Exercise therapy after ultrasound-guided corticosteroid injections in patients with subacromial pain syndrome: a randomized controlled trial

Karen Ellegaard, Robin Christensen, Sara Rosager, Cecilie Bartholdy, Søren Torp-Pedersen, Thomas Bandholm, Bente Danneskiold-Samsøe, Henning Bliddal, Marius Henriksen

Subacromial pain syndrome (SAPS) accounts for around 50 % of all cases of shoulder pain. The most commonly used treatments are glucocorticosteroid (steroid) injections and exercise therapy; however, despite treatment SAPS patients often experience relapse of their symptoms. Therefore the clinical effect of combining steroid and exercise therapy is highly relevant to clarify. The aim of this randomized controlled trial was to investigate if exercise therapy added to steroid injection in patients with SAPS will improve the effect of the injection therapy on shoulder pain. In this two-arm randomized trial running over 26 weeks, patients with unilateral shoulder pain (> 4 weeks) and thickened subacromial bursa (> 2 mm on US) were included. At baseline all participants received two steroid injections into the painful shoulder with an interval of one week. Subsequently they were randomized (1:1) to either 10 weeks exercise of the involved shoulder (intervention group) or exercise of the uninvolved shoulder (control group). The patients were re-examined after the exercise program (at week 13) and again at week 26. The primary outcome assessed after 26 weeks was change in shoulder pain analyzed using the intention-to-treat principle (non-responder imputation). Ninety-nine SAPS patients (58 female) participated (49 intervention/50 control). At both follow up visits (week 13 and 26) no statistically significant between-group differences in pain changes on a visual analog scale (mm) were seen (13 weeks: pain at rest 1.7 (95 % CI -3.6 to 7.0; P = 0.53); pain in activity 2.2 (95 % CI -6.5 to 10.9; P = 0.61), 26 weeks: rest 5.6 (95 % CI -0.9 to 12.1; P = 0.09); activity 2.2 (95 % CI -6.8 to 11.2; P = 0.62). The reduction in pain was most evident in the control group at all four pain measurements. The only difference between groups was seen by US examination at week 13, where fewer participants with impingement were observed in the intervention group compared with the controls (9 vs. 19 participants; P = 0.03). Exercise therapy in the painful shoulder in SAPS patients did not improve the effectiveness of steroid injections for shoulder pain in patients with unilateral SAPS and enlarged subacromial bursa on US examination. ClinicalTrials.gov ( NCT01506804 ). Registration date 5 May 2011.