Electromyographic analysis during the golf swing MARILYN
From the
*
PINK,*† MS, PT,
Biomechanics
FRANK W.
JOBE,‡ MD,
Six male and seven female professional golfers volunteered for this study. The data on their EMG shoulder activity have been reported previously as a comparison between men and women.’ Since there were no significant differences between the data for the men and for the women, the data could be combined for a detailed analysis of shoulder function. All of the golfers had normal shoulders and were righthanded. The average age for the men was 35 years old (range, 30 to 42 years), and the average age for the women was 32 years old (range, 24 to 44 years). The muscles monitored were the infraspinatus, supraspinatus, subscapularis, latissimus dorsi, pectoralis major, and anterior, middle, and posterior deltoids on both the right and left shoulders. The EMG signal was recorded through indwelling electrodes, and the signal was integrated and then quantified. The procedure for this has been reported earlier.44 Two 16 mm high speed motion picture cameras operating at 450 frames per second were positioned for superior and anterior views to record the subject’s performance. Each subject had four trials with the driver club. Marks were electronically placed on the film and the EMG data to allow
popular recreational and competitive sport,
with the market, yet there is limited scientific information on the mechanics of the swing. A review of the literature reveals information on foot mechanics,’ grip sizes,’ and some preliminary work has been done with EMG and motion analysis of the shoulder muscles.’,’5 The previous work with EMG led to specific exercises for golfers. Those exercises have been used by the three professional tours with good success, and have also been available to the general public.’ The feedback on these exercises has encouraged us to expand our database, quantify the data more precisely, and analyze the swing in greater detail. This should lead to development of an exercise program for optimal performance as well as for prevention and rehabilitation. a
PERRY,* MD
MATERIALS AND METHODS
Golf is a popular sport throughout the world, yet there is little in the literature that discusses the mechanics of the swing. The purpose of this study is to analyze the EMG activity in eight shoulder muscles of both the right and left arms during the golf swing. The results reveal that the infraspinatus and supraspinatus act predominantly at the extremes of shoulder range of motion, the subscapularis and pectoralis major during acceleration, the latissimus dorsi during forward swing, and the anterior deltoid during forward swing and followthrough. The middle and posterior deltoids appear to be relatively noncontributory, without any specific timing patterns. This data is an expansion of an earlier pilot study and allows us to more accurately develop an exercise program for optimal performance as well as for prevention and rehabilitation.
Golf is
AND JACQUELIN
Laboratory, Centinela Hospital Medical Center and the &Dag er; Kerlan-Jobe Orthopaedic Clinic, Inglewood, California
ABSTRACT
many &dquo;how to&dquo; books
of the shoulder
on
synchronization. The motion
broken down into the following five illustrated in the previous report:4 from ball address to the end of the backswing; takeway: forward swing: from the end of the backswing until the is horizontal; acceleration: from horizontal club to ball contact; early follow-through: from ball contact to horizontal was
segments, which 1. 2.
club 3. 4.
were
club; 5. late follow-through: from horizontal club to the end of motion. Means and standard deviations were calculated for each muscle during each phase. An analysis of variance (withinsubjects design with repeated measures) was done for each muscle to determine whether there were any statistically
t Address correspondence and repnnt requests to Manlyn Pink, MS/PT, Director, Biomechanics Laboratory, Centinela Hospital Medical Center, 555 East Hardy Street, Inglewood, CA 90307 137
138
significant differences between phases of the golf swing in that muscle. When statistically significant differences were found at P 0.05, a Tukey’s multiple comparisons test was done to determine where those differences lay (again, P =
=
and 18% MMT). Early and late follow-through had 74% MMT and 39% MMT. The right anterior deltoid (Fig. 6) showed significantly more activity during forward swing than during takeaway
0.05). RESULTS The
right infraspinatus (Fig. 1) showed significantly more activity during takeaway [27% manual muscle strength test (MMT)]than during acceleration or late follow-through (7% MMT and 9% MMT). Forward swing and early follow-
through had 13% MMT and 12% MMT. In the left infraspinatus, there was significantly more activity during early follow-through (61% MMT) than during the first three phases (14% MMT, 16% MMT, and 27% MMT). Late follow-through had 40% MMT. Thus, the infraspinatus appears to be most active at the extremes of the shoulder’s range of motion during the golf swing (as the club is the
highest on each side). The supraspinatus revealed a similar general patterni.e., most EMG activity at the extremes of shoulder motion (Fig. 2). The right supraspinatus had significantly more activity during takeaway (25% MMT) than during early and late follow-through (7% MMT during each phase). Forward swing and acceleration had 14% MMT and 12% MMT. In the left supraspinatus, more activity, but not of statistical significance, was seen during early and late follow-through (28% MMT in each phase) than during takeaway, forward swing, and acceleration (21% MMT, 21% MMT, and 18% MMT). The right subscapularis (Fig. 3) had significantly more activity during acceleration than during takeaway (68% MMT and 16% MMT). Early and late follow-through had high levels of activity (64% MMT and 56% MMT), as did the forward swing (49% MMT). The left subscapularis had 33% MMT during takeaway, 29% MMT during forward swing, 41% NI1VTT during acceleration, 23% MMT during early follow-through, and 35% MMT during late followthrough, but there were no significant differences. The right latissimus dorsi (Fig. 4) had significantly more activity during forward swing and acceleration (50% MMT and 47% MMT) than during takeaway (9% MMT). Early follow-through and late follow-through exhibited 39% MMT and 28% MMT. In the left latissimus dorsi there was more activity, but not of a significant level, during forward swing (46% MMT) than during any of the other phases (takeaway, 17% MMT; acceleration, 31% MMT; early follow-through, 32% MMT; late follow-through, 18% MMT). The right pectoralis major (Fig. 5) had significantly more activity during forward swing, acceleration, and early followthrough (64% MMT, 93% MMT, and 74% MMT) than during takeaway (12% MMT). Acceleration had significantly more activity than did late follow-through (93% MMT and 37% MMT). In the left pectoralis major there was significantly more activity during acceleration (93% MMT) than during takeaway and forward swing (21 % MMT
Figure
1.
Infraspinatus muscle activity.
Figure 2. Supraspinatus
muscle
activity.
Figure 3. Subscapularis
muscle
activity.
139
26% MMT through the progressive phases. There were no significant differences in the middle deltoid for the right or left sides (Fig. 7). There were low levels of activity for all phases of the activity (2% to 8% MMT). The posterior deltoid showed low activity levels and no significant differences nor pattern of activity throughout all phases for both the right and left sides (5% to 24% MMT;
Fig. 8). DISCUSSION
study supports the earlier findings of Jobe et a1.33 regarding rotator cuff function during the golf swing. Golf is not a strenuous arm activity in that it does not require extremes of strength or range of motion. Thus, there is a relatively small contribution from the three deltoids bilaterally. However, the golf swing is a rapid movement and the rotator cuff muscles must fire in synchrony to provide a coordinated, harmonious movement to protect the glenohumeral complex. This is reflected by the firing patterns of This
Figure 4.
Latissimus dorsi muscle
activity.
the cuff muscles.
Figure 5.
Pectoralis
major muscle activity.
Figure 7. Middle deltoid muscle activity.
Figure 6. Anterior deltoid muscle activity. (21% MMT and 5% MMT). The later phases had 10% MMT, 11% MMT, and 8% MMT. There were no significant differences in the left anterior deltoid; the activity levels were 13% MMT, 9% MMT, 10% MMT, 21% MMT, and
Figure 8. Posterior deltoid muscle activity.
140
There do seem to be specific roles for the different muscles. The infraspinatus and supraspinatus act in concert, both of them providing most activity when the club and ipsilateral shoulder are at the greatest height. At this point, the arm is in maximum external rotation and slight abduction. The EMG of the infraspinatus, which is primarily an external rotator, has relatively high amplitude compared to that of the supraspinatus, which is primarily an abductor. Together these muscles act to approximate and stabilize the
glenohumeral joint. The subscapularis
is most active during acceleration, which differentiates it from the infraspinatus and supraspinatus. The subscapularis is primarily an internal rotator. In these right-handed golfers, the right arm begins to move into internal rotation during forward swing and continues with that movement throughout the rest of the swing. At acceleration, it is moving in that direction the most forcefully. The left arm is in greater internal rotation than the right during takeaway, and thus there is more activity in the left at this point. The internal rotation movement is less pronounced in the left than in the right through the rest of the swing, thus the relative EMG amplitude is less. It is of interest to note that all of the right-handed golfers seen by this office for shoulder problems have had left shoulder rotator cuff problems. Also, all of these golfers have a strong pull-through on the left. It is possible that as this group has attempted to equalize the roles played by the right and left arms, they may have overdone the role of the left. Thus, microtrauma and instability ensued. Add to that any predisposition for hyperelasticity, and subluxation could easily follow. The significance of this is that a right-handed golfer must strengthen the rotator cuff muscles on the left as a
preventive
measure.
The power in the shoulder for the drive is coming from the latissimus dorsi and pectoralis major. The latissimus dorsi contributes its power earlier, during forward swing and acceleration, and the pectoralis major plays the major role during acceleration and follow-through. The reason for the earlier activity from the latissimus dorsi lies in its principal role of internal rotation, which occurs in forward swing. The pectoralis major follows as it adducts the arm a little later, along with the internal rotation. The pectoralis major is more active than the latissimus dorsi, and more active than any of the other muscles tested, most likely because of its dual role of adduction along with internal rotation to provide the power for the swing. The low levels of activity and the noncontributory role of the three deltoids has been mentioned previously.’ The only additional information from this study concerns the anterior deltoid, which is the most active in the right arm during forward swing, and in the left arm during follow-through. These phases have corresponding and reciprocal shoulder
heights. Thus, the anterior deltoid appears to be assisting with flexing and lifting the arm. Clinically, we have experienced the importance of strengthening the anterior deltoid at the end of a rehabilitation program. A professional golfer with long-term shoulder instability who had undergone an anterior capsulolabral reconstruction was seen in our office. He subsequently underwent an intense program of rotator cuff strengthening. He began playing a little golf; however, his stroke was not satisfactory nor was his endurance. It was noted that his anterior deltoid looked flat, and a specific strengthening program was initiated. He now is back to playing a full schedule with no problems. Thus, it is recommended that the anterior deltoid be specifically strengthened as part of the end stage of a rehabilitation program. There were no patterns of import in the other two heads of the deltoid. In addition, no timing differences between the three heads were found. This study reinforces the concept of synchrony of movement of the muscles about the shoulder girdle. The specific timing of each muscle reinforces their significant roles; each muscle plays its part at a given moment in order to assure rhythmic movement. Each muscle must be optimally strengthened for that rhythm to be smooth. CONCLUSIONS - The infraspinatus and supraspinatus act in concert at the extremes of shoulder motion as external rotators, abductors, and stabilizers. - The subscapularis is most active at acceleration when the arm is internally rotating. The latissimus dorsi and pectoralis major are the powerdrive muscles of the shoulder, with the latissimus dorsi responding before the pectoralis major, and with the pectoralis major contributing the most activity of all muscles tested as it assists the rotation and forceful adduction of the
arm.
The anterior deltoid is most active as it lifts and flexes the arm; the other two heads of the deltoid are relatively noncontributory to the golf swing and do not show any -
specific timing patterns. REFERENCES 1 2
3 4 5
6.
Chao EY, Cooney WP, Cahalan TD, et al: Biomechanics of golf swing and a comparison of club handle design. Biomed Sci Instrum 23. 23-27, 1987 Jobe FW, Moynes DR: 30 Exercises for Better Golf. Inglewood, CA, Champion Press, 1986 Jobe FW, Moynes DR, Antonelli DJ Rotator cuff function during a golf : 388-392, 1986 swing Am J Sports Med 14 Jobe FW, Perry J, Pink M: Electromyographic shoulder activity in men and women professional golfers. Am J Sports Med 17 782-787, 1989 Moynes DR, Perry J, Antonelli DJ, et al Electromyography and motion analysis of the upper extremity in sports Phys Ther 66. 1905-1911, 1986 Williams KR, Cavanagh PR. The mechanics of foot action during the golf swing and implications for shoe design. Med Sci Sports Exerc 15. 247-
255,1983
From the
*
PINK,*† MS, PT,
Biomechanics
FRANK W.
JOBE,‡ MD,
Six male and seven female professional golfers volunteered for this study. The data on their EMG shoulder activity have been reported previously as a comparison between men and women.’ Since there were no significant differences between the data for the men and for the women, the data could be combined for a detailed analysis of shoulder function. All of the golfers had normal shoulders and were righthanded. The average age for the men was 35 years old (range, 30 to 42 years), and the average age for the women was 32 years old (range, 24 to 44 years). The muscles monitored were the infraspinatus, supraspinatus, subscapularis, latissimus dorsi, pectoralis major, and anterior, middle, and posterior deltoids on both the right and left shoulders. The EMG signal was recorded through indwelling electrodes, and the signal was integrated and then quantified. The procedure for this has been reported earlier.44 Two 16 mm high speed motion picture cameras operating at 450 frames per second were positioned for superior and anterior views to record the subject’s performance. Each subject had four trials with the driver club. Marks were electronically placed on the film and the EMG data to allow
popular recreational and competitive sport,
with the market, yet there is limited scientific information on the mechanics of the swing. A review of the literature reveals information on foot mechanics,’ grip sizes,’ and some preliminary work has been done with EMG and motion analysis of the shoulder muscles.’,’5 The previous work with EMG led to specific exercises for golfers. Those exercises have been used by the three professional tours with good success, and have also been available to the general public.’ The feedback on these exercises has encouraged us to expand our database, quantify the data more precisely, and analyze the swing in greater detail. This should lead to development of an exercise program for optimal performance as well as for prevention and rehabilitation. a
PERRY,* MD
MATERIALS AND METHODS
Golf is a popular sport throughout the world, yet there is little in the literature that discusses the mechanics of the swing. The purpose of this study is to analyze the EMG activity in eight shoulder muscles of both the right and left arms during the golf swing. The results reveal that the infraspinatus and supraspinatus act predominantly at the extremes of shoulder range of motion, the subscapularis and pectoralis major during acceleration, the latissimus dorsi during forward swing, and the anterior deltoid during forward swing and followthrough. The middle and posterior deltoids appear to be relatively noncontributory, without any specific timing patterns. This data is an expansion of an earlier pilot study and allows us to more accurately develop an exercise program for optimal performance as well as for prevention and rehabilitation.
Golf is
AND JACQUELIN
Laboratory, Centinela Hospital Medical Center and the &Dag er; Kerlan-Jobe Orthopaedic Clinic, Inglewood, California
ABSTRACT
many &dquo;how to&dquo; books
of the shoulder
on
synchronization. The motion
broken down into the following five illustrated in the previous report:4 from ball address to the end of the backswing; takeway: forward swing: from the end of the backswing until the is horizontal; acceleration: from horizontal club to ball contact; early follow-through: from ball contact to horizontal was
segments, which 1. 2.
club 3. 4.
were
club; 5. late follow-through: from horizontal club to the end of motion. Means and standard deviations were calculated for each muscle during each phase. An analysis of variance (withinsubjects design with repeated measures) was done for each muscle to determine whether there were any statistically
t Address correspondence and repnnt requests to Manlyn Pink, MS/PT, Director, Biomechanics Laboratory, Centinela Hospital Medical Center, 555 East Hardy Street, Inglewood, CA 90307 137
138
significant differences between phases of the golf swing in that muscle. When statistically significant differences were found at P 0.05, a Tukey’s multiple comparisons test was done to determine where those differences lay (again, P =
=
and 18% MMT). Early and late follow-through had 74% MMT and 39% MMT. The right anterior deltoid (Fig. 6) showed significantly more activity during forward swing than during takeaway
0.05). RESULTS The
right infraspinatus (Fig. 1) showed significantly more activity during takeaway [27% manual muscle strength test (MMT)]than during acceleration or late follow-through (7% MMT and 9% MMT). Forward swing and early follow-
through had 13% MMT and 12% MMT. In the left infraspinatus, there was significantly more activity during early follow-through (61% MMT) than during the first three phases (14% MMT, 16% MMT, and 27% MMT). Late follow-through had 40% MMT. Thus, the infraspinatus appears to be most active at the extremes of the shoulder’s range of motion during the golf swing (as the club is the
highest on each side). The supraspinatus revealed a similar general patterni.e., most EMG activity at the extremes of shoulder motion (Fig. 2). The right supraspinatus had significantly more activity during takeaway (25% MMT) than during early and late follow-through (7% MMT during each phase). Forward swing and acceleration had 14% MMT and 12% MMT. In the left supraspinatus, more activity, but not of statistical significance, was seen during early and late follow-through (28% MMT in each phase) than during takeaway, forward swing, and acceleration (21% MMT, 21% MMT, and 18% MMT). The right subscapularis (Fig. 3) had significantly more activity during acceleration than during takeaway (68% MMT and 16% MMT). Early and late follow-through had high levels of activity (64% MMT and 56% MMT), as did the forward swing (49% MMT). The left subscapularis had 33% MMT during takeaway, 29% MMT during forward swing, 41% NI1VTT during acceleration, 23% MMT during early follow-through, and 35% MMT during late followthrough, but there were no significant differences. The right latissimus dorsi (Fig. 4) had significantly more activity during forward swing and acceleration (50% MMT and 47% MMT) than during takeaway (9% MMT). Early follow-through and late follow-through exhibited 39% MMT and 28% MMT. In the left latissimus dorsi there was more activity, but not of a significant level, during forward swing (46% MMT) than during any of the other phases (takeaway, 17% MMT; acceleration, 31% MMT; early follow-through, 32% MMT; late follow-through, 18% MMT). The right pectoralis major (Fig. 5) had significantly more activity during forward swing, acceleration, and early followthrough (64% MMT, 93% MMT, and 74% MMT) than during takeaway (12% MMT). Acceleration had significantly more activity than did late follow-through (93% MMT and 37% MMT). In the left pectoralis major there was significantly more activity during acceleration (93% MMT) than during takeaway and forward swing (21 % MMT
Figure
1.
Infraspinatus muscle activity.
Figure 2. Supraspinatus
muscle
activity.
Figure 3. Subscapularis
muscle
activity.
139
26% MMT through the progressive phases. There were no significant differences in the middle deltoid for the right or left sides (Fig. 7). There were low levels of activity for all phases of the activity (2% to 8% MMT). The posterior deltoid showed low activity levels and no significant differences nor pattern of activity throughout all phases for both the right and left sides (5% to 24% MMT;
Fig. 8). DISCUSSION
study supports the earlier findings of Jobe et a1.33 regarding rotator cuff function during the golf swing. Golf is not a strenuous arm activity in that it does not require extremes of strength or range of motion. Thus, there is a relatively small contribution from the three deltoids bilaterally. However, the golf swing is a rapid movement and the rotator cuff muscles must fire in synchrony to provide a coordinated, harmonious movement to protect the glenohumeral complex. This is reflected by the firing patterns of This
Figure 4.
Latissimus dorsi muscle
activity.
the cuff muscles.
Figure 5.
Pectoralis
major muscle activity.
Figure 7. Middle deltoid muscle activity.
Figure 6. Anterior deltoid muscle activity. (21% MMT and 5% MMT). The later phases had 10% MMT, 11% MMT, and 8% MMT. There were no significant differences in the left anterior deltoid; the activity levels were 13% MMT, 9% MMT, 10% MMT, 21% MMT, and
Figure 8. Posterior deltoid muscle activity.
140
There do seem to be specific roles for the different muscles. The infraspinatus and supraspinatus act in concert, both of them providing most activity when the club and ipsilateral shoulder are at the greatest height. At this point, the arm is in maximum external rotation and slight abduction. The EMG of the infraspinatus, which is primarily an external rotator, has relatively high amplitude compared to that of the supraspinatus, which is primarily an abductor. Together these muscles act to approximate and stabilize the
glenohumeral joint. The subscapularis
is most active during acceleration, which differentiates it from the infraspinatus and supraspinatus. The subscapularis is primarily an internal rotator. In these right-handed golfers, the right arm begins to move into internal rotation during forward swing and continues with that movement throughout the rest of the swing. At acceleration, it is moving in that direction the most forcefully. The left arm is in greater internal rotation than the right during takeaway, and thus there is more activity in the left at this point. The internal rotation movement is less pronounced in the left than in the right through the rest of the swing, thus the relative EMG amplitude is less. It is of interest to note that all of the right-handed golfers seen by this office for shoulder problems have had left shoulder rotator cuff problems. Also, all of these golfers have a strong pull-through on the left. It is possible that as this group has attempted to equalize the roles played by the right and left arms, they may have overdone the role of the left. Thus, microtrauma and instability ensued. Add to that any predisposition for hyperelasticity, and subluxation could easily follow. The significance of this is that a right-handed golfer must strengthen the rotator cuff muscles on the left as a
preventive
measure.
The power in the shoulder for the drive is coming from the latissimus dorsi and pectoralis major. The latissimus dorsi contributes its power earlier, during forward swing and acceleration, and the pectoralis major plays the major role during acceleration and follow-through. The reason for the earlier activity from the latissimus dorsi lies in its principal role of internal rotation, which occurs in forward swing. The pectoralis major follows as it adducts the arm a little later, along with the internal rotation. The pectoralis major is more active than the latissimus dorsi, and more active than any of the other muscles tested, most likely because of its dual role of adduction along with internal rotation to provide the power for the swing. The low levels of activity and the noncontributory role of the three deltoids has been mentioned previously.’ The only additional information from this study concerns the anterior deltoid, which is the most active in the right arm during forward swing, and in the left arm during follow-through. These phases have corresponding and reciprocal shoulder
heights. Thus, the anterior deltoid appears to be assisting with flexing and lifting the arm. Clinically, we have experienced the importance of strengthening the anterior deltoid at the end of a rehabilitation program. A professional golfer with long-term shoulder instability who had undergone an anterior capsulolabral reconstruction was seen in our office. He subsequently underwent an intense program of rotator cuff strengthening. He began playing a little golf; however, his stroke was not satisfactory nor was his endurance. It was noted that his anterior deltoid looked flat, and a specific strengthening program was initiated. He now is back to playing a full schedule with no problems. Thus, it is recommended that the anterior deltoid be specifically strengthened as part of the end stage of a rehabilitation program. There were no patterns of import in the other two heads of the deltoid. In addition, no timing differences between the three heads were found. This study reinforces the concept of synchrony of movement of the muscles about the shoulder girdle. The specific timing of each muscle reinforces their significant roles; each muscle plays its part at a given moment in order to assure rhythmic movement. Each muscle must be optimally strengthened for that rhythm to be smooth. CONCLUSIONS - The infraspinatus and supraspinatus act in concert at the extremes of shoulder motion as external rotators, abductors, and stabilizers. - The subscapularis is most active at acceleration when the arm is internally rotating. The latissimus dorsi and pectoralis major are the powerdrive muscles of the shoulder, with the latissimus dorsi responding before the pectoralis major, and with the pectoralis major contributing the most activity of all muscles tested as it assists the rotation and forceful adduction of the
arm.
The anterior deltoid is most active as it lifts and flexes the arm; the other two heads of the deltoid are relatively noncontributory to the golf swing and do not show any -
specific timing patterns. REFERENCES 1 2
3 4 5
6.
Chao EY, Cooney WP, Cahalan TD, et al: Biomechanics of golf swing and a comparison of club handle design. Biomed Sci Instrum 23. 23-27, 1987 Jobe FW, Moynes DR: 30 Exercises for Better Golf. Inglewood, CA, Champion Press, 1986 Jobe FW, Moynes DR, Antonelli DJ Rotator cuff function during a golf : 388-392, 1986 swing Am J Sports Med 14 Jobe FW, Perry J, Pink M: Electromyographic shoulder activity in men and women professional golfers. Am J Sports Med 17 782-787, 1989 Moynes DR, Perry J, Antonelli DJ, et al Electromyography and motion analysis of the upper extremity in sports Phys Ther 66. 1905-1911, 1986 Williams KR, Cavanagh PR. The mechanics of foot action during the golf swing and implications for shoe design. Med Sci Sports Exerc 15. 247-
255,1983
