Time- and risk-dependent effects of dual antiplatelet therapy targeting the P2Y12 receptor after CABG: A network meta-analysis of graft permeability and clinical outcomes.

Abstract

The optimal anticoagulation strategy following Coronary artery bypass grafting remains undetermined, particularly regarding the balance between graft patency, ischemic events, and bleeding across different follow-up periods and risk profiles. A systematic search was conducted in PubMed, Embase, Web of Science, and the Cochrane Database to identify randomized controlled trials comparing regimens of aspirin, DAPT (clopidogrel or ticagrelor), and rivaroxaban. The primary angiographic endpoints were saphenous vein graft(SVG) patency and the left internal mammary artery(LIMA) occlusion; clinical endpoints included myocardial infarction, major adverse cardiovascular events(MACE), major bleeding, and all-cause mortality. A random-effects network meta-analysis was conducted, reporting odds ratios (OR) and 95% CIs, with pre-specified 1-year follow-up stratification and multiple subgroup analyses.The primary analysis included 17 RCTs with 1-year follow-up (9,831 patients). SVG occlusion: Aspirin plus ticagrelor (OR = 0.41,95% CI:0.30-0.56) and aspirin plus clopidogrel (OR = 0.59, 95% CI: 0.44-0.78) were both significantly superior to aspirin monotherapy. MACE: Compared with aspirin monotherapy, no anticoagulation strategy showed a statistically significant difference. Major bleeding: Aspirin plus ticagrelor (OR = 2.53,95% CI:1.59-4.04) and rivaroxaban monotherapy (OR = 2.33, 95% CI: 1.01-5.37) significantly increased the risk. No strategy showed statistically significant differences in myocardial infarction or all-cause mortality. No clear conclusions could be drawn regarding LIMA occlusion due to the rarity of events. Subgroup analysis: In the subgroups of off-pump CABG (OR = 0.42, 95% CI: 0.19-0.90) and early initiation ≤ 48 h (OR = 0.47, 95% CI: 0.29-0.75), the combination of aspirin and ticagrelor significantly reduced MACE; however, the risk of bleeding increased with this regimen when initiation was delayed (> 48 h) (OR = 2.51, 95% CI: 1.52-4.16). Sensitivity analyses showed that the conclusions regarding SVG occlusion and major bleeding were highly robust (the direction and significance of ORs remained consistent across all sensitivity scenarios), whereas estimates for MACE and all-cause mortality were sensitive to specific studies. The net benefit of post-CABG antithrombotic therapy is time- and patient-specific. DAPT (particularly aspirin plus clopidogrel) offers the optimal balance between revascularization protection and bleeding risk; the ticagrelor regimen provides stronger ischemic protection but comes at a higher cost of bleeding; the rivaroxaban regimen is not suitable for routine use. An individualized antithrombotic strategy based on risk stratification is recommended.