The effectiveness of corticosteroid injections compared with physiotherapeutic interventions for adhesive capsulitis:...

To determine the effectiveness of corticosteroid injections compared with physiotherapeutic interventions for the treatment of adhesive capsulitis/frozen shoulder.

The electronic databases MEDLINE, EMBASE, CINAHL and AMED were searched up to Week 23 2009. Additional database searching included the Cochrane Central Register of Controlled Clinical Trials, the Physiotherapy Evidence Database (PEDro), the National Research Register and the National Recognition Information Centre, also up to Week 23 2009.

All English-language studies were eligible for inclusion if they showed evidence of random allocation of subjects to either a corticosteroid injection group or a physiotherapeutic intervention group. Studies were considered for inclusion if participants had a stated diagnosis of adhesive capsulitis/frozen shoulder or restriction of passive or active movement in two or more planes. The primary outcomes of interest were pain, external rotation and shoulder disability/function. Quality assessment was assessed using the PEDro scale. Standardised mean differences and effect estimates were calculated for the outcomes of pain, external rotation and shoulder disability at various follow-up periods.

Six studies were deemed eligible for inclusion in the final review. All had evidence of random allocation to either an injection group or a physiotherapeutic intervention group. There were some differences between the studies with regard to both the corticosteroid injections and physiotherapeutic interventions. Standardised mean differences and effect estimates were calculated for three of the included studies at various follow-up periods. There was a medium effect for corticosteroid injections compared with physiotherapeutic interventions for the outcomes of pain, passive external rotation and shoulder disability at 6 weeks. There was only a small effect in favour of corticosteroid injections for pain, passive external rotation and shoulder disability at 12 to 16 weeks and 26 weeks, and pain and shoulder disability at 52 weeks.

The results of this review suggest that corticosteroid injections have greater effect in the short term compared with physiotherapeutic interventions. This decreased over time, with only a small effect in favour of injections in the longer term. The results of this review must be interpreted with some caution due to the limited number of studies and differences in the interventions between the studies.