Management of frozen shoulder: a systematic review and cost-effectiveness analysis

Frozen shoulder is condition in which movement of the shoulder becomes restricted. It can be described as either primary (idiopathic) whereby the aetiology is unknown, or secondary, when it can be attributed to another cause. It is commonly a self-limiting condition, of approximately 1 to 3 years' duration, though incomplete resolution can occur.

To evaluate the clinical effectiveness and cost-effectiveness of treatments for primary frozen shoulder, identify the most appropriate intervention by stage of condition and highlight any gaps in the evidence.

A systematic review was conducted. Nineteen databases and other sources including the Cumulative Index to Nursing and Allied Health (CINAHL), Science Citation Index, BIOSIS Previews and Database of Abstracts of Reviews of Effects (DARE) were searched up to March 2010 and EMBASE and MEDLINE up to January 2011, without language restrictions. MEDLINE, CINAHL and PsycINFO were searched in June 2010 for studies of patients' views about treatment.

Randomised controlled trials (RCTs) evaluating physical therapies, arthrographic distension, steroid injection, sodium hyaluronate injection, manipulation under anaesthesia, capsular release or watchful waiting, alone or in combination were eligible for inclusion. Patients with primary frozen shoulder (with or without diabetes) were included. Quasi-experimental studies were included in the absence of RCTs and case series for manipulation under anaesthesia (MUA) and capsular release only. Full economic evaluations meeting the intervention and population inclusion criteria of the clinical review were included. Two researchers independently screened studies for relevance based on the inclusion criteria. One reviewer extracted data and assessed study quality; this was checked by a second reviewer. The main outcomes of interest were pain, range of movement, function and disability, quality of life and adverse events. The analysis comprised a narrative synthesis and pair-wise meta-analysis. A mixed-treatment comparison (MTC) was also undertaken. An economic decision model was intended, but was found to be implausible because of a lack of available evidence. Resource use was estimated from clinical advisors and combined with quality-adjusted life-years obtained through mapping to present tentative cost-effectiveness results.

Thirty-one clinical effectiveness studies and one economic evaluation were included. The clinical effectiveness studies evaluated steroid injection, sodium hyaluronate, supervised neglect, physical therapy (mainly physiotherapy), acupuncture, MUA, distension and capsular release. Many of the studies identified were at high risk of bias. Because of variation in the interventions and comparators few studies could be pooled in a meta-analysis. Based on single RCTs, and for some outcomes only, short-wave diathermy may be more effective than home exercise. High-grade mobilisation may be more effective than low-grade mobilisation in a population in which most patients have already had treatment. Data from two RCTs showed that there may be benefit from adding a single intra-articular steroid injection to home exercise in patients with frozen shoulder of < 6 months' duration. The same two trials showed that there may be benefit from adding physiotherapy (including mobilisation) to a single steroid injection. Based on a network of nine studies the MTC found that steroid combined with physiotherapy was the only treatment showing a statistically and clinically significant beneficial treatment effect compared with placebo for short-term pain (standardised mean difference -1.58, 95% credible interval -2.96 to -0.42). This analysis was based on only a subset of the evidence, which may explain why the findings are only partly supportive of the main analysis. No studies of patients' views about the treatments were identified. Average costs ranged from £36.16 for unguided steroid injections to £2204 for capsular release. The findings of the mapping suggest a positive relationship between outcome and European Quality of Life-5 Dimensions (EQ-5D) score: a decreasing visual analogue scale score (less pain) was accompanied by an increasing (better) EQ-5D score. The one published economic evaluation suggested that low-grade mobilisation may be more cost-effective than high-grade mobilisation. Our tentative cost-effectiveness analysis suggested that steroid alone may be more cost-effective than steroid plus physiotherapy or physiotherapy alone. These results are very uncertain.

The key limitation was the lack of data available. It was not possible to undertake the planned synthesis exploring the influence of stage of frozen shoulder or the presence of diabetes on treatment effect. As a result of study diversity and poor reporting of outcome data there were few instances where the planned quantitative synthesis was possible or appropriate. Most of the included studies had a small number of participants and may have been underpowered. The lack of available data made the development of a decision-analytic model implausible. We found little evidence on treatment related to stage of condition, treatment pathways, the impact on quality of life, associated resource use and no information on utilities. Without making a number of questionable assumptions modelling was not possible.

There was limited clinical evidence on the effectiveness of treatments for primary frozen shoulder. The economic evidence was so limited that no conclusions can be made about the cost-effectiveness of the different treatments. High-quality primary research is required.