Training & Testing
Game-based Training in Young Elite Handball Players
Authors
M. Buchheit1, 3, P. B. Laursen2, J. Kuhnle3, D. Ruch4, C. Renaud4, S. Ahmaidi 1
Affi liations
1
Key words ▶ small-sided games ● ▶ handball ● ▶ intermittent endurance ● index ▶ high intensity interval ● training
Faculté des sciences du sport, Laboratoire de Recherche EA -3300: Adaptations Physiologiques à l ′Exercice et Faculté Réadaptation à l′Eff ort, ort, Amiens, France 2 School of Exercise, Biomedical and Health Sciences, Edith Cowan Universit y, Joondalup, Australia 3 Sport Development and Analysis, Myorobie Association, Montvalezan, France 4 Picardie Handball League, Elite Player Training Center, Center, Pont Sainte Maxence, France
Abstract ▼
This study compared the e ff ect ect of high-intensity interval training (HIT) versus speci fic gamebased handball training (HBT) on handball performance parameters. Thirty-two highlytrained adolescents (15.5 ± 0.9 y) were assigned to either HIT (n = 17) or HBT (n = 15) groups, that performed either HIT or HBT twice per week for 10 weeks. The HIT consisted of 12– 12–24 × 15 s runs at 95 % of the speed reached at the end of the 30– 30 – 15 Intermittent Fitness Test (V IFT) interspersed with 15 s passive recovery, while the HBT consisted of small-sided handball games performed
Introduction ▼
accepted after revision October 1, 2008 Bibliography DOI 10.1055/s-0028-1105943 Published online: February 6, 2009 Int J Sports Med 2009; 30: 251– 251 –258 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0172-4622 Correspondence Dr. M. Buchheit
Faculté des sciences du sport Faculté Laboratoire de Recherche EA-3300: Adaptations Physiologiques à l′Exercice et Réadaptation à l′Eff ort ort Allé All ée P Grousset 80025 Amiens France Tel.: + 333/ 333/22 22//82 89 36 Fax: + 330 330/90 90//24 34 44
[email protected]
“Handball is a strenuous contact Olympic team sport that places emphasis on running, jumping, sprinting, arm throwing, hitting, blocking, and pushing”” [16]. pushing [16]. In addition to technical and tactical skills, it has recently been shown that anthropometric characteristics and high levels of strength, muscle power, and handball throwing velocity are the most important factors for successful participation at the elite handball level [17, 29]. 29]. Nevertheless, the importance of aerobic capacity for successful handball play should not be underestimated. Indeed, at least 90 % of the energy released during a 60-min match must be aerobically driven, and players run about 4 to 6 km at a mean intensity close to 80 –90 % of maximal heart rate [25]. [25] . Significant associations between maximal oxygen uptake (V̇ O2max) and playing level l evel have also als o been be en shown s hown [17, [1 7, 29] 29].. Although handball playing itself has a direct eff ect ect on most determinant qualities, it is today believed that elite handball players must perform additional handball-spe handball-speci cific conditioning, including strength, strength , power, speed, agility and high-intensity aerobic intermitte intermittent nt training [10, [ 10, 11, 16, 22]. 22] .
over a similar time period. Before and after training, performance was assessed with a counter movement jump (CMJ), 10 m sprint time (10 m), best (RSAbest) and mean (RSAmean) times on a repeated sprint ability (RSA) test, the V IFT and the intermittent endurance index (iEI). After training, RSAbest ( − 3.5 ± 2.7 %), %), RSAmean ( − 3.9 ± 2.2 %) %) and VIFT ( + 6.3 ± 5.2 %) %) were improved (P < 0.05), but there was no di ff erence erence between groups. In conclusion, both HIT and HBT were found to be eff ective ective training modes for adolescent handball players. However, HBT should be considered as the preferred training method due to its higher game-based specificity.
Evaluation of the physical performance characteristics of elite handball players should also be conducted with specific tests that replicate the specific movement patterns and physiological characteristics inherent to the game. Thus, using intermittent shuttle tests, such as the 30– 30 –15 Intermittent Fitness Test (30– (30 –15IFT, which elicits V̇ O2max while still taxing anaerobic pathways [7]), [7] ), the intermittent endurance index (iEI, which assesses intermittent aerobic endurance [8]) [8] ) and repeated repeat ed shuttle-sprint ability tests [10, 21] , may be of relevance. Training at or near V̇ O2max is thought to be an eff ective ective training stimulus to improv improve e aerobic fitness [26]. [26]. The most common way to achieve this is by performing high-intensity interval training [4, 23, 26]. 26]. High-intensity interval training is eff ective ective because intensity can be individualised and controlled using a reference workload, such as the running velocity associated with V̇ O2max (vV̇ O2max [4, 26] 26],) ,) or the speed reached at the end of the 30– 30–15IFT (V IFT), an alternative test for team sport athletes [6]. [6] . As run training can be perceived perceive d as being unpleasant by some handball players (personal observations), and because maintaining technical skills is primordial to
251
252 Training & Testing
successful handball performance [9, 16], small-sided gamebased handball-specific training may be an alternative means for improving a player’s aerobic capacity whilst still maintaining other important handball components, such as agility, reaction time, and hand-eye coordination. Through using this type of training, training time with the ball will be maximised, while training motivation should remain high [9]. Although acute physiological responses [18, 19, 28] and long term e ff ects on aerobic function [20] of small-sided soccer games are well documented, there is very little data on the e ff ectiveness of smallsided handball games [9]. Moreover, no data is presently available concerning the respective e ff ects of specific versus generic training on repeated sprint ability or intermittent endurance capacity. It has recently been shown that specially designed handball-specific games are an e ff ective means of achieving a high percentage of V̇ O2max during training, and may be perceived as being less painful by athletes compared with highintensity interval training [9] . Nevertheless, the negative correlation shown between individual cardiorespiratory fitness parameters and the percentage of V̇ O2max sustained during handball games [9] suggests that small-sided handball games might not represent an appropriate exercise stimuli for individual players (i.e., ceiling e ff ect for the fittest players and overload for those with poor aerobic capacities). As such, it is uncertain as to whether or not small-sided handball training is as e ffi cient a training stimulus as high-intensity interval training throughout the full macrocycle of the competitive season in adolescent players [11, 16]. In light of the lack of research in this area, the aim of the present study was to compare the e ff ects of high-intensity interval training versus specific small-sided handball training on the fitness performance characteristics of young elite male and female handball players. As the tight control of exercise training intensity using individualised high-intensity interval training is known to be a successful stimulus for enhancing aerobic performance, and because high-intensity interval training will also stress anaerobic pathways, we hypothesized that it would be more effi cient than game-based handball training at improving indices of (supra)maximal aerobic power (i.e., V IFT and time-toexhaustion during supramaximal intermittent runs). In contrast, due to the speci fic movement patterns required for successful handball play (e.g., repeated sprints and jumps, all interspersed with incomplete and active recovery) and because of the higher time spent at or near V̇ O2max during game-based handball training compared with high-intensity interval training [9] , we expected repeated sprint ability and the intermittent endurance index (iEI; time to exhaustion during intermittent runs) to be more enhanced by specific handball training.
Materials and Methods ▼
Subject recruitment Based on the assumption that a between group di ff erence in VIFT of 1 ± 0.9 km.h − 1 is meaningful [10], we used Minitab 14.1 Software (Minitab Inc, Paris, France) to determine that a sample size ≥ 19 subjects per group would provide a statistical power of 0.8 at an alpha level of 0.05. Thirty-eight trained female and male adolescent handball players, (girls, n =18 and boys, n =20) who trained 10 ± 2.1 h · wk − 1 in a regional centre, were recruited to participate in the study. Their maturational status was esti-
experienced investigator [24]. They were all free of cardiovascular and pulmonary disease and were not taking any medications. The study, which was approved by the local research ethics committee, conformed to the recommendations of the Declaration of Helsinki. Participants and their parents gave voluntary written informed consent to participate in the experiment.
Experimental overview and randomization Participants were divided into two training groups that performed, in addition to their usual training sessions, either short repeated high-intensity intermittent running (HIT; n = 19; girls = 9) or small sided handball games (HBT; n =19; girls = 9). Subjects within each group were matched according to their maturation level, initial athletic performance and years of practice (which was e ff ective for 2 × 12 players). The remaining 14 players were randomly added to each group while assuring that both groups displayed equivalent mean values for pre-cited maturation and performance parameters. Tests were performed on an indoor synthetic track two weeks prior to the commencement of training, and within two weeks following the training period. Ambient temperature for all testing and training ranged ▶ Table 1). Tests included a counter movement from 18 to 22 ° C ( ● jump (CMJ), a 10-m sprint time (10m), a repeated sprint ability test [10], a graded aerobic intermittent test (30–15IFT) [6, 7], and the assessment of intermittent endurance capacity via three maximal intermittent tests performed in a random and balanced order for each subject [8]. Players were familiarized with the exercise procedures prior to commencement of each test, were told not to perform intense exercise on the day prior to a test, and to consume their (usual) last meal at least 3 h before the scheduled test time. During all tests, HR was continuously recorded with a Polar S810 HR monitor (Polar Electro, Kempele, Finland) and an electrode transmitter belt (T61, Polar Electro, Kempele, Finland) fitted to the chest of each subject as instructed by the manufacturer.
Training intervention Subjects performed two speci fic training sessions per week (Monday and Wednesday for girls, Tuesday and Thursday for the boys), in addition to their usual training requirements, for five and four consecutive weeks (separated by one ‘recovery’ week). Training programs followed a typical periodized plan that incorporated progression, prevented overreaching and used a short ▶ Table 1). The tapering period to maximize final performance (● HIT program consisted of 6–12 min of intermittent running for 15 s (95 %VIFT) interspersed with 15 s of passive recovery, whereas specific HBT program consisted of small handball games (2 to ▶ Table 1) [9]. As described pre4 × 2-min 30 s to 4-min games, ● viously [9], small group play was organised in 4-a-side teams, excluding goalkeepers. Coaches encouraged players to achieve a high intensity of exercise during these games [19, 28]. Typical handball rules were simpli fied to avoid game breaks that would unnecessarily reduce exercise intensity. For example, dribbling and defence contacts were not allowed, infringements of minor technical rules (i.e., ‘walking’, ‘double dribble’) were not sanctioned, throw-on after a goal was immediately made by goalkeepers from their 6-m area, and coaches were always available to immediately replace the ball when it was kicked away from the playing area. Finally, all four players had to be in the opponent half of the court for a goal to be validated [9] . As the V IFT is always higher than one’s vV̇ O2max, HIT performed at 95 % of V IFT ̇
Training & Testing
Table 1 Training programs for game-based handball training (HBT) and high-intensity interval training (HIT) groups. HBT Girls Pre tests
Week -2 Week -1 Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Week 8 Week 9 Week 10 Week 11 Week 12
Training period
Post tests
HIT Boys
2× 2× 2× 2× 2×
[2 × 2′30 [2 × 2′45 [3 × 2′40 [2 × 3′15 [3 × 3′20
] ] ] ] ]
2× 2× 2× 1×
[2 × 3′15 [3 × 2′40 [3 × 2′40 [2 × 2′30
] ] ] ]
Girls
Boys
30–15IFT and explosive, sprint, RSA and throwing velocity tests iEI determination (Three intermittent runs to exhaustion) 2 × [2 × 3′15 ] 2 × [5′30 15 (90 %)-15 p] 2 × [7′ 15 (92 %)-15 p] 2 × [2 × 3′45 ] 2 × [6′ 15 (90 %)-15 p] 2 × [8′ 15 (92 %)-15 p] 2 × [3 × 3′] 2 × [9′ 15 (90 %)-15 p] 2 × [10′ 15 (92 %)-15 p] 2 × [2 × 4′15 ] 2 × [7′ 15 (90 %)-15 p] 2 × [9′ 15 (92 %)-15 p] 2 × [3 × 3′40 ] 2 × [11′ 15 (90 %)-15 p] 2 × [12′ 15 (92 %)-15 p] Regeneration* 2 × [2 × 3′30 ] 2 × [7′ 15 (92 %)-15 p] 2 × [7′30’’ 15 (93 %)-15 p] 2 × [2 × 4′15 ] 2 × [9′ 15 (92 %)-15 p] 2 × [9′ 15 (93 %)-15 p] 2 × [2 × 4′15 ] 2 × [9′ 15 (92 %)-15 p] 2 × [9′ 15 (93 %)-15 p] 1 × [2 × 2′45 ] 1 × [5′30 15 (95 %)-15 p] 1 × [6′ 15 (100 %)-15 p] 30-15IFT and explosive, sprint, RSA and throwing velocity tests iEI determination (Three intermittent runs to exhaustion)
Description of pre- and post-training test scheduling and details of training contents over the 10-week training period for girls and boys in handball speci fic (HBT) and high intensity interval run training (HIT) groups. Between handball games: 30 s of passive recovery. During HIT, ‘ p’ stands for passive recovery. Handball training in week 1 listed for girls as: ‘2 × [2 × 2′30 ]’ refers to 2 sessions per week, including 2 games of 2 min 30 s, interspersed with a 30-s passive recovery period. HIT training in week 1 listed for girls as: ‘2 × [5 ′30 15 (90 %)-15 p]’ refers to 2 sessions per week, consisting of 15-s runs at 90 % of V IFT interspersed with 15-s passive recovery periods, for a total duration of 5 × min
30 s. *For training content of the regeneration week, please refer to METHODS
[6]. Because time-to-exhaustion at V̇ O2max has been show to be slightly higher in adolescent males than in age-matched girls [3] , all specific training exercises (i.e., HIT or HBT) were on average ▶ Table 1). HIT and ~10 % shorter for girls compared to boys ( ● HBT programs were matched for exercise duration at each training session. Except for the HIT and HBT, all other training sessions amongst the subjects were kept similar in both groups (in addition to a weekly 1 h competitive handball game that participants performed with their own club, common total training time was 9.5 ± 2.1 h · wk − 1 and 8.5 ± 1.8 h · wk − 1 for boys and girls, respectively; including technical, tactical, strength and speed training). Strength training consisted of ~1 h of weight lifting for upper and lower limbs (65–85 % of 1RM). Speed training consisted of two 20-min sessions, including athletic drills, 8 –10 standing starts of 3–5 m sprints and ≈10 in-depth jumps ( < 15 cm). The remaining training time was devoted to handball training. Pilot studies conducted prior to the present study showed that technical and tactical training sessions can be qualified as intermittent and moderate-intensity aerobic exercise (45–75 % maximal oxygen uptake performed over 1 –1.5 h). During the recovery week (week 6; total training time = 2 ± 1.1 h · wk − 1), athletes were asked to perform two to three moderate training sessions (i.e., 40-min Fartlek training runs, where athletes were asked to run at a subjective 6–7 level on a 0–10 Borg scale, or any moderate recreational sport activity).
Lower limb explosive power test Lower limb explosive power was assessed using a vertical countermovement jump with flight time measured by an Optojump (Ergojump, Globus Italia, Codogne, Italy) to calculate jumping height (CMJ; cm). Since this method of measuring vertical power can be biased by some methodological flaws (notably landing with leg flexion), an experienced investigator validated each trial visually. CMJ was performed three times, separated by 4 5 s of passive recovery, and the best performance was recorded.
Speed tests Speed quality was evaluated by a 10 m standing-start run (10 m; s) recorded with photoelectric cells (Wireless Timing-Radio Controlled, Brower Timing System, Colorado, USA).
Repeated sprint ability The repeated sprint ability (RSA) test was assessed through 6 reps of maximal 2 × 15 m shuttle sprints ( ≈6 s) departing every 20 s [10]. During the 14-s recovery between sprints, subjects had a passive standing recovery. Three seconds before starting each sprint, subjects were asked to assume the ready position and await the start signal. Strong verbal encouragement was provided to each subject during all sprints. This test was adapted from previous cycling and running tests [15, 21, 30] that have been shown to be reliable and valid to estimate RSA [30] . Three scores were calculated for the RSA test: the best sprint time (RSAbest; s), usually the first sprint; the mean sprint time (RSAmean; s) and the percent sprint d ecrement (RSAdec; %) calculated as follows: 100–(total time/ideal time × 100); where the ideal time = 6 × RSAbest [30].
Maximal graded aerobic test Maximal aerobic performance for each subject was assessed using a 30–15 Intermittent Fitness Test (30–15IFT) [7]. The 30– 15IFT consists of 30-s shuttle runs interspersed with 15 -s passive recovery periods. This intermittent shuttle field test elicits peak HR and maximal oxygen uptake [7], and the reliable final running speed (VIFT; intraclass correlation coe ffi cient = 0.96; typical error = 0.33 (95 CI, 0.26–0.46) km.h − 1) has been shown to be an accurate tool for individualizing intermittent shuttle running exercise [6]. For this test, velocity was set at 8 km.h − 1 for the first 30-s run and speed was increased by 0.5 km.h − 1 every 30-s stage thereafter. Subjects were required to run back and forth between two lines set 40 m apart at a pace which was governed by a prerecorded beep. The prerecorded beep allowed the subject to adjust their running speed within a 3-m zone placed in the middle and at each extremity of the field. During the 1 5-s recovery
253
254 Training & Testing
closest line (at either the middle or end of the running area, depending on where their previous run had stopped); they would start the next run stage from this line. Subjects were instructed to complete as many stages as possible, and the test ended when the subject could no longer maintain the required running speed or when the subject was unable to reach a 3-m zone in time with the audio signal on three consecutive occasions. The velocity (km.h − 1) and a maximal HR attained during the last completed stage were determined as the subject ’s V IFT and HRmax, respectively. The field test was thought to be maximal if the following criteria were met: 1) HR max attained within 10 beats.min − 1 of the age-predicted maximum; and 2) volitional fatigue [6].
Maximal intermittent tests-to-exhaustion All tests were preceded by a supervised and standardized warmup consisting of 5 min running at 45 % of V IFT along with a few athletic drills and short bursts of progressive accelerations on the track [8]. Tests began two min after this warm-up. For the three intermittent shuttle exercise tests, subjects were asked to continually repeat as many intervals as possible [10] . The tests consisted of 15-s (15/15) runs at 100 % (Tlim100 %VIFT), 95 % (Tlim95 %VIFT) and 90 % (Tlim 90 %VIFT) of VIFT alternating with 15-s recovery periods ran at 45 % of V IFT. Although recovery intensity was originally passive [8], we chose an active recovery to reduce exercise duration [14] and save time for handball training content. During the 15-s exercise period, athletes were required to run back and forth over the 40-m area so that they covered the distance determined according to their V IFT (marked with cones). Run interval pace was provided by a digital timer that produced a sound every 15 s from the start to the end of the exercise interval. The recovery distance allowed participants to arrive at the starting line within 12–13 s. Two seconds prior to the commencement of each high intensity run, subjects were asked to assume the ready position and await the start signal. Subjects were verbally encouraged to run “to exhaustion” and were considered fatigued when: 1) they could no longer maintain the required running speed, 2) their maximal heart rate (HR) was greater than 85 % of its age-predicted maximal HR (HR max) and 3) their rating of perceived exertion was greater than 8 on a 10 point Borg scale [8]. Eff ective exhaustion time (s) was measured for each intermittent exercise interval, including the recovery period. Three participants stopped running without presenting these criteria and were asked to perform the test again the following week. The repeated test scores of these participants were included in the analysis.
in relative changes in performance were tested using a Student’s unpaired t-test. Student’s unpaired t -tests were also used to examine diff erences between groups for baseline and final measurements. For all analyses, the level of signi ficance was set at P < 0.05. Due to the limited sample size of the present study, our statistical analyses may have been slightly underpowered for some of the performance-related variables. In order to better interpret our results, when no signi ficant eff ects were observed, but a tendency towards signi ficance (P < 0.10) was apparent, eff ect sizes (ES) were calculated [13]. When calculating e ff ect sizes, the pre-training standard deviations for the HBT and HIT groups were applied. If the e ff ect size was large ( ≥ 0.80), but the statistical power was low, the likelihood of a type II error was noted. Linear regressions with Pearson’s coeffi cients were used to establish the respective relationships between improvements in physical performance indices. Since there was no interaction on performance between ‘gender’ and ‘period’ factors (all P > 0.41), relationships are presented with girls and boys pooled. Other polynomial regressions were rejected on the basis of importantly higher residuals. All statistical analyses were carried out using Minitab 14.1 Software (Minitab Inc, Paris, France). Data in text and tables are presented as means and standard deviations ( ± SD), while data in figures are presented as means and standard errors ( ± SE) for clarity.
Results ▼
Subjects Only subjects that participated in > 85 % of all training sessions were included in the final analysis. As a result, six out of the 38 subjects (15 %) were excluded from analysis. Only 32 players (girls: n =16, 15.2 ±0.9 y, 168.6 ± 5.9 cm, 63.1 ±8.1 kg and 24.3 ± 2.9 % body fat, and boys: n = 16, 15.7 ± 0.9 y, 179.6 ± 6.7 cm 71.8 ± 9.9 kg and 13.1 ± 3.6 % body fat) were included in the final analysis. Their maturational status was estimated at Tanner stage III (n = 5 and 7 for girls and boys, respectively), IV (n = 8 and 6) or V (n = 3 and 3). The final sample size for each training group was n = 17 (girls = 8) for HIT and n = 15 (girls = 8) for HBT. As a result of the subject exclusions, some players were no longer matched within both training groups; however, there was no significant diff erence between mean maturation level, initial athletic performance and years of practice between final groups before and after training. The b aseline anthropometric and outcome measures of the study drop-outs were not signi ficantly diff erent from those who completed the study.
Endurance Index for short intermittent runs
HR response to the training sessions
iEI was determined as previously described [8], from the plot of the percentage of V IFT versus the logarithm of exhaustion time (Ln(ET)), with the slope of the resulting linear relationship being ▶ Fig. 2). defined as iEI (●
Mean HR during the first training session in the HBT g roup was 86.8 ± 1.7 % HRmax, which was not significantly diff erent (P = 0.43) than the corresponding HIT group value (87.6 ± 1.1 % HRmax). Peak HR reached during HBT was 91.6 ± 1.2 % HRmax, which was significantly lower (P < 0.01) than during HIT (94.9 ± 0.5 % HRmax). Mean HR during the last HBT session was 85.9 ± 1.5 % HRmax, which was not di ff erent (P = 0.42) than the corresponding values in the HIT group (86.6 ± 1.5 % HRmax). Peak HR reached during the last HBT session was 91.1 ± 1.1 % HRmax, which was significantly lower (P < 0.01)) than with HIT (94.6 ± 1.2 % HRmax).
Statistical analyses The distribution of each variable was examined with the Shapiro-Wilk normality tests. Homogeneity of variance was veri fied by a Levene test. Since a gender interaction was not present when comparing pre- and post-training performances across genders, data were analyzed with girls and boys pooled using a 2-factor repeated measure ANOVA with one between factor (training type; HBT vs. HIT) and one within factor (period; pre-
Training & Testing
Table 2 Physical performance parameters measured in players before (Pre) and after (Post) the 10-wk training period in the handball speci fic (HBT) and high-intensity intermittent training (HIT) groups. HBT (n = 17) 10 m (s) CMJ (cm) RSAbest (s)* RSAmean (s)* RSAdec ( %)* Shoot (km.h − 1) VIFT (km.h − 1)* Tlim100 %VIFT (s)* Tlim95 %VIFT (s)* Tlim90 %VIFT (s)* iEI
HIT (n = 15)
Pre
Post
Pre
Post
2.03 ± 0.10 38.1 ± 7.9 6.19 ± 0.41 6.48 ± 0.40 4.10 ± 1.59 69.3 ± 6.9 18.4 ± 1.5 155 ± 66 274 ± 103 315 ± 133 − 14.5 ± 5.1
2.00 ± 0.09 39.3 ± 7.7 5.96 ± 0.31 6.18 ± 0.35 3.14 ± 0.92 71.3 ± 8.7 19.6 ± 1.4 196 ± 102 289 ± 104 439 ± 178 − 11.9 ± 3.8
2.02 ± 0.16 40.7 ± 9.7 6.27 ± 0.42 6.52 ± 0.42 3.43 ± 1.25 64.4 ± 7.5 17.9 ± 1.8 180 ± 96 275 ± 105 351 ± 111 − 13.8 ± 2.6
2.00 ± 0.14 42.0 ± 8.4 6.05 ± 0.37 6.30 ± 0.36 3.34 ± 1.03 68.5 ± 8.5 18.9 ± 1.3 245 ± 115 307 ± 117 411 ± 133 − 17.4 ± 5.3
Mean ( ± SD) values for 10 m sprint time (10 m), counter movement jump (CMJ), best time during the repeated sprint ability test (RSA best), mean of the sprint during the RSA test (RSAmean), RSA speed decrement (RSA dec), handball throw velocity (Shoot), velocity reached at the end of the 30 –15IFT test (VIFT), times-to-exhaustion running at 100 (Tlim100 %VIFT), 95 (Tlim 95 %VIFT) and 90 (Tlim 90 %VIFT) % of V IFT, and intermittent endurance index (iEI). *: significant ‘period’ eff ect for both groups pooled (P < 0.05)
Fig. 1 Values are means ± SE. Training-induced change in physical performance, 10-m sprint time (10 m), best time during the repeatedsprint ability test (RSA best), mean sprint time during the RSA test (RSAmean), countermovement jump (CMJ), throwing velocity (Shoot) and velocity reached at the end of the 30–15IFT (VIFT) for high-intensity interval training (HIT) and game-based handball training (HBT). RSA decrement (RSAdec) has been excluded for figure clarit y ( + 16.9 ± 70.1 vs. − 12.2 ± 43.3 % for HIT vs. HBT, respectively). *: significant ‘period’ eff ect for both groups pooled (P < 0.05). †: diff erence in relative changes between training groups with large e ff ect size (ES > 0.8).
Changes in physical performance after training Physical performance values before and after each training ▶ Table 2. The 2-factor ANOVA revealed period are presented in ● neither a ‘trainingtype’ eff ect (all P > 0.09), nor a ‘period’ × ‘training type’ interaction (all P >0.20) for any of the measured variables. A ‘period’ eff ect was only observed for RSA best (P = 0.03), RSAmean (P = 0.01), VIFT (P < 0.01), Tlim100 %VIFT (P = 0.04), Tlim95 %VIFT (P = 0.03) and Tlim90 %VIFT (P = 0.02). Relative changes for both training groups were not di ff erent (all P > 0.14, except for RSA mean ▶ Fig. 1). The e ff ect size shown for the di ff erence where P = 0.09, ● between training type, in terms of the relative changes for RSAmean, could be considered as large (E S = 0.79).
Fig. 2 Values are means ± SE. Times-to-exhaustion at 100, 95 and 90 % of VIFT (expressed logarithmically; Ln(ET)) and associated intermittent endurance index (iEI) before and after high-intensity interval training (HIT, triangles) or game-based handball training (HBT, circles).
Intermittent endurance index (iEI) determination Mean HR during the intermittent runs-to-exhaustion before and after training were 89.8 ± 0.7 and 89.4 ± 0.8 % HRmax during Tlim100 %VIFT, 89.4 ± 1.2 and 90.5 ± 0.8 % HRmax during Tlim95 %VIFT, and 90.3 ± 0.9 and 91.5 ± 1.0 % HRmax during Tlim90 %VIFT. There was no significant diff erence between HR values for all intermit-
255
256 Training & Testing
three exercise sets was strong for each individual subject, both ▶ Fig. 2 illusbefore and after training (0.94 < r2 < 1; P < 0.01). ● trates the iEI pooled gender values for both training groups before and after the training period.
While small-sided games have been used religiously by handball coaches for years [9], the present study is the first to investigate, using specific field tests and a controlled study design, a comparison of the e ff ectiveness of small-sided game-based handball training versus high-intensity interval run training on the physical performance capacities of elite young handball players. Due to the di ff erent responses inherent with speci fic game-based handball training and high-intensity interval training, we hypothesized that the high-intensity interval training group would obtain greater improvements in indices of (supra)maximal aerobic power, but that the game-based handball training group would be more efficient at enhancing repeated sprint ability and intermittent endurance. Our results show that both game-based handball training and high-intensity interval training were similarly eff ective at increasing most performance parameters in young elite players, but suggest that game-based handball training games should be used preferentially due to their higher quantity of game-based speci ficity.
with high-intensity interval training may have represented a strong enough stimulus to increase muscular power [2]. Results from the present study revealed a signi ficant increase in RSA after the training period when both groups were pooled (i.e., − 3.5 ± 2.7 % for RSAbest and − 3.9 ± 2.2 % for RSAmean). However, there was neither a signi ficant ‘training type’ nor ‘period’ × ‘training type’ interaction, suggesting that both methods were similarly e ff ective. Nevertheless, it is worth noting that relative changes tended to be slightly higher for handball train▶ Fig. 1, P = 0.09 and ES = 0.79). The lower sample size after ing (● training due to six drop outs reduced the statistical power of our analysis (i.e., statistical power = 0.25 for relative changes in RSAmean), so definitive conclusions regarding the eventual superiority of game-based handball training over high-intensity run training to improve RSA will need to be con firmed in a larger population. The present study is however the first to demonstrate the beneficial eff ects of specific small-sided game-based training on repeated sprint ability. It is likely that the high exercise intensity of the sprints, jumps and direction changes inherent with small-sided games created an e ffi cient stimulus that elicited improvement in RSA. It is also possible that improvement in agility and coordination in response to handball training improved the ability to change direction, which in turn increased repeated shuttle sprint performance [32]. The positive eff ects of high-intensity interval training on repeated sprint ability are in accordance with those previously reported using a similar protocol as that of the present study (short repeated intermittent runs, resulting in a signi ficant 2 % increase in mean repeated sprint performance [10]). However, our results are in opposition to another longitudinal 7-week study conducted in junior soccer players that showed no impact of high-intensity interval training on RSA [5] . It is possible that training backgrounds of the players, high-intensity interval training modality (i.e., straight line vs. shuttle runs, maximal vs. supramaximal intensity, work/rest ratio), and exercise intensity prescription might explain discrepancies in these results. Whereas high-intensity interval training consisted of 4 × 4 min of straight line running at 90 –95 % of maximal HR in the study by Bravo et al. [5], we used a reference speed (VIFT) to determine high-intensity interval training shuttle intensity [6] . Finally, the significant relationship shown between changes in 10-m sprint time and RSA mean (r = − 0.46, P = 0.01) suggests that improvements in speed via supplemental strength and velocity training sessions [12] should also be considered.
Eff ects on speed, explosive power and repeated sprint ability
Eff ects on maximal aerobic power and supramaximal intermittent performance
In the present study, both training regimens tended to be e ffi cient at improving muscular power-related variables that predict handball performance [17, 29]. There was no significant change in 10 m sprint time or jumping height following both ▶ Table 2 and ● ▶ Fig. 1). This is not surpristraining regiments (● ing, since both training programs were speci fically designed to improve aerobic capacities [9]. Nevertheless, in school-adolescents of similar age, high-intensity interval training has shown a concomitant 3 % [2] and 14 % [1] increase in vertical jump performance. While our players were highly trained school-adolescents, subjects from these previously mentioned studies likely had short training histories and moderate physical fitness levels. It is possible that the acceleration and deceleration phases, coupled with the maximal and supra-maximal velocities inherent
Because high-intensity interval training intensity can be controlled with much more precision than that of handball games [9], we hypothesized that improvements in maximal aerobic power would be greater in the high-intensity interval training group. Moreover, because high-intensity interval training is associated with a higher level of anaerobic participation than handball games (i.e., 11.6 vs. 8.9 mmol.l− 1 [9]), we also felt that high-intensity interval training would elicit a more pronounced impact on the long-term development of anaerobic capacity; this in turn, could improve time-to-exhaustion at a supramaximal intensity (i.e.; 100 % of V IFT, ~120 % vV̇ O2max [6]). By contrast, our results showed that both game-based handball training and high-intensity interval training were e ff ective at improving maximal aerobic power, as inferred from equivalent improve-
Relationships between physical performance indices The improvement of RSA best was signi ficantly correlated to the decrement in 10m sprint time (r = 0.37, P = 0.05), but there was no association with changes in CMJ (r = 0.06, P = 0.97). The decrease in RSAmean after training was signi ficantly correlated with the decrease in 10m sprint time (r = − 0.46, P = 0.01), the improvement of VIFT (r = − 0.40, P = 0.04) and T lim90 %VIFT (r = − 0.70, P < 0.001). Correlations between RSAmean and Tlim100 %VIFT and Tlim95 %VIFT were close to signi ficance (r = − 0.37, P = 0.07 and r = − 0.35, P = 0.08 for Tlim100 %VIFT and Tlim95 %VIFT, respectively). There was no association between changes in RSA mean and CMJ (r = − 0.22, P = 0.26). Improvement in Tlim100 %VIFT and Tlim95 %VIFT were strongly related to changes in V IFT (r = 0.58, P < 0.01 and r = 0.57, P = 0.03 for Tlim100 %VIFT and Tlim95 %VIFT, respectively). In contrast, there was no association between changes in Tlim 90 %VIFT and VIFT (r = 0.28, P = 0.22). There was no relationship between changes in all athletic performance indices and iEI (all P > 0.21).
Discussion ▼
Training & Testing
correlation with V̇ O2max [7], it is now apparent that V IFT, in addition to being a relevant tool for the individualization of interval training [6], can be used con fidently to assess maximal aerobic function [7]. While our comparison of high-intensity interval training versus speci fic handball training is novel, our findings parallel an investigation on soccer players that showed small-sided games to be as e ffi cient as high-intensity interval training for improving maximal oxygen uptake [20] . Finally, while obtainment of a high V IFT is associated with high blood lactate accumulation [7], which may be dependent on anaerobic capacity, present results do not support the view that anaerobic pathways might be di ff erently aff ected by high-intensity interval training or game-based handball training, as relative changes in maximal intermittent velocity and supramaximal endurance were similar. Further studies evaluating changes induced by both training regimens on anaerobic capacity via speci fic tests [27, 31] might assist with our understanding in this area.
aerobic training might be a su ffi cient training stimulus to improve maximal aerobic power and intermittent endurance in young handball players. Although the importance of exercise intensity individualization has been recognized for years as a determining factor of successful training, our findings put forward that spontaneous activity during small-sided games, with appropriate rules, can be at least equally e ff ective at enhancing aerobic and anaerobic capacities compared with more complex intermittent training protocols requiring reference tests. In young elite adolescent handball players, we thus recommend that game-based handball training be used over high-intensity interval run training due to its higher speci ficity and valuable impact on handball performance-related parameters.
Acknowledgements ▼
Eff ects on intermittent aerobic endurance V̇ O2max
Because percentage of time spent at or near has been shown to be significantly higher for game-based handball training versus high-intensity interval training (70 vs. 40 % for HBT vs. HIT, respectively [9]), we expected speci fic training to be more efficient than high-intensity interval training at improving timeto-exhaustion during intermittent runs and overall intermittent ▶ Table 1, ● ▶ Fig. 2) were much higher endurance. Values for iEI ( ● than those previously reported in young male adults with similar cardiorespiratory fitness levels (iEI = 6.1 ± 1.2, [8]). Nevertheless, since recovery intensity between intermittent runs was higher (45 % VIFT in the present study vs. passive in the study by Buchheit et al. [8]), it is not surprising that times-to-exhaustion were shorter [14] and consequently iEI, higher. As we also observed near-maximal HR values during all tests and good correlation coe fficients between %VIFT and times-to-exhaustion (0.94 < r2 < 1; P < 0.01), the present study con firms the validity of the iEI assessment when using intermittent runs with active recovery. Changes of all endurance parameters (i.e., times-toexhaustion at various intensities) were similar for both training ▶ Table 2). Thus, in contrast to our hypotheses, it is modalities (● likely that both training programs were similarly e ffi cient at improving intermittent endurance (inferred from changes in iEI). Nevertheless, game-based handball training tended (ES = 0.58) to lower iEI (suggestive of an enhanced endurance), whereas high-intensity interval training tended (ES = − 0.86) to increase the index (suggestive of a loss of intermittent endur▶ Fig. 2); these findings could be related to di ff erences in ance) (● time spent at or near V̇ O2max during each training condition. The equivalent e ff ectiveness of both training regimens on intermittent endurance is in line with the pre-cited study in soccer players [20]; relative changes in indirect indices of aerobic capacity (i.e., lactate threshold and running economy) were also similar for speci fic versus generic training regimens. To conclude, we have showed for the first time the respective long-term e ff ects of speci fic versus generic aerobic exercises on physical performance parameters of elite adolescent handball players. Indeed, performing small-sided handball games twice a week for ten weeks, in addition to usual training commitments, resulted in similar improvements to repeated sprint ability as repeated intermittent high-intensity run training. Our results also suggest that, whatever the exercise type considered (i.e., short vs. long interval training, maximal
The authors thank Irmant Cadjjiov for his assistance with manuscript preparation, and the subjects of the study for their enthusiastic participation.
References 1 Adeniran SA, Toriola AL. Eff ects of continuous and interval running programmes on aerobic and anaerobic capacities in schoolgirls aged 13 to 17 years. J Sports Med Phys Fitness 1988; 28: 260–266 2 Baquet G, Berthoin S , Gerbeaux M , Praagh E Van. High-intensity aerobic training during a 10 week one-hour physical education cycle: eff ects on physical fitness of adolescents aged 11 to 16. Int J Sports Med 2001; 22: 295–300 3 Berthoin S , Baquet G, Manteca F , Lensel-Corbeil G, Gerbeaux M . Maximal aerobic speed and running time to exhaustion for children 6 to 17 years old. Ped Exerc Sci 1996; 8: 234–244 4 Billat LV . Interval training for performance: a scienti fic and empirical practice. Special recommendations for middle- and long-distance running. Part II: anaerobic interval training. Sports Med 2001; 31: 75–90 5 Bravo DF , Impellizzeri FM , Rampinini E , Castagna C , Bishop D, Wislo ff U . Sprint vs. interval training in football. Int J Sports Med 2008; 29: 668–674 6 Buchheit M . The 30–15 Intermittent Fitness Test: accuracy for individualizing interval training of young intermittent sport players. J Strength Cond Res 2008; 22: 365–374 7 Buchheit M , Al Haddad H , Leprêtre PM , Millet G, Newton M , Ahmaidi S . Cardiorespiratory and cardiac autonomic responses to 30–15 Intermittent fitness test. J Strength Cond Res 2008 Dec 3. [Epub ahead of print] 8 Buchheit M , Laursen PB, Millet GP , Pactat F , Ahmaidi S . Predicting intermittent running performance: critical velocity versus endurance index. Int J Sports Med 2007; 29: 307–315 9 Buchheit M , Lepretre PM , Behaegel AL, Millet GP , Ahmaidi S . Cardiorespiratory responses during running and sport-speci fic exercises in handball players. J Sci Med Sport 2008 , [In press] (DOI:10.1016/j. jsams.2007.11.007) 10 Buchheit M , Millet GP , Parisy A, Pourchez S , Laursen PB, Ahmaidi S . Supramaximal training and post-exercise parasympathetic reactivation in adolescents. Med Sci Sports Exerc 2008 ; 40: 362–371 11 Cardoso Marques MA, Gonzalez-Badillo JJ . In-season resistance training and detraining in professional team handball pl ayers. J Strength Cond Res 2006; 20: 563–571 12 Christou M , Smilios I , Sotiropoulos K , Volaklis K , Pilianidis T , Tokmakidis SP . E ff ects of resistance training on the physical capacities of adolescent soccer players. J Strength Cond Res 2006; 20: 783–791 13 Cohen J . Statistical Power Analysis for the Behavioral Sciences. Hillsdale: Lawrence Erlbaum; 1988; 599 14 Dupont G, Moalla W , Guinhouya C , Ahmaidi S , Berthoin S . Passive versus active recovery during high-intensity intermittent exercises. Med Sci Sports Exerc 2004; 36: 302–308 15 Glaister M . Multiple sprint work: physiological responses, mechanisms of fatigue and the influence of aerobic fitness. Sports Med 2005; 35: 757–777
257
258 Training & Testing 16 Gorostiaga EM , Granados C , Ibanez J , Gonzalez-Badillo JJ , Izquierdo M . Eff ects of an entire season on physical fitness changes in elite male handball players. Med Sci Sports Exerc 2006 ; 38: 357–366 17 Gorostiaga EM , Granados C , Ibanez J , Izquierdo M . Diff erences in physical fitness and throwing velocity among elite and amateur male handball players. Int J Sports Med 2005; 26: 225–232 18 Hill-Haas S , Bishop D, Dawson B, Goodman C , Edge J . Eff ects of rest interval during high-repetition resistance training on strength, aerobic fitness, and repeated-sprint ability. J Sports Sci 2007; 25: 619–628 19 Ho ff J , Wislo ff U , Engen LC , Kemi OJ , Helgerud J . Soccer specific aerobic endurance training. Br J Sports Med 2002; 36: 218–221 20 Impellizzeri FM , Marcora SM , Castagna C , Reilly T , Sassi A, Iaia FM , Rampinini E . Physiological and performance eff ects of generic versus specific aerobic training in soccer players. Int J Sports Med 2006; 27: 483–492 21 Impellizzeri FM , Rampinini E , Castagna C , Bishop D, Ferrari Bravo D, Tibaudi A, Wislo ff U . Validity of a repeated-sprint test for football. Int J Sports Med 2008; 29: 899–905 22 Jensen J , Jacobsen ST , Hetland S , Tveit P . Eff ect of combined endurance, strength and sprint training on maximal oxygen uptake, isometric strength and sprint performance in female elite handball players during a season. Int J Sports Med 1997; 18: 354–358 23 Laursen PB, Jenkins DG. The scientific basis for high-intensity interval training: optimising training programmes and maximising performance in highly trained endurance athletes. Sports Med 2002; 32: 53–73 24 Leone M , Comtois AS . Validity and reliability of self-assessment of sexual maturity in elite adolescent athletes. J Sports Med Phys Fitness 2007; 47: 361–365
25 Loftin M , Anderson P , Lytton L, Pittman P , Warren B. Heart rate response during handball singles match-play and selected physical fitness components of experienced male handball players. J Sports Med Phys Fitness 1996; 36: 95–99 26 Midgley AW , Mac Naughton LR. Time at or near VO2max during continuous and intermittent running. A review with special reference to considerations for the optimisation of training protocols to elicit the longest time at or near VO2max. J Sports Med Phys Fitness 2006; 46: 1–14 27 Nummela A, Alberts M , Rijntjes RP , Luhtanen P , Rusko H . Reliability and validity of the maximal anaerobic running test. Int J Sports Med 1996; 17: S97–102 28 Rampinini E , Impellizzeri F , Castagna C , Abt G, Chamari K , Sassi A, Marcora S . Factors influencing physiological responses to s mall-sided soccer games. J Sports Sci 2007; 25: 659–666 29 Rannou F , Prioux J , Zouhal H , Gratas-Delamarche A, Delamarche P . Physiological profile of handball players. J Sports Med Phys Fitness 2001; 41: 349–353 30 Spencer M , Fitzsimons M , Dawson B, Bishop D, Goodman C . Reliability of a repeated-sprint test for field-hockey. J Sci Med Sport 2006; 9: 181–184 31 Vandewalle H , Peres G, Monod H . Standard anaerobic exercise tests. Sports Med 1987; 4: 268–289 32 Young WB, MacDowell MH , Scarlett BJ . Specificity of sprint and agility training methods. J Strength Cond Res 2001; 15: 315–319
