Secondary Logo

Journal Logo

Advanced Search
CLINICAL RESEARCH

Altered Cocontraction Patterns of Humeral Head Depressors in Patients with Subacromial Pain Syndrome: A Cross-sectional Electromyography Analysis

Overbeek, Celeste L. MD; Kolk, Arjen MD; de Groot, Jurriaan H. PhD; Visser, Cornelis P.J. MD, PhD; van der Zwaal, Peer MD, PhD; Jens, Axel BSc; Nagels, Jochem MD; Nelissen, Rob G.H.H. MD, PhD

Author Information

C.L. Overbeek, A. Kolk, J. H. de Groot, A. Jens, J. Nagels, R.G.H.H. Nelissen, Department of Orthopaedics, Leiden University Medical Center, Leiden, the Netherlands

C.L. Overbeek, A. Kolk, J. H. de Groot, Laboratory for Kinematics and Neuromechanics, Department of Orthopaedics and Rehabilitation, Leiden University Medical Center, Leiden, the Netherlands

C.P.J. Visser, Department of Orthopaedics, Haaglanden Medical Center, The Netherlands

P. v.d. Zwaal, Department of Orthopaedics, Alrijne Hospital, The Netherlands.

C. L. Overbeek MD, Department of Orthopaedics, Leiden University Medical Center, Post zone J11-R, PO box 9600, 2300 RC Leiden, the Netherlands, Email: [email protected]

All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.

Each author certifies that neither he or she, nor any member of his or her immediate family, have funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article.

The institution of one or more of the authors (R.G.H.H.N.) has received funding from the Dutch Arthritis Foundation (grant number 13-1-303).

Each author certifies that his or her institution approved or waived approval for the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research.

This study was performed at the Laboratory for Kinematics and Neuromechanics, Department of Orthopaedics and Rehabilitation, Leiden University Medical Center, Leiden, the Netherlands.

Clinical Orthopaedics and Related Research: August 2019 - Volume 477 - Issue 8 - p 1862-1868
doi: 10.1097/CORR.0000000000000745
  • Free

Abstract

Introduction

Chronic shoulder pain is the second most common musculoskeletal disorder in the general population, with prevalence rates ranging between 15% and 22% [1, 23, 26]. A specific anatomic basis for perceived symptoms, such as full-thickness rotator cuff tears or calcific tendinitis, is observed in many patients [11]. However, in approximately 29% to 34% of all patients with chronic shoulder pain, referred to here as subacromial pain syndrome (SAPS), the subacromial (suprahumeral) tissues are inflamed, but there is no structural anatomic cause that could explain persisting symptoms [7, 27]. The fact that altering bony shapes with surgical interventions yields unsatisfactory results comparable to those of physical therapy also suggests that behavioral and/or central pain mechanisms may play a more important role than previously assumed [3, 19].

In patients with SAPS, pain is frequently exacerbated during abduction, suggesting that motion-related (kinematic) factors contribute to the perpetuation of symptoms [7, 11]. Open MRI and radiographic studies have attributed this particular pain pattern, the painful arc, to insufficient humeral-head depression during abduction [9, 12]. We believe that patients with SAPS could benefit from mechanical unloading of the subacromial tissues during abduction by the active contribution of shoulder muscles that act as humeral-head depressors [13, 14, 22]. The craniocaudal position of the humerus relative to the scapula is directly determined by a balance of cranial forces generated by the shoulder abductors and caudal forces generated by cocontraction of the rotator cuff and arm adductors [16, 25]. In both research and clinical practice, there has been a focus on the rotator cuff, while the arm adductors, specifically the teres major, contribute the most to depression of the humerus during abduction [14, 16, 25].

Because of this, studying cocontraction of the arm adductors in patients with SAPS seems to be worthwhile; if adductor cocontraction is altered in patients with SAPS, this could indicate a treatable imbalance between the abductors and adductors. Accordingly, we asked: Do patients with SAPS have altered contraction patterns of the arm adductors during abduction compared with asymptomatic people?

Patients and Methods

In this study, we defined SAPS as shoulder pain lasting for longer than 3 months with no specific anatomic abnormalities that could explain the pain and could benefit from specific treatment (such as acromioclavicular osteoarthritis, calcific tendinitis, or full-thickness rotator cuff tears) [11]. Between April 2010 and September 2016, consecutive patients with a clinical diagnosis of SAPS who were referred to the outpatient clinics of the Leiden University Medical Center, Haaglanden Medical Center, or Alrijne Hospital were evaluated for inclusion in this cross-sectional cohort study (Trial registry number NTR2283) [6]. Patients were selected through clinical examination, radiographs, and MR arthrography.

The inclusion criteria were unilateral shoulder pain for at least 3 months, positive results of a Hawkins-Kennedy test (passive anteflexion of the shoulder to 90° with subsequent internal rotation of the shoulder to provoke subacromial pain) and Neer lidocaine impingement test (examining for immediate relief of pain after subacromial infiltration with lidocaine), and at least one of the following symptoms: pain during daily life activities with arm abduction, extension, and/or internal rotation; pain at night or incapability of lying on the shoulder; painful arc; diffuse pain during palpation of the greater tuberosity; scapular dyskinesis; and a positive full-can test, empty-can test, or Yocum test result [6]. Exclusion criteria were insufficient Dutch-language skills, age younger than 35 years or older than 60 years, inflammatory arthritis of the shoulder, clinical signs of glenohumeral or acromioclavicular osteoarthritis, previous fracture, dislocation or surgery of the shoulder, cervical radiculopathy, glenohumeral instability, decreased passive function (frozen shoulder), malignancy, full-thickness rotator cuff tears, calcific tendinitis, labrum or ligament pathology, pulley lesion, biceps tendinopathy, os acromiale, cartilage lesion, or a bony cyst [6]. All MR arthrography studies were evaluated by an experienced independent radiologist [7]. Of 85 patients who were referred with the clinical diagnosis of SAPS, 45 were excluded, leaving 40 patients for evaluation in this study (Fig. 1).

F1
Fig. 1:
A flow diagram of the participant inclusion process is shown.

To select control participants, we recruited the spouses of patients with musculoskeletal complaints at the outpatient clinic of Leiden University Medical Center between January 2016 and November 2016. Inclusion criteria were age between 35 and 60 years, no current or past shoulder concerns, no visit to a physician for shoulder-related concerns, and no past shoulder discomfort for more than 1 week. Exclusion criteria were impaired passive and active shoulder function during clinical examination, insufficient Dutch-language skills, prior shoulder surgery, injections, shoulder fracture or dislocation, radiculopathy, frozen shoulder, osteoarthritis or rheumatoid arthritis, and neurologic or muscle disease. No additional imaging was performed in the control group, because we only used imaging in the SAPS group to exclude specific anatomic conditions that would have explained a patient’s symptoms.

Forty patients with SAPS and 30 asymptomatic controls were compared. The SAPS and control groups were not different with respect to age, sex, and hand dominance (Table 1).

T1
Table 1.:
Baseline characteristics of patients with SAPS and controls

The research was conducted according to the principles of the Declaration of Helsinki (64th WMA General Assembly, Fortaleza, Brazil, October 2013) and in accordance with the Medical Research Involving Human Subjects Act. The review board of our institutional ethical medical commission approved this study (P09.227 & P15.046) and all participants provided written informed consent.

We used a standardized testing protocol [8]. In this study, we were interested in evaluating the activation patterns of muscles that directly act on the craniocaudal position of the humerus with respect to the scapula during abduction. In biomechanical evaluations and a recent systematic review on the topic, it has been shown that the deltoid muscle contributes the most to upward migration of the humerus during abduction [16, 25]. The arm adductors, specifically the latissimus dorsi, teres major, and, to a lesser extent, the pectoralis major, are the strongest humeral-head depressors during abduction [16, 25]. Other muscles that may contribute to humeral depression are the teres minor and the lower parts of the infraspinatus and subscapular muscles [25]. Because evaluating these muscles with EMG requires indwelling (fine wire) electrodes, we limited our evaluation to the deltoid, latissimus dorsi, teres major, and pectoralis major muscles.

With the target arm in external rotation at the side and attached to a one-dimensional force transducer at the wrist, we recorded activation of the latissimus dorsi, teres major, pectoralis major (clavicular part), and deltoid muscles (medial part) with surface EMG during rest and isometric abduction and adduction tasks (DelSys system Bagnoli-16, Boston, MA, USA; inter-electrode distance, 10 mm; bandwidth 20-450 Hz ) [8].

The EMG and force signals were analog-digitally converted and recorded simultaneously at a sample rate of 2500 Hz. Offline, the EMG signals were subtracted from the mean EMG signals for offset removal, rectified, and combined with the moving average over intervals of 0.1 seconds, using custom-made software in Matlab (MathWorks Inc., R2018b, Natick, MA, USA).

Participants first performed maximal abduction and adduction tasks to determine the maximum voluntary force. The maximum voluntary force was set as the lowest value of either the maximum voluntary force during isometric adduction or abduction. Subsequently, a target force of 60% with a tolerance of ± 3.75% of the maximum voluntary force was presented to the participants on a computer screen [8]. Finally, participants performed a 15-second isometric force task in abduction and adduction at equal target force levels for the purpose of computing a standardized measure of the degree of antagonistic versus agonistic activation. Measurements were performed twice, and both assessments were used in this study to reduce variability [6]. We quantified muscle activation with the AR using the following equation:

“Muscle” represents either the latissimus dorsi, teres major, pectoralis major, or deltoid muscle, and IP and OP indicate “in-phase” agonist activation and “out-phase” antagonist muscle activation, respectively, relative to the force task in abduction or adduction.

The AR ranges between -1 and 1 and indicates the task-related degree of antagonist activation relative to the same muscle’s degree of agonist activation. The AR equals 1 in case of sole agonist muscle activation and decreases with increasing cocontraction. An AR of -1 indicates sole antagonistic activation and no agonistic activation. Based on this AR, we assessed the difference in muscle activation patterns between patients with SAPS and asymptomatic controls. We also assessed the influence of the target force level on the activation ratio in the statistical analysis.

To prevent overestimation of the degree of cocontraction as assessed with the AR, the mean EMG amplitude during the agonistic task (the activity of the deltoid muscle during abduction and the activity of adductors during adduction) was verified to be twice the mean EMG amplitude of the 10% lowest EMG signals during the relative rest, abduction, and adduction tasks (a signal-to-noise ratio of ≥ 2.0). If this condition was not met or if EMG data were corrupt (because of loose electrodes or other technical problems), the ARs were excluded. In the SAPS group, one of 40 ARs of the latissimus dorsi (2.5%), one AR of the teres major (2.5%), one AR of the pectoralis major (2.5%), and one AR of the deltoid muscle (2.5%) had to be excluded because the twofold signal-to-noise ratio was not reached. For the same reason, four of 30 ARs of the latissimus dorsi (13%) and two ARs of the teres major (6.7%) had to be excluded in the control group. Furthermore, because of a disconnected EMG amplifier, three of 30 ARs of the deltoid muscle (10%), four ARs of the latissimus dorsi (13%), four ARs of the pectoralis major (13%) and three ARs of the teres major (10%) could not be used in the control group. Additionally, in the control group, one of 30 ARs of the latissimus dorsi (3%) and one AR of the pectoralis major (3%) could not be used because of a broken electrode.

Categorical data are described with numbers and percentages and continuous parameters are described with means and either 95% CIs, SDs, or medians with the 25th and 75th percentiles, depending on data distributions. Demographic data were compared using independent-samples t-tests or Mann-Whitney’s U test depending on the distribution of data. Linear mixed-model analyses assessed the differences in activation ratios between patients with SAPS and asymptomatic controls. There were separate analyses for each assessed muscle. The dependent variable was ARmuscle and the measurement moment was the repeated factor. Independent variables were the patient groups (SAPS or asymptomatic controls) and the target force level (60% maximum voluntary force). An effect size of 0.10 < AR < 0.20 was subjectively considered to be a modest effect size.

The patient and control populations were recruited in two different studies, and no a priori sample size analysis was performed for the AR. The statistical analysis was performed using SPSS® version 23 (IBM® Corp., Armonk, NY, USA). The results of the linear mixed-model analyses are presented as estimated regression coefficients, 95% CIs, and p values. A two-sided p value of 0.05 or less was considered statistically significant.

Results

Patients with SAPS showed less cocontraction of the teres major and more cocontraction of the pectoralis major than controls did. Patients with SAPS had a 0.11 higher AR of the teres major (95% CI, 0.01-0.21; p = 0.038), a 0.11 lower AR of the pectoralis major (95% CI, -0.18 to -0.04; p = 0.003), and a 0.12 lower AR of the deltoid muscle (95% CI, -0.17 to -0.06; p < 0.001) than controls did. In terms of effect size, these differences were considered to be modest. With the number of patients and controls available, there was no difference in the degree of latissimus dorsi cocontraction between the groups (difference = 0.05, 95% CI, -0.01 to 0.12; p = 0.120) (Table 2). The average activation ratios of the latissimus dorsi, teres major, pectoralis major, and deltoid muscle were 0.78 (SD 0.14), 0.53 (SD 0.23), 0.68 (SD 0.23), and 0.73 (SD 0.18) in patients with SAPS, and 0.73 (SD 0.19), 0.41 (0.27), 0.82 (0.08), and 0.85 (0.10) in controls (Fig. 2).

T2
Table 2.:
Difference in activation ratios between patients with SAPS and controls
F2
Fig. 2:
The activation ratios of four shoulder muscles in patients with SAPS and controls are shown. LD = latissimus dorsi; TM = teres major; PM = pectoralis major; DM = deltoid muscle

Discussion

In approximately 29% to 34% of all patients with SAPS, there is no anatomic explanation for symptoms, and behavioral aspects and/or central pain mechanisms may play a more important role than previously assumed [7, 27]. A possible behavioral explanation for pain in patients with SAPS is insufficient active depression of the humerus during abduction achieved by adductor cocontraction [13, 14, 22]. The adductor muscles, specifically the teres major, have the most important contribution to humeral depression; however, this subject has not been studied well in patients with SAPS [16, 22]. In the current study, we sought to determine whether patients with SAPS have altered cocontraction patterns of the arm adductors compared with asymptomatic controls, which could point towards a treatable imbalance between the abductor and adductor muscles. We found that patients with SAPS predominantly contracted with the pectoralis major, while controls did so with the teres major. To unload subacromial tissues, it may be more effective to cocontract with the teres major [14, 16].

Our study results have some limitations. First, the common chronic symptoms in the SAPS group were likely caused by a variety of anatomic factors. SAPS is a syndrome, not a specific anatomic diagnosis, and our conclusions need to be interpreted in light of this. A specific anatomic cause should be sought in patients presenting with shoulder pain, and treatment, other than nonsurgical therapies, should only be initiated when a specific, treatable anatomic basis is found for the symptoms [3]. Second, no a priori analysis was performed because the patient and control populations were recruited in two studies. Negative results may therefore have originated from underpowering. In light of the large amount of missing data for the latissimus dorsi in the control group (nine of 30 ARs; 30%), it is plausible that there is a difference between patients with SAPS and controls in cocontraction of the latissimus dorsi. We may have not been able to detect this difference because of underpowering. We do not consider it likely that missing data introduced a bias because the predominant cause was failure of the EMG equipment (20 of 26 total missing values; 77%). Third, we did not control for potential confounding variables such as sports participation and BMI. Although we selected our control group from the patients’ spouses, we cannot exclude the influence of these factors. Fourth, we only evaluated a selection of muscles that affect the craniocaudal position of the humerus the most [14, 16, 25]. Our conclusion may be supported by adding an analysis of other adductors, for example, the teres minor and lower parts of the infraspinatus and subscapularis.

Few studies have assessed the activation patterns of arm adductors during abduction tasks in patients with SAPS [2, 5, 10], and these studies contradicted one another, perhaps because of small sample sizes [5] or different testing positions [2]. In addition to altered adductor activation patterns, we observed a lower activation ratio of the deltoid muscle in patients with SAPS, originating from reduced activation during abduction. As suggested in previous studies [21, 24], it seems that patients with SAPS attempt to avoid pain by reducing abductor activation at the cost of function (that is, reduced target force level). Cocontraction with the teres major, as we observed in the control group, may protect the patient from pain while preserving function.

Our EMG assessment of muscles that determine the craniocaudal position of the humerus during abduction provides new insight regarding the function of the teres major. Patients with SAPS predominantly cocontracted with the pectoralis major, whereas controls did so with the teres major (a glenohumeral muscle). Both muscles contribute to glenohumeral stabilization. However, to reduce loading on subacromial tissues, it is more effective to use teres major cocontraction because this muscle is more capable of pulling the humerus downward (away from the acromion) than other muscles are [14, 16].

Recently, it has been shown that surgical interventions commonly yield unsatisfactory results in treating patients with SAPS, and physical therapy is preferable [3, 15, 19]. We believe that such nonsurgical approaches can be improved with targeted approaches [15]. Supporting our findings, other clinical studies have suggested that increasing cocontraction of the arm adductors is a viable treatment option for patients with SAPS [4, 13, 22]. To improve targeted treatment approaches that enhance teres major (and latissimus dorsi) cocontraction in patients with SAPS, we suggest performing trials in which EMG is used [17, 18, 20].

In this cross-sectional EMG evaluation, we found decreased cocontraction of the teres major and increased cocontraction of the pectoralis major in patients with SAPS. We based our study on the rationale that insufficient humeral depression during abduction leads to perpetuation of SAPS, by overloading of the subacromial tissues [7, 9, 12]. For depressing the humerus, increasing teres major cocontraction as observed in the control group could be more effective [14, 16, 22]. Future studies using EMG monitoring should assess if actively training teres major (and latissimus dorsi) cocontraction could be a target for physical therapy protocols for patients with SAPS.

Acknowledgments

We thank Pieter Bas de Witte MD, PhD for conducting a noteworthy part of the data collection.

References

1. Andersson HI, Ejlertsson G, Leden I, Rosenberg C. Chronic pain in a geographically defined general population: studies of differences in age, gender, social class, and pain localization. Clin J Pain. 1993;9:174-182.
2. Bandholm T, Rasmussen L, Aagaard P, Jensen BR, Diederichsen L. Force steadiness, muscle activity, and maximal muscle strength in subjects with subacromial impingement syndrome. Muscle Nerve. 2006;34:631-639.
3. Beard DJ, Rees JL, Cook JA, Rombach I, Cooper C, Merritt N, Shirkey BA, Donovan JL, Gwilym S, Savulescu J, Moser J, Gray A, Jepson M, Tracey I, Judge A, Wartolowska K, Carr AJ, Group CS. Arthroscopic subacromial decompression for subacromial shoulder pain (CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial. Lancet. 2017;391:329-338.
4. Beaudreuil J, Lasbleiz S, Aout M, Vicaut E, Yelnik A, Bardin T, Orcel P. Effect of dynamic humeral centring (DHC) treatment on painful active elevation of the arm in subacromial impingement syndrome. Secondary analysis of data from an RCT. Br J Sports Med. 2015;49:343-346.
5. de Witte PB, Henseler JF, van Zwet EW, Nagels J, Nelissen RG, de Groot JH. Cranial humerus translation, deltoid activation, adductor co-activation and rotator cuff disease - different patterns in rotator cuff tears, subacromial impingement and controls. Clin Biomech. 2014;29:26-32.
6. de Witte PB, Nagels J, van Arkel ER, Visser CP, Nelissen RG, de Groot JH. Study protocol subacromial impingement syndrome: the identification of pathophysiologic mechanisms (SISTIM). BMC Musculoskelet Disord. 2011;12:282.
7. de Witte PB, Overbeek CL, Navas A, Nagels J, Reijnierse M, Nelissen RG. Heterogeneous MR arthrography findings in patients with subacromial impingement syndrome - Diagnostic subgroups? J Electromyogr Kinesiol. 2016;29:64-73.
8. de Witte PB, van der Zwaal P, Visch W, Schut J, Nagels J, Nelissen RG, de Groot JH. Arm adductor with arm abduction in rotator cuff tear patients vs. healthy -- design of a new measuring instrument [corrected]. Hum Mov Sci. 2012;31:461-471.
9. Deutsch A, Altchek DW, Schwartz E, Otis JC, Warren RF. Radiologic measurement of superior displacement of the humeral head in the impingement syndrome. J Shoulder Elbow Surg. 1996;5:186-193.
10. Diederichsen LP, Norregaard J, Dyhre-Poulsen P, Winther A, Tufekovic G, Bandholm T, Rasmussen LR, Krogsgaard M. The activity pattern of shoulder muscles in subjects with and without subacromial impingement. J Electromyogr Kinesiol. 2009;19:789-799.
11. Diercks R, Bron C, Dorrestijn O, Meskers C, Naber R, de Ruiter T, Willems J, Winters J, van der Woude HJ; Dutch Orthopaedic Association. Guideline for diagnosis and treatment of subacromial pain syndrome: a multidisciplinary review by the Dutch Orthopaedic Association. Acta Orthop. 2014;85:314-322.
12. Graichen H, Bonel H, Stammberger T, Haubner M, Rohrer H, Englmeier KH, Reiser M, Eckstein F. Three-dimensional analysis of the width of the subacromial space in healthy subjects and patients with impingement syndrome. AJR Am J Roentgenol. 1999;172:1081-1086.
13. Graichen H, Hinterwimmer S, von Eisenhart-Rothe R, Vogl T, Englmeier KH, Eckstein F. Effect of abducting and adducting muscle activity on glenohumeral translation, scapular kinematics and subacromial space width in vivo. J Biomech. 2005;38:755-760.
14. Halder AM, Zhao KD, Odriscoll SW, Morrey BF, An KN. Dynamic contributions to superior shoulder stability. J Orthop Res. 2001;19:206-212.
15. Hanratty CE, McVeigh JG, Kerr DP, Basford JR, Finch MB, Pendleton A, Sim J. The effectiveness of physiotherapy exercises in subacromial impingement syndrome: a systematic review and meta-analysis. Semin Arthritis Rheum. 2012;42:297-316.
16. Hik F, Ackland DC. The moment arms of the muscles spanning the glenohumeral joint: a systematic review. J Anat. 2019;234:1-15.
17. Holtermann A, Mork PJ, Andersen LL, Olsen HB, Sogaard K. The use of EMG biofeedback for learning of selective activation of intra-muscular parts within the serratus anterior muscle: a novel approach for rehabilitation of scapular muscle imbalance. J Electromyogr Kinesiol. 2010;20:359-365.
18. Huang HY, Lin JJ, Guo YL, Wang WT, Chen YJ. EMG biofeedback effectiveness to alter muscle activity pattern and scapular kinematics in subjects with and without shoulder impingement. J Electromyogr Kinesiol. 2013;23:267-274.
19. Kolk A, Thomassen BJW, Hund H, de Witte PB, Henkus HE, Wassenaar WG, van Arkel ERA, Nelissen R. Does acromioplasty result in favorable clinical and radiologic outcomes in the management of chronic subacromial pain syndrome? A double-blinded randomized clinical trial with 9 to 14 years' follow-up. J Shoulder Elbow Surg. 2017;26:1407-1415.
20. Larsen CM, Juul-Kristensen B, Olsen HB, Holtermann A, Sogaard K. Selective activation of intra-muscular compartments within the trapezius muscle in subjects with subacromial impingement syndrome. A case-control study. J Electromyogr Kinesiol. 2014;24:58-64.
21. Michaud M, Arsenault AB, Gravel D, Tremblay G, Simard TG. Muscular compensatory mechanism in the presence of a tendinitis of the supraspinatus. Am J Phys Med. 1987;66:109-120.
22. Overbeek CL, Kolk A, Nagels J, de Witte PB, van der Zwaal P, Visser CPJ, Fiocco M, Nelissen R, de Groot JH. Increased co-contraction of arm adductors is associated with a favorable course in subacromial pain syndrome. J Shoulder Elbow Surg. 2018;27:1925-1931.
23. Picavet HS, Schouten JS. Musculoskeletal pain in the Netherlands: prevalences, consequences and risk groups, the DMC(3)-study. Pain. 2003;102:167-178.
24. Reddy AS, Mohr KJ, Pink MM, Jobe FW. Electromyographic analysis of the deltoid and rotator cuff muscles in persons with subacromial impingement. J Shoulder Elbow Surg. 2000;9:519-523.
25. Steenbrink F, de Groot JH, Veeger HE, van der Helm FC, Rozing PM. Glenohumeral stability in simulated rotator cuff tears. J Biomech. 2009;42:1740-1745.
26. United States Bone and Joint Initiative. The Burden of Musculoskeletal Diseases in the United States (BMUS), Fourth Edition. Rosemont, IL. Available at: http://www.boneandjointburden.org. Accessed February 12, 2019.
27. Vecchio P, Kavanagh R, Hazleman BL, King RH. Shoulder pain in a community-based rheumatology clinic. Br J Rheumatol. 1995;34:440-442.
© 2019 by the Association of Bone and Joint Surgeons