Pain and stiffness of the shoulder are common complaints in patients who visit family practices, but there is little consensus about the optimal therapy for this condition. The two principal treatments are local injection of a corticosteroid or referral for physical therapy. Van der Windt and colleagues conducted a randomized, controlled trial to compare the effectiveness of these two approaches.

Adults who presented to one of 60 Dutch general practices because of pain and restriction of movement in one shoulder were included in the study. Exclusion criteria included concurrent serious medical conditions and any history of surgery, significant trauma, recent injection or physical therapy to the shoulder. All patients were assessed by one physical therapist to confirm that they met the criteria for the study. Of the 109 patients enrolled in the study, 56 were randomly assigned to receive physical therapy and 53 were assigned to undergo treatment with corticosteroid injections.

Physical therapy consisted of 12 sessions, each lasting 30 minutes and including passive joint mobilization and exercise. Treatment could be adjusted depending on symptoms, and use of ice, hot packs or electrotherapy was permitted for relief of pain. Triamcinolone acetonide, in a dosage of 40 mg, was injected intra-articularly by the posterior route using standardized techniques. Patients in the injection group could receive up to three injections during the study period. Patients could continue medications that they were taking before the intervention started and could use analgesics for severe pain during the study. Patients were assessed at three, seven, 13, 26 and 52 weeks by an observer who was blinded to the treatment group assignment. The main outcome measures were self-report of pain and dysfunction plus clinical assessment, including measurement of mobility restriction using a digital inclinometer.

Treatment was assessed as successful at seven weeks in 40 patients treated with injections (77 percent) compared with 26 of those receiving physical therapy (46 percent). All outcome measures were significantly better in patients undergoing injection than in those undergoing physical therapy. By 26 and 52 weeks, only small, nonsignificant differences existed between the two groups. Approximately one half of all patients in each group reported adverse effects, especially pain, following treatment, but these were mild.

The authors conclude that injection with a corticosteroid provided superior treatment in patients with painful stiff shoulder syndrome.