|Home | About | Journals | Submit | Contact Us | Français|
Ann M. Cools, PhD, PT, contributed to conception and design; analysis and interpretation of the data; and drafting and final approval of the article. Ellen Geerooms, PT, and Dorien F. M. Van de Berghe, PT, contributed to acquisition and analysis and interpretation of the data and critical revision and final approval of the article. Dirk C. Cambier, PhD, PT, and Erik E. Witvrouw, PhD, PT, contributed to conception and design and critical revision and final approval of the article.
Context: During gymnastic exercises, considerable force output is required in the shoulder girdle muscles. Isokinetic performance of the scapular muscles in young, elite gymnasts has not been examined.
Objective: To compare the isokinetic muscle performance of the scapular muscles between elite adolescent gymnasts and nonathletic adolescents to identify differences in strength, endurance, and muscle balance based on high-level sport participation.
Design: Single-session, repeated-measures design.
Setting: University human research laboratory.
Patients or Other Participants: Sixteen young, elite gymnasts and 26 age-matched nonathletic subjects participated in the study.
Intervention(s): Linear protraction-retraction movement in the scapular plane at 2 velocities (12.2 cm/s and 36.6 cm/s).
Main Outcome Measure(s): Isokinetic strength and endurance values, peak force/body mass, work/body mass, fatigue index (difference between the work performed in the first third and the last third of the test), and protraction to retraction strength ratios.
Results: Elite gymnasts demonstrated higher values for the protraction peak force/body mass than the control group demonstrated (P < .05), and they demonstrated higher protraction to retraction ratios on the nondominant side than on the dominant side (P < .05 at low velocity, P < .001 at high velocity). Work/body mass and fatigue index values were not statistically different between the groups. Side differences (P = .003) for retraction strength with lower protraction to retraction ratios (P < .001) were apparent in the gymnast group on the dominant side.
Conclusions: Scapular muscle performance in elite, young gymnasts is characterized by increased protraction strength and altered muscular balance around the scapula compared with nonathletic adolescents.
The scapula plays a vital role in upper extremity function. The quality of movement depends upon the interaction between scapular and glenohumeral kinematics, especially in overhead sports where the demands on the shoulder are extremely high.1 Recently, researchers1,2 have described the function of the scapulothoracic joint in overhead movements as an important link in the sport-specific kinetic chain in which the scapula functions. The kinetic chain principle describes how the human body can be considered as a series of interrelated links or segments. Movement of 1 segment affects segments proximally and distally.2 With respect to the kinetic chain, power, velocity, and accuracy of the throwing or smashing movement depend upon the quality of movement and stabilization of each link. For instance, during the throwing movement, the final outcome is the result of forces generated from the ground-reaction forces that come from the contralateral lower limb and are transferred in a diagonal pattern through the trunk into the throwing arm. Optimal positioning of every segment within the chain and functional muscle activation patterns from proximal to distal are the key premises in this transfer of energy.2
Gymnasts are a group of athletes who use their arms extensively during their sport activity.3,4 However, the starting and ending points of the kinetic chain in these athletes substantially differ from those of throwers. Hence, the relative role of the links in the chain, including muscle activation patterns, is likely different. Indeed, a unique aspect of gymnastics is the regular use of the upper extremities to support body mass.5 Contrary to overhead throwing athletes who use their arms in an open kinetic chain, gymnasts use their upper extremities very often in closed kinetic chain activities with the hand supported on a floor, balance beam, or pommel horse. The additional task of weight bearing requires supplementary strength of the arm muscles and stability of all contributing joints.3–5
Strong scapular muscles are among the major prerequisites for optimal stability and functional movement of the scapulothoracic joint.1,6,7 Several investigators have studied scapular muscle adaptations in healthy overhead athletes who are involved in overhead sports, such as baseball8 or volleyball,9,10 that involve the primary use of their dominant arm. However, few investigators have examined scapular muscle performance in overhead athletes with bilateral use of their arms during their sports. Muscle strength of the scapular muscles and scapular position have been investigated only in swimming athletes.11–13 Recently, Cools et al14 developed an isokinetic protocol for the measurement of scapulothoracic protraction and retraction muscle strength. Normal nonathletic subjects and overhead athletes have been evaluated with this procedure.9,14,15 Scapular muscle adaptations in young gymnasts, however, have not been examined. Therefore, the purpose of our study was to compare the isokinetic muscle performance of the scapular muscles between elite adolescent gymnasts and nonathletic adolescents to identify differences in strength, endurance, and muscle balance based on high-level sport participation.
Sixteen young, elite gymnasts (3 males, 13 females) were recruited to participate in our study. They had a mean age of 12.8 ± 1.8 years (range, 11–17 years), mean height of 146.7 ± 12.7 cm (range, 135–176 cm), and mean mass of 37.1 ± 1.2 kg (range, 26–64 kg). Subjects were all members of the Sports School at the time of the investigation and had a mean training intensity of 30.25 ± 1.2 h/wk (range = 28–34 h/wk). Seven athletes participated in sport acrobatics, and 9 athletes were artistic gymnasts. Subjects were excluded from the study if they were unable to attend the normal training regimen because of upper extremity pain or injury.
Twenty-six age-matched nonathletic adolescents (3 males, 23 females) served as a control group. This group had a mean age of 13.5 ± 1.6 years (range, 11–17), mean height of 161.6 ± 8.2 cm (range, 148–175 cm), and mean mass of 48.3 ± 8.8 kg (range, 32–60 kg). Exclusion criteria for the control group were upper limb or cervical spine injuries within the year before the study and participation in overhead sports for more than 2 h/wk.
All subjects completed questionnaires regarding their training and athletic activities and their injury history. All subjects and their legal guardians gave their written informed consent to participate in this study. The study was approved by the ethical committee of Ghent University.
We followed the testing procedures used in a previous study.9 All isokinetic tests were performed using an isokinetic dynamometer (Biodex System 3; Biodex Medical Systems, Inc, Shirley, NY). The testing session started with a warm-up procedure consisting of shoulder movements in all directions, push-up exercises against the wall, and stretching exercises for the rotator cuff and scapular muscles. The dominant side was tested first in both groups. We determined arm dominance by identifying the arm that the subject used to throw a ball.16
For the testing procedure, the closed chain attachment was fixed to the isokinetic dynamometer in a horizontal position. The handgrip was inserted into the attachment-receiving tube with the neutral handle facing up to keep the glenohumeral joint in a neutral rotational position. The chair was rotated to 15°, and the dynamometer was rotated to 45° in the opposite direction (Figure). Each subject was assessed in the seated position with the arm horizontal in the scapular plane (30° anterior to the frontal plane). We stabilized the trunk by placing a strap diagonally from the contralateral shoulder across the chest and securing it with a buckle. Each participant was subjected to 3 isokinetic tests: low velocity, high velocity, and endurance. The first test involved a linear speed of 12.2 cm/s (5 repetitions at an angular velocity of 60°/s); the second, 36.6 cm/s (5 repetitions at an angular velocity of 180°/s); and the third, 36.6 cm/s (40 repetitions at an angular velocity of 180°/s). The resting period between tests was 10 seconds. Experimental procedures were modified from a previously published protocol14 consisting of the first 2 tests that we used. In the protocol,14 the test-retest reproducibility of this procedure was found to be good to excellent for the peak force values (intraclass correlation coefficient = 0.88–0.96). In our investigation, we added a third test comprising 40 repetitions at a linear speed of 36.6 cm/s to evaluate endurance and fatigue variables. The number of repetitions and the selected velocity were based on research regarding isokinetic endurance and fatigue analysis.17–20
To assess range of motion, we asked the subjects to perform a maximal protraction and a maximal retraction movement. Gravity correction was not performed because the movement occurred in a horizontal plane. The test started in a maximal retracted position, and the subjects were instructed to perform maximal protraction and retraction movements over the total range of motion. They also were instructed to keep their elbows extended during the test. All movements were performed in the concentric-concentric mode, which means that protraction and retraction movements were performed with concentric muscle activity. Subjects performed 5 familiarization trials before data collection, and verbal encouragement was given during testing. Visual feedback from the computer screen was not allowed.
After data collection, we used Biodex software (Biodex Medical Systems) to determine peak force (N), work (J), and work fatigue (%). To select dependent variables for further analysis, we used independent t tests with the α set at .05 to analyze group differences with respect to anthropometric characteristics. Results showed no significant age differences between groups (t40 = 1.377, P = .176); however, significant height (t40 = 4.633, P < .001) and mass (t40 = 3.758, P < .001) differences existed between the gymnasts and the controls. Therefore, absolute strength data were normalized as a percentage of body mass. Peak force/body mass (N/kg × 100) and work/body mass (J/kg × 100) during protraction and retraction measured at 2 velocities (12.2 and 36.6 cm/s) were selected as dependent variables for statistical analysis. In addition, protraction to retraction ratios were calculated for both velocities based on the peak-force data. Finally, from the last test (40 repetitions), fatigue index was taken into account for further interpretation. This is a ratio of the difference between the work performed during the first third and the last third of the test bout. Positive values (%) represent a decline in work, and negative values mean that the work in the last third is increased compared with the first third.
Because all data were distributed normally with equal variances, we used parametric tests for statistical analysis. A general linear model 2-way analysis of variance (ANOVA) with repeated-measures design was used for statistical analysis in which the within-subjects factor was side (2 levels) and the between-subjects factor was group (2 levels). Interaction effects of group and side, as well as main group effects, were of interest. In the presence of an interaction effect, group differences and side differences were tested post hoc at each level of the interacting variable using a Bonferroni adjustment. In the absence of interactions, main effects of group were analyzed. The α was set at .05. For each of the multiple pairwise comparisons, the Bonferroni adjustment was used with the α set at .025. All statistical analyses were performed with SPSS (version 12.0; SPSS Inc, Chicago, IL). Power analysis of the strength values was calculated at more than 80%. Computations regarding effect size were based on the results from a previous study on healthy adult subjects.14
Table 1 summarizes the descriptive data and the results from the post hoc Bonferroni adjustments for peak force/body mass for both groups, both sides, both testing velocities, and both movement directions. The general linear model 2-way ANOVA with repeated-measures design revealed significant main group effects for protraction at low velocity (F1 = 10.469, P = .002) and at high velocity (F1 = 9.061, P = .005). For the retraction movement, no significant interaction or main effects were found at low velocity; however, results indicated a significant group × side interaction effect at high velocity (F1 = 4.19, P = .006). Results from the pairwise comparisons indicated that, for protraction, gymnasts were stronger than the controls on their dominant side at both velocities. For retraction, no significant group differences were found. Regarding side differences within each group, gymnasts demonstrated significantly more retraction strength on their dominant side than on their nondominant side (P = .003).
The descriptive data and the results from the post hoc Bonferroni adjustments for work/body mass for both groups, both sides, both testing velocities, and both movement directions are presented in Table 2. For the protraction movement, a significant group × side interaction effect at low velocity (F1 = 4.646, P = .037) was noted. However, the post hoc Bonferroni tests did not confirm any significant group or side difference. For protraction at high velocity and retraction at both velocities, the general linear model 2-way, repeated-measures ANOVA showed no significant interaction or main effects.
Results from the descriptive analysis and results from the post hoc Bonferroni adjustments for both groups, both sides, and both testing velocities are presented in Table 3. For the agonist to antagonist ratio, the general linear model 2-way, repeated-measures ANOVA revealed significant main group effects (F1 = 4.316, P = .044) at low velocity and a significant group × side interaction effect at high velocity (F1 = 11.628, P = .001). Results from the post hoc tests showed that the gymnasts had a lower protraction to retraction ratio on their dominant sides than on their nondominant sides at high velocity (P < .001) and a higher protraction to retraction ratio on their nondominant sides at both velocities than the control group had (P = .021 at low velocity, P < .001 at high velocity).
The results of the descriptive analysis for this variable for both groups and both sides are presented in Table 4. The general linear model 2-way, repeated-measures ANOVA showed no significant interaction or main effects for either movement direction. This means that, although fatigue index data were slightly lower in the gymnast group than in the control group (with the exception of retraction on the nondominant side), these results did not reach statistical significance.
Our purpose was to determine whether, based on the high-level sport participation of the gymnasts, strength and endurance differences existed between young, elite gymnasts and nonathletic adolescents. Our study provides important information related to adaptations of scapular muscle training that occur in adolescent gymnasts participating at an elite level. This information will assist athletic trainers and physical therapists from both an injury-prevention and a rehabilitation perspective.
Our results show that the gymnast group was significantly stronger into protraction on the nondominant side at low velocity and on both sides at high velocity than the controls were. This increased protraction strength may be the result of the specific characteristics of the sport, in which an athlete frequently has to push off from weight bearing. Powerful protraction is a necessary condition for this movement. Showing higher values for protraction peak force/body mass, our results demonstrate that this muscle was stronger in the gymnasts than in the nonathletic adolescents.
No differences related to retraction peak force/body mass were apparent between groups, but side differences were evident based on arm dominance. Specifically at 36.6 cm/s, the gymnasts were significantly stronger on their dominant sides than on their nondominant sides. In view of the bilateral use of the arms in gymnastics, we did not expect side differences based on arm dominance in the gymnasts. However, we should take into account that, despite the characteristic of bilateral arm use, gymnasts exhibit more unilateral arm use than, for instance, swimmers or rowers. In addition to bilateral exercises, numerous gymnastic movements are performed on 1 hand. Natural dominance probably determines the arm that the gymnast prefers to use. In addition, some categories of gymnastics, in particular acrobatic gymnastics, require considerable strength in the dominant arm for activities, such as elevating the partner. Seven of the 16 gymnasts practiced acrobatic gymnastics, possibly explaining side differences in this population.
Not all researchers agree that differences in muscle strength are based on arm dominance in overhead athletes. Cools et al9 found no significant side differences in isokinetic scapular muscle strength in a healthy population of volleyball players. In contrast, other investigators have demonstrated increased muscle strength of the scapular depressors8 and in the middle and lower trapezius21 on the dominant side in professional baseball players. However, in these studies, isometric muscle strength was evaluated with a hand-held dynamometer, so comparison with our results should be performed with caution.
Our results revealed no significant group or side differences for the variable work/body mass. In spite of strength differences, work performed during the repetition of maximal peak force, which was normalized to body mass, seemed to be equal for both groups and both sides. We have no data with which to compare our results because this variable is rarely examined in isokinetic testing. However, work in relation to body mass provides the researcher or clinician with relevant information about muscle performance. Possibly, a difference in shape of the force-length curve may offer an explanation for this finding. A curve with a short, steep slope to a high peak-force value may result in similar work values as a curve with a long, flat slope to a smaller peak-force value.22 However, we did not examine the shape of the force-length curves, so our explanation remains hypothetical.
Agonist to antagonist ratios provide important information about the relative strength of 1 muscle group to another and, thus, about muscle balance around a joint or segment. During arm motion, the scapula performs rotatory and translatory movements, but these movements do not occur independent of one another.1,6,23 The general scapular movement pattern consists of progressive external and upward rotation of the scapula and movement from an anteriorly to a posteriorly tilted position as the humeral elevation angle increases.23,24 Simultaneous actions of the serratus anterior and the trapezius muscle as a force couple perform and control these movements.1 In the scapulothoracic joint, the protraction to retraction ratio is approximately 1 in a nonathletic adult population, meaning the protractors and the retractors are equal in strength.14 Our gymnasts exhibited overall higher protraction to retraction ratios than the control group exhibited. These differences reached statistical significance on the nondominant side at both velocities. These higher values are the result of the higher protraction peak force/body mass in the gymnast group. Interestingly, in the control group, we also found that the agonist to antagonist ratios were slightly higher than the standard of 1,14 particularly at low velocity. Compared with adults, adolescents seem to have relatively stronger protractors than retractors. The question arises whether the higher values in the gymnast group should be interpreted as a result of sport-specific adaptation or as a reflection of adolescent muscle characteristics. Further investigation of age-specific characteristics of scapulothoracic muscle performance in athletic and nonathletic subjects is imperative to answer this question.
In addition to evaluating muscle strength, we investigated muscle endurance. The fatigue index was selected as the dependent variable measuring the degree of muscle fatigue after 40 repetitions, and it indicated the endurance capacity of the muscles. Researchers11,17,25 have found that scapulothoracic muscles are susceptible to muscle fatigue and that muscle fatigue alters the scapular kinematics. Authors12,13,24 have shown that altered scapular kinematics are correlated with shoulder impingement. Therefore, analysis of scapular muscle endurance is clinically relevant. No standardized protocol for evaluating fatigue is available. The number of repetitions in a fatigue test varies from 20 to 100 repetitions.17–20 We chose a test bout of 40 repetitions to fatigue the scapular muscles. Because 180°/s is a frequently used testing velocity for endurance measurement, we chose the corresponding linear speed of 36.6 cm/s. In our investigation, the work fatigue index values were all positive. This means that, in both groups on both sides, the endurance test resulted in a decrease in work between the first third and the last third of the 40 repetitions. However, no significant group or side differences became apparent. The gymnast group seemed to have no more or no less endurance than the control group. Nevertheless, with the exception of retraction on the nondominant side, all values for work fatigue were slightly smaller in the gymnast group, reflecting less fatigue and more endurance capacity. The lack of significance may have resulted from the large SDs of this variable, suggesting much intersubject variability in endurance capacity.
Some questions arise from our results. What is the clinical relevance of the scapular muscle performance results? Are differences in strength and agonist to antagonist ratios the outcome of sport-specific adaptations, increasing the risk for injury? If they are, should the clinician try to alter them, or are these adaptations necessary for optimal sports performance and normal in view of the age of the population? Investigators1,4,8,9,26 do not have a unanimous opinion about this dilemma. According to some authors,1,8,26 adaptations should be corrected as early as possible in the athlete's career to prevent overuse injuries. In that case, the young gymnasts should thoroughly train the retractors to restore the muscle balance around the scapula. Others27 believe that structural and functional adaptations are necessary to protect the musculoskeletal system against injury imposed by the high-velocity movements, such as throwing or weight bearing in gymnastics. In view of that hypothesis, correcting the adaptations may increase the risk of overuse injuries.
Because of the relevance of the kinetic chain principles during daily movements and particularly during complex sport movements, local adaptations at the scapulothoracic joint may influence the quality of movement in other links of the kinetic chain and the quality of the final outcome. Stronger protractors possibly increase the energy transfer from the trunk into the upper extremities or vice versa, and they improve athletic performance.1 However, imbalances in the scapular muscles, with relative decrease in muscle strength of the trapezius muscle in relation to protractor strength, may jeopardize not only scapular stability but also trunk stability.28 Movement patterns that do not sequentially activate all portions of the kinetic chain or that leave out a portion or link, such as trunk extensions or rotation, can lead to injury and nonoptimal performance.2 Prospective studies of larger groups of gymnasts are needed to identify the possible influence of adaptations in the scapular muscles on the quality of gymnastic performance and on the risk for overuse injury in other body parts.
The major limitation of our study can be found in the composition of both groups. Dividing the gymnasts into subgroups of sex, age, and gymnastic discipline was not possible because the number of participants was rather low. As a result of this limitation, extrapolation of our results into a general gymnastic population should be performed with caution. Investigating possible strength differences based on adolescence, specific biological development, and the specific demands of their discipline would be interesting. Moreover, we chose to investigate a control group matched by age because this grouping variable was objective and safe to standardize. However, we should take into account that the high level of sport participation probably resulted in large morphologic and physiologic differences between the groups. Normalizing our strength results to the subject's body mass possibly only partially addressed this limitation. Future investigators should emphasize the effort to categorize young gymnasts with respect to nonathletic individuals based on biologic age differences rather than on pure anthropometric data.
In summary, scapular muscle performance in young, elite gymnasts is characterized by increased protraction strength and by altered muscular balance around the scapula compared with scapular muscle performance in nonathletic adolescents. Further research is necessary to determine whether these adaptations increase injury risk or are necessary in view of their high-level of sport performance.