Background: Dipeptidyl peptidase-4 (DPP-4) inhibitors and glucagon-like peptide-1 (GLP-1) agonists, both of which have an incretin effect, represent novel, effective therapeutic approaches to treating hyperglycemia in type 2 diabetes mellitus (T2DM). Their impact on cardiovascular (CV) outcomes, however, is still under investigation. Hypotheses: We assessed the impact of DPP-4 inhibitors and GLP-1 agonists on the risk of a composite of cardiovascular death, myocardial infarction, and stroke (MACE), and on all-cause mortality (ACM) in T2DM. Methods: We searched Medline, EMBASE, and Cochrane Library for randomized controlled trials (RCTs) published up to February 2010 that compared an incretin agent versus usual care, placebo, or other active agents for patients with T2DM, and reported a pre-specified CV outcome. We also searched a major trial registry, FDA and EMEA databases, and contacted authors about unclear CV outcome information. We undertook Der-Simonian and Laird random effect meta-analyses of MACE and all-cause mortality, using inverse-variance weights and continuity correction of 0.005. We excluded total-zero-event trials. We conducted subgroup analyses using a small number of pre-specified hypotheses, including study duration, type of control, treatment mode, and risk of bias. Results: Of 61 eligible trials, data on MACE was available in 33 trials of DPP-4 inhibitors (107 events /24341 patients), and 9 of GLP-1 agonists (12/3745), and ACM in 20 trials of DPP-4 inhibitors (45/17317), and 3 of GLP-1 (3/1404). There was a significant reduction in MACE (RR 0.53, 95% CI 0.33 to 0.85) between DPP-4 inhibitors and control, and a non-significant reduction in ACM (RR 0.46, 95% CI 0.19 to 1.07). No significant differences were found between GLP-1 agonists and control in MACE (RR 0.77, 95% CI 0.13 to 4.61) or ACM (RR 3.53,0 to >999). There was no significant heterogeneity detected in all analyses (p>0.1). Subgroup analyses did not show any significant differences in treatment effect, except that the effect on all-cause mortality differs by treatment mode (monotherapy vs. combination/add-on therapy, interaction p=0.04). Conclusion: DPP-4 inhibitors may reduce the risk of MACE and ACM. The effect of GLP-1 agonists on CV outcomes remains uncertain.