[Orthopedic Reviews 2012; 4:e18] [page 79]

Subacromial impingement syndromeMasood Umer, Irfan Qadir, Mohsin AzamOrthopaedic Surgery Department, AgaKhan University Hospital, Stadium Road,Karachi, Pakistan

Abstract

Subacromial impingement syndrome (SAIS)represents a spectrum of pathology rangingfrom subacromial bursitis to rotator cufftendinopathy and full-thickness rotator cufftears. The relationship between subacromialimpingement and rotator cuff disease in theetiology of rotator cuff injury is a matter ofdebate. However, the etiology is multi-factori-al, and it has been attributed to both extrinsicand intrinsic mechanisms. Managementincludes physical therapy, injections, and, forsome patients, surgery. No high-quality ran-domized controlled trials are available so far toprovide possible evidence for differences inoutcome of different treatment strategies.There remains a need for high-quality clinicalresearch on the diagnosis and treatment ofSAIS.

Introduction

Subacromial impingement syndrome (SAIS)is the most common disorder of shoulder,accounting for 44-65% of all complaints ofshoulder pain during a physician’s office visit.1

SAIS encompasses a spectrum of subacromialspace pathologies including partial thicknessrotator cuff tears, rotator cuff tendinosis, cal-cific tendinitis, and subacromial bursitis. Themain consequences of SAIS are functional lossand disability.2

Subacromial space is defined by the humer-al head inferiorly, the anterior edge and undersurface of the anterior third of the acromion,coracoacromial ligament and the acromioclav-icular joint superiorly. The height of spacebetween acromion and humeral head rangesfrom 1.0 to 1.5 centimeters as seen on radi-ographs. Interposed between these twoosseous structures are the rotator cuff ten-dons, the long head of the biceps tendon, thebursa, and the coracoacromial ligament. Anyabnormality that disturbs the relationship ofthese subacromial structures may lead toimpingement.3

Neer described three stages of impinge-ment.4 Stage-I impingement is characterizedby edema and hemorrhage of the subacromialbursa and cuff. It is typically found in patients

who are less than twenty-five years old. Stage-II impingement represents irreversiblechanges, such as fibrosis and tendinitis of therotator cuff, and is typically found in patientswho are twenty-five to forty years old. Stage-IIIimpingement is marked by more chronicchanges, such as partial or complete tears ofthe rotator cuff, and usually is seen in patientswho are more than forty years old. Given the high prevalence of this condition,

the aim of this review is to evaluate the differ-ent etiological theories that may explain SAIS.The different anatomical structures involvedin this type of impingement are described; theclinical findings are presented and treatmentguidelines are suggested.

Glenohumeral joint kinematicsThe glenohumeral joint possesses six

degrees of freedom, three rotations and threetranslations.5 With active in vivo glenohumer-al abduction in the scapular plane (approxi-mately 30-40º anterior to the frontal plane), thehumerus concomitantly externally rotates.External rotation is important for clearance ofthe greater tuberosity and its associated tis-sues as it passes under the coracoacromialarch, as well as for relaxation of the capsularligamentous constraints to allow maximumglenohumeral elevation.6

Translation of the humeral head in the mag-nitude of 1-3 mm in the superior directionoccurs in the first 30-60º of active glenohumer-al scapular plane elevation. After the initialphase of elevation in the scapular plane orfrontal plane abduction, the humeral headremains somewhat centered on the glenoidcavity with fluctuations between inferior andsuperior translations of typically less than 1mm. The glenohumeral joint demonstratesessentially ball and socket kinematics aboveapproximately 60º of glenohumeral elevation.7

Superior humeral translation that occursduring the initial phase of elevation appears tobe due in part to the cranially directed pull onthe head of the humerus by the deltoid mus-cle.5 Humeral head translations in the anteri-or-posterior directions have been less wellinvestigated. Anterior humeral head transla-tions in the magnitude of 2-5 mm have beendemonstrated during passive glenohumeralflexion. During active glenohumeral flexion,anterior humeral head translation of less than1 mm occurs over the course of motion.8

The height of the subacromial space, fromthe head of the humerus to the coracoacromi-al arch, is only 1.0-1.5 cm as seen on radi-ographs. Changes of this space occur in sub-jects with healthy shoulders; a decrease in thewidth of the acromio-humeral interval and anincrease in the contact between the inferioracromion and underlying subacromial tissuesoccurs during glenohumeral abduction.Contact pressure and force in the subacromial

space has also been demonstrated to increaseduring glenohumeral abduction. Theoretically,these changes in the subacromial space wouldbe accentuated with an increase in the normalsuperior and anterior humeral head transla-tion, leading to mechanical compression of thetissues in subacromial space during gleno-humeral motion.9

Scapulothoracic articulation kine-maticsScapula and thoracic cage form the scapu-

lothoracic articulation. This articulation isassessed kinematically either two-dimension-ally or three-dimensionally. The joint is typical-ly described with five degrees of freedom,three rotations and two translations.5

The scapula demonstrates a pattern ofupward rotation, external rotation, and poste-rior tilting during glenohumeral elevation. Thethree-dimensional analysis of scapular motionby van der Helm and Pronk describe scapularupward rotation occurring about an anterior-posterior axis, with the inferior angle of thescapula moving laterally;10 external rotationoccurring about a superior-inferior axis, withthe lateral border of the scapula moving poste-riorly; posterior tilt occurring about a medial-lateral axis, with the inferior angle movinganteriorly.Less well examined are scapular transla-

tions, depicted as scapular positions. Scapularpositions can be represented by clavicular rota-tions about the sternoclavicular joint in two

Orthopedic Reviews 2012; volume 4:e18

Correspondence: Irfan Qadir, OrthopaedicSurgery Department, Aga Khan UniversityHospital, Stadium Road, Room 211, Male hostel,Aga Khan University Hospital, Karachi 74800,Pakistan. Tel. +92.3007321600E-mail: muhamad_irfan_qadir@yahoo.com

Key words: subacromial impingement syndrome,subacromial bursitis, shoulder pain, rotator cufftendinopathy.

Contributions: MU, IQ, literature review andmanuscript preparation; MA, help in literatureresearch.

Conflict of interests: the authors report no poten-tial conflict of interests.

Received for publication: 1 February 2012.Accepted for publication: 16 April 2012.

This work is licensed under a Creative CommonsAttribution NonCommercial 3.0 License (CC BY-NC 3.0).

©Copyright M. Umer et al., 2012Licensee PAGEPress, ItalyOrthopedic Reviews 2012; 4:e18doi:10.4081/or.2012.e18

[page 80] [Orthopedic Reviews 2012; 4:e18]

different planes: clavicular elevation/depres-sion for superior/inferior translation and clav-icular protaction/retraction for anterior/poste-rior translation. The assumption is made thatmotion of the clavicle at the sternoclavicularjoint will be in direct relationship to scapulartranslation, because of the interposed rigidbone (clavicle) between these two joints andthe lack of significant motion occurring at theacromioclavicular joint. During glenohumeralelevation the clavicle retracts posteriorly andelevates, putting the scapula in essentially amore superior and posterior position.11

Subjects with subacromial impingementgenerally have decreased scapular posteriortilting, decreased upward rotation andincreased internal rotation compared tohealthy subjects. Weak or dysfunctional scapu-lar musculature, fatigue of the infraspinatusand teres minor and changes in thoracic andcervical spine posture have all demonstrated achange in scapular kinematics.12

Etiology

SAIS is an encroachment of the subacromi-al tissues as a result of narrowing of the sub-acromial space. Mechanisms of rotator cuff(RC) tendinopathy have been classicallydescribed as extrinsic, intrinsic or a combina-tion of both. Intrinsic impingement, theorizesthat partial or full thickness tendon tears occuras a result of the degenerative process thatoccurs over time with overuse, tension over-load, or trauma of the tendons.5 An alternativetheory is that of extrinsic impingement, whereinflammation and degeneration of the tendonoccur as a result of mechanical compression bystructures external to the tendon.4 A uniquesubset of extrinsic impingement, internalimpingement occurs due to compression of thearticular side rather than the bursal side of theRC tendons, between the posterior superiorglenoid rim and humerus when the arm is infull external rotation, abduction, and exten-sion. Although internal impingement can beconsidered an extrinsic mechanism, narrow-ing of the subacromial space is not a hallmarkfinding.12

Extrinsic ImpingementExtrinsic mechanisms of RC tendinopathy

that result in bursal sided RC tendon compres-sion due to narrowing of the subacromialspace include anatomical factors, biomechani-cal factors, or a combination. The acromio-humeral distance (AHD), a linear measurebetween the acromion and the humeral headused to quantify the subacromial space, hasbeen studied in patients with RC disease usingmagnetic resonance imaging,13 ultrasonogra-phy,14 and radiographs.13 AHD is normally

between 7 and 14 mm in healthy shoulders. Itis reduced in SAIS patients with the muscles atrest or during muscle activation which func-tionally narrow the subacromial space.Furthermore, AHD less than 7 mm with thearm at rest is a predictive indicator of lessfavorable surgical outcome.15

Anatomical factorsAnatomical factors that may excessively nar-

row the subacromial space and outlet to the RCtendons include variations in shape of theacromion, orientation of the slope/angle of theacromion or prominent osseous changes to theinferior aspect of the acromio-clavicular (AC)joint or coracoacromial ligament.12 A widelyused classification system for acromial shapeis flat (type I), curved (type II), or hooked (typeIII), which was developed from observations of139 shoulder specimens.3 Whether acromialshape is congenital or acquired with ageremains controversial.12 Another possible cul-prit of encroachment into the subacromialspace is thickening of the coracoacromial liga-ment. Significant relationships have beendemonstrated between acromion morphologyand patient’s self reported shoulder functionand the severity of the rotator cuff pathology.5

However, surgical decompression of coraco-acromial arch has only been demonstrated toproduce comparable level of successful out-come as non-operative treatment. This sug-gests that direct encroachment of the subacro-mial space by the coracoacromial arch soft tis-sue or bony changes is not the only mecha-nism of impingement.5,12 On the other hand,these coraco-acromial arch changes have sig-nificant effect on tendon injury when com-bined with overuse activity. Supporting thistheory of a requisite overuse exposure, symp-tomatic RC disease is more often present indominant than nondominant shoulders.16

Biomechanical factorsBiomechanical mechanism of extrinsic SAIS

is based on dynamic narrowing of the subacro-mial space leading to RC tendon compressionsecondary to superior translation of thehumeral head or aberrant scapular motion thatcauses the acromion to move inferiorly. Theseinclude shortening of the posterior-inferiorglenohumeral joint capsule and decreased RCmuscle performance.5,12

Posterior capsulePosterior capsular tightness may cause

changes in glenohumeral kinematics leadingto SAIS. When posterior capsular tightness wassurgically induced in cadavers, there was an inincrease in superior and anterior humeralhead translations during passive glenohumer-al flexion. Excessive superior and anteriorhumeral head translations can decrease the

size of the subacromial space, leading toincreased mechanical compression of the sub-acromial structures.9 Glenohumeral internalrotation range of motion and horizontal adduc-tion at 90° of elevation are reliable clinicalmeasures that potentially assess posterior cap-sule length. Furthermore, stretching toaddress impairments of posterior shouldertightness has been identified as an importantcomponent to rehabilitation for patients withRC tendinopathy.17

Scapular musculatureAberrant scapular muscle activity has been

identified in patients with SAIS and beendirectly linked to abnormal scapular kinemat-ics. Of particular interest are the relative con-tributions of the upper and lower serratusanterior muscles and trapezius muscles, foundto stabilize the scapula and induce scapularupward rotation, external rotation, and/or pos-terior tilt to potentially allow the humeral headto clear the acromion with elevation.18 Theseindividuals have decreased muscle perform-ance of the trapezius and serratus anterior interms of force output,19 muscle balance/ratios,19 electromyographical activity,18 andlatencies in activation.20

Relatively small changes in the muscle per-formance of the scapulothoracic muscles canalter the position of the scapula at a fixedangle of humeral elevation and, in theory,affect the length-tension relationship (pointon the length-tension curve) of the RC mus-cles and the subacromial space.12

SpineA relatively small increase in thoracic spine

flexion has resulted in a more elevated andanteriorly tilted scapula at rest, and lessupward rotation and posterior tilt duringglenohumeral elevation. An increase in tho-racic spine flexion has also resulted in adecrease in the amount of elevation of theglenohumeral joint and a decrease in theamount of force generated at 90º of gleno-humeral scapular plane abduction.21

Rotator cuff musculatureThe supraspinatus along with the other

rotator cuff muscles of teres minor, infraspina-tus, and subscapularis serve to maintain thecongruent contact between the humeral headand the glenoid fossa by producing a compres-sive force during glenohumeral movements.5

Weakness or dysfunctional rotator cuff muscu-lature can lead to changes in glenohumeraland scapulothoracic kinematics. Excessivesuperior translation of the humeral headresulting from rotator cuff weakness can leadto a decrease in the subacromial space duringelevation, and thus increased mechanical com-pression of the subacromial contents.22

Review

[Orthopedic Reviews 2012; 4:e18] [page 81]

Clinical Evaluation

HistoryAlthough impingement symptoms may arise

following trauma, the pain more typicallydevelops insidiously over a period of weeks tomonths. The pain is typically localized to theanterolateral acromion and frequently radiatesto the lateral mid-humerus. Patients usuallycomplain of pain at night, exacerbated by lyingon the involved shoulder, or sleeping with thearm overhead. Normal daily activities such ascombing one’s hair or reaching up into a cup-board become painful. Weakness and stiffnessmay also be encountered, but they are usuallysecondary to pain.2

Physical examinationIn their systematic analysis,

Papadonikolakis et al.23 concluded that thephysical findings used to diagnose theimpingement syndrome, i.e., the Neer sign(pain on forced flexion), the Hawkins sign(pain on internal rotation with the arm elevat-ed to 90_), and the Neer injection test (reliefof pain on the Neer sign after subacromialinjection of local anesthetic) may be sensitive,but are not specific. The average sensitivity(and standard deviation) of the Neer sign was76±11%, while the average specificity was36±22%. The respective values for theHawkins sign were 80±11% and 41±19%. Intheir meta-analysis, Hegedus et al.24 concludedthat neither the Neer nor the Hawkins signhad diagnostic utility for impingement.

ImagingStandard radiographs including internal and

external rotation anteroposterior, scapular Y,axillary, and Supraspinatus outlet views areimportant for the thorough evaluation ofshoulder pain. These plain radiographs mayshow characteristic changes of rotator cuff dis-ease, including subacromial osteophytes,sub-acromial sclerosis, cystic changes of thegreater tuberosity, and narrowing of theacromiohumeral distance, they are not defini-tive.15,25

Magnetic resonance imaging (MRI) pro-vides detail of potential sites of subacromialimpingement through the supraspinatus out-let. Ossification of the coracoacromial liga-ment (CAL) or presence of a subacromial spurcan be best identified in the sagittal obliqueplane; however, differentiation of a pathologicspur and the normal CAL can be difficult. MRIalso may demonstrate findings of subacromi-al/subdeltoid bursitis. Findings that indicatethis condition include bursal thickness >3mm, the presence of fluid medial to theacromioclavicular joint, and the presence offluid in the anterior aspect of the bursa.

Typically, MRI is performed with the armadducted; however, this position does notrecreate the position of impingement.26

TreatmentMany treatments are available for impinge-

ment syndrome such as physical therapy,shock-wave therapy, medication, and surgery.In the last decade, several (systematic)reviews on treatment for impingement syn-drome were published.27-31 These reviews com-pared the effectiveness of treatments on a vari-ety of outcome measures, including pain,range of movement, functional limitations,and return to work. Hence, the conclusion oneffectiveness of various treatments was prima-rily based on the combination of these out-come measures. There is strong evidence that extracorporeal

shock-wave therapy is no more effective thanplacebo,32,33 moderate evidence that ultrasoundtherapy is no more effective than placebo,34

and limited evidence that laser is no moreeffective than placebo with regard to function-al limitations.35With regard to the improve-ment in functional limitations there is limitedevidence that exercise is more effective thanno intervention,36 and moderate evidence thatexercise combined with manual therapy ismore effective than exercise alone.37

There is limited evidence for the effective-ness of the following interventions: exercise ismore effective than no intervention on func-tional limitations, oral diclofenac is moreeffective than analgesic injections, both onfunctional limitations and on ability to workafter 1 year. On the short term, arthroscopicacromioplasty is more effective than openacromioplasty with regard to functional limita-tions and return to work. However, moderateevidence exists that on the long term open andarthroscopic acromioplasty are equally effec-tive with regard to functional limitations.38

A systematic review by Dorrestijn et al. ledthe authors to conclude that according to thebest-evidence synthesis, however, there is noevidence from the available randomized con-trolled trials for differences in outcome in painand shoulder function between conservativelyand surgically treated patients with subacromi-al impingement syndrome.39 However, severalobservational studies report a significantlybetter outcome in operated-on patients whohad not responded to non-operative measuresand who had a short symptom duration com-pared with those who had prolonged symptomsbefore surgery.38

Summary

RC tendinopathy is a common disorder thatposes challenges for effective treatment.

Evidence suggests that extrinsic, intrinsic, andcombinations of biomechanical mechanismsplay a role. There are no significant differ-ences in outcome between conservatively andsurgically treated patients with subacromialimpingement syndrome. For most patientswith SAIS, nonsurgical treatment is success-ful. Surgical intervention is successful inpatients who fail nonsurgical treatment.Surgeon experience and intraoperative assess-ment may guide the method of surgical treat-ment. Studies have shown that many surgicalinterventions, including debridement andopen and arthroscopic acromioplasty, havebeen successful. However, there remains aneed for high-quality clinical research on thediagnosis and treatment of SAIS.

References

1. van der Windt DA, Koes BW, de Jong BA,Bouter LM. Shoulder disorders in generalpractice: incidence, patient characteris-tics, and management. Ann Rheum Dis1995;54:959-64.

2. Koester MC, George MS, Kuhn JE.Shoulder impingement syndrome. Am JMed 2005;118:452-5.

3. Bigliani LU, Levine WN. Subacromialimpingement syndrome. J Bone Joint SurgAm 1997;79:1854-68.

4. Neer CS 2nd. Impingement lesions. ClinOrthop Relat Res 1983:70-7.

5. Michener LA, McClure PW, Karduna AR.Anatomical and biomechanical mecha-nisms of subacromial impingement syn-drome. Clin Biomech (Bristol, Avon)2003;18:369-79.

6. Browne AO, Hoffmeyer P, Tanaka S, et al.Glenohumeral elevation studied in threedimensions. J Bone Joint Surg Br 1990;72:843-5.

7. Ludewig PM, Cook TM. Translations of thehumerus in persons with shoulderimpingement symptoms. J Orthop SportsPhys Ther 2002;32:248-59.

8. Wuelker N, Schmotzer H, Thren K, KorellM. Translation of the glenohumeral jointwith simulated active elevation. ClinOrthop Relat Res 1994:193-200.

9. Flatow EL, Soslowsky LJ, Ticker JB, et al.Excursion of the rotator cuff under theacromion. Patterns of subacromial con-tact. Am J Sports Med 1994;22:779-88.

10. van der Helm FC, Pronk GM. Three-dimen-sional recording and description ofmotions of the shoulder mechanism. JBiomech Eng 1995;117:27-40.

11. Karduna AR, McClure PW, Michener LA,Sennett B. Dynamic measurements ofthree-dimensional scapular kinematics: avalidation study. J Biomech Eng 2001;123:

Review

[page 82] [Orthopedic Reviews 2012; 4:e18]

184-90.12. Seitz AL, McClure PW, Finucane S, et al.

Mechanisms of rotator cuff tendinopathy:intrinsic, extrinsic, or both? Clin Biomech(Bristol, Avon) 2011;26:1-12.

13. Saupe N, Pfirrmann CW, Schmid MR, et al.Association between rotator cuff abnor-malities and reduced acromiohumeral dis-tance. AJR Am J Roentgenol 2006;187:376-82.

14. Cholewinski JJ, Kusz DJ, Wojciechowski P,et al. Ultrasound measurement of rotatorcuff thickness and acromio-humeral dis-tance in the diagnosis of subacromialimpingement syndrome of the shoulder.Knee Surg Sports Traumatol Arthrosc2008;16:408-14.

15. Weiner DS, Macnab I. Superior migrationof the humeral head. A radiological aid inthe diagnosis of tears of the rotator cuff. JBone Joint Surg Br 1970;52:524-7.

16. Yamaguchi K, Ditsios K, Middleton WD, etal. The demographic and morphologicalfeatures of rotator cuff disease. A compar-ison of asymptomatic and symptomaticshoulders. J Bone Joint Surg Am 2006;88:1699-704.

17. Myers JB, Oyama S, Wassinger CA, et al.Reliability, precision, accuracy, and validi-ty of posterior shoulder tightness assess-ment in overhead athletes. Am J SportsMed 2007;35:1922-30.

18. Diederichsen LP, Norregaard J, Dyhre-Poulsen P, et al. The activity pattern ofshoulder muscles in subjects with andwithout subacromial impingement. JElectromyogr Kinesiol 2009;19:789-99.

19. Cools AM, Witvrouw EE, Declercq GA, et al.Evaluation of isokinetic force productionand associated muscle activity in thescapular rotators during a protraction-retraction movement in overhead athleteswith impingement symptoms. Br J SportsMed 2004;38:64-8.

20. Moraes GF, Faria CD, Teixeira-Salmela LF.Scapular muscle recruitment patterns and

isokinetic strength ratios of the shoulderrotator muscles in individuals with andwithout impingement syndrome. JShoulder Elbow Surg 2008;17Suppl1:48S-53S.

21. Kebaetse M, McClure P, Pratt NA. Thoracicposition effect on shoulder range ofmotion, strength, and three-dimensionalscapular kinematics. Arch Phys MedRehabil 1999;80:945-50.

22. Reddy AS, Mohr KJ, Pink MM, Jobe FW.Electromyographic analysis of the deltoidand rotator cuff muscles in persons withsubacromial impingement. J ShoulderElbow Surg 2000;9:519-23.

23. Papadonikolakis A, McKenna M, Warme W,et al. Published evidence relevant to thediagnosis of impingement syndrome of theshoulder. J Bone Joint Surg Am 2011;93:1827-32.

24. Hegedus EJ, Goode A, Campbell S, et al.Physical examination tests of the shoul-der: a systematic review with meta-analy-sis of individual tests. Br J Sports Med2008;42:80-92.

25. House J, Mooradian A. Evaluation andmanagement of shoulder pain in primarycare clinics. South Med J 2011;103:1129-35.

26. Harrison AK, Flatow EL. Subacromialimpingement syndrome. J Am Acad OrthopSurg. 2011;19:701-8.

27. Checroun AJ, Dennis MG, Zuckerman JD.Open versus arthroscopic decompressionfor subacromial impingement. A compre-hensive review of the literature from thelast 25 years. Bull Hosp Jt Dis 1998;57:145-51.

28. Buchbinder R, Green S, Youd JM.Corticosteroid injections for shoulderpain. Cochrane Database Syst Rev 2003:CD004016.

29. Green S, Buchbinder R, Hetrick S.Physiotherapy interventions for shoulderpain. Cochrane Database Syst Rev2003:CD004258.

30. Green S, Buchbinder R, Glazier R, ForbesA. Interventions for shoulder pain.Cochrane Database Syst Rev 2000:CD001156.

31. Desmeules F, Cote CH, Fremont P.Therapeutic exercise and orthopedic man-ual therapy for impingement syndrome: asystematic review. Clin J Sport Med. 2003May;13(3):176-82.

32. Schmitt J, Haake M, Tosch A, et al. Low-energy extracorporeal shock-wave treat-ment (ESWT) for tendinitis of thesupraspinatus. A prospective, randomisedstudy. J Bone Joint Surg Br 2001;83:873-6.

33. Speed CA, Richards C, Nichols D, et al.Extracorporeal shock-wave therapy fortendonitis of the rotator cuff. A double-blind, randomised, controlled trial. J BoneJoint Surg Br 2002;84:509-12.

34. Downing DS, Weinstein A. Ultrasoundtherapy of subacromial bursitis. A doubleblind trial. Phys Ther 1986;66:194-9.

35. Vecchio P, Cave M, King V, et al. A double-blind study of the effectiveness of low levellaser treatment of rotator cuff tendinitis.Br J Rheumatol 1993;32:740-2.

36. Ludewig PM, Borstad JD. Effects of a homeexercise programme on shoulder pain andfunctional status in construction workers.Occup Environ Med 2003;60:841-9.

37. Bang MD, Deyle GD. Comparison of super-vised exercise with and without manualphysical therapy for patients with shoulderimpingement syndrome. J Orthop SportsPhys Ther 2000;30:126-37.

38. Faber E, Kuiper JI, Burdorf A, et al.Treatment of impingement syndrome: asystematic review of the effects on func-tional limitations and return to work. JOccup Rehabil 2006;16:7-25.

39. Dorrestijn O, Stevens M, Winters JC, et al.Conservative or surgical treatment forsubacromial impingement syndrome? Asystematic review. J Shoulder Elbow Surg2009;18:652-60.

Review