Isokinetic torque imbalances in the rotator cuff of the elite water polo player WILLIAM C.
McMASTER,* MD, SUSAN C. LONG, MA, AND VINCENT J. CAIOZZO, MS
From the Neuromuscular Research
Laboratory, Division of Orthopaedics, Department of Surgery, University of California, Irvine, California basis in the use of the shoulder adductor and internal rotation muscles. Several authors have evaluated rotator cuff muscle strength in baseball pitchers and have demonstrated a disparity in the balance between internal and external rotators. 1,Fowler (personal communication, July 1985) suggested the same for competitive swimmers. He contended that the external rotators have normal strength while the internal rotators are overdeveloped due to the repetitive mechanics of swimming. Such an imbalance could result in abnormal shoulder mechanics which might lead to injury, particularly in the lax shoulder, if it were to enhance joint subluxation during execution. Interestingly, the external rotator muscle groups are ignored in dry land training programs advocated for water polo players.&dquo; While there is no direct evidence that rotator cuff muscle strength imbalance leads to shoulder injury in water polo athletes, demonstrating the presence of the disparity would be the first step in such an investigation. It is the hypothesis of this study that rotator cuff muscle strength imbalances occur in the water polo player, and these deficits are speedspecific. The aim of this study is to test for the presence of these changes in the male water polo player and compare those findings to controls.
ABSTRACT The specific repetitive activity of water polo, like baseball pitching, emphasizes adduction and internal rotation. This study used the Cybex II to evaluate the isokinetic strength of the rotator cuff in elite water polo players and in a group of control subjects. The water polo players were significantly stronger than the controls. Of greater importance was the confirmation of imbalances in the rotator cuff force couples of adduction/abduction and external/internal rotation. These changes are similar to those reported for pitchers. The adductors in the water polo group had gained in relative strength resulting in an increase in the adduction/abduction ratio to about 2:1. The internal rotators had gained in relative strength resulting in a decrease in the external/internal ratio to about 0.6:1. For both force couples the differences are more apparent at a slow speed. Side-to-side differences were not significant.
Baseball pitching has been the cause celebre for the study of shoulder throwing mechanics and the resultant injuries from overuse.2~Whiting et al. 20 studied the throwing motion in water
MATERIALS AND METHODS
polo players using high-speed, two-plane, synchro-
nized photography and concluded that the motion is similar to baseball. Differences exist as a consequence of ball size and weight and the lack of a firm foundation resulting in decreased projectile release velocity and angular elbow acceleration compared to the baseball throw. The lack of a firm support base in water polo may result in the shift of more forces to the shoulder stabilizers, but there is no
The subjects were informed males who were healthy and free of shoulder complaints. Two groups were evaluated: a control group of college-aged noncompetitive males (N 10), and members of the U.S. Men’s National Water Polo Team in training for selection to the 1988 Olympics (N 15). Among the 15 water polo players tested, 8 were eventually selected as members of the 1988 U.S. Olympic Water Polo Team. Personal data collected for each subject consisted of age, height, weight, and dominant hand. Each subject was tested on the Cybex II Isokinetic Dy=
=
information available confirming this. Swimming and throwing have a common biomechanical * Address correspondence and reprnt requests to Wdliam C McMaster, MD, Orthopaedic Surgery, 1310 West Stewart Dnve, Suite 508, Orange, CA
namometer
(Cybex Co., Ronkonkoma, NY) for torque profollowing motions:
duction about the shoulder joint for the
92668. 72
73
internal rotation, external rotation, abduction, and adduction. The tests were performed on both right and left sides at speeds of 30 and 180 deg/sec. Prior to each test the Cybex II Isokinetic Dynamometer was calibrated to determine torque in foot pounds per millimeter of graph displacement. Produced torque values were selected using the angle-specific method. To identify the point of torque measurement, a magnetic switch was set to trigger and mark when the moment arm of the Cybex II Isokinetic Dynamometer passed the 45° position through each arc of motion. All torque values are consistently at 45° through each arc of motion being studied, hence angle-specific. To measure internal and external rotation, subjects were seated on the Cybex bench with the elbow resting on the input shaft of the Cybex. The humerus shaft was maintained collinear with the input shaft. The Cybex level arm was then adjusted to the length of the subject’s forearm with the elbow flexed to 90° and the shoulder positioned directly above the elbow. The arm starting position was perpendicular to the frontal plane of the body. The subject then rotated his arm internally for 90° in the horizontal plane. To measure external rotation, the arm movement began from a position parallel to the frontal plane of the body and rotated outwardly for 90° terminating in a position perpendicular to the frontal plane of the body. Measurements of abduction and adduction were made with the subject kneeling next to the Cybex with the input shaft aligned with the rotation center of the shoulder joint of the arm being measured. The lever arm was then adjusted to the length of the subject’s arm. Abduction was measured with the arm beginning in a position next to the body, then abducting 90°, and terminating perpendicular to the sagittal plane. Adduction was measured beginning with the arm in the 90° abducted position perpendicular to the sagittal plane with the test motion downward for 90°, terminating when the arm returned to the subject’s side. Gravity effects were calculated and corrections incorporated. The test sequence comprised an initial two trial attempts separated by a 10 second rest interval followed by three test efforts, also separated by a 10 second rest interval. The three test results of 45° angle-specific torque were averaged and standard deviations were calculated to the 95% limits. Torque values were calculated as well as torque ratios for the coupled motions of adduction/abduction and external/ internal rotation. Descriptive statistics were used to determine the mean and standard deviation for each of the vital and action data measured. A repeated measure analysis of variance was used to examine overall effect (BMDP2V, BMDP Statistic Software Inc., Los Angeles, CA). P < 0.05 was considered significant.
RESULTS The mean age, height, and weight for each group are given in Table 1. The angle-specific torque developed for each motion and speed are listed in Table 2. In controls the
TABLE 1 Mean physical characteristics of control and polo groups -
TABLE 2 Mean values (±SD) of angle-specific torque in the control and
polo
° P
McMASTER,* MD, SUSAN C. LONG, MA, AND VINCENT J. CAIOZZO, MS
From the Neuromuscular Research
Laboratory, Division of Orthopaedics, Department of Surgery, University of California, Irvine, California basis in the use of the shoulder adductor and internal rotation muscles. Several authors have evaluated rotator cuff muscle strength in baseball pitchers and have demonstrated a disparity in the balance between internal and external rotators. 1,Fowler (personal communication, July 1985) suggested the same for competitive swimmers. He contended that the external rotators have normal strength while the internal rotators are overdeveloped due to the repetitive mechanics of swimming. Such an imbalance could result in abnormal shoulder mechanics which might lead to injury, particularly in the lax shoulder, if it were to enhance joint subluxation during execution. Interestingly, the external rotator muscle groups are ignored in dry land training programs advocated for water polo players.&dquo; While there is no direct evidence that rotator cuff muscle strength imbalance leads to shoulder injury in water polo athletes, demonstrating the presence of the disparity would be the first step in such an investigation. It is the hypothesis of this study that rotator cuff muscle strength imbalances occur in the water polo player, and these deficits are speedspecific. The aim of this study is to test for the presence of these changes in the male water polo player and compare those findings to controls.
ABSTRACT The specific repetitive activity of water polo, like baseball pitching, emphasizes adduction and internal rotation. This study used the Cybex II to evaluate the isokinetic strength of the rotator cuff in elite water polo players and in a group of control subjects. The water polo players were significantly stronger than the controls. Of greater importance was the confirmation of imbalances in the rotator cuff force couples of adduction/abduction and external/internal rotation. These changes are similar to those reported for pitchers. The adductors in the water polo group had gained in relative strength resulting in an increase in the adduction/abduction ratio to about 2:1. The internal rotators had gained in relative strength resulting in a decrease in the external/internal ratio to about 0.6:1. For both force couples the differences are more apparent at a slow speed. Side-to-side differences were not significant.
Baseball pitching has been the cause celebre for the study of shoulder throwing mechanics and the resultant injuries from overuse.2~Whiting et al. 20 studied the throwing motion in water
MATERIALS AND METHODS
polo players using high-speed, two-plane, synchro-
nized photography and concluded that the motion is similar to baseball. Differences exist as a consequence of ball size and weight and the lack of a firm foundation resulting in decreased projectile release velocity and angular elbow acceleration compared to the baseball throw. The lack of a firm support base in water polo may result in the shift of more forces to the shoulder stabilizers, but there is no
The subjects were informed males who were healthy and free of shoulder complaints. Two groups were evaluated: a control group of college-aged noncompetitive males (N 10), and members of the U.S. Men’s National Water Polo Team in training for selection to the 1988 Olympics (N 15). Among the 15 water polo players tested, 8 were eventually selected as members of the 1988 U.S. Olympic Water Polo Team. Personal data collected for each subject consisted of age, height, weight, and dominant hand. Each subject was tested on the Cybex II Isokinetic Dy=
=
information available confirming this. Swimming and throwing have a common biomechanical * Address correspondence and reprnt requests to Wdliam C McMaster, MD, Orthopaedic Surgery, 1310 West Stewart Dnve, Suite 508, Orange, CA
namometer
(Cybex Co., Ronkonkoma, NY) for torque profollowing motions:
duction about the shoulder joint for the
92668. 72
73
internal rotation, external rotation, abduction, and adduction. The tests were performed on both right and left sides at speeds of 30 and 180 deg/sec. Prior to each test the Cybex II Isokinetic Dynamometer was calibrated to determine torque in foot pounds per millimeter of graph displacement. Produced torque values were selected using the angle-specific method. To identify the point of torque measurement, a magnetic switch was set to trigger and mark when the moment arm of the Cybex II Isokinetic Dynamometer passed the 45° position through each arc of motion. All torque values are consistently at 45° through each arc of motion being studied, hence angle-specific. To measure internal and external rotation, subjects were seated on the Cybex bench with the elbow resting on the input shaft of the Cybex. The humerus shaft was maintained collinear with the input shaft. The Cybex level arm was then adjusted to the length of the subject’s forearm with the elbow flexed to 90° and the shoulder positioned directly above the elbow. The arm starting position was perpendicular to the frontal plane of the body. The subject then rotated his arm internally for 90° in the horizontal plane. To measure external rotation, the arm movement began from a position parallel to the frontal plane of the body and rotated outwardly for 90° terminating in a position perpendicular to the frontal plane of the body. Measurements of abduction and adduction were made with the subject kneeling next to the Cybex with the input shaft aligned with the rotation center of the shoulder joint of the arm being measured. The lever arm was then adjusted to the length of the subject’s arm. Abduction was measured with the arm beginning in a position next to the body, then abducting 90°, and terminating perpendicular to the sagittal plane. Adduction was measured beginning with the arm in the 90° abducted position perpendicular to the sagittal plane with the test motion downward for 90°, terminating when the arm returned to the subject’s side. Gravity effects were calculated and corrections incorporated. The test sequence comprised an initial two trial attempts separated by a 10 second rest interval followed by three test efforts, also separated by a 10 second rest interval. The three test results of 45° angle-specific torque were averaged and standard deviations were calculated to the 95% limits. Torque values were calculated as well as torque ratios for the coupled motions of adduction/abduction and external/ internal rotation. Descriptive statistics were used to determine the mean and standard deviation for each of the vital and action data measured. A repeated measure analysis of variance was used to examine overall effect (BMDP2V, BMDP Statistic Software Inc., Los Angeles, CA). P < 0.05 was considered significant.
RESULTS The mean age, height, and weight for each group are given in Table 1. The angle-specific torque developed for each motion and speed are listed in Table 2. In controls the
TABLE 1 Mean physical characteristics of control and polo groups -
TABLE 2 Mean values (±SD) of angle-specific torque in the control and
polo
° P
