Abstract
Introduction
Complex Regional Pain Syndrome type 1 (CRPS1) is a potential complication, affecting the prognosis of functional joint recovery. Its incidence ranges from 2 to 40% depending on the series and the joints involved. Very few studies have evaluated the incidence of CRPS after shoulder surgery. The objective of our study was to determine the incidence of CRPS1 and to identify any pre-operative risk factors associated with its emergence after extra-articular subacromial space surgery.
Material and Methods
This is a retrospective single-centre study of patients who underwent surgery for a subacromial extra-articular shoulder pathology from January 2016 to December 2016 and included a follow-up period of at least 6 months. The primary inclusion criterion was developing a CRPS1 as defined by Veldman. A pre- and post-operative clinical assessment was performed based on the Constant (Cst) score.
Results
Among the 287 patients, with an average follow-up period of 6.5 months, included in the study, 38 (13%) presented with post-operative CRPS1. Treated hypothyroidism (OR = 3.79; 95% CI 1.58;9.07; p = 0.003), open surgery (OR = 2.92; 95% CI 1.35–6.32; p = 0.007) and the level of daily physical activity from the Cst score (OR = 0.088; 95% CI 0.79;0.97; p = 0.015) were found to be significantly associated with the onset of CRPS1.
Conclusion
CRPS1 affected more than 10% of patients who underwent surgery for a subacromial shoulder pathology. The current study identified hypothyroidism, open surgery, and pre-operative clinical status as risk factors for the onset of this complication. These parameters should, therefore, be taken into consideration during the patient's pre-operative consultation.
Keywords: Complex regional pain syndrome, CRPS, Acromioplasty, Rotator cuff repair, Arthroscopy, Sub-acromial surgery
Introduction
Complex Regional Pain Syndrome type 1 (CRPS1) is a potential complication that affects the clinical outcome of any surgical procedure. Its diagnosis is based on a combination of various clinical manifestations including severe pain, allodynia, oedema, local hyperthermia, trophic disorders and skin dyschromia [1–3]. Its incidence ranges from 2 to 40% depending on the series and the joints involved [4, 5]. Its frequency increases to more than 48% after a wrist fracture [6]. The complication entails additional medical costs, prolongation of rehabilitative care and above all it impacts the quality of life in the 6 to 12 months following the operation with a number of social, professional and psychological repercussions [4–10]. Shoulder surgery, in particular subacromial space surgery, is not exempt from this complication [10–12]. Yet the analysis of CRPS risk factors and the actual frequency of the complication are still controversial issues [9, 13–20].
The objective of the current study was to evaluate the incidence of CRPS1 after subacromial shoulder surgery and to identify potential risk factors associated with the onset of this complication.
Material and Methods
The current study is a single centre retrospective study of prospectively acquired data. Patients gave their informed consent to participate in this study. It completes all reference methodology criteria of the French National Commission for Informatics and Liberties (CNIL) and the University Hospital of Toulouse, France approved this study (RnIPH 2019-36) and confirmed that ethic requirements were totally respected.
The study included adult patients (> 18 years of age) who underwent shoulder surgery for a subacromial space pathology from the 1st of January 2016 to the 31st of December 2016. Surgeries encompassed rotator cuff tendon repair, tenotomy or tenodesis of the long biceps, acromioplasty, acromioclavicular arthroplasty or exeresis of calcification. All surgeries were performed by two fellowship-trained shoulder surgeons. All procedures were performed with the patient under general anaesthesia in the beach-chair position. The minimum post-operative follow-up period was expected to extend to 6 months.
Patients who underwent emergency surgery, who had a surgical history involving the affected shoulder, or whose surgery involved glenohumeral prosthetic arthroplasty, instability surgery, or osteosynthesis of the proximal end of the humerus were excluded.
Postoperative Protocol
Postoperative rehabilitation protocol depends on operative procedure:
-In the case of acromioplasty, acromioclavicular arthroplasty, calcific tendinitis or tenotomy of the long head of biceps, immobilization was short (< 8 days). Patients could start active rehabilitation immediately without restriction.
-In case of rotator cuff repair, shoulders were immobilized for six weeks in a sling. Isometric and pendulum exercises and relaxation of the muscles around the shoulder were initiated on the day after surgery. From the second week, the passive exercises were initiated, avoiding the provocation of pain. The active reeducation is started from the 6th week.
Clinical Assessment
Clinical assessments were performed at 45 days, 3 months, and 6 months after the surgery. The systematic physical examination focused on assessing active and passive mobility, as well as evaluating pain on an analogue scale.
The primary endpoint was post-operative onset of CRPS1 diagnosed according to the Veldman criteria [21], and was based on the presence of 4 of the following 5 symptoms: unexplained diffuse pain, the difference in skin colour with respect to the contralateral limb, diffuse oedema, the difference in skin temperature with respect to the contralateral limb, and limitation of joint motion range. Symptoms were required to persist or increase after mobilisation and the area affected was required to be larger than the operative site.
The secondary assessment criterion was the Constant (Cst) score performed pre-operatively and at least 6 months after surgery [22–24].
Statistical Analyses
Before doing any analyses, the power of the study for the hypothesis was assessed. We estimated that with 139 eligible patients and 15 predictors, we would have a statistical power level of 0.8, assuming a medium anticipated effect size (f2) of 0.15 and a significance level of 0.05.
Continuous variables were expressed by their mean and interquartile ranges (IQR); and categorical variables by their frequency and distribution.
We analysed the probability of developing CRPS at 6 months based on clinical (range of joint motion, mobility, strength, pain, constant score), demographic [gender, age, co-morbidity(ies)] and therapeutic (surgical intervention) variables.
Bivariate analyses of potential predictors of CRPS onset were performed using Student’s t-test (parametric) or the Wilcoxon’s rank-sum test (non-parametric) and the χ2 (parametric) or Fisher’s exact (non-parametric) tests. Simple logistic regression analysis between CRPS and individual potential predictor variables was performed to select variables (p value < 0.2) to include in the multiple logistic regression model. Collinearity between co-variables was evaluated using the variance inflation factor. Adjusted multiple logistic regression was used to identify any independent CRPS risk factors 6 months after extra-articular shoulder surgery. A top-down step-by-step procedure based on the minimisation of Akaike’s information criterion (i.e. generalised R2, c-statistic, and Akaike’s information criterion) [25] was used to select variables.
The analysis was performed with R 3.1.2 (R: A Language and Environment for Statistical Computing [R Foundation for Statistical Computing, Vienna, Austria; https://www.R-project.org). The confidence interval (CI) was set at 95% and p < 0.05 was considered significant.
Results
Cohort Characteristics
A total of 295 patients underwent surgery, 8 were lost to follow-up. The mean age at the time of surgery was 61 years (ranging from 24 to 87) and 48% (138) of patients were women. The Constant score was 48.7 points (ranging from 0 to 89) and pre-operative SSV was 42.4% (ranging from 0 to 100). An arthroscopy was performed in 71% of cases (n = 204). Epidemiological data of the study population are presented in Table 1.
Table 1.
Comparison of demographic, clinical and surgical data
| General population | Group I | Group II | p value | |
|---|---|---|---|---|
| No post-operative CRPS1 | Post-operative CRPS1 | |||
| n = 287 | n = 249 | n = 38 | ||
| Age at surgery, mean (SD) | 61 (11.1) | 61.4 (11.3) | 59.2 (9.3) | 0.12 |
| Female gender, n (%) | 138 (48%) | 114 (46%) | 24 (63%) | 0.05 |
| Dominant side, n (%) | 199 (69%) | 174 (70%) | 25 (66%) | 0.6 |
| Pre-operative assessment | ||||
| Constant Score (100 pts) average (SD) | 48.7 (13.4) | 53.6 (20.7) | 45.6 (16.7) | 0.01 |
| Pain (15 pts) | 7.7 (3) | 16.4 (6.1) | 15.5 (5.5) | 0.39 |
| Level of physical activity (10 pts) | 10.2 (3.7) | 11.4 (3.7) | 9.8 (3.2) | 0.02 |
| Mobility (40 pts) | 28.2 (7.5) | 28.5 (7.6) | 26.3 (7) | 0.03 |
| Anterior elevation (10 pts) | 7.5 (2.2) | 7.5 (2.2) | 7.1 (2.3) | 0.2 |
| Abduction (10 pts) | 6.8 (2.2) | 6.9 (2.2) | 6.2 (2) | 0.02 |
| External rotation (10 pts) | 7.3 (2.3) | 7,4 (2.3) | 6,8 (2.2) | 0.04 |
| Internal rotation (10 pts) | 6.5 (2.2) | 6,6 (2.3) | 6,1 (2) | 0.12 |
| Strength (25 pts) | 4.7 (3.6) | 4.8 (3.6) | 3.5 (3.4) | 0.007 |
| Comorbidities | ||||
| Diabetes type 1, n (%) | 4 (1%) | 4 (2%) | 0 | 1 |
| Diabetes type 2, n (%) | 13 (5%) | 12 (5%) | 1 (3%) | 1 |
| Dyslipidaemia, n (%) | 46 (16%) | 30 (12%) | 8 (21%) | 0.4 |
| Hypothyroidism, n (%) | 37 (13%) | 26 (10%) | 11 (29%) | 0.004 |
| Depression ± anxiety, n (%) | 56(20%) | 47 (19%) | 9 (24%) | 0.5 |
| History of CRPS1, n (%) | 4 (1%) | 3 (1%) | 1 (3%) | 0.4 |
| Surgical technique | ||||
| Appearance of lesions n (%) | 0.4 | |||
| Progressive | 177 (62%) | 151 (61%) | 26 (68%) | |
| Traumatic | 110 (38%) | 98 (39%) | 12 (32%) | |
| Occupational injury/illness, n (%) | 49 (17%) | 41 (16%) | 8 (21%) | 0.5 |
| Access n (%) | 0.02 | |||
| Open | 83 (29%) | 66 (27%) | 17 (45%) | |
| Arthroscopy | 204 (71%) | 183 (73%) | 21 (55%) | |
| Type of surgery, n (%) | ||||
| Acromioplasty | 250 (87%) | 215 (86%) | 35 (92%) | 0.4 |
| Tenotomy or tenodesis of the long biceps | 253 (88%) | 219 (88%) | 34 (89%) | 1 |
| Rotator cuff repair | ||||
| One tendon | 152 (53%) | 128(51%) | 24 (63%) | 0.2 |
| Several tendons | 52 (18%) | 48 (19%) | 4 (11%) | 0.2 |
| Resection of the distal 1/4 of the clavicle | 48 (17%) | 39 (16%) | 9 (24%) | 0.2 |
| Calcifications | 24 (8%) | 19 (8%) | 5 (13%) | 0.3 |
n absolute value, SD standard deviation
CRPS1 Risk Factors
At a mean follow-up period of 6.5 months, the incidence of CRPS1 was 13% (n = 38).
The logistic regression model identified several risk factors: open surgery (OR = 2.92; 95% CI 1.35–6.32; p = 0.007) and dysthyroidism (OR = 3.79; 95% CI 1.58;9.07; p = 0.003). A high level of daily physical activity on the pre-operative Constant score was a protective factor for developing CRPS1 (OR = 0.088; 95% CI 0.79;0.97; p = 0.015) (Table 2).
Table 2.
Results of multiple logistic regression analysis
| Variables | Adjusted odds ratio (IC 95%) | p value |
|---|---|---|
| Open surgery | 2.92 (1.35–6.31) | < 0.01 |
| Hypothyroidism | 3.79 (1.58–9.07) | < 0.01 |
| Pre-operative level of physical activity | 0.88 (0.79–0.97) | 0.01 |
CI Confidence Interval
Discussion
The objective of this study was to evaluate the incidence and pre-operative risk factors associated with the onset of type 1 complex regional pain syndrome after subacromial space surgery.
Our study identified a post-operative incidence of CRPS1 of almost 13%. This figure is consistent with a recent prospective study by Koorevaar et al. [10] which found an 11% incidence across all shoulder surgeries combined. Evans et al. [12] report a 5% incidence of the frozen shoulder following basic shoulder arthroscopy (acromioplasty ± acromioclavicular arthroplasty or tenotomy of the long biceps).
Several studies in the literature analysed the incidence and potential risk factors for developing CRPS1 in joints other than the shoulder. Rewhorn et al. [4] identified an CRPS1 incidence of 4% after foot or ankle surgery. Jellad et al. [26] reported a CRPS1 incidence of 32% in their study of distal radius end fractures treated orthopedically. In the case of elective hand surgery, the incidence varied between 2 and 40% depending on the type of surgery and the series [27–29].
We were able to demonstrate a significant increase in the risk of developing CRPS1 in thyroid disorder patients treated for hypothyroidism. Neither diabetes nor dyslipidaemia were identified as potential risk factors. Koreevar et al., however, showed that type 2 diabetes was predictive of CRPS1. In addition to all surgeries combined which also increased risk, Huang et al. [30] reported that the risk of developing frozen shoulder in the general population was increased by a factor of 1.22 in hyperthyroid patients. Diabetes and hypercholesterolemia were also identified as risk factors [9].
Our study did not confirm the female gender to be a risk factor for developing algoneurodystrophy. Several studies reported a higher incidence of CRPS1 in women compared to men, following orthopedically or surgically treated wrist fractures [7, 26]. Roh et al. [7] have indeed also shown that the incidence of CRPS1 in women surgically treated for a distal radius fracture was 11% compared to only 6% in men. According to Rewhorn et al. [4], women were more frequently affected since they accounted for 82 of the patients affected by CRPS1 in their study.
Rewhorn et al. [4] also identified anxiety and a history of depression as potential risk factors. Our study did not confirm this observation possibly due to the small size of this sub-population within our cohort.
With regards to the intensity of trauma, results remained divergent among the studies involving radius fractures [7, 26].
Conventional open surgery appeared to pose a greater risk than arthroscopy in our study. This is in contrast to the Koorevaar et al. study [10] which found that arthroscopy was a positive predictive factor. It would, however, appear that this relates to the type of surgery rather than the arthroscopy per se. In our study, CRPS was encountered after all types of assessed operative procedures but none of them could be identified as a specific risk factor. These different procedures were often associated with each other during the same surgical intervention, which makes it difficult to analyse each of them separately.
Evans et al. [12] identified a history of the frozen shoulder as a CRPS1 risk factor. This potential risk factor was not identified in our study. The small number of patients with a history of CRPS1 (n = 4) limited our ability to specifically quantify the contribution of this parameter to the onset of post-operative CRPS1.
We did not identify a significant association between the total pre-operative Cts score and post-operative CRPS1. However, the pre-operative “level of physical activity” was found to be significantly lower in patients affected post-operatively. This factor had a relatively low absolute value, which explains its lack of impact on the total score. We also found no correlation between pain, pre-operative mobility and post-operative CRPS1. However, patients with a stiff and tender shoulder, who were not deemed to require surgery, were not included in our study.
Patients who developed CRPS1 in our study experienced a significant clinical impact. The total Constant score and SSV were reduced by 28% and 38% respectively. Koorevar et al. [10] showed that for patients with post-operative frozen shoulder, the Constant score was reduced by 11% compared to the group of healthy patients. The DASH score increased by 18%, confirming the deterioration in the quality of life and the clinical importance of preventing the onset of this specific pathology.
Our study has a number of limitations. The first is related to its retrospective design. It leads to data loss. Furthermore, given the difference in the number of patients between the open surgery and the arthroscopy groups studied, it seems difficult to draw a definitive conclusion regarding the influence of one technique over the other on the incidence of CRPS1. Another limitation may be introduced using the Veldman criteria to define CRPS. According to Beerthuizen et al. [6], the incidence of CRPS1 after a fracture varies significantly depending on the diagnostic criteria used. Despite these issues, our study is the first to assess the risk factors for developing CRPS1 after extra-articular shoulder surgery, on a relatively large number of patients.
Conclusion
The incidence of CRPS1 after subacromial space surgery is over 13%. A medical history of hypothyroidism, open surgery, and an impact on the pre-operative level of daily physical activity appear to be potential risk factors for developing this post-operative complication. It is, therefore, important to provide accurate information on patients who fit this category during the pre-operative period. With regards to technical parameters, these could also be carefully considered to avoid this specific complication.
Compliance with Ethical Standards
Conflict of interest
The authors declare that they have no conflict of interests.
Ethical standard statement
This article does not contain any studies with human or animal subjects performed by any of the authors.
Informed consent
For this type of study informed consent is not required.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations
References
- 1.Field J. Complex regional pain syndrome: a review. Journal of Hand Surgery (European Volume) 2013;38(6):616–626. doi: 10.1177/1753193412471021. [DOI] [PubMed] [Google Scholar]
- 2.Iolascon G, de Sire A, Moretti A, Gimigliano F. Complex regional pain syndrome (CRPS) type I: historical perspective and critical issues. Clinical Cases in Mineral and Bone Metabolism. 2015;12(Suppl 1):4–10. doi: 10.11138/ccmbm/2015.12.3s.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bruehl S. Complex regional pain syndrome. BMJ. 2015;350:h2730. doi: 10.1136/bmj.h2730. [DOI] [PubMed] [Google Scholar]
- 4.Rewhorn MJ, Leung AH, Gillespie A, Moir JS, Miller R. Incidence of complex regional pain syndrome after foot and ankle surgery. Journal of Foot and Ankle Surgery. 2014;53(3):256–258. doi: 10.1053/j.jfas.2014.01.006. [DOI] [PubMed] [Google Scholar]
- 5.Li Z, Smith BP, Tuohy C, Smith TL, Koman LA. Complex regional pain syndrome after hand surgery. Hand Clinics. 2010;26:281–289. doi: 10.1016/j.hcl.2009.11.001. [DOI] [PubMed] [Google Scholar]
- 6.Beerthuizen A, Stronks DL, Van’t Spijker A, Yaksh A, Hanraets BM, Klein J, et al. Demographic and medical parameters in the development of complex regional pain syndrome type 1 (CRPS1): prospective study on 596 patients with a fracture. Pain. 2012;153:1187–1992. doi: 10.1016/j.pain.2012.01.026. [DOI] [PubMed] [Google Scholar]
- 7.Roh YH, Lee BK, Noh JH, Baek JR, Oh JH, Gong HS, et al. Factors associated with complex regional pain syndrome type I in patients with surgically treated distal radius fracture. Archives of Orthopaedic and Trauma Surgery. 2014;134:1775–1781. doi: 10.1007/s00402-014-2094-5. [DOI] [PubMed] [Google Scholar]
- 8.Anderson DJ, Fallat L. Complex regional pain syndrom of the lower extremity: a retrospective study of 33 patients. The Journal of Foot and Ankle Surgery. 1999;38(6):381–387. doi: 10.1016/S1067-2516(99)80037-4. [DOI] [PubMed] [Google Scholar]
- 9.Abboud JA, Kim JS. The effect of hypercholesterolemia on rotator cuff disease. Clinical Orthopaedics and Related Research. 2010;468:1493–1497. doi: 10.1007/s11999-009-1151-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Koorevaar RCT, Van’t Riet E, Ipskamp M, Bulstra SK. Incidence and prognostic factors for postoperative frozen shoulder after shoulder surgery: a prospective cohort study. Archives of Orthopaedic and Trauma Surgery. 2017;137(3):293–301. doi: 10.1007/s00402-016-2589-3. [DOI] [PubMed] [Google Scholar]
- 11.Brislin KJ, Field LD, Savoie FH. Complications after arthroscopic rotator cuff repair. Arthroscopy: The Journal of Arthroscopic and Related Surgery. 2007;23(2):124–128. doi: 10.1016/j.arthro.2006.09.001. [DOI] [PubMed] [Google Scholar]
- 12.Evans JP, Guyver PM, Smith CD. Frozen shoulder after simple arthroscopic shoulder procedures: What is the risk? The Bone and Joint Journal. 2015;97(B(7)):963–966. doi: 10.1302/0301-620X.97B7.35387. [DOI] [PubMed] [Google Scholar]
- 13.Eljabu W, Klinger HM, von Knoch M. Prognostic factors and therapeutic options for treatment of frozen shoulder: a systematic review. Archives of Orthopaedic and Trauma Surgery. 2016;136(1):1–7. doi: 10.1007/s00402-015-2341-4. [DOI] [PubMed] [Google Scholar]
- 14.Goebel A. Complex regional pain syndrome in adults. Rheumatology (Oxford, England) 2011;50(10):1739–1750. doi: 10.1093/rheumatology/ker202. [DOI] [PubMed] [Google Scholar]
- 15.Brunner F, Nauer M, Bachmann LM. Poor prognostic factors in complex regional pain syndrome 1: a Delphi survey. Journal of Rehabilitation Medicine. 2011;43(9):783–786. doi: 10.2340/16501977-0856. [DOI] [PubMed] [Google Scholar]
- 16.Itoi E, Arce G, Bain GI, Diercks RL, Guttmann D, Imhoff AB, et al. Shoulder stiffness: current concepts and concerns. Arthroscopy: The Journal of Arthroscopic and Related Surgery. 2016;32(7):1402–1414. doi: 10.1016/j.arthro.2016.03.024. [DOI] [PubMed] [Google Scholar]
- 17.Cucchi D, Marmotti A, De Giorgi S, Costa A, D’Apolito R, Conca M, et al. Risk factors for shoulder stiffness: current concepts. Joints. 2017;5(4):217–223. doi: 10.1055/s-0037-1608951. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cakir M, Samanci N, Balci N, Balci MK. Musculoskeletal manifestations in patients with thyroid disease. Clinical Endocrinology (Oxford) 2003;59(2):162–167. doi: 10.1046/j.1365-2265.2003.01786.x. [DOI] [PubMed] [Google Scholar]
- 19.Dilek B, Yemez B, Kizil R, Kartal E, Gulbahar S, Sari O, et al. Anxious personality is a risk factor for developing complex regional pain syndrome type I. Rheumatology International. 2012;32(4):915–920. doi: 10.1007/s00296-010-1714-9. [DOI] [PubMed] [Google Scholar]
- 20.Le HV, Lee SJ, Nazarian A, Rodriguez EK. Adhesive capsulitis of the shoulder: review of pathophysiology and current clinical treatments. Shoulder and Elbow. 2017;9(2):75–84. doi: 10.1177/1758573216676786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Veldman PH, Reynen HM, Arntz IE, Goris RJ. Signs and symptoms of reflex sympathetic dystrophy: prospective study of 829 patients. The Lancet (London, England) 1993;342(8878):1012–1016. doi: 10.1016/0140-6736(93)92877-V. [DOI] [PubMed] [Google Scholar]
- 22.Constant CR, Murley AH. A clinical method of functional assessment of the shoulder. Clinical Orthopaedics and Related Research. 1987;214:160–164. [PubMed] [Google Scholar]
- 23.Constant CR, Gerber C, Emery R, Søjbjerg JO, Gohlke F, Boileau P. A review of the Constant score: modifications and guidelines for its use. Journal of Shoulder and Elbow Surgery. 2008;17(2):355–361. doi: 10.1016/j.jse.2007.06.022. [DOI] [PubMed] [Google Scholar]
- 24.Roy JS, MacDermid JC, Woodhouse LJ. A systematic review of the psychometric properties of the Constant-Murley score. Journal of Shoulder and Elbow Surgery. 2010;19(1):157–164. doi: 10.1016/j.jse.2009.04.008. [DOI] [PubMed] [Google Scholar]
- 25.Harrell FE. Regression modeling strategies: with applications to linear models, logistic regression, and survival analysis. New York City: Springer Science & Business Media; 2013. [Google Scholar]
- 26.Jellad A, Salah S, Frih ZBS. Complex regional pain syndrome type I: incidence and risk factors in patients with fracture of the distal radius. Archives of physical medicine and rehabilitation. 2014;95:487–492. doi: 10.1016/j.apmr.2013.09.012. [DOI] [PubMed] [Google Scholar]
- 27.Reuben SS. Preventing the development of complex regional pain syndrome after surgery. Anesthesiology. 2004;101(5):1215–1224. doi: 10.1097/00000542-200411000-00023. [DOI] [PubMed] [Google Scholar]
- 28.Lichtman DM, Florio RL, Mack GR. Carpal tunnel release under local anesthesia: evaluation of the outpatient procedure. Journal of Hand Surgery. 1979;4(6):544–546. doi: 10.1016/S0363-5023(79)80007-6. [DOI] [PubMed] [Google Scholar]
- 29.MacDonald RI, Lichtman DM, Hanlon JJ, et al. Complications of surgical release for carpal tunnel syndrome. The Journal of Hand Surgery. 1978;3(1):70–76. doi: 10.1016/S0363-5023(78)80118-X. [DOI] [PubMed] [Google Scholar]
- 30.Huang S-W, Lin J-W, Wang W-T, Wu C-W, Liou T-H, Lin H-W. Hyperthyroidism is a risk factor for developing adhesive capsulitis of the shoulder: a nationwide longitudinal population-based study. Scientific Reports. 2014;4:4183. doi: 10.1038/srep04183. [DOI] [PMC free article] [PubMed] [Google Scholar]