inefficient protection. Further, when the effects of a series of aminoglycosides on NR1A/NR2B receptors were probed, no direct relationship emerged between the potency of an aminoglycoside and its ototoxicity.43 Thus, although nitric oxide production and swelling of dendrites may be part of the overall pattern of aminoglycoside ototoxicity, they probably cannot be considered a direct cause of hair cell loss.

ROS IN OTOTOXICITY Following the standoff in the early 1980s on the question of free radicals and ototoxicity,36,38 evidence from studies in vitro, in cell culture, and in vivo began to converge to the notion that ROS are causally related to aminoglycoside-induced damage. Manipulations of the endogenous cochlear antioxidant system were able to change the extent of aminoglycoside-induced damage. Depletion of glutathione in inner ear tissues enhanced, and restoration of glutathione levels attenuated aminoglycoside-induced ototoxicity in vivo.44–46 Complementing these indirect approaches were demonstrations that free radicals emerged in inner ear explants following the addition of aminoglycosides in explants of the inner ear.47,48 One of the difficulties in accepting the involvement of free radicals in the toxic actions of aminoglycosides was the absence of a rational explanation of how these drugs can catalyze the formation of ROS. However, in vitro experiments soon showed that aminoglycosides can form complexes with transition metals (with iron and copper in particular). These complexes are redox active (ie, can form free radicals49,50) in a reaction in which molecular oxygen is reduced to the superoxide radical at the expense of an electron donor such as arachidonic acid.51 Following the generation of superoxide, iron-catalyzed Fenton-type reactions can then promote the formation of other, more aggressive free radicals such as hydroxyls. The ability to catalyze free radical generation in this manner is shared by a variety of aminoglycoside antibiotics.52 Products of lipid peroxidation can be found in the inner ear following aminoglycoside treatment,53 which is compatible with the idea that arachidonic acid or a homologous polyunsaturated fatty acid participates in the reduction of oxygen to superoxide. Arachidonic acid, as an electron donor, also links these new observations to earlier findings that aminoglycoside antibiotics have a high binding affinity for polyphosphoinositide lipids, which correlates extremely well with their ototoxic potential. 54,55 Polyphosphoinositides contain arachidonic acid as the esterified lipid in their 2′-position. The ability of aminoglycosides to bind both a transition metal and a lipid would bring the redox catalyst iron and the electron donor arachidonate in close

Mechanisms for Aminoglycoside Ototoxicity: Basic Science Research O2 PUFA

e-

Aminoglycoside [Fe 2+/3+]

lipid peroxides •

NO•

superoxide O2

SOD

H2O + 1/2 O2

H2 O2

ONOO peroxynitrite

Fenton reaction [Fe2+]

hydroxyl radical •OH

Figure 9-2 Formation of free radicals by aminoglycoside antibiotics. The reduction of molecular oxygen to the superoxide radical is catalyzed by aminoglycosides and iron. The reducing electrons are derived from polyunsaturated fatty acids (PUFA; eg, arachidonic acid) yielding lipid peroxides, which can engage in a separate branch of free radical reactions. The slowly reacting superoxide radical is effectively dismutated to hydrogen peroxide by superoxide dismutase (SOD). Hydrogen peroxide, a proradical, can be inactivated by catalase to water and oxygen. However, nonenzymatic, iron-catalyzed “Fenton” reactions compete with catalase and can generate the aggressive hydroxyl radicals. If nitric oxide (NO) were produced by aminoglycosides it could react with excess superoxide to the very destructive peroxynitrite radical. Note that reactions are not written with all steps or in full stoichiometry.

proximity in a ternary complex, enabling an efficient electron transfer to oxygen and formation of superoxide (Figure 9-2). These or very similar mechanisms must also operate in vivo. Transgenic mice overexpressing superoxide dismutase (SOD), the enzyme that inactivates superoxide, show significantly lower threshold shifts after kanamycin treatment than do their nontransgenic littermates with normal levels of SOD.56 Likewise, in organotypic cultures from neonatal mice, the SOD mimetic M40403 significantly protected against damage induced by gentamicin.57 Conversely, in normal animals the levels of antioxidant enzymes such as superoxide dismutase, catalase, and glutathione peroxidase inversely correlated with their hearing loss after amikacin treatment.53 Adding persuasive evidence is the success of antioxidant therapy, which has been a highly successful interventional strategy against aminoglycoside ototoxi-

97

city. Antioxidants, when coadministered with aminoglycoside antibiotics, will significantly attenuate the expected functional and morphologic damage.58 This protective effect is shared by a wide variety of iron chelators, such as deferoxamine or dihydroxybenzoic acid, or radical scavengers such as mannitol, methionine, or lipoic acid.59–62 The only common denominator of these protective agents is their iron-chelating or antioxidant capacity, strongly suggesting free radical formation as a causative action in aminoglycoside ototoxicity. Most intriguing, from a therapeutic point of view, is the finding that salicylic acid is an effective protectant. Salicylate is the active ingredient of aspirin.63 In line with apoptotic and necrotic cell death in aminoglycoside ototoxicity is the fact that free radicals are powerful triggers of cell death. Further, a free radical–based mechanism could even explain the preferential destruction of OHCs and the base-to-apex gradient in the progression of hair cell damage. OHCs are intrinsically more sensitive to damage by free radicals than are IHCs and supporting cells, and along the cochlear spiral, basal outer hair cells are more sensitive than apical cells.64 Finally, we need to consider that aminoglycosides are both potentially cochleotoxic and vestibulotoxic, mostly against type 1 vestibular hair cells. Less attention has been paid to the potential underlying mechanisms in the vestibular system than in the cochlea. We can, however, safely assume that similar mechanisms of toxicity are operating in both parts of the inner ear. First, the mechanism of free radical formation by aminoglycosides has been established in vitro and is therefore independent of a specific tissue. Second, the protective antioxidant therapies that prevent destruction of cochlear hair cells also prevent the destruction of vestibular hair cells.65

SUMMARY Despite the discovery of aminoglycosides as a distinct antimicrobial class over 60 years ago, basic science research in aminoglycoside toxicity has only recently allowed us to formulate a rational hypothesis that can explain hair cell death induced by an ototoxic agent. • Because of similarities in their chemical structures and behaviors, results obtained from one aminoglycoside can usually be extrapolated to the entire group of these drugs. Overall, the phenomenon of aminoglycoside cochleotoxicity has been better studied than aminoglycoside vestibulotoxicity. • Aminoglycosides can cause both temporary and permanent impairment of cellular function. Temporary impairment is thought to occur from the competitive blockade of calcium channels required for the generation of receptor or action

98

Systemic Toxicity

potentials. Permanent impairment occurs when their uptake into the hair cells ultimately results in cell death from apoptosis and possibly cellular necrosis mechanisms. • The uptake of aminoglycosides into the hair cell is a necessary but not in itself a sufficient cause for ototoxicity. Recent basic science research has supported the formation of free radicals— ROS—derived from iron-aminoglycoside complexes in the cell as the ultimate triggers of aminoglycoside-induced cellular death. Free radical formation has been used convincingly to explain the preferential toxicity for OHCs versus IHCs and the base-to-apex gradient progression of hair cell damage in cochlear toxicity. • Future treatments using antioxidants may have a role in preventing cellular injury and death from aminoglycoside toxicity.

9.

10.

11.

12.

13.

ACKNOWLEDGMENTS This research is supported by grant DC03685 from the National Institute of Health. The author thanks Ms Andra Talaska for valuable comments on the manuscript.

14.

15.

REFERENCES 1. Hinshaw HC, Feldman WH. Streptomycin in treatment of clinical tuberculosis: a preliminary report. Mayo Clin Proc 1945;20:313–8. 2. Schatz A, Bugie E, Waksman SA. Streptomycin, a substance exhibiting antibiotic activity against gram-positive and gram-negative bacteria. Proc Soc Exp Biol Med 1944;55:66–9. 3. Wersäll J, Flock Å. Suppression and restoration of the microphonic output from the lateral line organ after local application of streptomycin. Life Sci 1964;3:1151–5. 4. Takada A, Schacht J. Calcium antagonism and reversibility of gentamicin-induced loss of cochlear microphonics in the guinea pig. Hear Res 1982;8:179–86. 5. Kossl M, Richardson GP, Russell IJ. Stereocilia bundle stiffness: effects of neomycin, A23187 and concanavalin A. Hear Res 1990;44:217–30. 6. Duan M, Agerman K, Ernfors P, Canlon B. Complementary roles of neurotrophon 3 and a Nmethyl-D-aspartate antagonist in the protection of noise and aminoglycoside-induced ototoxicity. Proc Nat Acad Sci U S A 2000;97:6939–40. 7. Tran Ba Huy P, Bernard P, Schacht J. Kinetics of gentamicin uptake and release in the rat: comparison of inner ear tissues and fluids with other organs. J Clin Invest 1986;77:1492–500. 8. Aran JM, Dulon D, Hiel H, et al. L’ototoxicité d’aminosides: resultats recent sur la captation et la clairance de la gentamicine par les cellules-

16.

17.

18.

19.

20.

21.

22

23.

sensorielles du limacon osseux. Rev Laryngol Otol Rhinol (Bord) 1993;114:125–8. Balogh K, Hiraide F, Ishii D. Distribution of radioactive dihydrostreptomycin in the cochlea. Ann Otol Rhinol 1970;79:641–52. Hayashida T, Normura Y, Iwamori M, et al. Distribution of gentamicin by immunofluorescence in the guinea pig inner ear. Arch Otorhinolaryngol 1985;242:257–64. De Groot JC, Meeuwsen F, Ruizendaal WE, Veldman JE. Ultrastructural localization of gentamicin in cochlea. Hear Res 1990;50:35–42. Hiel H, Schamel A, Erre JP, et al. Cellular and subcellular localisation of tritiated gentamicin in the guinea pig cochlea following combined treatment with ethacrynic acid. Hear Res 1992;57:157–65. Darrouzet J, Guilhaume A. Ototoxicité de la kanamycine au jour le jour. Étude expérimentale en microscopie électronique. Rev Laryngol Otol Rhinol (Bord) 1974;95:601–21. Hashino E, Sheor M. Endocytosis of aminoglycoside antibiotics in sensory hair cells. Brain Res 1995;704:135–40. Moestrup SK, Cui S, Vorum H, et al. Evidence that epithelial glycoprotein 330/megalin mediates uptake of polybasic drugs. J Clin Invest 1995;96: 1404–13. Ylikoski J, Pirvola U, Zhai S, et al. Aminoglycoside ototoxicity: high affinity receptors are expressed in secretory epithelia. Aud Neurosci 1997;3:415–24. Mizuta K, Saito A, Watanabe T, et al. Ultrastructural localization of megalin in the cochlear duct. Hear Res 1999;129:83–91. Richardson GP, Forge A, Kros CJ, et al. Myosin VIIA is required for aminoglycoside accumulation in cochlear hair cells. J Neurosci 1997;17:9506–19. Williams SE, Smith DE, Schacht J. Characteristics of gentamicin uptake in the isolated crista ampullaris of the inner ear of the guinea pig. Biochem Pharmacol 1987;36:89–95. Ohtsuki K, Ohtani I, Aikawa T, et al. The ototoxicity and the accumulation in the inner ear fluids of the various aminoglycoside antibiotics. Ear Res Jpn 1982;13:85–7. Dulon D, Aran J-M, Zajic G, Schacht J. Comparative pharmacokinetics of gentamicin, netilmicin and amikacin in the cochlea and the vestibule of the guinea pig. Antimicrob Agents Chemother 1986;30:96–100. Hiel H, Erre J, Aurousseau C, et al. Gentamicin uptake by cochlear hair cells precedes hearing impairment during chronic treatment. Audiology 1993;32:78–87. Leist M, Jäättelä M. Four deaths and a funeral: from caspases to alternative mechanisms. Nat Rev Mol Cell Biol 2001;2:1–10.

Mechanisms for Aminoglycoside Ototoxicity: Basic Science Research

24. Raphael Y, Altschuler RA. Scar formation after drug-induced cochlear insult. Hear Res 1991; 51:173–84. 25. Forge A. Outer hair cell loss and supporting cell expansion following chronic gentamicin treatment. Hear Res 1985;19:171–82. 26. Brown AM, McDowell B, Forge A. Acoustic distortion products can be used to monitor the effects of chronic gentamicin treatment. Hear Res 1989; 42:143–56. 27. Li L, Nevill G, Forge A. Two modes of hair cell loss from the vestibular sensory epithelia of the guinea pig. J Comp Neurol 1995;355:405–17. 28. Nakagawa T, Yamane H, Takayama M, et al. Apoptosis of guinea pig cochlear hair cells following aminoglycoside treatment. Eur Arch Otorhinolaryngol 1998;255:127–31. 29. Ylikoski J, Xing-Qun L, Virkkala J, Pirvola U. Blockade of c-Jun N-terminal kinase pathway attenuates gentamicin-induced cochlear and vestibular hair cell death. Hear Res 2002;166: 33–43. 30. Pirvola U, Xing-Qun L, Virkkala J, et al. Rescue of hearing, auditory hair cells, and neurons by CEP1347/KT-7515, an inhibitor of c-Jun N-terminal kinase activation. J Neurosci 2000;20:43–50. 31. Matsui JL, Haue A, Huss D, et al. Caspase inhibitors promote vestibular hair cells survival and function after aminoglycoside treatment in vivo. J Neurosci 2003;23:1111–22. 32. Bodmer D, Brors D, Pak K, et al. Rescue of auditory hair cells from aminoglycoside toxicity by Clostridium difficile toxin B, an inhibitor of the small GTPases Rho/Rac/Cdc42. Hear Res 2002;172:81–6. 33. Dehne N, Rauen U, de Groot H, Lautermann J. Involvement of the mitochondrial permeability transition in gentamicin ototoxicity. Hear Res 2002;169:47–55. 34. Dröge W. Free Radicals in the physiological control of cell function. Physiol Rev 2001;82:47–95. 35. Walker PD, Shah SV. Gentamicin enhanced production of hydrogen peroxide by renal cortical mitochondria. Am J Physiol 1987;253:C495–99. 36. Pierson MG, Møller AR. Prophylaxis of kanamycin-induced ototoxicity by a radioprotectant. Hear Res 1981;4:79–87. 37. Ramsammy LS, Josepovitz C, Ling K-Y, et al. Failure of inhibition of lipid peroxidation by vitamin E to protect against gentamicin nephrotoxicity in the rat. Biochem Pharmacol 1987;36:2125–32. 38. Bock GR, Yates GK, Miller JJ, Moorjani P. Effects of N-acetylcysteine on kanamycin ototoxicity in the guinea pig. Hear Res 1983;9:255–62. 39. Henley ChM, Gerhardt HJ, Schacht J. Inhibition of inner ear ornithine decarboxylase by neomycin invitro. Brain Res Bull 1987;19:695–8.

99

40. Basile AS, Huang JM, Xie C, et al. N-Methyl-Daspartate antagonists limit aminoglycoside antibiotic-induced hearing loss. Nat Med 1996;12: 1338–43. 41. Sha SH, Schacht J. Are aminoglycoside antibiotics excitotoxic? Neuroreport 1998;9:3893–5. 42. Puel JL, Pujol R, Tribillac F, et al. Excitatory amino acid antagonists protect cochlear auditory neurons from excitotoxicity. J Comp Neurol 1994;341: 241–56. 43. Harvey SC, Skolnick P. Polyamine-like actions of aminoglycosides at recombinant N-methyl- D aspartate receptors. J Pharmacol Exp Ther 1999; 291:285–91. 44. Hoffman DW, Whitworth CA, Jones-King KL, Rybak LP. Potentiation of ototoxicity by glutathione depletion. Ann Otol Rhinol Laryngol 1988;97:36–41. 45. Garetz SL, Altschuler RA, Schacht J. Attenuation of gentamicin ototoxicity by glutathione in the guinea pig in vivo. Hear Res 1994;77:81–7. 46. Lautermann J, McLaren J, Schacht J. Glutathione protection against gentamicin ototoxicity depends on nutritional status. Hear Res 1995;86:15–24. 47. Clerici WJ, Hensley K, DiMartino DL, Butterfield DA. Direct detection of ototoxicant-induced reactive oxygen species generation in cochlear explants. Hear Res 1996;98:116–24. 48. Hirose K, Hockenberry DN, Rubel EW. Reactive oxygen species in chick hair cells after gentamicin exposure in vitro. Hear Res 1997;104:1–14. 49. Priuska EM, Schacht J. Formation of free radicals by gentamicin and iron and evidence for an iron/gentamicin complex. Biochem Pharmacol 1995;50:1749–52. 50. Priuska EM, Clark K, Pecoraro V, Schacht J. NMR spectra of iron-gentamicin complexes and the implications for aminoglycoside toxicity. Inorg Chim Acta 1998;273:85–91. 51. Sha SH, Schacht J. Formation of reactive oxygen species following bioactivation of gentamicin. Free Radic Biol Med 1999;26:341–7. 52. Sha SH, Schacht J. Formation of free radicals by aminoglycoside antibiotics. Hear Res 1999;128: 112–8. 53. Klemens JJ, Meech RP, Hughes LF, et al. Antioxidant enzyme levels inversely covary with hearing loss after amikacin treatment. J Am Acad Audiol 2003;14:134–43. 54. Schacht J. Isolation of an aminoglycoside receptor from guinea pig inner ear tissues and kidney. Arch Otorhinolaryngol 1979;224:129–34. 55. Lodhi S, Weiner ND, Mechigian I, Schacht J. Ototoxicity of aminoglycosides correlated with their action on monomolecular films of polyphosphoinositides. Biochem Pharmacol 1980;29:597–601.

100

Systemic Toxicity

56. Sha SH, Zajic G, Epstein CJ, Schacht J. Overexpression of SOD protects from kanamycin-induced hearing loss. Audiol Neurootol 2001;6:117–23. 57. McFadden SL, Ding D, Salvemini D, Salci RJ. M40403, a superoxide dismutase mimetic, protects cochlear hair cells from gentamicin but not cisplatin toxicity. Toxicol Appl Pharmacol 2003;186:46–54. 58. Song BB, Schacht J. Variable efficacy of radical scavengers and iron chelators to attenuate gentamicin ototoxicity in guinea pig in vivo. Hear Res 1996;94:87–93. 59. Song BB, Anderson DJ, Schacht J. Protection from gentamicin ototoxicity by iron chelators in guinea pig in vivo. J Pharmacol Exp Ther 1997;282: 369–77. 60. Conlon BJ, Smith DW. Attenuation of neomycin ototoxicity by iron chelation. Laryngoscope 1998; 108:284–7.

61. Conlon BJ, Aran J-M, Erre J-P, Smith DW. Attenuation of aminoglycoside-induced cochlear damage with the metabolic antioxidant α-lipoic acid. Hear Res 1999;128:40–4. 62. Sha SH, Schacht J. Antioxidants attenuate gentamicin-induced free-radical formation in vitro and ototoxicity in vivo: D-methionine is a potential protectant. Hear Res 2000;142:34–40. 63. Sha SH, Schacht J. Salicylate attenuates gentamicininduced ototoxicity. Lab Invest 1999;79:807–13. 64. Sha SH, Taylor R, Forge A, Schacht J. Differential vulnerability of basal and apical hair cells is based on intrinsic susceptibility to free radicals. Hear Res 2001;155:1–8. 65. Song BB, Sha SH, Schacht J. Iron chelators protect from aminoglycoside-induced cochleo- and vestibulotoxicity in guinea pig. Free Radic Biol Med 1998;25:189–95.

CHAPTER 10

Macrolides Andrew R. Scott, BM, BS, MPhil, FRCS(ORL-HNS), and John A. Rutka, MD, FRCSC

Erythromycin is a macrolide antibiotic discovered in 1952 and used widely to treat a variety of bacterial infections, including upper and lower respiratory tract infections. 1 Ototoxicity has been reported with erythromycin and more recently with some of the newer macrolide antibiotics, including azithromycin and clarithromycin.

ERYTHROMYCIN Erythromycin obtained approval from the US Food and Drug Administration (FDA) in June 1964. Its mechanism of action is to inhibit protein synthesis in sensitive bacteria by reversibly binding to 50S ribosomal subunits. Nucleic acid synthesis is not affected.2 The first cases of erythromycin ototoxicity were reported in the literature by Mintz and colleagues in 1973.3 They described two women treated for pneumonia with intravenous erythromycin who developed a noticeable hearing loss. Audiometry revealed a 50 to 55 dB sensorineural hearing loss (SNHL). Erythromycin was discontinued, and the patients were switched to a cephalosporin antibiotic. In both cases the hearing returned to normal. A further case of reversible hearing loss in a woman treated with erythromycin was reported by Eckman and colleagues in 1975.4 In their case the erythromycin had been administered orally; again a normal audiogram was obtained on cessation of treatment. No audiogram was obtained during the period of perceived hearing loss. Further reports of reversible hearing loss soon followed. Karmody and Weinstein reported three cases.5 However, in one patient, the recovery was probably incomplete.6 Quinan and McCabe reported two cases, one of whom also complained of vertigo in addition to tinnitus and hearing loss.7 This report appears to be the first to document a possible vestibulotoxic as well as a cochleotoxic effect. Van Marion and colleagues reported two further cases of reversible hearing loss, and Lornoy and Steyaert reported one patient who also experienced vestibular symptoms.8,9 Although it had

been pointed out that the ototoxic effects of erythromycin could not necessarily be assumed to be reversible on cessation of therapy, this was the general impression that emerged from these early case reports.7 Findings of irreversible ototoxicity were first documented by Levin and Behrenth in 1986.10 Their patient was a 54-year-old woman who had experienced hearing loss and tinnitus after four doses of intravenous erythromycin (1 g every 6 hours). An audiogram taken 24 hours later demonstrated a sloping SNHL, worse in the high frequencies down to 65 dB at 8 kHz. After 1 year the audiogram had slightly improved to 45 dB at 8 kHz. Her tinnitus persisted. There was, however, as with many of these case reports, no pretreatment audiogram available for comparison, making it truly difficult to document any degree of reversibility from erythromycin toxicity. Dylewski also reported a case of irreversible hearing loss attributed to erythromycin. 11 Audiograms were performed at 5 and 23 weeks with no improvement. Further case reports of erythromycin ototoxicity without other obvious contributory factors have continued to be reported.12–25

PREDISPOSING FACTORS FOR ERYTHROMYCIN OTOTOXICITY Reports of erythromycin ototoxicity with associated or possible predisposing factors have also emerged. In 1979 Mery and Kanfer described ototoxicity in three patients with renal impairment.26 Audiograms were not provided, but a symmetrical high-frequency SNHL was described. Thompson and colleagues reported a case of reversible SNHL in an 18-year-old woman with severe renal failure.27 Taylor and colleagues reported on two patients who received erythromycin while on peritoneal dialysis. 28 In addition to hearing loss, one patient developed diplopia, and the other slurred speech and confusion. These symptoms recovered on cessation of erythromycin therapy and raised the question of a central nervous system effect of erythromycin.6 Further cases of erythromycin ototoxicity in

102

Systemic Toxicity

renal failure have been reported.29–31 Kanfer and colleagues carried out a study on the pharmacokinetics of erythromycin in renal failure.32 They compared chronic renal failure patients with normal subjects receiving a single dose of 1 g of erythromycin daily. Although the number of study participants was small, there was a trend toward higher maximum serum concentrations and increased overall bioavailability of erythromycin in the renal failure patients, which might predispose them to the toxic effects. Hepatic impairment and concomitant use of other ototoxic drugs have also been suggested to increase the potential for erythromycin ototoxicity. Umstead and Neuman reported two cases of reversible hearing loss in cancer patients with hepatic impairment who received erythromycin.33 Both were receiving other potentially ototoxic agents and both also had acute psychotic reactions. Although there are clearly several confounding factors, the hearing loss and psychoses appeared temporally related to the erythromycin treatment. Wallach and colleagues reported a case of erythromycininduced reversible ototoxicity in a patient who had suffered cisplatin-induced high-frequency hearing loss 9 months previously.34 Another group of patients who may have increased susceptibility to erythromycin toxicity are transplant recipients. Vasquez and colleagues performed a retrospective case note review of renal transplant patients treated with erythromycin for pneumonia. 35 They found hearing loss in 11 (32%) of 34 treatment courses of erythromycin. They also demonstrated a dose-related toxic effect. Hearing loss occurred more frequently in patients who received the higher dose of 4 g daily compared with those who received 2 g daily (53% vs 16%). Hearing loss was also significantly more common in those patients receiving longer courses (9.6 ± 4.7 days) of erythromycin compared with those treated with a shorter course (5.7 ± 3.6 days). Three cases of erythromycin ototoxicity have also been reported in liver transplant recipients.36

RISK FACTORS FOR ERYTHROMYCIN OTOTOXICITY Several papers have attempted to analyze risk factors for erythromycin ototoxicity. Hugues and colleagues reported on 30 patients who received erythromycin and in whom audiograms had been performed before and after treatment.37 Age and concurrent medical problems were recorded. Of interest, no hearing loss was found in any patient, including those with renal or hepatic problems. Swanson and colleagues performed a prospective case-control study in patients treated for community-acquired pneumonia.38 Thirty patients receiving erythromycin were compared with 15 patients receiving other antibiotics as the control group. Sequential audiograms were recorded by an audiologist

blinded to the treatment received. Serum erythromycin concentrations were measured for patients receiving erythromycin. They found evidence of ototoxicity in 5 of the 30 patients in the erythromycin group compared with none in the control group. Ototoxicity occurred only in patients receiving 4 g of erythromycin daily and not in those receiving 2 g daily. Ototoxicity was reversible and related to serum erythromycin concentration. The available evidence suggests that erythromycin ototoxicity is dose dependent and that patients with renal impairment or who have had a transplant are at increased risk. Hepatic dysfunction, advanced age, and female sex may also be risk factors (see also case reviews by Haydon and colleagues, Brummett, and Sacristan and colleagues).6,16,25 Azithromycin Azithromycin is a new-generation macrolide antibiotic that is often used for treating infections related to human immunodeficiency virus (HIV), including those caused by Mycobacterium avium complex (MAC). Azithromycin obtained FDA approval in November 1991. It has a similar mode of action to erythromycin, namely to bind to the bacterial 50S ribosomal subunit and inhibit protein synthesis.2 Wallace and colleagues reported three cases of ototoxicity in 21 HIV patients with MAC infections receiving prolonged courses of 500 mg of azithromycin daily.39 Symptoms occurred at 4 to 12 weeks into treatment and resolved within 2 to 4 weeks after cessation of therapy. Tseng and colleagues identified eight cases of potential azithromycin ototoxicity in a retrospective review of 46 patients from their HIV clinic who had received azithromycin (600 mg daily for a mean of 7.6 weeks). 40 Onset of ototoxicity followed 2 to 20 weeks of treatment and recovery occurred 2 to 11 weeks after discontinuing the azithromycin. Lo and colleagues reported a further case of hearing loss in a 35-year-old man with HIV and Mycobacterium infection treated with azithromycin.41 There were many possible confounding or contributory factors in these reports, including renal disease, hepatic disease, and concurrent drug therapy, and audiograms were not performed on all patients. However, these papers suggest a potential for ototoxicity when azithromycin is used at these higher doses for prolonged periods in this population of patients. There have also been reports of hearing loss in patients without HIV infection following azithromycin. Brown and colleagues studied adverse events in 39 elderly patients treated with 600 mg of azithromycin daily for mycobacterial lung disease.42 Among other effects, 10 (26%) of the patients developed hearing loss that was confirmed on baseline and repeat audiology. When the daily dose was halved to 300 mg in these patients, the adverse effects were reported to resolve.

Macrolides

Bizjak and colleagues described a case of complete deafness in a 47-year-old woman following an 8-day course of intravenous azithromycin for communityacquired pneumonia.43 They cautioned that intravenous dosing regimens can result in much higher serum concentrations of azithromycin and untoward toxic effects. Ress and Gross reported a case of irreversible SNHL in a 39-year-old woman.44 She was prescribed a 5-day course of azithromycin (500 mg orally for 1 day followed by 250 mg orally for 4 days) for a urinary tract infection. Her medical history was otherwise unremarkable, and she was not taking other medications. She experienced bilateral tinnitus after the first dose, which worsened after the second dose and was accompanied at that stage by a subjective hearing loss. The azithromycin was subsequently stopped. Audiometry demonstrated an asymmetrical high-frequency SNHL that had not improved on retesting 12 months later. No pretreatment audiogram was available. Magnetic resonance imaging (MRI) was normal. Mamikoglu and Mamikoglu replied to this paper with a letter to the editor in which they described two further cases of possible azithromycin ototoxicity occurring after short-term low-dose therapy in non–HIV-infected patients.45 The first patient developed a 70 dB loss in one ear after 3 days of 500 mg (once daily) of azithromycin. Hearing improved to 30 dB after 5 days in conjunction with steroid and antiviral therapy. Their second patient received treatment for an acute otitis media with a 5-day course of azithromycin (500 mg on day 1 followed by 250 mg daily for 4 days) when she developed tinnitus in the affected ear. Audiometry reportedly demonstrated a mixed (conductive and sensorineural) hearing loss with a mild to moderate sensorineural component as well as a 10 dB air-bone gap. Clearly it is difficult to demonstrate cause and effect in all of these cases, especially when there are confounding factors including local pathology in the affected ear and no pretreatment audiometry. These cases raise concerns of azithromycin ototoxicity at lower doses and after shorter treatment durations in otherwise healthy individuals. Clarithromycin Clarithromycin has a broad spectrum of activity against aerobic and anaerobic gram-positive and gramnegative bacteria and in particular most MAC microorganisms. It obtained FDA approval in October 1991 and has a similar mode of action to the other macrolides. Kolkman and colleagues in 2002 published the first case report to demonstrate ototoxicity as a result of clarithromycin.46 They described a 76-year-old man who developed hearing loss after 4 days of high-dose clarithromycin therapy for atypical pulmonary tuberculosis. The hearing loss improved subjectively on cessation of

103

clarithromycin but worsened again on reexposure. The dose was halved, and the hearing improved. There were also earlier studies suggesting a possible link between clarithromycin and hearing loss. In a study by Fernandez-Martin examining the treatment of Toxoplasma-associated encephalitis in patients with acquired immune deficiency syndrome (AIDS), they reported hearing loss in 2 of 13 patients. 47 Their patients were treated with a combination of clarithromycin (2 g daily) and pyrimethamine (75 mg daily) for 6 weeks. The two patients developed hearing loss 2 weeks into treatment, which was confirmed audiometrically. Another study looked at the efficacy of clarithromycin, either alone or in combination, for treating Mycobacterium-associated lung disease in 45 HIV-negative patients.48 This study identified 12 patients with hearing abnormalities pretreatment (although not all of the 45 patients were tested). Four of these described worsening hearing loss during treatment. The losses were not detailed further. None of the patients reported losses severe enough to require their treatment to be stopped. In a further study looking at the treatment of Mycobacterium lung disease in HIVnegative patients, three cases of ototoxicity were reported in 30 patients receiving a combination of clarithromycin (0.75–2 g daily), minocycline (200 mg daily), and clofazimine (100 mg daily).49 Although there are obviously confounding factors, such as concurrent drug therapy and illness, these studies nevertheless alert to the possibility of clarithromycin ototoxicity. Ketolides Ketolides represent a new class of macrolide antibiotic. They include telithromycin, which has been approved for clinical use, and ABT-773, which is currently in development. They share the same macrolactone ring structure as erythromycin, from which they are derived and to which they have a similar mode of action on the 50S ribosomal subunit, although they act in different ways on the nucleotides. Although the main site of ketolide and macrolide interaction is located at nucleotides A2058 and A2059 in domain V, the ketolides have an additional interaction at A752 in domain II. This additional binding site means that the ketolides have a higher affinity than do the macrolides for forming interactions with the ribosomes and therefore should delay emergence of drug resistance.50 To date, no cases of ototoxicity have been reported, including the eight double-blind, randomized comparative phase III trials (n = 2,045).50 Postmarketing surveillance studies will ultimately determine if ototoxicity occurs with this class of macrolide.

MECHANISMS OF ERYTHROMYCIN OTOTOXICITY The mechanisms of erythromycin ototoxicity are not fully understood. There is evidence for both central

104

Systemic Toxicity

and peripheral effects on the auditory pathways. In the first animal study to look into this phenomenon Stupp and colleagues in 1973 instilled a 0.1 mg solution of erythromycin into the middle ear space of guinea pigs three times in 1 day.51 The cochleas from the animals were harvested, then examined. Sensory hair cell loss was reported, especially in the lower turns. This finding suggests a peripheral toxic effect, although the route of administration is currently not applicable to humans. Brummett and colleagues reported on their findings of auditory brainstem response (ABR) changes with erythromycin.6 In their first report, anesthetized guinea pigs were administered erythromycin (125 mg/kg/h) and ABR was recorded to an 8 kHz tone pulse stimulus. The first effect was an increase in latency of the fourth wave, followed in sequence by the other waves. When the erythromycin infusion was stopped, the ABR returned to baseline in the reverse order. In their second report ABRs were recorded in patients receiving erythromycin. An increased latency in wave V was noted, which caused the authors to speculate that erythromycin ototoxicity may be caused by a central effect on the brainstem auditory pathways. Evidence for central nervous system effects of erythromycin is also supported by reports of acute psychiatric reactions to the drug, as mentioned previously.33,52 Conversely, in another report two patients who experienced erythromycin-induced hearing loss had abnormalities of waves I to III recorded on the ABR. The ABR returned to normal after erythromycin was stopped. Nonetheless these wave abnormalities would suggest a more peripheral effect.24 Further evidence for a direct toxic effect on inner ear structures has come from other animal studies. Liu and colleagues employed electrophysiological techniques in vitro to study the effects of erythromycin perfusion on each side of cells from the stria vascularis and also on each side of the homologous vestibular dark cells.53 They found reversible changes when the solution was applied to the basolateral cell surface in both models. They concluded that erythromycin has an inhibitory effect on ion transport by the stria vascularis and vestibular dark cells. Kobayashi and colleagues studied the effects of intravenous administration of erythromycin (100 and 150 mg/kg) on guinea pigs’ cochlear potentials.54 They recorded a transient reduction in endocochlear potential and cochlear microphonics at the first turn of the cochlea. Uzun and colleagues measured transiently evoked otoacoustic emissions from guinea pigs in their study.55 The effects of erythromycin (125 mg/kg intravenously), azithromycin (45 mg/kg orally,) and clarithromycin (75 mg/kg intravenously) were recorded. Azithromycin and clarithromycin caused a reversible reduction in the emission response, although, interestingly erythromycin did not.

The mechanism of macrolide ototoxicity is not determined, but certainly there is evidence from animal models that macrolides have direct effects on the inner ear as well as effects on higher central auditory pathways.

SUMMARY • Ototoxicity from erythromycin and class-related macrolides has usually been associated with reversible SNHL. Nevertheless, cases of irreversible SNHL have been reported. • Predisposing factors for ototoxicity appear to include having renal failure or hepatic impairment, being a transplant recipient, and using concomitant ototoxic agents. Risk factors include high-dose prolonged macrolide therapy. • Little is known about the effects of macrolides on the vestibular system. • Animal models of erythromycin ototoxicity have suggested that erythromycin impairs ion transportation at the level of the stria vascularis and may also affect higher central auditory pathways.

REFERENCES 1. McGuire JM, Bunch RL, Anderson RC, et al. “Ilotycin,” a new antibiotic. Antibiot Chemother 1952; 2:281–4. 2. Mosby’s drug consult. Elsevier; 2003. Available at: www3.us.elsevierhealth.com/DrugConsult/ (accessed Aug 2003). 3. Mintz U, Amir J, Pinkhas J, deVries A. Transient perceptive deafness due to erythromycin lactobionate. JAMA 1973;225:1122–3. 4. Eckman MR, Johnson T, Riess R. Partial deafness after erythromycin [letter]. N Engl J Med 1975; 292:649. 6. Karmody CS, Weinstein L. Reversible sensorineural hearing loss with intravenous erythromycin lactobionate. Ann Otol Rhinol Laryngol 1977; 86(1 Pt 1):9–11. 6. Brummett RE. Ototoxic liability of erythromycin and analogues. Otolaryngol Clin North Am 1993; 26:811–9. 7. Quinnan GV Jr, McCabe WR. Ototoxicity of erythromycin. Lancet 1978;1:1160–1. 8. van Marion WF, van der Meer JW, Kalff MW, Schicht SM. Ototoxicity of erythromycin. Lancet 1978;2:214–5. 9. Lornoy W, Steyaert J. Ototoxicity of erythromycin lactobionate. Acta Clin Belg 1979;34:111. 10. Levin G, Behrenth E. Irreversible ototoxic effect of erythromycin. Scand Audiol 1986;15:41–2. 11. Dylewski J. Irreversible sensorineural hearing loss due to erythromycin. Can Med Assoc J 1988;139: 230–1.

Macrolides

12. Meyer RD, Edelstein PH, Kirby BD, et al. Legionnaires’ disease: unusual clinical and laboratory features. Ann Intern Med 1980;93:242–3. 13. Beckner RR, Gantz N, Hughes JP, Farricy JP. Ototoxicity of erythromycin gluceptate. Am J Obstet Gynecol 1981;139:738–9. 14. Miller SM. Erythromycin ototoxicity. Med J Aust 1982;2:242–3. 15. Schwartz JL, Maggini GA. Erythromycin-induced ototoxicity substantiated by rechallenge. Clin Pharm 1982;1:374–6. 16. Haydon RC, Thelin JW, Davis WE. Erythromycin ototoxicity: analysis and conclusions based on 22 case reports. Otolaryngol Head Neck Surg 1984;92:678–84. 17. Rydberg J. Reversible hearing loss and erythromycin. Lakartidningen 1984;81:1308. 18. Schweitzer VG, Olson NR. Ototoxic effect of erythromycin therapy. Arch Otolaryngol 1984;110: 258–60. 19. Koegel L. Ototoxicity: a contemporary review of aminoglycosides, loop diuretics, acetylsalicylic acid, quinine, erythromycin, and cisplatinum. Am J Otol 1985;6:190–8. 20. Marti J, Mutio, L, Alonso A, et al. Reversible sensorineural deafness secondary to erythromycin. An Otorrinolaringol Ibero Am 1988;15:429–32. 21. Agusti C, Ferran F, Gea J, Picado C. Ototoxic reaction to erythromycin. Arch Intern Med 1991;151:380. 22. Kemp E, Keidar S, Brook JG. Sensorineural hearing loss with low dose erythromycin. BMJ 1991; 302:1341. 23. Whitener CJ, Parker JE, Lapp NL. Erythromycin ototoxicity: a call to heighten recognition. South Med J 1991;84:1214–6. 24. Sacristan JA, Angeles De Cos M, Soto J, et al. Ototoxicity in man: electrophysiologic approach. Am J Otol 1993;14:186–8. 25. Sacristan JA, Soto JA, De Cos MA. Erythromycininduced hypoacusis: 11 new cases and literature review. Ann Pharmacother 1993;27:950–5. 26. Mery JP, Kanfer A. Ototoxicity of erythromycin in patients with renal insufficiency. N Engl J Med 1979;301:944. 27. Thompson P, Wood RP II, Bergstrom L. Erythromycin ototoxicity. J Otolaryngol 1980;9:60–2. 28. Taylor R, Schofield IS, Ramos JM, et al. Ototoxicity of erythromycin in peritoneal dialysis patients. Lancet 1981;2:935–6. 29. Errick JK, Hyrciuk-Flaska L, Thies H. Probably erythromycin ototoxicity. Drug Intell Clin Pharm 1980;14:623–5. 30. Demaldent JE, Rolland A, Mongrolle Y. Ototoxic potential of erythromycin. Ann Otolaryngol Chir Cervicofac 1984;101:643–7.

105

31. Kroboth PD, McNeil MA, Kreeger A, et al. Hearing loss and erythromycin pharmacokinetics in a patient receiving hemodialysis. Arch Intern Med 1983;143:1263–5. 32. Kanfer A, Stamatakis G, Torlotin JC, et al. Changes in erythromycin pharmacokinetics induced by renal failure. Clin Nephrol 1987;27:147–50. 33. Umstead GS, Neumann KH. Erythromycin ototoxicity and acute psychotic reaction in cancer patients with hepatic dysfunction. Arch Intern Med 1986;146:897–9. 34. Wallach PM, Love SR, Fiorica JV, et al. Erythromycin associated hearing loss in a patient with prior cis-platinum induced ototoxicity. J Fla Med Assoc 1992;79:821–2. 35. Vasquez EM, Maddux MS, Sanchez J, Pollak R. Clinically significant hearing loss in renal allograft recipients treated with intravenous erythromycin. Arch Intern Med 1993;153:879–82. 36. Moral A, Navasa M, Rimola A, et al. Erythromycin ototoxicity in liver transplant patients. Transpl Int 1994;7:62–4. 37. Hugues FC, Laccourreye A, Lasserre MH, Toupet M. Cochlear toxicity of erythromycin in elderly patients. Therapie 1984;39:591–4. 38. Swanson DJ, Sung RJ, Fine MJ, et al. Erythromycin ototoxicity: prospective assessment with serum concentrations and audiograms in a study of patients with pneumonia. Am J Med 1992;92: 61–8. 39. Wallace MR, Miller LK, Nguyen MT, et al. Ototoxicity with azithromycin. Lancet 1994;343:241. 40. Tseng AL, Dolovich L, Salit IE. Azithromycinrelated ototoxicity in patients infected with human immunodeficiency virus. Clin Infect Dis 1997; 24:76–7. 41. Lo SH, Kotabe S, Mitsunaga L. Azithromycininduced hearing loss. Am J Health Syst Pharm 1999;56:380–3. 42. Brown BA, Griffith DE, Girard W, et al. Relationship of adverse events to serum drug levels in patients receiving high-dose azithromycin for mycobacterial lung disease. Clin Infect Dis 1997; 24:958–64. 43. Bizjak ED, Haug MT III, Schilz RJ, et al. Intravenous azithromycin-induced ototoxicity. Pharmacotherapy 1999;19:245–8. 44. Ress BD, Gross EM. Irreversible sensorineural hearing loss as a result of azithromycin ototoxicity. A case report. Ann Otol Rhinol Laryngol 2000; 109:435–7. 45. Mamikoglu B, Mamikoglu O. Irreversible sensorineural hearing loss as a result of azithromycin ototoxicity. A case report. Comment. Ann Otol Rhinol Laryngol 2001;110:102.

106

Systemic Toxicity

46. Kolkman W, Grooneveld JH, Baur HJ, Verschuur HP. Ototoxicity induced by clarithromycin. Ned Tijdschr Geneeskd 2002;146:1743–5. 47. Fernandez-Martin J, Leport C, Morlat P, et al. Pyrimethamine-clarithromycin combination for therapy of acute Toxoplasma encephalitis in patients with AIDS. Antimicrob Agents Chemother 1991; 35:2049–52. 48. Dautzenberg B, Piperno D, Diot P, et al. Clarithromycin in the treatment of Mycobacterium avium lung infections in patients without AIDS. Chest 1995;107:1035–40. 49. Roussel G, Iqual J. Clarithromycin with minocycline and clofazimine for Mycobacterium avium itracellulare complex lung disease in patients without the acquired immune deficiency syndrome. Int J Tuberc Lung Dis 1998;2:462–70. 50. Zhanel GG, Walters M, Noreddin A, et al. The ketolides: a critical review. Drugs 2002;62:1771–804.

51. Stupp H, Kupper K, Lagler F, et al. Inner ear concentrations and ototoxicity of different antibiotics in local and systemic application. Audiology 1973;12:350–63. 52. Cohen IJ, Weitz R. Psychiatric complications with erythromycin. Drug Intell Clin Pharm 1981; 15:3898. 53. Liu J, Marcus DC, Kobayashi T. Inhibitory effect of erythromycin on ion transport by stria vascularis and vestibular dark cells. Acta Otolaryngol 1996; 116:572–5. 54. Kobayashi T, Rong Y, Chiba T, et al. Ototoxic effect of erythromycin on cochlear potentials in the guinea pig. Ann Otol Rhinol Laryngol 1997; 106(7 Pt 1):599–603. 55. Uzun C, Koten M, Adali MK, et al. Reversible ototoxic effect of azithromycin and clarithromycin on transiently evoked otoacoustic emissions in guinea pigs. J Laryngol Otol 2001;115:622–8.

Topical Toxicity CHAPTER 11

Middle Ear Effects of Ototopical Agents Charles G. Wright, PhD, and Peter S. Roland, MD

Topical preparations containing antibiotics and antiinflammatory agents continue to be widely used for prophylaxis and treatment of external and middle ear infections. In addition to their primary ingredients, these preparations typically contain various solvents, penetrance enhancers, and preservatives that may have undesirable side effects. When placed in the external ear canal, topical preparations may enter the middle ear via tympanostomy tubes or tympanic membrane (TM) perforations, or they may be intentionally applied to the middle ear during surgery. Under those conditions, the constituents of otic drops can potentially cross the round window membrane of the cochlea and adversely affect the inner ear. Although inner ear toxicity from agents used in ototopical preparations has been the focus of many studies in laboratory animals, the effects of those materials on tissues of the middle ear have received much less attention. However, the work that has been completed to date indicates that some constituents of topical ear drops may produce middle ear injury, especially in cases in which the tympanic cavity is relatively free of effusion or active infection. This chapter briefly summarizes findings relating to antibiotics, steroids, and various other ingredients of topical preparations that have been investigated with specific regard to their effects on the middle ear.

ANTIBIOTICS One of the first publications to focus on the issue of middle ear toxicity caused by constituents of topical preparations was published in 1978 by Parker and James.1 These investigators assessed the effects of various antibiotics, antifungal compounds, anti-inflammatory agents, and solvents on the middle and inner ears of guinea pigs. The animals included in that study received 0.1 mL intratympanic injections of the test substances once daily for 5 days. Temporal bones were taken for anatomical study 10 days after the last injection. Of the nine antibiotics that were tested, only three

(penicillin G, carbenicillin, and colimycin) were found to be without significant inflammatory effects on the guinea pig middle ear. The other antibiotics assessed in the Parker and James study included gentamicin sulfate, chloramphenicol, bacitracin, and three tetracycline compounds, all of which produced middle ear inflammation, with chloramphenicol being the worst offender. The only aminoglycoside in the group, gentamicin, was reported to result in moderate mucosal inflammation and a thick mucoid secretion in all animals that were tested. In a more recent study by Barlow and colleagues, 0.3% gentamicin ophthalmic solution produced mild mucosal thickening as well as inflammatory cell infiltration and thickening of the guinea pig TM following middle ear instillation for 7 days.2 As noted above, Parker and James found penicillin G and carbenicillin to be free of inflammatory effects in the guinea pig middle ear.1 That feature, however, is apparently not shared by other penicillin derivatives. In a 1995 study by Jakob and colleagues it was found that ticarcillin, either alone or in combination with the β-lactamase inhibitor clavulanic acid, resulted in damage to cochlear tissues and provoked severe inflammatory changes in the chinchilla middle ear after administration of a single dose. 3 Those changes included hemorrhage within the TM and middle ear mucosa, fibrous adhesion formation, and the development of serous effusion. Dramatic thickening (on the order of 40×) of the TM was observed, including disruption of the integrity of the middle fibrous layer. In several animals sacrificed 4 weeks after drug administration, hyperplastic epidermis on the lateral side of the TM was found to penetrate the disrupted fibrous layer and reach the medial surface to produce middle ear cholesteatoma. As discussed below, propylene glycol is also capable of producing the rather specific structural alterations of the TM necessary to permit migration of epidermis through the intact TM. To our knowledge, ticarcillin is the only antibiotic studied to date that appears to affect the TM in such a way as to

108

Topical Toxicity

Figure 11-1 Histologic cross sections showing alterations of the chinchilla tympanic membrane and middle ear mucosa 1 week after application of a topical preparation containing 10% sulfacetamide. A, Thickening of the tympanic membrane with hyperplasia of its epidermal (Ep) and mucosal (Mc) layers. The highly active epidermis is producing large amounts of keratin (Kr) on the lateral surface. FL = fibrous layer. (Toluidine blue stain; ×400 original magnification.) B, Inflamed middle ear mucosa. Several widely dilated blood vessels are present in the thickened subepithelial tissue; one of these vessels is indicated by “V.” Acute inflammatory cells (IC) are seen lying immediately over the mucosa. Bn = bony wall of middle ear; MC = middle ear cavity. (Hematoxylin and eosin stain; ×100 original magnification.) Adapted from Brown OE et al.4

stimulate cholesteatoma formation in experimental animals. Over the last two decades, various broad-spectrum antibiotics have been tested in laboratory animals in an effort to identify antimicrobials that might be free of ototoxicity and also be effective in the management of middle ear infection. Sulfacetamide and ceftazidime are two such agents that appear to produce little or no inner ear toxicity in the highly sensitive chinchilla model. Sulfacetamide is a sulfonamide antibiotic effective against both gram-positive and gram-negative organisms, whereas ceftazidime is a third-generation cephalosporin with a broad antibacterial spectrum. Despite the absence of significant inner ear effects, studies by Brown and colleagues demonstrated that preparations containing either sulfacetamide sodium or ceftazidime resulted in mucosal inflammation, hemorrhage, and TM thickening in animals killed 1 week after singledose middle ear application (Figures 11-1 and 11-2).4,5 These effects, however, were largely resolved in animals kept for 1 month after instillation of either drug,

suggesting that these agents may not have long-term deleterious effects on the middle ear. The fluoroquinolone antibiotics are highly effective against most pathogens associated with otitis media, and these drugs show considerable promise for use in the formulation of topical preparations that might be used to treat both acute and chronic middle ear infections. An important representative of this class of compounds, ciprofloxacin, has been well documented to be free of inner ear toxicity. To date, however, little published information is available regarding its effects on the middle ear. Ofloxacin is another fluoroquinolone antibiotic that appears to be without toxic effects on the inner ear but has been found to produce mild to moderate middle ear inflammation in animal studies. Barlow and colleagues described mild mucosal and TM thickening with some inflammatory cell infiltration in guinea pigs sacrificed 1 week after a series of seven daily middle ear instillations of 1% ofloxacin.2 In an unpublished study from the authors’ laboratory, mucosal inflammation and hyperplasia of the TM epidermis were seen 1 week after a single middle ear application of a commercially available otic solution containing 0.3% ofloxacin (Figure 11-3A). Moderate mucosal changes together with some osteoneogenesis were still evident in these animals 4 weeks after the ofloxacin applications (Figure 11-3B). Cortisporin Otic Suspension (Burroughs Wellcome, Research Triangle Park, NC) is a topical preparation containing the antibiotics neomycin and polymyxin B, which is intended for use in the external ear canal but has been used in patients with tympanostomy tubes or TM perforations where it may enter the middle ear. In animal studies this preparation has been shown to result in severe middle and inner ear toxicity.2,6–8 There is some evidence that polymyxin B is capable of bonding to and inactivating bacterial endotoxin, thereby potentially reducing endotoxinmediated inflammation in otitis media.9 However, as is true for neomycin, polymyxin B is quite toxic to the inner ear, and readily crosses the round window membrane. In fact, when applied to the middle ears of chinchillas and primates at the same concentrations used in Cortisporin, polymyxin B has been found to be significantly more ototoxic than neomycin. 10 The major middle ear effects of combination preparations such as Cortisporin appear to be caused by their nonantibiotic constituents.

STEROIDAL ANTI-INFLAMMATORY AGENTS Several of the topical preparations that have been used in clinical management of ear disease contain one or more antibiotics in combination with a steroidal antiinflammatory agent, usually hydrocortisone or dexamethasone. The effects of steroids on middle ear tissues are complex, and their use in the treatment of otitis

Middle Ear Effects of Ototopical Agents

Figure 11-2 Mucosal hemorrhage associated with topical

administration of ceftazidime. A, Dissection of chinchilla middle ear showing mucosal hemorrhage (H) in the hypotympanum 7 days after application of 10% ceftazidime. C = apex of cochlea; TM = medial surface of tympanic membrane. (Unstained preparation; ×7 original magnification.) Adapted from Brown OE et al.5 B, Histologic cross section of middle ear mucosa in an area of hemorrhage 7 days after application of ceftazidime. The mucosa contains dilated blood vessels (one of which is indicated by “V”), and an accumulation of extravasated blood (RBCs) lies against the mucosal surface. (Toluidine blue stain; ×400 original magnification.)

media remains somewhat controversial. Steroids have been shown to up-regulate transepithelial sodium transport in the middle ear mucosa, which may result in improved fluid clearance from the middle ear.11 In addition to their ability to block the production of inflammatory mediators produced via the arachidonic acid cycle, corticosteroids appear to directly counteract the vasoactive effects of histamine on the mucosal vasculature, thereby reducing inflammation.12 Dexamethasone was one of the agents included in the Parker and James study, and they reported that it had no deleterious effects on the guinea pig middle ear when applied either in suspension form or in solution (as dexamethasone sodium phosphate).1 Conversely, it is well known that steroids can produce atrophic changes in human skin and may significantly inhibit wound healing, including the healing of TM perforations. Delayed healing of experimental TM perforations was demonstrated by Spandow and colleagues, who also found that hydrocortisone application was

109

associated with inflammatory cell infiltration of the lamina propria of the TM and hypertrophy of epidermal cells on the lateral aspect of the drum.13 In another study published in 1990, Spandow and colleagues confirmed the somewhat surprising finding that hydrocortisone may have proinflammatory effects when applied to the middle ears of experimental animals.14 They reported that application of hydrocortisone to the normal rat middle ear resulted in thickening of the mucosal epithelium and polymorphonuclear leukocyte infiltration of the subepithelial connective tissue.14 In this study, 20 µL of a 2% hydrocortisone solution was instilled into the middle ear once daily for 5 days, and animals were sacrificed at 5, 10, and 21 days after the last hydrocortisone application. Evidence of mucosal inflammation was still present, even after 21 days, and secondary infection was ruled out by middle ear culture. These findings have been supported by recent work from Nordang and colleagues, who also reported inflammation of the round window membrane and middle ear mucosa after hydrocortisone administration in the rat.15 Interestingly, they observed no inflammatory changes in the middle ears of animals given dexamethasone, a finding that parallels the older report from Parker and James. 1 It therefore appears that steroids may differ in their potential for producing inflammatory changes in the middle ear.

SOLVENTS AND PRESERVATIVES USED IN TOPICAL PREPARATIONS As noted above, topical otic preparations containing combinations of antibiotics such as neomycin and polymyxin B together with steroids and various other constituents have been found to injure the middle and inner ear in experimental studies. One of the first investigations to address this issue was that of Meyerhoff and colleagues in 1983.16 In that research, chinchillas were fitted with tympanostomy tubes, and Cortisporin Otic Suspension was placed in the external ear canals once per day for 5 days. Subsequent evaluation demonstrated that the otic drops entered the middle ear via the tympanostomy tubes, and the animals showed elevated auditory thresholds as well as sensory cell loss in the basal cochlear turn. In later work, it was shown that Cortisporin placed in the ear canals of animals with tympanostomy tubes also resulted in middle ear inflammation, including formation of effusion and granulation tissue as well as focal mucosal hemorrhage.17 When Cortisporin is placed directly in the chinchilla middle ear, it consistently produces severe cochlear hair cell loss; middle ear inflammation with extensive granulation tissue formation, fibrosis, bone erosion, and osteoneogenesis; and dramatic alterations of the TM (Figures 11-4, 11-5, and 11-6).6,7 After a single middle ear instillation of the otic drops, approximately 50% of animals

110

Topical Toxicity

Figure 11-4 Abnormalities typical of chronic inflammatory

disease associated with application of Cortisporin Otic Suspension and subsequent cholesteatoma development in the chinchilla middle ear. This specimen was taken 6 weeks after Cortisporin was applied to the tympanic cavity. The mucosal epithelium (ME) consists largely of ciliated columnar and goblet cells, and the subepithelial connective tissue (CT) is dramatically thickened. Inflammatory cells (IC) overlie the altered mucosa. Bn = bony wall of middle ear; MC = middle ear cavity. (Hematoxylin and eosin stain; ×66 original magnification.)

Figure 11-3 Cross sections of tympanic membrane and

middle ear mucosa illustrating the effects of a topical preparation containing 0.3% ofloxacin applied to the chinchilla middle ear. A, Hyperplasia of tympanic membrane epidermis (Ep) 1 week after ofloxacin application. FL = fibrous layer of TM. (Toluidine blue stain; ×100 original magnification.) B, Middle ear mucosa and underlying bone (Bn) 4 weeks after ofloxacin application. There is thickening and fibrosis of the subepithelial connective tissue (CT), and a layer of new bone (asterisk) has been deposited by osteoblasts at the interface between the bone and connective tissue (arrows). (Toluidine blue stain; ×50 original magnification.)

kept for 4 to 6 weeks before sacrifice have also been found to develop middle ear cholesteatoma.7 Subsequent work has shown that the middle ear inflammatory effects of Cortisporin are in large part a result of propylene glycol, which is included in the formulation as a solvent and penetrance enhancer. Propylene glycol is not severely toxic to the inner ear in concentrations that are present in topical preparations, but it has significant inflammatory effects in the middle ear.18–20 Its effects on the TM are of particular interest since it produces structural changes that lead to cholesteatoma formation in laboratory animals, such as chinchillas and rats.20–22 When propylene glycol solutions are placed in the middle ear, the epidermal and mucosal layers of the TM are quickly destroyed, leaving the middle fibrous layer denuded.23 As the epithelial layers regrow over the fibrous layer, they become hyper-

plastic and markedly thicker than normal (see Figure 11-6). Connective tissue surrounding the fibrous layer also becomes dramatically thickened, resulting in distortion and disruption of the fibrous layer so that hyperplastic epidermis on the lateral side of the drum can migrate entirely through the TM to reach the medial surface and continue to grow, leading to cholesteatoma development (see Figures 11-6 and 11-7). The process of epidermal migration may also produce TM perforations, but when it begins, the TM is usually intact.24 Once cholesteatoma is established in the middle ear it provokes additional inflammation as well as bone erosion just as is seen in human cholesteatoma.7,20 Taken together, these findings from laboratory animals suggest that considerable caution should be exercised in the clinical use of preparations containing propylene glycol in situations in which they may enter the middle ear. Various preservative and bacteriostatic agents have also been used in the formulations of topical medications for the ear and eye, and these may be an additional source of undesirable middle ear effects. One of these agents is benzalkonium chloride, which is sometimes included as a preservative. In their guinea pig study, Barlow and colleagues tested benzalkonium chloride at concentrations of 0.026 and 0.05%.2 At the 0.026% concentration, mild mucosal thickening was noted, whereas at the 0.05% level moderate to severe mucosal inflammation was found after a series of seven daily middle ear instillations. A dose-dependent effect was also observed for the TM, with specimens in


Figure 11-5 Middle ear cholesteatoma following Cortisporin application in the chinchilla. A, Low-power view of a periannular air cell in the middle ear showing cholesteatoma matrix (Ch) overlying granulation tissue (Gn) in an area where multinucleated osteoclasts are eroding bone (arrows). The cholesteatoma matrix consists of keratinizing (Kr) stratified squamous epithelium, which has invaded the middle ear cavity (MC). This cross section is from a specimen taken 6 weeks after Cortisporin Otic Suspension was applied to the middle ear. (Hematoxylin and eosin stain; ×25 original magnification.) Adapted from Wright CG et al.7 B, Higher power view of multinucleated osteoclasts (arrows) eroding bone (Bn) in an adjacent area of the same specimen. Ch = cholesteatoma matrix epithelium. (Hematoxylin and eosin stain; ×400 original magnification.)

the 0.05% group showing approximately twice the thickening seen in the 0.026% group of animals. Parker and James reported severe mucosal inflammation in guinea pig middle ears that received a series of five doses of an unspecified concentration of benzalkonium chloride.1 After application of 0.1% benzalkonium chloride to the guinea pig external ear canal, Monkhouse and colleagues noted severe inflammation and a marked increase in thickness of the epidermis of the medial portion of the ear canal.25 However, at the concentrations typically employed in commercially available topical preparations (0.0025–0.02%), it seems reasonable to expect that benzalkonium chloride would not have serious inflammatory effects on the middle ear mucosa.

111

Figure 11-6 Tympanic membrane changes resulting from

middle ear application of propylene glycol. A, Cross section of normal chinchilla tympanic membrane. At this relatively low magnification, it is difficult to discern the very thin epidermal and mucosal layers covering the fibrous layer, which is the most prominent structural feature of the tympanic membrane. (Toluidine blue stain; ×66 original magnification.) B, Greatly thickened tympanic membrane from an animal sacrificed 2 weeks after middle ear application of propylene glycol. The epidermal (Ep) and mucosal (Mc) layers are markedly hyperplastic, and the fibrous layer (FL) is distorted and broken (arrow) by rapid proliferation of surrounding connective tissue. (Toluidine blue stain; approximately ×66 original magnification.) Adapted from Masaki M et al.22 C, Similar specimen from an animal sacrificed 3 weeks after propylene glycol administration. At the arrow, hyperplastic epidermis (Ep) is penetrating a break in the fibrous layer (FL) to reach the medial portion of the tympanic membrane. Mc = mucosal layer. (Toluidine blue stain; approximately ×66 original magnification.) Adapted from Wright CG et al.23

ANTISEPTICS AND ANTIMYCOTICS Antiseptic agents employed during myringotomy and other surgical procedures may enter the middle ear and contact the mucosal epithelium, medial surface of the TM, and round window membrane. Of the many antiseptics that have been used in ear surgery, only a few have been specifically tested for possible middle ear toxicity. Parker and James included chlorhexidine acetate in their study and found that it caused cochlear

112

Topical Toxicity

studies of animals. Given the quite considerable variations in response between different species, it is often difficult to meaningfully extrapolate findings from the research laboratory to the clinical setting. This is especially true in view of the fact that these medications are generally used in patients with active ear disease in whom middle ear inflammation and effusion are already present, and those factors will certainly alter the ear’s response to topical drug treatment. Nonetheless, the animal studies do serve as a guide for identifying agents that are likely to be problematic, and they are useful in designing new and improved formulations for topical preparations. Until new medications are available that are entirely free of possible deleterious effects, currently available topical drugs should be used with appropriate attention to possible toxicity associated with their use. Figure 11-7 Dissection of chinchilla middle ear showing the

medial surface of the tympanic membrane (TM) with a cholesteatoma (Ch) in contact with the promontory (Prm) of the cochlea. This specimen was taken 3 weeks after propylene glycol was applied to the middle ear. An = bony annulus of tympanic membrane. (Unstained preparation; approximately ×10 original magnification.) Adapted from Masaki M et al.22

hair cell loss as well as moderate middle ear inflammation in guinea pigs.1 More recently, Igarashi and Oka investigated the effects of 0.05% chlorhexidine gluconate on the cat middle ear and found evidence of fairly minor mucosal damage, primarily involving the ciliated cells of the tympanic cavity.26 Gentian violet has occasionally been used as a disinfectant or antimycotic in the ear. This agent appears to be severely ototoxic if it reaches the tympanic cavity. In a recent study by L.W. Tom, in which gentian violet was administered to the guinea pig middle ear, animals were found to develop signs of vestibular dysfunction, indicating significant inner ear toxicity.27 Also, severe inflammation and new bone growth were observed in the middle ears of animals receiving gentian violet. Other antimycotics tested in this study included clotrimazole, miconazole, tolnaftate, and nystatin. All appeared to be free of deleterious effects, except that nystatin was noted to leave a persistent residue in the middle ear. Nystatin was reported to be without significant middle ear inflammatory effects in the Parker and James study, although two other antifungal agents, amphotericin B and griseofulvin, produced severe middle ear inflammation.1

CONCLUSION As is clear from the laboratory findings reviewed above, our knowledge of undesirable side effects from topical medications used in the ear is based almost entirely on

SUMMARY • Topical pharmaceutical preparations used in management of ear disease contain a variety of ingredients that may have undesirable side effects on the middle ear and inner ear. • Laboratory animal studies have shown that antibiotics, steroids, preservatives, and penetrance enhancers used in ototopical preparations are potentially injurious to middle ear tissues. • Although it is difficult to extrapolate the experimental findings to human patients, these preparations should be used with caution in the clinical setting. • Since currently available ototopical agents show significant potential for toxicity, there is a need to develop new formulations that remain effective but are less likely to adversely affect the middle and inner ear.

REFERENCES 1. Parker FL, James GWL. The effect of various topical antibiotic and antibacterial agents on the middle and inner ear of the guinea pig. J Pharm Pharmacol 1978;30:236–9. 2. Barlow DW, Duckert LG, Kreig CS, Gates GA. Ototoxicity of topical otomicrobial agents. Acta Otolaryngol 1994;115:231–5. 3. Jakob T, Wright CG, Robinson K, Meyerhoff WL. Ototoxicity of topical ticarcillin and clavulanic acid in the chinchilla. Arch Otolaryngol Head Neck Surg 1995;121:39–43. 4. Brown OE, Wright CG, Masaki M, Meyerhoff WL. Ototoxicity of vasocidin drops applied to the chinchilla middle ear. Arch Otolaryngol Head Neck Surg 1988;114:56–9. 5. Brown OE, Wright CG, Edwards LB, Meyerhoff WL. The ototoxicity of ceftazidime in the chinchilla


6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

middle ear. Arch Otolaryngol Head Neck Surg 1989;115:940–2. Wright CG, Meyerhoff WL. Ototoxicity of otic drops applied to the middle ear in the chinchilla. Am J Otolaryngol 1984;5:166–76. Wright CG, Meyerhoff WL, Burns DK. Middle ear cholesteatoma: an animal model. Am J Otolaryngol 1985;6:327–41. Vassalli L, Harris DM, Gradini R, Applebaum EL. Inflammatory effects of topical antibiotic suspensions containing propylene glycol in chinchilla middle ears. Am J Otolaryngol 1988;8:1–5. Darrow DH, Keithley EM. Reduction of endotoxin-induced inflammation of the middle ear by polymyxin B. Laryngoscope 1996;106:1028–33. Wright CG, Meyerhoff WL, Halama AR. Ototoxicity of neomycin and polymyxin B following middle ear application in the chinchilla and baboon. Am J Otol 1987;8:495–9. Tan CT, Escoubet B, Van den Abbeele T, et al. Modulation of middle ear epithelial function by steroids: clinical relevance. Acta Otolaryngol 1997; 117:284–8. Chan KH, Swarts JD, Tan L. Middle ear mucosal inflammation: an in vivo model. Laryngoscope 1994;104:970–80. Spandow OD, Hellstrom S, Schmidt S-H. Hydrocortisone delay of tissue repair of experimental tympanic membrane perforations. Ann Otol Rhinol Laryngol 1990;99:647–53. Spandow O, Hellstrom S, Anniko M. Structural changes in the round window membrane following exposure to Escherichia coli lipopolysaccharide and hydrocortisone. Laryngoscope 1990;100:995–1000. Nordang L, Linder B, Anniko M. Morphologic changes in round window membrane after topical hydrocortisone and dexamethasone treatment. Otol Neurotol 2003;24:339–43. Meyerhoff WL, Morizono T, Wright CG, et al. Tympanostomy tubes and otic drops. Laryngoscope 1983;93:1022–7.

113

17. Anderson RG, Wright CG, Meyerhoff WL. Inflammatory effects of otic drops on the middle ear. Int J Pediatr Otorhinolaryngol 1984;7:91–5. 18. Vernon J, Brummett R, Walsh T. The ototoxic potential of propylene glycol in guinea pigs. Arch Otolaryngol 1978;104:726–9. 19. Morizono T, Paparella MM, Juhn SK. Ototoxicity of propylene glycol in experimental animals. Am J Otolaryngol 1980;1:393–9. 20. Vassalli L, Harris DM, Gradini R, Applebaum EL. Propylene glycol-induced cholesteatoma in chinchilla middle ears. Am J Otolaryngol 1988; 4:180–8. 21. Huang CC, Shi GS, Yi ZX. Eperimental induction of middle ear cholesteatoma in rats. Am J Otolaryngol 1988;9:165–72. 22. Masaki M, Wright CG, Lee DH, Meyerhoff WL. Experimental cholesteatoma. Acta Otolaryngol 1989;108:113–21. 23. Wright CG, Bird LL, Meyerhoff WL. Tympanic membrane microstructure in experimental cholesteatoma. Acta Otolaryngol 1991;111: 101–11. 24. Wright CG, Bird LL, Meyerhoff WL. Development of tympanic membrane perforations associated with experimental cholesteatoma. In: Lim DJ, Bluestone CD, Klein JO, et al, editors. Recent advances in otitis media. Proceedings of the Fifth International Symposium. Hamilton (ON): Decker Periodicals; 1993. p. 457–9. 25. Monkhouse WS, Moran P, Freedman A. The histological effects on the guinea pig external ear of several constituents of commonly used aural preparations. Clin Otolaryngol 1988;13: 121–31. 26. Igarashi Y, Oka Y. Mucosal injuries following intratympanic applications of chlorhexidine gluconate in the cat. Arch Otorhinolaryngol 1988; 245:273–8. 27. Tom LW. Ototoxicity of common topical antimycotic preparations. Laryngoscope 2000;110:509–16.

CHAPTER 12

Topical Aminoglycoside Cochlear Ototoxicity Peter S. Roland, MD, and Charles G. Wright, PhD

The ototoxic and nephrotoxic effects of systemically administered aminoglycoside antibiotics have been recognized since the introduction of streptomycin for the treatment of tuberculosis in the 1940s.1 As other aminoglycoside drugs came into use they too were found to be associated with inner ear toxicity. It was subsequently discovered that these drugs may also damage the cochlea when topically applied in situations where they have access to the middle ear cavity. Much of the evidence for cochleotoxicity from topically administered aminoglycosides has come from experimental animal studies in which the drugs have been applied directly to the middle ear. As discussed below, such studies have consistently shown that these antibiotics are capable of producing severe inner ear damage, including largescale sensory cell loss, neural degeneration, and dramatically compromised auditory function. To what extent topically applied aminoglycosides may injure the human inner ear in the clinical setting has been a matter of considerable discussion and debate.2–6 This chapter reviews both laboratory and clinical findings regarding cochlear ototoxicity associated with topical application of aminoglycoside antibiotics. Topical antibiotic preparations have often been used to treat otorrhea after tympanostomy tube insertion and for management of acute or chronic otitis media in the presence of tympanic membrane perforations. Antibiotic preparations have also been administered prophylactically following tympanostomy tube placement to reduce the probability of postoperative otorrhea and to help prevent tube occlusion. Under any of these conditions, antibiotics applied to the external ear canal may enter the middle ear and contact the round window membrane, thereby gaining access to the inner ear. This has been a matter of concern since studies conducted in laboratory animals have demonstrated that aminoglycoside antibiotics used in topical otic preparations are potentially ototoxic. Animal studies relating to topical aminoglycoside ototoxicity have a long history. During the 1950s

Schucknecht reported that concentrated solutions of streptomycin applied to the middle ears of cats produced degeneration of cochlear as well as vestibular sensory cells.7 This research was part of an effort to develop an effective vestibular ablation procedure for relief of Meniere’s disease and was done within the context of the recognized vestibulotoxic properties of streptomycin. In addition to his experimental trials with cats, Schuknecht treated a group of Meniere’s disease patients with topical streptomycin and obtained clear evidence that the antibiotic made its way into the inner ear and damaged both the cochlea and the vestibular apparatus.7 To our knowledge, this was the first published study demonstrating topical aminoglycoside ototoxicity in human subjects. Schuknecht also observed severe ototoxicity in cats after a single middle ear application of streptomycin, whereas multiple doses were required to produce a comparable effect in human patients. He attributed this difference to greater exposure of the round window membrane in the cat middle ear relative to humans. Reports from animal studies appearing in the late 1960s and early 1970s began to focus on the ototoxic potential of other aminoglycosides, which were then being used for topical treatment of otitis externa and otitis media. In a 1969 study, Kohonen and Tarkkanen applied neomycin to the middle ears of guinea pigs in concentrations of 5, 10, 20, 50, and 100 mg/mL and found histologic evidence of cochlear hair cell injury at all concentrations above 10 mg/mL, with increasing amounts of sensory cell loss at the higher antibiotic concentrations.8 In a subsequent study that included several aminoglycoside antibiotics, Stupp and colleagues applied isomolar concentrations (0.1 M) of streptomycin, kanamycin, gentamicin, and neomycin to the middle ears of guinea pigs once daily for 3 days.9 The animals were terminated 2 days after the last drug application for anatomic assessment of cochlear hair cell damage and measurements of antibiotic concentration in perilymph. It was found that neomycin and

Topical Aminoglycoside Cochlear Ototoxicity

gentamicin produced complete destruction of sensory cells, and these drugs were also found to reach the perilymph in higher concentrations than did the other antibiotics tested. Similar results were later reported by Harada and colleagues, who placed 5 mg of neomycin directly on the round window membrane in guinea pigs and assessed perilymph concentrations of the drug and cochlear hair cell loss after various time intervals.10 These investigators found high concentrations of neomycin in the perilymph after 30 minutes of application, indicating that the antibiotic easily penetrated the round window membrane. Loss of hair cells was observed after neomycin application for 4 hours with increasing damage following longer periods of drug placement on the round window membrane. In an effort to approximate dosage schedules used in treatment of chronic otitis media, Brummett and colleagues applied 0.1 mL doses of neomycin at a concentration of either 5, 15, or 45 mg/mL to the middle ears of guinea pigs three times per day for 4 weeks.11 At the end of the drug administration period, the animals were kept for 30 days before they were evaluated electrophysiologically and cochlear tissues were taken for histologic study. These workers found dose-dependent effects on cochlear microphonics, and at the two higher neomycin concentrations, there was extensive loss of sensory cells in all cochlear turns. In an even longer-term study, which included evaluation of spiral ganglion cells, Zappia and Altschuler administered a single intratympanic dose of neomycin to guinea pigs and assessed cochlear histology at 10 weeks posttreatment.12 Extensive destruction of outer hair cells (OHCs) and inner hair cells (IHCs) was observed, together with spiral ganglion cell loss in all cochlear turns. The loss of ganglion cells was probably secondary to IHC loss. However, the possibility of a direct effect of the antibiotic on the spiral ganglion was not ruled out. In a study designed to compare the effects of single-dose, middle ear administration of neomycin and the nonaminoglycoside antibiotic polymyxin B in both chinchillas and baboons, Wright and colleagues found that although neomycin did damage cochlear sensory cells, it was significantly less toxic than polymyxin B when each antibiotic was administered at concentrations equal to those used in a commercially available ototopical preparation.13 Both antibiotics were found to be dramatically less ototoxic in the primate than in the rodent model. The studies summarized above are representative of a sizable body of laboratory work demonstrating that single aminoglycoside antibiotics applied to the middle ears of experimental animals are capable of crossing the round window membrane to reach the inner ear, where they adversely affect auditory function and cause structural damage of sensory cells and neural elements within the cochlea. In order to broaden the spectrum of antimicrobial activity for clinical management of ear

115

infections, aminoglycosides are often used together with other antibiotics and with anti-inflammatory agents in combination ototopical preparations such as Cortisporin Otic Suspension (containing neomycin, polymyxin B, and hydrocortisone) and Coly-Mycin S Otic Suspension (neomycin, polymyxin E, and hydrocortisone). According to manufacturers’ recommendations, otic preparations such as Cortisporin are to be used only in the external ear canal in patients with intact tympanic membranes. In actual practice, however, they are often administered to patients with patent tympanostomy tubes or with tympanic membrane perforations and are also used in intraoperative packing materials placed in the middle ear during surgery. Otic drops have also been recommended for prophylactic use in patients with tympanostomy tubes following water contamination of the ear canal.14 Since these preparations contain ingredients (including aminoglycoside antibiotics) with proven ototoxic potential, their clinical use in situations in which they can enter the middle ear has been questioned. As outlined above, several individual antibiotics such as neomycin, gentamicin, and the polymyxins have been shown to be ototoxic in experimental studies. It is therefore not particularly surprising that combination otic drops such as Cortisporin Otic Suspension are also injurious to the cochlea when topically applied to the middle ears of laboratory animals, as has been demonstrated in species ranging from guinea pigs to primates. One of the first such studies was done by Meyerhoff and colleagues, who placed Cortisporin Otic Suspension in the external ear canals of chinchillas that had been previously implanted with tympanostomy tubes.15 These investigators observed that the antibiotic preparation readily entered the middle ear and contacted the round window membrane within 30 minutes after placement in the ear canal. Following application of Cortisporin for 5 consecutive days in this animal model, there was elevation of auditory evoked potential thresholds and histologic evidence of OHC loss in the basal turn of the cochlea as illustrated in Figure 12-1. It was subsequently shown that a single application of 0.5 mL of Cortisporin directly to the chinchilla middle ear invariably results in complete destruction of all inner and OHCs throughout the cochlea (Figure 12-2), together with severe damage of the sensory epithelia of the vestibular apparatus.16 Electrophysiologic evidence for the ototoxic effects of combination otic drops in the chinchilla has also been reported by Morizono.17 Morizono applied 50 mL of Cortisporin Otic Suspension and Coly-Mycin S Otic Suspension directly to the round window membrane for 10 minutes and then rinsed it away. Over the next 24 hours compound action potential thresholds

116

Topical Toxicity

Figure 12-2 Scanning electron micrograph showing total loss of cochlear sensory cells in the third cochlear turn following direct middle ear application of Cortisporin Otic Suspension® in a chinchilla. Only phalangeal scars remain on the reticular lamina in the inner (IHC) and outer hair cell (OHC) areas. PL = pillar cell headplates. The scale bar at lower left indicates 10 µm.

Figure 12-1 Scanning electron micrographs illustrating outer hair cell (OHC) loss in the basal cochlear turn of a chinchilla following application of Cortisporin Otic Suspension to the external ear canal in the presence of a patent tympanostomy tube. A, Scattered OHC loss in the upper portion of the basal turn. B, More severe OHC loss in the lower basal turn of the same animal. Reproduced with permission from Meyerhoff WL et al.15

became elevated in association with application of both preparations, but the deleterious effects were found to be greater for the Coly-Mycin preparation, which, in addition to neomycin, contains polymyxin E rather than polymyxin B. Barlow and colleagues compared the morphological effects of Cortisporin Otic Suspension and 0.3% gentamicin ophthalmic solution on the middle and inner ears of guinea pigs following a series of seven daily middle ear instillations of the test agents.18 Extensive IHC and OHC loss was observed following the Cortisporin applications, whereas only minor sensory cell loss, which was not significantly different from saline controls, was seen with the gentamicin solution. In addition to laboratory findings obtained from work with rodent species, combination otic drops have also been shown to be ototoxic following topical middle ear administration in a primate model. After a single middle ear instillation of Cortisporin Otic Suspension in baboons, Wright and colleagues found IHC and OHC loss in all six animals included in their study (Figures 12-3 and 12-4)19; however, the sensory

cell loss was confined to the basal turn of the cochlea. The cochlear damage was therefore less severe than had been previously observed following Cortisporin administration in the chinchilla. The lesser degree of cochlear injury seen in the baboon was attributed to the thicker and more densely structured round window membrane in the primate, which probably affords a more effective barrier to penetration of ototoxic agents. Tobramycin is another aminoglycoside antibiotic found in topical ophthalmic solutions that have occasionally been used for treatment of ear infections. With regard to possible ototoxicity, tobramycin has received much less attention in animal studies than antibiotics such as neomycin. However, in a study in which the ototoxic properties of tobramycin were compared with those of gentamicin following systemic administration in guinea pigs, it was observed to be significantly less cochleotoxic than gentamicin.20 Also, Jinn and colleagues included Tobradex® (containing 0.3% tobramycin and 0.1% dexamethasone) in an in vitro study of isolated OHC toxicity from otic drops and found it to be largely free of adverse effects in their cell culture model.21 Although most of the aminoglycosides are unequivocally ototoxic in laboratory animals, reports of hearing loss associated with their topical use in human patients have been rather scattered and, for the most part, limited to single case reports. This is true despite the fact that ototopical antibiotics have been in widespread clinical use for many years. It is generally believed, therefore, that the human inner ear is less vulnerable to injury from these medications. Several factors are likely to contribute to this apparent difference between species. One of the more important of these is the fact that the human cochlea is relatively well protected from toxic agents


117

Figure 12-4 Dissection of the basal cochlear turn from a baboon 5 months after application of Cortisporin Otic Suspension® to the middle ear. The arrows indicate three of the several areas in which there is focal degeneration of the organ of Corti (OC) and of myelinated nerve fibers in the osseous spiral lamina (OSL). (Osmium tetroxide stain; ×12 original magnification.) Adapted from Wright CG et al.19

Figure 12-3 Light micrographs showing surface preparations of the organ of Corti from a baboon following middle ear application of Cortisporin Otic Suspension. A, Upper basal turn illustrating a normal sensory cell population. 1, 2, and 3 indicate the first, second, and third rows of outer hair cells (OHC). IHC = inner hair cell area. B, Midbasal turn showing scattered loss of OHCs. C, Lower basal cochlear turn; all OHCs destroyed. Only phalangeal scars remain in the OHC area. (All three micrographs, osmium tetroxide stain; ×400 original magnification.) Reproduced with permission from Wright CG et al.19

placed in the middle ear cavity. Most investigators agree that the round window membrane is the most significant soft tissue barrier between the middle ear and the inner ear and is the most important site of entry for ototopical agents. In the human temporal bone, the round window is positioned within a relatively deep niche, which, in more than 50% of cases, is partially or completely covered by a fold of mucous membrane (Figure 12-5).22 In the chinchilla and guinea pig, on the other hand, the round window membrane is fully and directly exposed to the middle ear space. In the animal species, therefore, materials in the middle ear can more easily come into direct contact with the round window membrane. The membrane itself is also a considerably more substantial barrier in the human ear than in animals. In rodents the round window membrane is quite thin, on the order of 12 to 16 microns in its central portion.19,23 Even in nonhuman primates, such as the baboon, the

membrane is only about 20 microns in thickness.19 In contrast, the human round window membrane averages 65 to 70 microns in thickness, and its middle, fibrous layer is quite substantial and densely structured.22,24 These differences in round window membrane morphology are illustrated in the histologic cross sections shown in Figure 12-6, and they probably play a role in the decreasing vulnerability to topical ototoxicity seen from rodents to nonhuman primates to man. In the presence of middle ear inflammation or infection, the round window membrane tends to undergo a series of changes leading to decreased permeability.25 In the acute phases of inflammation, the surface epithelial layer facing the middle ear cavity can be damaged, and permeability of the membrane may increase for a period of time. However, as the middle ear disease process becomes more chronic the membrane thickens and becomes increasingly fibrotic, making it a more effective physical barrier to drug penetration.24 Also, mucosal inflammation and edema, together with the presence of middle ear effusion, tend to reduce drug contact with the round window membrane surface. Thus, it might be expected that the risk of ototoxicity would be reduced in the presence of chronic middle ear disease, and that appears to be the case. Several authors have expressed views consistent with the idea that patients without active middle ear disease may be at greater risk for ototoxicity associated with use of topical antibiotics. Dumas and colleagues reported sensorineural hearing loss in a group of eight patients with tympanic membrane perforations who received aminoglycoside otic drops, and they stated that individuals who did not have middle ear inflammation or effusion were more likely to suffer adverse effects from topical drug treatment. 26 Lind and

118

Topical Toxicity

Figure 12-5 Human temporal bone cross section through the

round window niche area. The two arrows pointing toward the right indicate a sheet of mucous membrane that partially covers the opening of the niche. The round window membrane (RWM) lies in a protected position within the niche, which is relatively deep compared with that of laboratory animals. (Hematoxylin and eosin stain; original magnification ×25.)

Kristiansen also reported a patient who had a patent tympanostomy tube and an uninflamed middle ear and who was treated over a 10-day period for otitis externa with topical drops containing neomycin, gramicidin, and dexamethasone.27 This individual developed permanent, severe sensorineural hearing loss that appeared, with high probability, to be a result of the antibiotic medication.

Figure 12-6 Histologic cross sections showing increasing thickness and density of the normal round window membrane from chinchilla (A) to baboon (B) to human infant (C). (All three micrographs, toluidine blue stain; original magnification ×400.) Reproduced with permission from Wright CG et al.19

Nonetheless, if otic drops are used for extended periods of time, patients with active middle ear disease may also be at risk. Nomura, for example, described a patient with chronic suppurative otitis media who developed severe hearing loss during 7 months of treatment with otic drops containing neomycin.22 In that individual, the hearing loss progressed to total deafness after the antibiotic preparation was discontinued. A similar case was reported by Tommerup and Moller involving a patient with chronic otitis media who used otic drops containing framycetin (an aminoglycoside available in Europe) over an 11-month period and developed a severe-to-profound sensorineural hearing loss in the treated ear.28 In one of the only prospective studies that included multiple subjects, Podoshin and colleagues evaluated 150 patients with chronic otitis media, 124 of whom received a topical preparation containing neomycin, polymyxin B, and dexamethasone, whereas the remaining 26 patients were given dexamethasone alone.29 Evidence for minor sensorineural hearing loss (6 dB on average) was obtained in the antibiotic group following prolonged treatment (greater than 1 year). The average loss in the 26 patients given dexamethasone alone was 0.9 dB, and the difference between the groups proved to be significant at the p ≤ .025 level. In general, the available literature suggests that topical aminoglycosides are relatively safe for use in patients with chronic ear disease if treatment is of reasonably limited duration. For example, in a study of 44 pediatric patients with chronic suppurative otitis media and patent tympanostomy tubes, Merifield and colleagues found no audiometric changes after treatment with aminoglycoside drops for periods of 2 days to 2 weeks.30 (Their investigation included a total of 70 ears from the 44 patients, and bone-conduction thresholds were measured at 3, 4, and 6 kHz before and after treatment.) In another, larger study of 300 patients with otorrhea who received 0.3% gentamicin drops for up to 3 weeks, Gyde reported clinical success of the treatment in 271 of the patients with no evidence of ototoxic injury based on audiograms performed before, during, and after treatment.31 Although there have been occasional reports of human ototoxicity in connection with administration of topical antibiotics, many otologists are of the opinion that these drugs are highly effective and are safe if they are used with appropriate caution. Nearly 94% of the 2,235 otolaryngologists who participated in the 1993 survey by Lundy and Graham reported using topical antibiotic drops to treat otorrhea through ventilation tubes, and 84% used ototopical antibiotics in patients with draining tympanic membrane perforations.32 However, only 3.4% of the respondents believed they had ever observed irreversible sensorineural hearing loss caused by the medications. In a 1995 opinion


paper on topical antibiotic ototoxicity, Roland reviewed the relevant literature and expressed strong support for the safety of ototopical medications, particularly when used in patients with active middle ear disease.5 Other clinicians have continued to recommend prophylactic use of otic drops to reduce the risk of postoperative otorrhea and contend that the likelihood of ototoxicity is very small, even in patients with patent tympanostomy tubes and dry middle ears, provided that the duration of topical antibiotic treatment is kept short. For prophylaxis after tympanostomy tube placement, Baker and Chole specifically recommended the use of gentamicin drops rather than preparations containing multiple antibiotics and propylene glycol, which may carry a greater risk of ototoxicity.33 However, some otologists continue to stress the need for caution. Even though previous studies have indicated that topical drops containing gentamicin are relatively safe for clinical use, recent reports have emphasized that gentamicin preparations are not without significant ototoxic potential, especially when used in patients who do not have active middle ear disease (see Chapter 13, “Topical Aminoglycoside Vestibular Toxicity”). In a series of carefully conducted clinical studies,34–36 these investigators clearly demonstrated vestibulotoxicity associated with administration of gentamicin-containing drops, particularly when the topical preparations were applied for periods longer than 7 days. Although the toxicity does appear to be primarily vestibular rather than cochlear, some patients who received gentamicin drops did develop hearing loss, indicating that cochlear injury may also occur. On the basis of these findings, Bath and colleagues have stated that preparations containing aminoglycoside antibiotics should not be used in patients with healthy middle ears, and, if they are employed to treat otorrhea, the topical drops should be discontinued promptly after middle ear discharge has stopped.35 Other investigators have also suggested that the extent of cochlear injury produced by topical antibiotics may have been underestimated, and there are at least two reasons why that may be true.6,11,37 First, the earliest and most severe changes in auditory function are likely to occur at frequencies above those usually tested by conventional audiometry. Damage to the basal region of the organ of Corti, near the round window membrane, would therefore not be apparent during clinical evaluation. Second, it should be remembered that sensorineural hearing loss has been described as a possible sequela of chronic otitis media. In some patients the hearing loss might actually be caused by the antibiotics used to treat the otitis media rather than by the disease itself. Thus, there are numerous issues to be considered with regard to the possible role of topical aminoglycosides in human sensorineural hearing loss. Although

119

topical antibiotic preparations continue to be an important asset in clinical management of external and middle ear infections they do have some potential for inner ear injury. In their informative review on topical ototoxicity, Pickett and colleagues listed several strategies helpful in minimizing the risk of toxic effects during topical antibiotic therapy.6 Among these are the following: (1) use wicks to apply medications when treating otitis externa in the presence of tympanic membrane perforations, (2) keep the duration of treatment as short as possible, (3) monitor patients’ hearing function, especially when longer-term treatment is necessary, (4) choose topical medications that are less likely to affect the inner ear whenever possible. Until treatment methods are developed that are entirely free of possible toxicity, topical antibiotics should be administered with appropriate caution in order to avoid the possibility of hearing loss associated with their use.

SUMMARY • Laboratory studies have demonstrated that aminoglycoside antibiotics are capable of producing severe inner ear damage when topically applied to the middle ears of experimental animals. • The available evidence indicates that topical aminoglycosides are less likely to have ototoxic effects in human patients. • Although these drugs appear to be relatively safe for clinical application, they should be administered judiciously, with appropriate attention to the fact that under some conditions they are capable of damaging the inner ear.

REFERENCES 1. Hinshaw HC, Feldman WH. Streptomycin in the treatment of clinical tuberculosis: a preliminary report. Proc Mayo Clin 1945;20:313–8. 2. Mittelman H. Ototoxicity of “ototopical” antibiotics: past, present, and future. Trans Am Acad Ophthalmol Otolaryngol 1972;76:1432–43. 3. Wright CG, Meyerhoff WL. Ototopical agents: efficacy or toxicity in humans. Ann Otol Rhinol Laryngol 1988;97 Suppl 131:30–2. 4. Brummett RE, Morrison RB. The incidence of aminoglycoside antibiotic-induced hearing loss. Arch Otolaryngol Head Neck Surg 1990;116: 406–10. 5. Roland PS. Clinical ototoxicity of topical antibiotic drops. Otolaryngol Head Neck Surg 1994;110: 598–602. 6. Pickett BP, Shinn JB, Smith MFW. Ear drop ototoxicity: reality or myth? Am J Otol 1997;18: 782–91.

120

Topical Toxicity

7. Schuknecht HF. Ablation therapy in the management of Meniere’s disease. Acta Otolaryngol Suppl 1957;132:1–42. 8. Kohonen A, Tarkkanen J. Cochlear damage from ototoxic antibiotics by intratympanic application. Acta Otolaryngol 1969;68:90–7. 9. Stupp H, Kupper K, Lagler F, et al. Inner ear concentrations and ototoxicity of different antibiotics in local and systemic application. Audiology 1973; 12:350–63. 10. Harada T, Iwamori M, Nagai Y, Nomura Y. Ototoxicity of neomycin and its penetration through the round window membrane into the perilymph. Ann Otol Rhinol Laryngol 1986;95:404–7. 11. Brummett RE, Harris RF, Lindgren JA. Detection of ototoxicity from drugs applied topically to the middle ear space. Laryngoscope 1976;86:1177–87. 12. Zappia JJ, Altschuler RA. Evaluation of the effect of ototopical neomycin on spiral ganglion cell density in the guinea pig. Hear Res 1989;40:29–37. 13. Wright CG, Meyerhoff WL, Halama AR. Ototoxicity of neomycin and polymyxin B following middle ear application in the chinchilla and baboon. Am J Otolaryngol 1987;8:495–9. 14. Jaffe BF. Are water and tympanostomy tubes compatible? Laryngoscope 1981;91:563–4. 15. Meyerhoff WL, Morizono T, Wright CG, et al. Tympanostomy tubes and otic drops. Laryngoscope 1983;93:1022–7. 16. Wright CG, Meyerhoff WL. Ototoxicity of otic drops applied to the middle ear in the chinchilla. Am J Otolaryngol 1984;5:166–76. 17. Morizono T. Toxicity of ototopical drugs: animal modeling. Ann Otol Rhinol Laryngol 1990;99:42–5. 18. Barlow DW, Duckert LG, Kreig CS, Gates GA. Ototoxicity of topical otomicrobial agents. Acta Otolaryngol 1994;115:231–5. 19. Wright CG, Halama AR, Meyerhoff WL. Ototoxicity of an ototopical preparation in a primate. Am J Otolaryngol 1987;8:56–60. 20. Brummett RE, Himes D, Saine B, Vernon J. A comparative study of the ototoxicity of tobramycin and gentamicin. Arch Otolaryngol 1972;96:505–12. 21. Jinn TH, Kim PD, Russell PT, et al. Determination of ototoxicity of common otic drops using isolated cochlear outer hair cells. Laryngoscope 2001;111: 2105–8. 22. Nomura Y. Otological significance of the round window. Adv Otorhinolaryngol 1984;33:1–162. 23. Schachern PA, Paparella MM, Duvall AJ. The normal chinchilla round window membrane. Arch Otolaryngol 1982;108:550–4.

24. Sahni RS, Paparella MM, Schachern PA, et al. Thickness of the human round window membrane in different forms of otitis media. Arch Otolaryngol Head Neck Surg 1987;113:630–4. 25. Goycoolea MV, Munchow D, Schachern P. Experimental studies on round window structure: function and permeability. Laryngoscope 1988; 44 Suppl:1–20. 26. Dumas G, Bessard G, Gavend M, et al. Risque de surdite par instillations de gouttes auricularies contenant aminosides. Therapie 1980;35:357–63. 27. Lind O, Kristiansen B. Deafness after treatment with ear drops containing neomycin, gramicidin and dexamethasone. ORL J Otorhinolaryngol Relat Spec 1986;48:52–4. 28. Tommerup B, Moller K. A case of profound hearing impairment following the prolonged use of framycetin ear drops. J Laryngol Otol 1984; 98:1135–7. 29. Podoshin L, Fradis M, Ben David J. Ototoxicity of ear drops in patients suffering from chronic otitis media. J Laryngol Otol 1989;103:46–50. 30. Merifield DO, Parker NJ, Nicholson NC. Therapeutic management of chronic suppurative otitis media with otic drops. Otolaryngol Head Neck Surg 1993;109:77–82. 31. Gyde MC. When the weeping stopped. An otologist views otorrhea and gentamicin. Arch Otolaryngol 1976;102:542–6. 32. Lundy LB, Graham MD. Ototoxicity and ototopical medications: a survey of otolaryngologists. Am J Otol 1993;14:141–6. 33. Baker RS, Chole RA. A randomized clinical trial of topical gentamicin after tympanostomy tube placement. Arch Otolaryngol Head Neck Surg 1988;114:755–7. 34. Marais J, Rutka JA. Ototoxicity and topical eardrops. Clin Otolaryngol 1998;23:360–7. 35. Bath AP, Walsh RM, Bance ML, Rutka JA. Ototoxicity of topical gentamicin preparations. Laryngoscope 1999;109:1088–93. 36. Kaplan DM, Hehar SS, Bance ML, Rutka JA. Intentional ablation of vestibular function using commercially available topical gentamicinbetamethasone eardrops in patients with Meniere’s disease: further evidence for topical eardrop ototoxicity. Laryngoscope 2002;112:689–95. 37. Stringer SP, Meyerhoff WL, Wright CG. Ototoxicity. In: Paparalla MM, Shumrick DA, Gluckman JL, Meyerhoff WL, editors. Otolaryngology. Vol II. 3rd ed. Philadelphia (PA): WB Saunders; 1991. p. 1653–69.

CHAPTER 13

Topical Aminoglycoside Vestibular Toxicity Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), Vitaly Kisilevsky, MD, and John A. Rutka MD, FRCSC

Topical preparations containing aminoglycoside (eg, neomycin, framycetin, tobramycin, and gentamicin) are widely used by both otolaryngologists and primary care physicians for treatment of external otitis, acute otitis media (AOM) with perforation, chronic suppurative otitis media (CSOM), and post–tympanostomy tube otorrhea (PTTO). Their frequency of use is primarily related to their broad spectrum of antibacterial activity, in particular against Pseudomonas aeruginosa, which is the most commonly cultured organism in external otitis and CSOM,1–3 and their low per unit cost. Topical administration is also a favored route because it is convenient to use, it is effective, and other routes are either impractical or inappropriate because of poor gastrointestinal absorption.1 According to Browning and colleagues in 1988, approximately 5% of adults in the United Kingdom have CSOM, of whom at any given time approximately 2% will have an associated mucopurulent discharge indicative of an active infection.4,5 This in turn provides a rough estimate of the large number of patients possibly receiving topical antibiotic therapy for CSOM and the proportionate time and resources involved in the management of these patients.

THE CONTROVERSY: DOES CLINICAL TOPICAL OTOTOXICITY REALLY EXIST? Most product monographs and package inserts of aminoglycoside ototopical preparations caution that ototoxicity may occur if used in the presence of a tympanic membrane (TM) perforation, but this may be overstated.6 There is little doubt that topical antibiotics are effective and indicated in the treatment of CSOM.7,8 Acuin and colleagues in a meta-analysis of treatment for CSOM from 1966 to 1996 concluded that topical treatment with antibiotics or antiseptics was more effective than systemic antibiotics (six trials, odds ratio 0.46, 95% CI 0.30–0.69).8 They also concluded that combining topical and systemic antibiotics was not more effective than topical antibiotics alone. In addition, topical

quinolones were found to be more effective than nonquinolones (five trials, odds ratio 0.26, 95% CI 0.16 – 0.41). Although practice patterns vary, they will continue to evolve as safer topical antibiotic alternatives are introduced and as a result of medicolegal actions should topical ototoxicity occur. To demonstrate how practices have varied, a 1988 survey of UK general practitioners revealed that 66% would not prescribe topical aminoglycoside antibiotic ear drops in the presence of a TM perforation and that 85% would not give topical treatment in patients with discharging grommets (ventilation tubes).9 The primary reason for this reluctance was the perceived risk for ototoxicity and other complications. This concern, however, was apparently not shared by otolaryngologists in the United States. Lundy and Graham in 1992, for example, carried out a nationwide survey and reported that 84.1% of otolaryngologists felt comfortable prescribing ototopical aminoglycoside preparations in the presence of a discharging perforation and that 93.7% would use this treatment for PTTO. Eighty percent felt that the inherent risk of ototoxicity from CSOM was as great as, if not greater than, the risk involved from using aminoglycoside drops. A more recent questionnaire in 1999 carried out by Lancaster and colleagues among the consultant members of the British Association of Otorhinolaryngologists–Head and Neck Surgeons (BAO-HNS) revealed that 93% routinely used topical aminoglycoside-containing drops as part of their management of active mucosal “safe” CSOM. Only 48% believed there was a definite risk for ototoxicity when using these drops in ears with a known TM perforation. Of the 48% who acknowledged there was a risk, most of them justified their rationale for using these drops, feeling that the risk of ototoxicity is small and no greater than the risk imposed by active infection upon the inner ear.4 In 1994, Roland analyzed the reported incidence of ototopical ototoxicity and concluded that amino-

122

Topical Toxicity

glycoside-containing drops were exceedingly safe.10 This frequently cited report, however, appeared to have primarily focused on reported cochlear, not vestibular, toxicity. Considering that only one patient in this review had any form of vestibular assessment, and acknowledging that, for example, gentamicin’s toxicity is primarily vestibular, many cases of vestibulotoxicity might have been overlooked.6 Against this backdrop, guidelines and warnings from national regulatory committees regarding the potential toxic effects of topical aminoglycoside ear drops have been previously issued and continue to be updated as evidence mounts for topical ototoxicity. In 1981, the Committee on Safety of Medicines (CSM) and the Medicines Control Agency in the United Kingdom issued a warning of an increased risk of deafness associated with topical aminoglycosides for the treatment of otitis externa complicated by a perforated TM. In 1997, a further review by the CSM warned of similar risks of ototoxicity from aminoglycoside topical drops in any ear with a perforated TM.4,11 Similar precautions have been issued in Canada.12 In 1999, the National Formulary passed by an act of the British parliament formally limited the duration of topical aminoglycoside therapy to 7 days because of mounting evidence of ototoxicity.7 In 2002, Health and Welfare Canada advised against using all topical aminoglycoside agents in the presence of a TM defect or perforation because of increasing evidence for topical ototoxicity from surveillance and adverse drug reports.12 In the United States increasing concerns regarding ototoxicity, especially from topical aminoglycoside-containing preparations, led the American Academy of Otolaryngology–Head and Neck Surgery (AAO-HNS) to revisit, in March 2004, its previously endorsed position statement. A summary of the consensus panel’s recommendations is found in Appendix 2.

PORTAL OF ENTRY AND MECHANISM FOR OTOTOXICITY There is ample evidence that topical preparations of aminoglycosides, and gentamicin in particular, applied to the middle ear may enter the inner ear, especially in animal models. 13 The round window membrane (RWM) is the most likely portal of entry. Other proposed sites include the annular ligament of the stapes footplate and the otic capsule if congenital dehiscences or microfractures are present. Anatomical studies have shown that the RWM membrane in humans has an average thickness of 70 µm and is composed of three layers: outer cuboidal, middle fibrous-connective tissue, and inner mesothelial. All three layers contain micropinocytic vesicles for active substance transfer across the membrane.6 In addition, the RWM has been documented to be permeable to a wide variety of sub-

stances, including horseradish peroxidase, labeled electrolyte ions, I131, and radiolabeled albumin.14 In animal studies, several important principles regarding topical aminoglycoside absorption into the inner ear have been established. Harada and colleagues demonstrated that when neomycin was applied to the RWM the degree of cochlear damage appeared to be related to the duration of contact between neomycin and the RWM.15 In another study, Ikeda and Morizono demonstrated that the permeability of the RWM was increased in the presence of otitis media, making it more likely for ototoxic agents to gain entry into the inner ear.16 Moreover, when gentamicin was applied topically to the RWM, it appeared to enter the perilymph, albeit relatively slowly. Its clearance from the inner ear, however, appeared to be slower than its entry rate, suggesting that it may concentrate in the inner ear over time, allowing toxic levels to occur with prolonged use.6,17 Studies by Aran and coworkers from Bordeaux additionally identified the persistence of radiolabeled gentamicin in the inner ear 11 months after the initial treatment, suggesting a possible mechanism for increased toxicity from repeated treatment courses.18 There is also evidence for the systemic absorption of gentamicin when used topically in the management of CSOM. Lancaster and colleagues measured plasma gentamicin levels in 27 patients with active CSOM treated with the topical gentamicin-containing preparation Gentisone HC (Roche, Welwyn Garden City, Hertfordshire, UK). They noted that 7 (26%) of the patients studied had detectable plasma gentamicin levels and thereby concluded that systemic absorption of gentamicin from topical application could eventually be absorbed into the perilymph via the bloodlabyrinthine barrier.1 This observation provides an alternative route for the entry of a topical aminoglycoside into the perilymph, thus possibly potentiating any topical toxicity.

EXPERIMENTAL STUDIES IN ANIMALS There is little debate that topical aminoglycosides cause ototoxicity in animal models. Although all aminoglycosides may cause both cochleotoxicity and vestibulotoxicity, some (eg, gentamicin) appear more vestibulotoxic whereas others (eg, amikacin) appear more cochleotoxic.6 Middle ear placement of gentamicin in certain animal models has been shown to result in both histological and functional cochlear damage by primarily destroying outer hair cells (OHCs) in the basal turn, which corresponds to a high-frequency loss demonstrated by auditory brainstem response recordings and cochlear microphonics. 19,20 The effects of topical aminoglycosides on the vestibular system, however,

Topical Aminoglycoside Vestibular Toxicity

have not been as well investigated, despite the fact that gentamicin clinically is primarily vestibulotoxic.6 Regarding topical vestibular ototoxicity, Rudnick and colleagues demonstrated that maximum morphologic injury occurred in vestibular structures approximately 1 week after middle ear injections. This finding of delay correlates well with the delayed dizziness usually seen in humans following intentional topical aminoglycoside ototoxicity and after prolonged therapeutic use. 21 Omura and colleagues additionally demonstrated that intralabyrinthine injections of gentamicin markedly reduced posterior canal nerve ampullary potentials in a progressive fashion.6,22 Wanamaker and colleagues reported dose-related damage to the sensory hair cells in posterior crista and cochlea of Mongolian gerbils with transtympanic gentamicin injections.23 Although gentamicin is predominantly vestibulotoxic, the earliest detectable inner ear changes occur in the OHCs of the basal turn of the cochlea.24 This finding supports the theory that aminoglycosides gain access into the inner ear via the RWM; they would come into contact with hair cells in the basal turn before ascending through the cochlea to the labyrinth, where concentration may occur gradually because of slower clearing abilities.6,16 This would also explain the delayed presentation of vestibular toxicity that occurs usually after prolonged treatment. Extrapolating to humans from animal models that demonstrate ototoxicity with topical drops is less than ideal, given cross-species anatomical differences. The anatomical relationship between the TM and RWM, for example, differs significantly between humans and research animals such as guinea pigs and chinchillas.4 In humans, the RWM is deeper, thicker, and frequently covered by another mucous membrane or web formation.4 In addition, when the infected human ear is receiving treatment, pus or mucosal edema may cover the RWM and provide additional protection.

THE CASE FOR TOPICAL VESTIBULOTOXICITY: CLINICAL STUDIES IN MAN Case Reports and Series Although hearing loss and cochlear toxicity from topical aminoglycosides have been demonstrated sparingly in the English literature, 25–27 only recently has the vestibulotoxicity from these commercially available proprietary medications been documented.6,28,29 Perhaps the earliest association of vestibular dysfunction following topical gentamicin drop application was noted by LeLiever in 1985.30 In 1994 Longridge and colleagues reported two cases of acute vestibular deafferentation temporally related to Garasone (Schering Canada Inc, Pointe Claire, Quebec,

123

Canada) ear drops (1 mL contains 3 mg gentamicin sulfate and 1 mg betamethasone sodium phosphate). Both patients experienced oscillopsia, ataxia, and imbalance shortly after administration of the drops in the presence of a TM defect. Vestibular deafferentation was confirmed by ethyl chloride air caloric electronystagmography (ENG).31 In a slightly larger retrospective clinical series, Wong and Rutka in 1997 reported on five patients with TM perforations or defects who developed primary vestibulotoxicity after topical Garasone administration. The cases were instructional in that one patient developed acute vertigo after 3 weeks of treatment where the middle ear was dry. In another patient, unilateral highfrequency sensorineural hearing loss and vestibular loss were noted after prolonged treatment. The other three cases in the series each developed a bilateral vestibular loss confirmed by ENG following instillation of these drops over a prolonged treatment course of weeks to months.29 Topical vestibulotoxicity from another ototopical preparation, this time containing gentamicin and dexamethasone, was also reported by Abello and colleagues.32 All six reported cases presented with acute vertigo. Most had received topical treatment without medical supervision in the presence of a TM perforation. Treatment was prolonged and continued after their suppuration from the middle ear had cleared. Audiometric and vestibular examination revealed only vestibular involvement. Clinical recovery was spontaneous. Since the late 1990s much of the clinical incidence data linking topical aminoglycoside drops to vestibulotoxicity has arisen from the University of Toronto.33 In 1998, Marais and Rutka reported 9 (12 ears) welldocumented, incontrovertible cases of iatrogenic vestibulotoxicity in patients with TM perforations using topical Garasone ear drops. Six of the nine patients presented with an acute unilateral deafferentation, whereas three presented with a bilateral peripheral vestibular loss following treatment for bilateral middle ear pathology. This study underscored prolonged treatment in the presence of a dry middle ear as the major risk factor for the development of vestibulotoxicity. At the time of onset of vestibular symptoms, the mean duration of use of drops was 5.4 weeks, and affected patients had continued the drops on average for 18.4 days from cessation of otorrhea.6 A further update by Bath and colleagues documented topical ototoxicity from Garasone in 16 patients. All affected patients had either TM perforations or a tympanostomy (ventilation) tube in place. The earliest toxicity occurred around day 7 of drop use in an individual with a dry middle ear. This study included the case report of the intentional ablation of vestibular function in one patient with unilateral

124

Topical Toxicity

incapacitating Meniere’s disease using the same commercially available drops. The patient had a quantifiable normal pre-ablation caloric test response and had an absent response on caloric testing after treatment.28 Topical Gentamicin in the Treatment of Meniere’s Disease Vestibulotoxicity associated with systemic gentamicin administration has been a well-recognized phenomenon since its introduction in the 1960s. Paradoxically, because of this very property gentamicin has been used for the intentional topical ablation of vestibular function. Following initial reports by Beck and Schmidt,34 numerous authors have reported on the therapeutic use of intratympanic gentamicin in the treatment of unilateral Meniere’s disease.35–46 Both titration and fixed-schedule protocols evolved throughout this period in the hopes that undesirable side effects, primarily the worsening of sensorineural hearing, could be minimized. One of the initial fixed-schedule protocols for chemical ablation popularized by Nedzelski and colleagues applied topical concentrations of intratympanic gentamicin of approximately 25 mg/mL several times a day for 3 to 5 days. 47 Under this protocol patients invariably developed acute vertigo indicative of a vestibular deafferentation in a delayed fashion, sometimes as late as 1 to 2 weeks after treatment, suggesting a progressive concentration in the vestibular labyrinth. Because of concerns regarding the variability in dose response, however, treatment modalities appear to have become more conservative, with “low-dose” intratympanic gentamicin becoming more favored.48 Over the years the effects of topical gentamicin ablation therapy in the treatment of incapacitating Meniere’s disease have become well established, and consistent objective reductions in caloric activity have been demonstrated in the serial ENG testing of patients treated with this modality.47 Patients typically display a delayed onset of acute vertigo that parallels what is seen in cases of inadvertent topical vestibular toxicity. High-frequency hearing loss has also been identified in many as a common finding after gentamicin ablation therapy. 6 Although much higher concentrations of intratympanic gentamicin are employed for chemical ablation in Meniere’s disease than are found in commercially available drops (ie, around 25 mg/mL vs 3 mg/mL), the duration of treatment is shorter than the average duration of topical use identified in cases of inadvertent ototoxicity.6 Vestibular Ablation with Commercially Available Topical Gentamicin Drops in Meniere’s Disease Skepticism regarding the occurrence of topical vestibular toxicity was largely put to rest with the publication

of reports of intentional vestibular ablation in patients with Meniere’s disease using commercially available gentamicin ear drops. This provided irrefutable evidence for the potential vestibulotoxicity of topical gentamicin-containing preparations. As previously mentioned, Bath and colleagues in 1999 had reported the first case in the world literature of deliberate ablation of vestibular function in unilateral Meniere’s disease with gentamicin-containing ear drops (Garasone).8 Their initial report identified that Garasone ear drops, and by implication all topical aminoglycoside ear drops, could be vestibulotoxic in the presence of a TM defect. The patient had incapacitating unilateral Meniere’s disease with a normal pretreatment ENG caloric test and became symptom free after instilling Garasone (2 drops bid) into her ear using subsequent tragal pressure to “pump” the drops into the middle ear for 3 weeks. A ventilation tube had been previously inserted. During treatment, the patient experienced continued attacks of intermittent vertigo associated with decreased hearing from her Meniere’s disease. After finishing the prescribed course of ear drops, her attacks of vertigo stopped. Repeat ENG with air caloric stimulation at 10°C using the Dundas Grant method revealed absent vestibular function on the affected side. Kaplan and colleagues provided further evidence of topical ear drop ototoxicity in a prospective study of 20 patients who underwent intentional vestibular ablation for unilateral Meniere’s disease using Garasone ear drops.33 Tympanostomy tubes were inserted under local anesthesia, and patients were instructed to instill the same gentamicin-containing ear drops tid until they became acutely vertiginous for longer than 24 hours and then for an additional 2 days, or for 1 month if they did not become vertiginous, whichever came first. Although the methodology was slightly inconsistent (ENG with caloric testing was measured before treatment using bithermal water calorics and after treatment using air caloric tests owing to the tympanostomy tubes), most patients developed symptoms in the first 2 weeks after starting treatment (range 9 –28 days; mean 15 days). On repeat ENG testing, 15 (75%) patients were identified to have had a significant reduction in caloric excitability difference (ED) compared with pretreatment values. Ten (50%) patients had absent air caloric test responses on the treated side. These findings were almost identical to the ED results when a more concentrated solution of intratympanic gentamicin (25 mg/mL) was used initially for a shorter period. They concluded that the use of a less concentrated form of topical gentamicin for a longer period of time would yield the same results (Figure 13-1).7,33 Kaplan and colleagues, in their study of ototoxic events related to topical gentamicin therapy, which included a clinical history of prolonged vertigo lasting


125

individual. 6 Labyrinthine penetration by topical aminoglycosides is dependent on both the concentration placed in the middle ear and the duration of contact.17 Local factors that may prevent or influence the duration of contact of a topical aminoglycoside with the RWM include the following6,7: 1. The site and size of TM perforation 2. Presence of edematous mucosa, granulation tissue, pus, and debris, for example, that protect the RWM from drops 3. A patent eustachian tube that allows drainage of drops into nasopharynx 4. Barriers around the RWM, such as pseudomembranes, webs, and adhesions 5. Varying degrees of RWM thickness 6. The presence of exotoxins within the middle ear, which may accelerate diffusion and active transport by pinocytosis across the RWM 7. Inherent susceptibility or resistance to aminoglycosides

Figure 13-1 Pre- and posttreatment caloric responses in 20

Meniere’s disease patients treated with Garasone ablation protocol through a tympanostomy tube. Caloric excitability difference (ED) along x-axis. Adapted from Kaplan DM et al.33

days during the treatment, also found posttreatment signs of vestibular deafferentation (eg, the presence of post–head-shake nystagmus or a positive head-thrust, or Halmagyi, maneuver [see Chapter 2, “Physiology of the Vestibular System”]), in conjunction with changes in the pre- versus posttreatment audiometry and ENG caloric testing.33 Although falling short of a definitive doubleblind randomized control trial (RCT), this prospective cohort study provided conclusive evidence that commercially available gentamicin ototopical preparations penetrate the middle ear via a tympanostomy tube, where they can be absorbed into the inner ear and subsequently result in ototoxicity if used for a prolonged time.

TOPICAL OTOTOXICITY: INDIVIDUAL VARIABILITY? There is little doubt that when topical gentamicin is applied to the middle ear it can not only be absorbed into the inner ear, it can also result in severe vestibular and cochlear damage. Topical gentamicin (and by extension all other topical aminoglycosides) ototoxicity from ear drops probably occurs relatively infrequently, considering the number of prescriptions per year. Nevertheless it is not negligible. Many factors probably play a role in the development of ototoxicity and whether it will be clinically manifest in a given

Many patients, especially those with a unilateral peripheral vestibular loss, probably do not seek further investigation or treatment for ototoxicity because the condition reverses itself or the central nervous system compensates for it. Patients may also accept some dizziness as the inevitable consequence of an ear infection and do not report it.6 Despite seeking treatment, cases of bilateral topical vestibulotoxicity may not be readily recognized by physicians, as it is not widely appreciated that the features of a bilateral peripheral vestibular loss are those of ataxia, oscillopsia, and imbalance, not vertigo. Unfortunately, many physicians equate ototoxicity solely with hearing loss, not vestibular loss per se, and therefore tend not to ascribe much importance to complaints of imbalance.6

SUMMARY • Vestibular ototoxicity as a result of topical aminoglycoside use is relatively rare. It is not negligible, however, and it is documented in the world literature and warned against by regulatory bodies such as Health and Welfare Canada, the Committee for Safety of Medicines (United Kingdom), and the Medicine Control Agency (United Kingdom). • Although infrequent, topical vestibulotoxicity probably occurs far more commonly than is realized, given the widespread use of aminoglycoside-containing ear drops. • A bilateral peripheral vestibular loss presents with features of ataxia, oscillopsia, and imbalance, not vertigo. • Topical aminoglycoside ear drops should be used for as short a duration as possible. Patients should

126

Topical Toxicity

clearly understand that they should stop treatment as soon as the otorrhea has cleared. Dosage and method of administration should be clearly explained to patients, and they should be advised to stop treatment immediately if any dizziness, tinnitus, or hearing loss occurs. Patients with open middle ear spaces should be reviewed after 1 week to assess the need for further drops. • Nonototoxic topical preparations (eg, fluoroquinolones) are now widely available and have efficacy shown to be comparable with topical aminoglycosides. Increased use of these agents in the management of active CSOM is anticipated.49–51

REFERENCES 1. Lancaster JL, Mortimore S, McCormick M, Hart CA. Systemic absorption of gentamicin in the management of active mucosal chronic otitis media. Clin Otolaryngol 1999;24:435–9. 2. Yuen APW, Chau PY, Wei WI. Bacteriology of chronic suppurative otitis media: ofloxacin susceptibility. J Otolaryngol 1995;24:206–8. 3. Gehanno P, Cohen B. Effectiveness and safety of ofloxacin in chronic otitis media and chronic sinusitis in adult outpatients. Eur Arch Otorhinolaryngol 1993;250:13–4. 4. Lancaster JL, Makura ZGG, Porter G, McCormick M. Topical aminoglycoside in the management of active mucosal chronic suppurative otitis media. J Laryngol Otol 1999;113:10–2. 5. Browning GG, Gatehouse S, Calder IT. Medical management of active chronic otitis media: a controlled study. J Laryngol Otol 1988;102:491–5. 6. Marais J, Rutka JA. Ototoxicity and topical eardrops. Clin Otolaryngol 1998;23:360–7. 7. Rutka JA. Update on topical ototoxicity in chronic suppurative otitis media. Ear Nose Throat J 2002; 8 Suppl 1:18–9. 8. Acuin J, Smith A, Mackenzie I. Cochrane Ear, Nose and Throat Disorders Group. Interventions for chronic suppurative otitis media [systematic review]. Cochrane Database Syst Rev 2000;(2): CD000473. 9. Bickerton RC, Roberts C, Little JT. Survey of general practitioners’ treatment of the discharging ear. Br Med J 1988;296:1649–50. 10. Roland PS. Clinical ototoxicity of topical antibiotic drops. Otolaryngol Head Neck Surg 1994;110: 598–602. 11. Walby P, Stewart R, Kerr AG. Aminoglycoside eardrop ototoxicity: a topical dilemma? Clin Otolaryngol 1998;23:289–90. 12. Health Canada. Aminoglycoside eardrops and ototoxicity. Canadian Adverse Drug Reaction Newsletter 1997;7(2).

13. Smith SM, Myers MG. The penetration of gentamicin and neomycin into perilymph across the round window membrane. Otolaryngol Head Neck Surg 1979;87:888–91. 14. Goycoolea MV, Muchow D, Schachern P. Experimental studies of round window structure, function and permeability. Laryngoscope 1988;98 Suppl 44:1–20. 15. Harada T, Iwamori M, Nagai Y. Ototoxicity of neomycin and its penetration through the round window membrane into the perilymph. Ann Otol Rhinol Laryngol 1986;95(4 Pt 1):404–8. 16. Ikeda K, Morizono T. Changes in the permeability of the round window membrane in otitis media. Arch Otolaryngol Head Neck Surg 1998;114: 895–7. 17. Tran BA, Huy P, Meulemans A, et al. Gentamicin persistence in rat endolymph after two day constant infusion. Antimicrob Agents Chemother 1983;23:344–6. 18. Aran JM, Erre JP, Lima da Costa D, et al. Acute and chronic effects of aminoglycosides on cochlear hair cells. Ann N Y Acad Sci 1999;884:60–8. 19. Kohonen A, Tarkannen J. Cochlear damage from ototoxic antibiotics by intratympanic application. Acta Otolaryngol 1969;68:90–7. 20. Morozino T, Johnstone BM. Ototoxicity of topically applied eardrops: statistical analysis of electrophysiological measurement. Acta Otolaryngol 1975;80:389–93. 21. Rudnick MD, Ginsberg IA, Huber PS. Aminoglycosides ototoxicity following middle ear infection. Ann Otol Rhinol Laryngol 1980;89(6 Pt 4): i–iii, 1–28. 22. Omura R, Harada Y, Suzuki M. Structural and physiological change in the bullfrog semicircular canals due to gentamicin intoxication. Acta Otolaryngol Suppl 1989;468:41–7. 23. Wanamaker HH, Gruenwald L, Damm KJ, et al. Dose-related vestibular and cochlear effects of transtympanic gentamicin. Am J Otol 1998; 19:170–9. 24. Wright CG, Halama AR, Meyerhof WL. Ototoxicity of an ototopical preparation in a primate. Am J Otol 1987;8:56–60. 25. Lind O, Kristiansen B. Deafness after treatment with ear drops containing neomycin, geramicidin and germathasone. A case report. ORL J Otorhinolaryngol Relat Spec 1986;48:52–4. 26. Murphy KW. Deafness after topical neomycin. Br Med J 1970;2:114. 27. Tommerkup B, Moller K. A case of profound hearing impairment following the primary use of framycetin ear drops. J Laryngol Otol 1984; 98:1135–7.


28. Bath AP, Walsh RM, Bance ML, Rutka JA. Ototoxicity of topical gentamicin preparations. Laryngoscope 1999;109(7 Pt 1):1088–93. 29. Wong DLH, Rutka JA. Do aminoglycoside otic preparations cause ototoxicity in the presence of tympanic membrane perforations? Otolaryngol Head Neck Surg 1997;116:404–10. 30. Leliever WC. Topical gentamicin induced positional vertigo. Otolaryngol Head Neck Surg 1985;93:553–5. 31. Longridge NS. Topical gentamicin vestibular toxicity. J Otolaryngol 1994;23:444–6. 32. Abello P, Vinas JB, Vega J. Topical ototoxicity: review over a 6-year period. Acta Otorrinolaringol Esp 1998;49:353–6. 33. Kaplan DM, Hehar SS, Bance ML, Rutka JA. Intentional ablation of vestibular function using commercially available topical gentamicinbetamethasone eardrops in patients with Meniere’s disease: further evidence for topical eardrop ototoxicity. Laryngoscope 2002;112:689–95. 34. Beck C, Schmidt CL. Ten year experience with intratympanically applied streptomycin (gentamicin) in the therapy of morbus Meniere’s. Arch Otorhinolaryngol 1978;221:149–52. 35. Schmidt CL, Beck C. Treatment of morbus Meniere’s with intratympanically applied gentamicin. Laryngorhinootologie 1980;59:804–7. 36. Odkvist LM. Middle ear ototoxic treatment for inner ear disease. Acta Otolaryngol Suppl 1988; 457:83–6. 37. Youssef, Poe Ds. Intratympanic gentamicin injection for the treatment of Meniere’s disease. Am J Otol 1998;19:435–42. 38. Kaasinen S, Pryykko I, Ishizaki H. et al. Intratympanic gentamicin in Meniere’s disease. Acta Otolaryngol 1998;118:294–8. 39. McFeely WJ, Singleton GT, Rodriguezet FJ, Antonelli PJ. Intratympanic gentamicin in Meniere’s disease. Otolaryngol Head Neck Surg 1998;118:589–96. 40. Minor LM. Intratympanic gentamicin for control of vertigo in Meniere’s disease: vestibular signs that

127

specify completion of therapy. Am J Otol 1999; 20:209–19. 41. Silverstein H, Arruda J, Rosenberg SI, et al. Direct round window membrane application of gentamicin in the treatment of Meniere’s disease. Otolaryngol Head Neck Surg 1999;120:649–55. 42. Atlas JT, Parnes LS. Intratympanic gentamicin titration therapy for intractable Meniere’s disease. Am J Otol 1999;20:357–63. 43. Kaplan DM, Chen JM, Nedzelski JM, Shipp DB. Intratympanic gentamicin for the treatment of unilateral Meniere’s disease: a ten year experience using a predetermined treatment regimen. Laryngoscope 2000;110:1298–305. 44. Murofishi T, Halmagyi GM, Yavor RA. Intratympanic gentamicin in Meniere’s disease: result of therapy. Am J Otol 1997:18:52–7. 45. Harner SG, Kasperbauer JL, Facer GW, Beatty CW. Transtympanic gentamicin for Meniere’s syndrome. Laryngoscope 1998;108:1446–9. 46. Blakely BW. Update on intratympanic gentamicin for Meniere’s disease. Laryngoscope 2000; 100:236–40. 47. Nedzelski JN, Bryce GE, Pfleiderer AG. Treatment of Meniere’s disease with topical gentamicin: a preliminary report. J Otolaryngol 1992;21:95–101. 48. Bath AP, Walsh RM, Bance ML. Presumed reduction of vestibular function in unilateral Meniere’s disease with aminoglycoside eardrops. J Laryngol Otol 1999;113:916–8. 49. Gehanno P, Cohen B. Effectiveness and safety of ofloxacin in chronic otitis media and chronic sinusitis in adult outpatients. Eur Arch Otorhinolaryngol 1993;250:13–4. 50. Sumitsawan Y, Tharavichikul P, Prawatmuang W, et al. Ofloxacin otic solution as treatment of chronic suppurative otitis media and diffuse bacterial otitis externa. J Med Assoc Thai 1995;78:455–9. 51. Jones RN, Milazzo J, Seidlin M. Ofloxacin otic solution for treatment of otitis externa in children and adults. Arch Otolaryngol Head Neck Surg 1997; 123:1193–200.

CHAPTER 14

Chloramphenicol, Colymycin, and Polymyxin Leonard P. Rybak, MD, PhD, and Srinivasan Krishna, MD, MPH

Polymyxins, colymycin, and chloramphenicol (alone or in combination with other antibiotic agents, steroids, antifungals, and antiseptics) were historically among the first topical antibiotic agents to be used in the treatment of various ear diseases, ranging from external otitis and chronic suppurative otitis media to post– tympanostomy tube otorrhea.

POLYMYXINS History The polymyxins were discovered in 1947. They form a group of closely related antibiotic substances produced by strains of Bacillus polymyxa, an aerobic spore-forming rod found in soil. Colistin, also known as polymyxin E, is produced by Bacillus colistinus, a microorganism found in soil samples from Fukushima Prefecture in Japan.1 Mechanism of Action The polymyxins are cationic surface-active compounds (detergents) at physiologic pH. They are relatively simple, basic peptides with molecular weights of about 1,000. They produce their antibacterial effect by interacting with phospholipid components of the cytoplasmic membrane of susceptible bacteria, disrupting membrane integrity, causing osmotic imbalance and death of the bacterial organism.2 The antimicrobial activity of polymyxins is directed toward gram-negative bacteria, including Enterobacter sp, Escherichia coli, Klebsiella sp, Salmonella sp, Shigella sp, Pasteurella sp, Bordetella sp, Vibrio sp, Haemophilus sp, and Pseudomonas aeruginosa. Proteus sp, Providencia sp, and Serratia marcescens are usually resistant.2 Sensitivity to polymyxin B appears to be related to the content of phospholipid in the cell wall–membrane complex. The cell wall of certain resistant bacteria may prevent access of the drug to the cell membrane.1 Polymyxins are not active against Neisseria, gram-positive bacteria, most obligate anaerobes, or fungi.2

Polymyxin B binds to the endotoxin in the outer membrane of gram-negative bacteria and inactivates this molecule. 2 Polymyxins are poorly absorbed through the gastrointestinal tract, through mucous membranes, or through burns. Therapeutic Uses Parenteral forms of polymyxin B or colistimethate sodium are extremely limited in their indications. They are not considered drugs of choice for systemic infections. These drugs are primarily used for infections of the skin, mucous membranes, eye, and ear caused by organisms sensitive to polymyxin B. The drug is usually applied as an ointment, solution, or suspension. It may be curative for external otitis or corneal ulcers caused by P. aeruginosa. Because resistance to antibiotics can develop rapidly in P. aeruginosa, a prospective study of the susceptibilities of aural isolates in 231 consecutive children seen in the outpatient Pediatric Otolaryngology Department at Children’s Hospital of Pittsburgh during the years 1992 to 1993 was carried out. Only 18% of isolates were sensitive to neomycin. Conversely, 99.6% were sensitive to polymyxin B, 97.4% were sensitive to colistin, and 98.3% were sensitive to norfloxacin.3 Proprietary Agents Containing Polymyxins The American Medical Association drug evaluations2 list several otic preparations containing polymyxin B as one of the antimicrobial ingredients. These include the following: 1. Coly-Mycin S Otic with neomycin and hydrocortisone suspension contains colistin sulfate, neomycin sulfate, acetic acid, hydrocortisone acetate, thonzonium bromide, trimerosal, and polysorbate 80. 2. Cortisporin Otic Solution contains neomycin sulfate, polymyxin B, hydrocortisone, cupric

Chloramphenicol, Colymycin, and Polymyxin

3.

4.

5. 6.

7.

8.

sulfate, glycerin, propylene glycol, and potassium metabisulfate. Cortisporin Otic Suspension contains neomycin sulfate, polymyxin B, hydrocortisone, cetyl alcohol, propylene glycol, polysorbate 80, and thimerosal. Pedi-Otic Suspension contains the same ingredients as Cortisporin Otic Suspension with the exception of polysorbate 80, which is not present in Pedi-Otic Suspension. Otobiotic Solution contains polymyxin B, hydrocortisone, and propylene glycol. Otocort Solution contains neomycin sulfate, polymyxin B, hydrocortisone, propylene glycol, glycerin, and potassium metabisulfate. Otocort Suspension contains neomycin sulfate, polymyxin B, hydrocortisone, propylene glycol, cetyl alcohol, polysorbate 80, and thimerosal. (Several other proprietary ear drops contain neomycin, polymyxin B, and hydrocortisone.) Pyocidin-Otic Solution contains polymyxin B, hydrocortisone, and propylene glycol.

Toxicities Parenteral polymyxins may cause serious nephrotoxicity and neurotoxicity. These toxic side effects and the availability of effective and less toxic alternative antimicrobial agents have resulted in the limited usefulness of polymyxins for systemic use. Animal Studies Guinea pigs have been used to study the ototoxicity of neomycin, polymyxin B, and colymycin (colistin).4 Concentrations of 1 to 25 mg/mL of polymyxin B and 2 to 10 mg/mL of colistin were instilled into the middle ear. Polymyxin B at concentrations greater than 1 mg/mL resulted in hair cell degeneration. Higher concentrations caused total degeneration of the organ of Corti. The application of 5 mg/mL or greater concentrations of colistin resulted in slight to nearly complete destruction of hair cells in the majority of cochleas. A 0.1 M solution of polymyxin B was applied intratympanically in guinea pigs daily for 3 days. The cochlea was examined by light microscopy of surface preparations harvested 2 days following the last dose. A 50% hair cell loss was observed following this treatment. Polymyxin B has been shown to cause cochlear damage and loss of cochlear function in guinea pigs following application to the middle ear space. Solutions containing polymyxin B in concentrations similar to those found in commercially available otic drops were found to cause ototoxicity, as revealed by changes in the ability of the cochlea to generate the alternating current (AC) cochlear potential, and morphologic changes, as

129

demonstrated by the loss of inner and outer hair cells on surface preparations of the cochlea examined microscopically. These changes were observed only in the treated ears. The contralateral control ears were found to have normal morphology.5 Polymyxin B in combination with neomycin (Cortisporin Otic Suspension) was found to be highly ototoxic in the chinchilla. Following the application of 0.5 mL of Cortisporin Otic Suspension to the middle ear of these experimental animals, degeneration of all inner and outer hair cells throughout the cochlea, severe damage to the stria vascularis, and moderate to severe degeneration of the vestibular receptor organs were observed.6 Unfortunately, this study was performed with a mixture of at least two ototoxic drugs (neomycin and polymyxin B), and the solvent, propylene glycol, may also be ototoxic. Therefore, it is not clear whether the ototoxicity observed was primarily caused by polymyxin B, neomycin, or their combination with propylene glycol. Subsequent studies of the individual components, neomycin and polymyxin B, applied as solutions in the middle ear of the chinchilla and the baboon, were carried out. In both the rodent and the primate, the extent of hair cell loss and injury to the stria vascularis following application of polymyxin B alone was comparable to that previously reported after application of Cortisporin Otic Suspension.7 Although both species suffered hair cell loss after polymyxin B, the extent of the hair cell loss was much greater in the chinchilla than in the baboon.8 Polymyxin B was also perfused at a concentration of 1 mM through the scala tympani of the guinea pig cochlea, and cochlear potentials were monitored. Both cochlear microphonics (CM) and the endocochlear potential were altered, but in an independent manner. These data suggested that both the organ of Corti and the stria vascularis are involved in ototoxicity caused by polymyxin B.7 Colymycin applied to the middle ear of the chinchilla resulted in a dose-related elevation of compound action potential (CAP) thresholds that affected the higher frequencies more than the lower frequencies and progressed over time, resulting in hearing loss at all frequencies by 24 hours.9 A comparative study of various otic preparations has been carried out in the guinea pig.10 These preparations were instilled daily for 7 days into the bulla of juvenile guinea pigs. Fourteen days after treatment the animals were sacrificed, and the organ of Corti was examined using light and scanning electron microscopy. The average hair cell loss was 66% following the application of Cortisporin Otic Solution. Fosfomycin has previously been studied as a possible protective agent against polymyxin B ototoxicity.11 The former is an antibiotic that had been previously shown to reduce the ototoxicity of aminoglycoside

130

Topical Toxicity

antibiotics in animals. Two groups of chinchillas were tested. One group received polymyxin B solution alone in the middle ear cavity, whereas the second group was administered polymyxin B in combination with fosfomycin. Cochlear damage was dramatically reduced in the animals receiving the combination of fosfomycin with polymyxin B. Human Studies Only a few case reports document clinical evidence for ototoxicity attributable to topical otic preparations. Gyde performed a double-blind randomized study to compare the efficacy and safety of trimethoprimpolymyxin B (TP) and trimethoprim-sulfacetamidepolymyxin B (TSP) drops in the treatment of otorrhea.12 Sixty-eight patients were treated with varying pathologies and conditions such as external otitis, recurrent otitis media with tympanic membrane (TM) perforation, infected mastoid cavities, and postoperative tympanoplasty. The TP ototopical solution was successful clinically in 61% of cases, compared with 89% of cases treated with TSP, a statistically significant difference using the chi-square with Yates correction statistical method. There were no signs of ototoxicity in either group.12 In addition, Rakover and colleagues reported no change in auditory thresholds in children with tympanostomy tubes who were treated with neomycin or polymyxin B ear drops for 2 weeks.13 Merrifield and colleagues studied 44 pediatric patients with tympanostomy tubes who were treated with potentially ototoxic ear drops.14 Pre- and posttreatment bone-conduction audiometry was carried out. The ear drops used included Cortisporin Otic Suspension, a gentamicin ophthalmic solution, Cortisporin Ophthalmic Solution, Pedi-Otic Suspension, and ColyMycin S Otic. The duration of treatment ranged from 2 days to 2 weeks. No significant bone-conduction threshold shifts were detected in 70 ears tested. The authors concluded that no significant sensorineural hearing loss occurred as a result of otic drop treatment in the presence of ventilation tubes with purulent otitis media.14 Welling and colleagues treated 446 patients with neomycin, polymyxin B, or gentamicin drops after ventilation tube insertion.15 Only a single dose of ear drops was administered, but no auditory toxicity was detected. To date, 10 cases of ototoxicity have been attributed to the topical use of neomycin/polymyxin B drops in the presence of a TM perforation. Dumas and colleagues reported eight cases.16 Lindner and colleagues reviewed the charts of 134 patients with possible antibiotic-related ototoxicity.17 They found that two patients had bilateral profound sensorineural hearing loss attributable to excessive administration of ear drops containing framycetin (neomycin B) and polymyxin in the presence of a TM perforation. The

authors concluded that although ototopical preparations have been widely used in the presence of TM perforations, they appear to rarely induce sensorineural hearing loss.17 Recommendations for Safe Use When it is possible clinically, the physician should use topical antibiotic preparations that do not have the potential for ototoxicity when treating on open middle ear space or mastoid cavity. If it is necessary to use a potentially ototoxic preparation in an infected ear, its use should be discontinued promptly after resolution of the infection. In the latter instance, the patient should be forewarned of the potential risk of ototoxicity. If any symptoms of ototoxicity appear, such as hearing loss, new-onset tinnitus, or dizziness, the patient should promptly notify the physician so that appropriate measures can be instituted.

CHLORAMPHENICOL History Chloramphenicol was first isolated in 1947, from Streptomyces venezuelae, an organism isolated from a soil sample in Venezuela.18 It was synthesized in 1949, becoming one of the first completely synthetic antibiotics of importance to be produced commercially. Mechanism of Action Chloramphenicol inhibits the peptidyl transferase step of protein synthesis by binding reversibly to the 50S ribosomal subunit of bacterial cells.18 It can also inhibit mitochondrial protein synthesis in mammalian cells, with erythropoietic cells being particularly sensitive. Chloramphenicol has a wide antibacterial spectrum against aerobic and anaerobic gram-positive and gram-negative organisms. It is primarily bacteriostatic but may be bactericidal to certain species, such as Haemophilus influenzae, Neisseria meningitidis, Streptococcus pneumoniae, and Bacteroides fragilis. 18 Most strains of E. coli, Proteus sp, and Klebsiella pneumoniae are susceptible. P. aeruginosa is very resistant, whereas most Rickettsia spp, Mycoplasma spp, and Chlamydia spp are sensitive to chloramphenicol.19 The drug is rapidly absorbed from the gastrointestinal tract and is metabolized by the liver. It is well distributed in body fluids with concentrations in the cerebrospinal fluid reaching 60% of plasma concentrations in the presence of meningitis. Resistance to chloramphenicol is caused by a plasmid-encoded acetyltransferase that inactivates the drug.19 Therapeutic Uses Systemic indications for chloramphenicol are extremely limited because of its toxicities. It is generally reserved for patients refractory to less toxic anti-


bacterial agents. In the past, it has been used in the treatment of typhoid fever, bacterial meningitis, anaerobic infections (brain abscesses, abdominal infections, etc), rickettsial diseases, and brucellosis. Topical chloramphenicol is still used, either as a powder or as drops, in therapy of refractory ophthalmic and otologic infections, such as mastoid bowl infections, chronic refractory otitis media, and chronic refractory otitis externa, particularly in developing countries.20 It is particularly effective against B. fragilis.21 Proprietary Agents Containing Chloramphenicol Chloromycetin Otic contains 0.5% chloramphenicol and propylene glycol.21 Toxicities The most significant systemic adverse reaction is bone marrow suppression. This can be a dose-dependent toxicity resulting in anemia, leukopenia, or thrombocytopenia or an idiosyncratic response resulting in fatal pancytopenia.18 Animal Studies The toxicity of topical chloramphenicol has been studied extensively in animal experiments. In 1963, Patterson and Gulick found progressive loss of the CM responses in guinea pigs after topical application of a relatively low dose of chloramphenicol to the round window membrane for 30 minutes.22 The losses ranged from 20 to 30 dB and were irreversible. Gulick and Patterson in 1964 again reported that the decrease in CM potentials did not return to predrug levels for up to 60 hours after topical application of chloramphenicol to the round window membrane of cats.23 Thus the effects seem to be long lasting. Proud and colleagues in 1968 studied the effects of two different concentrations of chloramphenicol succinate (1.44 M vs 0.72 M) in a single application to the round window niche for 30 minutes in 22 guinea pigs.19 These were compared with eight control animals that had round window applications of sodium succinate solutions at the same concentrations and pH. All animals were euthanized at 3, 6, 9, and 24 hours after the application. Histopathology of the temporal bones was performed. In the experimental animals, the cochlea showed marked destruction of the outer and inner hair cells, as well as the supporting cells (Deiters’ cells, Claudius’ cells, Hensen’s cells, and a few inner sulcus cells) of the basal turn of the cochlea only, and not in the other turns.19,24,25 Additionally, the stria vascularis of the basal turn showed variable but consistent damage in the experimental animals but not in the control animals. The results were similar regardless of the concentration of the drug used or the time of sacrifice after application. The dosages of the drug used were reportedly much

131

lower than comparable human doses, and the time of exposure was much less than in clinical situations, and yet the toxicity was significant. Hence, the authors concluded that chloramphenicol succinate is not well suited for topical application to the middle ear.19 D’Angelo and colleagues applied chloramphenicol powder to the round window membrane of guinea pigs and measured the cochlear responses to various tonal stimuli. They found that the responses were markedly reduced in all frequencies.26 In 1966, Koide and colleagues published a study wherein they injected chloramphenicol solution into the middle ear of guinea pigs. After several days, the cochleas were harvested and studied. They found that the activity of oxidizing enzymes such as succinic dehydrogenase and diphosphopyridine nucleotide diaphorase was much reduced in the outer hair cells of the basal turn of the cochlea. The inner hair cells, supporting cells, and nerve fibers were not affected.24 In 1975, Morizono and colleagues observed that chloramphenicol dissolved in propylene glycol to attain final concentrations of 1% or greater and instilled in the tympanic bulla of guinea pigs for more than 1 day caused a decrease in CM potentials in both higher and lower frequencies.27 However, chloramphenicol sodium succinate dissolved in Ringer’s solution was not toxic until the 5% concentration was reached. They concluded that propylene glycol used as a solvent for chloramphenicol is ototoxic by itself and worsens the ototoxicity of chloramphenicol. Systemic chloramphenicol has also been shown to potentiate noise-induced hearing loss in rats.28,29 Human Studies Ototoxicity from systemic administration of chloramphenicol was first reported by Gargye and Dutta in 1959.30 They reported a case of irreversible suddenonset nerve deafness following high doses of intravenous chloramphenicol. Subsequently, Svenungsson and colleagues, in 1976, described eight cases of unilateral hearing loss following chloramphenicol therapy for meningitis.31 Iqbal and Srivatsav, in 1984, reported a case of a 20-year-old woman with bilateral profound sensorineural hearing loss following the systemic administration of chloramphenicol.32 Initially, unilateral hearing loss and tinnitus were noted about 2 months after completion of systemic therapy with chloramphenicol for typhoid. The hearing loss was gradually progressive. Subsequently, after a repeat course of the drug 2 months later, the other ear was also affected, and the patient finally developed profound bilateral sensorineural hearing loss. Joy and colleagues in 1960 reported optic and peripheral neuritis as a probable effect of prolonged chloramphenicol therapy,33 whereas Mortimer reported bilateral optic atrophy from its use.34

132

Topical Toxicity

The first clinical case report of hearing loss following topical application of chloramphenicol ear drops was from Matsumaru in 1964.35 He reported on a 45-year-old man who experienced a severe case of deafness, tinnitus, and dizziness immediately after application of a few drops of Chloromycetin ear drops (0.5% chloramphenicol, 1% solution of ethylaminobenzoate in 100% propylene glycol) for chronic suppurative otitis media. However, this report could not conclude that the ototoxicity was specifically from the chloramphenicol and not from the additives, such as the propylene glycol. Supiyaphun and colleagues in 2000 published a study comparing the safety and efficacy of 0.3% ofloxacin drops twice daily and that of oral amoxicillin thrice daily plus 1% chloramphenicol otic drops thrice daily, in a 2-week course for acute exacerbation of chronic suppurative otitis media.36 This was a prospective, randomized, single-blinded study of 80 adult patients at the Chulanglongkorn University Hospital in Thailand. All patients were 15 years of age or older and had otorrhea with a central TM perforation for at least 21 days. Pure tone and speech audiometry was performed in all patients before treatment and at day 14 posttreatment. Ototoxicity was defined as an elevation in bone-conduction threshold or speech reception threshold of greater than 5 dB or the presence of a high-frequency hearing loss following treatments. There was a significant deterioration in mean boneconduction threshold, from 22.8 ± 10.4 dB to 24.8 ± 10.4 dB (p = .007), in the amoxicillin plus chloramphenicol ears.39 The change in speech reception threshold was not statistically significant, however. There was a significantly higher rate of ototoxicity in the amoxicillin plus chloramphenicol group than in the ofloxacin group. Since systemic amoxicillin has not been shown to have any ototoxicity, the change in hearing may be attributable to the chloramphenicol otic drops. In addition, the chloramphenicol drops appeared to cause mild to moderate soreness in up to 37% of the patients in this study. Overall, topical chloramphenicol appears to have significant risk of ototoxicity, particularly if applied to the middle ear near the round window niche or membrane. This would preclude its use as a first-line ototopical agent in the treatment of chronic otitis media with a perforated TM. Safer alternatives are available. Topical chloramphenicol, however, may have a role in treating select situations of recalcitrant mastoid bowl infections, particularly after other agents have not been successful. Of cautionary note, two cases of fatal aplastic anemia and two cases of reversible bone marrow suppression following the use of topical chloramphenicol in the eye have been reported.37 This point merits consideration when using topical choramphenicol prepara-

tions in the ear. Skin reactions including erythema multiforme-like eruptions, have also been identified to result from topical chloramphenicol.38

SUMMARY • Polymxins, colymycin, and chloramphenicol (alone or in combination with other antibiotics, solvents, antifungals, and steroids) are commonly found in many commercially available ototopical medications. • Polymyxins have a broad spectrum of antibacterial activity that especially targets gram-negative organisms such as P. aeruginosa, E. coli, and Klebsiella sp. Chloromycetin appears to have limited activity against P. aeruginosa but is bactericidal toward H. influenzae, N. meningitidis, S. pneumoniae, and B. fragilis. • Animal studies have consistently demonstrated their potential for causing topical ototoxicity upon reaching the inner ear. Human studies and clinical series have demonstrated that topical ototoxicity occurs but is relatively infrequent unless used in a prolonged fashion (more than 7 days). • When clinically possible, topical preparations that do not have the potential for ototoxicity should be used to treat ear disease in the presence of a TM defect or perforation. • To date there have been no reports of bone marrow depression or aplastic anemia in patients receiving topical chloramphenicol for treatment of their ear disease. There have, however, been two reported cases of fatal aplastic anemia and two cases of reversible bone marrow suppression following the use of topical chloramphenicol in the eye.

REFERENCES 1. Sande MA, Mandell GL. Antimicrobial agents: tetracyclines, chloramphenical, erythromycin and miscellaneous antibacterial agents. In: Gilman AG, Rall TW, Nies AS, Taylor P, editors. Goodman and Gilman’s the pharmacological basis of therapeutics. 8th ed. New York: Pergamon Press; 1990. p. 1138. 2. American Medical Association. Drug evaluations annual. Chicago: American Medical Association; 1993. p. 1549–52. 3. Dohar JE, Kenna MA, Wadowsky RM. In vitro susceptibility of aural isolates of Pseudomonas aeruginosa to commonly used ototopical antibiotics. Am J Otol 1996;17:207–9. 4. Kohonen A, Tarkannen J. Cochlear damage from ototoxic antibiotics by intratympanic application. Acta Otolaryngol 1969;68:907. 5. Brummett RE, Harris RF, Lindgren JA. Detection of ototoxicity from drugs applied topically to the middle ear space. Laryngoscope 1976;86:1177–87.


6. Wright CG, Meyerhoff WL. Ototoxicity of otic drops applied to the middle ear in the chinchilla. Am J Otolaryngol 1984;5:166–76. 7. Halama AR, Wright CG, Meyerhoff WL. Ototoxicity of an ototopic preparation—experimental results and clinical facts. Acta Otorhinolaryngol Belg 1991;45:279–82. 8. Wright CG, Meyerhoff WL, Halama AR. Ototoxicity of neomycin and polymyxin B following middle ear application in the chinchilla and baboon. Am J Otol 1987;8:495–9. 9. Morizono T. Toxicity of ototopical drugs: animal models. Ann Otol Rhinol Laryngol Suppl 1990; 148:42–5. 10. Barlow DW, Duckert LG, Kreig CS, Gates GA. Ototoxicity of topical otomicrobial agents. Acta Otolaryngol 1995;115:231–5. 11. Leach JL, Wright CG, Edwards LB, Meyerhoff WL. Effect of topical fosfomycin on polymyxin B ototoxicity. Arch Otolaryngol Head Neck Surg 1990;116:49–53. 12. Gyde MC. A double-blind comparative study of trimethoprim-polymyxin B versus trimethoprimsulfacetamide-polymyxin B otic solutions in the treatment of otorrhea. J Laryngol Otol 1981; 95:251–9. 13. Rakover Y, Keywan K, Rosen G. Safety of topical ear drops containing ototoxic antibiotics. J Otolaryngol 1997;26:194–6. 14. Merrifield DO, Parker NJ, Nicholson NC. Therapeutic management of chronic suppurative otitis media with otic drops. Otolaryngol Head Neck Surg 1993;109:77–82. 15. Welling DB, Forrest LA, Goll F III. Safety of ototopical antibiotics. Laryngoscope 1995;105:472–4. 16. Dumas G, Bessard G, Gavend M, Charachon R. Risk of deafness following ototopical administration of aminoglycoside antibiotic. Therapie 1980; 35:357–63. 17. Lindner TE, Zwicky S, Brandle P. Ototoxicity of ear drops: a clinical perspective. Am J Otol 1995; 16:653–7. 18. Chambers HF. Antimicrobial agents: protein synthesis inhibitors and miscellaneous antibacterial agents. In: Goodman and Gilman’s the pharmacological basis of therapeutics. 10th ed. New York: Pergamon Press; 2001. p. 1246–50. 19. Proud GO, Mittelman H, Seiden GD. Ototoxicity of topically applied chloramphenicol. Arch Otolaryngol 1968;87:34–41. 20. Fairbanks DNF. Otic topical agents. Otolaryngol Head Neck Surg 1980;88:327–31. 21. American Medical Association. Drug evaluations annual. Chicago: American Medical Association; 1994. p 1574.

133

22. Patterson WC, Gulick WL. The effects of chloramphenicol upon the electrical activity of the ear. Ann Otol Rhinol Laryngol 1963;72:50–5. 23. Gulick WL, Patterson WC. The effects of chloramphenicol upon the electrical activity of the ear. Long term data. Ann Otol Rhinol Laryngol 1964;73:204–9. 24. Koide T, Hata A, Hando R. Vulnerability of the organ of Corti in poisoning. Acta Otolaryngol 1966;61:332. 25. Mittelman H. Ototoxicity of “ototopical” antibiotics: past, present, and future. Trans Am Acad Ophthalmol Otolaryngol 1972;76:1432–43. 26. D’Angelo EP, Patterson WC, Morrow RC. Chloramphenicol. Topical application in the middle ear. Arch Otolaryngol 1967;85:682–4. 27. Morizono T, Johnstone BM. Ototoxicity of chloramphenicol ear drops with propylene glycol as solvent. Med J Aust 1975;2:634–8. 28. Brown RD, Penny JE, Henley CM, et al. Ototoxic drugs and noise. Ciba Found Symp 1981; 85:151–71. 29. Henley CM, Brown RD, Penny JE, et al. Impairment in cochlear function produced by chloramphenicol and noise. Neuropharmacology 1984;23:197–202. 30. Gargye AK, Dutta DV. Nerve deafness following chloromycetin therapy. Indian J Pediatr 1959;26: 265. 31. Svenungsson B, Bengtsson E, Fluur E, Siegborn J. Hearing loss as a sequel to chloramphenicol and ampicillin treatment of Haemophilus influenzae meningitis. Scand J Infect Dis 1976;8:175–80. 32. Iqbal SM, Srivatsav CPB. Chloramphenicol ototoxicity. A case report. J Laryngol Otol 1984;98:523–5. 33. Joy JT, Scalettar R, Sodee DB. Optic and peripheral neuritis: probable effect of a prolonged chloramphenicol therapy. JAMA 1960;173:1731–4. 34. Mortimer AC. Bilateral optic atrophy following chloramphenicol therapy. JAMA 1953;151:1403. 35. Matsumuru H. A case report of anaphylaxis caused by application of 0.5% chloromycetin otic solution. Otologica Fukuoka 1964;10:48. 36. Supiyaphum P, Kerekhanjanarong V, Koranasophonepun J, Sastarasadhit V. Comparison of ofloxacin otic solution with oral amoxicillin plus chloramphenicol ear drops in the treatment of chronic suppurative otitis media with acute exacerbation. J Med Assoc Thai 2000;83:63–8. 37. Fraunfelder FT, Bagby GC Jr, Kelly DJ. Fatal aplastic anemia following topical administration of ophthalmic chloramphenicol. Am J Ophthalmol 1982;93:356–60. 38. Fisher AA. Erythema multiforme-like eruptions due to topical medications: Part II. Cutis 1986; 37:158, 160–1.

CHAPTER 15

Topical Antifungals Lawrence W. C. Tom, MD, Lisa M. Elden, MS, MD, and Roger R. Marsh, PhD

Andral and Gavarret in 1843 and Mayer in 1844 first described fungal infections of the external auditory canal. Virchow suggested the term “otomycosis” be used to describe this condition. Although otomycosis has been classically described as a fungal infection of the external auditory canal, it has been suggested that the term be expanded and redefined to include fungal infections of the middle ear and open mastoid cavities.1 Otomycosis occurs throughout the world, and its prevalence changes with location and climate. This condition is more common in tropical and subtropical environments and occurs more frequently when the weather is warm and humid. In temperate climates, it accounts for less than 10% of cases of external otitis.1,2 Many genera of fungi have been known to cause otomycosis. Aspergillus and Candida, well-known human pathogens, are the most common offending genera, with A. niger and C. albicans being the most common offending species. The exact prevalence of these pathogens varies by climatic region. Overall, Aspergillus spp are responsible for most cases of otomycosis.3,4 The difficulty in treating this annoying and stubborn condition has been well recognized. Multiple oral, topical, and intravenous agents and even radiation have been recommended to manage otomycosis.5–7 Many of these treatments have proven to be ineffective, unsafe, or both. For example, thimerosal (Merthiolate®) has been reported to be an effective agent but contains mercury. 8 The US Food and Drug Administration (FDA) has banned the use of mercury compounds as topical antiseptics because of their potential for systemic toxicity.9 The goals of management of otomycosis are to relieve symptoms, eliminate disease, and prevent recurrence. This is accomplished by identifying the infecting organisms, identifying and treating any predisposing factors, cleansing and drying the ear, and applying topical antifungal medications. Systemic antifungals may be necessary as a last resort in refractory cases.

Several well-recognized factors increase a patient’s susceptibility to fungal infections and their complications. Alterations in the immune state (primarily cellular immunity), systemic steroids, prolonged use of antibiotics, and diabetes can all increase the chance of developing fungal ear infections. Other factors that predispose the patient to otomycosis include the warmth, humidity, and configuration of the ear canal, the presence of open mastoid cavities, hearing aids with occlusive molds, trauma, and bacterial infections. The use of topical antibiotics or steroids, loss of cerumen, and water exposure have been previously implicated as predisposing factors, but it is not clearly established if they play a role in the development of otomycosis. Meticulous débridement is the mainstay of treatment of otomycosis. Cleansing removes the offending fungi, along with secretions and debris that provide nutrition for the organism, and allows topical medication to reach the affected area. Cleansing should be performed with the operating microscope. If irrigation is employed, microscopic examination should be performed after its completion. Débridement should be repeated frequently. The entire ear must be meticulously cleaned, with special attention directed to the pretympanic sulcus. The sulcus is the site of many early infections, and failure to clean this area may be responsible for recurrences.10 After the ear is cleaned, it should be dried with gentle mopping using a cotton-tip applicator or air. At the time of the initial cleaning, specimens may be sent for cultures in order to identify the fungal species. Although not necessary in all cases, the use of cultures will help direct therapy and is especially important for recurrent or resistant infections. Despite the need for a topical medication for otomycosis, there is no FDA-approved otic preparation that is specific for the treatment of otomycosis.11–14 Consequently, many agents with antimycotic properties have been used to treat otomycosis, but there are limited data available as to which agents are the most effective in treating this condition. Since manufacturers have not

Topical Antifungals

pursued approval for ototopical application, the ototoxic properties of these preparations have not been well studied. In many instances, patients with otomycosis also have communications between the external ear and the middle ear space, including those with pressure equalization (PE) tubes, tympanic membrane perforations, and open mastoid cavities. Wright and Meyerhoff have shown that antibiotic otic drops readily pass through a patent PE tube into the middle ear and onto the round window in chinchillas.15 Any topical antibacterial or antifungal medication can reach the cochlea by diffusion through the round window membrane (RWM). If the agent has ototoxic properties, temporary or permanent electrophysiologic changes within the inner ear or morphologic injury to the stria vascularis, hair cells, and supporting cells of the organ of Corti may occur. In general, studies designed to assess the ototoxicity of antifungal agents in humans have been flawed because they have been based on clinical impressions such as the lack of side effects and not on objective measurements of ototoxicity.1,14 Alternatively, electrophysiologic and histologic investigations of potential ototoxic effects of some topical antimycotic medications have been performed in animals.16–19 Although differences in the anatomy, methods of testing, and assessment make comparisons of ototoxicity between animals and humans difficult, the animal studies do have clinical relevance. If a preparation has been shown not to be ototoxic in animals, it is not likely to be ototoxic in humans. Knowledge of the presence or absence of potential ototoxicity of specific agents offers the clinician guidance regarding the choice of a particular topical antifungal medication when there is potential of exposure to the RWM. The choice as to which topical medication to use should be based on the susceptibility of the infecting species, the efficacy of the topical agent, and the potential ototoxic side effects. Antiseptics are not specific antifungal agents but achieve their therapeutic effects through secondary actions. Specific antifungal medications have been developed, but none is available as an otic preparation. Alcohol, acetic acid, boric acid, M-cresyl acetate, and gentian violet have been used to treat otomycosis for over 50 years.20 Clotrimazole, miconazole, econazole, nystatin, tolnaftate, potassium sorbate, and polysorbate are newer agents that have also been reported to be effective in the management of otomycosis.1,11,12,20–25

ANTISEPTICS Alcohol is a solvent and antiseptic that works by drying out the ear canal and may prevent the recurrence of otomycosis. Alcohol has been frequently combined with other substances to supplement their antimycotic actions. When it is applied to the middle ear, alcohol causes inflammation and pain and may be ototoxic.

135

Evaluating cochlear microphonics and endocochlear direct current potential, Morizono and Sikora have demonstrated that ethanol is ototoxic when applied to the middle ears of guinea pigs.26 Manning has suggested that most cases of otomycosis can be treated with topical therapy of mild acidic drops27 because fungal cells die when the pH is less than 4.28 Acidifying agents, such as acetic acid and boric acid, have long been used to treat this condition. Boric acid relieves pain, decreases inflammation, and destroys spores. It is available as a solution or powder, which also has a drying effect. In regions where specific antifungals are too expensive or unavailable, boric acid powder has been recommended as the treatment for otomycosis.29 It is not known whether boric acid causes ototoxicity. Acetic acid, however, has produced small auditory brainstem threshold shifts in guinea pigs, suggesting the potential for ototoxicity, and when acetic acid is combined with the solvent propylene glycol (VoSol®), the ototoxic effects have been shown to increase substantially.23 M-cresyl acetate solution (Cresylate®) is an antiseptic with antibacterial, antifungal, and anesthetic properties. In the ear canal it causes exfoliation of epithelium, reduction of pain, prevention of sporulation, and inhibition of secondary bacterial infection.5 In the past it has been widely used to manage otomycosis.13,28 Cresylate is often applied to the affected ear via a wick, which is then saturated with the solution. It is approved for use in the external ear but has some disadvantages. It has a bad odor, may burn, and is difficult to obtain. Cresylate is not recommended for use in the presence of a tympanic membrane perforation.30 Cresylate has been shown to be ototoxic in guinea pigs. In one study, auditory brainstem responses became significantly reduced 1 hour after it has been injected into the auditory bulla.17 Gentian violet, an aniline dye, is an antiseptic, which has anti-inflammatory, antibacterial, and antifungal properties, and has been used to treat otomycosis for over 60 years. Although it appears to be effective, it stains skin and clothing, which can result in noncompliance. Gentian violet has been shown to inhibit the growth of several species of Aspergillus, and in one study 81% of patients with otomycosis improved after treatment with gentian violet.1 Gentian violet, however, is ototoxic. Spandow and colleagues reported that when it is applied to the round window niche of rats, the latencies of auditory evoked potentials increase.16 In another study that was designed to evaluate histopathology of the cochlea following instillation of several antifungal agents, gentian violet was found to be significantly ototoxic if it enters the middle ear.19 This study was intended to dose 10 animals with gentian violet, but only 4 were dosed because the dye produced such dramatic, adverse changes. Three of the four

136

Topical Toxicity

guinea pigs developed torticollis, a behavioral response typical of severe vestibular injury. After the animals were sacrificed, three of the four had new bone growth that affected all surfaces of the middle ear; in two animals, the middle ear space was virtually obliterated by granular bone deposits. The cochlea was opened in one specimen, and the basal turn was filled with scar tissue. The intended histologic studies could not be performed because of the extreme new bone growth and cochlear fibrosis. Gentian violet should be used with great caution if there is a chance of its reaching the middle ear.

SOLVENTS Propylene glycol is a common ingredient of many otic preparations and serves as solvent carrier and penetrance enhancer. Glasgold and Boyd reported that a solution containing 99.9% propylene glycol and 0.1% dexamethasone effectively treated otomycosis and surmised that the hydrophilic and hydroscopic properties of the propylene glycol were responsible for the antifungal effects.31 Propylene glycol, however, has been shown to cause severe middle ear inflammation and cochlear toxicity in animals. Two studies by Morizono and colleagues demonstrated that there was a reduction in the cochlear microphonics of guinea pigs and chinchillas after weak concentrations of propylene glycol were instilled into the animals’ auditory middle ears.32,33 Further, in comparing chinchillas that had been treated with similar topical medications differing only in their concentrations of propylene glycol, Vassalli and colleagues found significant middle ear inflammatory changes, including fibrous adhesions and serous effusions and the development of cholesteatomas in ears treated with a 10.5% concentration of propylene glycol compared with those treated with a 2% concentration, which developed no lesions.34 They concluded that propylene glycol was the agent most likely responsible for these changes. To limit the risk of sensorineural hearing loss during ototopical therapy, Pickett and colleagues suggested that agents containing propylene glycol should be avoided unless there is clear clinical advantage.35 Preparations containing high concentrations of propylene glycol probably should not be used when there is potential access to the middle ear.

ANTIFUNGALS Although systemic and topical antifungal medications are readily available, the FDA has not approved these compounds for use in the ear. Despite this lack of approval, many antifungal medications are commonly used topically in the ear to treat otomycosis. Since the efficacy of these newer preparations has varied and has not been well studied, there is no consensus regarding which topical agent is most useful in eradicating this condition. In addition, several recommended drugs have limited availability or are no longer available.

The newer agents can be grouped into three general categories: polyenes, azoles, and miscellaneous.1,36 The polyenes are macrolides, which inhibit sterol synthesis in the cell membrane, leading to increased membrane permeability and death. They include amphotericin B, natamycin, and nystatin. The azoles are synthetic agents that reduce the concentration of ergosterol, an essential sterol in the normal fungal cytoplasmic membrane, and include clotrimazole, fluconazole, ketoconazole, bifonazole, and econazole. In the miscellaneous category are tolnaftate, polysorbate, and potassium sorbate, which have different mechanisms of action. Nystatin is the only commercially available topical polyene presently used to treat otomycosis. Amphotericin B is no longer available as a commercial, topical medication but is administered intravenously for serious fungal infections. Amphotericin B powder is used for the preparation of intravenous solutions. This medication can be obtained through a pharmacy either as powder or prepared into solution for instillation into the ear. It is not known whether amphotericin B is ototoxic. Natamycin is no longer available. Nystatin can be administered as a cream, ointment, or powder or can be made into a solution. Insufflation of nystatin powder into affected ears also helps to eliminate moisture. Oral nystatin may also supplement topical therapy.37 Nystatin is one of the most effective topical antifungal agents because it has a wide spectrum of activity. Although nystatin has not been reported to be ototoxic, it may be prudent to avoid its use when there may be access to the middle ear. In one study to determine cochlear hair cell toxicity, nystatin-treated animals were free of any hair cell damage, but a persistent residue of the medication or vehicle remained in the round window niche weeks after application.19 The significance of this finding is unknown but raises a concern for potential middle ear toxicity. Several azoles have been instilled into the ear to treat otomycosis. Bifonazole and econazole are no longer available. Miconazole is available as a cream that can be instilled into the ear. Stern and colleagues found that miconazole is effective in vitro, but it does not appear to be as potent as some other azoles.11 It does not appear to cause middle ear or cochlear injury.19 Clotrimazole is the most widely used topical azole, and it appears to be one of the most effective agents for the management of otomycosis. 1,11,12,14,23 It has an antibacterial effect, and this is an added advantage when treating mixed bacterial-fungal infections. Clotrimazole comes in a solution, lotion, and powder. Kwok and Hawke have reported that a single application of powder is superior to repeated applications of lotion.38 The powder coats the entire canal, delivers a higher concentration of medication, and decreases the humidity within the canal. In addition, the carrier vehicle for the liquid preparations of clotrimazole frequently

Topical Antifungals

irritates the underlying inflamed skin and exposed middle ear mucosa, causing pain. Pain is a common reason for noncompliance when clotrimazole liquid is applied in the presence of a tympanic membrane perforation or patent PE tube. Clotrimazole is relatively free of ototoxic effects. Marsh and Tom reported that after it was applied to the middle ear of guinea pigs there was a small shift in the auditory brainstem thresholds, suggesting a slight hearing loss.17 There has been no clinical evidence of clotrimazole ototoxicity, and histologic examinations have demonstrated that clotrimazole does not cause any cochlear hair cell loss.1,14,19 The systemic azoles, ketoconazole and fluconazole, have a limited but useful role in the management of otomycosis. Although they have a broad spectrum of activity, these agents are effective in treating Candida and not Aspergillus. Lucente has recommended that these drugs be used as adjuvants to topical therapy in refractory cases.28 Cohen and Thompson reported that they successfully managed 8 of 10 children with Candida infection of the middle ear or mastoid with oral ketaconazole.39 However, fluconazole has a more favorable pharmocologic profile and may be superior to ketaconazole in the management of Candida infections. Hepatotoxicity is a rare but potentially fatal side effect of both these medications, and they should be prescribed only after consultation with a physician who is familiar with their use.28 Tolnaftate is one of the most popular antifungal agents. It is available as a solution, cream, or powder and can easily be instilled into the ear. Tolnaftate acts by distorting hyphae and inhibiting the mycelial growth of susceptible fungi. It is a potent antifungal agent with a broad spectrum of activity. It has been recommended as an excellent choice for refractory cases of otomycosis and is well tolerated, with virtually no side effects. Its only disadvantage is that it may be ineffective against Candida sp.1 Tolnaftate is not ototoxic. Virtually no changes in auditory brainstem responses were found after instillation of tolnaftate solution into the middle ears of guinea pigs.17 When compared with untreated control animals, tolnaftate did not cause any middle ear injury, and there was no evidence of hair cell toxicity.19 Potassium sorbate and polysorbate are mold and yeast inhibitors. Fairbanks has reported that potassium sorbate is an effective topical agent for otomycosis.40 Although they are not available as primary active ingredients in any topical medication, potassium sorbate and polysorbate have been added to several well-known ototopical preparations such as Cortisporin TC Otic Suspension® and Cipro HC Otic Suspension® because of their antifungal properties. Studies do not exist that report on the ototoxic potential of either agent. However, polysorbate is presumed to be safe to use in the ear because Cipro HC Otic Suspension lacks ototoxicity.

137

APPLICATION There is no consensus regarding the best way to apply these topical agents. Application of ointment, insufflation of powder, and instillation of drops have all been recommended.2,28,38 Instillation of drops is the easiest method of application, but powder has advantages over drops because it can coat the surface, deliver a higher concentration of medication, and reduce humidity. The status of the ear should be considered when choosing the vehicle. Powder is preferred when the ear is wet and macerated and when there is an open mastoid cavity.4,21 Drops may be easier for the patient to apply but are more likely to enter the middle ear if there is a defect in the tympanic membrane. Drops may be beneficial if the infection involves the middle ear, but the medication may produce inflammatory changes, causing pain, discomfort, and local injury. Entrance into the middle ear also increases the chance that the drug may diffuse through the RWM into the inner ear causing cochlear ototoxicity. Insertion of gauze or a polyvinyl acetal wick into the ear canal or mastoid cavity and impregnating it with the medication is an excellent way to maintain continuous tissue contact and therapeutic levels while minimizing the potential for cochlear ototoxicity.

CONCLUSION Cleansing and drying of the ear and application of topical antifungal medications are important in the management of otomycosis. Although no FDA-approved otic preparation exists that is specific for the treatment of otomycosis, several antiseptics and antifungal agents are effective in managing this condition (Table 15-1). Table 15-1 Available Topical Agents Used to Treat Otomycosis Agent

Ototoxicity

Antiseptics

Alcohol

Yes

Acetic acid

Yes

Boric acid

Unknown

Gentian violet

Yes

M-cresyl acetate (Cresylate®)

Yes

Antifungals

Amphotericin B

Unknown

Clotrimazole

No

Miconazole

No

Nystatin

Probably not

Tolnaftate

No

138

Topical Toxicity

The decision as to which medication to use should be based on the susceptibility of the infecting species, efficacy of the agent, and, when a communication with the middle ear is present, the potential ototoxicity. Electrophysiologic and histologic studies in animals suggest that the antiseptics used to treat otomycosis are ototoxic, whereas the antifungals are nontoxic to the inner ear. Clotrimazole and tolnaftate are the most effective antifungal agents, and if the middle ear space communicates with the external ear, they are potentially the safest choices.

SUMMARY • Otomycosis commonly occurs in humid, tropical, and subtropical environments. Aspergillus and Candida spp appear to be most responsible. Risk factors when normal external and middle ear anatomy exists include defects in cellular immunity, systemic steroids, diabetes, and the prolonged use of topical antibiotic drops. The presence of an open mastoid cavity or the wearing of an occlusive hearing aid also appear to be predisposing factors. • Removal of infected debris, drying of the ear, and application of topical antifungals remain important in the management of otomycosis. Special attention should be focused on meticulous cleansing of the pretympanic sulcus, which is the site of many early infections and the area responsible for recurrences. • No FDA-approved otic solution for treatment of otomycosis exits at the present time. Antiseptics, solvents, and antifungals have all proven effective. Most animal studies, however, demonstrate significant inner ear toxicity from commonly used antiseptics should they reach the middle ear through a tympanic membrane defect, a point not to be overlooked in humans. Topical antifungals such as nystatin, tolnaftate, and those of the azole class (eg, clotrimazole or miconazole) appear to be safe without the risk for ototoxicity. • The decision to use a specific topical treatment for otomycosis should be based on the infective pathogen, its susceptibility, the efficacy of a known treatment, and whether a defect exists in the tympanic membrane.

REFERENCES 1. Paulose KO, Al Khalifa S, Shenoy P, Sharma RK. Mycotic infection of the ear (otomycosis): a prospective study. J Laryngol Otol 1989;103:30–5. 2. Mugliston T, O’Donoghue G. Otomycosis—a continuing problem. J Laryngol Otol 1985;99:327–33. 3. Hirsch BE. Disease of the external ear. Am J Otolaryngol 1992;13:145–55.

4. Selesnick SH. Otitis externa: management of a recalcitrant case. Am J Otol 1994;15:408–12. 5. Gill K. Otitis externa mycotica: comments concerning the prevalence, diagnosis and treatment of otomycosis. Arch Otolaryngol 1932;16:76–82. 6. Conley JJ. Evaluation of fungous disease of the external auditory canal. Arch Otolaryngol 1948; 47:721–45. 7. Ismail HK. Otomycosis. J Laryngol Otol 1962; 76:713–9. 8. Schneider ML. Merthiolate in treatment of otomycosis. Laryngoscope 1981;91:1194–5. 9. Status of certain additional over-the-counter Drug Category II and III active ingredients. Federal Register. April 22, 1998. 19799–802. 10. Beaney GPE, Broughton A. Tropical otomycosis. J Laryngol Otol 1967;81:987–97. 11. Stern JC, Shah MK, Lucente FE. In vitro effectiveness of 13 agents in otomycosis and review of the literature. Laryngoscope 1988;98:1173–7. 12. Lawrence TL, Ayers LW, Saunders WH. Drug therapy in otomycosis: an in vitro study. Laryngoscope 1978;88:1755–60. 13. Saunders WH. Otomycosis. In: Gates GA, editor. Current therapy in otolaryngology-head and neck surgery 1984–1985. Toronto: BC Decker; 1984. p. 1–2. 14. Bassiouny A, Kamel T, Moawad MK, et al. Broad spectrum antifungal agents in otomycosis. J Laryngol Otol 1986;100:867–73. 15. Wright CG, Meyerhoff WL. Ototoxicity of otic drops applied to the middle ear in the chinchilla. Am J Otolaryngol 1984;5:166–76. 16. Spandow O, Anniko M, Møller AR. The round window as access route for agents injurious to the inner ear. Am J Otolaryngol 1988;9:327–35. 17. Marsh RR, Tom LWC. Ototoxicity of antimycotics. Otolaryngol Head Neck Surg 1989;100:134–6. 18. Parker FL, James GWL. The effect of various topical antibiotic and antibacterial agents on the middle and inner ear of the guinea-pig. J Pharm Pharmacol 1978;30:236–9. 19. Tom LWC. The ototoxicity of common topical antimycotic preparations. Laryngoscope 2000; 110:509–16. 20. McBurney R, Searcy HB. Otomycosis: investigation of effective fungicidal agents in treatment. Ann Otol Rhinol Laryngol 1936;45:988–1008. 21. Smyth GDL. Fungal infections in the post-operative mastoid cavity. J Laryngol Otol 1962;76:797–821. 22. Damato PJ. Treatment of otomycosis due to Aspergillus niger with tolnaftate. St. Luke Hospital Gazette 1966;1:66–8. 23. Maher A, Bassiouny A, Moawad MK, et al. Otomycosis: an experimental evaluation of six antimyotic agents. J Laryngol Otol 1982;96:205–13.

Topical Antifungals

24. Liston SL, Siegel LG. Tinactin in the treatment of fungal otitis externa. Laryngoscope 1986;96:699. 25. Rohn GN, Meyerhoff WL, Wright CG. Ototoxicity of topical agents. Otolaryngol Clin North Am 1993;26:747–58. 26. Morizono T, Sikora MA. Ototoxicity of ethnol in the tympanic cleft in animals. Acta Otolaryngol 1981;92:33–40. 27. Manning SC. Mycoses. In: Paparella MM, Schumrick DA, Gluckman JL, et al, editors. Otolaryngology. 3rd ed. Philadelphia: WB Saunders; 1991. p. 595. 28. Lucente FE. Fungal infections of the external ear. Otolaryngol Clin North Am 1993;26:995–1006. 29. Than KM, Naing KS, Min M. Otomycosis in Burma and its treatment. Am J Trop Med Hyg 1980;29:620–3. 30. Fairbanks DNF. Pocket guide to antimicrobial therapy in head and neck surgery. Alexandria (VA): The American Academy of Otolaryngology-Head & Neck Surgery Foundation; 1991. p. 36. 31. Glasgold AI, Boyd JE. Otitis externa: control of bacterial and mycological infections—a preliminary report. Eye Ear Nose Throat Mon 1973;52:94–6. 32. Morizono T, Johnstone BM. Ototoxicity of chloramphenicol ear drops with propylene glycol as a solvent. Med J Aust 1975;2:634–8.

139

33. Morizono T, Paparella MM, Juhn SK. Ototoxicity of propylene glycol in experimental animals. Am J Otolaryngol 1980;1:393–9. 34. Vassalli L, Harris DM, Gradini R, et al. Inflammatory effects of topical antibiotic suspensions containing propylene glycol in chinchilla middle ears. Am J Otolaryngol 1988;8:1–5. 35. Pickett BP, Shinn JB, Smith MFW. Ear drop ototoxicity: reality or myth? Am J Otol 1997; 18:782–91. 36. Neibart E, Gumprecht J. Antifungal agents and the treatment of fungal infections of the head and neck. Otolaryngol Clin North Am 1993;26:1123–31. 37. Jahn AF, Hawke M. Infections of the external ear. In: Cummings CW, Frederickson JM, Harker LA, et al, editors. Otolaryngology–head and neck surgery. 2nd ed. St. Louis (MO): Mosby Year Book; 1993. p. 2787–94. 38. Kwok P, Hawke M. Clotrimazole power in the treatment of otomycosis. J Otolaryngol 1987; 16:398. 39. Cohen CR, Thompson JW. Otitic candidiasis in children: an evaluation of the problem and effectiveness of ketoconazole in 10 patients. Ann Otol Rhinol Laryngol 1990;99:427–31. 40. Fairbanks DNF. Otic topical agents. Otolaryngol Head Neck Surg 1980;88:327–31.

CHAPTER 16

Surgical Disinfectants and Antiseptics Andrew R. Scott, BM, BS, MPhil, FRCS(ORL-HNS), Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL), and John A. Rutka, MD, FRCSC

Potential ototoxicity of topical skin preparations was first highlighted by Bicknell in 1971.1 Fourteen patients from a series of 97 who underwent myringoplasty experienced a severe sensorineural hearing loss within the first 6 months of the operation, 13 having dead ears. The only common factor appeared to be the use of chlorhexidine (0.5% in 70% spirit) for preoperative sterilization. All cases had dry perforations, and the presumed mechanism of ototoxicity was that the sterilizing agent had entered the inner ear via the round window membrane (RWM).

CHLORHEXIDINE Chlorhexidine is an ingredient in various preparations available for preoperative hand and general skin disinfection in concentrations varying between 0.015% and 5%, depending on the preparation and its intended use; the higher concentrations are generally for hand washing of the surgical team.2 It is available in aqueous solution, in 70% alcohol, or combined with cetrimide and has a broad spectrum of antimicrobial activity.2 The chlorhexidine molecule is positively charged and acts by disrupting the cell membrane and precipitating the cytoplasm of the susceptible microbe.3 Since the publication of Bicknell’s original clinical paper in 1971, several animal studies have been performed to examine the potential ototoxicity of chlorhexidine. Aursnes examined morphological changes in the organ of Corti and also, in a separate study, on the vestibular neuroepithelium of guinea pigs exposed to chlorhexidine via topical application to the middle ear. Damage was observed in the sensory epithelia of both sites and was related to the concentration of chlorhexidine, the duration of exposure, and the time lapse after exposure.4,5 Similar studies were carried out by Igarashi and colleagues on cats.6,7 Two different concentrations of chlorhexidine (2% and 0.05%) were applied, and saline was used as a control substance in the contralateral ears. In the cochlear study, nine animals were sacrificed and examined immediately after completion of

application of the test solution whereas another three were looked at 4 weeks later. Again, damage to the sensory epithelium was seen and was related to the concentration of chlorhexidine applied and the time lapse after exposure (ie, the injury had progressed in the period following exposure). Specifically, hair cell degeneration with loss of hair bundles was observed on electron microscopy. Further, the damage was more apparent in the lower cochlear turns, as would be expected if the portal of entry was via the RWM. The vestibular damage included degeneration of afferent nerve endings with dilatation of the nerve terminals and degenerative changes of the mitochondria within the nerve terminals. These changes were evident on both type 1 and type 2 vestibular hair cells. In a further study to examine the ototoxicity of antiseptics in an animal model, an electrophysiologic approach was employed.8 Perez and colleagues used auditory brainstem responses (ABRs) to measure auditory function and the relatively new technique of shortlatency vestibular evoked potentials (VEPs) to measure vestibular function. VEPs can be measured in response to angular and linear acceleration. The first peak of the VEP represents the compound action potential (CAP) of the vestibular nerve and is therefore dependent on (and hence a measure of) end-organ function.8 Rats were used as the animal model, and each initially underwent a right-sided surgical labyrinthectomy. Subsequent testing could then be performed on animals with only unilateral cochleovestibular function, simplifying interpretation of the VEPs and avoiding masking issues with ABRs. In addition to the test solutions, saline was used as a negative control and gentamicin (known to be ototoxic) as a positive control to check experimental conditions. The rats were randomly assigned to receive topical application of one of the test solutions or controls. The chlorhexidine group received a dose of 0.5% chlorhexidine in aqueous solution. In all five animals that received chlorhexidine the VEPs and ABRs were completely abolished, similar to the findings

Surgical Disinfectants and Antiseptics

with gentamicin and at postmortem examination. All in the saline group preserved normal function. This model, therefore, clearly demonstrates cochlear and vestibular loss of function following chlorhexidine application. Nonetheless, direct extrapolation to humans is always difficult owing to possible species effects and anatomical differences, particularly in the region of the RWM.

ALCOHOLS Alcohols are used alone or in combination with other disinfectant agents for skin preparation. They are also used as excipients for a wide variety of agents. One of the problems in assessing ototoxicity of many preparations is identifying which of the components is potentially ototoxic. Further, some evidence shows that topical alcohol on its own may be toxic to the inner ear. When Perez and colleagues evaluated VEPs and ABRs following topical antiseptics (described above), 70% ethyl alcohol was also evaluated. Complete loss of both ABR and VEP was observed in two of five animals, and a loss or elevated ABR threshold was observed in the remaining three (VEPs remaining normal). This clearly raises concern for all alcohol-based preparations.

IODINE Iodine-containing skin preparations are numerous and may be aqueous or alcohol based, with or without a detergent. In 1982 Morizono and Sikora examined the ototoxicity of two commonly used povidone-iodine preparations applied topically to chinchilla middle ears.9 They compared povidone-iodine scrub, which contains a detergent (0.75% available iodine), and povidone-iodine solution, which does not (1% available iodine). Ringer’s solution was used as a diluent and control. The main outcome measures were CAPs recorded from the round window in response to tone bursts (a method of assessing cochlear dysfunction); these were recorded during the first 2 hours in one group of animals and at 24 hours in another group. A toxic effect on cochlear function was reported with the povidone-iodine scrub even at a 1:100 dilution. The povidone-iodine solution produced measurable toxic effects at concentrations greater than 1:4. Further, the cochlea appeared to be affected in a gradient manner, with the higher frequencies of the basal turn being most susceptible. This fits with the presumed route of toxicity via the RWM. The authors postulated that the presence of detergent in the scrub may allow the solution to diffuse through the RWM more easily, thus accounting for the increased effects. However, the possibility that the detergent is directly toxic must be considered.10 Again, extrapolating these results to humans is difficult. The chinchilla RWM is six times thinner than that of humans, and hence the risk to humans may be overstated.9 In another animal study by Aursnes in

141

1982, topical iodine toxicity was assessed morphologically in guinea pigs. Evidence of damage to the organ of Corti and also to vestibular receptors was observed with alcohol-based iodine preparations (70% alcohol) but not with aqueous preparations.11 The potential direct toxicity of alcohol must also be considered and indeed may have accounted for all of the damage seen. Povidone-iodine was also evaluated in the previously described study by Perez using VEPs and ABRs as outcome measures.8 In all five animals exposed to povidone-iodine solution (10%), there was no apparent effect on VEPs, with only a minimal effect, if any, on ABRs. The effect on ABR was a prolongation of the latency of the first wave, without change in amplitude. The authors discussed that this may be accounted for by a minimal conductive hearing loss secondary to thickening of the middle ear mucosa observed in this test group. Overall, the evidence for topical ototoxicity of iodine in skin preparation solutions appears to be low but conflicting at this time.

QUATERNARY AMMONIUM COMPOUNDS Quaternary ammonium compounds have been used as biocides since the 1930s.12 Benzalkonium chloride, benzethonium chloride, and cetrimide are members of this group of compounds used for skin disinfection. Cetrimide is probably best known in combination with chlorhexidine as Savlon. Benzalkonium chloride and benzethonium chloride are also used as preservatives in some drug preparations. In particular, benzalkonium chloride is a component of some ototopical preparations including Floxin Otic, Betnesol, Predsol, Gentisone HC, Genticin, Garamysin, Predsol-N, Betnesol-N, and Vista-Methasone.2,3 The potential ototoxicity of benzalkonium chloride and benzethonium chloride was tested in an animal model by Aursnes in 1982.13 Each compound was administered topically to guinea pig middle ears in a concentration of 0.1% in both aqueous and alcohol bases (70%). When examined morphologically, using phase contrast microscopy, damage to the neuroepithelium of both cochlear and vestibular parts of the inner ear was observed. The extent of damage was related to the duration of exposure to the test compound as well as the time from exposure to sacrifice. Possible ototoxicity to Savlon has been reported in the veterinary literature. Vestibular dysfunction was observed in 12 dogs and 3 cats following topical ear treatment by the veterinary practitioner.14 In eight of the cases it was confirmed that the ear canal had been rinsed with Savlon in the presence of a tympanic membrane perforation. Of course, ototoxicity may have been a result of the chlorhexidine component. However, the authors conducted their own study on the cetrimide (15%) component alone.14 This was instilled into unilateral middle ears of two guinea pigs. Signs of

142

Topical Toxicity

vestibular dysfunction on the treated side were observed after 2.5 hours. These included rotation of the head with the treated ear down, deviation of the eyes toward the treated side, nystagmus with a fast component away from the treated side, and a tendency to circle toward the treated side. The signs resolved within 24 hours. Cetrimide ototoxicity is suggested.

PROPYLENE GLYCOL Propylene glycol is one of the most common solvents used as a vehicle for drug delivery. It is used in some topical ear preparations, including chloramphenicol ear drops, Molcer, Audax, Audicort, and Tri-Adcortyl.2 Morizono and colleagues addressed the question of ototoxicity of propylene glycol in an animal model.15 They looked at the functional effects as well as the morphologic effects of propylene glycol in varying concentrations on the guinea pig and chinchilla ear. Cochlear microphonics were measured in the guinea pigs, and endocochlear potentials were measured in guinea pigs and chinchillas. Ringer’s solution was used as the diluent and as a control. They found that a 50% solution or stronger always caused a reduction in the cochlear microphonics. The effects at lower concentrations varied among the guinea pigs, but one animal had recordable changes with only a 10% solution, although it was applied for 6 days. Dose-related changes in the endocochlear potentials were also noted. Histological changes in the inner ear included destruction and ossification of the scala tympani and scala vestibuli and distortion of the organ of Corti. Inflammation of the middle ear mucosa was also reported. In a previous similar study, Morizono reported an irreversible reduction in cochlear microphonics with a 10% solution of propylene glycol after only 24 hours.16 Other morphologic studies have been conflicting. Stupp and colleagues and Parker and James reported inner ear hair cell loss following topical application of propylene glycol to the middle ear.17,18 Vernon and colleagues, however, found only middle ear changes and no hair cell loss above that found in the control animals.19 It is obviously possible to have loss of function despite normal morphology; the physiologic outcomes may be more sensitive.

ACETIC ACID Acetic acid is a weak acid that is often included in otic preparations for its antiseptic properties; for example it is a component of aluminum acetate, Otomize, and VoSol.2,20 In 1989, Ikeda and Morizono investigated the ototoxicity of acetic acid in an animal model.20 The effects of VoSol (2% acetic acid, 3% propylene glycol diacetate, 0.02% benzethonium chloride, and 0.015% sodium acetate dissolved in propylene glycol), 2% acetic acid (dissolved in perilymph and adjusted to pH 4 with NaOH), and a strong nonorganic acid (perilymph

adjusted to pH 4 with HCl) were compared. Changes in endocochlear potentials and endolymph pH were recorded. VoSol and the acetic acid solution caused a depression of endocochlear potential and marked acidification of endolymph whereas the HCl solution did not. Further, the effect of VoSol was significantly more than that of acetic acid alone, suggesting a direct or synergistic action of the propylene glycol component. The authors postulated that the effects of acetic acid were a result of its ability to cross the RWM and interfere with enzymatic processes in the stria vascularis.

SKIN PREPARATION: RECOMMENDATIONS The choice of ototopical agent in treating infection can only be made based on the perceived benefit balanced against the known risk for the specific patient circumstances encountered. But what of prophylactic use of preparations to attempt to sterilize the ear prior to surgery? Most surgeons would accept that it is desirable to reduce the bacterial load around the operative site as much as is safely possible, and it seems sensible to continue to apply skin preparation solutions to the pinna and skin surrounding the ear. However, it is not necessary to instill or apply these solutions to the ear canal if there is a risk of them entering the middle ear either via a perforation or during the procedure, particularly if they are potentially ototoxic. To this end, it is standard practice and teaching to occlude the ear canal with a temporary barrier (ie, cotton wool) when the skin preparation is applied to prevent it entering the ear canal. Anecdotally, such practice does not appear to increase the number of postoperative wound infections. However, some surgeons feel it necessary to attempt to disinfect the ear canal prior to surgery, and usually an aqueous iodine solution is recommended.21–23 Certainly on the available evidence (mostly animal), aqueous iodine preparations have the least potential for ototoxicity. Another strategy is to flush the ear canal first with the disinfectant and then with saline to remove it.24 This seems reasonable when the tympanic membrane is known prior to surgery to be intact. The available evidence indicates that all of the major types of skin preparation solutions have the potential to be ototoxic. Chlorhexidine should be avoided, and aqueous solutions are preferable to alcohol-based ones.

SUMMARY • Animal models have demonstrated that nearly all commonly used topical disinfectants, antiseptics, and excipients appear to be ototoxic should they enter into the middle ear. Based on extrapolation from these experiments and on the few series reports in humans, all chlorhexidene and alcohol-based preparations should be avoided during ear surgery when a tympanic membrane

Surgical Disinfectants and Antiseptics

perforation is present unless a temporary barrier is applied to prevent middle ear exposure. • The resultant ototoxicity appears directly related to the concentration, duration of exposure, and the time lapse from exposure of the agent used. • Of all the solutions available at the time of writing, aqueous iodine preparations are the safest of all topically applied agents in humans.

REFERENCES 1. Bicknell PG. Sensorineural deafness following myringoplasty operations. J Laryngol Otol 1971; 85:957–61. 2. British Medical Association, Royal Pharmaceutical Society of Great Britain. British National Formulary. BNF45 (March 2003). London: Pharmaceutical Press; 2003. 3. Mosby’s drug consult. Elsevier; 2003. Available at: http://www3.us.elsevierhealth.com/DrugConsult/ (accessed August 2003). 4. Aursnes J. Vestibular damage from chlorhexidine in guinea pigs. Acta Otolaryngol 1981;92:89–100. 5. Aursnes J. Cochlear damage from chlorhexidine in guinea pigs. Acta Otolaryngol 1981;92:259–71. 6. Igarashi Y, Suzuki J. Cochlear ototoxicty of chlorhexidine gluconate in cats. Arch Otorhinolaryngol 1985;242:167–76. 7. Igarashi Y, Oka Y. Vestibular ototoxicity following intratympanic applications of chlorhexidine gluconate in the cat. Arch Otorhinolaryngol 1988; 245:210–7. 8. Perez R, Freeman S, Sohmer H, Sichel JY. Vestibular and cochlear ototoxicity of topical antiseptics assessed by evoked potentials. Laryngoscope 2000;110:1522–7. 9. Morizono T, Sikora MA. The ototoxicity of topically applied povidone-iodine preparations. Arch Otolaryngol 1982;108:210–3. 10. Austin DF. Ototoxicity of detergents. Arch Otolaryngol 1982;108:808. 11. Aursnes J. Ototoxic effect of iodine disinfectants. Acta Otolaryngol 1982;93:219–26. 12. Russell AD. Introduction of biocides into clinical practice and the impact on antibiotic-resistant bacteria. J Appl Microbiol 2002;92 Suppl:121S–35S.

143

13. Aursnes J. Ototoxic effect of quaternary ammonium compounds. Acta Otolaryngol 1982;93:421–33. 14. Galle HG, Venker-van Haagen AJ. Ototoxicity of the antiseptic combination chlorhexidine/cetrimide (Savlon): effects on equilibrium and hearing. Vet Q 1986;8:56–60. 15. Morizono T, Paparella MM, Juhn SK. Ototoxicity of propylene glycol in experimental animals. Am J Otolaryngol 1980;1:393–9. 16. Morizono T, Johnstone BM. Ototoxicity of chloramphenicol ear drops with propylene glycol as solvent. Med J Aust 1975;2:634–8. 17. Stupp H, Kupper K, Lagler F, et al. Inner ear concentrations and ototoxicity of different antibiotics in local and systemic application. Audiology 1973;12:350–63. 18. Parker FL, James GW. The effect of various topical antibiotic and antibacterial agents on the middle and inner ear of the guinea-pig. J Pharm Pharmacol 1978;30:236–9. 19. Vernon J, Brummett R, Walsh T. The ototoxic potential of propylene glycol in guinea pigs. Arch Otolaryngol 1978;104:726–9. 20. Ikeda K, Morizono T. The preparation of acetic acid for use in otic drops and its effect on endocochlear potential and pH in inner ear fluid. Am J Otolaryngol 1989;10:382–5. 21. Jackson CG. Principles of temporal bone and skull base surgery. In: Glasscock ME, Gulya AJ, editors. Surgery of the ear. 5th ed. Hamilton (ON): BC Decker Inc; 2003. p. 271. 22. Shea MC Jr. Tympanoplasty: the undersurface graft technique–transcanal approach. In: Brackmann DE, Shelton C, Arriaga MA, editors. Otologic surgery. 2nd ed. Philadelphia: WB Saunders Co; 2001. p. 106. 23. Jackson CG, Glasscock ME, Strasnick B. Tympanoplasty: the undersurface graft technique– postauricular approach. In: Brackmann DE, Shelton C, Arriaga MA, editors. Otologic surgery. 2nd ed. Philadelphia: WB Saunders Co; 2001. p. 116. 24. Mandolis S. Closure of tympanic membrane perforations. In: Glasscock ME, Gulya AJ, editors. Surgery of the ear. 5th ed. Hamilton (ON): BC Decker Inc; 2003. p. 407.

Interventions CHAPTER 17

Genetic Factors in Aminoglycoside Ototoxicity Nathan Fischel-Ghodsian, MD

Aminoglycosides, which include gentamicin, streptomycin, and tobramycin, are a group of clinically important antibiotics that in Western countries are mainly used in hospitalized patients with aerobic gramnegative bacteria or in patients with tuberculosis.1,2 In East Asia, aminoglycosides are often used as first-line outpatient therapy for relatively minor infections, such as otitis media and bronchitis. They are highly polar cations, are generally not thought to be metabolized, and are excreted almost entirely by glomerular filtration. Their main toxicities involve the auditory, vestibular, and renal systems. 1,2 The renal impairment is usually reversible, whereas the vestibular and auditory ototoxicity is frequently irreversible. Although all of the aminoglycosides are capable of affecting both cochlear and vestibular function, some (streptomycin and gentamicin) produce predominantly vestibular damage, others (amikacin, neomycin, kanamycin) predominantly cochlear damage, whereas tobramycin, for example, affects both equally.1 In the United States, approximately 4 million courses of aminoglycosides are administered annually.3 It is estimated that at least 2 to 5%, and in some studies up to 25%, of patients treated with these antibiotics develop clinically significant hearing loss.4–6

OVERVIEW OF THE PATHOLOGIC AND SUSPECTED MECHANISMS FOR AMINOGLYCOSIDE CELLULAR TOXICITY Structural and Functional Changes in Aminoglycoside Ototoxicity Animal and human temporal bone studies implicate the cochlear neuroepithelium as the primary site of injury associated with aminoglycoside ototoxicity,7,8 although occasionally neural destruction without cochlear hair cell damage has been described.9 In general, the damage is first seen in the outer hair cells of the basal turn, and in more severe cases degenerative changes extend into the upper cochlear turns and inner

hair cells.10 At the same time varying degrees of atrophy of the stria vascularis can be observed.10 Some investigators describe the appearance of swollen mitochondria in the supranuclear portion of the cell as one of the earliest ultrastuctural changes within the hair cells.11 Others have observed an increase in lysosomes, proliferation of the endoplasmic reticulum, and formation of Hensen’s bodies as the first ultrastructurally identifiable changes.12 Later changes include the appearance of multivesicular bodies, clumping of nuclear chromatin, formation of giant, fused stereocilia, and, finally, extrusion of cell content.11 Associated with the morphologic changes are functional changes, the earliest of which is the decrease in cochlear microphonics.13 This is an apparently reversible phenomenon owing to blockage of ion-selective transduction channels in the hair cells.14 Later elevations of auditory nerve thresholds are observed, with the eventual occurrence of permanent functional impairment.5,15 Mechanisms for Aminoglycoside-Induced Cellular Toxicity Aminoglycoside toxicity is thought to affect mainly kidney and ear because the drugs are concentrated in renal tubular cells and in the perilymph and endolymph of the inner ear.16,17 The precise mechanism of hair cell toxicity in the ear is unclear. The first step is the best understood: the positively charged aminoglycosides are attracted to the negatively charged glycocalyx present on the apical surface of the hair cells; the drugs are then bound to the stereocilia, which, in competition with Ca2+, leads to a reversible interference with transduction channels.14,18,19 The second step is thought to be the entrance of the drugs into the cell, possibly from the basal side, and interference with the biochemical machinery of the cell. The precise mechanism of this step, however, is speculative; that is, the binding of the drug to phospholipids may or may not cause membrane damage, or interference with the protein-producing or processing machinery of the cell, or disturbed nucleic

Genetic Factors in Aminoglycoside Ototoxicity

acid synthesis.18,20,21 More recently, an activation of gentamicin via the formation of an iron-gentamicin complex followed by free radical formation within the cell as the damage-producing intermediary has been described (see Chapter 9, “Mechanisms for Aminoglycoside Ototoxicity: Basic Science Research”).22 The wide variety of structural and functional abnormalities described with aminoglycoside toxicity underscores the need to identify those factors that are primary events and those that are secondary responses to injury. Identification of the genetic basis of many inherited and acquired diseases could serve as a model for identifying primary events. For example, sickle cell disease with its various clinical signs of anemia, predisposition to severe infections, severe pain, strokes, and other various abnormalities, could be reduced to a single point mutation in the β-globin gene.23 Similarly, the complex symptoms and pathologic and biochemical changes of some cancers could be reduced to single genetic events, with Philadelphia chromosome–positive chronic myelocytic leukemia and acute promyelocytic leukemia as other examples.24,25 The identification of primary events is an important step to develop preventive and therapeutic modalities.26

GENETIC SUSCEPTIBILITY FOR AMINOGLYCOSIDE OTOTOXICITY Evidence for Genetic Factors prior to Molecular Testing Although anecdotal reports of patients who lost hearing after a single dose of aminoglycosides are not uncommon, formal evidence for individual susceptibility to aminoglycoside ototoxicity is difficult to document in the literature. All retrospective studies lack detailed information on some of the parameters related to drug toxicity, such as drug dose, renal function at the time of drug administration, peak and trough blood levels of the drug during administration, kind and doses of concomitant medications with possible ototoxicity, and details of the underlying disease treated with the antibiotic.1,27 One prospective study implicated length of treatment, bacteremia, fever level, and liver function with ototoxicity while, surprisingly, finding no correlation with plasma levels of the drug.6 The question of individual variability in susceptibility was not addressed. Nevertheless, the existence of families with multiple individuals with ototoxic deafness induced by aminoglycoside exposure was noticed early on. The first families with more than two members with streptomycin-induced hearing loss were initially described in the Japanese literature in 1957.28 Prazic and colleagues described in 1964 a family with four sisters who developed hearing loss after streptomycin injections.29 In 1971 Tsuiki and Murai described 16 families in

145

which two or more members had aminoglycoside ototoxicity.30 Konigsmark and Gorlin in 1976 summarized most existing descriptions of familial aminoglycoside ototoxicity and concluded that inheritance of the predisposition was probably autosomal dominant with incomplete penetrance. However, they also noted that no male-to-male transmission had been seen and concluded that the inheritance could be multifactorial.31 In 1989 Higashi reviewed the literature and concluded that the most likely explanation for the maternal inheritance observed was on the basis of a mitochondrial deoxyribonucleic acid (DNA) defect.28 In 1991 Hu and colleagues in China described another 36 families with maternally transmitted predisposition to aminoglycoside ototoxicity and concluded also that a mitochondrial defect might be responsible for this predisposition.32 Additional evidence for a genetic basis for aminoglycoside susceptibility comes from animal studies. Macaque monkeys are apparently immune to the ototoxic effects of dihydrostreptomycin, whereas patas monkeys appear to be highly sensitive to that drug.33 The mechanism of this differential susceptibility has not been established but very likely is because of genetic differences in these two species. The model emerging from these data seems to indicate that at very high total cumulative dose and drug levels most individuals will exhibit toxicity, whereas at lower cumulative drug levels, aminoglycoside ototoxicity results from genetic susceptibility in at least some cases. The genetic susceptibility factors can be at multiple levels, including drug uptake, drug interactions within the cell, or at the level of tissue response to the injury. Identification of the First AminoglycosideSusceptibility Mutation: The A1555G Mutation in the 12S Ribosomal Ribonucleic Acid Gene The pattern of maternal transmission of susceptibility to aminoglycosides was the first hint that mutations in the mitochondrial chromosomes could be the molecular basis for this susceptibility. Since the bactericidal effect of aminoglycosides occurs through altering the fidelity of translation of the genetic code34 and since mitochondria are evolutionarily derived from bacteria, in 1993 we proposed that susceptible individuals had altered protein synthesis in the mitochondria when exposed to aminoglycosides.35 We also proposed that the mitochondrial 12S ribosomal ribonucleic acid (rRNA) gene was the most likely gene to harbor such mutations since the small rRNA had been shown to bind to aminoglycosides and to harbor resistance mutations in bacteria and in the mitochondria of yeast and Tetrahymena.36–38 In 1993 the same group analyzed the mitochondrial genome of three Chinese families with 15 members with

146

Interventions

aminoglycoside ototoxicity and a pattern of inheritance compatible with maternal inheritance. We identified a A1555G mutation in the 12S rRNA gene in all three families and did not find this sequence change in 278 control individuals.35 Interestingly, this mutation lay exactly in the region of the gene for which the resistance mutations in yeast and Tetrahymena had been described and in which aminoglycoside binding had been previously documented in bacteria.36–38 In addition, the mutation made the mitochondrial RNA gene in this region more similar to the bacterial rRNA gene.35 Subsequently, the same mutation was found in families with aminoglycoside ototoxicity from all ethnic backgrounds and geographic origins.39–44 Aminoglycoside ototoxicity owing to the A1555G mutation has also been described in a relatively small number of individuals of different ethnic backgrounds with no family history.39,45,46 Although the mutation is nearly always found in homoplasmy, two reports from Spain describe several families with the A1555G mutation in a heteroplasmic state.42,47 It needs to be stressed that the A1555G mutation appears to be a deafness-predisposing mutation in a broader sense. Although aminoglycosides appear to be a major trigger for the phenotypic expression of this mutation, families with this mutation have been described with hearing loss without exposure to aminoglycosides,35,43 as well as families with hearingimpaired members of whom only some were exposed to aminoglycosides.40,42,43,48 In this regard we postulated that instead of environmental factors such as aminoglycosides, allelic variants of nuclear genes can also interact with the A1555G mutation in such a way as to precipitate hearing loss.49,50 A major locus for such a modifier gene has been localized to chromosome 8,51,52 and two additional modifier genes have recently been implicated through a candidate gene approach.53,54 Other Mitochondrial Predisposing Mutations Since the A1555G mutation in the mitochondrial 12S rRNA gene accounts for only a minority of aminoglycoside ototoxicity, other susceptibility mutations possibly can be found in the same gene. DNA from 35 Chinese patients with sporadic aminoglycoside ototoxicity and without the A1555G mutation was analyzed for sequence variations in the 12S rRNA gene. Three sequence changes were found; only one of them, an absence of a thymidine at position 961 with varying numbers of cytosines inserted (∆961Cn), appeared likely to be a pathogenic mutation.55 Analysis of 34 similar patients in the United States of varying ethnic backgrounds did not reveal this mutation, but an Italian family with 5 maternally related members who all became deaf after aminoglycoside treatment were found to have the ∆961Cn mutation.56 This sequence change was not found in 799 control individuals.55 A

recent report based on a newborn screening program in Texas has found a significantly higher rate of the ∆961Cn mutation, with seven positive results of 1,173 unselected samples.57 A different study on 721 samples from a repository of deaf probands in the United States revealed a frequency of 1.2% for the ∆961Cn mutation.58 It is not clear how many of these probands, if any, were exposed to aminoglycosides. At this stage the evidence for the ∆961Cn mutation as a predisposing mutation for aminoglycoside ototoxicity is equivocal, but prudence would suggest that individuals with this mutation should avoid these antibiotics, in particular if there is a family history of hearing loss after aminoglycoside administration. A third mitochondrial mutation predisposing to aminoglycoside ototoxicity was recently identified in a large Chinese pedigree with maternally inherited hearing loss. 59 Family members in the maternal line demonstrated hearing ranging from severe hearing loss to normal hearing and when exposed to aminoglycosides had a subsequent severe to profound hearing loss. Mutational analysis of the mitochondrial 12S rRNA gene did not reveal the A1555G or ∆961Cn mutations. However, a homoplasmic C to T transition mutation was identified at position 1494 in the 12S rRNA gene and was not seen in 300 ethnically matched individuals. Interestingly, and convincingly, this mutation leads to a nucleotide pairing 1494T-A1555 in the penultimate loop of the 12S rRNA in the aminoglycoside binding region, very similar to the base pairing generated by the A1555G mutation, 1494C-G1555. In both cases the mitochondrial gene became more similar to the bacterial gene and thus presumably more susceptible to aminoglycoside toxicity. Nuclear Inherited Predisposing Genes Mutations in the mitochondrial 12S ribosomal rRNA gene are found in only a minority of patients with aminoglycoside ototoxicity, thus leaving open the possibility that nuclear-encoded genes involved in the uptake, intracellular trafficking and binding, and disposal of aminoglycosides can also be predisposing genes to ototoxicity. It is also possible that the regeneration capability of the sensory epithelium is dependent on the genetic makeup of the individual and that some of the distinctions between reversible and irreversible ototoxicity are dependent on such regeneration differences.60–62 In an effort to clone some of these putative nuclear factors, Fischel-Ghodsian and colleagues decided to implement a yeast genetic approach. In analogy to the earlier observation that aminoglycoside-resistance mutations were found in the mitochondrial small rRNA gene, they undertook to clone genes that when overexpressed in yeast would lead to aminoglycoside resistance. Another advantage of identifying genes that


could function as an aminoglycoside “sink,” inactivating the function of the antibiotic, would be their potential therapeutic potential. Exhaustive screening of 35 yeast genome equivalents for genes that, when overexpressed, confer neomycin resistance led to the identification of eight such genes.63,64 Two of these genes seem to lower the intracellular concentration of active aminoglycosides by chemical modification (an acetyltransferase gene) and another through pumping the drug (efflux mechanism) out of the cell (a member of the adenosine triphosphate binding cassette transporters, which includes the P-glycoprotein involved in chemotherapy resistance). The mechanism of aminoglycoside resistance remains unknown for the six other genes. It is also interesting that none of the genes selected plays any role in oxidative stress reduction, which is one of the toxicity mechanisms proposed.22 The genes selected were good candidates for identification of the human homologues and for sequence analysis in patients. The presumed mode of inheritance in patients is autosomal recessive since in a dominant model male-to-male transmission would have been observed. Loss of function appears to be more commonly associated with autosomal recessive mutations, and it is easy to envision models where reduced function of a P-glycoprotein–like molecule can lead to increased cellular damage. These genes might also provide a basis for the development of therapeutic options to prevent or treat aminoglycoside ototoxicity, possibly with local administration of ear drops in the future. A similar approach was taken by Li and colleagues when showing that mutant alleles of MTO1, encoding a mitochondrial protein of unknown function, manifest respiratory-deficient phenotype only when coupled with the yeast mutation corresponding to the A1555G mutation.65 A complementary effort to clone nuclear genes with aminoglycoside-susceptibility mutations uses a candidate gene approach. Candidate genes, such as the human homologues of the genes identified in yeast models described above, are identified and tested for mutations in patients with aminoglycoside ototoxicity. Additional candidate genes are based on genetic studies in Escherichia coli that implicate ribosomal protein subunits S4, S5, and S12 in aminoglycoside sensitivity.66–68 Single amino acid changes in these proteins lead to drugresistant or drug-dependent phenotypes. Other ribosomal proteins have been cross-linked to streptomycin, and although these might also be candidates for aminoglycoside action, the genetic studies more directly define genes that functionally interact with these antibiotics. The human homologues of S4, S5, and S12 in the mitochondrial ribosome would then be the best candidate genes in which to search for susceptibility mutations. Eukaryotic counterparts to two of these candidate genes

147

have been cloned. These mitochondrial ribosomal protein genes are the yeast nam9 gene (the E. coli S4 homologue)69 and the Drosophila technical knockout (tko) gene (the E. coli S12 homologue).70 As a first step, Johnson and colleagues cloned the human mitochondrial ribosomal protein S1271 and screened 40 aminoglycoside-sensitive patients without mitochondrial mutations. No mutation was identified in the coding region, promoter, or 3´ untranslated region of the gene.

PATHOPHYSIOLOGY OF AMINOGLYCOSIDE OTOTOXICITY Notably, the A1555G and C1494T mutations lie exactly in the region of the gene for which the resistance mutations in yeast and Tetrahymena have been described and in which aminoglycoside binding has been documented in bacteria. 36–38 In addition, the mutation makes the mitochondrial RNA gene in this region more similar to the bacterial rRNA gene.35 Since aminoglycosides are concentrated within cochlear cells and remain there for prolonged periods,17 it has been proposed that susceptible individuals with the A1555G mutation have increased binding to aminoglycosides leading to altered protein synthesis in the mitochondria.36 The tissue specificity is presumably owing to the concentration of the drug in those cells. Subsequent binding experiments have proven that increased binding to the mitochondrial 12S rRNA occurs.72 However, when examining lymphoblastoid cell lines of individuals with the A1555G mutation, exposure of the cell lines to high concentrations of neomycin or paromomycin led to a decreased rate of growth in glucose medium and reduced synthesis of mitochondrial proteins, but no mutant proteins were detected.73,74 Similar results of decreased protein synthesis but no mutant proteins were obtained in Japan using mitochondrial transfer from human skin fibroblast line with the A1555G mutation to p0 HeLa cells exposed to very high levels of streptomycin.75 This may indicate that the effect of aminoglycosides in these cell lines could be nonspecific and be different than in the cochlea, perhaps because of different transport of the antibiotic into the mitochondria.

CLINICAL RELEVANCE OF THE GENETIC PREDISPOSITION TO AMINOGLYCOSIDE OTOTOXICITY Prevalence of Inherited Susceptibility to Aminoglycoside Ototoxicity In China, use of aminoglycosides is widespread, Hu and colleagues found that 22% of all deaf-mutes in one district of the city could trace the cause of hearing loss to aminoglycoside use.32 Of those, 28% had other relatives with aminoglycoside ototoxicity.32 In all familial cases of aminoglycoside ototoxicity examined to date, a mito-

148

Interventions

chondrial susceptibility mutation has been identified.35,39–44,48,56,59 In two different populations of sporadic Chinese patients, 4 of 78 and 2 of 36 patients were found to have a predisposing mitochondrial mutation, implying that 5 to 6% of sporadic Chinese patients have this mutation.39,45,55 Combining these numbers from the familial and sporadic cases, one-third of patients with aminoglycoside ototoxicity in China appear to have the A1555G mutation, and a tiny minority appear to have the C1494T or ∆961Cn mutation. In the United States, where aminoglycosides are used much more sparingly, it is rare to find families with several members who lost their hearing after aminoglycosides. A medical record review in two US institutions led to the identification of 41 patients with hearing loss after the administration of aminoglycosides. Molecular analysis revealed that seven of them (17%) had a mitochondrial susceptibility mutation. Four of these patients had some family history of maternal relatives with the same diagnosis.46 The lower frequency in the United States compared with China may be because of the higher frequency of severe concomitant disease leading to hearing loss in the United States, such as renal disease or meningitis, the lower stringency used in selecting patients, or ethnic differences. Clinical Phenotype of Individuals with the A1555G Mutation In the initial studies with Chinese families, the most common history of aminoglycoside ototoxicity indicated that patients with the mitochondrial susceptibility mutation developed severe hearing loss after small doses and with relatively immediate onset. However, in one study of 41 patients in the United States, the entrance criteria for aminoglycoside ototoxicity had been defined very loosely.46 Surprisingly, three of the seven affected patients had only tinnitus and very mild or no initial hearing loss and later had progressive hearing loss over many years, in one case leading to a diagnosis 17 years after aminoglycoside administration.46 A search for another sequence change in the same gene, which might modulate the severity of the hearing loss, led to only one possible sequence change, a homoplasmic C1525T change, in one of the three patients.46 In addition to this clinical variability in the phenotypic expression of aminoglycoside ototoxicity, there is also some evidence that the ototoxicity in patients with the A1555G mutation is restricted to the cochlea and that the vestibular system is not affected. One patient with profound hearing loss after streptomycin had, on extensive testing, normal vestibular function.76 Less extensive testing on the four patients with the A1555G mutation did not reveal any vestibular abnormality. This tissue specificity of the A1555G mutation remains not well understood.49,50

Prevention and Therapy of Aminoglycoside Ototoxicity The most critical clinical issue remains to increase the awareness of physicians and other health care providers of the existence of these susceptibility mutations. It is tragic to see reports of 3 Chinese families with 15 members who became deaf after aminoglycoside exposure, 12 Spanish families with 40 family members, and 1 Italian family with 5 family members; even in the United States, 4 of the 7 patients had other family members with aminoglycoside ototoxicity. Proper family history taking can significantly reduce the frequency of aminoglycoside-induced hearing loss. At a minimum every individual set to receive aminoglycosides should be asked for such a family history, and family members of an individual who became deaf after aminoglycosides should be warned that they are at risk for aminoglycoside ototoxicity. The use of alternative antibiotics is usually possible. In countries where molecular tests are available, testing for the three known susceptibility mutations should be performed. Identification of such mutations will allow targeted counseling of family members and is highly likely to increase compliance as well. A negative result does not rule out a familial cause but, because of the putative autosomal recessive inheritance, will put mainly siblings of the patient at risk. Population screening for disease-susceptibility mutations has not yet become a fully integrated part of clinical practice. However, as the price for such tests comes down, and the number of useful tests increases, it becomes possible to hope that the early screening panels for healthy individuals will include screening for the aminoglycoside-susceptibility mutations. Such panels should definitely include the A1555G mutation. Although it is not yet clear whether screening for the ∆961Cn and C1494T mutations is appropriate and cost effective, one recommends erring on the side of testing. In the longer term, prevention of aminoglycoside ototoxicity might also be possible in a more direct way. The elucidation of the molecular mechanism of aminoglycoside ototoxicity might lead to the identification of those parts of the drug that are necessary for the bactericidal activity and those that are responsible for the toxicity. It will then be possible to create new aminoglycosides in which the toxicity part has been altered. Currently no therapies exist to treat aminoglycoside ototoxicity after the damage has occurred. Approaches using iron-chelating drugs, N-methyl-Daspartate antagonists, and neurotrophic factors have been promising in animal models,23,77,79 and it is likely that these drugs interfere with downstream effects of the primary events. In the longer term it might be possible to locally administer molecules that bind aminoglycosides and prevent their toxic activities in the cochlea. This could be done after hearing loss is


documented but before irreversible hair cell destruction occurs. Such an approach, in combination with mitotic regeneration of the sensory epithelium after hair cell loss, opens the possibility that aminoglycoside ototoxicity can be preventable and treatable.60–62

SUMMARY • Ototoxicity is the major irreversible toxicity of aminoglycosides and occurs in a dose-dependent and idiosyncratic fashion. Research has now clearly demonstrated that there is a genetic basis to aminoglycoside ototoxicity in at least some cases. • Three mitochondrial mutations (the first, the A1555G mutation, discovered in 1993) have been identified in the mitochondrial 12S rRNA gene, which account for ototoxicity in 17 to 33% of cases. Although it is likely that mutations in nuclear genes predispose to aminoglycoside ototoxicity as well, none of these have yet been identified. • The clinical phenotype associated with the mitochondrial mutations can be atypical (ie, not occurring after small dose exposure or with relative immediate onset), in that in some patients the hearing loss is only noticed years after the aminoglycoside exposure and in that the vestibular system may remain preserved. • Ototoxicity associated with these mutations is preventable through a combination of taking family histories and molecular screening. • Future research may further elucidate the genetic factors predisposing to aminoglycoside ototoxicity and result in the development of nontoxic aminoglycoside analogs or in treatment strategies that prevent irreversible cochlear damage.

REFERENCES 1. Sande MA, Mandell GL. Antimicrobial agents. In: Gilman AG, Rall TW, Nies AS, Taylor P, editors. Goodman and Gilman’s the pharmacological basis of therapeutics. 8th ed. Elmsford (NY): Pergamon Press, Inc.; 1990. p. 1098–116. 2. Lortholary O, Tod M, Cohen Y, Petitjean O. Aminoglycosides. Med Clin North Am 1995; 79:761–98. 3. Price KE. Aminoglycoside research 1975-1985: prospects for development of improved agents. Antimicrob Agents Chemother 1986;29:543–8. 4. Prazic M, Salaj B. Ototoxicity with children caused by streptomycin. Audiology 1975;14:173–6. 5. Ryback LP. Drug ototoxicity. Annu Rev Pharmacol Toxicol 1986;26:79–99. 6. Moore RD, Smith CR, Lietman PS. Risk factors for the development of auditory toxicity in patients receiving aminoglycosides. J Infect Dis 1984; 149:23–30.

149

7. McGee TM, Olszewski J. Streptomycin sulfate and dihydrostreptomycin toxicity. Behavioral and histopathologic studies. Arch Otolaryngol 1962;75:295–311. 8. Wersall J. Structural damage of the organ of Corti and the vestibular epithelia caused by aminoglycoside antibiotics in the guinea pig. In: Lerner SA, Matz GJ, Hawkins JE Jr, editors. Aminoglycoside ototoxicity. Boston: Little, Brown and Company; 1981. p. 197–214. 9. Hinojosa R, Lerner SA. Cochlear neural degeneration without hair cell loss in two patients with aminoglycoside ototoxicity. J Infect Dis 1987; 156:3449–55. 10. Duvall AJ, Wersall J. Site of action of streptomycin upon inner ear sensory cells. Acta Otolaryngol 1964;57:581–98. 11. Darrouzet J, Guilhaume A. Ototoxicite de la kanamycin au jour le jour. Etude experimentale en microscopie electronique. Rev Laryngol Otol Rhinol 1974;95:601–21. 12. De Groot JC, Huizing EH, Veldman JE. Early structural effects of gentamicin cochleotoxicity. Acta Otolaryngol 1991;111:273–80. 13. Nuttall AL, Marques DM, Lawrence M. Effects of perilymphatic perfusion with neomycin on the cochlear microphonic potential in the guinea pig. Acta Otolaryngol 1977;83:393–400. 14. Kroese ABA, Das A, Hudspeth AJ. Blockage of the transduction channels of hair cells in the bullfrog’s sacculus by aminoglycoside antibiotics. Hear Res 1989;37:203–17. 15. Puel JL, Lenoir M, Uziel A. Dose-dependent changes in the rat cochlea following aminoglycoside intoxication. I. Physiological study. Hear Res 1987;26:191–7. 16. Vrabec DP, Cody DT, Ulrich JA. A study of the relative concentrations of antibiotics in the blood, spinal fluid and perilymph in animals. Ann Otol Rhinol Laryngol 1965;74:689. 17. Henley CM, Schacht J. Pharmacokinetics of aminoglycoside antibiotics in blood, inner-ear fluids and tissues and their relationship to ototoxicity. Audiology 1988;27:137–46. 18. Lim DJ. A review: effects of noise and ototoxic drugs at the cellular level in the cochlea. Am J Otolaryngol 1986;7:73–99. 19. Williams SE, Zenner HP, Schacht J. Three molecular steps of aminoglycoside ototoxicity demonstrated in outer hair cells. Hear Res 1987;30: 11–8. 20. Jarlstedt J, Bagger-Sjoback D. Gentamicin-induced changes in RNA content in sensory and ganglionic cells in the hearing organ of the lizard Calotes versicolor: a cytochemical and morphological investigation. Acta Otolaryngol 1977;84:361–9.

150

Interventions

21. Stockhorst W, Schacht J. Radioactive labeling of phospholipids and proteins by cochlear perfusion in the guinea pig and the effect of neomycin. Acta Otolaryngol 1977;83:401–9. 22. Song BB, Schacht J. Variable efficacy of radical scavengers and iron chelators to attenuate gentamycin ototoxicity in guinea pigs in vivo. Hear Res 1996;94:87–93. 23. Ingram VM. Gene mutations in human haemoglobin: the chemical difference between normal and sickle cell haemoglobin. Nature 1957; 180:326. 24. Groffen J, Stephenson JR, Heisterkamp N, et al. Philadelphia chromosomal breakpoints are clustered within a limited region, bcr, on chromosome 22. Cell 1984;36:93–9. 25. de The H, Chomienne C, Lanotte M, et al. The t(15;17) translocation of acute promyelocytic leukaemia fuses the retinoic acid receptor alpha gene to a novel transcribed locus. Nature 1990; 347:558–61. 26. Castaigne S, Chomienne C, Daniel MT, et al. Alltrans retinoic acid as a differentiation therapy for acute promyelocytic leukemia. I. Clinical results. Blood 1990;76:1704–9. 27. Stringer SP, Meyerhoff WL, Wright CG. Ototoxicity. In: Paparella WM, Shumrick DA, Gluckman JL, Meyerhoff WL, editors. Otolaryngology. 3rd ed. Philadelphia (PA): WB Saunders Co.; 1991. p. 1653–69. 28. Higashi K. Unique inheritance of streptomycininduced deafness. Clin Genet 1989;35:433–6. 29. Prazic M, Salaj B, Subotic R. Familial sensitivity to streptomycin. J Laryngol Otol 1964;78:1037–43. 30. Tsuiki T, Murai S. Familial incidence of streptomycin hearing loss and hereditary weakness of the cochlea. Audiology 1971;10:315–22. 31. Konigsmark BW, Gorlin RJ. Genetic and metabolic deafness. Philadelphia (PA): WB Saunders Co.; 1976. p. 364–5. 32. Hu D-N, Qiu W-Q, Wu B-T, et al. Genetic aspects of antibiotic induced deafness: mitochondrial inheritance. J Med Genet 1991;28:79–83. 33. Stebbins WC, McGinn CS, Feitosa AG, et al. Animal models in the study of ototoxic hearing loss. In: Lerner SA, Matz GL, Hawkins JE Jr, editors. Aminoglycoside ototoxicity. Boston: Little, Brown and Company; 1981. p. 5–25. 34. Hornig H, Woolley P, Luhrmann R. Decoding at the ribosomal A site: antibiotics, misreading and energy of aminoacyl-tRNA binding. Biochimie 1987;69:803–13. 35. Prezant TR, Agapian JV, Bohlman MC, et al. Mitochondrial ribosomal RNA mutation associated with both antibiotic-induced and non-syndromic deafness. Nat Genet 1993;4:289–94.

36. Gravel M, Melancon P, Brakier-Gingras L. Crosslinking of streptomycin to the 16S ribosomal RNA of Escherichia coli. Biochemistry 1987;26:6227–32. 37. Li M, Tzagaloff A, Underbrink-Lyon K, Martin NC. Identification of the paromomycin-resistance mutation in the 15S rRNA gene of yeast mitochondria. J Biol Chem 1982;257:5921–8. 38. Spangler EA, Blackburn EH. The nucleotide sequence of the 17S ribosomal RNA gene of Tetrahymena thermophila and the identification of point mutations resulting in resistance to the antibiotics paromomycin and hygromycin. J Biol Chem 1985;260:6334–40. 39. Hutchin T, Haworth I, Higashi K, et al. A molecular basis for human hypersensitivity to aminoglycoside antibiotics. Nucleic Acids Res 1993; 21:4174–79. 40. Matthijs G, Claes S, Longo-Mbenza B, Cassiman JJ. Non-syndromic deafness associated with a mutation and a polymorphism in the mitochondrial 12S ribosomal RNA gene in a large Zairean pedigree. Eur J Hum Genet 1996;4:46–51. 41. Pandya A, Xia X, Radnaabazar J, et al. Mutation in the mitochondrial 12S rRNA gene in two families from Mongolia with matrilineal aminoglycoside ototoxicity. J Med Genet 1997;34:169–72. 42. El-Schahawi M, deMunain L, Sarrazin AM, et al. Two large Spanish pedigrees with non-syndromic sensorineural deafness and the mtDNA mutation at nt 1555 in the 12SrRNA gene: evidence of heteroplasmy. Neurology 1997;48:453–6. 43. Estivill X, Govea N, Barcelo A, et al. Familial progressive sensorineural deafness is mainly due to the mtDNA A1555G mutation and is enhanced by treatment with aminoglycosides. Am J Hum Genet 1998;62:27–35. 44. Shohat M, Fischel-Ghodsian N, Legum C, Halpern GJ. Aminoglycoside induced deafness in an Israeli Jewish family with a mitochondrial ribosomal RNA gene mutation. Am J Otolaryngol 1999;20:64–7. 45. Fischel-Ghodsian N, Prezant TR, Bu X, Oztas S. Mitochondrial ribosomal RNA mutation associated with aminoglycoside ototoxicity. Am J Otolaryngol 1993;14:399–403. 46. Fischel-Ghodsian N, Prezant TR, Chaltraw W, et al. Mitochondrial gene mutations: a common predisposing factor in aminoglycoside ototoxicity. Am J Otolaryngol 1997;18:173–8. 47. del Castillo FJ, Rodriguez-Ballesteros M, Martin Y, et al. Heteroplasmy for the 1555A>G mutation in the mitochondrial 12S rRNA gene in six Spanish families with non-syndromic hearing loss: implications for genetic diagnosis and counseling. J Med Genet 2003;40:632–6. 48. Casano RAMS, Bykhovskaya Y, Johnson DF, et al. Hearing loss due to the mitochondrial A1555G


mutation in Italian families. Am J Med Genet 1998;79:388–91. 49. Fischel-Ghodsian N. Mitochondrial genetics and hearing loss—the missing link between genotype and phenotype. Proc Soc Exp Biol Med 1998; 218:1–6. 50. Fischel-Ghodsian N. Mitochondrial mutations and hearing loss—paradigm for mitochondrial genetics. Am J Hum Genet 1998;62:15–9. 51. Bykhovskaya Y, Estivill X, Taylor K, et al. Candidate locus for a nuclear modifier gene for maternally inherited deafness. Am J Hum Genet 2000;66: 1905–10. 52. Bykhovskaya Y, Yang H, Taylor K, et al. Linkage and linkage dysequilibrium mapping of a nuclear modifier gene for maternally inherited deafness. Genet Med 2001;3:177–80. 53. Bykhovskaya Y, Mengesha E, Wang D, et al. Analysis of candidate genes for nuclear-encoded modifiers in families with the mitochondrial A1555G mutation and non-syndromic deafness [abstract]. Am J Hum Genet 2003;73 Suppl: 2122. 54. Bykhovskaya Y, Mengesha E, Wang D, et al. Human mitochondrial transcription factor B1 as a modifier gene for hearing loss associated with the mitochondrial A1555G mutation. [Submitted] 55. Bacino CM, Prezant TR, Bu X, et al. Susceptibility mutations in the mitochondrial small ribosomal RNA gene in aminoglycoside induced deafness. Pharmacogenetics 1995;5:165–72. 56. Casano RAMS, Johnson DF, Hamon M, et al. Inherited susceptibility to aminoglycoside ototoxicity: genetic heterogeneity and clinical implications. Am J Otolaryngol 1999;20:151–6. 57. Tang HY, Hutcheson E, Neill S, et al. Genetic susceptibility to aminoglycoside ototoxicity: how many are at risk? Genet Med 2002;4:336–45. 58. Arnos K, Xia XJ, Norris G, et al. Relative frequencies of the mitochondrial A1555G and 961 delT mutations in the 12SrRNA gene in a large sample of deaf probands from the United States [abstract]. Am J Hum Genet 2003;73:2196. 59 Zhao H, Li R, Wang Q, et al. Maternally inherited aminoglycoside-induced and non-syndromic deafness associated with the novel C1494T mutation in the mitochondrial 12S rRNA gene in a large Chinese family. Am J Hum Genet [In press] 60. Forge A, Li L, Corwin JT, Nevill G. Ultrastructural evidence for hair cell regeneration in the mammalian inner ear. Science 1993;259:1616–9. 61. Warchol ME, Lambert PR, Goldstein BJ, et al. Regenerative proliferation in inner ear sensory epithelia from adult guinea pigs and humans. Science 1993;259:1619–22.

151

62. Lefebvre PP, Malgrange B, Staecker H, et al. Retinoic acid stimulates regeneration of mammalian auditory hair cells. Science 1993;260: 692–5. 63. Prezant TR, Chaltraw W, Fischel-Ghodsian N. Identification of an overexpressed yeast gene which prevents aminoglycoside toxicity. Microbiology 1996;142:3407–14. 64. Johnson DF, Prezant TR, Lubavin B, et al. Isolation of overexpressed yeast genes which prevent aminoglycoside ototoxicity. Hear Res 1998;120: 62–8. 65. Li X, Li R, Lin X, Guan MX. Isolation and characterization of the putative nuclear modifier gene MTO1 involved in the pathogenesis of deafnessassociated mitochondrial 12S rRNA A1555G mutation. J Biol Chem 2002;277:27256–64. 66. Davies J, Nomura M. The genetics of bacterial ribosomes. Annu Rev Genet 1972;6:203–34. 67. Wittmann HG, Apirion D. Analysis of ribosomal proteins in streptomycin resistant and dependent mutants isolated from streptomycin independent Escherichia coli strains. Mol Gen Genet 1975; 141:331–41. 68. Van Acken U. Protein chemical studies on ribosomal proteins S4 and S12 from ram (ribosomal ambiguity) mutants of Escherichia coli. Mol Gen Genet 1975;140:61–8. 69. Boguta M, Dmochowska A, Borsuk P, et al. Nam9 nuclear suppressor of mitochondrial ochre mutations in Saccharomyces cerevisiae codes for a protein homologous to S4 ribosomal proteins from chloroplasts, bacteria, and eucaryotes. Mol Cell Biol 1992;12:402–12. 70. Royden CS, Pirrotta V, Jan LY. The tko locus, site of a behavioral mutation in D. melanogaster, codes for a protein homologous to prokaryotic ribosomal protein S12. Cell 1987;51:165–73. 71. Johnson DF, Hamon M, Fischel-Ghodsian N. Cloning and characterization of the human mitochondrial ribosomal S12 gene. Genomics 1998; 52:363–8. 72. Hamasaki K, Rando RR. Specific binding of aminoglycosides to a human rRNA construct based on a DNA polymorphism which causes aminoglycoside-induced deafness. Biochemistry 1997;36:12323–8. 73. Guan M, Fischel-Ghodsian N, Attardi G. Biochemical evidence for nuclear gene involvement in phenotype of non-syndromic deafness associated with mitochondrial 12S rRNA mutation. Hum Molec Genet 1996; 5:963–72. 74. Guan M-X, Attardi G, Fischel-Ghodsian N. A biochemical basis for the inherited susceptibility to aminoglycoside ototoxicity. Hum Mol Genet 2000;9:1787–93.

152

Interventions

75. Inoue K, Takai D, Soejima A, et al. Mutant mtDNA at 1555 A to G in the 12S rRNA gene and hypersusceptibility of mitochondrial translation to streptomycin can be co-transferred to p0 HeLa cells. Biochem Biophys Res Commun 1996;223:496–501. 76. Braverman I, Jaber L, Levi H, et al. Audio-vestibular findings in patients with deafness caused by a mitochondrial susceptibility mutation and precipitated by an inherited nuclear mutation or aminoglycosides. Arch Otolaryngol Head Neck Surg 1996;122:1001–4.

77. Song BB, Anderson DJ, Schacht J. Protection from gentamicin ototoxicity by iron chelators in guinea pig in vivo. J Pharmacol Exp Ther 1997;282: 369–77. 78. Basile AS, Huang J-M, Xie C, et al. N-Methyl-Daspartate antagonists limit aminoglycoside antibiotic-induced hearing loss. Nat Med 1996;2: 1338–43. 79. Ernfors P, Duan ML, ElShamy WM, Canlon B. Protection of auditory neurons from aminoglycoside toxicity by neurotrophin-3. Nat Med 1996;2:463–7.

CHAPTER 18

Audiologic Monitoring for Ototoxicity Kathleen C. M. Campbell, PhD

Clinically, audiologic monitoring for ototoxicity is generally performed for one of two purposes. The most common purpose is to detect ototoxic changes before hearing in the speech frequency range, and thus communication, is affected. When changes are detected early, the physician can consider alternative treatment protocols, possibly with less ototoxic medications. The second purpose is to monitor changes, when and if they occur, even when it is known that the treatment regimen cannot be safely altered, as in some cancer treatments. In the second case, the purpose of monitoring is to assist the patient and family in maintaining communication as hearing loss develops. This assistance may include counseling, communication strategies, amplification, and assistive listening devices. Certainly, preventing hearing loss and maintaining communication is a major quality-of-life issue, particularly for patients and families dealing with serious, and possibly life-threatening, illness. Audiologic monitoring for ototoxicity is also a very active area of research. Several clinical and research centers are comparing different monitoring and analysis techniques for various patient populations. Further new drugs are being developed that appear to have excellent therapeutic efficacy without ototoxic side effects.1 Other drugs are being developed specifically to prevent ototoxicity when delivered either before or in combination with ototoxic drugs.2–8 Currently there are three main approaches to audiologic monitoring for ototoxicity: the basic audiologic assessment, high-frequency audiometry, and otoacoustic emissions (OAEs). Depending on the tests’ purpose and patient considerations, they may be used separately or in combination. It should be noted that all of these approaches require a baseline evaluation, preferably prior to any ototoxic drug administration, so that later results have a meaningful basis for comparison. Given the high incidence of preexisting hearing loss in the population at large, any assessment for ototoxicity without a baseline evaluation will be difficult to

interpret in regard to cause.9 In those cases, the patient or family may inaccurately attribute a long-standing, but newly diagnosed, hearing loss to the current or recent medical treatment.

THE BASIC AUDIOLOGIC ASSESSMENT The basic audiologic assessment is an important part of most ototoxicity monitoring programs.10 The basic audiologic assessment may not detect the early ototoxic changes detected by high-frequency audiometry and OAEs, but it assesses the patient’s hearing in the speech frequency range for communication, assesses word recognition ability, and detects whether a conductive component is contributing to any hearing loss or change in hearing over time. Although rare, occasionally an ototoxic medication may selectively cause lowor midfrequency hearing loss, best evaluated by testing in the conventional frequency range. If the patient’s treatment protocol cannot be altered even if ototoxic hearing loss is detected, the ototoxicity monitoring protocol may include only the basic audiologic assessment with monitoring in the conventional frequency range to determine when the patient needs assistance with any development of hearing loss. At baseline, the basic audiologic assessment should include pure tone air-conduction thresholds for the conventional frequency range (250–8000 Hz), the frequency range needed for normal speech perception. Standard modified Hughson-Westlake threshold procedures should be routinely employed. 11 Word recognition (speech discrimination) measures should be included, using 50-word lists. If air-conduction thresholds are greater than or equal to 10 dB HL, bone-conduction threshold testing should be included to determine if any air-bone gaps exist, which would indicate a conductive component. Tympanometry at baseline is also recommended to assist in determining if a middle ear disorder may be present. Conductive hearing losses can be common in pediatric populations, among infectious disease patients, such as those

154

Interventions

receiving aminoglycoside antibiotics, and in patients immunosuppressed by chemotherapy regimens. Therefore, any possible conductive component must be parceled out to determine if later threshold changes are sensorineural and possibly attributable to the ototoxic medication or conductive and attributable to other changes. After the baseline evaluation, basic assessment follow-up evaluations will generally comprise only pure tone air-conduction threshold tests unless a change in hearing threshold is noted. If a significant change is noted, the entire basic assessment should be repeated, including bone-conduction threshold testing and tympanometry, to check for a conductive or middle ear component to the loss. Also word recognition testing should be repeated to assist in determining the communicative impact on the patient. If 50-word lists have been used, standardized interpretation criteria can be used to determine if the change is significant.12 Sometimes, because patient populations receiving ototoxic medications may tire easily, audiologists prefer to use 25-word lists or “half-lists”; however, interpreting significant change for word recognition is then more difficult. At the baseline assessment, patients should also be counseled to avoid noise exposure during and, in the case of cisplatin and aminoglycoside antibiotics, for several months following drug administration. Noise exposure can exacerbate the ototoxicity of both aminoglycosides13–15 and cisplatin16–19 although prior noise exposure may not potentiate cisplatin ototoxicity.20 Over the years, several “significant change criteria” have been suggested for determining ototoxic change. Although simple test-retest variability for each threshold should not exceed 5 dB, that criterion is too stringent for determining ototoxic threshold shift, particularly in these patient populations. Early studies of ototoxic threshold shift proposed criteria of either 15 dB or greater at one or more frequencies21 or 20 dB or greater at any one frequency.22 However, Brummet and Morrison reported that these thresholds were exceeded even in control subjects over time.23 Similarly, Meyerhoff and colleagues in a study of patients receiving tobramycin or vancomycin reported that 15 dB shifts at a single frequency or an average of 5 dB shifts across frequencies occurred equally in both negative and positive directions, indicating random variability.24 The most widely used and validated criteria for determining ototoxic threshold shift were published by the American Speech-Language-Hearing Association (ASHA).25 Significant ototoxic change must meet one of the following three criteria: (1) 20 dB or greater decrease at any one test frequency, (2) 10 dB or greater decrease at any two adjacent frequencies, or (3) loss of response at three consecutive frequencies where responses were previously obtained. Changes are always computed relative to baseline measures and must be

confirmed by repeat testing, generally within 24 hours. These criteria minimize random variability by using adjacent test frequencies. It has been demonstrated that these criteria are sensitive to ototoxic change and have not been shown to yield false-positive findings for airconduction threshold testing in either the conventional or high-frequency ranges.4,26,27

HIGH-FREQUENCY AUDIOMETRY AND OTOACOUSTIC EMISSIONS High-frequency audiometry comprises air-conduction threshold testing for the frequencies above 8000 Hz. Threshold testing in the extended high-frequency (EHF) range, between 10000 Hz and 20000 Hz, can usually detect aminoglycoside-induced or cisplatininduced ototoxicity before changes in the conventional frequency range occur.28–34 EHF testing can generally detect these changes early because most ototoxic changes first affect the basal turn, or high-frequency region, of the cochlea before progressing to lowerfrequency regions.35–37 High-frequency audiometry is now well established and widely used in ototoxicity monitoring programs. Test procedures are similar to testing in the conventional frequency range. High-frequency audiometry was first proposed and studied several decades ago. 38–41 However, lack of commercially available equipment and standard calibration references limited its clinical acceptance for many years. Further, even among individuals with normal hearing in the conventional frequency range, wide differences exist for EHF thresholds.42 Because of that high intersubject variability, no standard EHF threshold reference has been developed as it has for the conventional frequency range (ie, dB HL). However, ototoxicity monitoring compares each subject’s hearing thresholds to his or her own threshold at baseline and not to thresholds of other subjects. Consequently, high intersubject threshold variability at baseline is not an issue for ototoxicity monitoring; only intrasubject variability is. When EHF testing equipment was first being developed, intrasubject variability was also problematic because of the transducer effects. The transducers first used for the EHF highly directional waveforms resulted in acoustic variances when coupled to the ear. Numerous investigations addressed these issues and resulted in significant improvements.42–46 Currently, intrasubject variability over time for EHF, using standardized clinical procedures and commercially available equipment, is no higher than for air-conduction threshold testing in the conventional frequency range.1,26,47–51 Further, high-frequency audiometry can be conducted in a quiet hospital room if necessary. 52 Obtaining EHF auditory brainstem response (ABR) thresholds to monitor ototoxicity bedside using a portable unit has been investigated but has not obtained widespread

Audiologic Monitoring for Ototoxicity

clinical use.53,54 However, EHF ABR testing remains an area of investigation.55 High-frequency audiometry may not be applicable to all patients. Patients with hearing loss in the conventional frequency range may not have measurable hearing in the EHF range.56,57 Even when accounting for hearing thresholds in the conventional frequency range, the elderly have disproportionately poorer EHF thresholds,58,59 which may limit monitoring capabilities in that frequency range.58 Some individuals may have better hearing in the EHF than in conventional frequency ranges, but these cases are rare.60 Because patients receiving ototoxic medications are usually seriously ill, sometimes extended audiologic behavioral testing requiring careful attention and patient responses are problematic. Consequently, several investigators have been working to develop abbreviated EHF monitoring protocols using a restricted frequency range.27,53,54,61 OAEs are another option for monitoring ototoxicity. OAEs are acoustic signals generated by the cochlear outer hair cells and transmitted from the cochlea through the middle ear to the ear canal, where they can be detected and recorded with a sensitive low-noise microphone (see Chapter 1, “Anatomy and Physiology of the Cochlea”). Although many normal ears generate OAEs spontaneously, spontaneous OAEs (SOAEs) are not widely used clinically. SOAEs are frequently absent even in normal ears and are generally absent in hearing-impaired ears, thus providing a limited basis for monitoring. The OAEs most commonly used clinically are transient OAEs (TOAEs) and distortionproduct OAEs (DPOAEs). Both TOAEs and DPOAEs are elicited in response to an acoustic stimulus. TOAEs are elicited in response to a series of transient signals, usually clicks, and generally elicit a broad-spectrum cochlear response. DPOAEs are elicited in response to a series of two simultaneous tones. The DPOAEs’ primary tones are called F 1 (lower frequency) and F2 (higher frequency) but elicit a cochlear response of a different frequency, predominantly 2 F1 – F2. OAEs have several advantages over behavioral measures. They are quick, are generally inexpensive, and require no behavioral response from the patient. Thus, they can be obtained on even comatose patients. Therefore, OAEs can be advantageous for ototoxicity monitoring because many of these patients are too ill to comfortably provide reliable behavioral responses over long time periods. Further, like high-frequency audiometry, TOAE 62–65 and DPOAE 66–68 responses change before hearing thresholds in the conventional frequency range. Thus, they detect ototoxicity early, providing the physician and patient with the option of changing the treatment protocol before irreversible changes in the speech frequency range occur. Currently, it is not known whether high-frequency audiometry or

155

OAEs consistently provide the earliest indication of ototoxic change. Research in that area continues. Although both TOAEs and DPOAEs change before hearing thresholds in the conventional frequency range, DPOAEs appear to provide an earlier indication than do TOAEs. 68 This advantage probably exists because DPOAEs can measure at higher frequencies than can TOAEs, thus being more sensitive to the cochlear frequency areas first affected. Further, DPOAEs can often be recorded in the presence of more severe sensorineural hearing loss than can TOAEs, rendering more patients eligible for OAE monitoring.69–71 DPOAEs can also provide some indication of degree and configuration of hearing loss if those data cannot be obtained behaviorally.72,73 One of the primary advantages of high-frequency audiometry over OAEs is that the significant change criteria for EHF testing are well established with excellent specificity and sensitivity.25 A variety of significant change criteria have been proposed for interpreting OAEs,68,74 but none yet have universal acceptance. The sensitivity and specificity of these criteria now need to be documented on large-scale patient populations. Both high-frequency audiometry and OAEs will be problematic in patients with hearing loss, particularly the elderly, because there may be no responses or limited responses available for monitoring.56,57,75,76 However Ress and colleagues found that DPOAEs could be recorded in a greater number of patients than could EHF thresholds and that they were equally sensitive in detecting ototoxic change in those patients who could be tested using both measures.67 The mean age of subjects in their study was 62 years. Thus, presbycusis was probably a factor for several of their subjects. One disadvantage of OAEs is that they cannot be reliably recorded in the presence of otitis media.77 As previously discussed, the patient populations receiving ototoxic medications may be particularly susceptible to otitis media, which can interfere with OAE ototoxicity monitoring.78 Thus, OAEs should probably not be the sole method of ototoxicity monitoring because interruptions in monitoring may occur whenever otitis media is present. High-frequency audiometry can be conducted in the presence of otitis media, but it cannot then be assumed that any changes in the EHF range are secondary to ototoxicity. Whenever changes in any responses are noted, a complete assessment is needed to determine if a conductive component may be contributing to the observed differences.

PEDIATRIC TESTING Children pose special challenges for ototoxicity monitoring. Hearing preservation is particularly critical in this patient population because they are still acquiring speech and language and have many decades of life ahead of them. Their communication abilities will

156

Interventions

greatly affect the quality of their current and future life. However, young children, particularly when they are ill, may not be able to provide sufficient behavioral information for monitoring, particularly for both conventional and high-frequency audiometry. OAEs, requiring no behavioral responses, seem ideal, but otitis media78 or crying65 can preclude reliable recordings for ototoxicity monitoring. ABR testing, particularly with EHF capabilities, holds promise but is currently timeconsuming and may require sedation for toddlers. Repeated sedation for monitoring may be inadvisable for this population. Nonetheless, an experienced audiologist, employing various test techniques, including behavioral, electrophysiologic, and OAE measures, can generally monitor these patients successfully.

OTOTOXICITY TESTING FOR CLINICAL TRIALS Currently the US Food and Drug Administration (FDA) does not have specific Good Clinical Practice (GCP) guidances for monitoring ototoxicity in patients receiving new drugs. A detailed protocol, however, by Campbell and colleagues in 2003 for a phase I clinical trial of a new glycopeptide antibiotic that monitors for both cochleotoxicity and vestibulotoxicity appears especially promising in its applicability.1 All audiologic methods were submitted to the FDA in advance of phase I study data collection, and results were accepted in the subsequent reports. That study included air-conduction testing in the conventional and EHF ranges, bone-conduction testing as indicated, and tympanometry and word recognition at baseline. Patient inclusion and exclusion criteria and replicability criteria were very strict to avoid false-positive or false-negative findings. The Dizziness Handicap Inventory (DHI) additionally was employed to monitor for potential vestibulotoxicity. This study may serve as a helpful model for others performing clinical trials of new drugs in which ototoxicity monitoring is needed.

OTOTOXICITY MONITORING SCHEDULES It is preferable to obtain the baseline evaluation prior to the patient receiving ototoxic medications and before the patient is connected to intravenous or monitoring equipment that may produce high ambient acoustic or electrical noise levels. For platinum-based chemotherapy patients, this is generally possible because the chemotherapy is prescheduled. Because cisplatin can cause marked hearing loss following a single administration, predrug baseline testing is essential.79 For infectious disease patients, aminoglycoside administration may occur on an emergency basis, so prior audiologic assessment may not be possible. Fortunately, even kanamycin, one of the most ototoxic aminoglycosides, generally does not cause demonstrable cochleotoxicity for at least 72 hours after administration.80,81 Consequently, baseline testing for aminoglycosides may occur

prior to or within the first 2 days after drug initiation. However, predrug baseline testing is optimal. Follow-up visits should occur just prior to each round of platinum-based chemotherapy, after any temporary threshold shift has had time to recover and before the patient is connected to intravenous or monitoring equipment. This schedule will also allow patients to be tested when they are feeling at their best, thus providing more reliable behavioral responses. Because platinum-based chemotherapy can cause delayed hearing loss, a follow-up test should also occur a few months after chemotherapy cessation. Generally this testing can be coordinated with a medical followup visit. If the patient also received head and neck radiation, monitoring for the next year or two is advisable because hearing loss may continue to progress. For aminoglycoside antibiotics, weekly or biweekly monitoring is generally recommended. Because aminoglycosides can also cause delayed hearing loss, followup testing should also be scheduled a few months after drug discontinuation. Ototoxicity monitoring is most commonly performed for patients receiving multiple-dose platinumbased chemotherapy and long-term (generally more than 5 days) aminoglycoside antibiotic administration. However, ototoxicity may occur secondary to a wide variety of agents, and monitoring protocols may need to be designed accordingly. For some drugs, the incidence of ototoxicity is so low that prospective monitoring may not have been scheduled. Consequently, the patient may not be referred for audiologic testing until hearing loss is suspected. In those cases, the same criteria for ototoxic change can be applied if baseline data are available. If baseline data are not available, interpretation can be problematic.

MONITORING FOR TINNITUS AND VESTIBULOTOXICITY Tinnitus is a common side effect of many ototoxic drugs,82 particularly cisplatin,34 but currently no formal monitoring procedures have been developed specifically for tinnitus. Most studies do not report how they assessed tinnitus and apparently rely primarily on patient self-report. Further, whether or not tinnitus onset precedes EHF threshold shift or changes in OAEs has not been formally investigated, although tinnitus is frequently considered an early indicator of ototoxicity. When monitoring patients for ototoxicity, formally questioning them about any tinnitus symptoms at each appointment would be advisable. Although the vestibulotoxicity of some drugs, particularly certain aminoglycosides, is well established,82 no standardized, widely accepted guidelines for vestibulotoxicity monitoring exist (see Chapter 19, “Monitoring Vestibular Ototoxicity”). Some vestibular testing protocols would be impractical for weekly


monitoring, particularly for this patient population. A recent study by Campbell and colleagues used the DHI to monitor for vestibular or balance changes in a phase I study of a new glycopeptide.1 The DHI, a simple questionnaire, was administered just prior to each audiologic assessment for ototoxicity monitoring. The DHI was not initially designed for ototoxicity monitoring but is the most commonly used and best validated selfassessment scale for dizziness and dysequilibrium.83–89 In the Campbell and colleagues’ study, the DHI served as a quick, noninvasive, cost-effective method of monitoring vestibular and balance function and yielded no false-positive or false-negative findings in that patient population.1 However, the glycopeptide in that study was not found to be either cochleotoxic or vestibulotoxic. For the purposes of vestibulotoxicity monitoring, the DHI could probably be abbreviated by eliminating questions irrelevant for that purpose.

OTHER CONSIDERATIONS: ENVIRONMENTAL CHEMICALS Although it is well known that environmental chemicals such as organic solvents, asphyxiant gases, and heavy metals can cause hearing and vestibular disorders,90–92 workers exposed to these solvents may not be monitored for ototoxicity. In some cases, these workers may be monitored for noise-induced hearing loss in their environment, with the potential concomitant role of chemical ototoxins overlooked. It may be advisable to consider regular monitoring for ototoxicity in workers routinely exposed to these agents.

SUMMARY • A wide array of audiologic test techniques are available to monitor ototoxicity in various patient populations. Test techniques employed and the testing schedule may vary according to the drug involved, the patient’s age, and the ability to perform behavioral testing. • The purpose of audiologic monitoring is to detect early changes in hearing from ototoxic agents that might influence continued treatment and to assist both the patient and family members in maintaining communication should a hearing loss occur. • Pretreatment baseline audiometry should be performed whenever possible. • As most ototoxic agents affect high-frequency hearing initially, EHF audiometry and OAEs (preferably DPOAEs) may provide the earliest evidence for impending ototoxicity, before conventional speech frequencies are affected. • Audiologic monitoring for platinum-based chemotherapy should occur prior to each round of treatment and for the 1 to 2 years afterward if

157

there has been concomitant radiation to the head and neck. For aminoglycosides, monitoring is recommended every 1 to 2 weeks while on therapy and for at least 4 to 6 months after drug discontinuation. Patients exposed to environmental chemicals should have their hearing assessed on a regular basis. • Ototoxicity monitoring is a very active area of research. Current techniques are constantly being compared, and new techniques are being investigated. In the future, there is hope that otoprotective agents will prevent ototoxicity in most patients.

REFERENCES 1. Campbell KCM, Kelly E, Targovnik N, et al. Audiologic monitoring for potential ototoxicity in a phase I clinical trial of a new glycopeptide antibiotic. J Am Acad Audiol 2003;14:157–69. 2. Campbell KCM, Rybak LP, Meech RP, Hughes L. DMethionine provides excellent protection from cisplatin ototoxicity in the rat. Hear Res 1996; 102:90–8. 3. Campbell KCM, Meech RP, Rybak LP, Hughes LP. D-Methionine protects against cisplatin damage to the stria vascularis. Hear Res 1999;138:13–28. 4. Campbell KCM, Meech, RP, Rybak LP, Hughes LF. The effect of D-methionine on cochlear oxidative state with and without cisplatin administration: mechanisms of otoprotection. J Am Acad Audiol 2003;14:144–56. 5. Kopke R, Liu W, Gabaizedeh R, et al. Use of organotypic cultures of Corti’s organ to study the protective effects of antioxidant molecules on cisplatin-induced damage of auditory hair cells. Am J Otol 1997;18:559–71. 6. Sha S, Schacht J. Antioxidants attenuate gentamicininduced free radical formation in vitro and ototoxicity in vivo: D -methionine is a potential protectant. Hear Res 2000;142:34–40. 7. Doolittle ND, Muldoon LL, Brummett RE, et al. Delayed sodium thiosulfate as an otoprotectant against carboplatin-induced hearing loss in patients with malignant brain tumors. Clin Cancer Res 2001;7:493–500. 8. Blakley BW, Cohen JI, Doolittle ND, et al. Strategies for prevention of toxicity caused by platinumbased chemotherapy: review and summary of the annual meeting of the Blood-Brain Barrier Disruption Program, Gleneden Beach, Oregon, March 10, 2001. Laryngoscope 2002;112:1997–2001. 9. Campbell KCM, Kalkanis J, Glatz FR. Ototoxicity: mechanisms, protective agents, and monitoring. Curr Opin Otolaryngol Head Neck Surg 2000; 8:436–40.

158

Interventions

10. Campbell KCM, Durrant J.D. Audiologic monitoring for ototoxicity. Otolaryngol Clin North Am 1993;26:903–14. 11. Carhart R, Jerer JF. Preferred method for clinical determination of pure-tone thresholds. J Speech Hear Res 1959;24:330–45. 12. Thornton A, Raffin MJM. Speech discrimination scores modeled as a binomial variable. J Speech Hear Res 1977;21:507–18. 13. Dayal VS, Kokshanian A, Mitchell DP. Combined effects of noise and kanamycin. Ann Otol Rhinol Laryngol 1971;80:1–6. 14. Brown JJ, Brummett RE, Meikle MB, Vernon J. Combined effects of noise and neomycin: cochlear changes in the guinea pig. Acta Otolaryngol 1978; 86:394–400. 15. Brown JJ, Brummett RE, Fox KE, Bendrick TW. Combined effects of noise and kanamycin. Arch Otolaryngol 1980;106:744–50. 16. Sharma RP, Edwards IR. cis-Platinum: subcellular distribution and binding to cytosolic ligands. Biochem Pharmacol 1983;32:2665–9. 17. Bhattacharyya TK, Dayal VS. Ototoxicity and noise-drug interaction. J Otolaryngol 1984;13: 361–6. 18. Boettcher FA, Henderson D, Gratton MA, et al. Synergistic interactions of noise and other ototraumatic agents. Ear Hear 1987;8:192–212. 19. Gratton MA, Salvi RJ, Kamen BA, Saunders SS. Interaction of cisplatin and noise on the peripheral auditory system. Hear Res 1990;50:211–23. 20. Laurell G, Borg E. Cisplatin ototoxicity in previously noise-exposed guinea pigs. Acta Otolaryngol 1986;101:66–74. 21. Thompson PL, Northern JL. Audiometric monitoring of patients treated with ototoxic drugs. In: Lerner SA, Matz GJ, Hawkins JE, editors. Aminoglycoside ototoxicity. Boston: Little, Brown & Co; 1981. p. 237–48. 22. Reddel RR, Kefford RF, Grant JM, et al. Ototoxicity in patients receiving cisplatin: importance of dose and administration. Cancer Treat Rep 1982; 66:9–23. 23. Brummett RE, Morrison RB. The incidence of aminoglycoside antibiotic induced hearing loss. Arch Otolaryngol Head Neck Surg 1990;116:406–10. 24. Meyerhoff WL, Maale GE, Yellin W, et al. Audiologic threshold monitoring of patients receiving ototoxic drugs. Ann Otol Rhinol Laryngol 1989; 98:950–4. 25. American Speech-Language-Hearing Association. Guidelines for the audiologic management of individuals receiving cochleotoxic drug therapy. ASHA 1994;36 Suppl 12:11–9. 26. Frank T. High frequency (8 to 16 kHz) reference thresholds and intrasubject threshold variability

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38. 39. 40.

41.

42.

43.

44.

relative to ototoxicity criteria using a Sennheiser HAD 200 earphone. Ear Hear 2001;22:161–8. Fausti SA, Henry JA, Helt WJ, et al. An individualized, sensitive frequency range for early detection of ototoxicity. Ear Hear 1999;20:497–505. Jacobson EJ, Downs MP, Fletcher JL. Clinical findings in high frequency thresholds during known ototoxic drug usage. J Auditory Res 1969;9:379–85. Fausti SA, Rappaport BZ, Schechter MA, et al. Detection of aminoglycoside ototoxicity by high frequency auditory evaluation: selected case studies. Am J Otolaryngol 1984;5:177–82. Fausti SA, Schechter MA, Rappaport BZ, et al. Early detection cisplatin ototoxicity: selected case reports. Cancer 1984b;53:224–31. Fausti SA, Henry JA, Shaffer HI. High-frequency audiometric monitoring for early detection of aminoglycoside ototoxicity. J Infect Dis 1992; 165:1026–32. Rappaport BZ, Fausti SA, Schechter MA, et al. Detection of ototoxicity by high-frequency auditory evaluation. Semin Hear 1985;6:369–77. Tange RA, Dreschler WA, van der Hulst RJ. The importance of high-tone monitoring for ototoxicity. Arch Otorhinolaryngol 1985;242:77–81. Kopelman J, Budnick AS, Kramer MB, et al. Ototoxicity of high-dose cisplatin by bolus administration in patients with advanced cancers and normal hearing. Laryngoscope 1988;98(8 Pt 1):858–64. Fee WE. Aminoglycoside ototoxicity in the human. Laryngoscope 1980;90(10 Pt 2 Supplement 24): 1–19. Wright CG, Schaefer SD. Inner ear histopathology in patients treated with cisplatin. Laryngoscope 1982;92:1408–13. Schuknecht HF. Disorders of intoxication. In: Pathology of the ear. Philadelphia: Lea & Febiger; 1993. p. 255–77. Fletcher H. Speech and hearing. New York: Van Nostrand; 1929. Fletcher JL. Reliability of high frequency thresholds. J Auditory Res 1965;5:133–7. Rosen S, Plester D, El-Mofty A, Rosen H. High frequency audiometry in presbycusis. Arch Otolaryngol 1964;79:18–32. Zilis T, Fletcher JL. Relation of high frequency thresholds to age and sex. J Auditory Res 1966; 6:189–98. Northern JL, Ratkiewicz B. The quest for highfrequency normative data. Semin Hear 1985; 6:331–9. Fausti SA, Frey RH, Erickson DA, et al. A system for evaluating auditory function from 8000–20,000 Hz. J Acoust Soc Am 1979;66:1713–8. Stelmachowicz PG, Gorga MP, Cullen JK. A calibration procedure for the assessment of thresholds


45.

46.

47.

48.

49.

50. 51.

52.

53.

54.

55.

56. 57.

58.

59.

60. 61.

above 8000 Hz. J Speech Hear Res 1982;25: 618–23. Tonndorf J, Kurman B. High frequency audiometry. Ann Otol Rhinol Laryngol 1984;93(6 Pt 1): 576–82. Valente M, Valente M, Goeble J. High-frequency thresholds: circumaural versus insert earphone. J Am Acad Audiol 1992;3:410–8. Fausti SA, Frey RH, Rappaport BZ, Schechter MA. High frequency audiometry with an earphone transducer. Semin Hear 1985;6:347–57. Dreschler WA, van der Hulst RJ, Tange RA, Urbanus NA. The role of high frequency audiometry in early detection of ototoxicity. Audiology 1985;24: 387–95. Feghali JG, Bernstein RS. A new approach to serial monitoring of ultra-high frequency hearing. Laryngoscope 1991;101:825–9. Frank T, Dreisbach LE. Repeatability of high frequency thresholds. Ear Hear 1991;12:294–5. Frank T. High-frequency hearing thresholds in young adults using a commercially available audiometer. Ear Hear 1990;11:450–4. Valente M, Potts LG, Valente M, et al. High frequency thresholds: sound suite versus hospital room. J Am Acad Audiol 1992;3:287–94. Fausti SA, Frey RH, Henry JA, et al. Early detection of ototoxicity using high frequency, tone-burst evoked auditory brainstem responses. J Am Acad Audiol 1992;3:397–404. Fausti SA, Frey RH, Henry JA, et al. Portable stimulus generator for obtaining high-frequency (8–14 kHz) auditory brainstem response responses. J Am Acad Audiol 1992;3:166–75. Fausti SA, Flick CL, Bobal AM, et al. Comparison of ABR stimuli for the early detection of ototoxicity: conventional clicks compared with high frequency clicks and single frequency tonebursts. J Am Acad Audiol 2003;14:239–50. Osterhammel D. High frequency audiometry. Clinical aspects. Scand Audiol 1980;9:249–56. Kujansuu E, Rahko T, Punnonen R, Karma P. Evaluation of the hearing loss associated with cisplatin treatment by high-frequency audiometry. Gynecol Oncol 1989;33:321–2. Stelmachowicz PG, Beauchaine KA, Kalberer A, Jesteadt W. Normative thresholds in the 8-20 kHz range as a function of age. J Acoust Soc Am 1989; 86:1384–91. Wiley TL, Cruikshanks KJ, Nondahl DM, et al. Aging and high-frequency hearing sensitivity. J Speech Lang Hear Res 1998;41:1061–72. Berlin CI. Unusual forms of residual highfrequency hearing. Semin Hear 1985;6:389–95. Dreschler WA, van der Hulst RJ, Tange RA, Urbanus NA. Role of high frequency audiometry

62.

63.

64.

65.

66.

67.

68.

69.

70. 71.

72.

73.

74.

75.

76.

159

in the early detection of ototoxicity. II clinical aspects. Audiology 1989;28:211–20. Plinkert PK, Krober S. Fruherkennung einer Cisplatin-Ototoxizitat durch evosierte otoakustische Emissionen. Laryngorhinootologie 1991;70:457–62. Beck A, Maurer J, Welkoborsky HJ, Mann W. Changes in transitory evoked otoacoustic emissions in chemotherapy and with cisplatin and 5FU. HNO 1992;40:123–7. Zorowka PG, Schmitt HJ, Gutjahr P. Evoked otoacoustic emissions and pure tone threshold audiometry in patients receiving cisplatinum therapy. Int J Pediatr Otorhinolaryngol 1993;25:73–80. Stavroulaki P, Apostolopoulos N, Dinopoulo D, et al. Otoacoustic emissions—an approach for monitoring aminoglycoside induced ototoxicity in children. Int J Pediatr Otorhinolaryngol 1999;50:177–84. Mulheran M, Degg C. Comparison of distortion product OAE generation between a group requiring frequent gentamicin therapy and control subjects. Br J Audiol 1997;31:5–9. Ress BD, Sridhar KS, Balkany TJ, et al. Effects of cis-platinum chemotherapy on otoacoustic emissions: the development of an objective screening protocol. Otolaryngol Head Neck Surg 1999; 121:693–701. Lonsbury-Martin BL, Martin GK. Evoked otoacoustic emissions as objective screeners for ototoxicity. Semin Hear 2001;22:377–91. Wier CC, Pasanen EG, McFadden D. Partial dissociation of spontaneous otoacoustic emissions and distortion products during aspirin use in humans. J Acoust Soc Am 1988;84:230–7. Norton SJ. Cochlear function and otoacoustic emissions. Semin Hear 1992;13:1–14. Probst R, Lonsbury-Martin BL, Martin GK. A review of otoacoustic emissions. J Acoust Soc Am 1991;20:2021–7. Lonsbury Martin BL, Martin GK. The clinical utility of distortion-product otoacoustic emissions. Ear Hear 1990;11:144–54. Martin GK, Ohlms LA, Franklin DJ, et al. Distortion product emissions in humans. III: Influence of sensorineural hearing loss. Ann Otol Rhinol Laryngol Suppl 1990;147:30–42. Katbamna B, Homnick DN, Marks JH. Effects of chronic tobramycin treatment on distortion product otoacoustic emissions. Ear Hear 1999;20: 393–402. Lonsbury-Martin BL, Cutler WM, Martin GK. Evidence for the influence of aging on distortionproduct otoacoustic emissions in humans. J Acoust Soc Am 1991;89:1749–59. Stover L, Norton SJ. The effects of aging on otoacoustic emissions. J Acoust Soc Am 1993;94: 2670–81.

160

Interventions

77. Owens JJ, McCoy MJ, Lonsbury-Martin BL, Martin GK. Influence of otitis media on evoked otoacoustic emissions in children. Semin Hear 1992; 13:53–66. 78. Allen GC, Tiu C, Koike K, et al. Transient-evoked otoacoustic emissions in children after cisplatin chemotherapy. Otolaryngol Head Neck Surg 1998;118:584–8. 79. Durrant JD, Rodgers G, Meyers EN, Johnson JT. Hearing loss risk factor for cisplatin ototoxicity? Observations. Am J Otol 1990;11:375–7. 80. Brummett RE. Animal models of antibiotic ototoxicity. Rev Infect Dis 1983;5 Suppl 2: S294–303. 81. Brummett RE, Fox KE. Studies of aminoglycoside ototoxicity in animal models. In: Whelton A, Neu HC, editors. The aminoglycosides. New York: Marcel Decker, Inc.; 1982. p. 419–51. 82. Seligman H, Podoshin L, Ben-David J, Fradis M. Drug-induced tinnitus and other hearing disorders. Drug Saf 1996;14:198–212. 83. Jacobson GP, Newman CW. The development of the Dizziness Handicap Inventory. Arch Otolaryngol Head Neck Surg 1990;116:424–7. 84. Jacobson GP, Newman CW, Hunter L, Balzer GK. Balance Function Test correlates of the Dizziness Handicap Inventory. J Am Acad Audiol 1991; 2:253–60.

85. Fielder H, Denholm SW, Lyons RA, Fielder CP. Measurement of health status in patients with vertigo. Clin Otolaryngol 1996;21:124–6. 86. Enloe LJ, Shields RK. Evaluation of health-related quality of life in individuals with vestibular disease using disease-specific and general outcome measures. Phys Ther 1997;77:890–903. 87. Cowand JL, Wrisley DM, Walker M, et al. Efficacy of vestibular rehabilitation. Otolaryngol Head Neck Surg 1998;118:49–54. 88. Whitney SL, Hudak MT, Marchetti GF. The Activities-Specific Balance Confidence Scale and the Dizziness Handicap Inventory: a comparison. J Vestib Res 1999;9:253–9. 89. Jacobson GP, Calder JH. Self-perceived balance disability/handicap in the presence of bilateral peripheral vestibular system impairment. J Am Acad Audiol 2000;11:76–83. 90. Rybak LP. Hearing: the effects of chemicals. Otolaryngol Head Neck Surg 1992;106:677–86. 91. Morioka I, Kuroda M, Miyashita K, Takeda S. Evaluation of organic solvent ototoxicity by the upper limit of hearing. Arch Environ Health 1999; 54:341–6. 92. Sulkowski WJ, Kowalska S, Matja W, et al. Effects of occupational exposure to a mixture of solvents on the inner ear: a field study. Int J Occup Med Environ Health 2002;15:247–56.

CHAPTER 19

Monitoring Vestibular Ototoxicity Vitaly E. Kisilevsky, MD, R. David Tomlinson, PhD, Paul J. Ranalli, MD, FRCPC, and Narayanan Prepageran, MBBS, FRCS(Ed), FRCS(Glas), MS(ORL)

The goal of vestibular monitoring when potentially ototoxic medications are administered should be to provide information about the vestibular system that would allow for timely intervention before an irreversible vestibulotoxic event occurs. In everyday life, maintaining one’s balance depends on information from visual, vestibular, and proprioceptive systems that are centrally integrated at the level of the vestibulocerebellum.1 Recent advances in vestibular testing provide a variety of measurement tools for assessing these systems. Laboratory tests commonly used to evaluate the balance system include electronystagmography (ENG), rotational chair (sinusoidal and pseudorandom), sacculocolic testing, vestibular evoked potentials, and computerized dynamic posturography (CDP). Along with these quantitative or so-called “objective” tests for physical performance measurements, there are several clinical bedside tests and subjective self-reported measures. A flaw all vestibular-monitoring paradigms share (especially those involving quantitative testing) is an inability to recognize subtle changes in function indicative of impending vestibulotoxicity that does not always correlate with changes in the ability to perform one’s daily activities. Conversely, one benefit of subjective measures is their direct “real-time” relation to daily function. Recognition that an individual might be experiencing early signs or symptoms of vestibulotoxicity is important as any window of time for recovery is often short. For this reason, proper objective and subjective monitoring of vestibular function may help recognize toxic effects and prevent permanent damage.

ANATOMICAL BASIS FOR VESTIBULAR MONITORING Maintaining body equilibrium and posture in daily activity is a complex function involving multiple sensory organs. Specific reflexes related to the vestibular system result in stereotypic motor responses for eye movements, postural control, and perceptual outputs.

In short, multiple sensory inputs from the vestibular end-organs, the visual system, and the somatosensory and proprioceptive systems are integrated at the level of the brainstem and cerebellum and are subsequently influenced by the cerebral cortex. Second-order neurons in the vestibular nuclei form specific vestibulospinal and vestibulocerebellar tracts. Connection between vestibular and oculomotor nuclei, in part via the medial longitudinal fasciculus, forms the anatomic basis of the vestibulo-ocular reflex (VOR) (see Chapter 2, “Physiology of the Vestibular System”).

NEURAL BASIS OF THE VOR Traditionally, the VOR has been believed to depend on two pathways: a direct pathway known as the threeneuron arc and an indirect pathway thought to involve the reticular formation. This is now known to be an oversimplification. Two different neuron types in the vestibular nuclei are involved in the generation of the VOR: position-vestibular-pause (PVP) cells and floccular target neurons (FTNs). Each forms its own threeneuron arc. PVP cells receive input from the labyrinth and project to the appropriate oculomotor nuclei. They do not receive input from the cerebellum. Unlike PVP cells, FTNs receive direct input from the cerebellar flocculus. Whereas the PVP cell pathway is relatively unmodifiable, the FTNs change their firing profiles whenever a change in VOR performance is required. For example, if a subject wears 2× magnifying glasses, since the size of the seen world is doubled, the VOR gain must also double. This gain increase is almost completely a result of changes in FTN, rather than PVP, firing profiles. Basically, following adaptation to the magnifying glasses, the same rotation produces a much larger change in FTN firing than is observed prior to the adaptation. Compensation for peripheral lesions requires a similar gain increase. Since the flocculus is known to be required for proper compensation following peripheral vestibular lesions, FTNs are thought to

162

Interventions

CBLM LR

FTN

VI

LSC

RSC

LPC

RPC

VIII

PVP Figure 19-2 The right superior canal (RSC) and the left posFigure 19-1 Basic wiring of the horizontal VOR. CBLM =

cerebellum; FTN = floccular target neuron; LR = lateral rectus muscle; PVP = position-vestibular-pause cell; VI = VIth nucleus; VIII = VIIIth nerve.

play a major role. The basic layout from the horizontal VOR is shown in Figure 19-1. For simplification, only the direct excitatory pathways are shown.

NEUROPHYSIOLOGICAL BASIS FOR COMPENSATION FOLLOWING UNILATERAL VESTIBULAR LOSS Several facts are necessary to understand the consequences of unilateral peripheral vestibular lesions (as might occur from topical ototoxicity). The primary afferent neurons arising from the canals and otoliths are spontaneously active (“tonic” activity). As a result, horizontal rotation to the right results in an increase in firing in the right lateral canal (a branch of the superior vestibular) nerve and a corresponding decrease in firing in the left lateral canal nerve. Because of the way the information is processed in the vestibular nuclei, any difference between the firing rates on the two sides is interpreted as a rotation, and vestibular eye movements will be generated to compensate for this rotation. The six canals on the two sides form three functional pairs. The two lateral canals form one pair, whereas the superior canal on one side and the posterior canal on the other side form another pair. Since there are two superior and two posterior canals in total, these four canals form two functional pairs (Figure 192). In each case, a rotation of the head in the plane of the canals results in an increase in firing in one member of the pair and a decrease in the other member. Immediately following complete unilateral vestibular lesions, two main things happen: (1) the VOR gain drops to about half of its normal value because of the loss of input from one side and (2) a spontaneous nystagmus (away from side of lesion) develops because there is now a difference in firing between the lesioned side and the normal side. The

terior canal (LPC) are parallel and so form a functions pair, as do the right posterior canal (RPC) and the left superior canal (LSC).

reason for the gain decrease can easily be understood by looking at Figure 19-3. Since type I secondary vestibular neurons, such as PVP cells, receive excitation from the ipsilateral and inhibition from the contralateral labyrinth, they effectively respond to the difference in firing between the two VIIIth nerves. Following a unilateral lesion, when the head moves, only the healthy labyrinth will respond, and so the change in firing in the PVP cell will only be half as large as normal. Both of these deficits need to be corrected for full compensation. In addition, this compensation requires the integrity of the cerebellum. Spontaneous nystagmus will largely resolve without any intervention simply because the brain knows that a rotation cannot be going on indefinitely. As a result, the imbalance in activity levels in the two vestibular nuclei that exists following a unilateral lesion largely disappears within a few days as the tonic activity in the nucleus spontaNORMAL +

I

_

I

II +

+

LESION +

I

_

II +

I

+

X Figure 19-3 Type I secondary vestibular neurons receive

excitatory input from the ipsilateral labyrinth and inhibition from the contralateral labyrinth via type II neurons. Following a lesion (X), the type II neuron no longer receives any input and so the type I neuron contralateral to the lesion will generate half as large a response as normal when the head moves.

Monitoring Vestibular Ototoxicity

neously recovers. Recalibration of the VOR, however, takes somewhat longer. Further, this recalibration can occur only if the brain knows that something is still wrong. It knows this because the low VOR gain results in image motion on the retina (“retinal blur”) whenever the head moves. Thus compensation requires both head movements and intact vision. Again, this step requires cerebellar integrity. Experiments have shown that FTNs receive input from the cerebellum, whereas PVP cells do not. Not surprisingly, since the cerebellum is required for VOR compensation, FTNs show large changes in their firing profiles when the VOR gain is changed, whereas PVP cells change very little. In summary, the VOR is based on two parallel pathways through the vestibular nuclei, a highly modifiable FTN pathway and a relatively fixed PVP cell pathway. Compensation is believed to require that all three components—the FTNs, the PVP cells, and the cerebellum—be intact.

WHY COMPENSATION IS NEVER TRULY COMPLETE Despite the ability of the brain to compensate for vestibular lesions, this compensation is incomplete. Halmagyi and Curthoys showed that when very rapid head turns (head thrusts) were used to invoke the VOR, rotations toward the lesioned side evoked eye movements of much lower gain than rotations toward the healthy side.2 Further, this asymmetry shows little or no recovery with time. Likely this maintained asymmetry causes many patients with unilateral lesions to continue to complain that they experience disorientation or residual imbalance during rapid head movements, even though their rotation test results are normal. Since regular rotation testing shows no such asymmetry, what is different about these rapid head thrusts used by Halmagyi and Curthoys? Over the last few years, several experimenters have attempted to answer this question, but the success has been incomplete. The head thrust is different from regular rotation testing because the frequencies and accelerations are much higher in head thrusts than can be attained using rotation chairs. Currently both of these factors are believed to play a role. High-frequency stimuli, particularly if those stimuli involve high accelerations, exhibit large asymmetries, whereas low-frequency stimuli do not. The reasons are beyond the scope of this chapter, but they may be related to the high-frequency performance of the FTN pathway. Despite these limitations, compensation in healthy individuals for unilateral lesions is quite good, largely because each canal in a functional pair is able to sense rotation in both directions. Thus, within limits, the information sensed by the remaining healthy canal is able to replace that lost through the lesion. However, this presupposes that the remaining side is healthy. Should a bilateral lesion occur, compensation will be

163

poor as there is no other source of information with the necessary high-frequency performance. Although visual following mechanisms and even the cervico-ocular reflex may be augmented to help, neither is able to replace the lost vestibular input.

PATHOPHYSIOLOGIC BASIS FOR VESTIBULAR MONITORING Vestibular toxicity can vary from a minimal, clinically undetectable disturbance to a devastating complete bilateral loss of vestibular function. The degree of vestibular loss for the most part depends on the extent of cellular damage within the vestibular end-organ. With vestibular ototoxicity, the initial and most extensive hair cell damage occurs in the apex of the cristae and the striolar regions of the maculae. Further hair cell loss extends toward the peripheral vestibular receptors, and there is additional damage to the otoconial membrane and the otolith structures themselves.3 When streptomycin, for example, is given to animals, hair cell damage is most pronounced in the central part of crista, type I hair cells being more vulnerable than type II hair cells.4 Takumida and colleagues, Meza and colleagues, and Norris have demonstrated that aminoglycoside antibiotics attack and first destroy the stereocilia and the sensory cells of the cristae without damaging the attached first-order neurons.5–7 This destructive action on the sensory cells without damage to the rest of the vestibular system has been verified by experiments on cats with streptomycin by Norris and Shea8 and in guinea pigs with gentamicin by Kimura and colleagues.9 Morphological studies of the human vestibular organ after intratympanic treatment with gentamicin have demonstrated different stages of degeneration of hair cells. Many hair cells were totally missing, and a common finding demonstrated type I sensory cells to be vacuolated more than type II cells. In addition, neural elements showed signs of degeneration with severe swelling of regions of the nerve terminals. No clear synaptic areas with normal morphology were observed.10 A human temporal bone study of aminoglycoside ototoxicity by Tsuji and colleagues demonstrated significant hair cell loss, especially type I hair cells, in all three cristae, but not for the maculae.11 There was no significant loss of Scarpa’s ganglion cells. In contrast to aminoglycosides, heavy metal intoxication results in damage to the central vestibular system. Gozdzik-Zolnierkiewicz and Moszynski found segmental demyelination and axonal degeneration of the VIIIth nerve in young guinea pigs with chronic lead intoxication.12 Examination of the central nervous system (CNS) in patients with mercury poisoning revealed a selective sensitivity of the granule-cell layer of the neocerebellum and demyelination of the subcortical and brainstem white matter.13

164

Interventions

CLINICAL FEATURES OF VESTIBULAR TOXICITY REQUIRING MONITORING The clinical features that make monitoring of vestibular function for ototoxicity a challenge include its gradual or apparently sudden development, its delayed onset from beginning of treatment, the possible spontaneous reversibility of vestibular symptoms, and the striking difference in clinical presentation in patients with unilateral and bilateral vestibular loss. Systemic ototoxicity typically causes a bilateral, simultaneous, symmetric loss of vestibular function manifested by symptoms and signs of ataxia and oscillopsia (visual blurring with head movement) but without a history of vertigo (defined as a hallucination of movement). The absence of vertigo in these patients may be confusing if one is looking for typical symptoms of acute unilateral vestibular disorder. As such, a bilateral process would not cause significant asymmetry of electrical activity at the level of the vestibular nucleus in the brainstem (ie, an individual would not perceive any hallucination of movement at rest), and the bilateral process might be slow enough to allow some central compensatory mechanism to occur. Not all patients, however, are able to compensate for a bilateral vestibular loss, and interruptions in the VOR and vestibulospinal tract inputs can have a devastating effect on their daily activities.* As demonstrated in animal studies, the entry of ototoxic drugs such as aminoglycosides into the labyrinthine fluids is slow for perilymph and extremely slow for endolymph. As a result, it takes time for the drug to reach significant concentrations, which may account for the delay in onset of clinical symptoms.14 Once in the inner ear fluids, the drug is slowly eliminated from the perilymph but persists for extended periods of time in the endolymph. This further explains why ototoxicity may continue even after the drug is withdrawn. The pattern of degeneration in aminoglycoside toxicity seems to depend on the dosing schedule, total cumulative dosage, and route of administration. High but brief peak plasma levels apparently yield markedly lower inner ear tissue concentrations than do plateau plasma levels.15 As pointed out by Black and Pesznecker,3 the true incidence of vestibular ototoxicity may be underestimated as the initial ototoxic destruction of hair cells occurs well outside the normal active and passive head movement frequency range for the vestibular system. *There is no real way to compensate for a complete bilateral loss as there is simply no other major source for the missing vestibular information. People may learn to live without it, but the compensation is mainly psychological rather than physiological. Thus, a person with a complete bilateral loss has no VOR, at least as long as the head movements are passive. However, compensation may occur to a limited extent if the lesions are incomplete.

Subclinical vestibular dysfunction from ototoxic therapy, especially resulting from posterior cristae (semicircular canal) or otolithic injury, is also difficult to document. Most patients who receive potentially vestibulotoxic medications systemically are severely ill, hospitalized, or on bed rest. Laboratory vestibular testing can rarely be performed because of the patient’s general condition. Unfortunately, symptoms of vestibular ototoxicity are usually first noted only when patients start to ambulate without help. They demonstrate an ataxic gait and lose their balance in the dark and when moving quickly. They may need to hold on to a wall for support. In the absence of vertigo, these symptoms are often assumed to be a result of an underlying metabolic, orthopedic, neurologic, or possibly psychiatric disorder. Considering the differences in the clinical presentation of bilateral and unilateral vestibular loss, the guidelines for vestibular monitoring should be tailored for the specific therapeutic protocol a patient is receiving. Prepageran and colleagues have reported several cases of systemic vestibulotoxicity in patients properly monitored for plasma-level aminoglycoside concentrations.16 They found that an increased risk for gentamicin ototoxicity included a prolonged duration of treatment as the only risk factor (greater than 14 days). This confirms the finding of other investigators that vestibulotoxicity may be seen even when serum antibiotic levels remain within the therapeutic range.17 The prolonged and hence total cumulative dose of gentamicin (typically more than 2.5 g) has more impact on probability of ototoxicity than does peak serum level concentration. Several other factors influencing the risk for systemic ototoxicity include the dosage regimen (single vs multiple), concurrent diuretic therapy, renal impairment, and the underlying disease process (Table 19-1). The development of vestibulotoxicity following topical treatment demands a high index of clinical suspicion and an ability to distinguish whether the vestibuTable 19-1 Risk Factors for Aminoglycoside Vestibulotoxicity

Drug related

Known vestibulotoxic agent, prolonged (>10–14 days) treatment

Patient related

Age Renal insufficiency (before or during treatment) Genetic susceptibility, repeated treatment courses

Physician related

Concurrent diuretic and other known ototoxic drug administration Failure to recognize early symptoms of vestibular loss Failure to recognize ototoxicity as the cause of vestibular loss


lar signs are a result of toxicity or the underlying ear disease. Kisilevsky and colleagues reported symptoms and signs of vestibulotoxicity on average after 18 days in patients treated with topical gentamicin for their middle ear condition.18 The duration was somewhat shorter (11 days) in patients with unilateral Meniere’s disease (with normal middle ears) who underwent chemical ablation of vestibular function with commercially available topical gentamicin administration. The risk of development of topical ototoxicity varied, depending on the condition of the middle ear, duration of treatment, and dosage regimen. As demonstrated by Inoue and colleagues, low-dose gentamicin treatment apparently caused less damage to the vestibular system than did highly concentrated topical gentamicin.19 The timing of the appropriate clinical and laboratory vestibular testing can sometimes be based on the known pharmacokinetics of an ototoxic medication and the treatment protocol used. Experience with topical aminoglycosides for ablation of vestibular function in Meniere’s disease demonstrates that when a highly concentrated gentamicin was employed for chemical ablation, the duration for the development of ototoxicity was briefer than in the treatment with commercially available drops.20,21 Recognition and prevention of possible ototoxic effects depend on timely reassessment during treatment.

LABORATORY EVALUATION OF VOR The goal of VOR testing is to determine whether the labyrinthine end-organ and the CNS pathways are functioning normally. A normal response is obtained only when all components of the reflex arc are intact. As demonstrated in avian experiments by Goode and colleagues,22 the VOR can be eliminated with aminoglycoside antibiotic treatment and apparently recovers as hair cells regenerate. Recovery of the VOR highly correlates with the regeneration of type I hair cells. To

165

date, no human data, however, can be found to conclusively support these findings (Table 19-2).

ENG Various systems are currently available for recording eye movements. Historically, ENG is the most available and widely used vestibular test. The principle of ENG is based on the measurement of changes in the corneoretinal potential that occurs with eye movements with respect to electrodes placed around the eyes. Recordings are usually made with a three-electrode system using differential amplifiers. The difference in electrical potential between these electrodes is amplified, graphed, and digitally stored. The ENG test battery includes (1) oculomotor testing for evaluation of CNS function, (2) measurement of spontaneous nystagmus, (3) positional testing for conditions such as benign positional vertigo or atypical positional nystagmus, and (4) caloric testing. The caloric test, despite its limitations, remains the time-honored gold standard for evaluating a unilateral vestibular deficit. The caloric test uses a nonphysiologic stimulus (water or air) to induce endolymphatic flow in the semicircular canals (SCCs) by creating a temperature gradient from one side of the canal to the other. The caloric response is provoked by two mechanisms: endolymph convection, which depends on head position and accounts for about 80% of the response, and direct thermal effect, which is independent of head position and accounts for approximately 20% of the response.23 The caloric test is specifically a test of the horizontal SCC since this canal develops the largest temperature gradient, being anatomically nearest to the external auditory canal. Being the only vestibular test allowing stimulation of each labyrinth separately, the caloric test has proved to be highly sensitive for unilateral peripheral vestibular lesions. The use of caloric testing, however, is limited in

Table 19-2 Relative Advantages and Limitations of Laboratory Vestibular Tests Laboratory Vestibular Test Advantages

Disadvantages

ENG caloric

Possibility of separate side stimulation Availability Highly sensitive for unilateral loss

Tests mainly lateral SCC Nonphysiologic stimulus Low sensitivity in bilateral loss Tests only low frequencies

Rotating chair

Allows for high-frequency testing Physiologic stimulus Allows differentiation between central and peripheral impairment

Bilateral simultaneous stimulation only Stimulation in horizontal plane only Lower than in caloric side sensitivity

CDP

Assesses overall balance performance Directly reflects ability of everyday activity May identify abnormalities in case of normal ENG caloric Useful in rehabilitation and quantification of overall balance

Unable to localize site of lesion within vestibular system Possibility of “learning curve” in case of malingering

CDP = computerized dynamic posturography; ENG = electronystagmography; SCC = semicircular canal.

166

Interventions

cases of bilateral vestibular loss, resulting from, for example, systemic ototoxicity. Thus, it is not surprising that in patients with bilateral vestibular loss, caloric responses are usually symmetrically depressed. 24 Another limitation of caloric testing is that it only allows for low-frequency stimulation. Standard caloric stimulation, in fact, is equivalent to a very low-frequency rotation of 0.002 to 0.004 Hz.25 This frequency falls far below the normal physiologic range of the vestibular system. Although low-frequency VOR responses can be severely affected by acute and chronic ototoxicity, both the sensitivity and specificity of the caloric test for symmetric loss of vestibular function remain poor.

a period of about a year, VOR high-frequency gains recovered to within normal limits. In a study by O’Leary and colleagues, patients exposed to gentamicin ototoxicity were tested with active head movements to determine their high-frequency (2–6 Hz) VOR responses.31 VOR improvement correlated with patients’ reported reduction of symptoms. In more advanced systems the use of the magnetic scleral coil eye movement recordings provides the most accurate recordings, especially during high-frequency rotation. As demonstrated by Prepageran and colleagues, this method allows for documentation of symptomatic high-frequency vestibular loss undiagnosed by other modalities.32

ROTATIONAL TESTING

INTEGRATED VISUAL-VESTIBULARPROPRIOCEPTIVE ASSESSMENT

Rotational chair testing offers the opportunity to test the VOR with a range of rotational stimuli, using a physiologic stimulus (rotation) that is quantifiable. The chair is usually rotated at frequencies ranging from 0.01 to 1.2 Hz. The most advanced systems have the capability of delivering high-torque rotation at frequencies of 1.0 to 4.0 Hz to evaluate high-frequency function. Unlike caloric testing, the rotatory stimulus to the SCC affects both labyrinths simultaneously. Eye movement responses to acceleration are used to detect asymmetries in the vestibular system. The velocity and amplitude of the nystagmus are measured to calculate the gain and phase that are used as the main output measures. Gain is the ratio of the output (eye velocity) to the input (chair velocity). Phase describes the temporal (timing) relationship between the input and output. Significant reductions in gain are consistent with a bilateral peripheral vestibular loss and are useful in confirming and quantifying results obtained from caloric testing. Standard rotational test battery includes tests of optokinetic nystagmus (OKN), the VOR, and visual–vestibular interaction. Baloh and colleagues reported that lesions of the peripheral vestibular system characteristically impair only the VOR, whereas lesions of the central system impair OKN and visual–vestibular interaction.26 As reported by Minor, the caloric test identified the side of unilateral vestibular hypofunction more often (in 56%) than did rotational chair testing (in 20%).27 Caloric responses taken together with rotatory chair data provided information about the vestibulotoxic effects of treatment not obtainable from a single modality. In 1980 Tomlinson and colleagues reported their experience with the high-frequency stimulation rotation test.28 Use of this technique demonstrated better correlation with symptoms in comparison with caloric testing in patients with a bilateral peripheral loss.29 Black and colleagues, however, have reported larger decline in VOR function at lower stimulus frequencies compared with the high frequencies in subjects suffering from aminoglycoside-induced toxicity.30 Over

Observation of a patient’s ability to maintain balance has been used for more than a century to evaluate vestibular function. The Romberg test, in which the patient’s ability to stand with open eyes is compared with performance with eyes closed, is the classic example. A computer-controlled moving platform—computerized dynamic posturography (CDP)—provides quantitative evaluation of both sensory and motor components of postural control along with the brain interaction of sensory inputs. This test measures an individual’s balance while visual and somatosensory cues are intact or altered, creating conflicting multisensory inputs. Investigation of the relationship between VOR function and CDP performance by Crane and Demer in 1998 demonstrated that gaze velocity measurements during CDP are sensitive to vestibular loss.33 These results support the work of Hamid, who reported differences in VOR gain in patients with vestibular loss and vestibular hypofunction.34 In patients with a unilateral vestibular deficit, VOR gain was normal or hyperactive, whereas total vestibular loss was accompanied by minimal or absent VOR gain. Pospiech and colleagues reported the use of CDP for the evaluation of balance in persons exposed to ototoxic substances.35 They found a statistically significant increase in abnormal results in exposed workers than in a normal control group. Dimitri and colleagues recently included CDP in the multivariate vestibular testing for bilateral Meniere’s disease and aminoglycoside ototoxicity.36 They reported an increase in test sensitivity in identifying bilateral vestibular hypofunction when ENG, sinusoidal harmonic acceleration, and CDP were simultaneously used. In addition, CDP can be used to measure physical therapy outcomes in patients with bilateral vestibular loss. Since there is a statistically significant correlation between improvement in CDP composite scores and clinical changes, CDP evaluation may have an application in vestibular rehabilitation.36


Because information obtained from CDP is based on the vestibulospinal reflex and stimulation of SCC and otolith organs simultaneously, this test alone cannot delineate the site of lesion within the vestibular system. Use of CDP in vestibular monitoring of ototoxicity is therefore primarily limited to screening and for the monitoring of vestibular rehabilitation.

BEDSIDE CLINICAL VESTIBULAR TESTS Limitations in laboratory vestibular testing increase the practical role of bedside clinical tests for monitoring vestibular toxicity. Patients with a bilateral vestibular loss often complain of objects “jumping” or “jiggling” with head movements. This symptom of oscillopsia is caused by the bilateral loss of VOR function and subsequent inability to stabilize visual images on the retina. Excessive retinal slip not only causes oscillopsia but also impairs visual acuity.

DYNAMIC VISUAL ACUITY TESTING (OSCILLOPSIA TESTING) Evaluation of dynamic visual acuity (DVA) assesses the detection of poor VOR function. A particularly helpful development in the assessment of oscillopsia can be found in the works of Longridge and Mallinson.37,38 They proposed a dynamic illegible E (DIE) test for monitoring patients receiving aminoglycoside antibiotics. During the oscillopsia test, the head is passively oscillated at a frequency of about 1 to 2 Hz. The patient is asked to read the lowest possible line on a Snellen chart at rest and during passive head rotation. A drop in acuity of three lines or more is considered significant for a bilateral vestibular loss, as might occur from systemic ototoxicity.39 Computerized DVA testing was later developed and allows for more precise measurement by controlling for head velocity. As reported by Herdman and colleagues, a computerized DVA test improved the accuracy of the assessment of vestibular deficits.40 This test was reliable in distinguishing between normal subjects and patients with vestibular loss and was useful in determining unilateral versus bilateral vestibular hypofunction. It is important to take into account, however, the normal changes in VOR gain that may be present in patients who wear glasses. With myopic lenses, VOR gain is reduced, and with hyperopic lenses, VOR gain is increased. When using this test in patients with longstanding vestibular loss, the possibility of compensation should also be considered. Tian and colleagues reported the effectiveness of DVA in detection and lateralization of unilateral vestibulopathy.41 They found 90% sensitivity and 100% specificity of this technique at high frequency and acceleration (2,800°/s2) in detection of unilateral vestibular deafferentation.

167

Post–Head-Shake Nystagmus The discovery of nystagmus after repetitive head shaking is a well-recognized clinical sign that was first noted by Barany in 1907.42 Observation of post–headshake nystagmus (HSN) after vigorous horizontal head shake has been ascribed to the activation of a latent vestibular asymmetry. If the integrity of vestibular system is intact, the central mechanism of velocity storage system helps to sustain slow-phase velocity when the vestibular stimulus is of low frequency. In patients with unilateral lesions, horizontal post-HSN may be observed with fixation eliminated by Frenzel glasses. The slow phase is initially directed toward the impaired ear and the fast phase away from the side of the lesion. Hain reported that HSN was a sensitive indication of the existence and location of a unilateral vestibular lesion.43 Asawavichiangianda and colleagues found statistically significant correlation between the presence of HSN and peripheral vestibular dysfunction versus psychogenic dizziness. 44 The results of their study indicate, however, that HSN was neither specific nor sensitive for vestibular dysfunction. High-Frequency Horizontal Head Thrust Halmagyi and Curthoys in 1988 described a horizontal head thrust test for clinical diagnosis of unilateral vestibular loss.2 During this maneuver, the patient’s head is quickly turned while the patient is asked to focus on a midline target. In case of unilateral canal paresis, one large or several small oppositely directed, compensatory refixation saccades may be elicited by rapid horizontal head rotation toward the affected side. This finding is usually supported by the patient’s complaint of blurred vision during quick head movements toward the affected side. In studying the relationship of HSN and head thrust to caloric testing, Harvey and colleagues found low sensitivity for both HSN and head thrust for detecting peripheral vestibular disease.45 However, the positive predictive value for both tests when combined was 80%. Thus, when both tests are positive, there is an 80% chance of a true vestibular lesion. In addition, they found that both HSN and head thrust could accurately predict the side of the lesion in patients with vestibular pathology. Unlike the postHSN test, the high-frequency horizontal head thrust was able to provide evidence of bilateral vestibular impairment when present.

SUBJECTIVE EVALUATIONS Health care providers should always consider and take seriously vocalized complaints by the patient concerning feelings of dizziness and imbalance. Vague complaints are generally ignored until a vestibulotoxic event occurs. Health care providers often do not anticipate complaints such as oscillopsia (visual blurring of

168

Interventions

head movement), for example, in an individual receiving a potentially ototoxic agent systemically. The Dizziness Handicap Inventory (DHI) is a validated subjective measure that easily allows patients to evaluate ongoing well-being.46 The aforementioned DIE test is another example of a quasiobjective test that can easily be performed at the bedside and may provide early information concerning impending vestibulotoxicity.37,38 In the absence of a specific vestibular monitoring test that can provide real-time information concerning the vestibular system, vigilance must be constant. Reversibility for vestibulotoxicity depends on early recognition and cessation of the causative agent.

SUMMARY • No one protocol exists that can monitor, equally and safely, all variants of vestibulotoxicity. Adequate monitoring ideally would include a determination of baseline vestibular function before the start of a potentially ototoxic treatment and sequential testing during therapy on a regular basis (ie, weekly for aminoglycosides). • Possibility for the development of vestibulotoxicity following potentially ototoxic treatment demands a high index of clinical suspicion. Prevention of possible ototoxic effects depends on timely assessment during treatment that realistically involves the need for frequent clinical tests of vestibular function (ie, DVA testing, high-frequency head thrust). A patient’s subjective reports of changes in well-being should never be ignored. • If ototoxicity occurs, repeated testing may provide sufficient information regarding any spontaneous vestibular improvement and, if CDP is used, the efficacy of vestibular rehabilitation.

REFERENCES 1. Baloh R, Honrubia V. Clinical neurophysiology of the vestibular system. Contemporary neurology series. Philadelphia: FA Davis Co; 1979. p. 47–60. 2. Halmagyi GM, Curthoys IS. A clinical sign of canal paresis. Arch Neurol 1988;45:737–9. 3. Black FO, Pesznecker SC. Vestibular ototoxicity. Clinical consideration. Otolaryngol Clin North Am 1993;26:713–36. 4. Lindeman H. Regional differences in sensitivity of the vestibular sensory epithelia to ototoxic antibiotics. Acta Otolaryngol 1969;67:177. 5. Takumida M, Bagger-Sjoback D, Harada Y, et al. Sensory hair fusion and glycocalyx changes following gentamicin exposure in the guinea pig vestibular organs. Acta Otolaryngol 1989;107: 39–47.

6. Meza G, Lopez I, Paredes M, et al. Cellular target of streptomycin in the internal ear. Acta Otolaryngol (Stockh) 1989;107:406–11. 7. Norris CH. Application of streptomycin to the lateral semicircular canal. Trans Am Otol Soc 1987; 75:84–8. 8. Norris CH, Shea JJ. Selective chemical vestibulectomy. Am J Otol 1990;11:395–400. 9. Kimura RS, Iverson NA, Southard RE. Selective lesions of the vestibular labyrinth. Ann Otol Rhinol Laryngol 1988;97:577–84. 10. Bagger-Sjoback D, Bergenius J, Lundberg AM. Inner ear effects of topical gentamicin treatment in patients with Meniere’s disease. Am J Otolaryngol 1990;11:406–10. 11.` Tsuji K, Velaquez-Villasenor L, Rauch SD, et al. Temporal bone studies of the human peripheral vestibular system. Ann Otol Rhinol Laryngol Suppl 2000;181:20–5. 12. Gozdzik-Zolnierkiewicz T, Moszynski B. VIIIth nerve in experimental lead poisoning. Acta Otolaryngol 1969;68:85. 13. Hunter D, Russel DS. Focal cerebral and cerebellar atrophy in a human subject due to organic mercury compounds. J Neurol Neurosurg Psychiatr 1954;17:235. 14. Tran Ba Huy P, Manuel C, Meulemans A, et al. Pharmacokinetics of gentamicin in perilymph and endolymph of the rat as determined by radioimunoassay. J Infect Dis 1981;43:476–86. 15. Tran Ba Huy P, Deffrennes D. Influence of dosage regimen on drug uptake and correlation between membrane binding and some clinical features. Acta Otolaryngol 1988;105:511–5. 16. Prepageran N, Kisilevsky V, Rutka A. Systemic gentamicin ototoxicity: dosing regimes, risk factors and medico-legal concerns. J Otolaryngol 2004. [In press] 17. Dayal VS, Chait GE, Fenton SA. Gentamicin vestibulotoxicity. Long-term disability. Ann Otol Rhinol Laryngol 1979;88:36–9. 18. Kisilevsky V, Prepageran N, Rutka J. Intentional chemical ablation of vestibular function using commercially available gentamicin ear drops in Meniere’s disease versus inadvertent topical gentamicin ototoxicity: what’s the difference? J Otolaryngol 2004. [In press] 19. Inoue H, Uchi Y, Nogami K, Uemura T. Low-dose intratympanic gentamicin treatment of Meniere’s disease. Eur Arch Otorhinolaryngol 1994;251 Suppl 1:S12–4. 20. Kaplan DM, Hehar SS, Bance ML, Rutka JA. Intentional ablation of vestibular function using commercially available topical gentamicinbetamethasone eardrops in patients with Meniere’s


21.

22.

23. 24. 25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

disease: further evidence for topical eardrop ototoxicity. Laryngoscope 2002;112:689–95. Nedzelski JM, Schessel DA, Bruce GE, Pfleiderer AG. Chemical labyrinthectomy: local application of gentamicin for the treatment of unilateral Meniere’s disease. Am J Otol 1992;13:18–22. Goode CT, Carey JP, Fuchs AF, Rubel EW. Recovery of the vestibulocolic reflex after aminoglycoside ototoxicity in domestic chickens. J Neurophysiol 1999;81:1025–35. Coats AC, Smith SY. Body position and the intensity of caloric nystagmus. Acta Otolaryngol 1967;63:515. Fee WE. Aminoglycoside ototoxicity in the human. Laryngoscope 1980;10 Pt 2 Suppl 24:1–19. Furman JM, Wall CIII, Kamerer DB. Alternate and simultaneous binaural bithermal caloric testing: a comparison.Ann Otol Rhinol Laryngol 1988;97:359. Baloh RW, Sakala SM, Yee RD, et al. Quantitative vestibular testing. Otolaryngol Head Neck Surg 1984;92:145–50. Minor LB. Intratympanic gentamicin for control of vertigo in Meniere’s disease: vestibular signs that specify completion of therapy. Am J Otol 1999; 20:209–19. Tomlinson RD, Saunders GE, Schwarz DWF. Analysis of human vestibulo-ocular reflex during active head movements. Acta Otolaryngol 1980; 94:53–60. Larsby B, Hyden D, Odkvist LM, et al. Caloric and rotatory tests in patients with uni- and bilateral vestibular loss. Acta Otolaryngol Suppl 1984;412: 111–2. Black FO, Peterka RJ, Elardo SM. Vestibular reflex changes following aminoglycoside induced ototoxicity. Laryngoscope 1987;97:582–6. O’Leary DP, Davis LL, Li S. Predictive monitoring of high-frequency vestibulo-ocular reflex rehabilitation following gentamicin ototoxicity. Acta Otolaryngol Suppl 1995;520 Pt 1:202–4. Prepageran N, Kisilevsky V, Tomlinson RD, et al. Symptomatic high frequency vestibular loss: a newly recognized clinical syndrome? Acta Otolaryngol 2004;124:1–7. Crane BT, Demer JL. Gaze stabilization during dynamic posturography in normal and vestibulopathic humans. Exp Brain Res 1998;122:235–46. Hamid MA. Clinical patterns of dynamic posturography. In: Arenberg IK, editor. Dizziness and

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

169

balance disorders: an interdisciplinary approach to diagnosis, treatment and rehabilitation. Amsterdam/New York: Kugler Publishing; 1993. Pospiech L, Przerwa-Tetmajer E, Gawron W. The evaluation of the balance organ in workers occupationally exposed to organic solvents and tinctorial dusts. Med Pr 1998;49:363–9. Dimitri PS, Wall III C, Rauch SD. Multivariate vestibular testing: thresholds for bilateral Meniere’s disease and aminoglycoside ototoxicity. J Vestib Res 2002;11:391–404. Longridge NS, Mallinson AI. A discussion of the dynamic illegible “E” test: a new method of screening for aminoglycoside vestibulotoxicity. Otolaryngol Head Neck Surg 1984;92:671–7. Longridge NS, Mallinson AI. The dynamic illegible E (DIE) test: a simple technique for assessing the ability of the vestibulo-ocular reflex to overcome vestibular pathology. J Otolaryngol 1987;16: 97–103. Chambers BR, Mai M, Barber HO. Bilateral vestibular loss, oscillopsia and the cervico-ocular reflex. Otolaryngol Head Neck Surg 1985;93:403–7. Herdman SJ, Tusa RJ, Blatt P, et al. Computerized dynamic visual acuity test in the assessment of vestibular deficits. Am J Otol 1998;19:790–6. Tian JR, Shubayev I, Demer JL. Dynamic visual acuity during passive and self-generated transient head rotation in normal and unilaterally vestibulopathic humans. Exp Brain Res 2002;142:486–95. Barany R. Untersuchungen uber Verhalten des Vestibularaparates bei Kopftraumen und ihre practische Bedeuntung. Verhanduren der Deutschen Otol Gessellschaft 1907;252–66. Hain TC. Head-shaking nystagmus in patients with unilateral peripheral vestibular lesions. Am J Otolaryngol 1987;8:36–46. Asawavichiangianda S, Fujimoto M, Mai M, et al. Significance of head-shake nystagmus in the evaluation of the dizzy patient. Acta Otolaryngol Suppl 1999;540:27–33. Harvey SA, Wood DJ, Feroah TR. Relationship of the head impulse test and head-shake nystagmus in reference to caloric testing. Am J Otol 1997;18: 207–13. Jacobson GP, Newman CW. The development of the Dizziness Handicap Inventory. Arch Otolaryngol Head Neck Surg 1990;116:424–7.

CHAPTER 20

Ototoxic Damage to Hearing: Otoprotective Therapies Thomas R. Van De Water, PhD, and Leonard P. Rybak, MD, PhD

In response to ototoxic damage by either an aminoglycoside antibiotic or a chemotherapeutic agent, there are in general two ways by which damaged sensorineural cells of the cochlea can die. The damaged sensorineural cells of the auditory receptor can be eliminated by either necrosis or apoptosis or by a mixture of these two processes within the injured cochlea. These two cell death processes have distinct histologic and biochemical profiles that can be used to identify them. Necrosis generally occurs in areas of the organ of Corti that have sustained the greatest degree of damage. The histologic features of necrotic cell death include swelling of the mitochondria and the nucleus, dissolution of cellular organelles, and lysis of the affected cochlear sensory cell with degradation of its deoxyribonucleic acid (DNA). Because necrotic cell death occurs rapidly and only in those sensory cells that have sustained a high level of internal damage, it is extremely difficult to prevent necrosis through treatment of damaged cochlear sensory cells. The best approach to prevent cell loss from necrosis is to prevent the damage from occurring to the sensory cells of the cochlea. The second form of cell death that eliminates damaged cochlear sensory cells is apoptosis, also known as programmed cell death. However the term “programmed cell death” in general refers to the apoptosis that occurs naturally during normal development and is a normal physiologic process for reducing the number of cells within a maturing organ to a physiologically relevant number and also for sculpting the shape of body parts and organs (eg, transition of a limb paddle to a hand with distinct digits). Here, the term “apoptosis” refers to cell death that is unwanted and occurs in response to ototoxic damage to inner ear sensory cells. In the process of apoptosis of damaged cochlear sensory cells, a biochemical cascade of intracellular signaling activates cell death molecules that are present within normal cells in pro-forms that are inactive (eg, procaspase-3). During the process of apoptosis the cytoplasm condenses, both the ribosomes and the

mitochondria aggregate, the nuclear chromatin condenses and aggregates, and as the cell dies small cellular fragments termed apoptotic bodies form. One of the delineating characteristics of apoptotic cell death is the enzymatic cleavage of the affected cells’ DNA into 180 bp internucleosomal fragments, also termed DNA laddering. Other characteristics of apoptosis are generation of reactive oxygen species (ROS) and other free radicals, intracellular acidification, reduction of or a complete loss in the membrane potential of the cell’s mitochondria, activation of caspases (eg, caspase-3), and externalization of phosphatidyl serine residues. Apoptosis can occur over a series of days after the initial insult and is often the result of cumulative damage occurring within a cochlear sensory cell owing to ongoing generation of ROS and other free radicals, release of cytochrome-c from damaged mitochondria, and activation of intracellular cell death molecules (eg, caspases) within the damaged cell (Figure 20-1).1,2 Unlike necrotic cell death of cochlear sensory cells, the apoptosis of injured cochlear sensory cells can be lessened or in some cases completely prevented by interventional otoprotective therapies that target the process of apoptosis within the affected cochlear sensory cells. These interventional therapies are known collectively as otoprotection and have been applied as part of treatments aimed at lessening ototoxic damage to the cochlear neurosensory cells and to hearing from both aminoglycosides and chemotherapeutic agents.

AMINOGLYCOSIDE ANTIBIOTICS: OTOPROTECTION Aminoglycosides are highly effective antimicrobial agents whose efficacy and use in the treatment of lifethreatening aerobic gram-negative bacterial infections have been severely limited because of both nephrotoxic and ototoxic side effects at high dosages or when large cumulative dose levels are achieved after extended usage. Aminoglycosides cause a loss of hearing acuity through the generation of ROS and other free radicals

Ototoxic Damage to Hearing: Otoprotective Therapies

OTOTOXIN

Extrinsic Cell Death Receptor Pathway

171

Intrinsic Mitochondrial Cell Death Pathway

Injury Oxidative stress

TNF family receptors (eg, Fas)

Stress kinase pathways (eg, c-jun)

Procaspase-8 Active caspase-8 (high concentration)

Active caspase-8 (low concentration) BID Apoptosome Smac/Diablo

dATP APAF-1

BIM

BCL2

MAPK/JNK Mitochondrion

BCIXL

Cyto C

Procaspase-9

ROS HNE

Cyto C

BAX

Active caspase-8 Activation of downstream caspases (eg, caspase-3)

IAPs

Apoptotic substrates (eg, degradation of proteins and DNA)

Apoptosis Figure 20-1 This flow chart depicts both the extrinsic cell death receptor pathway and the intrinsic mitochondrial cell death pathway that can be activated by exposure to an ototoxin. The extrinsic pathway involves the activation of receptors of a member of the tumor necrosis factor (TNF) family such as activation of Fas receptors with a Fas ligand. This brings attached procaspase-8 molecules into close association and through autocatalysis results in their conversion into active caspase-8 molecules. At high concentration caspase-8 can act directly on downstream effector procaspase molecules and activate these effector caspases (eg, caspase-3) that affect the apoptosis of the injured sensory cell. Lower levels of activated caspase-8 can cross over into the intrinsic pathway via the activation of the proapoptotic member of the Bcl-2 family BID. The intrinsic pathway involves oxidative stress and the formation of reactive oxygen species (ROS) causing lipid peroxidation damage to cellular membranes with the formation of a natural toxin, ie, 4-hydroxyl-2,3-nonenal (HNE), that activates the mitogen-activated protein kinase (MAPK)/c-jun-N-terminal kinase (JNK) cell death signal cascade, resulting in damage to the mitochondrial membranes and formation of pores in the outer mitochondrial membrane by a proapoptotic member of the Bcl-2 family (ie, BAX). Two antiapoptotic members of the Bcl-2 family, ie, BCL2 and BCL XL , both try to stabilize this membrane and inhibit pore formation by BAX. BIM, a proapoptosis member of the Bcl-2 family, acts to block the antiapoptotic effects of BCL 2 and BCL XL. Pore formation releases cytochrome c (Cyto C) from the inner compartment of the damaged mitochondrion, and once Cyto C is within the cytoplasm of the injured sensory cell it forms complexes with procaspase-9 and APAF-1 in the presence of dATP to form the apoptosome, which converts procaspase-9 into active caspase-9. Activated caspase-9 activates downstream effector procaspase molecules, and these molecules (eg, caspase-3) cleave apoptotic substrates within the cell, affecting its death via apoptosis. Both activator caspases and effector caspases can be inhibited by naturally occurring inhibitor of apoptosis molecules (IAPs), but these molecules themselves are the targets of activated effector caspase molecules.

that directly damage the auditory hair cells.3–9 Because aminoglycosides remain an important class of antibiotics that are still used to treat difficult life-threatening infections and to treat such diseases as cystic fibrosis (inhalation therapy; 16% of treated patients develop a bilateral sensorineural hearing loss), 10 many compounds have been tried as otoprotectant molecules to prevent the ototoxic side effects of these antibiotics. This chapter discusses many of the otoprotective

molecules used to protect hearing and to prevent the loss of auditory sensory cells via aminoglycosideinduced apoptosis of these cells.

SPIN-TRAPPING AGENTS Alpha-phenyl-tert-butyl-nitrone It is well known that external ear canal antibiotic drops that contain aminoglycoside antibiotics have the

172

Interventions

potential to cause a hearing loss if they enter the middle ear cavity through an opening in the tympanic membrane.11–16 Alpha-phenyl-tert-butyl-nitrone (PBN) is a spintrap molecule that can effectively trap and inactivate ROS and other free radicals. PBN given systemically and when applied to the round window membrane (RWM) prior to application of an aminoglycoside otic drop medication was able to prevent the ototoxic effect of this preparation as determined by a loss of sensitivity of the cochlear action potential in the highfrequency range.17 This is the only report on the use of a spin-trap molecule for this application, and to date there have been no further studies to determine if this approach could be used in a clinically relevant application. This report provides evidence that otic drop preparations containing aminoglycoside antibiotics (eg, neomycin) generate ROS and other free radicals when applied to the RWM and that this oxidative stress plays a role in the ototoxic side effect of these otic drop preparations.

ANTIOXIDANTS The primary function of antioxidant molecules is either to prevent the formation of or to detoxify ROS and other free radical molecules formed in affected cells during oxidative stress. By limiting a cell’s exposure to ROS and other free radicals, the internal damage within a cell is also limited; the level of internal damage determines whether a cell will self-repair or commit to a program of suicide (ie, apoptosis). Glutathione Glutathione (GSH) is an endogenous thiol-containing amino acid that is part of every cell’s natural defense against excessive oxidative stress. GSH can detoxify ROS and other toxic molecules (eg, 4-hydroxy-2,3nonenal; HNE) created within a cell under oxidative stress. GSH molecules are present within mitochondria and are critical to the cell’s ability to neutralize the effect of oxidative stress and therefore critical to a hair cell’s viability when subjected to the oxidative stress created by exposure to an aminoglycoside antibiotic. Two articles that initially showed a relationship between nutritional status of laboratory animals, their GSH levels, and increased susceptibility of the cochlea to ototoxic damage from exposure to aminoglycoside antibiotics created an interest in GSH as an otoprotectant molecule.18,19 Studies with isolated outer hair cells in vitro demonstrated that GSH attenuated the ototoxic effect of gentamicin metabolites on these sensory cells. 20,21 A study that looked at the relationship between diet and GSH levels within cochlear tissues clarified the issue of the protective effect of GSH against aminoglycoside-induced hearing loss.22 This study showed that supplementing the diet of amino-

glycoside-exposed laboratory animals with the monoethyl ester of GSH (GSHe) was effective only in protecting against hearing loss if the animals were nutritionally deprived (ie, low-protein diet). Similarly, if the animals were on a normal protein diet, and therefore had normal levels of GSH within their cochlear tissues, there was no otoprotective effect achieved by GSHe supplementation of their diet. GSH levels within the outer hair cells (OHCs) of the cochlea are distributed in a base-to-apex pattern that mirrors the susceptibility of the OHCs to the toxic effects of aminoglycosides, with the lowest levels of GSH found in the OHCs in the base of the cochlea and the highest levels in the most apical OHCs.23 This recent observation supports the importance of the natural otoprotective ability of GSH within the cochlea. The higher the level of GSH within an OHC, the greater its ability to resist the ototoxic effect of an aminoglycoside antibiotic because there is a base-to-apex gradient in OHC susceptibility to aminoglycosides. For GSH to function as an otoprotectant molecule it must be intracellular; it is not cell permeable. The GSH molecule must be esterified (ie, GSHe) in order for it to pass through the cell membrane to act as an otoprotectant molecule. Local application of GSHe to the cochlea via the RWM may be an effective approach for using this otoprotective molecule against aminoglycoside ototoxicity, but as yet there are no animal studies with GSHe. Methionine Methionine (Met) is a thiol containing naturally occurring essential amino acid that possesses both antioxidant and metal-chelating properties. It has been shown to prevent gentamicin-mediated formation of free radicals in a cell-free in vitro test, in cell cultures, and when injected systemically to significantly attenuate the amount of gentamicin-induced free radical formation and hearing loss.24 Protection of hearing by systemic administration of D-Met was partial but did extend to all frequencies tested (ie, 3–18 kHz), and therefore the potential of methionine as an otoprotectant therapy against aminoglycoside ototoxicity may warrant additional investigation. Whether a high level of systemically administered D-Met would interfere with the antimicrobial efficacy of the aminoglycosides has yet to be determined. Iron Chelators: Deferoxamine and 2,3Dihydroxybenzoate Two well-characterized iron chelators, deferoxamine (DFO) and 2,3-dihydroxybenzoate (DHB), have been demonstrated to be highly effective in the prevention of gentamicin-induced hearing loss in a guinea pig model of ototoxicity.25 This study also showed that treating the guinea pigs with these iron chelators did not diminish the effective plasma levels of available gentamicin within


the treated animals. A later study using electron paramagnetic resonance spectroscopy and a spin-trapping agent (5,5-dimethyl-pyrroline-N-oxide; DMPO) demonstrated in cochlear tissue isolates that treatment of these isolates with DFO prevented the formation of ROS and other free radicals normally generated by exposure of this cochlear tissue to ototoxic levels of gentamicin and kanamycin.4 A series of studies that followed have shown that iron chelators can be highly effective otoprotective agents that substantially diminish the ototoxic side effects of aminoglycoside treatment on both hair cell integrity and the oxidative stress-induced hearing loss that results from aminoglycoside treatment. The use of these iron chelators had no effect on either the levels or antimicrobial efficacy of aminoglycoside antibiotics.26–30 The main drawback of using iron chelators as otoprotectant molecules against the ototoxic effects of aminoglycoside antibiotics is their toxic side effects when used at levels high enough to be otoprotective. Salicylate Salicylates are among the most commonly used drugs, especially in the form of aspirin (ie, acetylsalicylate) that is rapidly converted to salicylate in serum within 15 to 30 minutes of ingestion. Salicylate (2-hydroxybenzoate) is a potent antioxidant and free radical scavenger that when interacting with free radicals (ie, hydroxyl radicals) forms the chelator molecule 2,3-dihydroxybenzoic acid, which converts to the highly effective iron chelator DHB. A study of the otoprotective capability of salicylate when coadministered with gentamicin in guinea pigs demonstrated that this treatment provided both a high level of protection against drug-induced hearing loss and protection of hair cell integrity.31 Salicylate treatment did not interfere with either the serum levels of the administered aminoglycoside or with its antimicrobial efficacy and the levels required for protection of hearing corresponded to levels presently used in patients receiving aspirin therapy for arthritis. The results of a clinical trial performed in China are being evaluated for aspirin otoprotection against aminoglycoside ototoxicity (for a more in-depth discussion of these results see Chapter 9, “Mechanisms for Aminoglycoside Toxicity: Basic Science Research”). Tanshinone Tanshinone is an extract of Salviae miltiorrhizae that contains diterpene quinones and phenolic acids that possess potent antioxidant properties. Produced in China, tanshinone is a traditional herbal medicine available over the counter and known by the name Danshen. The results of an initial study of the otoprotective properties of tanshinone both in vitro and in vivo against aminoglycoside ototoxicity are very promising, with suppression of the formation of ROS and other free radicals by gentamicin in an in vitro

173

system and protection against both hearing loss and loss of auditory hair cells following kanamycin exposure in vivo.32 Tanshinone was also demonstrated not to interfere with the antimicrobial action of the aminoglycoside antibiotic kanamycin. These findings are very encouraging because this is a widely used herbal medicine with strong antioxidant and otoprotective properties that does not interfere with the antimicrobial efficacy of the aminoglycoside antibiotic tested. Further studies are warranted. Superoxide Dismutase Superoxide dismutase (SOD) is a naturally occurring antioxidant defense molecule that protects cells from an excess of superoxide produced during oxidative stress by dismutating superoxide into hydrogen peroxide molecules that can be inactivated through the action of the antioxidant molecule catalase. SOD has a molecular weight greater than 30 kDa; because of its large size it is difficult to use in direct application as an otoprotective agent against the ototoxic effects of aminoglycosides. An additional consideration is that the half-life of SOD in the body is very short, approximately 6 minutes. Evidence for its otoprotective capability comes from both transgenic animals that were genetically engineered to overexpress Cu/Zn SOD (SOD1) and from gene therapy experiments where a vector was used to overexpress either SOD1 or Mn SOD (SOD2). Overexpression of SOD1 in a transgenic mouse conveyed protection against kanamycin-induced hearing loss in the overexpressing animals.33 Adenoviral vector overexpression of SOD2 was more effective as an otoprotective therapy (ie, protecting both hair cells and hearing) against aminoglycoside ototoxicity than was the adenoviral vector overexpression of SOD 1.34 These gene therapy results are encouraging; however, many questions remain to be addressed before gene therapy can be considered for application as an inner ear therapy in the clinic. An in vitro study with organ of Corti explants from neonatal mice has demonstrated that an SOD mimetic, M40403 (a manganese-based nonpeptidyl molecule), can provide some protection of the auditory hair cells in the explants from the ototoxic effect of gentamicin.35 These results with the SOD mimetic were encouraging, but this approach needs to be tested in an animal model to determine efficacy in vivo and if this type of therapy would have any unwanted side effects.

INHIBITORS OF CELL DEATH PATHWAYS The mitogen-activated protein kinase (MAPK) cell death signaling pathway was first implicated in aminoglycoside-induced auditory hair cell loss in an in vitro study that reported the use of an inhibitor of this pathway to protect the hair cell in aminoglycosidechallenged organ of Corti explants.

174

Interventions

CEP 1347, A MAPK Pathway Inhibitor CEP 1347 is a derivative of the indolocarbazole K252a and blocks the MAPK signal pathway at the level of the mixed lineage kinases (eg, MLK-3). The in vitro results of an initial study with CEP 1347 have shown that this MAPK pathway inhibitor can protect auditory hair cells within neomycin-challenged neonatal rat organ of Corti explants from the toxic effect of this aminoglycoside antibiotic.36 The results of a second experimental series studying in vivo aminoglycoside-induced hearing and hair cell losses has shown that CEP 1347 when administered systemically can partially protect both hair cells and hearing from the ototoxic effect of gentamicin.37 This in vivo study also demonstrated the protection of vestibular hair cells against the vestibulotoxic effect of gentamicin. This type of therapeutic approach is encouraging; however, no data are available to determine the affect of CEP 1347 administration on the antimicrobial action of gentamicin and any possible deleterious affects of long-term systemic administration of a MAPK pathway inhibitor (ie, CEP 1347). c-Jun N-Terminal Kinase Inhibitory Peptide c-Jun N-terminal kinase inhibitory peptide (D-JNKI-1) is a chemically synthesized cell-permeable JNK ligand that blocks JNK-mediated activation of its target molecules, that is, c-Jun. D-JNKI-1 is an efficient inhibitor of the action of all three JNK isoforms and is made by linking the 20 amino acid terminal JNK-inhibitory sequence (ie, the JNK-binding domain of JIP-1/IB1) to a 10–amino acid HIV-TAT transporter sequence. The results of in vitro and in vivo studies demonstrate that local application of the D-JNKI-1 peptide can protect both auditory hair cells and hearing from the ototoxic effects of neomycin.38 This study confirms the original observation that the MAPK–JNK cell death signal pathway is involved in the apoptosis of oxidative stressinjured auditory hair cells and that the blocking of the MAPK cell death signal pathway is an effective way to prevent aminoglycoside-induced hair cell and hearing losses. At present it is not known if treatment of the inner ear with D -JNKI-1 peptide would effect the antimicrobial efficacy of an aminoglycoside antibiotic, but since the D-JNKI-1 treatment was via local application it is highly unlikely that it would have a systemic effect on the actions of an aminoglycoside. Caspase Inhibitors Caspases are a family of cysteine proteases that are present within the cells of normal healthy tissue in inactive (procaspase) forms. The procaspase configuration of a caspase is quiescent, and in order for a caspase to become active it requires cleavage of its prodomain. After the prodomain has been cleaved from the procaspase molecule, the caspase molecule is activated and can

now act upon and cleave a specific tetrapeptide sequence (this sequence varies for each member of the caspase family) in targeted proteins, which can include both the cytoskeletal and nuclear proteins of an oxidative stress-damaged auditory hair cell. Many members of the caspase family of proteases (eg, caspases-3, -8, and -9) have been proven to participate in the regulation and execution of apoptosis of oxidative stress-damaged hair cells, which can be the result of a cell’s response to a high level of internal injury (eg, aminoglycosideinduced damage). Caspases were found to participate in the apoptotic cell death of gentamicin-damaged vestibular hair cells. Explants of utricular maculae excised from both adult guinea pigs and adult gerbils were exposed to ototoxic levels of gentamicin and protected by addition of a broad-spectrum pancaspase inhibitor (either z-VAD-fmk or BAF) to the culture medium.39 The explants with the pancaspase inhibitor treatment showed a significant level of protection against gentamicin-initiated loss of vestibular hair cells and a significant reduction in the number of apoptotic hair cell nuclei. This finding of pancaspase protection of vestibular hair cells from aminoglycoside-initiated apoptosis was confirmed in another in vitro study that used post-hatch white leghorn chickens for a source of utricular maculae explants and neomycin as the ototoxic drug.40 The pancaspase inhibitors added to the culture medium were the same used in the previous study (z-VAD-fmk and BAF), and both of these inhibitors were highly effective in preventing aminoglycoside-induced death of hair cells. The first proof that specific members of the caspase family of cysteine proteases are involved in the apoptosis of both auditory and vestibular hair cells after an ototoxic insult have been reported in two recent in vitro studies, avian basilar papilla explants and mouse utricular explants. The study of chick basilar papilla explants exposed to an ototoxic level of gentamicin used fluorescent-labeled peptide substrates for caspases-3, -8, and -9 to detect their activation within the ototoxin-exposed basilar papilla explants.41 The results of this in vitro study show that gentamicin-damaged auditory hair cells degenerate and undergo apoptosis in a caspase-dependent manner and that the initiator caspases-8 and -9 and a downstream effector caspase, caspase-3, are activated in aminoglycoside-damaged auditory hair cells. This study used the general caspase inhibitor z-VAD-fmk to prevent caspase activation, so no conclusions could be drawn about the action or efficacy of specific caspase inhibitors in this ototoxin damage model. However, the adult mouse utricular explant–neomycin study reported the use of a combination of the following: (1) in situ substrate detection for specific caspases; (2) immunolabeling of activated caspases; and (3) caspase inhibitors with known specificity.42 Initially, a


pancaspase inhibitor was used to confirm that caspases were involved in the apoptosis of these ototoxindamaged vestibular hair cells. The results of this study confirm the activity of casapses-8, -9, and -3 in the apoptosis of ototoxin-damaged vestibular hair cells; additionally, this study shows that caspase-8 plays only a minor role in the process of apoptosis of aminoglycoside-damaged hair cells, whereas caspases-9 and -3 were found to have major roles in the apoptosis of these damaged hair cells. When specific caspase inhibitors were used to prevent hair cell death, the caspase-8 inhibitor (z-IETD-fmk) had no significant effect on either preventing neomycin-induced hair cell death or the downstream activation of procaspase-3. In contrast, the caspase-9–specific inhibitor (z-LEHD-fmk) prevented both neomycin-induced hair cell death and the downstream activation of procaspase-3 in these ototoxinexposed utricular explants. Currently, there are no reported studies to determine whether downstream effector caspases-6 and -7 participate in the process of apoptotic cell death of hair cells that eliminates these aminoglycoside-damaged sensory cells. All of the caspase inhibitor studies discussed in this section were performed in vitro, so it is not currently known whether these irreversible inhibitors of caspases (eg, z-VAD-fmk, a pancaspase inhibitor) will be effective when delivered in vivo (eg, perfused into the scala tympani). Caspases-9 and -3 appear to be essential components in the apoptotic cell death of aminoglycosidedamaged inner ear hair cells, whereas caspase-8 has been shown to play only a minor role in this cell death process. These in vitro results are encouraging; however, there are as yet no animal studies demonstrating that caspase inhibitors can work in the intact animal to protect against aminoglycoside-induced hair cell loss and the loss of hearing. The action of caspase inhibitors on the antimicrobial action of aminoglycosides also needs to be defined.

GROWTH FACTORS Glial Cell Line–Derived Neurotrophic Factor Glial cell line–derived neurotrophic factor (GDNF) is a member of the transforming growth factor-beta (TGF-β) family. In both in vitro and in vivo studies GDNF has been shown to partially protect hair cells for aminoglycoside ototoxicity.43 It has also been shown to have a neuroprotective effect on oxidative stressed VIIIth nerve ganglion neurons in vitro.43 In addition, the use of an adenoviral vector to overexpress GDNF in the inner ear has been shown to be as effective as GDNF protein in partially protecting the auditory hair cells, hearing, and vestibular hair cells from the ototoxic effects of aminoglycosides (eg, gentamicin) in vivo.44,45 These results with GDNF are promising because this

175

growth factor appears to offer a partial level of otoprotection to both the hair cells and the neurons of the cochlea. Studies are needed to determine that GDNF does not interfere with the antimicrobial action of the aminoglycosides and to develop a safe and effective route for the delivery of GDNF. Direct infusion into the scala tympani and delivery via a gene therapy vector are at present not approved methods of delivery of otoprotective agents into the cochlea of a patient. Members of the Neurotrophin Family: BrainDerived Neurotrophic Factor, Neurotrophin Type 3, and Neurotrophin Type 4/5 The auditory neurons of the spiral ganglion depend on the auditory hair cells that they innervated for their health and survival because the hair cells supply trophic support for these neurons. The results of gene nullmutation experiments have demonstrated that both brain-derived neurotrophic factor (BDNF) and neurotrophin type-3 (NT-3) are important neurotrophins for the maturation and survival of VIIIth nerve neurons, with BDNF suggested to be most important for the neurons of Scarpa’s ganglion and NT-3 most important for the spiral ganglion neurons.46 However, in vivo neurotrophin infusion experiments have determined that either of these neurotrophins (ie, BDNF or NT-3) is adequate to support the survival of spiral ganglion neurons following an aminoglycoside-induced loss of auditory hair cells.47,48 There is some indication from the results of in vitro studies that BDNF, NT-3, and NT-4/5 can have a neuroprotective capability when VIIIth nerve neurons are subjected to neurotoxic molecules.49–51 An in vivo study combined NT-3 with an Nmethyl-D-aspartate (NMDA) antagonist (ie, MK 801) to prevent hearing loss from exposure to amikacin, and this approach provided partial protection to the hearing receptor.52 The main application of BDNF and NT-3 therapy appears to be protecting and supporting the health and survival of the auditory neurons. The action of neurotrophins on the antimicrobial action of aminoglycoside antibiotics is currently unknown and needs to be determined. Transforming Growth Factor-Alpha Transforming growth factor-alpha (TGF-α) is a member of the epithelial growth factor (EGF) family and has been suggested to be involved in the repair of damaged hair cells. An in vitro study using both hair cell counts and real-time confocal Ca ++ imaging have demonstrated that TGF-α can protect auditory hair cells from the ototoxic effect of neomycin.53 This observation, although robust, was made in isolated organ of Corti explants. Experiments have never been attempted in vivo, and there is no information of the possible effect

176

Interventions

of TGF-α treatment on the antimicrobial action of any of the aminoglycoside antibiotics.

AMINOGLYCOSIDE OTOPROTECTION RESEARCH TO DATE: SOME CONCLUSIONS Several otoprotective agents have been shown to be effective against aminoglycoside ototoxicity. Some of these agents have been demonstrated not to interfere with the antimicrobial actions of the aminoglycoside antibiotics, but many of these otoprotective agents have not been evaluated for possible interference by these compounds. Novel routes of drug administration will be required for some of the otoprotective compounds to be effective. At present only two routes of administration are currently available and approved. These are direct injection through the tympanic membrane into the middle ear cavity and delivery to the RWM via a Silverstein MicroWick; the middle ear–RWM catheter produced by Durect Corporation and approved by the US Food and Drug Administration (FDA) is no longer commercially available. 54–56 Clinical trials are in progress in China, testing the efficacy of aspirin against aminoglycoside ototoxicity; the initial results are encouraging (see Chapter 9, “Mechanisms for Aminoglycoside Toxicity: Basic Science Research”), but more trials will be needed to test other promising agents such as D-Met and D-JNKI-1 peptide to determine the dose, timing, and route of administration. The herbal medicine approach to otoprotection needs additional exploration as suggested by the very encouraging results of an initial animal study with the Chinese herbal antioxidant compound tanshinone.32

CHEMOTHERAPEUTIC AGENTS: CISPLATIN OTOPROTECTION Cisplatin is a highly effective chemotherapeutic agent used to treat a variety of soft tissue neoplasms. To achieve therapeutic cures, the doses of cisplatin have been escalated. Unfortunately, ototoxicity, nephrotoxicity, and neurotoxicity can occur. In some series, every patient treated with cisplatin had hearing loss (see Chapter 6, “Ototoxicity of Platinum Compounds”).57 Thiol Compounds Several potentially protective drugs have been tested against cisplatin ototoxicity because of their efficacy in preventing nephrotoxicity. Several agents tested initially in animals contain thiol groups. These include sodium thiosulfate, D-Met, L-Met, diethyldithiocarbamate (DDTC), methylthiobenzoic acid, lipoic acid, L-N acetylcysteine, and amifostine. Sodium thiosulfate has been shown to protect against cisplatin ototoxicity in guinea pigs and in hamsters.58,59 Unfortunately, this drug neutralized the antitumor effect of cisplatin. To avoid this problem, a study

was conducted using intracochlear administration of sodium thiosulfate. Perfusion of sodium thiosulfate into the cochleae of guinea pigs completely prevented cisplatin-induced hearing loss as revealed by no change in compound action potential threshold and by no change in distortion-product otoacoustic emission audiograms. Cochlear hair cells and the stria vascularis were well preserved in the animals receiving sodium thiosulfate protection.60 On the other hand, chronic RWM application of sodium thiosulfate using an osmotic minipump implanted in guinea pigs treated with systemic cisplatin provided no protection against ototoxicity.61 Intracochlear perfusion of thiourea also provided partial protection against cisplatin ototoxicity in guinea pigs. Ears treated with thiourea had significantly fewer outer hair cells lost, although the brainstem auditory evoked potential threshold did not differ from that of animals treated with cisplatin alone.62 Thus, thiourea appeared to be less effective as a protective agent against cisplatin ototoxicity. Another sulfur-containing drug, mesna (sodium 2-mercaptoethane sulfonate), also interferes with the antitumor activity of cisplatin. The free thiol group of mesna reacts with cisplatin in the circulation, forming an inactive complex. Therefore, this drug has not been pursued as a chemoprotector against cisplatin in clinical trials.63 DDTC has been shown to effectively prevent cisplatin ototoxicity in rats. Animals pretreated with DDTC had significantly smaller elevations of auditory brainstem response thresholds compared with animals receiving cisplatin alone.64 It was also effective in preventing hair cell destruction in hamsters treated with cisplatin.59 However, patients given DDTC as a protective agent against cisplatin toxicity experienced side effects. These unpleasant side effects included numbness in the arm into which it was being infused, diaphoresis, chest discomfort, flushing, agitation, and elevation of systolic blood pressure. The combination of DDTC and cisplatin treatment in melanoma B16-bearing mice was compared with the antitumor efficacy of cisplatin alone. The efficacy of cisplatin against this tumor was reduced by half in mice administered DDTC in combination with cisplatin.65 Amifostine protects against cisplatin nephrotoxicity. However, it was not found to be effective against cisplatin ototoxicity in hamsters.59 It is metabolized to an active metabolite that is selectively taken by normal tissues. However, because amifostine has unpleasant side effects, including transient hypotension, nausea, and vomiting, it is not an ideal protective agent. Experimental studies in rats have shown that 4methylthiobenzoic acid (MTBA) is an excellent protective agent against cisplatin ototoxicity. In vitro studies


demonstrated that adding MTBA to organotypic cultures of the organ of Corti prevented hair cell loss attributed to cisplatin exposure.66 Pretreatment of rats with this agent prior to cisplatin prevented elevation of auditory brainstem response thresholds and loss of outer hair cells in the cochlea.67 An added benefit associated with this protective agent was prevention of weight loss and protection against cisplatin-induced nephrotoxicity. 68 Boogaard and colleagues demonstrated that MTBA did not interfere with the antineoplastic efficacy of cisplatin.69 To our knowledge, no clinical trials of MTBA protection against cisplatin ototoxicity have yet been performed. D-Met was found to provide excellent protection against cisplatin-induced hearing loss (elevation of brainstem auditory evoked potential threshold) and loss of outer hair cells in rats.70 It was also found to prevent damage to the stria vascularis.71 D-Met provided cytoprotection against cisplatin toxicity without compromising antitumor activity in an animal model of ovarian cancer.72 Rats bearing an aggressive form of breast cancer were protected against both cisplatin ototoxicity and nephrotoxicity. It was suggested that both L- and D-Met reduced the ability of cisplatin to kill an aggressive form of breast cancer in vitro and in vivo.73 However, when L-Met was delivered to the RWM of systemically treated rats bearing this aggressive form of breast cancer, there was excellent protection against ototoxicity without compromising the chemotherapeutic efficacy of cisplatin.74 Similarly, application of D-Met to the RWM prior to administering cisplatin to the RWM in chinchillas provided excellent protection against hair cell loss and prevented elevation of auditory brainstem response thresholds.75 Chronic administration of D-Met to the RWM of guinea pigs using an implanted osmotic minipump in animals treated with systemic cisplatin prevented changes in distortionproduct otoacoustic emissions on days 3 and 4 of treatment. Unfortunately, no differences in morphology could be seen on electron microscopy of the cochlea of guinea pigs treated with D-Met versus saline on the RWM in combination with systemic cisplatin.61 The intratympanic administration of D-Met is a potentially useful route of providing this protective agent because it would not be likely to interfere with the antitumor efficacy of cisplatin. A study of cisplatin pharmacokinetics suggested that cisplatin concentrations in blood might be reduced by the administration of D-Met.76 However, extremely large doses of D-Met were employed in that study, so the significance of these findings is unclear. Alpha-lipoic acid is an endogenous antioxidant with free radical scavenging properties that can act as a chelator for heavy metals and as a potent therapeutic agent against oxidative tissue injury. Pretreatment of rats with lipoic acid prior to cisplatin revealed dose-

177

dependent protection against cisplatin ototoxicity. Elevation of auditory brainstem response thresholds was reduced, and the antioxidant system in cochlear tissues (glutathione and antioxidant enzymes) was preserved in protected animals.77 Ebselen acts as a mimic for glutathione peroxidase and as a scavenger for peroxynitrite radicals. Rats pretreated with Ebselen prior to cisplatin had nearly complete protection against hearing loss, glutathione depletion, and lipid peroxidation in the cochlea.77 Phosphonic Acid Antibiotics Fosfomycin is a phosphonic acid derivative that has antibiotic activity. It was initially reported to protect against cisplatin ototoxicity in guinea pigs.78 However, subsequent studies have shown no protection against cisplatin effects on the inner ear of experimental animals.59 Antioxidants Sodium salicylate has been found to protect rats against cisplatin ototoxicity and nephrotoxicity without compromising the antitumor action in rats bearing a highly metastatic form of breast cancer. The loss of outer hair cells was prevented, as was the elevation of brainstem auditory evoked potential thresholds in cisplatin-treated rats pretreated with sodium salicylate. The chemotherapeutic efficacy of cisplatin on suppression of tumor mass and cancer cell metastasis was unaffected by salicylate. The mechanism of salicylate action is unclear. It may act as a scavenger of free radicals generated by cisplatin, and it may prevent the activation of the transcription factor nuclear factor kappa B (NF-κB).79 Salicylate could also be acting as an iron chelator. Vitamin E (α-tocopherol) is a slow-acting free radical scavenger that has been shown to prevent cisplatin nephrotoxicity and endothelial cell damage. In both guinea pigs and rats, cisplatin ototoxicity was reduced by pretreatment with vitamin E.80,81 Hair cell loss was prevented, and auditory function was preserved. Local administration of a water-soluble form of vitamin E (Trolox) prevented the ototoxicity of cisplatin applied to the RWM of guinea pigs.82 L-N-Acetylcysteine (L-NAC) is an antioxidant used clinically as an antidote against acetaminophen poisoning and as a mucolytic agent to clear the airways. In vitro experiments with organ of Corti organotypic cultures have shown that this protective agent protects the outer hair cells and spiral ganglion cells against cisplatin damage in a dose-dependent manner. This drug may directly scavenge free radicals and is a precursor for the natural antioxidant glutathione.83 The antitumor efficacy of cisplatin was reduced by half in melanoma B16-bearing mice when L-NAC was administered in combination with cisplatin.65 This raises concerns that this protective agent could diminish the

178

Interventions

antineoplastic efficacy of cisplatin in patients. Glutathione ethyl ester, but not glutathione, was found to reduce the auditory brainstem response threshold shifts and outer hair cell loss observed in the rat after cisplatin. However, this protection was only partial.84 Aminoguanidine is an inhibitor of inducible nitric oxide synthase. It also is an antioxidant that can react with and scavenge hydroxyl radicals. Pretreatment of rats with aminoguanidine reduced the ototoxicity of cisplatin. It significantly reduced the production of malondialdehyde, a marker for lipid peroxidation, in the cochlear tissues of rats receiving cisplatin. It reduced the elevation of auditory brainstem responses compared with those recorded in rats treated with cisplatin alone, but it did not reduce the amount of nitric oxide produced. Therefore, it was concluded that aminoguanidine may act more as a free radical scavenger than as an inhibitor of nitric oxide synthesis in the cochlea exposed to cisplatin.85 Peptides The peptides α-melanocyte–stimulating hormone (αMSH) and the nonmelanotropic adrenocorticotropic hormone (ACTH)/MSH (4-9) analog Org 2766 have been shown to ameliorate cisplatin ototoxicity. Both outer hair cell survival and recovery of the compound action potential threshold are significantly enhanced by both peptides.86 Adenosine Receptor Agonists Adenosine receptor agonists have been found to protect against cisplatin ototoxicity in chinchillas. Cisplatin applied to the RWM of the chinchilla produced significant loss of hair cells in the cochlea with concomitant elevation of auditory brainstem response thresholds. Pretreatment with adenosine A1 receptor agonist Rphenylisopropyladenosine (R-PIA) or 2-chloro-Ncyclopentyladenosine (CCPA) prevented hair cell loss, reduced brainstem auditory evoked potential threshold elevations, and reduced the increase in malondialdehyde (an index of lipid peroxidation). The effect of RPIA was abrogated by application of the adenosine A1 receptor antagonist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX). Not only did the application of the adenosine A2 receptor agonist 2-[4-(2-p-carboxyethyl) phenylamino]-5′N-ethylcarboxamidoadenosine (CGS) not protect against cisplatin ototoxicity, it exacerbated it.87 R-PIA also potentiated the protective effect of L-Nacetylcysteine on hair cells in the organotypic culture of the organ of Corti exposed to cisplatin. Nearly complete protection of hair cells was afforded by the combined application of these two agents in combination with cisplatin.66 These findings suggest that adenosine A1 receptor agonists might be administered on the RWM for protection against cisplatin ototoxicity.

Inhibitors of Cell Death Pathways The in vitro application of inhibitors of enzymes in the cell death pathway, caspases, has shown promise in protection against cisplatin toxicity for hair cells in vitro. Treatment of cisplatin-exposed cochlear explants, both prior to and during cisplatin administration, with inhibitors of caspase-1 or caspase-3, resulted in a near total prevention of DNA degradation in the hair cells, as indicated by terminal deoxynucleotidyl transferasemediated deoxyuridine triphosphate nick end labeling (TUNEL). These findings correlated with an increase in hair cell survival in the explants treated with caspase inhibitors in combination with cisplatin. However, damage to the stereocilia on the hair cells did occur in cultures protected by caspase inhibitors. Separate cultures of spiral ganglion cells treated with cisplatin required the presence of both caspase inhibitor and neurotrophins to prevent apoptotic cell death.88 Pifithrin is a low-molecular-weight inhibitor of TP53 (ie, p53). It has been found to provide significant protection against cisplatin-induced cochlear and vestibular hair cell loss in cochlear and utricular organotypic cultures. Control cultures were devoid of TP53 immunolabeling, TP53 protein on Western blots, and caspase-1 and caspase-3 labeling. Cisplatin exposure increased expression of TP53, caspase-1 and caspase-3 labeling, and apoptosis of cochlear and vestibular hair cells. Adding pifithrin to cisplatin-treated cultures resulted in a dose-dependent increase in hair cell survival, suppression in TP53 expression, and caspase-1 and -3 labeling. Temporary suppression of TP53 with pifithrin affords significant protection against the ototoxic and vestibulotoxic effects of cisplatin.89 Many tumors are already deficient in TP53, so TP53 inhibition may not interfere with the antitumor action of cisplatin in those cases. Future studies need to focus on the safety of TP53 inhibition in combination with cisplatin chemotherapy. Cultured auditory neurons exposed to cisplatin undergo cell death following exposure to cisplatin. A proposed molecular pathway from cellular insult to apoptosis is the activation and expression of immediate early genes, especially c-Jun. The activity of c-Jun is regulated by phosphorylation and activation of various components of the stress kinase pathway. The phosphorylation cascade required for c-Jun activation is mediated, in part, by c-Jun N-terminal kinase (JNK). The treatment of these neurons with curcumin, an upstream inhibitor of JNK–c-Jun interaction, rescued auditory neurons from cell death caused by cisplatin. Treatment with c-Jun antisense oligonucleotide also protected neurons from cisplatin-induced cell death.90 Whether such treatments will protect auditory hair cells from cisplatin-induced cell death remains to be determined.


The transduction of neurotrophin-3 using a viral vector successfully protected spiral ganglion cells from cisplatin ototoxicity both in vitro and in vivo in aged mice.91 The ability to deliver neurotrophins to the inner ear in this manner suggests that neurotrophin-based gene therapy may be a useful preventive treatment for ototoxic injury in the future.

CARBOPLATIN OTOPROTECTION Carboplatin is unique among ototoxic agents in causing preferential destruction of the inner hair cells in chinchillas, but in guinea pigs outer hair cell loss occurs and inner hair cells remain intact.92 Chinchillas treated with systemic carboplatin were found to have 84% loss of inner hair cells. Animals pretreated with D-Met prior to carboplatin treatment had a significantly lower loss of inner hair cells compared with chinchillas treated with carboplatin alone.93 In guinea pigs, sodium thiosulfate administered 1 to 8 hours after carboplatin decreased carboplatin-induced ototoxicity but was not effective when given 24 hours after carboplatin. In a lung cancer cell line in vitro, sodium thiosulfate combined with a tumoricidal dose of carboplatin completely blocked the cell killing.94 Carboplatin has been used to treat brain tumors. The blood-brain barrier is first opened with mannitol, then carboplatin is administered intravenously. This protocol results in hearing loss in 79% of patients. However, when sodium thiosulfate was administered intravenously 2 or 4 hours after carboplatin, when the blood-brain barrier had closed, patients were protected against hearing loss.95,96 Because thiosulfate is given after the blood-brain barrier has closed, there appears to be no interference with the antitumor effect of carboplatin.95

PLATINUM OTOPROTECTION RESEARCH TO DATE: SOME CONCLUSIONS Several protective agents have been shown to be effective against cisplatin and carboplatin ototoxicity. Some of these agents also interfere with the antitumor effect of these drugs. Novel routes of drug administration may overcome some of these problems that may limit the usefulness of chemoprotection. Additional clinical trials are needed to determine the dose, timing, and route of administration of protective agents to develop an effective protocol for chemoprotection against platinum toxicity while preserving the desired therapeutic effect.97 A schematic diagram of cell death pathways that may be activated and participate in the loss of auditory sensory cells in response to ototoxic damage caused by aminoglycoside antibiotics and chemotherapeutic agents is presented in Figure 20-1.

179

SUMMARY • Ototoxic injury to cochlear hair cells results in cellular death from apoptosis or necrosis. Both processes have distinct histologic and biochemical profiles. Because cellular necrosis is difficult to prevent, interventional strategies broadly known as otoprotection have been directed for the most part to lessen and if possible prevent apoptosis from occurring. Current research for the prevention of aminoglycoside and platinum-based chemotherapy ototoxicty appears promising from both in vitro studies and a few animal models studied to date. • Several otoprotective agents have been shown to be effective against aminoglycoside ototoxicity. Agents under investigation have included spintrapping agents (alpha-phenyl-tert-butylnitone), antioxidants (glutathione and methionine, iron chelators, salicylates, tanshinone, and SOD), inhibitors of cell death pathways and certain growth factors (ie, GDNF, BDNF, neurotrophins, and TGF-α). • Novel routes of drug administration will be required for some compounds to be effective against aminoglycoside ototoxicity. Delivery of otoprotective agents via the round window membrane has significant advantages whereby middle ear instillations can be used to prevent ototoxic injury from occurring. • Studied otoprotective agents for cisplatin ototoxicity have included thiol compounds, fosfomycin, certain antioxidants (eg, salicylates vitamin E), peptides, adenosine receptor agonists, and inhibitors of cell death pathways (eg, pifithrin, curumin, neurotrophin-3). Human studies are anticipated in the not too distant future. • Intratympanic administration strategies for otoprotection have the advantage of providing inner ear protection while not interfering with the antitumor efficacy of platinum-based chemotherapy. Future research protocols in humans will likely include this methodology. Additional trials are needed to determine the dose, timing, and route of administration for otoprotective agents against platinum-based compounds.

REFERENCES 1. Lefebvre PP, Malgrange B, Lallemend F, et al. Mechanisms of cell death in the injured auditory system: otoprotective strategies. Audiol Neurootol 2002;7:165–70. 2. Van De Water TR, Lallemend F, Eshraghi AA, et al. Caspases, the enemy within, their role in oxidative stress-induced apoptosis of inner ear sensory cells. Otol Neurotol 2004. [In press]

180

Interventions

3. Clerici WJ, DiMartino DL, Prasad MR. Direct effects of reactive oxygen species on cochlear outer hair cell shape in vitro. Hear Res 1995;84:30–40. 4. Clerici WJ, Hensley K, DiMartino DL, Butterfield DA. Direct detection of ototoxicant-induced reactive oxygen species generation in cochlear explants. Hear Res 1996;98:116–24. 5. Hirose K, Hockenbery DM, Rubel EW. Reactive oxygen species in chick hair cells after gentamicin exposure in vitro. Hear Res 1997;104:1–14. 6. Takumida M, Popa R, Anniko M. Free radicals in the guinea pig inner ear following gentamicin exposure. ORL J Otorhinolaryngol Relat Spec 1999;61:63–70. 7. Kopke R, Allen KA, Henderson D, et al. A radical demise. Toxins and trauma share common pathways in hair cell death. Ann N Y Acad Sci 1999; 884:171–91. 8. Evans P, Halliwell B. Free radicals and hearing. Cause, consequence, and criteria. Ann N Y Acad Sci 1999;884:19–40. 9. Lopez-Gonzalez MA, Delgado F, Lucas M. Aminoglycosides activate oxygen metabolites production in the cochlea of mature and developing rats. Hear Res 1999;136:165–8. 10. Mulheran M, Degg C, Burr S, et al. Occurrence and risk of cochleotoxicity in cystic fibrosis patients receiving repeated high-dose aminoglycoside therapy. Antimicrob Agents Chemother 2001;45:2502–9. 11. Morizono T. Toxicity of ototopical drugs: animal modeling. Ann Otol Rhinol Laryngol 1990;99: 42–5. 12. Barlow DW, Duckert LG, Kreig CS, et al. Ototoxicity of topical otomicrobial agents. Acta Otolaryngol (Stockh) 1994;115:231–5. 13. Roland PS. Clinical ototoxicity of topical antibiotic drops. Otolaryngol Head Neck Surg 1994;110: 598–602. 14. Roland PS, Stewart MG, Hannley M, et al. Consensus panel on role of potentially ototoxic antibiotics for topical middle ear use: introduction, methodology, and recommendations. Otolaryngol Head Neck Surg 2004;130:S51–S56. 15. Roland PS, Rybak L, Hannley M, et al. Animal ototoxicity of topical antibiotics and the relevance to clinical treatment of human subjects. Otolaryngol Head Neck Surg 2004;130:S57–S78. 16. Matz G, Rybak L, Roland PS, et al. Ototoxicity of ototopical antibiotic drops in humans. Otolaryngol Head Neck Surg 2004;130:S79–S82. 17. Hester TO, Jones RO, Clerici WJ. Protection against aminoglycoside otic drop-induced ototoxicity by a spin trap: I. acute effects. Otolaryngol Head Neck Surg 1998;119:581–7. 18. Hoffman DW, Whitworth CA, Jones KL, Rybak LP. Nutritional status, glutathione levels and ototoxi-

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

city of loop diuretics and aminoglycoside antibiotics. Hear Res 1987;31:217–22. Hoffman DW, Jones-King KL, Whitworth CA, Rybak LP. Potentiation of ototoxicity by glutathione depletion. Otol Rhinol Laryngol 1988; 97:36–41. Garetz SL, Rhee DJ, Schacht J. Sulfhydryl compounds and antioxidants inhibit cytotoxicity to outer hair cells of a gentamicin metabolite in vitro. Hear Res 1994;77:75–80. Zenner HP, Keiner S, Zimmerman U. Specific glutathione-SH inhibition of toxic effects of metabolized gentamicin on isolated guinea pig hair cells. Eur Arch Otorhinolaryngol 1994;251: 84–90. Lautermann J, McLaren J, Schacht J. Differential vulnerability of basal glutathione protection against gentamicin ototoxicity depends on nutritional status. Hear Res 1995;86:15–24. Sha SH, Taylor R, Forge A, Schacht J. Differential vulnerability of basal and apical hair cells is based on intrinsic susceptibility to free radicals. Hear Res 2001;155:1–8. Sha SH, Schacht J. Antioxidants attenuate gentamicin-induced free radical formation in vitro and ototoxicity in vivo: D-methionine is a potential protectant. Hear Res 2000;142:34–40. Song BB, Schacht J. Variable efficacy of radical scavengers and iron chelators to attenuate gentamicin ototoxicity in guinea pig in vivo. Hear Res 1996;94:87–93. Song BB, Anderson DJ, Schacht J. Protection from gentamicin ototoxicity by iron chelators in guinea pig in vivo. J Pharmacol Exp Ther 1997;282:369–77. Conlon BJ, Perry BP, Smith DW. Attenuation of neomycin ototoxicity by iron chelation. Laryngoscope 1998;108:284–7. Song BB, Sha SH, Schacht J. Iron chelators protect from aminoglycoside-induced cochleo- and vestibulo-toxicity. Free Radic Biol Med 1998; 25:189–95. Sinswat P, Wu WJ, Sha HS, Schacht J. Protection from ototoxicity of intraperitoneal gentamicin in guinea pig. Kidney Int 2000;58:2525–32. Dehne N, Rauen U, de Groot H, Lautermann J. Involvement of the mitochondrial permeability transition in gentamicin ototoxicity. Hear Res 2002;169:47–55. Sha SH, Schacht J. Salicylate attenuates gentamicin-induced ototoxicity. Lab Invest 1999; 79:807–13. Wang AM, Sha SH, Lesniak W, Schacht J. Tanshinone (Salviae miltiorrhizae extract) preparations attenuate aminoglycoside-induced free radical formation in vitro and ototoxicity in vivo. Antimicrob Agents Chemother 2003;47:1836–41.


33. Sha SH, Zajic G, Epstein CJ, Schacht J. Overexpression of copper/zinc-superoxide dismutase protects from kanamycin-induced hearing loss. Audiol Neurootol 2001;6:117–23. 34. Kawamoto K, Sha SH, Minoda R, et al. Antioxidant gene therapy can protect hearing and hair cells from ototoxicity. Mol Ther 2004;9:173–81. 35. McFadden SL, Ding D, Salvemini D, Salvi RJ. M40403, a superoxide dismutase mimetic, protects cochlear hair cells from gentamicin, but not cisplatin toxicity. Toxicol Appl Pharmacol 2003; 186:46–54. 36. Pirvola U, Xing-Qun L, Virkkala J, et al. Rescue of hearing, auditory hair cells, and neurons by CEP1347/KT7515, and inhibitor of c-Jun N-terminal kinase activation. J Neurosci 2000;20:43–50. 37. Ylikoski J, Xing-Qun L, Virkkala J, Pirvola U. Blockade of c-Jun N-terminal kinase pathway attenuates gentamicin-induced cochlear and vestibular hair cell death. Hear Res 2002;166:33–43. 38. Wang J, Van De Water TR, Bonny C, et al. A peptide inhibitor of c-Jun N-terminal kinase protects against both aminoglycoside and acoustic traumainduced auditory hair cell death and hearing loss. J Neurosci 2003;23:8596–607. 39. Forge A, Li L. Apoptotic death of hair cells in mammalian vestibular sensory epithelia. Hear Res 2000; 139:97–115. 40. Matsui JI, Ogilvie JM, Warchol ME. Inhibition of caspases prevents ototoxic and ongoing hair cell death. J Neurosci 2002;22:1218–27. 41. Cheng AG, Cunningham LL, Rubel EW, et al. Hair cell death in the avian basilar papilla: characterization of the in vitro model and caspase activation. J Assoc Res Otolaryngol 2003;4:91–105. 42. Cunningham LL, Cheng AG, Rubel EW. Caspase activation in hair cells of the mouse utricle exposed to neomycin. J Neurosci 2002;22:8532–40. 43. Kuang R, Hever G, Zajic G, et al. Glial cell linederived neurotrophic factor. Potential for otoprotection. Ann N Y Acad Sci 1999;884:270–91. 44. Yagi M, Magal E, Sheng Z, et al. Hair cell protection from aminoglycoside ototoxicity by adenovirusmediated overexpression of glial cell line-derived neurotrophic factor. Hum Gene Ther 1999; 10:813–23. 45. Suzuki M, Yagi M, Brown JN, et al. Effect of transgenic GDNF expression on gentamicin-induced cochlear and vestibular toxicity. Gene Ther 2000; 7:1046–54. 46. Ernfors P, Van De Water TR, Loring J, Jaenisch R. Complimentary roles of BDNF and NT-3 in vestibular and auditory development. Neuron 1995;14:1153–64. 47. Staecker H, Kopker R, Malgrange B, et al. NT-3 and/or BDNF therapy prevents loss of auditory

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

181

neurons following loss of hair cells. Neuroreport 1996;7:889–94. Ernfors P, Duan ML, ElShamy WM, Canlon B. Protection of auditory neurons from aminoglycoside toxicity by neurotrophin-3. Nat Med 1996;2: 463–7. Zheng JL, Stewart RR, Gao WQ. Neurotrophin-4/5 enhances survival of cultured spiral ganglion neurons and protects them from cisplatin neurotoxicity. J Neurosci 1995;15:5079–87. Zheng JL, Gao WQ. Differential damage to auditory neurons and hair cells by ototoxins and neuroprotection by specific neurotrophins in rat cochlear organotypic cultures. Eur J Neurosci 1996;8:1897–905. Zheng JL, Stewart RR, Gao WG. Neurotrophin4/5, brain-derived neurotrophic factor, and neurotrophin-3 promote survival of cultured vestibular ganglion neurons and protect them against neurotoxicity of ototoxins. J Neurobiol 1995;28:330–40. Duan M, Agerman K, Ernfors P, Canlon B. Complementary roles of neurotrophin 3 and a Nmethyl-D-aspartate antagonist in the protection of noise and aminoglycoside-induced ototoxicity. Proc Natl Acad Sci U S A 2000;97:7597–602. Staecker H, Dazert S, Malgrange B, et al. Transforming growth factor alpha treatment alters intracellular calcium levels in hair cells and protects them from ototoxic damage in vitro. Int J Dev Neurosci 1997;15:553–62. Sala T. Transtympanic administration of aminoglycosides in patients with Ménière’s disease. Arch Otorhinolaryngol 1988;245:293–6. Seidman MD, Van De Water TR. Pharmacologic manipulation of the labyrinth with novel and traditional agents delivered to the inner ear. Ear Nose Throat J 2003;82:276–80,282–3,287–8. Silverstein H. Use of a new device, the MicroWickTM, to deliver medication to the inner ear. Ear Nose Throat J 1999;78:24–7. Benedetti Pancini P, Greggi S, Scambia G, et al. Efficacy and toxicity of very high-dose cisplatin in advanced ovarian carcinoma: 4-year survival analysis and neurological follow-up. Int J Gynecol Cancer 1993;3:44–53. Otto WC, Brown RD, Gage-White L, et al. Effect of cisplatin and thiosulfate upon auditory brainstem responses of guinea pigs. Hear Res 1988;35:79–85. Kaltenbach JA, Church MW, Blakley BW, et al. Comparison of five agents in protecting the cochlea against the ototoxic effects of cisplatin in the hamster. Otolaryngol Head Neck Surg 1997; 117:493–500. Wang J, Faulconbridge RVL, Fetoni A, et al. Local application of sodium thiosulfate prevents

182

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

Interventions

cisplatin-induced hearing loss in the guinea pig. Neuropharmacology 2003;45:380–94. Wimmer C, Mees K, Stumpf P, et al. Round window application of D -methionine, sodium thiosulfate, brain-derived neurotrophic factor and fibroblast growth factor-2 in cisplatin-induced ototoxicity. Otol Neurotol 2004;25:33–40. Ekborn A, Laurell G, Ehrsson H, Miller J. Intracochlear administration of thiourea protects against cisplatin-induced outer hair cell loss in the guinea pig. Hear Res 2003;181:109–15. Verschraagen M, Boven E, Ruijter R, et al. Pharmacokinetics and preliminary clinical data of the novel chemoprotectant BNP7787 and cisplatin and their metabolites. Clin Pharmacol Ther 2003;74:157–69. Rybak LP, Ravi R, Somani SM. Mechanism of protection by diethyldithiocarbamate against cisplatin ototoxicity: antioxidant system. Fundam Appl Toxicol 1995;26:293–300. Konstantinov S, Topashka-Ancheva M, Karaivanova M, et al. Antitumor, nephrotoxic and clastogenic effect of cis-DDP with DDTC or NAC. Neoplasma 1994;42:253–8. Kopke RD, Liu W, Gabaizadeh R, et al. The use of organotypic cultures of Corti’s organ to study the protective effects of antioxidant molecules on cisplatin-induced damage of auditory hair cells. Am J Otol 1997;18:559–71. Kamimura T, Whitworth CA, Rybak LP. Effect of 4-methylthiobenzoic acid on cisplatin-induced ototoxicity in the rat. Hear Res 1999;131:117–27. Rybak LP, Husain K, Whitworth C, Somani SM. Dose dependent protection by lipoic acid against cisplatin-induced ototoxicity in rats: antioxidant defense system. Toxicol Sci 1999;47:195–202. Boogaard PJ, Lempers EL, Mulder GJ, Meerman JHN. 4-Methylthiobenzoic acid reduces cisplatin nephrotoxicity in rats without compromising antitumor activity. Biochem Pharmacol 1991;41: 1997–2003. Campbell KCM, Rybak LP, Meech RP, Hughes L. D-Methionine provides excellent protection from cisplatin ototoxicity in the rat. Hear Res 1996; 102:90–8. Campbell KCM, Meech RP, Rybak LP, Hughes LF. D-Methionine protects against cisplatin damage to the stria vascularis. Hear Res 1999;138:13–28. Cloven NG, Re A, McHale M, et al. Evaluation of D-methionine as a cytoprotectant in cisplatin treatment of an animal model for ovarian cancer. Anticancer Res 2000;20:4205–10. Reser D, Rho M, Dewan D, et al. L- and D-methionine provide equivalent long term protection against CDDP-induced ototoxicity in vivo, with partial in vitro and in vivo retention of antineoplastic activity. Neurotoxicology 1999;20:731–48.

74. Li G, Frenz DA, Brahmblatt S, et al. Round window membrane delivery of L-methionine provides protection from cisplatin ototoxicity without compromising chemotherapeutic efficacy. Neurotoxicology 2000;10:1–14. 75. Korver KD, Rybak LP, Whitworth CA, Campbell KM. Round window application of D-methionine provides complete cisplatin otoprotection. Otolaryngol Head Neck Surg 2002;126:683–9. 76. Ekborn A, Laurell G, Johnstrom P, et al. D-Methionine and cisplatin ototoxicity in the guinea pig: D -methionine influences cisplatin pharmacokinetics. Hear Res 2002;165:53–61. 77. Rybak LP, Whitworth C, Somani S. Application of antioxidants and other agents to prevent cisplatin ototoxicity. Laryngoscope 1999;109:1740–4. 78. Schweitzer VG, Dolan DF, Davidson T. Amelioration of cisplatin-induced ototoxicity by fosfomycin. Laryngoscope 1986;96:948–58. 79. Li G, Sha S-H, Zotova E, et al. Salicylate protects hearing and kidney function without compromising its oncolytic action. Lab Invest 2002;82: 585–96. 80. Teranishi M-A, Nakashima T, Wakabayashi T. Effects of alpha-tocopherol on cisplatin-induced ototoxicity in guinea pigs. Hear Res 2001;151: 61–70. 81. Kalkanis JG, Whitworth C, Rybak LP. Vitamin E reduces cisplatin ototoxicity. Laryngoscope 2004; 114:538–42. 82. Teranishi M, Nakashima T. Effects of trolox, locally applied on round windows, on cisplatin-induced ototoxicity in guinea pigs. Int J Pediatr Otorhinolaryngol 2003;67:133–9. 83. Feghali JG, Liu W, Van De Water TR. L-N-Acetyl cysteine protection against cisplatin-induced auditory neuronal and hair cell toxicity. Laryngoscope 2001;111:1147–55. 84. Campbell KCM, Larsen DL, Meech RP, et al. Glutathione ester, but not glutathione, protects against cisplatin-induced ototoxicity in a rat model. J Am Acad Audiol 2003;14:124–33. 85. Kelly TC, Whitworth CA, Husain K, Rybak LP. Aminoguanidine reduces cisplatin ototoxicity. Hear Res 2003;186:10–6. 86. Hamers FPT, Wijbenga J, Wolters FLC, et al. Cisplatin ototoxicity involves organ of Corti, stria vascularis and spiral ganglion: modulation by alpha-MSH and ORG 2766. Audiol Neurootol 2003;8:305–15. 87. Whitworth CA, Ramkumar V, Jones B, et al. Protection against cisplatin ototoxicity by adenosine agonists. Biochem Pharmacol 2004:67:1801–7. 88. Liu W, Staecker H, Stupak H, et al. Caspase inhibitors prevent cisplatin-induced apoptosis of auditory sensory cells. Neuroreport 1998;9:2609–14.


89. Zhang M, Liu W, Ding D, Salvi R. Pifithrin-alpha suppresses p53 and protects cochlear and vestibular hair cells from cisplatin-induced apoptosis. Neuroscience 2003;120:191–205. 90. Scarpidis U, Madnani D, Shoemaker C, et al. Arrest of apoptosis in auditory neurons: implications for sensorineural preservation in cochlear implantation. Otol Neurotol 2003;24:409–17. 91. Bowers WJ, Chen X, Guo H, et al. Neurotrophin-3 transduction attenuates cisplatin spiral ganglion neuron ototoxicity in the cochlea. Mol Ther 2002;6:12–8. 92. Takeno S, Harrison RV, Mount RJ, et al. Induction of selective inner hair cell damage by carboplatin. Scanning Microsc 1994;8:97–106. 93. Lockwood DS, Ding DL, Wang J, Salvi RJ. D-methionine attenuates inner hair cell loss in carboplatintreated chinchillas. Audiol Neurootol 2000;5:263–6. 94. Neuwelt EA, Brummett RE, Remsen LG, et al. In vitro and animal studies of sodium thiosulfate

183

as a potential chemoprotectant against carboplatin-induced ototoxicity. Cancer Res 1996;56: 706–9. 95. Neuwelt EA, Brummett RE, Doolittle ND, et al. First evidence of otoprotection against carboplatin-induced hearing loss with a two compartment model in patients with CNS malignancy. J Pharmacol Exp Ther 1998;286;77–84. 96. Doolittle ND, Muldoon LL, Brummett RE, et al. Delayed sodium thiosulfate as an otoprotectant against carboplatin-induced hearing loss in patients with malignant brain tumors. Clin Cancer Res 2001;7:493–500. 97. Blakley BW, Cohen JI, Doolittle ND, et al. Strategies for prevention of toxicity caused by platinum-based chemotherapy: review and summary of the annual meeting of the blood-barrier disruption program, Gleneden Beach, Oregon, March 10, 2001. Laryngoscope 2002;112: 1997–2001.

Therapeutic Uses of Ototoxic Effects CHAPTER 21

Systemic Treatment of Bilateral Meniere’s Disease Sumit K. Agrawal, MD, and Lorne S. Parnes, MD, FRCSC

DEFINITION OF MENIERE’S DISEASE Meniere’s disease can be defined as the idiopathic syndrome of endolymphatic hydrops. The Committee on Hearing and Equilibrium of the American Academy of Otolaryngology–Head and Neck Surgery (AAO-HNS) published guidelines for reporting the results of treatment of Meniere’s disease in 1972 and 1985. These guidelines were most recently updated in 1995. Although diagnosis can be confirmed only by postmortem histopathologic analysis of temporal bones, the AAO-HNS guidelines can be used clinically to diagnose Meniere’s disease and classify it as definite, probable, or possible. The criteria for definite Meniere’s disease consist of (1) two or more definitive spontaneous episodes of vertigo lasting 20 minutes or longer, (2) audiometrically documented hearing loss on at least one occasion, (3) tinnitus or aural fullness, and (4) the exclusion of other causes. Probable Meniere’s disease requires only one episode of vertigo along with the three other criteria. Possible Meniere’s disease requires that other causes be excluded while consisting of either episodic vertigo of the Meniere’s type without associated hearing loss or sensorineural hearing loss, fluctuating or fixed, with chronic dysequilibrium.1

PATHOPHYSIOLOGY Numerous theories exist regarding the etiology and pathophysiology of Meniere’s disease. An overaccumulation of endolymph is thought to result in endolymphatic hydrops, which in turn causes a distortion in the membranous labyrinth.2 The most accepted theory is the inadequate absorption of endolymph by the endolymphatic sac.3 This is supported by studies examining animal models,4,5 the finding of perisaccular ischemia and fibrosis in patients with hydrops,6 and computed tomography (CT) and magnetic resonance imaging (MRI) studies suggesting hypoplastic endolymphatic drainage systems.7–10

Although temporal bone studies of patients with Meniere’s disease almost always show endolymphatic hydrops, the presence of endolymphatic hydrops may be secondary to other causes and occur in the absence of clinical Meniere’s disease.11,12 The mechanism by which endolymphatic hydrops causes vertigo and hearing loss is much debated in the literature. Schuknecht postulated that membrane ruptures within the inner ear cause leakage of potassium-rich endolymph into the sodium-rich perilymph (ie, the “Na+–K+ intoxication theory”), which subsequently alters endocochlear potentials, resulting in a profound depolarization with deactivation of electrical activity along the VIIIth nerve; the acute unilateral loss of electrical activity thereby causes the hearing loss and vertigo experienced by patients.13 With the repair of membrane ruptures, the acute attack would subside and hearing and balance would be restored.14

DEMOGRAPHICS The incidence of Meniere’s disease varies from 7.5 to 157 per 100,000; no significant difference in race or sex has been shown.15 The peak incidence occurs in patients 40 to 60 years of age. 3 The incidence of bilateral Meniere’s disease is highly variable in the literature, with ranges of 2 to 78%.16 This is likely because the stage of Meniere’s disease varies at the time of diagnosis and different diagnostic criteria are used (AAO-HNS guidelines vs audiometric data only). Whereas some authors have found a decreased likelihood of bilateral involvement if it has not occurred within 5 years, others have found that the incidence of bilateral disease increases over time.17–19 Kitahara surveyed 15 institutions in Japan and found that the incidence of bilateral disease rose from 9.1% in the first year to 41.5% after 20 years.20

NATURAL HISTORY The natural history of Meniere’s disease is quite variable. Friberg and colleagues studied 161 patients over

Systemic Treatment of Bilateral Meniere’s Disease

9 years and found patients on average had a 50 dB average pure-tone hearing loss, a 53% speech discrimination score, and a 50% caloric response reduction.17 The mean frequency of attacks decreased over 20 years, and the incidence of bilateral disease increased to 47%. Patients can often have several attacks followed by long remissions. One study found spontaneous cessations of vertigo in 57% of patients after 2 years and 71% of patients after 8.3 years.21

HISTORY OF STREPTOMYCIN-INDUCED VESTIBULOTOXICITY Streptomycin sulfate was first developed in 1944 by Shatz and colleagues.22 In 1945, Hinshaw and Feldman reported the first clinical trial using this drug in tuberculosis patients.23 Treatment of these patients with streptomycin led to the realization that the drug was ototoxic, and detailed accounts of vestibular disturbances in these patients were published.23–25 Although streptomycin is metabolized to dihydrostreptomycin, which by itself causes severe vestibulotoxicity and cochlear damage,26 the effects on the vestibular system were shown to occur prior to the effects on cochlear function.27 In 1946, Hawkins considered these characteristics and suggested the use of streptomycin sulfate in order to prevent vertigo in Meniere’s disease.28 Fowler was the first to use intramuscular streptomycin sulfate (ISS) for vertigo in 1948.29 He treated four patients, who all had a resolution of their vertigo. In the following 3 years, Van Deinse,30 Hamberger and colleagues,31 Hanson,32 and Ruedi33 all treated 3 to 5 patients and had response rates of 66 to 100%. In 1956, Shuknecht was the first to systematically use ablative therapy with ISS in eight patients with Meniere’s disease.34 Since that time he and his colleagues have published on 20 patients followed over 25 years.35–38 Although there was a 95% response rate in terms of vertigo control with no significant hearing loss, ablative therapy caused patients to develop severe oscillopsia and ataxia for several months. Planned subtotal ablation using titration ISS was first introduced by Graham and colleagues in 1982 and later by Silverstein in 1983.39–42 This has also been shown to be effective in controlling vertigo but with a decreased incidence and severity of ataxia and oscillopsia compared with ablative therapy.

MECHANISM OF ACTION IN MENIERE’S DISEASE Aminoglycosides inhibit ribosomal protein synthesis in order to exert their antibiotic properties (see Chapter 8, “Clinical Aminoglycoside Ototoxicity”). In the inner ear, aminoglycosides have been shown to affect both the sensory neuroepithelium and endolymph-secreting cells. In the vestibular apparatus, aminoglycosides primarily affect the hair cells of the crista ampullaris and the macula of the saccule and utricle.43,44 Cochleotoxic

185

effects of aminoglycosides have been linked to degeneration of neural cells, ganglion cells, and the organ of Corti, as well as to the loss of hair cells and supporting structures. 45,46 In terms of endolymph production, which has been linked to the dark cells of the crista ampullaris, 47 aminoglycosides have been shown to damage the secretory epithelium before the sensory neuroepithelium.48–50 Therefore, aminoglycosides may actually reverse endolymphatic hydrops in addition to ablating the vestibular function; this may explain why certain patients experience a relief from vertigo with subtotal ablation (intact caloric responses).51 In the search for aminoglycosides with a wide therapeutic index, gentamicin was found to be relatively less cochleotoxic than streptomycin and less cumulative in the inner ear than tobramycin.52

INDICATIONS FOR SYSTEMIC THERAPY Current indications for ISS include (1) active bilateral Meniere’s disease and (2) Meniere’s disease in an only hearing ear.28,41,53,54 In bilateral incapacitating Meniere’s disease, only a limited number of options are available. For example, endolymphatic sac surgery could technically be performed on both sides with minimal risk to hearing. However, the efficacy has been reported to be as low as 50%, and long-term benefits have been questioned.55–58 Vestibular nerve sections have also been performed in bilateral Meniere’s disease but resulted in oscillopsia.59 Meniere’s disease in the only hearing ear also poses a challenge, and vestibular nerve sections have been contraindicated because of the significant risk to hearing.60 Intratympanic gentamicin therapy is now widely used, but the average rate of hearing loss is 30%.61 Toth and Parnes found that a weekly titration method significantly reduced the rate of hearing loss62; however, a certain subset of patients may develop a sudden, severe, and irreversible hearing loss. 63 Labyrinthectomy is obviously contraindicated in both of these scenarios as it would destroy both hearing and vestibular function.

REVIEW OF TREATMENT REGIMENS Two main protocols have been used to administer ISS. The first, ablative ISS, was introduced by Schuknecht in 1956 and was based upon dosing regimens used for treating tuberculosis.34 Intramuscular injections of streptomycin sulfate were administered, 1 g bid, until there was no response to ice water caloric stimulation. Patients were hospitalized, and the total dose ranged from 21 to 72 g. All patients developed a profound ataxia, wide-based gait, and oscillopsia but slowly resumed normal function over 2 to 9 months. In 1984, Graham and colleagues proposed titration ISS in an effort to prevent oscillopsia and ataxia.39 Baseline studies were conducted including (1) air, bone, and speech audiometry, (2) electronystagmography (ENG),

186

Therapeutic Uses of Ototoxic Effects

and (3) blood work including complete blood count, blood urea nitrogen, and creatinine. Intramuscular injections of streptomycin sulfate were administered, 1 g bid, for 5 days (total 10 g), and repeat baseline studies were performed 3 days following. According to Graham, treatment should be suspended “when symptoms cease, when vestibular function subsides rapidly, should hearing decline, or if the patient becomes oscillopsic.”28 If the patient did not meet these criteria, treatment continued for 3 days (total 6 g), and then 2 weeks were allowed to pass before repeating baseline studies. The ISS treatment continued, as necessary, in 2-day increments (total 4 g), with a minimum of 2 weeks before retesting.28 The incidence and severity of ataxia and oscillopsia using titration ISS were lower than with ablative ISS and appeared to be equally effective for vertigo control. The frequency of monitoring with different protocols has varied in the literature. Schuknecht realized that unmonitored daily dosing would lead to a rapid decline in vestibular function and oscillopsia and that intermittent monitoring of inner ear function was necessary.34 Initial protocols repeated studies only once a total of 20 g of streptomycin was given; however, doses less than this can control vertigo or cause oscillopsia.39,40,60 It is now advocated to increase monitoring frequency and decrease dosage in patients with reduced vestibular or renal function.53 Older patients often require a smaller total dose of streptomycin to suppress vestibular responses, and patients who already have reduced caloric responses may be more difficult to titrate.28

RESULTS Although systemic streptomycin treatment for Meniere’s disease was initiated over 50 years ago, the world literature is still limited to a small number of case series. Comparison of these studies is difficult as they often use different protocols, have varying follow-up time, have different criteria for Meniere’s disease, and fail to report important details. Balyan and colleagues compiled an excellent table summarizing their literature review.54 Table 21-1 has been adapted from their work with minor revisions and updates. Ablative ISS therapy used a total streptomycin dose of 8 to 90 g. The majority of studies had a 100% relief from vertigo, with the overall range being 50 to 100%. Nearly all patients developed severe ataxia and oscillopsia, which slowly resolved over the following 9 months. The rates of mild residual dysequilibrium were quite high and ranged from 50 to 100%. Hearing was preserved or improved in nearly all studies, with only Wilson and Schuknecht revealing a hearing loss in four ears.38 Titration ISS used less total streptomycin than ablative ISS, ranging from 5 to 70 g. There was excellent relief of vertigo attacks, with the total range of 66 to

100%. The incidence and severity of post-treatment dysequilibrium, ataxia, and oscillopsia were less than the ablative ISS group, with the majority of studies quoting an incidence of 50%. Hearing loss was noted in only one study by Langman and colleagues, where cochlear function was reduced in 13 ears.60 As the minimum followup was 2 years, it is difficult to ascertain whether this was a direct result of streptomycin treatment or secondary to the natural progression of the disease. Posttreatment caloric responses were not described.

COMPLICATIONS OF THERAPY In 1952, the New England Journal of Medicine published a riveting account by a physician recounting his own experience with vestibular dysfunction caused by streptomycin therapy for tuberculous arthritis.64 Side effects often experienced during the initial titration include pain at the injection site; numbness and tingling around the mouth, fingers, and toes; marked nausea; transient oscillopsia (even with active caloric responses); and skin rash.28,41 Following ablative therapy, Silverstein states that all patients experience a direction-changing nystagmus, dysequilibrium, profound ataxia, oscillopsia, and wide-based gait.41 No details regarding the direction-changing nystagmus were provided by the author. Intuitively, one would not expect nystagmus to be present as simultaneous bilateral vestibular ablation should not result in vestibular asymmetry. Wilson and Schuknecht found that following vestibular ablation, all 20 patients were able to resume normal activity within 2 to 9 months, contrary to the clinical experience of most physicians.38 Patients who have been inadvertently ablated by systemic aminoglycosides for treatment of a severe infection are usually permanently disabled unless they recover vestibular function as demonstrated by regained caloric responses. One possible explanation could be that Wilson and Schuknecht’s patients were able to recover some vestibular function over time. Jackson and Arcieri reviewed 69 patients suffering from gentamicininduced ototoxicity and found that labyrinthine dysfunction was reversible in more than 50% of patients if the drug was promptly discontinued.65 However, if Wilson and Schuknecht’s patients did recover their vestibular function over time, it is somewhat surprising that they did not have a recurrence of their Meniere’s symptoms over their 25-year follow-up.38 This is perhaps secondary to the aminoglycoside effect on the dark cells of the crista ampullaris and the resulting decrease in endolymph production. During titration therapy, one-third of patients may develop oscillopsia in the early stages of treatment despite active bithermal responses. This oscillopsia almost always disappears with the discontinuation of the drug; however, if oscillopsia is present with ablated

13.5–89 13.5–89

5

3

8

8

2

4

13

Ruedi, 195133

Schuknecht, 195634

Gunther, 195967

Jatho, 196368

Graybiel et al, 196737

Singleton and Schuknecht, 196836 15

20

Hanson, 195132

Wilson and Schuknecht, 198038

Silverstein, 198441

3

LaRouere et al, 199353 16–25

10–32

5–50

20–70

10–60

15–34

30–60

21–54

30–61

30–58

21–72

8–40

40–86

55–90

Adapted from Balyan et al.54 NR = not reported *Reported in number of ears. †Only 3 patients treated for vertigo, the remainder for hearing loss.

13

19

Langman et al, 199060

Balyan et al, 199854

7

Moretz et al, 198769

20

7†

Silverstein, 198441

Graham and Kemink, 198458

8

Graham et al, 198439

Titration

22.5–54

4

Hamberger et al, 194931

40–56

3

Van Deinse, 194930

25–33

Total Dose (g)

4

Total Cases

Fowler, 194829

Ablative

Authors

Table 21-1 Literature Review

13 (100)

2 (66)

12 (63)

7 (100)

18 (90)

2 (66)

8 (100)

13 (100)

19 (95)

15 (100)

4 (100)

2 (100)

8 (100)

8 (100)

2 (66)

5 (100)

4 (100)

2 (66)

2 (50)

Relief of Vertigo (%)

4 (31)

3 (100)

9 (47)

4 (57)

11 (55)

NR

5 (62)

13 (100)

10 (50)

15 (100)

4 (100)

NR

NR

8 (100)

0

3 (60)

2 (50)

NR

2 (50)

Residual Dysequilibrium, Ataxia, or Oscillopsia (%)

Unchanged (16),* improved (4)*

Improved (3)

Unchanged (17),* improved (8),* worsened (13)*

Unchanged (6), improved (1)




Improved (5), NR (8)



Unchanged (1),* improved (4)*

Unchanged (2)

Improved (8)

Improved (5), unchanged (3)

NR



NR

Unchanged (2), NR (2)

Hearing (no. of patients)

2–9

1.8–5

2–9.3

0.5–3.5

1–5

1.1–1.5

0.4–2

NR

1–16

0.6–1.2

2.1–4

10

Up to 3

1–4.9

0.2–0.5

1.5–4

0.2–0.5

0.4

0.5–1

Follow-up (yr)

Systemic Treatment of Bilateral Meniere’s Disease 187

188


caloric responses on ENG, it is generally permanent.28 The ataxia experienced following titration therapy is much less severe than with ablative therapy and has a quicker resolution.41 Mild persistent dysequilibrium has been found to be extremely common following ISS therapy and is felt to be a universal side effect.53 Black and colleagues found that rotational chair testing was not predictive of rapid vestibular decline but may help to predict long-term dysequilibrium.66 Therefore, using rotatory chair testing during ISS may help to prevent long-term dysequilibrium.53 In most studies (see Table 21-1), hearing was found to be stable or to improve during treatment with ISS. This hearing gain has been found to be transient, and ISS likely does not change the natural progression of Meniere’s disease.41,53 Although Wilson and Schucknecht38 and Langman and colleagues60 found patients with a progressive hearing loss during long-term followup, it is difficult to differentiate whether this is secondary to the ISS or to the underlying disease process.53 Despite more than 50 years of experience with ISS therapy for Meniere’s disease, scarcely more than 100 cases are documented in the literature worldwide. Considering that bilateral Meniere’s disease is fairly common, clinicians seem reluctant to use ISS therapy as their first line of treatment, likely for concern of inducing permanent dysequilibrium.

SUMMARY • ISS therapy is an important therapeutic consideration in treating patients with incapacitating bilateral Meniere’s disease or patients with Meniere’s disease affecting their only hearing ear. • Ablative ISS therapy has been found to be very effective in controlling vertigo, but post-treatment ataxia and oscillopsia are often severe. Titration ISS therapy has been shown to be equally efficacious in controlling disabling vertigo and reduces the incidence and severity of post-treatment vestibular dysfunction. • Although persistent mild dysequilibrium is common, the world literature suggests that it is well tolerated by most patients. In most cases, patients experience either a transient improvement or no change in hearing following streptomycin therapy. Overall, ISS-associated hearing loss is rare.

REFERENCES 1. Committee on Hearing and Equilibrium guidelines for the diagnosis and evaluation of therapy in Meniere’s disease. American Academy of Otolaryngology-Head and Neck Foundation, Inc. Otolaryngol Head Neck Surg 1995;113:181–5.

2. Hallpike CS, Carins H. Observations on the pathology of Meniere’s syndrome. J Laryngol Otol 1938;53:625. 3. Paparella MM. The cause (multifactorial inheritance) and pathogenesis (endolymphatic malabsorption) of Meniere’s disease and its symptoms (mechanical and chemical). Acta Otolaryngol 1985;99:445–51. 4. Fukuda S, Keithley EM, Harris JP. The development of endolymphatic hydrops following CMV inoculation of the endolymphatic sac. Laryngoscope 1988;98:439–43. 5. Kimura RS. Experimental blockage of the endolymphatic duct and sac and its effect on the inner ear of the guinea pig. A study on endolymphatic hydrops. Ann Otol Rhinol Laryngol 1967; 76:664–87. 6. Yazawa Y, Kitahara M. Immunofluorescent study of the endolymphatic sac in Meniere’s disease. Acta Otolaryngol Suppl 1989;468:71–6. 7. Albers FW, Van Weissenbruch R, Casselman JW. 3DFT-magnetic resonance imaging of the inner ear in Meniere’s disease. Acta Otolaryngol 1994; 114:595–600. 8. Dreisbach J, Seibert C, Arenberg IK. Patency and visibility of the vestibular aqueduct in Meniere’s disease. A comparison between conventional multidirectional tomography and reformatted high resolution computed tomographic scanning of the temporal bone. Otolaryngol Clin North Am 1983;16:103–13. 9. Tanioka H, Zusho H, Machida T, et al. Highresolution MR imaging of the inner ear: findings in Meniere’s disease. Eur J Radiol 1992;15:83–8. 10. Valvassori GE, Dobben GD. Multidirectional and computerized tomography of the vestibular aqueduct in Meniere’s disease. Ann Otol Rhinol Laryngol 1984;93:547–50. 11. Horner KC. Functional changes associated with experimentally induced endolymphatic hydrops. Hear Res 1993;68:1–18. 12. Rauch SD, Merchant SN, Thedinger BA. Meniere’s syndrome and endolymphatic hydrops. Doubleblind temporal bone study. Ann Otol Rhinol Laryngol 1989;98:873–83. 13. Schuknecht HF. The pathophysiology of Meniere’s disease. Am J Otol 1984;5:526–7. 14. Kimura RS, Schuknecht HF. Effect of fistulae on endolymphatic hydrops. Ann Otol Rhinol Laryngol 1975;84:271–86. 15. Pfaltz C, Thomsen J. Symptomatology and definition of Meniere’s disease. In: Pfaltz C, editor. Controversial aspects of Meniere’s disease. New York: Thieme; 1986. p. 2–7. 16. Balkany TJ, Sires B, Arenberg IK. Bilateral aspects of Meniere’s disease: an underestimated clinical

Systemic Treatment of Bilateral Meniere’s Disease

17.

18.

19.

20.

21.

22.

23.

24. 25.

26. 27.

28.

29.

30. 31.

32.

33. 34. 35.

entity. Otolaryngol Clin North Am 1980;13: 603–9. Friberg U, Stahle J, Svedberg A. The natural course of Meniere’s disease. Acta Otolaryngol Suppl 1984;406:72–7. Gacek RR, Gacek MR. Comparison of labyrinthectomy and vestibular neurectomy in the control of vertigo. Laryngoscope 1996;106:225–30. Stahle J, Friberg U, Svedberg A. Long-term progression of Meniere’s disease. Acta Otolaryngol Suppl 1991;485:78–83. Kitahara M. Bilateral aspects of Meniere’s disease. Meniere’s disease with bilateral fluctuant hearing loss. Acta Otolaryngol Suppl 1991;485:74–7. Silverstein H, Smouha E, Jones R. Natural history vs. surgery for Meniere’s disease. Otolaryngol Head Neck Surg 1989;100:6–16. Schatz A, Bugie E, Waksman SA. A substance exhibiting antibiotic activity gram-positive and gram-negative bacteria. Proc Soc Exp Biol Med 1944;55:66–9. Hinshaw HS, Feldman WH. Streptomycin treatment of clinical tuberculosis: a preliminary report. Proceedings of the Staff Meetings of the Mayo Clinic, Rochester. 1945;20:313–8. Fowler EP, Seligman E. Otic complications of stretomycin therapy. JAMA 1947;133:87–91. Jongkees LBW, Hulk J. The action of streptomycin on vestibular function. Acta Otolaryngol 1950; 38:225–32. Hawkins JE, Lurie MH. The ototoxicity of streptomycin. Ann Otol 1952;61:789–806. Serles W. Streptomycinschaden im Elektronystagmogramm. Wochenschr Ohrenkheilk 1966;100: 251. Graham MD. Bilateral Meniere’s disease. Treatment with intramuscular titration streptomycin sulfate. Otolaryngol Clin North Am 1997;30:1097–100. Fowler EP. Streptomycin treatment for vertigo. Trans Am Acad Ophthalmol Otolaryngol 1948; 52:293–301. Van Deinse JB. Medical treatment for Meniere’s disease. Ned Tijdschr Geneeskd 1949;31:2619–27. Hamberger CA, Hyden H, Koch H. Streptomycin bei der Meniereschen Krankheit. Arch Ohren Nasen Kehlkopfheilkd 1949;155:667–82. Hanson HV. The treatment of endolymphatic hydrops (Meniere’s disease) with streptomycin. Ann Otol Rhinol Laryngol 1951;60:676–91. Ruedi K. Therapeutic and toxic effect of streptomycin in otology. Laryngoscope 1951;61:613–36. Schuknecht HF. Ablation therapy for the relief of Meniere’s disease. Laryngoscope 1956;66:859–70. Schuknecht HF. Ablation therapy for the relief of Meniere’s disease. Acta Otolaryngol Suppl 1959; 132:14–42.

189

36. Singleton EF, Schuknecht HF. Streptomycin sulfate in the management of Meniere’s disease. Otolaryngol Clin North Am 1968;1:531–9. 37. Graybiel A, Schuknecht HF, Fregly AR, et al. Streptomycin in Meniere’s disease. Long-term follow-up. Arch Otolaryngol 1967;85:156–70. 38. Wilson WR, Schuknecht HF. Update on the use of streptomycin therapy for Meniere’s disease. Am J Otol 1980;2:108–11. 39. Graham MD, Sataloff RT, Kemink JL. Titration streptomycin therapy for bilateral Meniere’s disease: a preliminary report. Otolaryngol Head Neck Surg 1984;92:440–47. 40. Graham MD, Kemink JL. Titration streptomycin therapy for bilateral Meniere’s disease: a progress report. Am J Otol 1984;5:534–5. 41. Silverstein H. Streptomycin treatment for Meniere’s disease. Ann Otol Rhinol Laryngol Suppl 1984; 112:44–8. 42. Silverstein H, Hyman SM, Feldbaum J, Silverstein D. Use of streptomycin sulfate in the treatment of Meniere’s disease. Otolaryngol Head Neck Surg 1984;92:229–32. 43. Koegel L Jr. Ototoxicity: a contemporary review of aminoglycosides, loop diuretics, acetylsalicylic acid, quinine, erythromycin, and cisplatinum. Am J Otol 1985;6:190–9. 44. Wersall J, Hawkins JE. The vestibular sensory epithelia in the cat labyrinth and their reactions in chronic streptomycin intoxication. Acta Otolaryngol 1962;54:1–22. 45. Wanamaker HH, Gruenwald L, Damm KJ, et al. Dose-related vestibular and cochlear effects of transtympanic gentamicin. Am J Otol 1998;19:170–9. 46. Zheng Y, Schachern PA, Sone M, Paparella MM. Aminoglycoside ototoxicity. Otol Neurotol 2001; 22:266–8. 47. Hawkins JE. Vestibular ototoxicity. In: Naunton RF, editor. The vestibular system. New York: Academic Press; 1975. p. 321–49. 48. Nakai Y, Hilding D. Vestibular endolymphproducing epithelium. Electron microscopic study of the development and histochemistry of the dark cells of the crista ampullaris. Acta Otolaryngol 1968;66:120–8. 49. Sparwald E, Merck W, Leupe M. [Restitution of the dark and sensory cells of the guinea pig crista ampullaris after streptomycin intoxication]. Arch Klin Exp Ohren Nasen Kehlkopfheilkd 1973; 204:17–26. 50. Pender DJ. Gentamicin tympanoclysis: effects on the vestibular secretory cells. Am J Otolaryngol 1985;6:358–67. 51. Hellstrom S, Odkvist L. Pharmacologic labyrinthectomy. Otolaryngol Clin North Am 1994; 27:307–15.

190


52. Lange G. Gentamicin and other ototoxic antibiotics for the transtympanic treatment of Meniere’s disease. Arch Otorhinolaryngol 1989;246:269–70. 53. LaRouere MJ, Zappia JJ, Graham MD. Titration streptomycin therapy in Meniere’s disease: current concepts. Am J Otol 1993;14:474–7. 54. Balyan FR, Taibah A, De Donato G, et al. Titration streptomycin therapy in Meniere’s disease: long-term results.Otolaryngol Head Neck Surg 1998;118:261–6. 55. Glasscock ME 3rd, Jackson CG, Poe DS, Johnson GD. What I think of sac surgery in 1989. Am J Otol 1989;10:230–3. 56. House WF, Owens FD. Long-term results of endolymphatic subarachnoid shunt surgery in Meniere’s disease. J Laryngol Otol 1973;87:521–7. 57. Schuknecht HF. Pathology of Meniere’s disease as it relates to the sac and tack procedures. Ann Otol Rhinol Laryngol 1977;86:677–82. 58. Graham MD, Kemink JL. Surgical management of Meniere’s disease with endolymphatic sac decompression by wide bony decompression of the posterior fossa dura: technique and results. Laryngoscope 1984;94:680–3. 59. Fisch UP. Excision of Scarpa’s ganglion. Arch Otolaryngol 1973;97:147–9. 60. Langman AW, Kemink JL, Graham MD. Titration streptomycin therapy for bilateral Meniere’s disease. Follow-up report. Ann Otol Rhinol Laryngol 1990;99:923–6.

61. Berryhill WE, Graham MD. Chemical and physical labyrinthectomy for Meniere’s disease. Otolaryngol Clin North Am 2002;35:675–82. 62. Toth AA, Parnes LS. Intratympanic gentamicin therapy for Meniere’s disease: preliminary comparison of two regimens. J Otolaryngol 1995; 24:340–4. 63. Nedzelski JM, Chiong CM, Fradet G, et al. Intratympanic gentamicin instillation as treatment of unilateral Meniere’s disease: update of an ongoing study. Am J Otol 1993;14:278–82. 64. JC. Living without a balancing system. N Engl J Med 1952;246:458–60. 65. Jackson GG, Arcieri G. Ototoxicity of gentamicin in man: a survey and controlled analysis of clinical experience in the United States. J Infect Dis 1971;124 Suppl:130. 66. Black FO, Peterka RJ, Elardo SM. Vestibular reflex changes following aminoglycoside induced ototoxicity. Laryngoscope 1987;97:582–6. 67. Gunther H. Erfahrungen bei der Streptomycinbehandlung der Meniereschen Krankheit. Z Laryngol Rhinol Otol 1959;38:319–25. 68. Jatho K. Vestibulare Defektzustander nach Streptomycin Behandlung der Meniereschen Krankheit. HNO 1963;11:15–8. 69. Moretz WH Jr, Shea JJ Jr, Orchik DJ, et al. Streptomycin treatment in Meniere’s disease. Otolaryngol Head Neck Surg 1987;96:256–9.

CHAPTER 22

Intratympanic Gentamicin in the Treatment of Meniere’s Disease Brian W. Blakley, MD, PhD, FRCSC

The history of intratympanic aminoglycoside use is an interesting demonstration of how medical treatments are adopted. There are no large randomized controlled trials and few p values to show that this therapy is better than older methods. There are as many effective protocols as there are authors in the literature and no clear indication that any protocol is better than another. Still, the use of intratympanic aminoglycosides has essentially eliminated the need for surgery for vestibular disorders. This can be interpreted to represent a victory for good clinical judgment over decisions from the p values of evidence-based medicine. Ten or 15 years ago, this chapter would probably not have been published. In North America, treatment focused on vestibular neurectomy, labyrinthectomy, and endolymphatic sac surgery for Meniere’s disease when it did not respond to conservative therapy. The notion of placing toxic agents into the middle ear was considered risky because an optimal protocol had not been established. The optimal protocol has still to be established, but our understanding of topical aminoglycoside therapy in Meniere’s disease has certainly increased.

HISTORY Streptomycin was the first aminoglycoside developed. In 1944, it was marketed after validation by the first nationally coordinated, privately funded drug evaluation in history (see Chapter 8, “Clinical Aminoglycoside Ototoxicity”).1 The ototoxicity of streptomycin soon became apparent, but there was usually no other choice, so the drug became widely used. Fowler was the first to exploit this toxicity systemically to treat patients with Meniere’s disease. 2 In 1957, Schuknect used intratympanic streptomycin injections to treat Meniere’s disease in eight patients.3 Gentamicin was introduced to the market in 1964 and is currently the most widely used aminoglycoside. It is particularly popular in underdeveloped countries because it is inexpensive. Although intratympanic treatment had been used in Europe for some time,4,5 the technique was not

widely used in North America until Nedzelski and colleagues popularized it in the1990s.6,7 Intratympanic Aminoglycosides in the Treatment of Meniere’s Disease The current literature on intratympanic aminoglycosides is relatively large and growing. In 1997, 11 papers addressed the clinical methods of intratympanic aminoglycoside administration.8 By 2000, there were 18 papers.9 A literature review as of August 2003 identified 41 papers that addressed the clinical use of intratympanic aminoglycosides. Some general comments can be made. First, the use of intratympanic streptomycin has been abandoned in favor of intratympanic gentamicin (ITG). Possible reasons for this preference are that gentamicin causes less hearing loss, is less painful to inject, and is easier to obtain. The first two reasons do not have strong research support, and the recent increase in tuberculosis has resulted in greater availability of streptomycin. Nevertheless gentamicin is well established, so it will likely remain the intratympanic drug of choice. Almost all authors start with the 40 mg/mL stock solution of gentamicin, and most buffer it to give 20 to 27 mg/mL. The middle ear is filled, so only about 0.75 mL (or 20 mg of gentamicin) is used. Second, for every clinician’s philosophy of treatment there is an effective protocol. In medical literature, the average response or the most likely outcome has been traditionally used to characterize treatments or approaches. There is a growing realization in medical research that variability may be as important as the mean. Small variability suggests consistency. Large variability may be associated with uncertainty and lack of understanding, even if the mean is the same. Herein the impact of variability measured simply by the statistical standard deviation of the mean is considered. Third, widely varying protocols yield similar results. The usual dose–response relation that exists for systemic drugs does not appear to apply to ITG. This

192


Table 22-1 Control of Vertigo and Hearing Loss Percentage Vertigo Control (%) (from 41 Publications)

Percentage Hearing Loss (%) (from 33 Publications)

Mean (SD)

85 (8.7)

22 (19)

Range

56–100

0–95

Different authors used different criteria for vertigo control (from 41 publications) and hearing loss (from 33 publications). Some authors did not report hearing loss. In many papers the percentages were calculated from data provided in the paper. The standard deviation relative to the mean percentage for vertigo control (8.7 vs 85%) is much smaller than that for hearing loss (19 vs 22%). These data suggest more inconsistent results for hearing loss than for vertigo control.

bothers some clinicians, but why should there be only one valid paradigm? Fourth, not everyone is the same. No author has published perfect results for all patients. What does this tell us? Could it be that individual differences mandate different protocols? Can we ever chemically dissect vestibular from cochlear hair cells? Several genetically related special sensitivities to systemic aminoglycosides have been described. Are these important in ITG (see Chapter 17, “Genetic Factors in Aminoglycoside Ototoxicity”)? Questions about ITG in the Literature What is the “best way” to deliver ITG therapy? The literature on ITG contains a wide variety of methods, which all claim success. Not a single paper indicates that the ITG is not effective in controlling vertigo. Although some may find the variety of doses and methods confusing and upsetting, the message probably is that many protocols can be effective and have similar safety records. Table 22-1 contains summary data regarding the success of controlling vertigo and the percentage of hearing loss from the 41 papers reviewed.3,4,6,10–47 Some authors have published one or more updates of their experience. An attempt was made to include only those papers with actual data and eliminate redundant reports from the same authors. Weaknesses of the ITG literature include the following: 1. Different definitions of vertigo control and hearing loss are employed. Although it is optimal to use the classification of the American Academy of Otolaryngology-Head and Neck Surgery (AAO-HNS),48 several papers were published before these guidelines were available, and some of the best papers come from Europe, where these guidelines are less commonly used. Some authors simply did not use the guidelines. 2. There is no consensus regarding terminology, testing, outcome, administration method, or many other factors.

3. The studies are typically one person’s observations about his or her own patients reflecting his or her personal bias. “Cutting-edge” clinicians who reported unusually large numbers of patients may be overapplying the diagnosis, which invalidates their research results. Although it would be nice if all studies followed the same format, this is not the independent, creative nature of research. In fact, the inconsistencies in methods are probably not too important when 41 studies are available. For example, in Table 22-1 the percentage of vertigo control is around 85%, with a standard deviation that is only about 10% of the mean. This variability seems small, so the results are fairly consistent. For hearing loss the story is somewhat different. The standard deviation for hearing loss is almost as large as the mean. This implies much variability in hearing loss. Hearing loss is a main concern with ITG that needs attention. There appears to be no particular relationship between hearing loss and control of vertigo. Eleven of 41 papers indicated follow-up times greater than 1 year, and 10 of these had follow-up times greater than 2 years. Of these, five studies reported recurrence rates averaging 21%, with a standard deviation of 8.3%, that were treated successfully with ITG.

DOSE The “dose” of gentamicin is often thought of in the same manner as a systemic dose—a number of milligrams, concentration, or rate to be applied. Papers in the literature employed cumulative doses of gentamicin from 0.24 to 720 mg. Some of the tiny amounts reported as effective might be difficult to believe. Although commercially available topical gentamicin had been used for years in ears with perforations with few apparent problems, it is possible that these small amounts could cause problems. Imamura and Adams found that 2 mg of gentamicin applied to guinea pig round windows resulted in “overt degeneration,” was too destructive for research purposes, and had to be diluted.49 The “effective dose” seems to vary by greater than three orders of magnitude. Application of systemic philosophy to ITG is flawed. ITG administration

Intratympanic Gentamicin in the Treatment of Meniere’s Disease

should be considered as topical treatment. In topical skin treatments, for example, the “dose” consists of applications of medication concentrations empirically found to be effective. The treatment is repeated until the desired effect is achieved.

METHOD OF DELIVERY AND THERAPEUTIC END POINT The many basic delivery techniques can be classified as follows: 1. Direct injections (n = 25). Direct injection is defined as an injection with a needle through the tympanic membrane without any temporary device. Direct injections are the most popular and simplest of the delivery techniques. Advantages include simplicity and the ability to titrate therapy and stop treatment if needed. 2. Catheters and tympanostomy tube assemblies (n = 7) of various designs that were left in place during treatment without pump or other device. The gentamicin was delivered intermittently into the catheter or tube, and in two cases pressure was applied afterward. Sometimes the catheter was attached to the tube for each instillation; other times the catheter was left intact, and the patient wore the assembly between applications. Catheters and tubes were more prominent in papers for which total ablation was the goal. If a catheter was used, gentamicin was likely to be administered several times per day. Catheters appear more appropriate if total ablation is the goal because other end points are likely to be overshot. The effects of gentamicin on the cochlea may take weeks or even months to evolve.32,50,51 When the full effect develops, hearing loss is likely. 3. Pumps (n = 2) placed into the inner ear or on the round window membrane. Pumps included direct infusion into the scala tympani, permeable membranes contacting the round window, and, in one case, an insulininfusion pump. The pump placement requires surgical implantation, which most patients see as a disadvantage. It has been suggested that the round window should be cleaned or inspected to allow the gentamicin to enter properly, but the evidence that this is reasonable is lacking. One author recommended that the round window be obliterated before intratympanic therapy. The tendency is to administer a larger amount of gentamicin if a pump is used, as opposed to intermittent injections. This is acceptable when total ablation is the goal.

193

4. Surgical placement of sponge or wicks (n = 2) of a gentamicin-soaked wick or device directly against the round window without an additional pump was used by two authors. Surgery is required, which may be perceived as a disadvantage. Sponges, wicks, and pump applications do not easily permit titration or cessation of therapy if allergy, hearing loss, or other reasons arise. 5. There were five papers for which the method was not clear. Some authors hoped to ablate function. Others wished only to reduce vestibular function. The end point philosophies can be classified as follows: a.

b.

c.

d.

e.

First sign or symptom of inner ear disturbance (n = 8). Criteria included significant imbalance or hearing loss and rarely vertigo or nystagmus or some nonspecific clinical assessment. Of the five studies that reported relapse rates, all used the injection method. For three, the first sign of inner ear symptom was the end point, and for two the end point was unclear. Given time, vestibular and cochlear hair cells may regenerate, 52,54 and function may even recover.55,57 Cessation of spells (n = 3). This end point becomes problematic in patients who have infrequent spells. Table 22-2 shows that this was used only by authors who performed injections and did not require devices for a specified duration of treatment. Ablation (n = 5) as assessed by caloric testing was the goal for at least two papers. It is likely that others with unclear methods had total ablation in mind as well. Three papers contrasted total ablation with subtotal ablation. One paper reported that there was no difference in outcome, but the other two reported that ablated patients did better. Completion of a preset course (n = 8), such as a certain number of injections or insertion of a wick or sponge, which does not permit cessation of therapy. The number of injections to be employed and the amount of gentamicin in the wick was arbitrary. Considering the variability in hearing loss and the lack of evidence that more applications provide better vertigo control, employing a preset number of injections or treatments seems illogical. The final outcome was inconsistent or unclear in 20 papers.

194


Table 22-2 Methods of Delivery and Philosophy of End Point (n = 41) First Symptom

Cessation of Spells

Ablation

Preset Course

Unclear

Total Method

Injections

6

3

0

5

11

25

Catheter/tube

1

0

1

1

4

7

Pump

1

0

0

0

1

2

Sponge, wick

0

0

0

2

0

2

Unclear

0

0

1

0

4

5

Total by end point

8

3

2

8

20

41

The number of papers using various delivery methods and the philosophy for treatment end point for clinical papers that applied ITG to human patients are displayed. For papers that used more than one method or end point the largest number of patients used of each is shown. The method of delivery of gentamicin to the inner ear is partially dependent on the end point. Methods that are difficult to stop (wick, pump, sponge) are more likely to be used if total ablation is the goal of treatment.

Table 22-2 shows the combinations of method and end point. Analysis of variance with the Wilks lambda adjustment indicated no significant overall effect (p = .54) of method or end point on the percentage of either vertigo control or hearing loss. Although several authors used the repeated caloric testing or serial audiograms in end point determination, evidence supporting better results for objective testing over clinical assessment was absent or weak.

DISCUSSION ITG therapy is indicated essentially only for the acute attacks of Meniere’s disease. Chronic imbalance, tinnitus, aural fullness, and hearing loss are inappropriate uses for this therapy. It appears that differences in methods and end point of treatment may not be important determinants of clinical outcome. ITG results in control of vertigo spells in about 85% of patients, but the degree of hearing loss varies. By definition, hearing loss is present in Meniere’s disease, and it is expected to fluctuate. ITG does not seem to change that. Nevertheless, there are several reports of moderate losses becoming profound with ITG therapy. There are no reports yet that the systemic genetic susceptibilities to aminoglycosides58,59 have adverse effects in ITG therapy. The 1555G mutation affects hearing loss presumably by reducing the adenosine triphosphate generation process in mitochondria. In future, it is possible that protective agents may be concomitantly applied to try to reduce ITG-induced cochlear toxicity. Currently at least 30 compounds reduce ototoxicity for various agents. One or more of these agents will likely be useful to protect the auditory system and permit the desired effect on the vestibular hair cells. These protective agents are typically thiols (—SH) containing compounds that act as antioxidants. It is probably not a coincidence that oxygen and sulfur

are in the same chemical family in the periodic table. It is plausible that some of these antioxidants will have therapeutic effects on hair cells60 in addition to their protective effects against ototoxicity. The genetic susceptibilities are not changeable, but oxidation states may be. Probably when we better understand the biochemistry of oxidation in vivo, we will be able to exploit this knowledge for use in ITG, offering toxicity and protection in sophisticated ways to retain or augment cochlear function yet induce the hoped for vestibular deafferentation. Treatment of conservative failures of Meniere’s disease has progressed from primarily surgical to primarily medical in the past 10 to 15 years. This has been good for patients, and it should continue. Applied biochemical and nonsurgical therapy to the inner ear appears to be the way of the future.

CONCLUSION How do we decide which protocol is best? Philosophers often credit William of Ockham with first insisting that the simplest course is usually the best. If there are no advantages to other methods then the simplest, safest, and cheapest method is chosen.

SUMMARY • ITG treatment is currently indicated for Meniere’s disease that is significant, is disabling, and has failed conservative medical therapy as determined by good clinical judgment. • Before treatment it is important to counsel patients that ITG is destructive and that function cannot be regenerated once lost. The clinical recurrence rate may be about 20%, and retreatment may be required. Hearing may become worse in about 25 to 30% and can be profound in the affected ear.


• The main goal of ITG is long-term control of acute vertigo spells. Permanent or transient imbalance, tinnitus, or other complications such as pain or drug allergy may occur. • Injection techniques are best performed with the smallest spinal needle available through an aural speculum through the tympanic membrane, filling the middle ear, at weekly intervals or less often. Topical anesthetic such as injected lidocaine or topical phenol may be used but may cause as much discomfort as the actual needle. • ITG therapy should be stopped when the patient describes imbalance or hearing loss or some other new sign or symptom of inner ear dysfunction. • Quantitative measurements of cochleovestibular function should be obtained with audiometry and electronystagmographic calorics before and after treatment. • Follow-up for 1 year posttreatment would be recommended on clinical grounds.

11.

12.

13.

14.

15.

16.

REFERENCES 1. Weinstein L. Antimicrobial agents, streptomycin, gentamicin and other aminoglycosides. In: Goodman LS, Gilman A, editors. The pharmacological basis of therapeutics. 5th ed. New York: Macmillan; 1975. p. 1167–82. 2. Fowler E. Streptomycin treatment of vertigo. Trans Am Acad Ophthalmol Otolaryngol 1948;52: 239–301. 3. Schuknect H. Ablation therapy in Meniere’s disease. Acta Otolaryngol Suppl (Stockh) 1957;132: 1–42. 4. Lange G. [27 year experience with transtympanic aminoglycoside treatment of Meniere’s disease]. Laryngorhinootologie 1995;74:720–3. 5. Beck C, Schmidt CL. 10 years of experience with intratympanally applied streptomycin (gentamycin) in the therapy of Morbus Meniere. Arch Otorhinolaryngol 1978;221:149–52. 6. Nedzelski JM, Bryce GE, Pfleiderer AG. Treatment of Meniere’s disease with topical gentamicin: a preliminary report. J Otolaryngol 1992;21:95–101. 7. Nedzelski JM, Chiong CM, Fradet G, et al. Intratympanic gentamicin instillation as treatment of unilateral Meniere’s disease: update of an ongoing study. Am J Otol 1993:14:278–82. 8. Blakley BW. Clinical forum: a review of intratympanic therapy. Am J Otol 1997;18:520–6; discussion 527–31. 9. Blakley BW. Update on intratympanic gentamicin for Meniere’s disease. Laryngoscope 2000;110(2 Pt 1):236–40. 10. Abou-Halawa AS, Poe DS. Efficacy of increased gentamicin concentration for intratympanic

17.

18.

19.

20.

21.

22.

23. 24.

25.

195

injection therapy in Meniere’s disease. Otol Neurotol 2002;23:494–502; discussion 502–3. Atlas JT, Parnes LS. Intratympanic gentamicin titration therapy for intractable Meniere’s disease. Am J Otol 1999;20:357–63. Charabi S, Thomsen J, Tos M. Round window gentamicin mu-catheter—a new therapeutic tool in Meniere’s disease. Acta Otolaryngol Suppl 2000; 543:108–10. Corsten M, Marsan J, Schramm D, Robichaud J. Treatment of intractable Meniere’s disease with intratympanic gentamicin: review of the University of Ottawa experience. J Otolaryngol 1997; 26:361–4. Driscoll CL, Kasperbauer JL, Facer GW, et al. Lowdose intratympanic gentamicin and the treatment of Meniere’s disease: preliminary results. Laryngoscope 1997;107:83–9. Hirsch BE, Kamerer DB. Intratympanic gentamicin therapy for Meniere’s disease. Am J Otol 1997;18:44–51. Hoffer ME, Kopke RD, Weisskopf P, et al. Use of the round window microcatheter in the treatment of Meniere’s disease. Laryngoscope 2001;111 (11 Pt 1):2046–9. Hoffer ME, Kopke RD, Weisskopf P, et al. Microdose gentamicin administration via the round window microcatheter: results in patients with Meniere’s disease. Ann N Y Acad Sci 2001;942:46–51. Hoffmann F, Beck C, Stratulat S. [Subablative gentamycin therapy in Meniere’s disease]. HNO 1993;41:296–300. Hone SW, Nedzelski J, Chen J. Does intratympanic gentamicin treatment for Meniere’s disease cause complete vestibular ablation? J Otolaryngol 2000;29:83–7. Kaasinen S, Pyykko I, Ishizaki H, Aalto H. Intratympanic gentamicin in Meniere’s disease. Acta Otolaryngol 1998;118:294–8. Kaplan DM, Nedzelski JM, Chen JM, Shipp DB. Intratympanic gentamicin for the treatment of unilateral Meniere’s disease. Laryngoscope 2000; 110:1298–305. Kaplan DM, Hehar SS, Bance ML, Rutka JA. Intentional ablation of vestibular function using commercially available topical gentamicinbetamethasone eardrops in patients with Meniere’s disease: further evidence for topical eardrop ototoxicity. Laryngoscope 2002;112:689–95. Laitakari K. Intratympanic gentamycin in severe Meniere’s disease. Clin Otolaryngol 1990;15:545–8. Light JP, Silverstein H, Jackson LE. Gentamicin perfusion vestibular response and hearing loss. Otol Neurotol 2003;24:294–8. Magnusson M, Padoan S. Delayed onset of ototoxic effects of gentamicin in treatment of

196

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.


Meniere’s disease. Rationale for extremely low dose therapy. Acta Otolaryngol 1991;111:671–6. Marks S, Arenberg IK, Hoffer ME. [Round window microcatheter administered microdose of gentamycin: an alternative in the treatment of tinnitus in patients with Meniere’s disease]. Laryngorhinootologie 2000;79:327–31. Marzo SJ, Leonetti JP. Intratympanic gentamicin therapy for persistent vertigo after endolymphatic sac surgery. Otolaryngol Head Neck Surg 2002; 126:31–3. McFeely WJ, Singleton GT, Rodriguez FJ, Antonelli PJ. Intratympanic gentamicin treatment for Meniere’s disease. Otolaryngol Head Neck Surg 1998;118:589–96. Minor LB. Intratympanic gentamicin for control of vertigo in Meniere’s disease: vestibular signs that specify completion of therapy. Am J Otol 1999; 20:209–19. Mondain M, Mouchet F, Marlier F, et al. [Chemical labyrinthectomy: results and applications]. Ann Otolaryngol Chir Cervicofac 1998;115:234–42. Murofushi T, Halmagyi GM, Yavor RA. Intratympanic gentamicin in Meniere’s disease: results of therapy. Am J Otol 1997;18:52–7. Odkvist LM. Middle ear ototoxic treatment for inner ear disease. Acta Otolaryngol Suppl 1989; 457:83–6. Parnes LS, Riddell D. Irritative spontaneous nystagmus following intratympanic gentamicin for Meniere’s disease. Laryngoscope 1993;103:745–9. Perez N, Martin E, Garcia-Tapia R. Intratympanic gentamicin for intractable Meniere’s disease. Laryngoscope 2003;113:456–64. Pfleiderer AG. The current role of local intratympanic gentamicin therapy in the management of unilateral Meniere’s disease. Clin Otolaryngol 1998;23:34–41. Pyykko I, Ishizaki H, Kaasinen S, Aalto H. Intratympanic gentamicin in bilateral Meniere’s disease. Otolaryngol Head Neck Surg 1994;110:162–7. Quaranta A, Aloisi A, De Benedittis G, Scaringi A. Intratympanic therapy for Meniere’s disease. Highconcentration gentamicin with round-window protection. Ann N Y Acad Sci 1999;884:410–24. Rauch SD, Oas JG. Intratympanic gentamicin for treatment of intractable Meniere’s disease: a preliminary report. Laryngoscope 1997;107:49–55. Schoendorf J, Neugebauer P, Michel O. Continuous intratympanic infusion of gentamicin via a microcatheter in Meniere’s disease. Otolaryngol Head Neck Surg 2001;124:203–7. Silverstein H, Arruda J, Rosenberg SI, et al. Direct round window membrane application of gentamicin in the treatment of Meniere’s disease. Otolaryngol Head Neck Surg 1999;120:649–55.

41. Thomsen J, Charabi S, Tos M. Preliminary results of a new delivery system for gentamicin to the inner ear in patients with Meniere’s disease. Eur Arch Otorhinolaryngol 2000;257:362–5. 42. Toth AA, Parnes LS. Intratympanic gentamicin therapy for Meniere’s disease: preliminary comparison of two regimens. J Otolaryngol 1995; 24:340–4. 43. Watanabe S, Kato I, Takahashi K, et al. Indications and results of gentamycin injection into the middle ear of patients with meniere’s disease. Acta Otolaryngol Suppl 1995;519:282–5. 44. Yamazaki T, Hayashi M, Hayashi N, Kozaki H. Intratympanic gentamycin therapy for Meniere’s disease placed by tubal catheter with systemic isosorbide. Arch Otorhinolaryngol 1988;245:170–4. 45. Yamazaki T, Hayashi M, Komatsuzaki A. Intratympanic gentamicin therapy for Meniere’s disease placed by a tubal catheter with systematic isosorbide. Acta Otolaryngol Suppl 1991;481:613–6. 46. Youssef TF, Poe DS. Intratympanic gentamicin injection for the treatment of Meniere’s disease. Am J Otol 1998;19:435–42. 47. Wu IC, Minor LB. Long-term hearing outcome in patients receiving intratympanic gentamicin for Meniere’s disease. Laryngoscope 2003;113: 815–20. 48. Monsell EM. New and revised reporting guidelines from the Committee on Hearing and Equilibrium. American Academy of Otolaryngology-Head and Neck Surgery Foundation, Inc. Otolaryngol Head Neck Surg 1995;113:176–8. 49. Imamura S, Adams J. Distribution of gentamicin in guinea pig inner ear after local systemic application. J Assoc Res Otolaryngol 2003;4:176–95. 50. Imamura S, Adams J. Changes in cytochemistry of sensory and nonsensory cells in gentamicintreated cochleas. J Assoc Res Otolaryngol 2003; 4:196–218. 51. Lopez I, Honrubia V, Lee SC, et al. The protective effect of brain-derived neurotrophic factor after gentamicin ototoxicity. Am J Otol 1999;20: 317–24. 52. Lopez-Gonzalez MA, Guerrero JM, Torronteras R, et al. Ototoxicity caused by aminoglycosides is ameliorated by melatonin without interfering with the antibiotic capacity of the drugs. J Pineal Res 2000;28:26–33. 53. Husmann KR, Morgan AS, Girod DA, Durham D. Round window administration of gentamicin: a new method for the study of ototoxicity of cochlear hair cells. Hear Res 1998;125:109–19. 54. Warchol ME, Lambert PR, Goldstein BJ, et al. Regenerative proliferation in inner ear sensory epithelia from adult guinea pigs and humans. Science 1993;259:1619–22.


55. Black FO, Peterka RJ, Elardo SM. Vestibular reflex changes following aminoglycoside induced ototoxicity. Laryngoscope 1987;97:582–6. 56. De Waele C, Meguenni R, Freyss G, et al. Intratympanic gentamicin injections for Meniere disease: vestibular hair cell impairment and regeneration. Neurology 2002;59:1442–4. 57. Woolley SM, Wissman AM, Rubel EW. Hair cell regeneration and recovery of auditory thresholds following aminoglycoside ototoxicity in Bengalese finches. Hear Res 2001;153:181–95.

197

58. Guan MX, Fischel-Ghodsian N, Attardi G. A biochemical basis for the inherited susceptibility to aminoglycoside ototoxicity. Hum Mol Genet 2000;9:1787–93. 59. Van Camp G, Smith RJ. Maternally inherited hearing impairment. Clin Genet 2000;57:409–14. 60. Seidman MD. Effects of dietary restriction and antioxidants on presbyacusis. Laryngoscope 2000; 110(5 Pt 1):727–38.

Medicolegal Concerns CHAPTER 23

Medicolegal Aspects of Ototoxicity Peter Rhatican, JD, Sloan H. Mandel, LLB, and John A. Rutka, MD, FRCSC

Editors’ Note: This chapter is an introduction to the law as it applies regarding the concepts of negligence, standard of care, harm, and disclosure. Clinical cases of ototoxicity are presented to which Mr Peter Rhatican and Mr Sloan Mandel, both esteemed and well-respected trial lawyers in the United States and Canada, respectively, respond. Although their responses are in agreement for the most part, their dialogues demonstrate a diversity of opinion from a legal perspective; law, like medicine, has few absolutes. In order to provide full disclosure to the reader, our contributing lawyers were presented with information that might be considered the equivalent of an expert opinion that can be found in Appendix 1. Although the law occasionally differs between these countries, overall, one is more impressed with the similarities across the borders rather than the differences.

MEDICOLEGAL CONCEPTS The majority of medical malpractice suits against physicians are based on a claim of negligence. Although the law does not demand perfection, allegations of negligence extend not only to acts the physician is said to have committed in error but also to the steps it is suggested the physician should have taken but failed to take. The alleged omission on the part of the physician usually constitutes the majority of claims for negligence. For a medicolegal action based on a claim of negligence to be successful, four elements must be established or proven1: 1. There must be a duty of care owed to the patient. 2. There must be a breach of duty of that care. 3. The patient must have suffered some harm or injury. 4. The breach of duty of care must have caused or materially contributed to the harm or injury.

STANDARD OF CARE Although physicians are not expected to be correct every time, they are expected to exercise reasonable care, skill, and judgment in arriving at a diagnosis. The physician is also required to properly treat the patient in accordance with current and accepted standards of care, refer the patient, or obtain a consultation when the patient is not responding to treatment or is beyond the competence of the treating physician. The physician is also expected to arrange for coverage when absent and to adequately instruct patients about their treatment and follow-up care. In order to determine if there has been a breach of duty to the patient the courts have generally considered the standard of care to be that required of a reasonable prudent physician colleague in similar circumstances. It is important to realize that medical science has not yet reached the stage where the law ought to presume that all treatment afforded a patient must have a successful outcome and that anything less suggests negligence. The appropriate measure is, therefore, the level of reasonableness and not a standard of perfection. Breach of duty or care toward the patient is usually not enough to demonstrate negligence. It must be demonstrated that the patient suffered some harm or injury as a result of the actions or inactions of the physician. For the purpose of definition, “Harm is defined broadly as an unexpected or normally avoidable outcome that negatively affects the patient’s health and/or quality of life which normally occurs (or occurred) in the course of health care treatment and is not due directly to the patient’s illness.”2 Medicolegal actions are never entirely black and white because there are usually several mitigating circumstances, including other factors that could have caused or contributed to the same result for which the physician could not be faulted. For this reason the plaintiff must therefore establish on a balance of probability

Medicolegal Aspects of Ototoxicity

(ie, within reasonable medical certainty) that the alleged breach of duty contributed to the injury sustained. Defining “Standard of Care” Sloan Mandel Regardless of whether you live in the United States or Canada, the legal principles involved are plain enough but are not always easy to apply to a particular set of circumstances. Every medical practitioner must bring to the task a reasonable degree of skill and knowledge and must exercise a reasonable degree of care. Peter Rhatican In addition, if we further define the standard of care as a species of obligation, when the health care provider enters into a professional relationship with the patient, the obligation assumed by the provider is to exercise in treating the patient the degree of care, knowledge, and skill ordinarily possessed and exercised in similar situations by a normal, prudent practitioner of the same experience and standing. The definition of this duty is surprisingly not what was learned in medical school or, for that matter, what is contained in medical books or treatises within the field. Medicine is active as well as it is traditional. Although anatomy may not change, the technology, methods, and procedures applied to that anatomy do. Because of these advances from research, clinical studies, and the experimental components inherent in the field of medicine, one must look to the periodicals and journals to witness and discern the evolving nature of one’s field. More importantly the standard of care is often found within the interaction and interplay that occurs daily between health care providers. The sharing of information with peers in the professional setting or quasi-educational environment at continuing medical education lectures and seminars provides the opportunity to familiarize one with the current thinking in the field. “Standard of Care” versus “Community Standard of Care” Sloan Mandel Some defense counsel have suggested that a lesser standard of care is required from a rural practitioner than from a physician practicing in an urban center. To the extent that such a premise exists (which is debatable), the submission is no longer well received by our courts. If the physician in a rural setting has less access to necessary resources, this may be considered by the court. This does not imply that a lesser standard of care is required. Further, historical disadvantages to the rural physician have been greatly reduced by advances in transportation (eg, air ambulance), technology (eg, the Internet), and timely access to specialized care.

199

Peter Rhatican When the “standard” is at issue in court, the standard of care is initially expressed by “an expert.” For the expert’s opinion to be acceptable, testimony must relate to generally accepted medical standards, not merely to the standards personal to the expert. Community standards of care are not and should not be distinguished from the generally accepted standard of medical practice. Knowledge, experience, and training know no political boundaries. The combination of personal experience, learned knowledge from persons with adequate training and experience, and current information imparted through recognized periodicals, journals, and updated texts provides to the practitioner the only boundaries (or evidence) of what constitutes the accepted conduct within the discipline of medicine. The community standard concept conjures the notion that Los Angeles, California, has a different standard from Toronto, Ontario. This image does not belong in the minds of providers and should be disregarded. Nevertheless, we should never lose sight that although medicine is not an exact science, there are common understandings relative to its practice. The standard of care is measured neither by the personal opinion of the providers nor by the behavior of providers in their practice community. Rather, a universal approach is preferred over the limitation established by personal views and political boundaries. The generally accepted standard of care does not exist independently of the health care provider’s exercise of “reasonable judgment.” An evaluation of a provider’s reasonable judgment must be measured by an objective standard and not based on a provider’s good faith or honesty. Motivation has no role in determining negligence. For the exercise of judgment to meet the required standard of care the following circumstances must be met: 1. The selection of a differential diagnosis from those contained in a medically reasonable differential 2. The selection among acceptable and medically reasonable causes of treatment that applies to the presenting diagnosis. Once the selection of a reasonable diagnosis and medically reasonable course of treatment is made then the application of knowledge, care, and skill of the provider becomes the hallmark in determining whether or not there exists a breach of duty to the patient. For example, no issue relating to the exercise of reasonable judgment exists once the acceptable surgical procedure is selected should a surgical mishap occur or where medication is selected from recognized alternatives should the patient sustain an adverse reaction. The standard of care is objective, whereas the exercise of

200

Medicolegal Concerns

judgment takes into account choices that fall within the reasonable medical options.

DISCLOSURE When a physician becomes aware while treating a patient that a harm has occurred in the course of receiving treatment that can be reasonably be expected to affect the patient’s health or quality of life, then it is the physician’s obligation to inform the patient about the harm sustained. Reluctance to disclose such information is understandable; the physician may feel guilt and shame, have concerns that the patient will become more distressed and lose confidence in the physician’s competence, and fear legal repercussions. Although more disclosure may mean there will be some lawsuits that otherwise might never see the light of day, a recent study suggests that patient dissatisfaction about medical harm is frequently generated by physician attitude and denial rather than the adverse effect itself.3 The majority of patients want a simple and honest explanation of what happened and, where appropriate, an apology. When patients get neither or are rejected they feel doubly wronged and not infrequently seek legal counsel. Ototoxicity Ototoxicity can be broadly defined as the tendency of certain substances to cause functional impairment and cellular damage to the tissues of the inner ear and especially to the end organs of the cochlear and vestibular divisions of the 8th cranial nerve that can occur from systemic or topical administration.4 The following medicolegal cases illustrate ototoxicity that occurred during the treatment of a medical condition. Case 1: Systemic Gentamicin Ototoxicity In September 1995 a 63-year-old white female with type 2 non–insulin-dependent diabetes was admitted to hospital for the first time with a cellulitis of her right foot and a suspected osteomyelitis of her right fourth toe that followed minor trauma. There was a previous history of peripheral vascular disease, hypercholesterolemia, and hypertension. There was general consensus among the residents and the attending staff internist to treat her infection aggressively intravenously for 6 weeks with triple antibiotic therapy, which included cefazolin, metronidazole, and gentamicin. A direct aspirate from the infected interphalangeal joint came back 72 hours later demonstrating 3+ β-hemolytic streptococci sensitive to cefazolin, metronidazole, gentamicin, and penicillin, among others. While in hospital her clinical condition improved on treatment. Parameters of renal function and moni-

tored serum gentamicin levels remained within the expected range, and she was discharged home 2 weeks later on clindamycin 600 mg PO bid and gentamicin 260 mg once daily to be provided intravenous by a visiting nurse. The family physician agreed to assume care of the patient at this point. Arrangements were made for weekly serum creatinine and serum gentamicin trough levels to be taken and for the family physician to be informed of these results. Between days 15 and 28 of treatment, serum creatinine levels rose from 66 to 122 mmol/L (normal 50 to 115 mmol/L). Serum trough gentamicin levels rose from 2.0 to 3.4 mg/L (normal < 2 mg/L). Nothing was done to alter the treatment regimen. On treatment day 29 the patient vocalized complaints of generalized “dizziness” to the visiting nurse. Despite repeated complaints that something was “not right,” treatment continued until day 35 when she became acutely imbalanced and ataxic. The patient now complained of visual blurring, her horizon appeared unsteady, and she was unable to stand. No hearing loss was noted subjectively. The gentamicin was subsequently discontinued by the family physician. Unfortunately, her condition has not improved. When assessed 2 years later the patient continues to be wheelchair bound despite having undergone attempts at vestibular rehabilitation. Her clinical examination points to a bilateral vestibular loss (bilateral positive head thrust maneuver,5 positive oscillopsia test,6 gait ataxia, etc), which has been confirmed with electronystagmography (ENG) calorics. A bilateral caloric reduction indicative of a bilateral peripheral (inner ear) vestibular loss has been identified. An intracranial magnetic resonance imaging (MRI) scan was unremarkable. Upon questioning, the patient denies having been told about the adverse effects of the medications she had been placed on and the risks of treatment. Evidence of Medical Negligence Peter Rhatican I believe there is evidence of medical negligence here. Arguably the family physician and the visiting nurse deviated from recognized standards applicable to their respective disciplines. The hospital and its staff internist and residents did not appear to deviate from reasonable medical standards. The hospital, of course, would be liable only under the “respondent superior” theory: namely, the employer takes a hit for its employees. Sloan Mandel Given this particular patient’s case history, there are facts that would be considered by appropriately qualified medical experts to support a finding of medical negligence. Osteomyelitis can be a severe and limb-threatening infection (taking into account the heightened risk this


patient’s diabetes and peripheral vasculature posed); it would have been appropriate for the medical team to treat this infection aggressively. There is no question that they would have been criticized if they had not. Until the cultures came back and the strain of bacteria was identified, it was appropriate to treat this woman’s infection with triple antibiotic therapy. However, when it was known 72 hours later that the bacteria was sensitive to a variety of antibiotics (including cefazolin, metronidazole, penicillin, and gentamicin) there ought to have been a review regarding what antibiotic therapy was best suited to treat this particular patient’s infection. Following that review the most responsible physician ought to have been consulted and the patient advised of the risks and benefits associated with the various alternatives. If the physician were prudent, the discussion would have been charted with reasonable detail. The failure to discuss the various risks and benefits associated with the treatment alternatives might possibly give rise to a “double-barreled” negligence claim. The patient might allege, for example, that the hospital physicians failed to obtain her informed consent when prescribing gentamicin therapy, especially after the microbiology results returned 72 hours later. Apart from the failure to obtain an informed consent, should the patient allege that the treatment team breached the standard of care and that the breach resulted in avoidable harm, the court would very likely consider the following facts: • The patient’s underlying condition typically required 6 weeks of antibiotic therapy. • There were less risky treatments available without risk for ototoxicity. • The patient was not advised about those symptoms that may develop as a result of gentamicin therapy. • There is reportedly a 1 to 2% overall risk for ototoxicity when a patient receives systemic gentamicin therapy. The risk is relative, however, and significantly increases when gentamicin therapy is in excess of 14 days and when evidence of renal impairment occurs that is typically associated with elevated serum trough (preinjection) gentamicin levels. Peter Rhatican The family physician and visiting nurse take on a different standard relative to the influence their conduct had on the ultimate injury sustained by the patient. As a baseline, the family physician had a duty to obtain the prior laboratory results from the hospital. After that the family physician had the duty to monitor the weekly reports for any fluctuation to detect trends toward toxicity. The facts reveal significant changes over the ensuing 2 weeks (weeks 3 and 4 of treatment),

201

namely the serum creatinine levels rose from 66 to 122 mmol/L (normal 50 to 114 mmol/L) and the serum trough gentamicin levels rose from 2.0 to 3.4 mg/L (theraputic range to minimize ototoxicity would be < 2 mg/L). These rising levels should have alerted the family physician that the patient was slipping into renal failure and was certainly at a heightened risk for ototoxicity. The experience, knowledge, and skill ordinarily possessed and exercised in similar situations by the average member of the medical profession mandated that the family physician alter the treatment or in the alternative consult on an urgent basis the internist who had initially prescribed the therapy or perhaps a nephrologist and an otolaryngologist in order to determine the danger the patient faced. The failure of the family physician to act suggests a departure from accepted standards. In analysis of the doctor’s culpability one must recognize the maxim that you cannot act on something you are unaware of. It has been established that the visiting nurse, either out of ignorance or forgetfulness, failed to make known to the physician the patient’s complaints of generalized “dizziness” on day 29 of treatment. Complaints of acute imbalance, ataxia, an unsteady horizon, and visual blurring with head movement, all symptoms of a bilateral vestibular loss, presented in the fifth week (day 35 of treatment). The family physician finally responded to the gentamicin menace and appropriately discontinued the drug. One wonders if events might have unfolded differently if the visiting nurse had stressed to the family physician her concern about the patient when she first began experiencing dizziness. Although the profession of nursing has continued to expand and evolve, it primarily remains a profession of fact gatherers and fact reporters. Notwithstanding, the ultimate interpretation and conclusion relative to those facts still remain with the family physician. In my opinion the failure to gather the facts or, having gathered the facts, failing to report these facts to the most responsible physician levies a departure from accepted nursing standards. The nurse’s responsibility in this case was to administer the gentamicin, but implicit in this act, the nurse was also required to be aware of gentamicin’s adverse effects, especially those pertaining to ototoxicity and nephrotoxicity. For this reason the nurse would also be held accountable for what happened. Sloan Mandel One of the reasons patients may be discharged from hospital to home nursing care has to do with cost. It is certainly cheaper to treat a patient at home than in hospital. Presumably if this patient had been kept in hospital one might reasonably argue that she would have been monitored more diligently (as she appeared

202


to be during the first 2 weeks of treatment). There might have been a few more checks and balances in place that could have allowed for a timely discontinuation of her gentamicin before toxicity occurred and to minimize the development of her bilateral vestibular loss. Although other forms of therapy (including hospitalization) might be more expensive, cost containment is not a defense for inappropriate care. In the leading American authority of Wickline v. the State of California, the California Court of Appeal stated the following: While we recognize, realistically, that cost consciousness has become a permanent feature of the health care system, it is essential that cost limitation programs not be permitted to corrupt medical judgment. In the Canadian case of Law Estate v. Simice, the British Columbia Superior Court noted the following: If it comes to a choice between a physician’s responsibility to his or her individual patient and his or her responsibility to the medicare system overall, the former (the patient) must take precedence in a case such as this. Sending this patient out into the community without the appropriate infrastructure and reporting network to react timely to changes indicating pending ototoxicity was dangerous. Any savings from an extended hospitalization have certainly been lost by the cost to society of this patient’s current clinical status, not to mention the costs involved in a medical malpractice suit. Legal Research Sloan Mandel After having acquired the available medical records (eg, from past hospitalizations and other treating physicians) and receiving information from the patient directly, the first step in any medical malpractice investigation would be to retain an appropriately qualified medical expert to review the available information and to provide guidance to counsel. Often the medical expert will be in a position to refer you to authoritative medical texts, literature, and case studies. Within your office, nonmedical personnel may also conduct research via the Internet, medical libraries, and reported legal cases. Peter Rhatican The initial interview with the client and family is very extensive. One records not only the relevant personal information but also the client’s past medical history, including major illnesses, major hospitalizations, previous treating physicians, medications (past and present), and family medical history.

Your expert on liability in this case of systemic gentamicin ototoxicity must be familiar with standards applicable to infectious diseases and internal medicine generally (if we assume the staff physician at the hospital was an internist). In order to flesh out the extent of the damages and make sure the damages are not attributable to other medical conditions it would be necessary to have the damages assessed by specialists. For example, an infectious diseases expert could address the relationship of her osteomyelitis with the patient’s underlying diabetes and peripheral vascular disease. The recognized and medically reasonable courses of treatment coupled with the potential adverse reactions to these treatment modalities are better explained by those who specialize in that area. Clearly, an opinion regarding symptomatic bilateral vestibular loss would be appropriate from a neurologist or an otoneurologist. Both would be in the best position to educate the judge or jury in matters involving the diagnosis, treatment, and prognosis as well as how as a practical matter the resultant disability impacts on the individual’s quality of life and for how long. Responsibility of Health Care Professionals Peter Rhatican Clearly all the health care professionals in this case (ie, the hospital physicians, the family physician, and the visiting nurse) shared in the responsibility for the health and welfare of the patient. Nevertheless, the theories of liability differ somewhat (ie, informed consent vs medical liability), and in that difference you will find a differential in the apportionment of liability. Sloan Mandel As a general rule when addressing the standard of care of the involved health care professionals, the court generally wishes to compare apples with apples and oranges with oranges. More particularly, if we agree that the family practitioner breached the standard of care then expert evidence should be given by a family practitioner to address the care provided (eg, Were levels requisitioned with appropriate frequency? Was there timely follow-up once levels were taken?). In this case the internist who prescribed gentamicin after 72 hours (when the culture results came back demonstrating a bacterial organism sensitive to less toxic antibiotics) would also be a target defendant in my opinion. A physician of equal qualification should be retained to comment upon the quality of care provided by this individual. The laboratory that measured the serum creatinine and gentamicin trough levels may also be a potential target (ie, how quickly did the laboratory respond in notifying the treating physician of the elevated levels, and what recommendations were given, if any?)


A pharmacologist or toxicologist might also be appropriate to retain. A nurse should also be retained to address the quality of care provided by the visiting nurse and to review the nursing protocols for intravenous aminoglycoside delivery in a home setting (ie, Was she aware or should she have been aware of the elevated levels, and what should she have done when the patient vocalized complaints of dizziness?). Recommendations for Physicians and Allied Health Care Professionals Sloan Mandel My recommendations to the health care team would be as follows: • Know and discuss with your patient the risks and benefits associated with different treatment alternatives. • Review your recommendations when new information is known about the patient’s condition. • If the family physician agrees to assume outpatient care, then the family physician must ensure that the patient understands the risks associated with the proposed care and that the patient is appropriately monitored (through timely and frequent contact with the patient, the laboratory, and the nurse). • The family physician and the nurse should be knowledgeable about the risks associated with the therapy proposed and the signs and symptoms of potential concern. • The laboratory should have a written protocol regarding what efforts are to be taken when abnormal levels are discovered. Case 2: Topical Gentamicin Drop Ototoxicity A 48-year-old white male ironworker involved in skyscraper construction inadvertently perforated his left tympanic membrane (TM) with a cotton swab while cleaning his ear in March 1997. At the time he experienced some pain and a little bleeding from the ear canal. He was aware of a mild hearing loss but had no dizziness or imbalance. He saw an otolaryngologist 3 days later having developed a slightly purulent discharge from the ear. On examination the otolaryngologist noted a safe central TM perforation. The ear canal appeared moist, and a scant amount of discharge was noted through the perforation. A diagnosis of a traumatic central TM perforation was made with secondary infection. Audiometry revealed a mild right conductive hearing loss in the left ear; the sensorineural (inner ear) reserve appeared normal. The patient was prescribed an oral aminopenicillin and two standard bottles of commercially available topical gentamicin sulfate drops (3 mg/mL). He

203

was asked to use the drops, 3 drops tid for 7 days, and then to stop. The patient used the drops as recommended but continued of his own volition to use them beyond 7 days. The discharge had stopped 4 days into treatment. Drops continued to be used until day 15 of treatment, when he became acutely vertiginous, nauseated, and imbalanced. The patient stopped the drops and was reviewed 2 weeks later by the same otolaryngologist who had originally treated him. Examination now revealed the TM perforation to have healed. The otoneurologic examination was normal apart from a positive left high-frequency, head thrust maneuver5 and right-beating post–head-shake nystagmus.7 Repeat audiometry was essentially normal apart from a slight sensorineural loss at 8000 Hz in the left ear. The previous mild conductive hearing loss had resolved. An electronystagmogram revealed an absent left caloric response. An intracranial MRI scan was normal. Although his balance slowly improved, he was unable to return to his job as an ironworker for 6 months. At present his only symptoms appear to be slight unsteadiness when moving his head quickly. Patient Harm Leading to Medicolegal Action Peter Rhatican There is no question the patient suffered harm as a result of the treatment he received. However, the clinical indications (ie, the development of a secondary infection following a traumatic TM perforation) supported the treatment that was prescribed. Even if topical fluoroquinolone drops without risk of ototoxicity had been widely available at the time (which they were not in 1997), the dosage and administration of the prescribed aminoglycoside in this case history would be considered reasonable. The impact of the patient’s noncompliance needs to be considered. When the patient did not discontinue the drops after the purulent discharge stopped, against an amended medical order, he played brinkmanship with his health and contributed significantly to his ultimate injury. Failure to follow the instructions of the prescribing physician (whether not taking or extending its use) is still considered noncompliance. In my opinion it was the physician’s reasonable medical judgment that insulated against liability, not necessarily the outcome of the treatment chosen. In this case there is no question that the patient’s sequelae were related more to the duration the drops were used, not to the initial choice of medication. Sloan Mandel I add one practical consideration. Although the patient suffered harm, if we are to fully assume that he was fully capable of returning to work, driving a car, and engag-

204


ing in most of his premorbid activities, then the harm this patient has suffered may not justify the costs and risks associated with pursuing a medicolegal action. Patient’s Case for Physician Responsibility Peter Rhatican When the physician gives instructions to a patient who is capable of understanding the instructions, the physician has the right to expect the patient’s compliance. One might argue that the physician failed to educate the patient about the risks for topical aminoglycoside ototoxicity, but in 1997 the world literature still quoted a low apparent relative risk of 1 in 10,000 for its development. Nevertheless, the physician did not mention this to the patient, nor was a follow-up arranged for at the end of the prescribed treatment course—both of which might have prevented the development of topical ototoxicity. If I was arguing this case on behalf of the patient, I would plead that the drops were used in what would have been considered an “off-label” indication (ie, in the treatment of middle ear infection) that did not have US Food and Drug Administration approval and that the physician failed to disclose what harm noncompliance could have resulted in. The “devil is in the details,” and it is important to remember that the communication and follow-up stages of treatment are no less compulsory in the physician–patient relationship. These pleas might represent fertile ground for building a case against the treating physician. Sloan Mandel One would also submit that the treating physician ought not to have prescribed a quantum of medication that exceeded the 7-day dosage. In doing so the physician unnecessarily exposed the patient to potentially avoidable harm. Standard of the Times Sloan Mandel With respect to the issue of “standard of care,” the year in which this case history occurred is of significant importance. By 2000, topical fluoroquinolone drops were widely available and were known to be equally as effective as topical gentamicin, without the risk of ototoxicity. One could argue that it was a breach of care not to prescribe (or at least discuss the use of) the less risky form of therapeutic treatment. Peter Rhatican Conversely, the generally recognized standard of care is not necessarily time bound. By 1997–98 there was growing recognition, if not necessarily widespread acceptance, within the entire medical community that topical gentamicin drops reaching the middle ear could

result in clinical ototoxicity, for example. Around this time fluoroquinolone drops were also being introduced as an effective alternative to the aminoglycosides without risk of ototoxicity. The question of timing ultimately begs the question: Were the standards of care met when advising and treating the patient? The generally expected scope of physician knowledge as to the specific injury potential of topical aminoglycosides (vestibular vs cochlear) is a material fact that would certainly need to be ascertained relative to the year. Case 3: Topical Toxicity from Surgical Disinfectants and Preparatory Solutions In an effort to standardize surgical preparatory solutions essentially as a cost-saving measure, the hospital in question requested that all surgeons move to a 2% chlorhexidine solution. The change went through the standard process, which involved a literature review, and was passed by the pharmacy committee at the hospital, which included representatives from every surgical department or division before implementation. Otolaryngologists on staff at the hospital were informed of the change by the head of the department in a written communication. An aqueous solution of iodine had been used previously for ear surgery. Approximately 3 months later, staff otolaryngologists identified that six of their patients had sustained a profound unilateral sensorineural hearing loss following routine tympanoplasty surgery. During the same time frame, 15 similar operations had been performed. The discovery took place during scheduled departmental morbidity and mortality rounds. In the previous 5 years there had been no sensorineural hearing loss documented as directly related to the procedure itself. As concerned physicians they immediately discontinued the use of chlorhexidine and disclosed to all the patients by writing who had undergone tympanoplasty surgery during that time frame their concern that any hearing impairment postsurgery might have resulted from the surgical preparatory solution used. Of interest, the one surgeon who routinely used a temporary barrier (cotton batten) to prevent the preparatory solution from entering the middle ear had no reported cases of sensorineural hearing loss. Preventable Profound Hearing Loss Peter Rhatican Logic dictates that where there is evidence in the world literature of extreme inner ear toxicity in mammals to chlorhexidine and a few studies in humans implicating chlorhexidine with profound unilateral sensorineural deafness, there is little to support its appropriateness in otologic surgery.


The hearing loss was preventable. The physicians could have certainly decided it was not in the best interest of the patient to change the old protocol for the surgical preparatory solutions. Having changed the protocol, however, an instruction for the use of a temporary barrier in those cases involving perforation of the TM should have been mandated. Their inability to recognize the inherent toxicity of chlorhexidine put the whole group of patients at risk for total hearing loss in the presence of a TM perforation, which was unlikely to cause the same complication on its own. Apportioning Blame Sloan Mandel Although the hospital, the surgeons, and the members of the pharmacy committee would all be named as defendants to the litigation, the hospital and the members of the pharmacy committee would be the primary targets. If the surgeon was following hospital protocol (a protocol, I might add, that would need to have been approved by the pharmacy committee at the hospital), it might be difficult to fault the surgeon given this particular scenario. It may be said, however, that the surgeon ought to have used a temporary barrier. Nevertheless, it is not clear as to whether or not the failure to do so would be considered a breach of care. Peter Rhatican There is no question the hospital must answer for its medical decisions, and it certainly has the right to rely upon its committees in making changes in pharmaceutical protocols. Nonetheless, the hospital has a corresponding duty to the patients by making the hospital a safe facility for the treatment it rendered. The otolaryngologists must certainly shoulder some of the responsibility as they acted in concert with the hospital in establishing a protocol best suited for the hospital financially. By defective reasoning, they placed all surgical patients with a TM defect to a greater risk of inner ear injury than from the untreated natural history. Although they would likely be insulated from individual liability by virtue of the fact they were providing a service to the hospital most likely in compliance with their contract for surgical privileges, the fact that as a group they claimed surprise to find six patients with profound sensorineural hearing loss after the use of chlorhexidine was approved is indicative of their joint failure to raise objections on behalf of their joint patients. Appropriateness of Surgeon Warning Sloan Mandel It was the ethical and fiduciary obligation of the surgeons to issue a warning once discovered. If the hospital had reasonable assurances from the surgeons that

205

each potentially affected patient would receive an appropriate warning, then there would have been no reason to issue a second warning. Peter Rhatican In general the premise that a cost-saving effort was implemented at the cost of increasing the risk of permanent injury to a class of patients is never a satisfactory substitute for good medicine. If this premise were true, then surely a warning to the patient by both the surgeon and the hospital was warranted. Had the surgeon recognized the potential for chlorhexidine toxicity, then timely actions could have been taken to reduce or eliminate the risk (ie, using a temporary barrier). Case 4: Cisplatin Ototoxicity A 25-year-old male university student was recently diagnosed with metastatic testicular carcinoma. Investigations confirmed liver involvement. He underwent a right orchiectomy and following surgery was scheduled for combination chemotherapy, which included high-dose cisplatin at 75 mg/m2 for 5 days with three repeat cycles. He was advised by his oncologists that the adverse effects of cisplatin therapy include nephrotoxicity, neuropathy, and hearing loss from cochlear injury. Prechemotherapeutic audiometry appeared normal. Within 48 hours of chemotherapy he was identified to have a significant hearing loss of more than 15 dB between 4000 and 8000 Hz and noticed some difficulty hearing in competing background noise. Frank discussions ensued with his oncologists, and he was advised that although his cisplatin dosage could be reduced it could conversely affect the expected remission rate (ie, this cancer was definitely curable). The patient decided to continue his present treatment course, and his hearing continued to further deteriorate. His cancer appeared to respond, and remission was achieved after the cycles of chemotherapy had been completed. His hearing further deteriorated, and he was left with a 40 to 60 dB sensorineural hearing loss over a 2000 to 8000 Hz range. Aural rehabilitation was arranged, and he was fitted with hearing aids. He is able to continue his university studies but continues to have difficulty in competing background noise situations. Patient Grounds for Negligence Sloan Mandel Before undertaking a preliminary liability investigation, I would not offer an opinion regarding the success of the claim. In order to rule out the possibility of negligence, it would be necessary to undertake an appropriate investigation at significant cost. Hearing loss seems to be an unfortunate but generally accepted risk of the necessary treatment, so the patient would not be successful in his claim. However,

206


if the patient wished to pursue and fund the cost of an investigation, then I would ensure that the appropriate experts are retained to provide the necessary opinions.

disclosing the potential harm to hearing, on balance, a reasonably prudent patient would most likely have accepted the treatment.

Peter Rhatican The use of cisplatin would have been considered necessary by the oncologist to treat a life-threatening malignancy. The toxic side effects would have been generally accepted within that setting. The notion that the physician may be held liable under these circumstances is unsupportable. The physician fully explained the deleterious side effects of the chemotherapy, including the potential for hearing loss from cochlear injury. The physician also revealed to the patient that once the hearing loss manifested itself, reducing the dosage of the chemotherapeutic agent would likely reduce the success of remission. The patient actively participated in his therapy management to the extent of recognizing the downside of hearing loss in this race for survival. There are no viable grounds to assert that there was a deviation from recognized standards of medical care. There is no lack of informed consent. In this difficult condition, both the physician and the patient became painfully aware of the sacrifices that might be necessary to survive this imminently treatable malignancy.

Sloan Mandel Even if this patient had not been advised about the potentially cochleotoxic effects of cisplatin, I doubt he would be successful in his claim. In determining whether or not a patient will be successful in an informed consent case the Canadian courts apply a modified objective test. More particularly a judge or jury will be asked to determine what a reasonable person in this patient’s particular position would have done knowing the risks involved. In this particular circumstance a court or jury would place great weight on the fact that the underlying condition, if untreated, is terminal. Provided that there were not other less risky equally efficacious treatment options available, it is likely that the “reasonable patient” would have elected cisplatin treatment in any event.

Informed Consent and Medicolegal Action: What if Hearing Loss due to Treatment Had Not Been Discussed? Peter Rhatican In my home state of New Jersey there would be grounds for a medicolegal action. The New Jersey Supreme Court has explained that the foundation for the physician’s duty to disclose in the first place is found in the idea that it is the prerogative of the patient, not the physician, to determine for himself the direction in which his interests seem to be.8 The choice of the treatment plan is for the patient to make and is not one of pure medical judgment committed to the physician’s direction. In this scenario the exclusion of facts to the patient renders the physician in a tenuous position. Clearly the duty requires the physician to disclose. However, one must then evaluate whether or not under the circumstances (ie, a treatable malignancy in a young man) a reasonable individual would have refused chemotherapy even if the risks were known. Although the physician was wrong in not

REFERENCES 1. Evans KG. A medicolegal handbook for physicians in Canada. 5th ed. Ottawa: Canadian Medical Protective Association; 2002. 2. Policy Statement #1-03. Disclosure of harm: College of Physicians and Surgeons of Ontario. Dialogue 2003;11(3):1–12. 3. Wu AW. Handling hospital errors: is disclosure the best defence? Ann Intern Med 1999;131:970–2. 4. Hawke M, Jahn AF. Diseases of the ear: clinical and pathologic aspects. Philadelphia (PA): Lea & Febiger; 1987. 5. Halmagyi M, Curthoys IS, Aw ST, et al. The human vestibulo-ocular reflex after unilateral vestibular deafferentation. The results of high acceleration impulsive testing. In: Sharpe J, Baraber HO, editors. The vestibular–ocular reflex and vertigo. New York: Raven Press; 1993. p. 45–54. 6. Chambers BR, Mai M, Barber HO. Bilateral vestibular loss, oscillopsia and the cervico-ocular reflex. Otolaryngol Head Neck Surg 1985; 93:403–7. 7. Fujimoto M, Mai M, Rutka J. A study in the phenomenon of head shaking nystagmus: its presence in a dizzy population. J Otolaryngol 1993; 22:376–401. 8. Largey v Rothman, 110 NJ 204 at 214 (1988).

APPENDIX 1

Information Provided to Lawyers

CASE 1: SYSTEMIC GENTAMICIN OTOTOXICITY

Multiple versus Single Daily Dosing

Overview of Systemic Gentamicin Ototoxicity Gentamicin is a member of the aminoglycoside class of antibiotics. It is a bactericidal agent that is active against some gram-positive and many gram-negative (eg, Pseudomonas, Escherichia coli, and Klebsiella species) bacteria not usually sensitive to commonly prescribed antibiotics, such as penicillin, cephalosporin, and macrolides—hence its continued use. It is poorly absorbed through the gastrointestinal tract and must be administered intravenously or intramuscularly. Gentamicin is excreted via the kidney. It is relatively inexpensive (which hospital pharmacies love) compared with the newer antibiotics available (eg, third-generation cephalosporins and fluoroquinolones).1–3 Antimicrobial resistance is relatively low (which microbiologists love).4

In patients with normal renal function gentamicin was conventionally given every 8 hours (multiple daily dosing). By the mid-1990s meta-analyses revealed that a combined dose could be given once a day (single daily dosing) without loss of effect and without placing the patient at risk for increasing nephrotoxicity.1,4,9 The meta-analyses could not, however, address whether single daily dosing protocols resulted in a greater degree of ototoxicity.10 Single daily dosing had one important advantage over multiple daily dosing—its ease of administration made it ideal for outpatient and homecare administration (publicly funded hospitals, insurance companies, and patients liked the fact that more expensive inpatient treatment could be minimized). But is treatment with systemic gentamicin always in the best interest of the patient, even if it can be delivered on an outpatient basis?

Clinical Indications Gentamicin has been used in the treatment of infective disorders such as osteomyelitis, infective endocarditis, and abdominal sepsis.1 All of these conditions typically require a prolonged treatment course in patients who often are very ill and initially in an intensive care setting. Toxicity Gentamicin’s major side effects are nephrotoxicity, ototoxicity, and neuromuscular blockade.5 It is recommended that gentamicin be avoided in patients who are pregnant and in those who have renal failure. The ototoxicity with systemic gentamicin is usually vestibular in nature, presenting with imbalance, ataxia, and oscillopsia (visual blurring with head movement).6,7 Gentamicin vestibulotoxicity tends to be permanent (although it is estimated that if gentamicin is stopped immediately upon onset of symptoms there is approximately a 50% chance of reversibility).8

Risk Factors To date, risks for ototoxicity are usually quoted at 1 to 2% for all those receiving systemic gentamicin.8,11 However, we have identified several risk factors that place an individual at greater risk for ototoxicity. Risk factors for gentamicin vestibulotoxicity include the following12,13: 1. Prolonged treatment course > 14 days with a cumulative dosage > 2.5 g 2. Development of renal failure during treatment (ie, rising creatinine levels) 3. Use of concomitant ototoxic agents, such as diuretics and macrolide antibiotics (eg, erythromycin) 4. Elevated serum trough (preadministration levels) antibiotic levels 5. Previous history of ototoxicity Of all the risk factors, prolonged treatment course appears to be the most important.6 There had been

208

Appendix 1

great hopes that monitoring serum gentamicin levels could prevent ototoxicity from occurring.14 Although monitoring may reduce the risk somewhat, there is a body of evidence in the world literature of patients who had normal serum levels and did not develop renal toxicity or other risk indications yet who went on to develop gentamicin vestibulotoxicity.6 Note that at the present time there are equally efficacious and safer antibiotics to use without risk for ototoxicity (albeit more expensive and as such possibly precluded from home or outpatient treatment). 1. Evidence Supporting an Action of Medical Negligence In the absence of other causes the patient has been diagnosed with a bilateral peripheral vestibular loss from ototoxicity. Consider the following: a.

b.

c.

d.

e.

f.

g.

Osteomyelitis can be a severe and limbthreatening infection, especially in diabetics and those with peripheral vascular disease. Until cultures were known, it would have been appropriate to treat with triple antibiotic therapy. Although diabetics with chronic osteomyelitis tend to have polymicrobial infections, a deep aspirate from the joint in a first-time infection would be evidence that the streptococcus cultured was responsible for her infection.15,16 Cultures demonstrated that the streptococcus cultured was sensitive to all agents independently and even to straightforward penicillin. Treatment for a suspected osteomyelitis usually requires a minimum of 6 weeks of antibiotic therapy.16,17 The patient was not told about the wellrecognized adverse effects of gentamicin (ie, if the patient were advised at first onset of symptoms it may have resulted in an earlier discontinuation of her gentamicin with possible spontaneous recovery of vestibular function). Toxicity is the result of prolonged administration of gentamicin and is evidenced by the development of renal impairment during treatment.

2. Legal Research a. How do you research a case like this to determine its merits (what resources do you use)? b. How do you go about getting an expert opinion?

c.

How many expert opinions would you need in this case (do you need an infectious disease expert’s opinion, a microbiologist’s opinion, a neurotoxicologist’s opinion)?

3. Responsibility of Health Care Professionals a. The hospital internist prescribed triple antibiotic therapy when cultures a few days later revealed a bacteria that safer agents could have treated without relying on gentamicin. b. The family physician failed to act (ie, by discontinuing the gentamicin or altering its dose) on rising levels of serum creatinine and trough levels of gentamicin that suggested the patient was slipping into renal failure. c. The visiting nurse failed to take seriously the complaints of the patient (she was also unaware that gentamicin was primarily vestibulotoxic, not cochleotoxic) and did not notify the treating physician. 4. Recommendations for Physicians and Allied Health Care Professionals

CASE 2: TOPICAL GENTAMICIN OTOTOXICITY Overview of Topical Aminoglycoside Drop Ototoxicity Topical aminoglycoside drops (with or without a steroid component) have been used in the topical treatment of ear disease owing to their efficacy, low resistance, and low per unit cost.18 Because of their activity against gram-negative organisms they have been specifically indicated for the treatment of external otitis and chronic otitis media (where gram-negative polymicrobial infections are the norm). 19–21 Until fluoroquinolones such as ciprofloxacin were introduced, there were no oral antibiotic agents available for the majority of gram-negative organisms. Topical treatment ideally delivers a high concentration of the antibiotic to areas that are difficult for systemic antibiotics to reach tissue levels adequate to fight infection. Clinical Indications Topical aminoglycoside drops are typically prescribed for the treatment of a discharging ear. This is despite there being no official US Food and Drug Administration (FDA) approval for a middle ear indication. Use would therefore represent an “off-label” indication. Toxicity There has been sufficient documentation in animal models that topical aminoglycosides are toxic to the


inner ear. In humans, instillation of concentrated gentamicin into, for example, the middle ear (at concentrations often > 10 times that found in commercially available topical drops) has been therapeutically used to cause vestibular deafferentation in the treatment of Meniere’s disease.23 Nevertheless, up until 1997 or so little evidence in the world literature supported a significant concern regarding topical aminoglycoside ototoxicity in humans treated for middle ear sepsis.24,25 In a 1992 survey most US otolaryngologists acknowledged that although there was a risk for ototoxicity they felt comfortable prescribing aminoglycoside drops in the presence of a tympanic membrane (TM) perforation on the basis that any untreated middle ear sepsis carried a higher risk for ototoxicity than did the treatment itself.26 Introduction of Fluoroquinolone and Recognition of Topical Ototoxicity In 1997–98 topical fluoroquinolone drops were introduced into the US market. These agents were effective and carried no risk for ototoxicity. They were, however, usually more expensive. One of the medications, ofloxacin, even received FDA approval for middle ear use. Around the same time studies in the US and European literature began demonstrating inadvertent ototoxicity from commercially available gentamicincontaining drops. The toxicity, however, was identified to be primarily vestibular in nature, not cochlear (similar to that seen in systemic ototoxicity and perhaps the main reason why it had been missed for so long).25,27,28 Risk Factors for Topical Aminoglycoside Ototoxicity Risk factors for topical gentamicin ototoxicity (the topical agent most studied) appear to be directly related to duration of treatment (> 7 days), especially when used in the presence of a dry TM perforation.25,27,29 For this reason it has been suggested that patients requiring topical aminoglycoside drops have weekly reviews to determine the need for continued treatment and should be instructed to stop drops once the discharge has stopped.18 1. Harm Leading to a Medicolegal Action a. Although he missed 6 months of work he is now able to perform most of his premorbid activities (driving a car, returning to work, etc). b. He has only slight imbalance with fast movements of his head. c. His physical examination suggests a loss of left vestibular activity, which is confirmed with electronystagmography testing. d. There was no evidence to suggest another pathology apart from the gentamicin drops

209

(the infection had stopped many days prior, there was no evidence of a viral upper respiratory tract infection that might have resulted in a vestibular loss, etc). 2. Responsibility of Patient versus Lack of Treating Physician Follow-up a. Topical treatment was indicated and the duration of recommended treatment (7 days) would be considered reasonable. b. The treating physician did not warn the patient of ototoxicity (rate of cochlear toxicity, not vestibular, has been quoted as 1 in 10,000)—was there a lack of informed consent? 24 c. The drops were used in what would be considered an “off-label” indication (no FDA approval but within the community of practicing otolaryngologists it would be considered standard of care to use the drops).24,26 3. Year of Incident a. Since 1997–98 there has been a general awareness that topical aminoglycoside drops are ototoxic when used in a prolonged fashion in the presence of a dry middle ear. b. In 1997–98 topical fluoroquinolone drops were introduced that did not carry risk for ototoxicity and were demonstrated to be equally effective even if used in an “offlabel” fashion.

CASE 3: TOPICAL TOXICITY FROM SURGICAL DISINFECTANTS AND PREPARATORY SOLUTIONS Overview of Ototoxicity from Topical Disinfectants Chlorhexidine is an agent available for preoperative hand washing and general skin disinfection in preparations that range in concentration from 0.015 to 5%.30 It is positively charged and acts by disrupting the cell membrane and precipitating the cytoplasm of susceptible bacteria.31 Ototoxicity Animal models have demonstrated how extremely toxic chlorhexidine and other alcohols can be to the inner ear if they reach the middle ear through a TM perforation. Toxicity appears to be directly related to the concentration and the contact time with the round window membrane of the inner ear. Toxicity appears to occur within hours and is both cochlear and vestibular.32–35 A 1971 article by Bicknell identified profound unilateral deafness in 14 of 97 (14%) patients who had undergone routine tympanoplasty surgery; the only common

210

Appendix 1

factor was a 0.5% chlorhexidine solution in 70% spirit.36 (Otherwise there have been only a few sporadic case reports of disinfectant ototoxicity in humans.) Prevention There is no evidence-based medicine to demonstrate that use of disinfectants prevents infection from occurring in middle ear surgery, although it makes good sense to reduce bacterial load wherever possible. Only aqueous-based (not alcohol-based) iodine solutions are relatively safe to use in the presence of a TM perforation.37–39 When the TM is intact there need be no worry about the preparatory solution used. When a TM perforation is present, a temporary barrier (a plug of cotton) is often used to prevent the solution from entering the middle ear. 1. Prevention of Profound Hearing Loss There appears to be no agreed upon standard regarding a temporary barrier to prevent preparatory solution from entering the middle ear. Some surgical texts have specifically addressed this issue and have stated that preparatory solutions in the middle ear should be avoided whenever possible. Others have not addressed this issue. Clinical experience points to aqueous iodine solutions being the safest in humans. 2. Apportioning Blame a. The hospital wished to proceed to a standardized preparatory solution, one reason for which was to save money. b. The recommended change passed through the pharmacy committee at the hospital following a review of the literature (not necessarily specific to ear surgery); representatives included members of all departments, including otolaryngology. c. The surgeon would have had the last best chance to avoid toxicity by using a temporary barrier. d. In general, the odds of a complete sensorineural hearing loss as a complication of tympanoplasty surgery are 1 in 1,000 operations.38–40 e. Reports of toxicity from preparatory solutions in the ear are quite rare; one large retrospective series was written up more than 30 years ago.36 3. Appropriateness of Surgeons Issuing Warning

CASE 4: CISPLATIN OTOTOXICITY Overview of Cisplatin Ototoxicity Cisplatin is a chemotherapeutic agent that has typically resulted in durable remissions in solid cancers such as

ovarian and testicular cancer and significant action in cancer of the head and neck, cervix, prostate, bladder, lung, esophagus, uterus, and brain.41 Its systemic pharmacodynamics, antitumor specificity, and cellular mechanisms of antitumor action are well understood.41,42 Ototoxicity The cellular and molecular interactions leading to toxic dose-limiting side effects are not as well understood.42 Toxicities include permanent high-frequency sensorineural hearing loss (range of 20 to 90% in those receiving cisplatin) and peripheral neuropathy as well as dose-related renal impairment from tubular necrosis and interstitial nephritis.43–47 There is significant variability in presentation and susceptibility to cisplatin-mediated ototoxicity. However, it is generally accepted that high-dose therapy may be more cochleotoxic. Accompanying tinnitus (unwanted hear noise) may be transient or permanent and is estimated to occur in 7% of clinical trials (range 2 to 36%).41 Despite these risks cisplatin administration is considered necessary to treat life-threatening conditions, and as such, the toxic side effects are often accepted under the circumstances. Prevention Hearing loss appears to be permanent despite treatments with prehydration or mannitol administration.41,48 In other words, there is no treatment that will universally protect the inner ear from cisplatin toxicity, although some research agents are quite promising.41,42,48 1. Grounds for Negligence The physician demonstrated full disclosure of the risks and benefits of chemotherapy (including hearing loss), and it was the patient’s ultimate decision to carry on with therapy for a very treatable malignancy. 2. Role of Informed Consent

REFERENCES 1. Begg EJ, Barclay ML. Aminoglycosides—50 years on. Br J Clin Pharmacol 1995;39:597–603. 2. Davis BD. Mechanism of the bactericidal action of aminoglycosides. Microbiol Rev 1987;57:341–50. 3. Bertino JS, Rorschafer JC. Editorial response. Single daily dosing of aminoglycosides. A concept whose time has not yet come. Clin Infect Dis 1997;24:820–3. 4. Levison ME. New dosing regimen for aminoglycoside antibiotics. Ann Intern Med 1992;117:693–4. 5. John JF. What price success? The continuing saga of the toxic:therapeutic ratio in the use of aminoglycoside antibiotics. J Infect Dis 1988;158:1–6.


6. Halmagyi M, Fattore C, Curthoys IS, Wade S. Gentamicin vestibulotoxicity. Otolaryngol Head Neck Surg 1994;111:571–4. 7. Black FO, Pesznecker S. Vestibular ototoxicity: clinical considerations. Otolaryngol Clin North Am 1993;26:713–36. 8. Jackson GG, Arcieri G. Ototoxicity of gentamicin in man: a survey and controlled analysis of clinical experience in the United States. J Infect Dis 1971;124 Suppl:130–7. 9. Bailey TC, Little JR, Littenberg B, et al. A metaanalysis of extended-interval dosing versus multiple daily dosing of aminoglycosides. Clin Infect Dis 1997;24:786–95. 10. Aminoglycosides once daily? [editorial] Drug Ther Bull 1997;35(5):36–7. 11. Arcieri GM, Falso F, Smith HM, Hobson CB. Clinical research with gentamicin: incidence of adverse effects. Med J Aust 1970;Suppl:30–4. 12. Rutka J, Alberti PW. Toxic and drug-induced disorders in otolaryngology. Otolaryngol Clin North Am 1984;17:761–74. 13. Rutka J. Disorders of immunosuppression. In: Gray RJ, Rutka JA, editors. Recent advances in otolaryngology. 6th ed. London (UK): Churchill Livingstone; 1988. p. 223–43. 14. Lerner SA, Matz GJ. Suggestions for monitoring patients with aminoglycoside antibiotics. Otolaryngol Head Neck Surg 1978;87:222–8. 15. Wheat LJ, Allan SD, Henry M, et al. Diabetic foot infections: bacterial analysis. Ann Intern Med 1986;146:1935–40. 16. Lipsky BA, Pecoraro RE, Wheat L. The diabetic foot: soft tissue and bone infection. Infect Dis Clin North Am 1990;4:409–31. 17. Gilbert DN, Moellering RC, Sande MA, editors. The Sanford guide to antimicrobial therapy. 29th ed. Vienna (VA): Antimocrobial Inc; 1999. 18. Bath AP, Walsh RM, Bance ML, Rutka JA. Ototoxicity of topical gentamicin preparations. Laryngoscope 1999;109:1088–93. 19. Lancaster JL, Mortimore S, McCormick M, Hart CA. Systemic absorption of gentamicin in the management of active mucosal chronic otitis media. Clin Otolaryngol 1999;24:435–9. 20. Acuin J, Smith A, MacKenzie I. Interventions for chronic suppurative otitis media [systematic review]. Cochrane Ear Nose and Throat Disorders Group. Cochrane Database of Systematic Reviews, 2003;(1). 21. Hannley MT, Denneny JC, Holzer SS. Use of ototopical antibiotics in treating 3 common ear conditions. Otolaryngol Head Neck Surg 2000;122: 934–40. 22. Roland PS, Rutka J, Haynes DS. Antibiotic ototoxicity: pathology, management and medicolegal

23.

24.

25.

26.

27. 28.

29.

30. 31. 32.

33. 34.

35.

36.

37.

38.

39.

211

implications. Ear Nose Throat J 2003;82 (1 Suppl):1–16. Nedzelski JM, Schessel DA, Bryce GE, Pfleiderer AG. Chemical labyrinthectomy: local application of gentamicin for the treatment of unilateral Meniere’s disease. Am J Otol 1992;13:18–22. Roland PS. Clinical ototoxicity of topical antibiotic drops. Otolaryngol Head Neck Surg 1994;110: 598–602. Wong DL, Rutka JA. Do aminoglycoside otic preparations cause ototoxicity in the presence of tympanic membrane perforations? Otolaryngol Head Neck Surg 1997;116:404–10. Lundy LB, Graham MD. Ototoxicity and ototopical medications: a survey of otolaryngologists. Am J Otol 1993;14:141–6. Marais J, Rutka J. Ototoxicity of topical ear drops. Clin Otolaryngol 1998;23:360–7. Walby P. Stewart R, Kerr AG. Aminoglycoside ear drop ototoxicity: a topical dilemma? Clin Otolaryngol 1998;23:219–40. Kaplan DM, Hehar SS, Bance ML, Rutka JA. Intentional ablation of vestibular function using commercially available topical gentamicinbetamethasone ear drops in patients with Meniere’s disease. Further evidence for topical ear drop ototoxicity. Laryngoscope 2002;112:689–95. British National Formulary 45, March 2003. Mosby’s drug consult. Elsevier; 2003. Available at: http://www3.us.elsevierhealth.com/Drug Consult/. Aursnes J. Vestibular damage from chlorhexidine in guinea pigs. Acta Otolaryngol 1981;92: 89–100. Aursnes J. Cochlear damage from chlorhexidine in guinea pigs. Acta Otolaryngol 1981;92:259–71. Igarashi Y, Suzuki J. Cochlear ototoxicity of chlorhexidine gluconate in cats. Arch Otol Rhinol Laryngol 1985;242:167–76. Igarashi Y, Oka Y. Vestibular ototoxicity following intratympanic applications of chlorhexidine gluconate in the cat. Arch Otol Rhinol Laryngol 1988;245:210–7. Bicknell PG. Sensorineural deafness following myringoplasty operations. J Laryngol Otol 1971; 85:957–61. Jackson CG. Principles of temporal bone and skull base surgery. In: Glasscock ME, Gulya AJ, editors. Surgery of the ear. 5th ed. Hamilton (ON): BC Decker Inc; 2003. p. 263–88. Shea MC. Tympanoplasty: the undersurface graft technique-transcanal approach. In: Brackmann DE, Shelton C, Arriaga MA, editors. Otologic surgery. 2nd ed. Philadelphia (PA): WB Saunders Co; 2001. Jackson CG, Glasscock ME, Strasnick B. Tympanoplasty: the undersurface graft technique-

212

40.

41. 42.

43.

Appendix 1

postauricular approach. In: Brackmann DE, Shelton C, Arriaga MA, editors. Otologic surgery. 2nd ed. Philadelphia (PA): WB Saunders Co; 2001. Mandolis S. Closure of tympanic membrane perforations. In: Glasscock ME, Gulya AJ, editors. Surgery of the ear. 5th ed. Hamilton (ON): BC Decker Inc; 2003. p. 400–19. Schweitzer VG. Ototoxicity of chemotherapeutic agents. Otolaryngol Clin North Am 1993;26:759–89. Schweitzer VG. Cisplatinum-induced ototoxicity: effect of pigmentation and inhibitory agents. Larygngoscope 1993;103 Suppl 59:1–52. Gratton MA. The interaction of cisplatinum and noise on the peripheral auditory system. Hear Res 1990;50:211–24.

44. Saferstein RL. Cisplatinum nephrotoxicity. Am J Kidney Dis 1986;8:356–67. 45. Daugaard G. Cisplatinum nephrotoxicity: review. Cancer Chemother Pharmacol 1989;25:1–9. 46. Walsh TJ, Clark AW, Parhad IM, Green WR. Neurotoxic effects of cisplatinum therapy. Arch Neurol 1982;39:719–20. 47. Forastiere AA, Takasugi BJ, Baker SR, et al. High dose cisplatinum in advanced head and neck cancer. Cancer Chemother Pharmacol 1987; 19:155–8. 48. Fausti SA, Henry JA, Schaffer HI, et al. Highfrequency monitoring for early detection of cisplatin ototoxicity. Arch Otolaryngol Head Neck Surg 1993;119:661–6.

APPENDIX 2

Summary of The 2004 AAO-HNS Consensus Panel Recommendations

SUMMARY OF THE 2004 AMERICAN ACADEMY OF OTOLARYNGOLOGY—HEAD AND NECK SURGERY CONSENSUS PANEL RECOMMENDATIONS REGARDING POTENTIALLY OTOTOXIC ANTIBIOTICS FOR MIDDLE EAR USE In 2003 Jonas Johnson, then president of the American Academy of Otolaryngology—Head and Neck Surgery (AAO-HNS), charged a panel of physicians with the responsibility to develop a consensus position on the use of potentially ototoxic antibiotic topical drops in the treatment of ear disease based on evidence-based medicine. Previous AAO-HNS position statements adopted in 1994 and reaffirmed in 1998 had recognized the appropriateness of using available topical preparations, including those containing aminoglycosides, in the treatment of external and middle ear disorders despite their potential for ototoxicity, provided the benefits of therapy outweighed the risks and no equally effective, less potentially hazardous treatments were available. The need to revisit this specific issue was based on the perceived increase in claims against physicians alleging topical iatrogenic ototoxic injury, political developments and practice guidelines in countries such as Canada and the United Kingdom limiting the use of potentially ototoxic topical antibiotics, the introduction of nonototoxic preparations (specifically topical fluoroquinolones) to the marketplace, and the fact that no systematic evidence-based review of these issues had been performed. All consensus documents such as this without exception will need to be revisited from time to time. This dynamic process is necessary if recommendations are to remain germane to the practice of medicine: former recommendations may change to reflect subsequent evidence from published clinical and basic science research. Readers wishing a more detailed explanation of the consensus panel’s recommendations are requested to review the March 2004 supplement to Otolaryngology— Head and Neck Surgery: Official Journal of the American Academy of Otolaryngology—Head and Neck Surgery.1

RECOMMENDATIONS OF THE CONSENSUS PANEL ON OTOTOPICAL ANTIBIOTICS 1. When possible, topical antibiotic preparations free of potential ototoxicity should be preferred over ototopical preparations that have the potential for ototoxic injury if the middle ear and mastoid are open. 2. If used, potentially ototoxic antibiotic preparations should be used only in infected ears. Use should be discontinued shortly after the infection has resolved. 3. If potentially ototoxic antibiotic drops are prescribed for use in the open middle ear or mastoid, the patient should be warned of the risk of ototoxicity. 4. If potentially ototoxic antibiotics are prescribed, the patient should be specifically instructed to call the physician or return to his or her office if the patient develops the following: • Dizziness or vertigo • Hearing loss (or additional hearing loss if hearing impairment was part of the original condition) • Tinnitus 5. If the tympanic membrane is known to be intact and the middle ear and mastoid are closed, then the use of potentially ototoxic preparations presents no risk of ototoxic injury.

REFERENCE 1. Roland PS, Stewart MG, Hannley M, et al. Consensus panel on the role of potentially ototoxic antibiotics for topical middle ear use: introduction, methodology and recommendations. Otolaryngol Head Neck Surg 2004;130(3 Suppl):S51–6.

APPENDIX 3

Major Groups of Agents Recognized to be Ototoxic in Humans

AMINOGLYCOSIDES

LOOP DIURETICS

amikacin dihydrostreptomycin gentamicin kanamycin neomycin netilmicin streptomycin tobramycin

ethacrynic acid furosemide

ANTISEPTIC/DISINFECTANT AGENTS

MACROLIDES azithromycin clarithromycin erythromycin

NONSTEROIDAL ANTI-INFLAMATORY DRUGS (NSAIDS)

alcohol chlorhexidine

TOPICAL ANTIBIOTICS

ARSENICALS

all aminoglycosides chloramphenicol polymyxin

CYTOTOXIC AGENTS bleomycin carboplatin cisplatin nitrogen mustard vinca alkaloids

IRON-CHELATING AGENTS (FE 2+) deferoxamine

TRADITIONAL AGENTS acetylsalicylic acid (ASA) quinine

VANCOMYCIN (in conjunction with aminoglycosides, primarily)

Index

A ABR, 13f, 154 chlorhexidine, 140 lead, 36 Acetic acid, 135, 142 for otomycosis, 137t Acetylcysteine, 176 ACTH/MSH, 178 Actin fibers, 7 Acute otitis media (AOM), 121 Adaptation, 24 Adenosine receptor agonists, 178 A1555G mutation clinical phenotype of, 148 in 12S ribosomal ribonucleic acid gene, 145–146 Alcohol, 135, 141, 209–210 for otomycosis, 137t Alexander’s law, 24 Alpha-lipoic acid, 177 Alpha-melanocyte-stimulating hormone (alpha-MSH), 178 Alpha-MSH, 178 Alpha-phenyl-tert-butyl-nitrone, 171–172 Alpha-tocopherol, 177 Aluminum acetate, 142 AMD-473, 68 clinical status of, 62t American Academy of Otolaryngology– Head and Neck Surgery Consensus Panel Recommendations ototopical antibiotics, 213 potentially ototoxic antibiotics for middle ear use, 213 American Speech Language Hearing Association (ASHA) ototoxic threshold shift criteria, 154 Amifostine, 176 Amikacin, 84 availability of, 83t Aminoglycosides, 170–171, 215. See also Topical aminoglycosides activity of, 93 availability of, 83t cell death, 94–95 clinical indications for, 82–83

free radicals, 97f inducing cellular toxicity mechanisms for, 144–145 multiple-daily vs. once daily dosing, 86–87 nephrotoxicity of, 84–85 neuromuscular blockade, 86 otoprotection research, 176 ototoxicity of, 82–89 genetic factors in, 144–149 genetic susceptibility to, 145–147 mechanisms for, 93–98 nitric oxide, 96 pathophysiology of, 147 prevention and therapy of, 148–149 risk factors for, 86t ROS, 96–97 structural and functional changes in, 144 pharmacokinetics of, 94 recommendations for, 88–89 renal monitoring, 87–88 serum monitoring, 87–88 structure of, 93 toxicity of, 84–86 tympanostomy tubes, 118 vestibulotoxicity risk factors for, 164t in vitro activity, 83 AMPA, 9t Amphotericin B for otomycosis, 137t ototoxicity of, 112 Anemia with carboplatin, 67 with cisplatin, 61 Antibiotics. See also Aminoglycosides middle ear effects of, 107–111 ototopical American Academy of Otolaryngology–Head and Neck Surgery Consensus Panel Recommendations, 213 topical, 215 Antifungals, 136 for otomycosis, 137t

topical, 134–138 application of, 137 for otomycosis, 137t Antimycotics, 111–112 Antioxidants, 172–173, 177–178 Antiseptics, 111–112, 135–136, 140–143, 215 for otomycosis, 137t AOM, 121 Apoptosis, 94–95, 170 free radical formation, 95–96 Arachidonic acid metabolism salicylates, 31 Arsenicals, 215 ASHA ototoxic threshold shift criteria, 154 Aspergillus, 134 Aspirin, 28 ATP receptors, 9t Audax, 142 Audicort, 142 Audiologic assessment, 153–154 Auditory brainstem response (ABR), 13f, 154 chlorhexidine, 140 lead, 36 Auditory electrical pathway furosemide, 44–45 Auditory potentials salicylates, 30 Azithromycin, 102–103 Azoles, 136 Azosemide ototoxicity of, 46

B Bacteroides fragilis, 130 BAO-HNS, 121 Basilar membrane, 1, 5, 14 mechanics of, 13–14 Bechterew’s nucleus, 22 Bedside clinical vestibular tests, 167 Benzalkonium chloride, 110, 141 Benzethonium chloride, 141 Bifonazole, 136

216

Index

Bladder cancer cisplatin for, 55, 55t Blame, 205, 210 Bleomycin for testicular cancer, 51t Boric acid, 135 for otomycosis, 137t Bottcher cells, 5, 6 Brain-derived neurotrophic factor, 175 Brainstem potentials, 13 British Association of Otorhinolaryngologists–Head and Neck Surgeons (BAO-HNS), 121 Bumetanide ototoxicity of, 45–46 C Caloric tests, 165 advantages and limitations of, 165t Candida, 134, 137 CAP, 12, 13 Carbenicillin middle ear effects of, 107–111 Carboplatin, 66–67 chemistry, 66 clinical pharmacology, 66 clinical status of, 62t clinical use of, 66 indications for, 66 mechanism of action, 66 otoprotection, 179 ototoxicity of, 67 pathophysiology of, 67 toxicity of nonotological manifestations of, 66–67 Caspase inhibitors, 174–175 Caspases, 178 CDP, 161 tests advantages and limitations of, 165t Ceftazidime mucosal hemorrhage associated with, 109f Cell death, 170 aminoglycosides, 94–95 pathways, 95f free radical formation, 95–96 mitogen-activated protein kinase (MAPK) signaling pathway, 173–175 pathways inhibitors of, 173–174, 178–179 Central vestibular system, 22 CEP 1347, 174 Cerebrospinal fluid (CSF) composition of, 2t Cervical cancer cisplatin for, 54 Chemotherapeutic agents, 176–177 Children ototoxicity monitoring, 155–156 Chlamydia, 130 Chloramphenicol, 130–132 animal studies, 131 ear drops, 142

historical development of, 130 human studies, 131–132 proprietary agents containing, 131 therapeutic use of, 130–131 Chlorhexidine, 140–141, 209–210 ABR, 140 Chlorhexidine gluconate ototoxicity of, 112 Chloromycetin Otic, 131 Cholesteatoma Cortisporin Otic Suspension, 110f, 111f Chronic suppurative otitis media (CSOM), 121 Cinchona bark, 33 Ciprofloxacin middle ear effects of, 108 Cisplatin, 178 for bladder cancer, 55, 55t for cervical cancer, 54 chemistry of, 61 clinical pharmacology, 50–52, 61 clinical status of, 62t clinical uses of, 50–57, 61 for epithelial ovarian cancer, 52–54, 52t intraperitoneal, 54 indications for, 61 interactions with other ototraumatic agents, 65–66 leukopenia with, 61 for lymphoma, 56 mechanism of action, 50, 61 nephrotoxicity of, 50–51 neurotoxicity of, 50 for non-small cell lung cancer, 56 otoprotection, 176–177 ototoxicity of, 61–63, 205, 210 characteristics of, 64 clinical presentation of, 60t epidemiology of, 63–64 historical development of, 60 pathophysiology of, 64–65 predisposing factors, 60t risk factors, 65 pharmacokinetics, 50 for small cell lung cancer, 55–56 for squamous cell head and neck cancer, 54–55 for testicular cancer, 51t testis germ cell tumors, 51–52 toxicity of, 50 nonotological manifestations of, 61 C-Jun N-terminal protein kinase (JNK), 95 C-Jun N-terminal protein kinase inhibitory peptide (D-JNK-1), 174 Clarithromycin, 103 Clinical topical ototoxicity, 121–122 Clotrimazole, 136 for otomycosis, 137t ototoxicity of, 112 CM, 12–13 Cochlea anatomy of, 1–10 blood supply, 4 salicylates, 31

duct, 10 electrophysiology, 12–13 fluid spaces, 2–4 lateral wall, 4–5 neural structure, 8–10 physiology of, 10–16 position of, 1f potentials, 12–13 scalae, 2f standing current in, 11f vasculature, 3f, 4 Cochlea microphonics (CM), 12–13 Cochlear potentials salicylates, 30 Colymycin animal studies, 129 Coly-Mycin S cochlear ototoxicity, 115 Coly-Mycin S Otic, 128 Compensation, 24, 163 Compound action potential (CAP), 12, 13 Computerized dynamic posturography (CDP), 161 tests advantages and limitations of, 165t Conductive hearing loss, 153 Cortisporin Otic Solution, 128–129 Cortisporin Otic Suspension, 109 basal cochlear turn, 117f cholesteatoma, 110f, 111f cochlear ototoxicity, 115 middle ear effects of, 108 organ of Corti, 117f outer hair cell, 116f propylene glycol, 110 Cresylate, 135 for otomycosis, 137t Crista, 21f CSF composition of, 2t CSOM, 121 Cyclooxygenase pathways salicylates, 31 Cytotoxic agents, 215

D DDTC, 176 Deafness mercury, 35–36 Deferoxamine, 76–78, 172–173 ototoxicity of, 76–78 mechanism of, 78 minimizing, 77 Deiters’ cell, 6, 6f, 7, 11 Deiters’ nucleus, 22 Desferrioxamine. See Deferoxamine Dexamethasone middle ear effects of, 109 ototoxicity of, 116 DHI, 168 Diethyldithiocarbamate (DDTC), 176 2,3-dihydroxybenzoate, 172–173 Disclosure, 200

Index Distortion product otoacoustic emissions (DPOAE), 15f, 155 Diuretics loop, 215 Dizziness Handicap Inventory (DHI), 168 D-JNK-1, 174 D-Met, 176, 177 Domoic acid, 9t DPOAE, 15f, 155 Dynamic visual acuity, 26 Dynamic visual acuity testing (oscillopsia testing), 167

FTN, 161 Furosemide ototoxicity of, 43–45 administration route, 44 biochemical changes, 45 bolus dosing, 44 clinical manifestations of, 43–44 mechanisms of, 44–45 morphologic and histologic changes, 45 rate of infusion, 44

E Ebselen, 177 Econazole, 136 EHF, 154 Electronystagmography (ENG), 161 Endocochlear potential ethacrynic acid, 43 furosemide, 44–45 Endolymph, 11 Endolymphatic sac fluid composition of, 2t Endolymph electrolytes ethacrynic acid, 43 ENG, 161 ENG caloric tests, 165 advantages and limitations of, 165t ENG tests, 165–166 Environmental chemicals, 157 ERCC1, 69 Erythromycin, 101–102 ototoxicity of mechanisms of, 103–104 predisposing factors for, 101–102 risk factors for, 102 Escherichia coli, 130 Ethacrynic acid ototoxicity of, 42–43 clinical manifestations of, 42 mechanisms of, 42–43 morphologic and histologic changes, 43 Etoposide for testicular cancer, 51t Excision repair cross-complementing 1 (ERCC1), 69 Extended high-frequency (EHF), 154 Extrinsic cell death receptor pathway flow chart, 171f

G Garasone ear drops Meniere’s disease, 125f vestibular ablation, 124 vestibulotoxicity of, 123 GDNF, 175 Genetic factors in aminoglycoside ototoxicity, 144–149 clinical relevance of, 147–149 Gentamicin. See also Topical gentamicin availability of, 83t cochlear ototoxicity, 114–115 indications for, 207 intratympanic delivery of, 192–193 dose of, 193 end point, 193–194 multiple vs. single daily dosing, 207 otorrhea, 118 risk factors, 207–208 systemic ototoxicity case study, 200 information provided to lawyers, 207–208 toxicity of, 207 Gentian violet, 135–136 for otomycosis, 137t ototoxicity of, 112 Glial cell line-derived neurotrophic factor (GDNF), 175 GluR, 9t Glutamate, 9t Glutathione, 172 Griseofulvin ototoxicity of, 112 Growth factors, 175–176

F Floccular target neurons (FTN), 161 Fluconazole, 137 Fluoroquinolone, 209 middle ear effects of, 108 Fosfomycin animal studies, 129–130 Framycetin chronic otitis media, 118 Free radicals aminoglycosides, 97f

H Habituation, 24 Haemophilus influenzae, 130 Hair cells, 22–23, 23f glycogen metabolism ethacrynic acid, 43 Half-lists, 154 Halmagyi maneuver, 25–26, 25t Head and neck cancer cisplatin for, 54–55 Head shake test, 25t, 26 Health care professionals responsibility of, 202–203, 208

217

Hearing ototoxic damage, 170–179 toxicity criteria for, 63t Hearing loss A1555G mutation in 12S ribosomal ribonucleic acid gene, 146 conductive, 153 control of, 192t mercury, 35–36 profound preventable, 204–205, 210 quinine, 34 salicylates, 29, 32 sensorineural with deferoxamine, 76, 77t ethacrynic acid, 42 Heavy metals, 35–36 Hensen’s cells, 6 High-frequency audiometry, 154–155 High-frequency head thrust (Halmagyi maneuver), 25–26, 25t High-frequency horizontal head thrust, 167 Horizontal vestibulo-ocular reflex (VOR) wiring of, 162f Hughson-Westlake threshold procedures, 153 Hydrocortisone middle ear effects of, 109 Hypoalbuminemia furosemide, 45 I IAC, 21 Ifosfamide for testicular cancer, 51t IHC, 1, 6, 7f, 11 stereocilia, 8 Indacrinone ototoxicity of, 46 Informed consent, 206 Inherited factors in aminoglycoside ototoxicity, 144–149 Inner hair cells (IHC), 1, 6, 7f, 11 stereocilia, 8 Integrated visual-vestibular-proprioceptive assessment, 166–167 Internal auditory canal (IAC), 21 Intratympanic gentamicin delivery of, 192–193 dose of, 193 end point, 193–194 Iodine, 141 Iron-chelating agents, 76–78, 172–173, 215 J JNK, 95 K Kainic acid, 9t Kanamycin availability of, 83t cochlear ototoxicity, 114–115 interacting with cisplatin, 65–66

218

Index

Ketoconazole, 137 Ketolides, 103 Klebsiella pneumoniae, 130 L Laboratory vestibular tests advantages and limitations of, 165t Labyrinthine artery, 4 Lateral olivocochlear bundles, 6, 10 Lawyers information provided to, 207–210 Lead ABR, 36 ototoxicity of, 36 Left posterior canal (LPC), 162f Legal research, 202, 208 Leukopenia with cisplatin, 61 Linear acceleration receptor, 22f Lipoic acid, 176 L-Met, 176, 177 L-N-acetylcysteine (L-NAC), 177 Loop diuretics, 215 ototoxicity of, 42–46 LPC, 162f Lymphoma cisplatin for, 56 M Macrolides, 101–104, 215 Macular stimulation, 23f Malaria Plasmodium falciparum, 33 MAPK cell death signaling pathway, 173–175 M-cresyl acetate solution (Cresylate), 135 for otomycosis, 137t Medial olivocochlear bundles, 6, 10 Medical negligence, 200–202 evidence supporting action of, 208 Medicolegal action harm leading to, 209 Membranous labyrinth, 1, 3f Meniere’s disease defined, 184 demographics of, 184 Garasone ear drops, 125f vestibular ablation, 124 intratympanic gentamicin, 191–195 history of, 191 mechanism of, 185 natural history of, 184–185 pathophysiology of, 184 systemic treatment of, 184–189 complications of, 186–188 indications for, 185 review of, 185–186 topical gentamicin, 124 vestibulotoxicity, 165 Mercury, 35–36 deafness, 35–36 hearing loss, 35–36 Mesna, 176 Metabotropic receptors, 9t

Methionine, 172 Methylthiobenzoic acid, 176 4-methylthiobenzoic acid (MTBA), 176–177 Miconazole for otomycosis, 137t ototoxicity of, 112 Micromonospora, 82 Mitogen-activated protein kinase (MAPK) cell death signaling pathway, 173–175 Molcer, 142 Monitoring vestibular ototoxicity, 161–168 Motion physiology of, 23f MTBA, 176–177 Mycobacterium avium, 102 Mycobacterium tuberculosis streptomycin for, 84 Mycoplasma, 130 N Nausea with carboplatin, 67 with cisplatin, 61 Neck cancer cisplatin for, 54–55 Necrosis, 94–95, 170 Nedaplatin, 67–68 clinical pharmacology, 67 clinical status of, 62t clinical use of, 67 indications for, 67 mechanisms of action, 67 toxicity of, 67 nonotological, 67 Negligence patient grounds for, 205–206 Neisseria meningitidis, 130 Neomycin availability of, 83t cochlear ototoxicity, 114–115 ototoxicity of, 116 Neomycin/polymyxin B drops, 130 Netilmicin availability of, 83t Neural integrator, 22 Neuromuscular blockade aminoglycosides, 86 Neurotrophin family, 175 Neurotrophin type-3, 175 Neurotrophin type 4/5, 175 NMDA receptors, 9t Node of Ranvier, 8 Noise aminoglycosides ototoxicity, 154 cisplatin ototoxicity, 154 Nonmelanotropic adrenocorticotropic hormone (ACTH)/MSH, 178 Non-NMDA glutamate receptors, 9t Non-small cell lung cancer cisplatin for, 56 Nonsteroidal anti-inflammatory drugs (NSAID), 215 ototoxicity of, 33

Norepinephrine salicylates, 31 NSAID, 215 ototoxicity of, 33 Nuclear inherited predisposing genes, 146–147 Nystagmus, 24 characteristics of, 25t optokinetic, 166 post-head-shake, 167 vestibular-induced fixation of, 26 Nystatin, 136 for otomycosis, 137t ototoxicity of, 112 O OAE. See Otoacoustic emissions (OAE) Ofloxacin middle ear effects of, 108 tympanic membrane, 110f OHC. See Outer hair cells (OHC) OKN, 166 Optokinetic nystagmus (OKN), 166 Organ or Corti, 5–7, 6f, 7f, 11, 14 neurotransmitters within, 9t receptors within, 9t Oscillopsia test, 25t, 26 Oscillopsia testing, 167 Osseous labyrinth, 1 Osseous spiral lamina, 11 Otitis media acute, 121 chronic suppurative, 121 Otoacoustic emissions (OAE), 10, 154–155 mechanics of, 14–15 salicylates, 30–31 Otomize, 142 Otomycosis, 134 acetic acid for, 137t alcohol for, 137t boric acid for, 137t clotrimazole for, 137t debridement, 134 gentian violet for, 137t miconazole for, 137t nystatin for, 137t tolnaftate for, 137t Otoprotective therapies, 170–179 Ototopical agents middle ear effects of, 107–112 Ototopical antibiotics American Academy of Otolaryngology– Head and Neck Surgery Consensus Panel Recommendations, 213 Ototoxic agents, 215 Ototoxicity audiologic monitoring for, 153–157 clinical topical, 121–122 defined, 93–94 medicolegal aspects of, 198–206 monitoring schedules, 156 pediatric monitoring, 155–156 testing for clinical trials, 156

Index Outer hair cells (OHC), 1, 6, 6f, 11 stereocilia, 5–7, 12 Ovarian cancer cisplatin for, 52–54, 52t paclitaxel-carboplatin vs. paclitaxelcisplatin, 53t Oxaliplatin, 68 chemistry of, 68 clinical pharmacology of, 68 clinical status of, 62t clinical use of, 68 indications for, 68 mechanism of action, 68 toxicity of, 68 Ozolinone ototoxicity of, 46 P Paclitaxel-carboplatin vs. paclitaxel-cisplatin for ovarian cancer, 53t Paromomycin availability of, 83t Patient harm medicolegal action, 203–204 Pediatric ototoxicity monitoring, 155–156 Penicillin G middle ear effects of, 107–111 Peptides, 178 Perilymph, 3 electrolytes ethacrynic acid, 43 Peripheral vestibular system, 21–22 anatomic organization of, 21f Physician follow-up vs. patient responsibility, 209 Physician responsibility patient’s case for, 204 Pifithrin, 178 Pillar cells, 6, 7 Piretanide ototoxicity of, 46 Plasma fluid composition of, 2t Plasmodium falciparum malaria, 33 Platinum compounds chemotherapeutic discovery history of, 60–61 clinical status of, 62t in clinical trial, 68 otoprotection, 69 otoprotection research, 179 ototoxicity of, 60–69 toxicity profiles of, 63t tumor resistance, 68–69 Polyenes, 136 Polymyxin B, 116 animal studies, 129 Polymyxin E, 116 Polymyxins, 128–130 animal studies, 129–130 historical development of, 128 mechanism of action, 128

proprietary agents containing, 128–129 recommendations for safe use, 130 therapeutic use, 128 toxicity of, 129 Polysorbate, 137 Position-vestibular-pause (PVP), 161 Post-head-shake nystagmus, 167 Post-tympanostomy tube otorrhea (PTTO), 121 Potassium sorbate, 137 Preparatory solutions topical toxicity, 204, 209–210 Profound hearing loss preventable, 204–205, 210 Propylene glycol, 135, 136, 142 tympanic membrane, 111f Prostaglandin salicylates, 31 Proteus, 130 Pseudomonas, 83, 84 Pseudomonas aeruginosa, 121, 130 PTTO, 121 Pure tone air-conduction thresholds, 153 Purkinje’s cells, 22 PVP, 161 P2X2, 9t Q Quaternary ammonium compounds, 141–142 Quinine hearing loss, 34 ototoxicity of, 33–34 biochemistry of, 34 historical overview of, 33–34 manifestations of, 34 mechanisms of, 34–35 morphology of, 34 physiology of, 34 pharmacokinetics, 34 Quisqualic acid, 9t R Rickettsia, 130 Right superior canal (RSC), 162f Rotating chair tests advantages and limitations of, 165t Rotational testing, 166 Round window membrane topical gentamicin, 122 Round window membrane (RWM), 14 antifungals, 135 cross section, 118f Round window niche cross section, 118f RSC, 162f RWM, 14 antifungals, 135 cross section, 118f S Salicin, 28 Salicylates, 173 arachidonic acid metabolism, 31 hearing loss, 29, 32

219

historical overview, 28 norepinephrine, 31 ototoxicity of, 28–32 biochemistry of, 31–32 manifestations of, 32–33 mechanisms of, 29 pathology of, 29–30 physiology of, 30–31 pharmacokinetics, 28–29 prostaglandin, 31 Salviae miltiorrhizae, 173 Satraplatin, 68 clinical status of, 62t Savlon, 141 Scala media, 2, 2f fluid composition of, 2t Scala tympani, 2, 2f, 14 fluid composition of, 2t Scala vestibuli, 2f, 14 fluid composition of, 2t Scarpa’s ganglion, 21 SCC, 21 Schwalbe’s nucleus, 22 Semicircular canals (SCC), 21 Sensorineural hearing loss (SNHL) with deferoxamine, 76, 77t ethacrynic acid, 42 Sensory cells, 1 Skin preparation, 142 Small cell lung cancer cisplatin for, 55–56 SNHL with deferoxamine, 76, 77t ethacrynic acid, 42 SOAE, 14 SOD, 173 Sodium salicylate, 28, 177 Sodium thiosulfate, 69, 176–177 Solvents, 136 SP, 12, 13 Spectinomycin availability of, 83t Speech discrimination, 153 Spinal vestibular nucleus, 22 Spin-trapping agents, 171–172 Spiral ligament, 5 Spiral modiolar artery, 4 Spiral modiolar vein, 4 Spontaneous otoacoustic emissions (SOAE), 14 Squamous cell head and neck cancer cisplatin for, 54–55 Standards of care, 198–200 vs. community standard of care, 199–200 defined, 199 Steroidal anti-inflammatory agents, 108–109 Streptococcus pneumoniae, 130 Streptomyces, 82 Streptomycin availability of, 83t

220

Index

cochlear ototoxicity, 114–115 for Mycobacterium tuberculosis, 84 Streptomycin-induced vestibulotoxicity, 185 Streptomycin sulfate for Meniere’s disease, 185–186 Strial marginal cells, 11 Stria vascularis, 5f, 11 Sulfacetamide middle ear effects, 108 Summating potential (SP), 12, 13 Superoxide dismutase (SOD), 173 Surgeon warning, 205 Surgical disinfectants, 140–143 topical toxicity, 204, 209–210 Systemic gentamicin ototoxicity case study, 200 information provided to lawyers, 207–208 T Tanshinone, 173 Tectorial membrane, 5, 12 Testicular cancer bleomycin for, 51t chemotherapy of, 51t cisplatin for, 51–52, 51t etoposide for, 51t metastatic seminomas, 52 Tetrahymena, 145, 146 Thiol compounds, 176–177 Thrombocytopenia with carboplatin, 66–67 with cisplatin, 61 with nedaplatin, 67 Tinnitus monitoring for, 156–157 quinine, 34 salicylates, 32–33 TOAE, 14, 155 Tobradex ototoxicity of, 116 Tobramycin availability of, 83t ototoxicity of, 116 Tolnaftate, 137 for otomycosis, 137t ototoxicity of, 112 Topical aminoglycoside experimental studies, 122–123 ototoxicity of individual variability, 125 vestibulotoxicity clinical studies, 123–125

Topical aminoglycosides cochlear ototoxicity, 114–119 drop ototoxicity, 208–209 ototoxicity of risk factors for, 209 round window membrane, 122 vestibular toxicity of, 121–126 Topical antibiotics, 215 Topical antifungals, 134–138 application of, 137 for otomycosis, 137t Topical gentamicin drop ototoxicity, 203 Meniere’s disease, 124 ototoxicity of, 208–209 clinical indications, 208 round window membrane, 122 vestibular ablation Meniere’s disease, 124–125 vestibular ototoxicity, 123 Topical preparations ototoxicity of, 121–122 solvents used in, 109–110 Torsemide ototoxicity of, 46 Transforming growth factor-alpha, 175–176 Transient evoked otoacoustic emissions (TOAE), 14, 155 Tri-Adcortyl, 142 Trimethoprim-sulfacetamide-polymyxin B (TSP) human studies, 130 TSP human studies, 130 Tunnel of Corti, 7 Tympanic membrane cholesteatoma, 112f ofloxacin, 110f propylene glycol, 111f sulfacetamide, 108f U Utriculoendolymphatic valve of Bast, 3 V Vancomycin, 215 VEPT chlorhexidine, 140 Vertigo control of, 192t quinine, 34 Vestibular dysfunction clinical manifestations of, 24–25

Vestibular evoked potentials (VEPT) chlorhexidine, 140 Vestibular function clinical tests of, 25–26, 25t Vestibular loss unilateral neurophysiological basis for compensation, 162–163 unilateral peripheral, 24–25 Vestibular monitoring anatomical basis for, 161–162 pathophysiologic basis for, 163 Vestibular nerve, 22 Vestibular neurons, 162f Vestibular ototoxicity monitoring, 161–168 Vestibular system anatomy of, 21–23 applied physiology, 23 peripheral, 21–22 anatomic organization of, 21f physiology of, 20–27 schematic representation of, 20f Vestibular tests advantages and limitations of, 165t bedside clinical, 167 Vestibulo-ocular reflex (VOR), 20 gain, 24 horizontal wiring of, 162f laboratory evaluation of, 165 neural basis of, 161–162 suppression test, 25t, 26 Vestibulotoxicity monitoring for, 156–157, 164–165 streptomycin-induced, 185 Vinblastine ototoxicity of, 79 Vinca alkaloids ototoxicity of, 78–79 Vincristine ototoxicity of, 78–79 Vinorelbine ototoxicity of, 79 Visual loss bilateral peripheral, 25 Vitamin E (alpha-tocopherol), 177 VOR. See Vestibulo-ocular reflex (VOR) VoSol, 142 W Willow bark, 28 Word recognition, 153 Word recognition testing, 154

Ototoxicity

Recommend Documents