Nationwide observational study of antithrombotic treatment strategies after coronary artery bypass graft surgery in patients with atrial fibrillation

Abstract

Abstract Background Atrial fibrillation (AF) is common following coronary artery bypass graft surgery (CABG). Antiplatelet therapy (APT) is recommended after CABG to reduce graft failure. In the case of concomitant AF, this recommendation may instead include oral anticoagulation (OAC) guided by the CHA2DS2-VASc score and bleeding risk. Guidelines remain unclear on optimal antithrombotic treatment in AF patients after CABG. Purpose To assess comparative effectiveness and safety of different antithrombotic treatment strategies in AF patients after CABG. Methods In this nationwide observational study, we identified patients with AF in relation to isolated CABG surgery (before or ≤3 months after CABG) from January 2006 to January 2022 using national administrative registries. Patients alive and claiming a minimum of 1 antithrombotic prescription during the first 3 months after CABG were categorised in 1) APT alone, 2) OAC alone, and 3) OAC+APT. Incidence rates per 100 person-years were calculated for the effectiveness endpoint; a composite of all-cause mortality, stroke, and myocardial infarction, and its individual components, and for the safety endpoint; a composite of bleeding leading to hospitalisation. We also assessed the comparative effectiveness including its individual components and safety of APT and OAC+APT compared with OAC 3 to 12 months after CABG by multivariable Cox proportional-hazards models. Results We identified 33,763 patients who underwent CABG, of whom 6,008 (17.8%) patients had AF, were alive, and collected at least 1 antithrombotic drug ≤3 months after CABG. Temporal trends in the antithrombotic treatment strategies are illustrated in Figure 1. The mean age was 71 years, 18% were female, and 87% had a CHA2DS2-VASc score ≥2 for men and ≥3 for women. A total of 2,845 (47.3%) patients were treated with OAC+APT, 1,958 (32.6%) with APT alone, and 1,205 (20.1%) with OAC alone. Patients treated with APT more often had had a myocardial infarction and 87% had a CHA2DS2-VASc score ≥2 for men and ≥3 for women, whereas patients with OAC+APT more often had valve disease compared with OAC. The incidence rates per 100 patient-years (IR) of the effectiveness endpoint were 4.5 in patients with OAC+ACT, 4.8 with APT, and 4.9 with OAC. For the safety endpoint, the IR were 3.6 with OAC+ACT, 2.1 with APT, and 2.0 with OAC. No differences were found between treatment strategies for the effectiveness endpoint. Compared with patients with OAC, mortality was lower with OAC+APT (HR 0.66 [95% CI 0.45-0.98]) and APT (HR 0.53 [95% 0.34-0.84]), and bleeding risk was higher with OAC+APT (HR 1.69 [95% CI 1.08-2.64]) (Figure 2). Conclusion(s) Every third patient with AF were not treated with OAC after CABG. Compared with OAC alone, mortality was reduced in patients with OAC+APT and APT alone, and bleeding risk was higher in patients with OAC+APT. Future studies are warranted on optimal antithrombotic regimens in AF patients after CABG.Temporal trends in antithromboticsForest plot