In swimming, performance depends on many factors DRY-LAND STRENGTH TRAINING VS. ELECTRICAL STIMULATION IN SPRINT SWIMMING PERFORMANCE

Transcription

1 DRY-LAND STRENGTH TRAINING VS. ELECTRICAL STIMULATION IN SPRINT SWIMMING PERFORMANCE SÉBASTIEN GIROLD, 1 CHADI JALAB, 1 OLIVIER BERNARD, 1 PIERRE CARETTE, 1,2 GILLES KEMOUN, 1,2,3 AND BENOIT DUGUÉ 1 1 Laboratory of Exercise-Induced Physiological Adaptations (EA 3813), University of Poitiers, Poitiers, France; 2 Department of Physical and Rehabilitation Medicine, Mile trie Hospital, CHU de Poitiers, France; and 3 Saint Marie Foundation Hospital, Paris, France ABSTRACT Girold, S, Jalab, C, Bernard, O, Carette, P, Kemoun, G, and Dugué, B. Dry-land strength training vs. electrical stimulation in sprint swimming performance. J Strength Cond Res 26(2): , 2012 This study was undertaken to compare effects of dry-land strength training vs. an electrical stimulation program on swimmers. Twenty-four national-level swimmers were randomly assigned to 3 groups: dry-land strength training program (S), electrical stimulation training program (ES), and control (C) group. The training program lasted 4 weeks. The subjects were evaluated before training, at end of training program, and 4 weeks later. The outcome values ascertained were peak torque during arm extension at different velocities (from 260 to 180 s 21 ) using an isokinetic dynamometer and performance, stroke rate, and stroke length during a 50-m front crawl. A significant increase in swimming velocity and peak torque was observed for both S and ES at end of training and 4 weeks later. Stroke length increased in S group but not in ES group. However, no significant differences in swimming velocity between S and ES groups were observed. No significant changes occurred in C group. Programs combining swimming training with dry-land strength or electrical stimulation programs led to a similar gain in sprint performance and were more efficient than swimming alone. KEY WORDS weight training, stroke technique, muscular strength INTRODUCTION In swimming, performance depends on many factors wherein some technical and physiological factors play a predominant role. For distances ranging from 50 to 200 m, propulsive efficiency has a major impact Address correspondence to Dr. Sébastien Girold, s.girold@gmail.com. 26(2)/ Ó 2012 National Strength and Conditioning Association on performance (51). This efficiency is mainly because of involvement of muscle contractile qualities and muscle strength of upper limbs (24,43). The lower limbs only participated very slightly (;12%) in propulsion of swimmer (30,49). Several conditioning training methods using swim bench, weight training, or in-water resistance have been described to improve swimmers physical abilities (8,18,48). At high levels of practice and ability, program efficiency mainly depends on training session intensity (4,37) and specificity of coming event (33,46). Generally, gains in sprint performance are consistent within % (8,43,48). For example, Strass et al. (48) showed that press and draw exercises with barbells over a 6-week period at a rate of 4 sessions per week and an intensity of % in relation to 1 maximal repetition led to significant 4.4 and 2.1% increases in performance over 25 and 50 m, respectively. Hawley et al. (24) and Smith et al. (45) demonstrated a strong relationship (r = 0.82 and 0.93, respectively; p, 0.05) between upper body strength with respect to peak torque and sprint swimming performance over 25 yd and 50 m. Pichon et al. (40) analyzed effects of a 3-week electrical stimulation training program on strength of latissimus dorsi muscle and sprint swimming performance. They observed a significant increase in peak torque of arm extensors of, on average, 10 24% at different velocities (from 260 to 360 s 21 ). They also observed an increase in sprint swimming performance of 1.3% for arm-stroking with a pull buoy over 25 m and 1.4% over a 50-m freestyle sprint. The increases in peak torque, measured in eccentric condition (260 s 21 ), were correlated with increased performance. These results show that an electrostimulation program focused on latissimus dorsi could increase sprint swimming performance. An electromyography study undertaken by Clarys (6) revealed extensive involvement of latissimus dorsi in front crawl throughout propulsive phase, in lowering and adducting arm, and in orienting propulsive surfaces. This study also demonstrated importance of triceps brachii, deltoideus, and teres major muscles. Surprisingly, although efficiency of dry-land strength and electrical stimulation training methods on sprint swimming performance are both reported, to our knowledge, VOLUME 26 NUMBER 2 FEBRUARY Copyright National Strength and Conditioning Association Unauthorized reproduction of this article is prohibited.

2 Strength vs. Electrostimulation in Swimming no randomized comparative studies have thus far been performed. Moreover, re is currently no information available regarding how long effects of se training techniques may last. Thus, purpose of this study was to compare effects of a combined dry-land strength training and swimming program with those of a combined electrical stimulation program with same swimming program and to assess wher observed effects last after training period has ended. In a context of time-consuming training, addition of short-term effective methods to regular training may be attractive. Although it is believed that more demanding training is, greater benefit will be, one must also consider training volume of athletes in its entirety to avoid overtraining-related disturbances. METHODS Experimental Approach to Problem The main purpose of this study was to investigate wher dry-land and electrical stimulation strength training methods were able to enhance performances of swimmers in 50-m sprint. It was thus decided to organize exercises such as pull-ups and draws with pulley in dry-land strength training conditions and to use of an electrical stimulation strength training program providing by an electrical stimulator for electrical stimulation strength training. In dry-land strength training and in electrical stimulation training, short and intensive sets were performed to increase maximal strength and anaerobic power, respectively, and thus to improve sprint abilities over 50 m during a 4-week training period. Subjects A group of 24 national-level competitive swimmers (12 men, 12 women; mean 6 SD, age years, height cm, body mass kg,armspan cm) took part in this TABLE 1. Main characteristics of S, ES, and C groups of French national 1 swimmers and ir weekly training volume.* S group study. The swimmers signed an informed consent form and participated in study on a voluntary basis. This study was approved by our University Committee on Human Research. Swimmers trained an average of hwk 21 in water and an average of 3 3 1hwk 21 on dry land based on a specific conditioning program, in same swimming club, and under same conditions. Globally, training volume was around 30 kmwk 21 for each of our swimmers. They were all swimmers sprinters and ir personal records in 50 m were seconds (75.5% of world record). They were all French national 1 level swimmers who had been able to maintain ir level at least for last for 4 years. Procedures Before randomization, matching was performed according gender, age, and performance (50-m sprint). Then, swimmers were randomly divided into 3 groups: (a) dry-land strength (S), (b) electrical stimulation (ES), and (c) control (C) groups. The average of best performance (expressed in seconds as mean 6 SD) in 50-m freestyle for every group were , , and , respectively. These times correspond to 78.5, 75.25, and 73.78% better world record for this distance; respectively. Each group performed 10 swimming training sessions per week for 4 weeks and same aerobic dominant work. Normal dry-land training was substituted with an extra dryland strength training program for strength group (see below) and with an extra electrical stimulation training program for electrical stimulation group (see below). The control group did not perform any dry-land training but followed same swimming training program as ES and S groups. Therefore, training volume was exactly same for S and ES (Table 1). The swimmers were asked to follow ir usual eating habits and not to add any dietary supplements. They got instructions as to how to behave during entire study and were asked to keep a diary (to control hydration status, food intake, mood, and amount ES group C group Age (y) 21.1 (1.4) 19.5 (1.5) 24.2 (4.6) Height (m) 1.76 (0.08) 1.70 (0.11) 1.75 (0.07) Body mass (kg) 69.1 (6.5) 62.1 (9.8) 69.3 (7.4) Arm span (m) 1.78 (0.1) 1.74 (0.12) 1.75 (0.11) Swimming (hwk 21 ) Dry-land strength training (hwk 21 ) 0.75 Electrical stimulation (hwk 21 ) 0.75 Training volume (hwk 21 ) *S = strength group; ES = electrically stimulated; C = control. Results are expressed as mean (SDs). 498 of sleep). This study was carried out far from any important competition and not over a too long period to keep head coach seasonal training program and to assure compliance of swimmers and of ir trainer. Moreover, we also wanted to investigate wher effects of our intervention were maintained afterward. Therefore, we had to consider an 8-week period. Therefore, we decided to carry out this study in a nonspecific macrocycle at

3 beginning of second part of national French swimming season far from important competition and on a relatively short period to allow swimmers to work on or technical and physiological points in or part of cycle. In practice, training program took place in April for 4 weeks, during second macrocycle of season. Our athletes were preparing for ir national championships in June. During this macrocycle, re were no important competitions, and strength training intervention ended 6 weeks before championships. Description of Strength Training and Electrostimulation Sessions The strength training program for S concentrated on increasing muscular strength of upper limbs, principally latissimus dorsi. The strength training sessions were 15 minutes long and were preceded by a 10-minute warm-up using a skipping rope, which was considered as an all-around exercise. There were 3 strength training sessions per week, which were performed on Tuesday, Thursday, and Friday morning 1 hour before swimming training sessions. The program was same for each session. The program included 3 sets of 3 exercises with a 2 minutes of rest between each set. A maximum of 6 repetitions was performed for each exercise. The exercises included pull-ups and draws with pulleys. To perform exercises under most specific conditions, y were performed with hand in a pronation grip and distance between 2 hands set at shoulder width. The swimmers were instructed to pull on bar until it reached level of chest when performing pull-ups and first draw exercises. For second draw exercise, swimmers were instructed to pull on handle until it reached level of hips. For pull-ups, swimmers own body weight was used as resistance. For draw exercises, swimmers were seated on a bench with ir legs locked under a bar to avoid any movement of lower limbs. The intensity of training varied between 80 and 90% of maximal load for draw exercises in relation to 1RM tests performed before start of study. For each repetition, swimmers were instructed to perform concentric phase of movement as fast as possible, maintain a 3-second isometric contraction and return to initial position by slowing movement (eccentric phase). The exercise rate was approximately 1 movement every 6 seconds. The electrical stimulation training program for ES group was performed at same time as strength training program after same 10-minute warm-up using a skipping rope. The swimmers sat on a bench with ir arms along length of body and elbows in contact with trunk to avoid any movements. Electrostimulation was performed with a Compex Ò Mi Sport 500 (Compex Medical SA, Ecublens, Switzerland). Both latissimi dorsi muscles were simultaneously stimulated. The electrodes used for electrostimulation were 2 mm thick, self-adhesive, and of elastomer type. Double-pole stimulation was performed for each latissimus dorsi using 3 electrodes of different surface areas, 2 smalls electrodes of 22 cm 2 (4 cm cm), and 1 large electrode of 66 cm 2 (12 cm cm). The negative electrode ( large one) was placed 2 finger widths away from its proper anatomically described landmark. The positive electrodes were placed along spine at a slight angle. To stimulate paravertebral muscles, se electrodes were placed 2 finger widths from spinous processes. Pulse currents of 100 Hz in frequency lasting 300 microseconds were used. The contraction time was 5 seconds, and rest time was 15 seconds. The number of contractions per session was 45, for a total duration of 15 minutes. The intensity of muscle contraction was controlled by subjects and was designed to correspond on average to 80 90% of maximal voluntary contraction according to ir sensations. As previously mentioned, swimming training program was same for all 3 groups and consisted of 10 sessions of 2 hwk 21. These sessions comprised a combination of dominant aerobic work consisting of long sets at a moderate intensity, a short recovery period in front crawl, and technical work in medley. The training volume was 6, m per session. Swimming Velocity Swimming velocity was measured before training program (W0), after 4 weeks (W4) of training, and 4 weeks after end of training program (W8). These measurements were performed under same conditions to evaluate wher training effects lasted. The measurements were made on same day of week and at same time of day and were based on a 50-m front crawl, which was performed with a diving start following a starter s instructions, at maximal speed in a 25-m pool and about 15 minutes after a 2,000-m standardized warm-up. The swimmers performed ir 50 m one by one, from slowest to fastest as it happened in competition, at a rate of approximately one departure every seconds. Timing was manual. Technical Parameters All 50-m trials were video recorded with a digital cam Sony Ò minidv (HDR-HC7, Sony, Japan) at a frequency of 25 Hz. Stroke rate and length were measured with video analyzer software Pinnacle Ò (Studio Pinnacle System, Inc, Mountain View, CA, USA). A minimum of 3 complete stroke cycles was analyzed during each 25 m of 50 m over a distance of 10 m, which corresponded to field of camera. The mean of total number of cycles recorded during 50 m was used to calculate stroke rate and length. Swimming velocity was calculated based on stroke length and stroke time ratio. The camera was placed 12.5 m from edge of pool such that recording started 7.5 m after departure and lasted until 17.5 m of each 25 m to avoid any influence of starting and turning actions on swimming velocity. The camera was placed in a Plexiglas VOLUME 26 NUMBER 2 FEBRUARY

4 Strength vs. Electrostimulation in Swimming waterproof box (SPKHCC.CE8, Sony, Japan) at a depth of 0.15 m. A ruler designating every meter was placed in field of camera at a depth of 0.15 m in swimmers lane for calibration. Muscle Strength Measurements The extension peak torque (Newton meter) of dominant arm was measured with an isokinetic dynamometer (CON- TREX Ò, Medimex Factory, Tassin la Demi Lune, France) at W0, W4, and W8. Before measurements, a 10-minute standardized warm-up and familiarization period was performed with apparatus at several submaximal velocities in concentric conditions (60 and 180 s 21 ), isometric conditions, and eccentric conditions (260 s 21 ). These different angular velocities were selected because y are most representative for testing swimmers shoulder torque (35,36). The angular range movement in concentric and eccentric conditions was performed at an average of 160 from start to end of movement. In isometric condition, angle between arm and trunk was set at 90. The measurements took place at end of each week, 24 hours after last training session. The swimmers laid down and were strapped at shoulders and pelvis. The arm was maintained parallel to CON-TREX s arm lever. The spindle of motor was positioned in line with center of rotation of elbow joint. The subjects were asked to perform 5 maximal efforts. The best performance was retained. A 1-minute 30-second rest period separated each test. For isometric action, effort lasted 5 seconds, with a 2-minute rest period between repetitions. The angle between arm and trunk was set at 90. Intraclass correlations (Rs) of physical strength measurements were assessed using a coefficient of variation of 1.9% between 2 measurements. Body Weight No changes in body weight and body composition were observed during course of training. Statistical Analyses All results are presented with ir mean and SD. The sample size calculation regarding 50-m test was performed as previously described (18). Using a difference between groups of 1.0 seconds, an SD of 0.68 seconds, a beta value of 0.80, and an alpha value of 0.05, 6 volunteers per group should be a sufficient number of subjects to detect a significant difference at end of training program, should this difference exist. Analysis of variance (2-way repeated measurements (groups [S, ES, and C] 3 measures [W0, W4, and W8]) followed by Tukey-Kramer posthoc test were used to compare main characteristics, performance, muscular strength, stroke rate, and stroke length, of 3 groups before and after training. For whole group, correlation coefficients were calculated between performance and different parameters measured. The Statistica 5.0 (SAS Institute Inc, Cary, NC, USA) program was used. A p value of 0.05 was considered statistically significant. RESULTS TABLE 2. Swimming performance and arm peak torque characteristics of S, ES, and C groups of French national 1 swimmers before start of training.* S group ES group Before training (W0), re were no significant differences in swimming velocity, technical parameters, muscular strength, or morphological parameters between 3 groups (Table 2). C group 50-m Performance (s) (1.90) (1.27) (2.20) Stroke length (m) 2.12 (0.12) 2.05 (0.01) 2.08 (0.03) Stroke rate (cyclemin 21 ) 54.7 (4.1) 52.9 (3.5) 53.5 (2.4) Peak torque of arm extensors in 92.5 (33.5) 76.4 (1.7) 82.8 (28.2) isometric (Nm 21 ) Peak torque of arm extensors in 98.4 (27.4) 82.2 (9.6) 96.8 (28.6) concentric at 60s 21 (Nm 21 ) Peak torque of arm extensors in 88.3 (13.6) 84.7 (6.7) (29.5) concentric at 180s 21 (Nm 21 ) Peak torque of arm extensors in eccentric at 260s 21 (Nm 21 ) (34.4) 90.2 (1.2) (28.3) *S = strength group; ES = electrically stimulated; C = control. Results are expressed as mean (SDs). 500 Effects of Training on Swimming Velocity A significant improvement in mean swimming velocity for 50 m front crawl (p, 0.05) at end of training period (W4) was observed in S and ES groups when comparing results with those obtained from same groups at W0 ( and %, respectively) and C group at W0 and W4. At W8, delayed posttest demonstrated significant improvement (p, 0.05) was still exhibited by both S and ES groups. However, no significant differences were observed between S and ES groups at any time point. No changes in performance were observed in C group throughout study (Figure 1). Effects of Training on Stroke Length and Stroke Rate After 4 weeks of training, stroke length was significantly (p, 0.05) increased during 50minSbutnotinESandC. At W8, delayed posttest demonstrated that re was still significant improvement

5 Figure 1. Changes in performances over 50 m before training (W0), after 4 weeks of training (W4) and 4 weeks after end of training (W8) in French national 1 level swimmers. S = dry-land strength group; ES = electrical stimulation group; C = control groups. *Denotes significance (p, 0.05). Values are expressed as mean 6 SE. (p, 0.05) in stroke length in S but not in ES and C. There was no significant difference in technical parameter variation, expressed as a percentage of baseline, between 3 groups after 4-week training period (Table 3). Effects of Training on Muscle Strength After 4 weeks of training, peak torque was significantly (p, 0.05) increased in concentric condition at 60 s 21 in S and ES, but not in C. Peak torque was significantly (p, 0.05) increased in concentric condition at 180 s 21 in S and ES but not in C. Peak torque was significantly (p, 0.05) TABLE 3. Stroke rate and length in W0, W4, and W8 in S, ES, and C groups of French national 1 swimmers.* Stroke rate (cyclemin 21 ) increased in isometric and eccentric conditions at 260 s 21 in ES but not in S and C. At W8, delayed posttest demonstrated a significant (p, 0.05) increase in concentric condition at 60 s 21 in S and ES but not in C. Peak torque was significantly (p, 0.05) increased in concentric condition at 180 s 21 in S and ES but not in C. Peak torque was significantly (p, 0.05) increased in isometric and eccentric condition at 260 s 21 in ES but not in S and C. When compared with baseline, increases in peak torque after 4 weeks of training differed in concentric condition at 60 and 180 s 21 between S ( %; ) and C ( %; %; p, 0.05), respectively, and between ES ( %; ) and C ( %; %; p, 0.05), respectively, but not between S and ES. When compared with baseline, increases in peak torque after 4 weeks of training differed in isometric and eccentric conditions at 260 s 21 between S ( ; %) and ES ( ; %; p, 0.05), respectively, and between ES ( ; %) and C ( ; %; p, 0.05), respectively but not between S and C. Performance variation after 4 weeks of training, expressed as a percentage of baseline, correlated with stroke length (r = 0.96; p, 0.05) and peak torque in concentric condition at 60 and 180 s 21 Stroke length (m) W0 W4 W8 W0 W4 W8 S 54.7 (4.1) 55.9 (2.7) 55.1 (1.8) 2.05 (0.01) 2.11 (0.01) 2.11 (0.08) ES 52.9 (3.5) 54.3 (2.6) 53.9 (2.8) 2.12 (0.12) 2.17 (0.14) 2.17 (0.16) C 53.5 (2.4) 53.9 (2.3) 53.6 (2.1) 2.08 (0.03) 2.10 (0.02) 2.11 (0.05) *S = strength group; ES = electrically stimulated; C = control; W0 = week before training; W4 = after 4 weeks of training; W8 = 4 weeks after end of training. Data are expressed as mean (SD). Significant difference between W0 and W4, p, Significant difference between W0 and W8, p, (r = 0.97; 0.97; p, 0.05) variations in S, but this relationship did not occur in ES and C. In addition, peak torque variations in eccentric condition at 260 s 21 (r = 0.99; p, 0.05) occurred in ES but not in S and C. None of parameters correlated with performance in C after 4 weeks of training (Table 4). Gender Effect There was a similar gender percentage in each group, and re were no significant VOLUME 26 NUMBER 2 FEBRUARY

6 Strength vs. Electrostimulation in Swimming TABLE 4. Peak torque changes (%) in W4 and W8 in S, ES, and C groups of French national 1 swimmers.* differences in training effects between men and women among 3 groups. DISCUSSION Isometric Concentric Eccentric 0s 21 60s s 21 60s 21 W4 S 2.7 (2.1) 11.2 (13.6) 16.9 (11.7) 4.6 (6.1) ES 13.5 (10.9) 14.8 (7.2) 13.9 (5.6) 22.9 (6.6) C 1.6 (0.9) 2.2 (7.4) 1.2 (0.7) 6.1 (1.5) W8 S 2.6 (1.1) 10.1 (12.1) 16.8 (12.8) 4.2 (7.9) ES 12.9 (9.1) 14.2 (6.8) 13.1 (6.7) 20.8 (8.2) C 1.7 (1.2) 2.4 (5.1) 1.3 (1.4) 6.3 (1.6) *S = strength group; ES = electrically stimulated; C = control; W4 = after 4 weeks of training; W8 = 4 weeks after end of training. Data are expressed as mean (SD). Significantly different at p, The main findings of this study are as follows: (a) The 2 methods combining swimming and dry-land strength or swimming and ES were more efficient than swimming program alone in increasing sprint performance. Muscle strength increased in S and ES, whereas stroke length increased only in S. (b) No differences in performance were observed between S and ES. (c) The training effects were maintained 4 weeks after end of training program in S and ES groups. After 4 weeks of training, no significant differences were observed in performance gain over 50 m between S and ES (Figure 1). The 2% increase in 50-m performance in S group are comparable with 2.1% gain reported by Strass et al. (48) over 50 m after a 6-week dry-land (strength) training period and to 3.6% gain reported by Sharp et al. (43) over 25 yd after an 8-week dry-land (swim bench) training period. The 1.7% increase in 50-m performance in ES are close to 1.3 and 1.4% gain reported by Pichon et al. (40), who used similar methods of electrical stimulation, for 25-m arm stroking with a pull buoy and 50-m freestyle, respectively. The S and ES training methods were more efficient for increasing sprint performance than swimming program alone. These results are in agreement with those previously reported by Pichon et al. (40) and Tanaka et al. (50), indicating that combination of swimming and dry-land resistance training was more effective than swim training alone for improving swim performance. The ES group exhibited a stronger increase in eccentric condition than did S group (Table 4). In ES group, % increase in arm extensor isometric strength is close to 24.1% gain reported by Pichon et al. (40) after 3 weeks of training. This suggests that ES training may be a useful means for developing muscular strength and swimming velocity in swimmers sprints. These findings are consistent with previous reports confirming that brief periods of ES training have beneficial effects on muscle strength (2,27,32,41) and specific abilities of highly skilled athletes (28,31,40). It is generally accepted that neural adaptations predominate in short-term strength training (42) and ES training (7,14,32). For instance, Maffiuletti et al. (28) recently suggested that ES training would increase neural drive from supraspinal centers, resulting in a greater number of recruited motor units. Therefore, strength gains observed after present ES training during concentric test (60 and 180 s 21 ) but more likely during eccentric (260 s 21 ) maximal voluntary isokinetic contractions could be partly attributed to neural adaptations. Neverless, such a conclusion is only valid for a given angular velocity, and strength gains obtained for ES group would be primarily attributed to neural adaptations preferentially affecting fast-twitch fibers. Indeed, fast-twitch fibers have been suggested to be preferentially recruited during eccentric contractions ([16,39]; for a opposite view, see [47]) and increasingly recruited at high concentric velocities (9,13,17). Moreover, effectiveness of supplementing training with electrical stimulation is based on concept that fast-twitch fibers are activated first and to a greater extent than that predicted by Henneman s size principle (i.e., voluntary motor unit recruitment follows a progressive recruitment of small, typically slow, motor units followed in order of increasing size to larger, typically fast, motor units) (12,14,15,20,44). Therefore, ES can be used to activate fast motor units in addition to slow ones (i.e., disorderly recruitment), even at relatively low force levels (20,26). In conclusion, majority of evidence suggests that ES-induced motor unit recruitment is nonselective and that muscle fibers are recruited without obvious sequencing related to fiber types (20). This unique feature has important implications for use of ES in context of rehabilitation and sport training (26). Nardone et al. (38) indicated that fast-twitch fibers are preferentially recruited during submaximal exercise in eccentric condition. Moreover, Enoka (14) and Delitto et al. (11) stated that fast-twitch fibers could be preferentially

7 recruited during electrostimulation. This statement was confirmed by study performed by Andersen et al. (1) and Maffiuletti et al. (29), who observed a significant change in relative content of myosin heavy chain isoforms of 22% for type 2A muscle fibers after short-term electrostimulation resistance training. These data could explain results obtained in this study. The peak torque gains measured in concentric conditions for S and ES are in agreement with results of Häkkinen et al. (21). Indeed, y indicated that a training program in concentric and isometric conditions increases concentric and isometric strength, respectively, but not strength of or forms of contraction. In contrast, Häkkinen et al. (21) observed that an eccentric training program induced an increase in physical strength at each contraction regimen. Indeed, swimmers who performed strength training program (S) in our study were asked to generate maximal intensity during concentric phase of movement and to maintain ir effort during isometric and eccentric phases for 3 seconds, which was certainly too short to obtain significant adaptation to se forms of contraction. In addition, ability of electrical stimulation to enhance eccentric strength, as reported by several studies in or sports (20,27,32), by preferentially recruiting fast-twitch fibers could explain why ES peak torques increased for each condition. The stroke length increased only in S (Table 3). This result is in agreement with those values reported by Craig et al. (10), indicating that stroke length increases were probably related to ability to develop force necessary for overcoming resistance to forward movement. In this study, significant increase in physical strength in concentric conditions was correlated with performance increase in S. Thus, this strength gain may explain increase in stroke length. Moreover, peak torque gain was most important in concentric condition at 180 s 21, which is in agreement with our previous results (18) and underscores that, to be efficient, a 50-m swimmer has to generate maximal strength at high movement speed. In this study, improvement in 50-m performance in S was correlated with improvement in concentric strength at 60 and 180 s 21, and in ES, same gain was correlated with gain in eccentric strength at 260 s 21. These data are in agreement with those of previous studies (23,24,48) showing a strong relationship between power of upper limbs and sprint swimming performance. Moreover, relationship between gain in performance and gain in eccentric strength for ES confirms results of Pichon et al. (40). Indeed, y reported that variations in peak torque of arm extensors, measured in eccentric condition at 260 s 21, were related to variations in performance (r =0.77;p, 0.05) after a 3-week period of electrostimulation training. This relationship could be explained, on one hand, by ability of electrical stimulation to enhance eccentric strength (20,27,32), so importance of fast-twitch fibers in strength development during eccentric contractions. On or hand, sprint swimming performance requires both explosive force and fast-twitch fibers. The gain in performance for S was also correlated with increase in stroke length. Several studies have demonstrated importanceofstrokelengthinswimming(5,8).ina50-m sprint, stroke length is a key factor even if it is less than that observed for or swimming distances. As previously mentioned, 50-m swimmer has to generate maximal strength at a high stroke rate to be efficient (18), and physical strength will directly influence stroke length (10). All training effects were maintained 4 weeks after end of training program. These results confirm those reported by Häkkinen et al. (22), indicating that no change in strength and power occur during short periods of detraining. They also confirm results obtained by Gondin et al. (19) concerning preservation of maximal strength after 4 weeks of detraining subsequent to an 8-week electrical stimulation training program. Lemmer et al. (25) also reported that strength loss only occurred after 12 weeks of detraining subsequent to a 9-week strength training program. Both S and ES groups demonstrated significant improvements in performance as a result of ir specific training. Though improvement may stem from different signaling pathways, neir program provided such a different stimulus that outcome was different. However, a longer period of training might lead to more variations in improvement. Furr research is required to assess this question. PRACTICAL APPLICATIONS This study shows that methods combining swimming and dry-land S or swimming and ES were more efficient for increasing sprint performance in a 50-m front crawl than swimming program alone. No differences in performance gain were observed between dry-land strength training and electrical stimulation training methods. These training methods are both interesting and can be used during entire swimming season. However, electrical stimulation misses many of inherent characteristics of tissue stimulation related to connective tissue and its role and compression and strain rates. In addition, little work has been done concerning connective tissues, which play an important role in kinetic chain function. However, during periods of high training volume and competition (or when swimmer suffers from fatigue because of accumulation of high training load or in case of influenza, etc.), electrical stimulation may be a good replacement for dry-land strength training. The use of electrical stimulation instead of strength training during periods of high training volume and competitions may help to lower risk of injury because of an overload of movement on shoulder joint. Indeed, swimmer s shoulder is most exposed joint in water, and frequent injuries such as shoulder tendonitis are reported (34). More recently, Brushoj et al. (3) pointed out that shoulder pain is most common musculoskeletal complaint in competitive swimmers, principally consecutive to labral pathology and subacromial impingement. VOLUME 26 NUMBER 2 FEBRUARY

8 Strength vs. Electrostimulation in Swimming ACKNOWLEDGMENTS The authors wish to thank swimmers for ir voluntary participation and ir enthusiasm in training program. The authors warmly thank The club Stade Poitevin Natation and its board for ir very inspiring atmosphere, ir interest in applied research, and for making this study possible. REFERENCES 1. Andersen, JL and Aagaard, P. Myosin heavy chain IIX overshoot in human skeletal muscle. Muscle Nerve 23: , Brocherie, F, Babault, N, Cometti, G, Maffiuletti, N, and Chatard, JC. Electrostimulation training effects on physical performance of ice hockey players. Med Sci Sports Exerc 37: , Brushoj, C, Bak, K, Johannsen, HV, and Fauno, P. Swimmers painful shoulder arthroscopic findings and return rate to sports. Scand J Med Sci Sports 17: , Chatard, J and Mujika, I. Training load and performance in swimming. In: Biomechanics and Medicine in Swimming VIII. K.L. Keskinen, P.V. Komi, and A.P. Hollander, eds. Finland: University of Jyväskylä, pp Chatard, JC, Collomp, C, Maglischo, E, and Maglischo, C. Swimming skill and stroking characteristics of front crawl swimmers. Int J Sports Med 11: , Clarys, J. Hydrodynamics and electromyography: Ergonomic aspects in aquatics. Appl Ergon 16: 11 24, Colson, S, Martin, A, and Van Hoecke, J. Re-examination of training effects by electrostimulation in human elbow musculoskeletal system. Int J Sports Med 21: , Costill, D, Sharp, R, and Troup, J. Muscle strength: Contribution to sprint swimming. Sports Med 21: 29 34, Coyle, EF, Feiring, DC, Rotkis, TC, Cote, RW III, Roby, FB, Lee, W, and Wilmore, JH. Specificity of power improvements through slow and fast isokinetic training. J Appl Physiol 51: , Craig, AB Jr, Skehan, PL, Pawelczyk, JA, and Boomer, WL. Velocity, stroke rate, and distance per stroke during elite swimming competition. Med Sci Sports Exerc 17: , Delitto, A, Rose, SJ, McKowen, JM, Lehman, RC, Thomas, JA, and Shively, RA. Electrical stimulation versus voluntary exercise in strengning thigh musculature after anterior cruciate ligament surgery. Phys Ther 68: , Delitto, A and Snyder-Mackler, L. Two ories of muscle strength augmentation using percutaneous electrical stimulation. Phys Ther 70: , Dudley, GA and Harris, RT. Use of electrical stimulation in strength and power training. In: Strength and Power in Sport. P.V. Komi, ed. Boston, MA: Blackwell Scientific, pp Enoka, RM. Muscle strength and its development: New perspectives. Sports Med 6: , Feiereisen, P, Duchateau, J, and Hainaut, K. Motor unit recruitment order during voluntary and electrically induced contractions in tibialis anterior. Exp Brain Res 114: , Friden, J. Changes in human skeletal muscle induced by long-term eccentric exercise. Cell Tissue Res 236: , Froese, EA and Houston, ME. Torque-velocity characteristics and muscle fiber type in human vastus lateralis. J Appl Physiol 59: , Girold, S, Maurin, D, Dugue, B, Chatard, JC, and Millet, G. Effects of dry-land vs. resisted- and assisted-sprint exercises on swimming sprint performances. J Strength Cond Res 21: , Gondin, J, Guette, M, Ballay, Y, and Martin, A. Neural and muscular changes to detraining after electrostimulation training. Eur J Appl Physiol 97: , Gregory, CM and Bickel, CS. Recruitment patterns in human skeletal muscle during electrical stimulation. Phys Ther 85: , Häkkinen, K and Komi, PV. Effect of different combined concentric and eccentric muscle work regimens on maximal strength development. J Human Mov Stud 7: 33 34, Häkkinen, K, Pakarinen, A, Kyrolainen, H, Cheng, S, Kim, DH, and Komi, PV. Neuromuscular adaptations and serum hormones in females during prolonged power training. Int J Sports Med 11: 91 98, Hawley, JA and Williams, MM. Relationship between upper body anaerobic power and freestyle swimming performance. Int J Sports Med 12: 1 5, Hawley, JA, Williams, MM, Vickovic, MM, and Handcock, PJ. Muscle power predicts freestyle swimming performance. Br J Sports Med 26: , Lemmer, JT, Hurlbut, DE, Martel, GF, Tracy, BL, Ivey, FM, Metter, EJ, Fozard, JL, Fleg, JL, and Hurley, BF. Age and gender responses to strength training and detraining. Med Sci Sports Exerc 32: , Maffiuletti, NA. Physiological and methodological considerations for use of neuromuscular electrical stimulation. Eur J Appl Physiol 110: , Maffiuletti, NA, Cometti, G, Amiridis, IG, Martin, A, Pousson, M, and Chatard, JC. The effects of electromyostimulation training and basketball practice on muscle strength and jumping ability. Int J Sports Med 21: , Maffiuletti, NA, Dugnani, S, Folz, M, Di Pierno, E, and Mauro, F. Effect of combined electrostimulation and plyometric training on vertical jump height. Med Sci Sports Exerc 34: , Maffiuletti,NA,Zory,R,Miotti,D,Pellegrino,MA,Jubeau,M, and Bottinelli, R. Neuromuscular adaptations to electrostimulation resistance training. Am J Phys Med Rehabil 85: , Maglischo, EW. Swimming Fastest. Human Kinetics, Malatesta, D, Cattaneo, F, Dugnani, S, and Maffiuletti, NA. Effects of electromyostimulation training and volleyball practice on jumping ability. J Strength Cond Res 17: , Martin, L, Cometti, G, Pousson, M, and Morlon, B. Effect of electrical stimulation training on contractile characteristics of triceps surae muscle. Eur J Appl Physiol Occup Physiol 67: , McCafferty, WB and Horvath, SM. Specificity of exercise and specificity of training: A subcellular review. Res Q 48: , McMaster, WC. Swimming injuries. An overview. Sports Med 22: , McMaster, WC, Long, SC, and Caiozzo, VJ. Shoulder torque changes in swimming athlete. Am J Sports Med 20: , Miyashita, M and Kanehisa, H. Dynamic peak torque related to age, sex, and performance. Res Q 50: , Mujika, I, Chatard, JC, Busso, T, Geyssant, A, Barale, F, and Lacoste, L. Effects of training on performance in competitive swimming. Can J Appl Physiol 20: , Nardone, A, Romano, C, and Schieppati, M. Selective recruitment of high-threshold human motor units during voluntary isotonic lengning of active muscles. JPhysiol409: , Nardone, A and Schieppati, M. Shift of activity from slow to fast muscle during voluntary lengning contractions of triceps surae muscles in humans. J Physiol 395: ,

9 40. Pichon, F, Chatard, JC, Martin, A, and Cometti, G. Electrical stimulation and swimming performance. Med Sci Sports Exerc 27: , Romero, JA, Sanford, TL, Schroeder, RV, and Fahey, TD. The effects of electrical stimulation of normal quadriceps on strength and girth. Med Sci Sports Exerc 14: , Sale, DG. Neural adaptation to resistance training. Med Sci Sports Exerc 20: S135 S145, Sharp, RL, Troup, JP, and Costill, DL. Relationship between power and sprint freestyle swimming. Med Sci Sports Exerc 14: 53 56, Sinacore, DR, Delitto, A, King, DS, and Rose, SJ. Type II fiber activation with electrical stimulation: A preliminary report. Phys Ther 70: , Smith, DJ, Norris, SR, and Hogg, JM. Performance evaluation of swimmers: Scientific tools. Sports Med 32: , Stewart, AM and Hopkins, WG. Seasonal training and performance of competitive swimmers. J Sports Sci 18: , Stotz, PJ and Bawa, P. Motor unit recruitment during lengning contractions of human wrist flexors. Muscle Nerve 24: , Strass, D. Effects of maximal strength training on sprint performance of competitive swimmers. In: Swimming Science V, International Series on Sport Sciences (Vol. 18.). B. E. Ungerechts, K. Wilke, and K. Reischle, eds. IL: Human Kinetics Books, pp Swaine, IL. Arm and leg power output in swimmers during simulated swimming. Med Sci Sports Exerc 32: , Tanaka, H, Costill, DL, Thomas, R, Fink, WJ, and Widrick, JJ. Dryland resistance training for competitive swimming. Med Sci Sports Exerc 25: , Toussaint, HM and Hollander, AP. Energetics of competitive swimming. Implications for training programmes. Sports Med 18: , VOLUME 26 NUMBER 2 FEBRUARY