Background To evaluate the effects of bariatric surgery on patients with type 2 diabetes through comparing Asian and non-Asian populations according to follow-up time. Methods In this meta-analysis of randomized controlled trials (RCTs), we searched PubMed databases for relevant articles. A meta-analysis was undertaken of 37 eligible RCTs. Results Diabetes remission rates among Asian populations were similar to those of non-Asian populations at 1 year (95% confidence interval (CI) 0.519 [0.417, 0.619] vs 0.571 [0.316, 0.794], respectively) but were found to be higher in Asian populations at 2 years compared to non-Asian populations (95% CI 0.672 [0.469, 0.826] vs 0.563 [0.220, 0.855], respectively). Asian populations showed greater reductions in HbA1c (- 3.512 [- 2.726, - 4.299], - 3.001 [- 1.433, - 4.569], and - 3.345 [- 2.267, - 4.423]) than non-Asian populations (- 2.129 [- 1.821, - 2.438], - 2.301 [- 1.881, - 2.722], and - 2.107 [- 1.941, - 2.273]) at 1, 2, and 5 years, respectively. Asian populations also showed greater reductions of fasting plasma glucose than non-Asian populations, and these effects were prominent at 3- and 5-year follow-up (Asian populations, - 83.297 [- 73.264, - 93.331]; non-Asian populations, - 54.288 [- 37.409, - 71.166] at 3 years; Asian populations, - 99.014 [- 82.239, - 115.788]; non-Asian populations, - 45.756 [- 21.892, - 69.621] at 5 years). Conclusions Bariatric surgery is more successful in diabetes remission and glucose outcome improvement in Asian populations with type 2 diabetes than in non-Asian populations, even over long-term follow-up periods. Bariatric surgery can be considered an effective treatment option and may present an appropriate opportunity to improve the prognosis for Asian patients with type 2 diabetes.