Background: Shoulder subluxation and pain are common complications of stroke that impair upper limb function. Objectives: This study conducted a systematic review and network meta-analysis to compare multiple therapeutic interventions for post-stroke shoulder subluxation, establishing an evidence-based hierarchy of treatment efficacy to optimize rehabilitation strategies and guide clinical practice. Methods: A comprehensive search was conducted using the MEDLINE, EMBASE, Cochrane, Scopus, and Web of Science databases until 8 August 2025. Randomized controlled trials evaluating treatments for shoulder subluxation, including neuromuscular electrical stimulation (NMES), Kinesio taping, corticosteroid injections, slings, repetitive peripheral magnetic stimulation, and electroacupuncture, were included. The follow-up duration in the included trials ranged from 1 to 12 weeks. Effect sizes were calculated using standardized mean differences with a random-effects model, and treatment rankings were determined using surface under the cumulative ranking curve (SUCRA). Results: Thirteen studies including 402 patients were analyzed. NMES was the most effective intervention for reducing subluxation distance (SUCRA: 84.9), while corticosteroid injections provided the greatest pain relief at rest (SUCRA: 73.6). Kinesio taping was most effective for functional recovery, as measured by the Fugl–Meyer Assessment (SUCRA: 98.5), and for pain relief during activity (SUCRA: 87.7). Conclusions: Our network meta-analysis suggests that different interventions are optimal for specific aspects of post-stroke shoulder impairment. NMES most effectively reduces subluxation distance, whereas corticosteroid injections are most effective for alleviating pain at rest. Kinesio taping appears superior for enhancing functional recovery and reducing pain during movement. These findings, based on short-term follow-up durations (1–12 weeks), provide an evidence-based ranking of interventions to support multimodal rehabilitation and inform clinical decision-making. The observed heterogeneity across studies underscores the need for standardized treatment protocols and rigorous long-term investigations.
Park et al. (2025) studied this question.