Objective. To perform a Cochrane Systematic Review of benefits and harms of botulinum toxin for shoulder pain. Methods. We included clinical trials of adults with shoulder pain (population), comparing botulinum toxin (intervention) to placebo or other therapies (comparison), and reporting benefits or harms (outcomes). We calculated relative risk (RR) for categorical outcomes and mean differences (MD) for continuous outcomes. Results. Six randomized controlled trials (RCT) with 164 patients all comparing single botulinum toxin type A injections to placebo were included. Five RCT in patients with post-stroke shoulder pain found that an intramuscular injection of botulinum toxin type A significantly reduced pain at 3-6 months (MD -1.2 points on 0-10 scale, 95% CI -2.4 to -0.07) and improved shoulder external rotation at 1 month (MD 9.8°, 95% CI 0.2° to 19.4°). Number of adverse events did not differ between groups (RR 1.46, 95% CI 0.6 to 24.3). One RCT in arthritis-related shoulder pain showed that single intraarticular botulinum toxin type A injection reduced pain (MD -2.0 on 0-10 scale, 95% CI -3.7 to -0.3) and shoulder disability (MD -13.4 on 0-100 scale, 95% CI -24.9 to -1.9) and improved shoulder abduction (MD 13.8°, 95% CI 3.2° to 44.0°) at 1 month, compared with placebo. Serious adverse events did not differ between groups (RR 0.35, 95% CI 0.11, 1.12). Conclusion. With evidence from few studies with small sample sizes and medium to high risk of bias, botulinum toxin type A injections decreased pain and improved shoulder function in patients with chronic shoulder pain due to spastic hemiplegia or arthritis. The Journal of Rheumatology Copyright © 2011. All rights reserved.