238

Shoulder pain in wheelchair athletesThe role of muscle imbalance

ROBERT S. BURNHAM,*† MD, LAURA MAY,‡ BHSc, PT, EWEN NELSON,* BSc,ROBERT STEADWARD,* PhD, AND DAVID C. REID,§ MD

From the *Rick Hansen Centre, University of Alberta, the ‡Glenrose Rehabilitation Hospital, andthe §Glen Sather University of Alberta Sports Medicine Clinic, Edmonton, Alberta, Canada

ABSTRACT

Shoulder rotator cuff impingement syndrome is a com-mon and disabling problem for the wheelchair athlete.In this study we investigated the role of shoulder

strength imbalance as a factor for the development ofthis syndrome. Nineteen paraplegic male athletesunderwent clinical and isokinetic examination of bothshoulders with peak torque values measured in abduc-tion, adduction, and internal and external rotation.

Twenty athletic, able-bodied men without shoulderproblems were tested as controls. Ten (26%) of theparaplegic athletes had rotator cuff impingement syn-drome. The results of the isokinetic testing demon-strated that 1) the paraplegics’ shoulders were strongerthan the controls in all directions (P < 0.05); 2) thestrength ratio of abduction:adduction was higher forparaplegic athletes (P < 0.05); 3) paraplegics’ shoulderswith rotator cuff impingement syndrome were weakerin adduction and external and internal rotation than the

paraplegic athletes without impingement syndrome (P< 0.05); and 4) paraplegics’ shoulders with rotator cuffimpingement syndrome had higher abduc-tion:adduction and abduction:internal rotation strengthratios than the shoulders of paraplegics without im-pingement syndrome (P < 0.05). We concluded thatshoulder muscle imbalance, with comparative weak-ness of the humeral head depressors (rotators andadductors), may be a factor in the development andperpetuation of rotator cuff impingement syndrome inwheelchair athletes.

The high incidence of shoulder injuries in wheelchair sports,particularly basketball and road racing, has been well doc-

umented. 2,4,9,11,12 This is reflected in the injury statisticsfrom the 1988 Canadian Paralympic Team competing inSeoul, South Korea.’ Despite being recognized as a commonand disabling problem, little has been written in the sportsmedicine literature regarding the cause, prevention, or treat-ment of shoulder problems in wheelchair athletes.The orthopaedic literature contains research that identi-

fies shoulder problems to be prevalent among wheelchair-dependent nonathletes, 16,20 with the rotator cuff as the com-mon site of trauma. Bayley et aU found that 31 of 94 (33%)wheelchair-bound paraplegics had chronic, persistent shoul-der pain that was clinically diagnosed as rotator cuff im-pingement syndrome. Twenty-three underwent shoulder ar-thrography and 15 of the 23 (65%) had rotator cuff tears; 5had avascular necrosis of the humeral head. The mean onsetof the shoulder pain was 13 years after spinal cord injury.This suggests that the trauma to the rotator cuff did notoccur at the time of the spinal cord injury but was relatedto subsequent shoulder stresses. Pressure transducers placedinto the acromiohumeral space of 5 paraplegics withoutshoulder pain documented pressures up to 2.5 times themean arterial values generated by upper limb weightbearingduring transfers. Although 3 patients with rotator cuff tearsin that series underwent surgical treatment, conservativemanagement was recommended because of the significantfunctional impairment that resulted from prolonged post-operative arm immobilization.

Isokinetic shoulder strength balance has been previouslystudied in athletes involved in baseball pitching,’ swim-ming,5 and waterpolo.13 Significant differences in shoulderstrength balance as compared to controls were found. Theseincluded stronger shoulder internal rotators as compared toexternal rotators, and stronger adductors as compared toabductors. The possible role of these imbalances in the

development of tendinitis and anterior subluxation has beensuggested.

Biomechanically, any factors that preserve the acromio-

t Address correspondence and reprint requests to: R. S. Burnham, MD,Rick Hansen Centre, W1-67 Van Vliet Complex, Edmonton, Alberta, CanadaT6G 2H9.

239

humeral space could minimize the pressure and wear on therotator cuff. The deltoid muscle, when working alone, pullsthe humeral head cephalad, thus encroaching on the acrom-iohumeral space that houses the rotator cuff and subacro-mial bursaY Normally, this force is counterbalanced by the45° to 55° downward pull provided by the oblique rotatorcuff muscles-subscapularis, infraspinatus, and teres mi-nor.&dquo; Additionally, some of the powerful shoulder adduc-tors-latissimus dorsi, teres major, and lower fibers of pec-toralis major-exert a downward pull on the humerus whenthe trunk is fixed.&dquo; Therefore, a shoulder strength imbal-ance, with comparative weakness of the oblique rotator cuffmuscles and shoulder adductors in relation to the deltoid,could theoretically allow impingement of the soft tissuestructures within the acromiohumeral space.The purpose of this study was to 1) compare the isokinetic

shoulder strength of paraplegic male wheelchair athleteswith those of able-bodied individuals, and 2) compare theisokinetic strength of shoulders in paraplegic male athleteswith and without clinical evidence of rotator cuff impinge-ment syndrome.

METHOD

Nineteen paraplegic male wheelchair athletes voluntarilyparticipated after an explanation of the nature of the study.Twenty moderately fit, athletic, able-bodied male subjectswithout shoulder complaints functioned as the control

group. The mean age of the paraplegic athletes was 29 years,whereas the able-bodied controls averaged 25 years. Theparaplegia was traumatic in origin in 16 cases, 2 were relatedto polio, and 1 to spina bifida. The average duration ofparaplegia was 14 years. The level of the paraplegia was high(T1-5) in 4, midlevel (T6-12) in 11, and low (L-1 and below)in 4.

A clinical history and physical examination with specificreference to the shoulder were performed on each paraplegicsubject. A shoulder was defined as having rotator cuff im-pingement syndrome if it was painful to the athlete and hadat least two of the following clinical signs on physical ex-amination :

1. painful arc of abduction;2. pain in the &dquo;impingement positions&dquo; 7, 15;3. pain with resisted shoulder abduction, external rotation,

or forward flexion;4. tenderness to palpation over the greater tuberosity,

lesser tuberosity, or bicipital groove;5. wasting of the supra- or infraspinous fossae.The athletes then underwent an isokinetic shoulder

strength evaluation using a Cybex 340 dynamometer (CybexCo., Ronkonkoma, NY). Both shoulders of each athlete weretested in abduction/adduction (45° lean position) and inter-nal/external rotation (supine position) at a speed of 60 deg/sec following the protocol of Reid et al. 18,19 Special care wasnecessary to ensure adequate trunk stabilization, particu-larly in the high paraplegics. Descriptive statistics were usedto describe the demographic data of the wheelchair athletes.

Analysis of variance with Scheffe post-hoc tests were usedto compare peak torque and strength ratio values of thewheelchair and able-bodied athletes’ shoulders as well as thewheelchair athletes’ shoulders with and without rotator cuff

impingement syndrome. A significance level of P < 0.05 was

used.

RESULTS

On average, the paraplegic athletes were spending 96 hoursper week in their wheelchairs and were involved in wheel-chair sport 13 hours per week. Wheeling (including basket-ball and roadracing) made up 85% of the wheelchair sportactivity and weight training made up the remaining 15%.Ten of 38 shoulders (26%) of paraplegic athletes were in-volved with rotator cuff impingement syndrome. Six wereon the dominant side and 4 were on the nondominant side.No significant dominant versus nondominant side differ-

ence was found in absolute shoulder strength or strengthbalance among the paraplegic athletes. The paraplegic ath-letes were significantly stronger in shoulder abduction, ad-duction, and external and internal rotation compared to theable-bodied athletes (Fig. 1). Specifically, they were 62%stronger in abduction, 23% stronger in adduction, 61%stronger in external rotation, and 57% stronger in internalrotation. The ratio of shoulder abduction to adduction

strength was significantly higher in the paraplegic athletesas compared to the able-bodied athletes (Fig. 2). This sug-gests a relative weakness of shoulder adduction as compared

Figure 1. Isokinetic shoulder strength: paraplegic versus able-bodied athletes.

Figure 2. Isokinetic shoulder strength balance: paraplegicversus able-bodied athletes.

240

to abduction in the paraplegic athletes. No significant dif-ference in strength balance was found between the para-plegic and able-bodied athletes when comparing abductionto external rotation, abduction to internal rotation, andexternal rotation to internal rotation.Ten of the 38 wheelchair athletes’ shoulders met the

clinical criteria for rotator cuff impingement syndrome. Theabsolute peak torque strength and strength balance ratiosof these 10 shoulders were compared with the 28 wheelchairathletes’ shoulders without impingement syndrome. Theshoulders of paraplegics with impingement syndrome weresignificantly weaker in adduction, and in internal and exter-nal rotation, compared to those without impingement syn-drome (Fig. 3). The pattern of relative weakness of adductionas compared to abduction, which was seen previously in theparaplegic group as a whole, was even more exaggerated inthe group with impingement syndrome. The abduction tointernal rotation strength ratio was significantly higher inthe impingement group indicating a relative weakness of theinternal rotators (Fig. 4). No significant difference was foundin absolute abduction strength or in the balance betweenthe external and internal rotation strength.The strength balance of the paraplegic shoulders without

rotator cuff impingement syndrome did not differ signifi-cantly from that of the able-bodied controls.

DISCUSSION

It was not surprising that the wheelchair athletes werestronger around the shoulders than the able-bodied athletes.

Figure 3. Isokinetic shoulder strength: paraplegics’ shoulderswith versus without impingement syndrome.

Figure 4. Isokinetic shoulder strength balance: paraplegics’shoulders with versus without impingement syndrome.

The upper extremities are their limbs of propulsion andweightbearing. Considering that they are averaging over 96hours per week in a wheelchair and 13 hours of wheelchair

sport training, increases in shoulder girdle muscle strengthwould be expected.The observation that 10 of the 38 shoulders (26%) were

involved with rotator cuff impingement syndrome empha-sizes the troublesome nature of this disorder for the para-plegic athlete. Potential causes of this syndrome could bemyriad.This study suggests that muscle imbalance may be one of

those causes. The pattern of the imbalance found in theparaplegic athletes is much different than that described inswimmers, water polo players, and baseball pitchers, whowere found to have relatively stronger adductors and inter-nal rotators.5,S,13 In contrast, relative weakness of shoulderadduction, as compared to abduction, was found in theparaplegic athlete group as a whole. This pattern was evenmore exaggerated in the paraplegic athletes with rotator cuffimpingement syndrome. The presumption of this testing hasbeen that abduction strength reflects predominantly deltoidmuscle function. As indicated previously, without the bal-ancing force of the &dquo;shoulder depressors,&dquo; deltoid pulls thehumeral head cephalad. This constricts the acromiohumeralspace and may cause shoulder rotator cuff impingement. Itis generally accepted that it is the shoulder oblique rotatorsthat fulfill this important function. The potential role of themore powerful shoulder adductors (latissimus dorsi, teresmajor, and lower fibers of pectoralis major) in protecting theacromiohumeral space has not previously been emphasized.It is possible that relative shoulder adductor weakness hasa causative role in the development of rotator cuff impinge-ment syndrome and that adductor strengthening could bean effective prophylactic and active treatment modality.Resistance adduction training can be done simply and in-expensively using surgical tubing or dental dam. More elab-orate machinery can also be used by the paraplegic (Fig. 5).Preferably most of the adduction strengthening exercisesshould be performed with the arms below the level of theshoulders to minimize the impingement positioning.

Paraplegics’ shoulders affected by impingement syndromedemonstrate significant weakness of external and internalrotation strength, as well as reduced internal rotation

strength balance, compared to abduction. It is not clearwhether this deficit is the cause of, or the result of, theshoulder rotator cuff impingement syndrome. Nonetheless,it represents a functional deficit, and strength training ofthe shoulder internal and external rotators, as well as the

adductors, is recommended as part of the rehabilitationtreatment of a wheelchair athlete with shoulder rotator cufftendinitis. This is incorporated once the acute inflammatoryphase of the tendinitis settles, as described in the rehabili-tation program outlined by Mendoza et a1.14 for able-bodiedathletes with shoulder tendinitis. Again, the rotation train-ing can be performed easily and inexpensively using surgicaltubing or dental dam (Fig. 6). We recommend commencingthe rotation exercises with the shoulder in neutral (full

241

Figure 5. Training for adductors.

adduction) to minimize any mechanical impingement. Grad-ually, more diagonal rotational exercises can be performed.

Undoubtedly, the cause of rotator cuff impingement syn-drome in this group of athletes is multifactorial. With the

repetitive upward axial loading of the shoulder that occurswith upper extremity weightbearing, inadequate shoulderdepressor strength appears to be one important predisposingfactor. Other possible factors would include chronic overuseand repetitive impingement positioning. The shoulders ofthe wheelchair-bound individual rarely get rest and recoverytime simply because of the demands of wheelchair propul-sion. Superimposed on this, the wheelchair athlete trainsintensively, and many are involved throughout the year witha variety of wheelchair sports. Careful attention to the

structuring of training schedules with adequate recoverytime is particularly important for the wheelchair athlete.Because the wheelchair-bound individual is lower to the

ground, reaching up into shoulder flexion and abduction isa common arm position for everyday activities. Additionally,sporting activities such as shooting a basketball and someweight training maneuvers (i.e., inclined bench presses, la-tissimus pull downs) require repetitive shoulder positioningin the impingement positions. It has also been observed thata shoulder internal rotation and scapular protraction posi-tion is common for the wheelchair athlete both at rest andwith heavy wheeling (Fig. 7). This also can predispose toimpingement of the greater tuberosity under the coracoacro-mial arch.6 Treatment for these factors include trainingprogram modifications so that work in the shoulder impinge-ment positions is minimized (i.e., limiting the number ofrepetitive foul shots per practice session, doing shorter arc

Figure 6. Shoulder rotator training using surgical tubing.

Figure 7. Internal rotation shoulder impingement position withwheeling.

242

latissimus pull downs, and using a declined instead of in-clined bench press). Posture training to encourage scapularretraction and avoiding the protracted scapula/internallyrotated shoulder position may also be helpful.The findings and recommendations of this paper may also

have application to wheelchair dependent nonathletes andable-bodied athletes involved in arm dominant sports whoare prone to develop rotator cuff impingement syndrome.

CONCLUSIONS

1. Paraplegic athletes are isokinetically stronger in shoul-der abduction, adduction, and internal and external rotationthan their able-bodied counterparts.

2. The paraplegic athletes demonstrated an imbalance ofabduction to adduction strength with comparative weaknessof shoulder adduction.

3. Rotator cuff impingement syndrome was commonamong the paraplegic athletes examined.

4. Paraplegics’ shoulders clinically involved with impinge-ment syndrome demonstrated significant weakness of ad-duction, external rotation, and internal rotation as comparedto the uninvolved shoulders.

5. The imbalance of abduction to adduction strength withcomparative weakness of shoulder adduction was signifi-cantly more severe in the paraplegics’ shoulders affected byimpingement syndrome. Additionally, the affected shouldersdemonstrated a comparative weakness of internal rotationas compared to abduction.

6. Shoulder adduction and external and internal rotation

strengthening may be an important modality for the preven-tion and treatment of shoulder rotator cuff impingementsyndrome among paraplegic athletes.

REFERENCES

1. Bayley JC, Cochran TP, Sledge CB: The weight-bearing shoulder: Theimpingement syndrome in paraplegics. J Bone Joint Surg 69A: 676-678,1987 prevention implications. Physician Sportsmed 14(9): 97-105, 1986

2. Bloomquist LE: Injuries to athletes with physical disabilities&mdash;preventionimplications. Physician Sportsmed 14(9): 97-105, 1986

3. Burnham RS, Newell E, Steadward R: Sports medicine for the physicallydisabled&mdash;the Canadian team experience at the 1988 Seoul ParalympicGames. Clin J Sport Med 1: 193-196, 1991

4. Ferrara MS, Davis RW: Injuries to elite wheelchair athletes. Paraplegia 28:335-341,1990

5. Fowler PJ: Upper extremity swimming injuries, in The Upper Extremity inSports Medicine. St. Louis, CV Mosby, 1990, pp 891-896

6. Hawkins RJ, Hobeika PE: Impingement syndrome in the athletic shoulder.Clin Sports Med 2: 391-405, 1983

7. Hawkins RJ, Kennedy JC: Impingement syndrome in athletes. Am J SportsMed 8: 151-158, 1980

8. Hinton RY: Isokinetic evaluation of shoulder rotational strength in highschool baseball pitchers. Am J Sports Med 16: 274-279, 1988

9. Hoeberigs JH, Verstappen FTJ: Muscle soreness in wheelchair basket-ballers. Int J Sports Med 5 (suppl): 177-179, 1984

10. Hollinshead WH, Jenkins DB: Functional Anatomy of the Limbs and Back.5th edition. Philadelphia, WB Saunders Company, 1981, pp 83-111

11. Martinez SF: Medical concerns among wheelchair road racers. PhysicianSportsmed 17(2): 63-68, 1989

12. McCormack DAR, Reid DC, Steadward RD, et al: Injury profiles in wheel-chair athletes: Results of a retrospective survey. Clin J Sport Med 1: 35-40, 1991

13. McMaster WC, Long SC, Caiozzo VJ: Isokinetic torque imbalances in therotator cuff of the elite water polo player. Am J Sports Med 19: 72-75,1991

14. Mendoza FX, Nicholas JA, Sands A: Principles of shoulder rehabilitation inthe athlete, in The Upper Extremity in Sports Medicine. St. Louis, CVMosby Co, 1990, pp 262

15. Neer CS, Welsh RP: The shoulder in sports. Orthop Clin North Am 8: 583-591, 1977

16. Nichols PJR, Norman PA, Ennis JR: Wheelchair users shoulder? Scand JRehabil Med 11: 29-32, 1979

17. Perry J: Shoulder anatomy and biomechanics. Clin Sports Med 2: 247-270, 1983

18. Reid DC, Oedekoven G, Kramer KF, et al: Isokinetic muscle strengthparameters for shoulder movements. Clin Biomech 4(2): 97-104, 1989

19. Reid DC, Saboe L, Burnham RS: Current research of selected shoulderproblems. Clin Phys Ther 9: 225-237, 1991

20. Wing PC, Tredwell SJ: The weightbearing shoulder. Paraplegia 21: 107-113, 1983