Effects of seven types of exercise in the treatment of rotator cuff-related shoulder pain (RCRSP): a systematic review and Bayesian network meta-analysis.

backgroundRotator cuff-related shoulder pain (RCRSP) is the most common cause of shoulder pain and represents one of the most common musculoskeletal conditions affecting the shoulder. Despite ample evidence supporting the effectiveness of exercise therapy for RCRSP, no consensus exists regarding the optimal exercise intervention. This systematic review aims to evaluate and compare the efficacy of specific versus conventional exercise interventions for improving pain and dysfunction in patients with RCRSP.

methodsThe MEDLINE, Embase, Cochrane Library, CINAHL, SPORTiscus, and Web of Science databases were searched up to June 2025. Randomized controlled trials that diagnosed with RCRSP or a synonymous condition (e.g., rotator cuff tendinopathy, subacromial impingement, subacromial pain syndrome, subacromial bursitis, long-head biceps tendinopathy, or partial-thickness rotator cuff tear) were included, comparing at least two different types of exercise interventions and reporting on the effects associated with pain, dysfunction, related self-reported outcome metrics. A Bayesian network meta-analysis (NMA) was performed to compare seven exercise interventions, with treatments ranked using the surface under the cumulative ranking curve (SUCRA).

resultsThe Shoulder Pain Score Analysis Network contained 15 studies (n = 913; 45% female); the Shoulder Dysfunction Index Analysis Network contained 16 studies (n = 947; 47% female). In terms of shoulder dysfunction improvement, concentric strengthening training (CST) demonstrated a significant advantage in most comparisons and was identified as the most effective exercise intervention program for the treatment of RCRSP, eccentric strengthening training (ECT) and traditional training (TT) were moderately effective, and significantly superior to motor control exercise (MCE) and scapula-focused training (SFT). HLT, which had negative effect sizes in most comparisons, appeared to be the worst exercise intervention program. In terms of shoulder pain improvement, CST also has the highest probability of being ranked first, but comparisons between the other types of interventions were largely non-significant.

conclusionCST performed best in improving both shoulder pain and shoulder dysfunction. When patients with rotator cuff-related shoulder pain are not suitable for CST, ECT and MCE can be used as alternatives for the treatment of RCRSP.

trial registrationPROSPERO Registration No. CRD420251022110.