Comparison of physiotherapy, manipulation, and corticosteroid injection for treating shoulder complaints in general practice: randomised, single blind study

Abstract Objective: To compare the efficacy of physiotherapy, manipulation, and corticosteroid injection for treating patients with shoulder complaints in general practice. Design: Randomised, single blind study. Setting: Seven general practices in the Netherlands. Subjects: 198 patients with shoulder complaints, of whom 172 were divided, on the basis of physical examination, into two diagnostic groups: a shoulder girdle group (n = 58) and a synovial group (n = 114). Interventions: Patients in the shoulder girdle group were randomised to manipulation or physiotherapy, and patients in the synovial group were randomised to corticosteroid injection, manipulation, or physiotherapy. Main outcome measures: Duration of shoulder complaints analysed by survival analysis. Results: In the shoulder girdle group duration of complaints was significantly shorter after manipulation compared with physiotherapy (P < 0.001). Also the number of patients reporting treatment failure was less with manipulation. In the synovial group duration of complaints was shortest after corticosteroid injection compared with manipulation and physiotherapy (P < 0.001). Drop out due to treatment failure was low in the injection group (17%) and high in the manipulation group (59%) and physiotherapy group (51%). Conclusions: For treating shoulder girdle disorders, manipulation seems to be the preferred treatment. For the synovial disorders, corticosteroid injection seems the best treatment. Key messages Many patients with shoulder complaints are treated in general practice, but there has been little evaluation of different treatments In this single blind randomised trial we investigated the effect of corticosteroid injection, manipulation, and physiotherapy on the duration of shoulder complaints among patients treated in general practice Patients were divided into two diagnostic groups: those with complaints originating from the synovial structures and those whose complaints originated from the shoulder girdle The first group was treated by injection, manipulation, or physiotherapy: those receiving injection showed quickest recovery and only 17% of patients dropped out because of treatment failure, compared with 51% in physiotherapy group and 59% in manipulation group Patients with a shoulder girdle disorder were treated by manipulation or physiotherapy: duration of complaints was significantly shorter after manipulation, and there was only 20% drop out in this group compared with 45% in physiotherapy group