Comparisons of Non-Surgical Interventions for Frozen Shoulder: A Systematic Review and Network Meta-Analysis

Background: Frozen shoulder is a common musculoskeletal disorder, characterised by pain and movement restriction. Non-surgical approaches are the mainstay therapy for frozen shoulder. Our aim was to compare the effect of different non-surgical interventions, and to identify potential moderating variables of treatment effect. Methods: The following databases were searched, from their inception to October 18, 2018: PubMed, EMBASE, Cochrane Library, and Web of Science. The search was supplemented by manual review of relevant reference lists. Randomised controlled trials (RCTs) of non-surgical interventions, with a clear comparison group and measured outcomes, were selected. All records were screened by two independent reviewers, and the quality of studies assessed. Study characteristics and outcome data were extracted for pairwise and network meta-analyses. Measured outcomes included pain, shoulder function in activities of daily living, and range of motion (ROM). The mean change in value from baseline was the primary outcome measure, with the pooled standardised mean difference (SMD), and its 95% credible interval (CrI), used to compare outcomes between two groups. Findings: Of 2953 records identified, 92 RCTs were eligible, evaluating 32 non-surgical interventions, in 5946 patients. Of these, intraarticular injection improved pain (steroid injection, SMD 1·68 [95% CrI, 1·03-2·34]; capsular distension, SMD 2·68 [1·32-4·05]), and shoulder function (steroid injection: SMD 2·16 [1·52-2·81]; distension: SMD 2·89 [1·71-4·06]) to a greater extent than either a placebo or conventional physical therapy. Subgroup analyses revealed steroid injection combined with conventional physical therapy yielding improvement during the painful freezing phase, with joint manipulation yielding more benefits in the adhesive phase. Disease stage, female sex, and diabetes were identified as moderating factors of treatment outcome. Interpretation: Intra-articular injection is the prevailing non-surgical treatment for frozen shoulder. Factors (disease stage, sex, and diabetes) should be considered for individualised optimal options. Funding Statement: This work was funded by the grants from National Natural Science Foundation of China (No.81902278), and Zhejiang Provincial Natural Science Foundation (No. LY15H170002) Declaration of Interests: The authors declare no competing interests. Ethics Approval Statement: This systematic review was developed according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) extension statement, incorporating network meta-analyses. The study protocol has been submitted to PROSPERO (registration number CRD42018112971).