- Email: [email protected]
Shoulder strength and range of motion in elite female cricket fast bowlers with and without a history of shoulder pain
Journal of Science and Medicine in Sport (2008) 11, 575—580
ORIGINAL PAPER
Shoulder strength and range of motion in elite female cricket fast bowlers with and without a history of shoulder pain Max C. Stuelcken a,∗, Karen A. Ginn b, Peter J. Sinclair a a b
Discipline of Exercise and Sports Science, University of Sydney, NSW, Australia Discipline of Biomedical Science, University of Sydney, NSW, Australia
Received 18 January 2007 ; received in revised form 30 May 2007; accepted 7 June 2007 KEYWORDS Cricket; Shoulder; Pain; Range of motion; Muscle strength
Summary This study aimed to determine the prevalence of shoulder pain in female cricket fast bowlers and compare the shoulder rotation range of motion and strength of those bowlers with and without a history of shoulder pain. The active range of motion and isokinetic strength of the shoulder internal and external rotators was assessed in the bowling and non-bowling shoulders of 26 elite Australian female fast bowlers. Twelve bowlers had a history of shoulder pain. There were significant bilateral differences in external rotation range of motion for those bowlers without a history of shoulder pain (p < 0.05), and in internal rotation range of motion for both the total cohort and those bowlers with a history of shoulder pain (p < 0.05). There were no bilateral differences in average torques or average torque ratios, nor were there any differences in rotation range of motion, torques or torque ratios in the bowling shoulder between bowlers with and without a history of shoulder pain. There was an association between concentric internal rotation torque for the bowling shoulder and years of fast bowling (rs = 0.45). Given the relatively small number of elite female cricketers, future research in this field will need to recruit fast bowlers from other cricket playing nations to increase the power of studies and provide more confidence in the statistical outcomes. © 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
Introduction Chronic shoulder injuries were a major problem for the Australian women’s cricket squad on a recent overseas tours1 and anecdotal evidence based on clinical files indicates that female fast bowlers ∗
Corresponding author. E-mail address: [email protected] (M.C. Stuelcken).
sustain an unexpectedly high proportion of the shoulder injuries in elite female cricketers. Therefore, at the request of Cricket Australia, this study investigating shoulder pain in elite female fast bowlers was undertaken. A cricket fast bowler must perform two distinct unilateral overhead skills — bowling and throwing — and both produce high loads at the shoulder.2,3 The muscles of the rotator cuff have a crucial role
1440-2440/$ — see front matter © 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jsams.2007.06.007
576 in providing dynamic support at the shoulder joint, and an imbalance in the eccentric external rotation (ER) strength relative to the concentric internal rotation (IR) strength4 or a decrease in IR range of motion (ROM)5 in the dominant shoulder have been associated with shoulder pain in overhead athletes. Only two studies have investigated shoulder rotation strength or ROM in fast bowlers.6,7 In 30 male club level bowlers, Mabasa et al.6 assessed isokinetic strength with the humerus in 45◦ abduction in the scapular plane and reported concentric IR peak torque was significantly higher in the bowling shoulder compared to the non-bowling shoulder at test velocities of 60, 180 and 300◦ s−1 . As there was no bilateral difference in ER strength, these bowlers had a significantly lower concentric ER:IR peak torque ratio in their bowling shoulder. In 21 male provincial level fast bowlers, Aginsky et al.7 studied both isokinetic strength and ROM in the bowling shoulder with the humerus in 90◦ abduction and found that bowlers with a history of shoulder injury had significantly greater body massnormalised concentric IR peak torque at 180◦ s−1 than bowlers with no history of shoulder injury, but found no significant differences in passive IR or ER ROM between these bowlers. The authors suggested that the imbalance in normalised rotation torques between the internal and external rotators in bowlers with a history of shoulder injury indicated a predisposition to impingement syndrome. There is no published data on the ROM and strength of the shoulder rotator muscle groups in female fast bowlers. The aims of this study were to: (1) determine the prevalence of shoulder pain in current elite Australian female cricket fast bowlers; (2) describe the shoulder rotation ROM and strength profiles of these bowlers; (3) compare the shoulder rotation ROM and strength profiles of those bowlers with and without a history of shoulder pain; (4) investigate the association between the shoulder rotation ROM and strength profile for the bowling shoulder and bowling history in these bowlers. This information will contribute to the development of specific training and rehabilitation programmes for female cricketers.
Methods Participants Twenty-six elite female fast bowlers (mean ± S.D.: age 22.5 ± 4.5 years, height 170.6 ± 5.0 cm, mass 66.2 ± 7.5 kg) participated in the study and each
M.C. Stuelcken et al. provided written consent prior to commencement. Participants were assessed during the first month of the competitive season using a protocol that was approved by the Human Research Ethics Committee at the University of Sydney.
Questionnaire A self-administered questionnaire was developed for this study. Data included demographic information, cricket experience and history of shoulder pain that was attributable to or aggravated by bowling or throwing. A bowler was considered to have a history of pain in the bowling shoulder if selfreported pain had occurred (i) in the previous 12 months or (ii) at any stage in the bowler’s career AND could be verified by records kept by team medical support staff. Episodes of self-reported shoulder pain that did not meet these criteria were disregarded to minimise recall bias.8 If, at the commencement of the study, a bowler was experiencing shoulder pain while playing cricket its intensity was rated on a five point scale (very mild, mild, moderate, severe, very severe).
Range of motion Bilateral active shoulder rotation ROM was assessed with the subject positioned supine on a plinth with the upper limb to be tested in 90◦ shoulder abduction in the coronal plane, 90◦ elbow flexion, and the forearm in neutral pronation/supination. Two examiners were involved in all ROM measurements. One examiner monitored the movement and ensured the correct position was maintained while the other obtained the measurements. During IR, the end of range was determined as the point at which the posterolateral acromion began to rise off the plinth. In the measurement of IR ROM, techniques requiring visual inspection or manual stabilisation of the scapular have demonstrated equally acceptable reliability.9 During ER, it was ensured the scapula did not tilt inferiorly and retract as the end of range was approached and that the spine did not hyperextend. This method of measuring ER ROM has previously demonstrated good reliability.9 A long-armed goniometer (OEC Orthopaedic Ltd.) with a spirit level attached to the stationary vertical arm was used for all measurements. With the olecranon and ulnar styloid processes marked to assist consistent placement, the goniometer axis was positioned over the olecranon and the moving arm was aligned along the lateral aspect of the ulna and with the styloid process.
Shoulder pain in cricketers
Strength Isokinetic testing of the shoulder rotator muscle groups was performed on a Kin-Com dynamometer (Chattecx Corp., Chattanooga, TN, USA) using gravity correction. Each subject was seated with their trunk upright. The humerus was in 45◦ abduction and 30◦ horizontal flexion and aligned with the rotational axis of the dynamometer. The elbow was supported in 90◦ of flexion and the forearm was in neutral pronation/supination. Adjustable cuffs were used to secure both the wrist to the resistance pad and the proximal forearm to the dynamometer arm. Velcro straps were placed horizontally across the chest and pelvis to stabilise the trunk. This protocol has previously been shown to be reliable.10 A practice trial consisting of one set of five repetitions at 50% of maximum voluntary effort was performed before each test to familiarise the subject with the procedure. Each test consisted of one set of five continuous concentric—eccentric cycles at a velocity of 90◦ s−1 . The range through which testing was performed was individually determined from the results of the rotation ROM tests with 10◦ subtracted from each end of the subject’s rotation range for safety. There was a 2 min rest between IR and ER strength tests. Verbal encouragement was provided. Torque measurements within 10◦ of each end of the subject’s test range were removed from the analysis to allow time for the dynamometer arm to accelerate to the required velocity and to eliminate impact artifacts as it changed direction. The mean of the three repetitions with the highest average torques for each muscle group, contraction type and side were used for analysis.
Statistical analysis Due to both the presence of outliers in the data set, determined using the median absolute deviation statistic,11 and violations in the assumption of normality, tested using the Shapiro—Wilk statistic and Fisher’s measures of skewness and kurtosis, non-parametric statistical tests were undertaken using SPSS for Windows (version 11.0). Data were analysed for the total cohort and for two subgroups that consisted of those bowlers with (Pain subgroup) and without (No Pain subgroup) a history of shoulder pain. Bilateral differences were analysed using the Wilcoxon signed-rank tests and between-group differences for the bowling shoulder were analysed using the Mann—Whitney U test. To support these tests, a non-parametric estimator of the standardised-difference effect size (dbw ) was calculated using the bi-weight standard deviation
577 in the denominator.12 The associations between rotation ROM and strength in the bowling shoulder and bowling history (number of years as a fast bowler, whether they started fast bowling before puberty (age 13), and whether they bowled for Australian representative teams during the previous three seasons) were tested using Spearman’s rank order correlation and rank bi-serial correlation. For all statistical tests the significance level was set at p < 0.05.
Results Twelve (46%) female fast bowlers in this study had a history of shoulder pain and eight of these had experienced at least one episode of shoulder pain in the previous 12 months. At the commencement of the study, 5 of these 12 bowlers reported they were experiencing shoulder pain while playing cricket. For four of these five bowlers this was the latest episode in an ongoing problem of shoulder pain while for one bowler this was her first episode of shoulder pain. The intensity of this pain was rated as mild (2), moderate (2) and moderate-severe (1). There were no significant differences between the Pain and No Pain subgroups in terms of age, height, mass or bowling experience (Supplementary File). The active shoulder rotation ROM results are presented in Table 1. The ER ROM was significantly greater in the bowling shoulder than it was in the non-bowling shoulder for the No Pain subgroup (p = 0.044; dbw = 0.53). The IR ROM was significantly less in the bowling shoulder compared to the non-bowling shoulder for both the total cohort (p = 0.023; dbw = 0.57) and the Pain subgroup (p = 0.015; dbw = 1.40). There were no bilateral differences in total rotation ROM for the total cohort or either subgroup (p > 0.05). There were no differences between the Pain and No Pain subgroups for rotation ROM in the bowling shoulder, nor were there any associations between ROM in the bowling shoulder and bowling history (p > 0.05). The shoulder rotation torque and torque ratio results are presented in Table 2 and Supplementary File, respectively. There were no bilateral differences in torques or torque ratios for either the total cohort or the Pain and No Pain subgroups (p > 0.05). There were no differences between the Pain and No Pain subgroups for any of the torques or torque ratios in the bowling shoulder (p > 0.05). There was a significant association between concentric IR torque for the bowling shoulder and years of fast bowling (rs = 0.45, p = 0.020).
578
Table 1
Active shoulder rotation range of motion for elite female fast bowlers
Variable
Pain subgroupa (n = 12)
All bowlers (n = 26) Bowling shoulder
Internal rotation External rotation Total rotation
*
43.5 (7.3) (30.0—56.0) 86.0 (11.4) (60.0—103.0) 129.6 (11.7) (106.0—149.0)
Non-bowling shoulder 47.8 (10.4) (20.0—69.0) 82.7 (10.4) (64.0—102.0) 130.4 (11.3) (105.0—151.0)
Bowling shoulder *
42.8 (5.5) (35.0—50.0) 82.5 (13.0) (60.0—100.0) 125.3 (13.6) (106.0—147.0)
No Pain subgroupb (n = 14) Non-bowling shoulder
Bowling shoulder
Non-bowling shoulder
49.4 (5.3) (42.0—56.0) 82.0 (10.9) (64.0—97.0) 131.4 (9.3) (118.0—151.0)
44.1 (8.7) (30.0—56.0) 89.1 (9.3)* (67.0—103.0) 133.2 (8.7) (115.0—149.0)
46.4 (13.4) (20.0—69.0) 83.2 (10.4) (64.0—102.0) 129.6 (13.1) (105.0—146.0)
Values are mean (S.D.) and (range); range of motion is expressed in degrees. a Pain subgroup, Bowlers with a history of shoulder pain at the commencement of the study. b No Pain subgroup, Bowlers with no history of shoulder pain at the commencement of the study. * Significantly different to the non-bowling shoulder, p < 0.05.
Table 2
Mean shoulder rotation torques for elite female fast bowlers
Variable
No Pain subgroupb (n = 14)
Bowling shoulder
Non-bowling shoulder
Bowling shoulder
Non-bowling shoulder
Bowling shoulder
Non-bowling shoulder
23.4 26.5 13.5 17.3
25.2 29.4 14.5 18.1
22.6 23.6 12.8 15.0
23.6 28.3 14.7 17.4
24.2 29.0 14.0 19.2
26.6 30.3 14.4 18.7
(7.1) (13.5—40.7) (10.7) (11.0—59.9) (3.4) (7.4—20.3) (6.1) (7.7—31.2)
(8.8) (11.4—39.1) (11.9) (9.9—55.7) (3.1) (10.3—21.4) (7.1) (8.9—38.1)
(7.4) (8.3) (3.7) (5.6)
(13.5—34.6) (11.1—36.7) (7.4—20.3) (7.7—25.9)
Values are mean (S.D.) and (range); torque is expressed in N m. IR, internal rotation; ER, external rotation. a Pain subgroup, Bowlers with a history of shoulder pain at the commencement of the study. b No Pain subgroup, Bowlers with no history of shoulder pain at the commencement of the study.
(9.1) (9.3) (2.6) (5.9)
(13.9—39.1) (9.9—42.7) (10.6—18.8) (9.6—29.7)
(6.9) (15.3—40.7) (12.2) (11.0—59.9) (3.1) (9.1—19.3) (6.0) (7.9—31.2)
(8.7) (11.4—36.9) (14.0) (12.8—55.7) (3.6) (10.3—21.4) (8.2) (8.9—38.1)
M.C. Stuelcken et al.
Concentric IR Eccentric IR Concentric ER Eccentric ER
Pain subgroupa (n = 12)
All bowlers (n = 26)
Shoulder pain in cricketers
Discussion The retrospective survey found that 46% of our sample of female fast bowlers had a history of shoulder pain and this is comparable to the reported prevalence of shoulder pain in male cricket fast bowlers.7,13 It is also similar to the prevalence of shoulder pain reported in elite female badminton players (47%14 ), elite male volleyball players (46%15 ) and male professional tennis players (44%5 ) and indicates that female fast bowlers are just as likely to experience shoulder pain as other overhead athletes who are considered to be at high risk of shoulder injury. The chronic nature of many of the shoulder injuries in the Australian women’s squad has been identified as a major problem by medical support staff as they have had a detrimental affect on overall team performance, especially during a recent overseas tour.1 Although suitable for a descriptive study of injury prevalence, a retrospective design does have a number of limitations. Firstly, there is the potential for recall bias.8 This was minimised through the use of a strict definition of shoulder pain and by accessing medical records. The use of medical records, however, may also present problems because a member of the team’s support staff may not have been present at every training session or a bowler may have preferred to consult a private practitioner.8 Secondly, the study is a survey of ‘survivors’ whereby those bowlers who had suffered serious shoulder injuries may no longer be playing cricket at the elite level. Therefore, the current study may have under-reported the true extent of the problem of shoulder pain in female fast bowlers. A prospective longitudinal investigation would be required to clarify this issue. The No Pain subgroup had significantly more ER ROM in the bowling shoulder compared to the nonbowling shoulder. Increased ER ROM is probably attributable to a combination of bony16 and soft tissue17 adaptations. A greater ER range may facilitate optimal throwing mechanics and minimise the injury potential at the shoulder by allowing a larger rotation range over which force can be generated during the acceleration phase of the throw.18 The increased ER ROM in the bowling shoulder of the No Pain subgroup would support this hypothesis. Although the ER ROM was on average 6.6◦ greater in the bowling shoulder of the No Pain subgroup compared to the Pain subgroup, this difference was not statistically significant. A larger sample size would be needed to increase statistical power and determine if pain-free shoulder function is associated
579 with increased shoulder ER ROM in elite female fast bowlers. The IR ROM in the bowling shoulder of the Pain subgroup was similar to that in the bowling shoulder of the No Pain subgroup. Therefore, the significant bilateral difference in IR ROM in the Pain subgroup did not indicate a restriction in IR ROM in the bowling shoulder. Rather, it was due to relatively greater IR ROM in the non-bowling shoulder of the Pain subgroup compared to the No Pain subgroup. As such, the significant bilateral difference in IR ROM in the Pain subgroup is unlikely to be related to their history of shoulder pain. The female fast bowlers demonstrated no bilateral differences in shoulder torques or torque ratios. As no association has been found between the concentric strength of the shoulder internal rotators and ball velocity in fast bowling,19 an increased concentric IR strength may not be directly related to any performance benefit in this group of elite female fast bowlers. Although there were no statistically significant differences in torques between the Pain and No Pain subgroups, there was a tendency for the Pain subgroup to have less eccentric ER strength in the bowling shoulder (p = 0.072). The shoulder is particularly vulnerable during the deceleration phase of both bowling20 and throwing.3 If the eccentric strength of the external rotators is unable to effectively oppose the action of the internal rotators during this phase, translational movements of the humeral head may increase20 and lead to a greater potential for injury.4 Interestingly, the ‘dynamic control ratio’ (EccER:ConcIR) for both the Pain (0.70) and No Pain (0.83) subgroups was well below one, a level considered necessary for appropriate muscle balance to maintain adequate dynamic stability at the shoulder joint.4 Therefore, the results of this study may suggest a need for training and rehabilitation programmes to address this ratio in the bowling shoulder. There was an association between the concentric strength of the internal rotators in the bowling shoulder and years of fast bowling in this group of elite female fast bowlers. This association may indicate that the acceleration phase of bowling or throwing provides the stimulus for a small strength gain in the internal rotators over time. The major limitation of this study was the small sample size which reduced the statistical power to detect significant differences. The sample of 26 fast bowlers did, however, represent all but one of Australia’s elite female bowlers at the time of testing. Therefore, to address this issue, the sample size could only be increased by recruiting elite female fast bowlers from other cricket playing nations.
580
M.C. Stuelcken et al.
Practical implications • Elite Australian female cricket fast bowlers are just as likely to experience shoulder pain as other elite overhead athletes. • Female fast bowlers with no history of shoulder pain exhibited more external rotation range of motion in their bowling shoulders compared to their non-bowling shoulders. • Effective shoulder training and rehabilitation programmes are required to deal with the recurrent problem of shoulder pain in the Australian women’s cricket squad.
Appendix A Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jsams.2007.06.007.
References 1. Ross L. Physiotherapy WCA tour report world cup; 2000. 2. Ferdinands RED. Three-dimensional biomechanical analysis of fast bowling in cricket. Ph.D. thesis. University of Waikato; 2004. 3. Fleisig GS, Barrentine SW, Escamilla RF, et al. Biomechanics of overhand throwing with implications for injuries. Sports Med 1996;21:421—37. 4. Wang H-K, Cochrane T. Mobility impairment, muscle imbalance, muscle weakness, scapular asymmetry and shoulder injury in elite volleyball athletes. J Sports Med Phys Fitness 2001;41:403—10. 5. Vad VB, Gebeh A, Dines D, et al. Hip and shoulder internal rotation range of motion deficits in professional tennis players. J Sci Med Sport 2003;6:71—5.
6. Mabasa XM, Stewart AV, Fleishman C. Isokinetic strength of shoulder internal and external rotators in cricket bowlers. S Afr J Physiother 2002;58(2):35—8. 7. Aginsky KD, Lategan L, Stretch RA. Shoulder injuries in provincial male fast bowlers—–predisposing factors. S Afr J Sports Med 2004;16(1):25—8. 8. Gabbe BJ, Finch CF, Bennell KL, et al. How valid is a self reported 12 month sports injury history? Br J Sports Med 2003;37:545—7. 9. Ginn KA, Cohen ML. Conservative treatment for shoulder pain: prognostic indicators of outcome. Arch Phys Med Rehabil 2004;85:1231—5. 10. Mandalidis DG, Donne B, O’Regan M, et al. Reliability of isokinetic internal and external rotation in the scapular plane. Isokinet Exerc Sci 2001;9:65—72. 11. Wilcox RR, Keselman HJ. Modern robust data analysis methods: measures of central tendency. Psychol Methods 2003;8:254—74. 12. Grissom RJ, Kim JJ. Effect sizes for research. A broad practical approach. London: Lawrence Erlbaum Associates; 2005. 13. Orchard J, James T. Cricket Australia injury report; 2003. 14. Fahlström M, Yeap JS, Alfredson H, et al. Shoulder pain—–a common problem in world-class badminton players. Scand J Med Sci Sports 2006;16:168—73. 15. Wang H-K, Cochrane T. A descriptive epidemiological study of shoulder injury in top level English male volleyball players. Int J Sports Med 2001;22:159—63. 16. Osbahr DC, Cannon DL, Speer KP. Retroversion of the humerus in the throwing shoulder of college baseball pitchers. Am J Sports Med 2002;30:347—53. 17. Mihata T, Lee YS, McGarry MH, et al. Excessive humeral external rotation results in increased shoulder laxity. Am J Sports Med 2004;32:1278—85. 18. Wilk KE, Meister K, Fleisig G, et al. Biomechanics of the overhead throwing motion. Sports Med Arthrosc Rev 2000;8:124—34. 19. Loram LC, McKinon W, Wormgoor S, et al. Determinants of ball release speed in schoolboy fast-medium bowlers in cricket. J Sports Med Phys Fitness 2005;45: 483—90. 20. Myers P, O’Brien BP. Cricket: injuries, rehabilitation and training. Sports Med Arthrosc Rev 2001;9:124—36.
Available online at www.sciencedirect.com
ORIGINAL PAPER
Shoulder strength and range of motion in elite female cricket fast bowlers with and without a history of shoulder pain Max C. Stuelcken a,∗, Karen A. Ginn b, Peter J. Sinclair a a b
Discipline of Exercise and Sports Science, University of Sydney, NSW, Australia Discipline of Biomedical Science, University of Sydney, NSW, Australia
Received 18 January 2007 ; received in revised form 30 May 2007; accepted 7 June 2007 KEYWORDS Cricket; Shoulder; Pain; Range of motion; Muscle strength
Summary This study aimed to determine the prevalence of shoulder pain in female cricket fast bowlers and compare the shoulder rotation range of motion and strength of those bowlers with and without a history of shoulder pain. The active range of motion and isokinetic strength of the shoulder internal and external rotators was assessed in the bowling and non-bowling shoulders of 26 elite Australian female fast bowlers. Twelve bowlers had a history of shoulder pain. There were significant bilateral differences in external rotation range of motion for those bowlers without a history of shoulder pain (p < 0.05), and in internal rotation range of motion for both the total cohort and those bowlers with a history of shoulder pain (p < 0.05). There were no bilateral differences in average torques or average torque ratios, nor were there any differences in rotation range of motion, torques or torque ratios in the bowling shoulder between bowlers with and without a history of shoulder pain. There was an association between concentric internal rotation torque for the bowling shoulder and years of fast bowling (rs = 0.45). Given the relatively small number of elite female cricketers, future research in this field will need to recruit fast bowlers from other cricket playing nations to increase the power of studies and provide more confidence in the statistical outcomes. © 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
Introduction Chronic shoulder injuries were a major problem for the Australian women’s cricket squad on a recent overseas tours1 and anecdotal evidence based on clinical files indicates that female fast bowlers ∗
Corresponding author. E-mail address: [email protected] (M.C. Stuelcken).
sustain an unexpectedly high proportion of the shoulder injuries in elite female cricketers. Therefore, at the request of Cricket Australia, this study investigating shoulder pain in elite female fast bowlers was undertaken. A cricket fast bowler must perform two distinct unilateral overhead skills — bowling and throwing — and both produce high loads at the shoulder.2,3 The muscles of the rotator cuff have a crucial role
1440-2440/$ — see front matter © 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jsams.2007.06.007
576 in providing dynamic support at the shoulder joint, and an imbalance in the eccentric external rotation (ER) strength relative to the concentric internal rotation (IR) strength4 or a decrease in IR range of motion (ROM)5 in the dominant shoulder have been associated with shoulder pain in overhead athletes. Only two studies have investigated shoulder rotation strength or ROM in fast bowlers.6,7 In 30 male club level bowlers, Mabasa et al.6 assessed isokinetic strength with the humerus in 45◦ abduction in the scapular plane and reported concentric IR peak torque was significantly higher in the bowling shoulder compared to the non-bowling shoulder at test velocities of 60, 180 and 300◦ s−1 . As there was no bilateral difference in ER strength, these bowlers had a significantly lower concentric ER:IR peak torque ratio in their bowling shoulder. In 21 male provincial level fast bowlers, Aginsky et al.7 studied both isokinetic strength and ROM in the bowling shoulder with the humerus in 90◦ abduction and found that bowlers with a history of shoulder injury had significantly greater body massnormalised concentric IR peak torque at 180◦ s−1 than bowlers with no history of shoulder injury, but found no significant differences in passive IR or ER ROM between these bowlers. The authors suggested that the imbalance in normalised rotation torques between the internal and external rotators in bowlers with a history of shoulder injury indicated a predisposition to impingement syndrome. There is no published data on the ROM and strength of the shoulder rotator muscle groups in female fast bowlers. The aims of this study were to: (1) determine the prevalence of shoulder pain in current elite Australian female cricket fast bowlers; (2) describe the shoulder rotation ROM and strength profiles of these bowlers; (3) compare the shoulder rotation ROM and strength profiles of those bowlers with and without a history of shoulder pain; (4) investigate the association between the shoulder rotation ROM and strength profile for the bowling shoulder and bowling history in these bowlers. This information will contribute to the development of specific training and rehabilitation programmes for female cricketers.
Methods Participants Twenty-six elite female fast bowlers (mean ± S.D.: age 22.5 ± 4.5 years, height 170.6 ± 5.0 cm, mass 66.2 ± 7.5 kg) participated in the study and each
M.C. Stuelcken et al. provided written consent prior to commencement. Participants were assessed during the first month of the competitive season using a protocol that was approved by the Human Research Ethics Committee at the University of Sydney.
Questionnaire A self-administered questionnaire was developed for this study. Data included demographic information, cricket experience and history of shoulder pain that was attributable to or aggravated by bowling or throwing. A bowler was considered to have a history of pain in the bowling shoulder if selfreported pain had occurred (i) in the previous 12 months or (ii) at any stage in the bowler’s career AND could be verified by records kept by team medical support staff. Episodes of self-reported shoulder pain that did not meet these criteria were disregarded to minimise recall bias.8 If, at the commencement of the study, a bowler was experiencing shoulder pain while playing cricket its intensity was rated on a five point scale (very mild, mild, moderate, severe, very severe).
Range of motion Bilateral active shoulder rotation ROM was assessed with the subject positioned supine on a plinth with the upper limb to be tested in 90◦ shoulder abduction in the coronal plane, 90◦ elbow flexion, and the forearm in neutral pronation/supination. Two examiners were involved in all ROM measurements. One examiner monitored the movement and ensured the correct position was maintained while the other obtained the measurements. During IR, the end of range was determined as the point at which the posterolateral acromion began to rise off the plinth. In the measurement of IR ROM, techniques requiring visual inspection or manual stabilisation of the scapular have demonstrated equally acceptable reliability.9 During ER, it was ensured the scapula did not tilt inferiorly and retract as the end of range was approached and that the spine did not hyperextend. This method of measuring ER ROM has previously demonstrated good reliability.9 A long-armed goniometer (OEC Orthopaedic Ltd.) with a spirit level attached to the stationary vertical arm was used for all measurements. With the olecranon and ulnar styloid processes marked to assist consistent placement, the goniometer axis was positioned over the olecranon and the moving arm was aligned along the lateral aspect of the ulna and with the styloid process.
Shoulder pain in cricketers
Strength Isokinetic testing of the shoulder rotator muscle groups was performed on a Kin-Com dynamometer (Chattecx Corp., Chattanooga, TN, USA) using gravity correction. Each subject was seated with their trunk upright. The humerus was in 45◦ abduction and 30◦ horizontal flexion and aligned with the rotational axis of the dynamometer. The elbow was supported in 90◦ of flexion and the forearm was in neutral pronation/supination. Adjustable cuffs were used to secure both the wrist to the resistance pad and the proximal forearm to the dynamometer arm. Velcro straps were placed horizontally across the chest and pelvis to stabilise the trunk. This protocol has previously been shown to be reliable.10 A practice trial consisting of one set of five repetitions at 50% of maximum voluntary effort was performed before each test to familiarise the subject with the procedure. Each test consisted of one set of five continuous concentric—eccentric cycles at a velocity of 90◦ s−1 . The range through which testing was performed was individually determined from the results of the rotation ROM tests with 10◦ subtracted from each end of the subject’s rotation range for safety. There was a 2 min rest between IR and ER strength tests. Verbal encouragement was provided. Torque measurements within 10◦ of each end of the subject’s test range were removed from the analysis to allow time for the dynamometer arm to accelerate to the required velocity and to eliminate impact artifacts as it changed direction. The mean of the three repetitions with the highest average torques for each muscle group, contraction type and side were used for analysis.
Statistical analysis Due to both the presence of outliers in the data set, determined using the median absolute deviation statistic,11 and violations in the assumption of normality, tested using the Shapiro—Wilk statistic and Fisher’s measures of skewness and kurtosis, non-parametric statistical tests were undertaken using SPSS for Windows (version 11.0). Data were analysed for the total cohort and for two subgroups that consisted of those bowlers with (Pain subgroup) and without (No Pain subgroup) a history of shoulder pain. Bilateral differences were analysed using the Wilcoxon signed-rank tests and between-group differences for the bowling shoulder were analysed using the Mann—Whitney U test. To support these tests, a non-parametric estimator of the standardised-difference effect size (dbw ) was calculated using the bi-weight standard deviation
577 in the denominator.12 The associations between rotation ROM and strength in the bowling shoulder and bowling history (number of years as a fast bowler, whether they started fast bowling before puberty (age 13), and whether they bowled for Australian representative teams during the previous three seasons) were tested using Spearman’s rank order correlation and rank bi-serial correlation. For all statistical tests the significance level was set at p < 0.05.
Results Twelve (46%) female fast bowlers in this study had a history of shoulder pain and eight of these had experienced at least one episode of shoulder pain in the previous 12 months. At the commencement of the study, 5 of these 12 bowlers reported they were experiencing shoulder pain while playing cricket. For four of these five bowlers this was the latest episode in an ongoing problem of shoulder pain while for one bowler this was her first episode of shoulder pain. The intensity of this pain was rated as mild (2), moderate (2) and moderate-severe (1). There were no significant differences between the Pain and No Pain subgroups in terms of age, height, mass or bowling experience (Supplementary File). The active shoulder rotation ROM results are presented in Table 1. The ER ROM was significantly greater in the bowling shoulder than it was in the non-bowling shoulder for the No Pain subgroup (p = 0.044; dbw = 0.53). The IR ROM was significantly less in the bowling shoulder compared to the non-bowling shoulder for both the total cohort (p = 0.023; dbw = 0.57) and the Pain subgroup (p = 0.015; dbw = 1.40). There were no bilateral differences in total rotation ROM for the total cohort or either subgroup (p > 0.05). There were no differences between the Pain and No Pain subgroups for rotation ROM in the bowling shoulder, nor were there any associations between ROM in the bowling shoulder and bowling history (p > 0.05). The shoulder rotation torque and torque ratio results are presented in Table 2 and Supplementary File, respectively. There were no bilateral differences in torques or torque ratios for either the total cohort or the Pain and No Pain subgroups (p > 0.05). There were no differences between the Pain and No Pain subgroups for any of the torques or torque ratios in the bowling shoulder (p > 0.05). There was a significant association between concentric IR torque for the bowling shoulder and years of fast bowling (rs = 0.45, p = 0.020).
578
Table 1
Active shoulder rotation range of motion for elite female fast bowlers
Variable
Pain subgroupa (n = 12)
All bowlers (n = 26) Bowling shoulder
Internal rotation External rotation Total rotation
*
43.5 (7.3) (30.0—56.0) 86.0 (11.4) (60.0—103.0) 129.6 (11.7) (106.0—149.0)
Non-bowling shoulder 47.8 (10.4) (20.0—69.0) 82.7 (10.4) (64.0—102.0) 130.4 (11.3) (105.0—151.0)
Bowling shoulder *
42.8 (5.5) (35.0—50.0) 82.5 (13.0) (60.0—100.0) 125.3 (13.6) (106.0—147.0)
No Pain subgroupb (n = 14) Non-bowling shoulder
Bowling shoulder
Non-bowling shoulder
49.4 (5.3) (42.0—56.0) 82.0 (10.9) (64.0—97.0) 131.4 (9.3) (118.0—151.0)
44.1 (8.7) (30.0—56.0) 89.1 (9.3)* (67.0—103.0) 133.2 (8.7) (115.0—149.0)
46.4 (13.4) (20.0—69.0) 83.2 (10.4) (64.0—102.0) 129.6 (13.1) (105.0—146.0)
Values are mean (S.D.) and (range); range of motion is expressed in degrees. a Pain subgroup, Bowlers with a history of shoulder pain at the commencement of the study. b No Pain subgroup, Bowlers with no history of shoulder pain at the commencement of the study. * Significantly different to the non-bowling shoulder, p < 0.05.
Table 2
Mean shoulder rotation torques for elite female fast bowlers
Variable
No Pain subgroupb (n = 14)
Bowling shoulder
Non-bowling shoulder
Bowling shoulder
Non-bowling shoulder
Bowling shoulder
Non-bowling shoulder
23.4 26.5 13.5 17.3
25.2 29.4 14.5 18.1
22.6 23.6 12.8 15.0
23.6 28.3 14.7 17.4
24.2 29.0 14.0 19.2
26.6 30.3 14.4 18.7
(7.1) (13.5—40.7) (10.7) (11.0—59.9) (3.4) (7.4—20.3) (6.1) (7.7—31.2)
(8.8) (11.4—39.1) (11.9) (9.9—55.7) (3.1) (10.3—21.4) (7.1) (8.9—38.1)
(7.4) (8.3) (3.7) (5.6)
(13.5—34.6) (11.1—36.7) (7.4—20.3) (7.7—25.9)
Values are mean (S.D.) and (range); torque is expressed in N m. IR, internal rotation; ER, external rotation. a Pain subgroup, Bowlers with a history of shoulder pain at the commencement of the study. b No Pain subgroup, Bowlers with no history of shoulder pain at the commencement of the study.
(9.1) (9.3) (2.6) (5.9)
(13.9—39.1) (9.9—42.7) (10.6—18.8) (9.6—29.7)
(6.9) (15.3—40.7) (12.2) (11.0—59.9) (3.1) (9.1—19.3) (6.0) (7.9—31.2)
(8.7) (11.4—36.9) (14.0) (12.8—55.7) (3.6) (10.3—21.4) (8.2) (8.9—38.1)
M.C. Stuelcken et al.
Concentric IR Eccentric IR Concentric ER Eccentric ER
Pain subgroupa (n = 12)
All bowlers (n = 26)
Shoulder pain in cricketers
Discussion The retrospective survey found that 46% of our sample of female fast bowlers had a history of shoulder pain and this is comparable to the reported prevalence of shoulder pain in male cricket fast bowlers.7,13 It is also similar to the prevalence of shoulder pain reported in elite female badminton players (47%14 ), elite male volleyball players (46%15 ) and male professional tennis players (44%5 ) and indicates that female fast bowlers are just as likely to experience shoulder pain as other overhead athletes who are considered to be at high risk of shoulder injury. The chronic nature of many of the shoulder injuries in the Australian women’s squad has been identified as a major problem by medical support staff as they have had a detrimental affect on overall team performance, especially during a recent overseas tour.1 Although suitable for a descriptive study of injury prevalence, a retrospective design does have a number of limitations. Firstly, there is the potential for recall bias.8 This was minimised through the use of a strict definition of shoulder pain and by accessing medical records. The use of medical records, however, may also present problems because a member of the team’s support staff may not have been present at every training session or a bowler may have preferred to consult a private practitioner.8 Secondly, the study is a survey of ‘survivors’ whereby those bowlers who had suffered serious shoulder injuries may no longer be playing cricket at the elite level. Therefore, the current study may have under-reported the true extent of the problem of shoulder pain in female fast bowlers. A prospective longitudinal investigation would be required to clarify this issue. The No Pain subgroup had significantly more ER ROM in the bowling shoulder compared to the nonbowling shoulder. Increased ER ROM is probably attributable to a combination of bony16 and soft tissue17 adaptations. A greater ER range may facilitate optimal throwing mechanics and minimise the injury potential at the shoulder by allowing a larger rotation range over which force can be generated during the acceleration phase of the throw.18 The increased ER ROM in the bowling shoulder of the No Pain subgroup would support this hypothesis. Although the ER ROM was on average 6.6◦ greater in the bowling shoulder of the No Pain subgroup compared to the Pain subgroup, this difference was not statistically significant. A larger sample size would be needed to increase statistical power and determine if pain-free shoulder function is associated
579 with increased shoulder ER ROM in elite female fast bowlers. The IR ROM in the bowling shoulder of the Pain subgroup was similar to that in the bowling shoulder of the No Pain subgroup. Therefore, the significant bilateral difference in IR ROM in the Pain subgroup did not indicate a restriction in IR ROM in the bowling shoulder. Rather, it was due to relatively greater IR ROM in the non-bowling shoulder of the Pain subgroup compared to the No Pain subgroup. As such, the significant bilateral difference in IR ROM in the Pain subgroup is unlikely to be related to their history of shoulder pain. The female fast bowlers demonstrated no bilateral differences in shoulder torques or torque ratios. As no association has been found between the concentric strength of the shoulder internal rotators and ball velocity in fast bowling,19 an increased concentric IR strength may not be directly related to any performance benefit in this group of elite female fast bowlers. Although there were no statistically significant differences in torques between the Pain and No Pain subgroups, there was a tendency for the Pain subgroup to have less eccentric ER strength in the bowling shoulder (p = 0.072). The shoulder is particularly vulnerable during the deceleration phase of both bowling20 and throwing.3 If the eccentric strength of the external rotators is unable to effectively oppose the action of the internal rotators during this phase, translational movements of the humeral head may increase20 and lead to a greater potential for injury.4 Interestingly, the ‘dynamic control ratio’ (EccER:ConcIR) for both the Pain (0.70) and No Pain (0.83) subgroups was well below one, a level considered necessary for appropriate muscle balance to maintain adequate dynamic stability at the shoulder joint.4 Therefore, the results of this study may suggest a need for training and rehabilitation programmes to address this ratio in the bowling shoulder. There was an association between the concentric strength of the internal rotators in the bowling shoulder and years of fast bowling in this group of elite female fast bowlers. This association may indicate that the acceleration phase of bowling or throwing provides the stimulus for a small strength gain in the internal rotators over time. The major limitation of this study was the small sample size which reduced the statistical power to detect significant differences. The sample of 26 fast bowlers did, however, represent all but one of Australia’s elite female bowlers at the time of testing. Therefore, to address this issue, the sample size could only be increased by recruiting elite female fast bowlers from other cricket playing nations.
580
M.C. Stuelcken et al.
Practical implications • Elite Australian female cricket fast bowlers are just as likely to experience shoulder pain as other elite overhead athletes. • Female fast bowlers with no history of shoulder pain exhibited more external rotation range of motion in their bowling shoulders compared to their non-bowling shoulders. • Effective shoulder training and rehabilitation programmes are required to deal with the recurrent problem of shoulder pain in the Australian women’s cricket squad.
Appendix A Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jsams.2007.06.007.
References 1. Ross L. Physiotherapy WCA tour report world cup; 2000. 2. Ferdinands RED. Three-dimensional biomechanical analysis of fast bowling in cricket. Ph.D. thesis. University of Waikato; 2004. 3. Fleisig GS, Barrentine SW, Escamilla RF, et al. Biomechanics of overhand throwing with implications for injuries. Sports Med 1996;21:421—37. 4. Wang H-K, Cochrane T. Mobility impairment, muscle imbalance, muscle weakness, scapular asymmetry and shoulder injury in elite volleyball athletes. J Sports Med Phys Fitness 2001;41:403—10. 5. Vad VB, Gebeh A, Dines D, et al. Hip and shoulder internal rotation range of motion deficits in professional tennis players. J Sci Med Sport 2003;6:71—5.
6. Mabasa XM, Stewart AV, Fleishman C. Isokinetic strength of shoulder internal and external rotators in cricket bowlers. S Afr J Physiother 2002;58(2):35—8. 7. Aginsky KD, Lategan L, Stretch RA. Shoulder injuries in provincial male fast bowlers—–predisposing factors. S Afr J Sports Med 2004;16(1):25—8. 8. Gabbe BJ, Finch CF, Bennell KL, et al. How valid is a self reported 12 month sports injury history? Br J Sports Med 2003;37:545—7. 9. Ginn KA, Cohen ML. Conservative treatment for shoulder pain: prognostic indicators of outcome. Arch Phys Med Rehabil 2004;85:1231—5. 10. Mandalidis DG, Donne B, O’Regan M, et al. Reliability of isokinetic internal and external rotation in the scapular plane. Isokinet Exerc Sci 2001;9:65—72. 11. Wilcox RR, Keselman HJ. Modern robust data analysis methods: measures of central tendency. Psychol Methods 2003;8:254—74. 12. Grissom RJ, Kim JJ. Effect sizes for research. A broad practical approach. London: Lawrence Erlbaum Associates; 2005. 13. Orchard J, James T. Cricket Australia injury report; 2003. 14. Fahlström M, Yeap JS, Alfredson H, et al. Shoulder pain—–a common problem in world-class badminton players. Scand J Med Sci Sports 2006;16:168—73. 15. Wang H-K, Cochrane T. A descriptive epidemiological study of shoulder injury in top level English male volleyball players. Int J Sports Med 2001;22:159—63. 16. Osbahr DC, Cannon DL, Speer KP. Retroversion of the humerus in the throwing shoulder of college baseball pitchers. Am J Sports Med 2002;30:347—53. 17. Mihata T, Lee YS, McGarry MH, et al. Excessive humeral external rotation results in increased shoulder laxity. Am J Sports Med 2004;32:1278—85. 18. Wilk KE, Meister K, Fleisig G, et al. Biomechanics of the overhead throwing motion. Sports Med Arthrosc Rev 2000;8:124—34. 19. Loram LC, McKinon W, Wormgoor S, et al. Determinants of ball release speed in schoolboy fast-medium bowlers in cricket. J Sports Med Phys Fitness 2005;45: 483—90. 20. Myers P, O’Brien BP. Cricket: injuries, rehabilitation and training. Sports Med Arthrosc Rev 2001;9:124—36.
Available online at www.sciencedirect.com