Reduction in adiposity is considered a cornerstone in the treatment of obstructive sleep apnea (OSA). Exercise training is one of the interventions used in the treatment of OSA. However, it remains unclear whether exercise training alleviates OSA by improving body composition. This study aimed to verify the effect of exercise training on body composition and the apnea-hypopnea index (AHI) in patients with OSA. This study adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and was registered in PROSPERO. PubMed/Medline, Scopus, and Cochrane databases were used for systematic review. Analyses were performed using Review Manager software (version 5.3; R Foundation for Statistical Computing, Vienna, Austria). Exercise training significantly reduced body mass index (mean difference = -0.55 kg/m(2); 95% CI = -0.95 to -0.15), body fat percentage (mean difference = -1.19%; 95% CI = -2.03 to -0.35), neck circumference (mean difference = -0.55 cm; 95% CI = -1.08 to-0.02), total body weight (mean difference = -2.13 kg; 95% CI = -4.30 to 0.04), and AHI-8.06 events/hour; 95% CI = -10.47 to-5.66). Although body composition was improved by exercise, this improvement was clinically small. Moreover, improvement in AHI by exercise may not be entirely due to improvements in adiposity. (C) 2021 Elsevier B.V. All rights reserved.