IntroductionAn increasing number of studies have investigated the effect of exercise on flow-mediated dilation (FMD) in people with type 2 diabetes mellitus (T2DM), while the findings were controversial. The primary aim of this systematic review and meta-analysis was to investigate the effect of exercise on FMD in T2DM patients, and the secondary aim was to investigate the optimal type, frequency, session duration, and weekly time of exercise for T2DM patients.MethodsSearches were conducted in PubMed, Cochrane Library, Scopus, Web of Science, Embase and EBSCO databases. The Cochrane risk of bias tool (RoB2) in randomized trial and Physiotherapy Evidence Database (PEDro) scale were used to assess the methodological quality of the included studies.ResultsFrom the 3636 search records initially retrieved, 13 studies met the inclusion criteria. Our meta-analysis revealed that exercise had a significant effect on improving FMD in T2DM patients [WMD, 2.18 (95% CI, 1.78-2.58), p < 0.00001, I2 = 38%], with high-intensity interval training (HIIT) being the most effective intervention type [HIIT, 2.62 (1.42-3.82); p < 0.0001; aerobic exercise, 2.20 (1.29-3.11), p < 0.00001; resistance exercise, 1.91 (0.01-3.82), p = 0.05; multicomponent training, 1.49 (0.15-2.83), p = 0.03]. In addition, a higher frequency [> 3 times, 3.06 (1.94-4.19), p < 0.00001; ≤ 3 times, 2.02 (1.59-2.45), p < 0.00001], a shorter session duration [< 60 min, 3.39 (2.07-4.71), p < 0.00001; ≥ 60 min, 1.86 (1.32-2.40), p < 0.00001], and a shorter weekly time [≤ 180 min, 2.40 (1.63-3.17), p < 0.00001; > 180 min, 2.11 (0.82-3.40), p = 0.001] were associated with larger improvements in FMD.ConclusionThis meta-analysis provides clinicians with evidence to recommended that T2DM patients participate in exercise, especially HIIT, more than 3 times per week for less than 60 min, with a target of 180 min per week being reached by increasing the frequency of exercise.Systematic review registrationhttps://www.crd.york.ac.uk/prospero, identifier CRD42023466575.