Coronary Revascularization in Patients With Diabetes: A Meta-Analysis of Randomized Controlled Trials and Propensity-Matched Studies.

OBJECTIVE: Patients with diabetes have poorer outcomes with coronary artery disease (CAD) and pose a unique clinical population for revascularization. We performed a pairwise meta-analysis of randomized trials (RCTs) and propensity-matched observational studies (PMS) to compare the clinical outcomes of coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) in patients with diabetes. METHODS: A comprehensive literature search was performed to identify RCT and PMS studies comparing CABG with PCI in patients with diabetes with concurrent CAD. Studies were pooled using the random-effects model to perform a pairwise meta-analysis. Primary outcomes included long-term all-cause mortality, cardiac mortality, myocardial infarction (MI), major adverse cardiac and cerebrovascular events (MACCE), and repeat revascularization. Meta-regression was used to explore the effects of baseline risk factors on primary outcomes with moderate to high heterogeneity. RESULTS: A total of 18 RCTs and 9 PMS with 28,846 patients were included. PCI was associated with increased long-term all-cause mortality (risk ratio [RR] = 1.34, P < 0.001), cardiac mortality (RR = 1.52, P < 0.001), MI (RR = 1.51, P = 0.009), MACCE (RR = 1.65, P < 0.001), and repeat revascularization (RR = 2.48, P < 0.001) compared with CABG. There was no difference in long-term stroke between the 2 groups (RR = 0.95, P = 0.82). At meta-regression, a greater proportion of female patients in studies was associated with a decreased protective benefit for CABG for long-term all-cause mortality but an increased protective benefit for long-term MI and repeat revascularization. CONCLUSIONS: Revascularization of patients with diabetes using CABG is associated with significantly reduced long-term mortality, MI, MACCE, and repeat revascularizations. Future studies exploring the influence of gender on revascularization outcomes are necessary to elucidate the ideal treatment modality in patients with diabetes.