Variation in preoperative antithrombotic strategy, severe bleeding, and use of blood products in coronary artery bypass grafting: results from the multicentre E-CABG registry.

Aims: No data exists on inter-institutional differences in terms of adherence to international guidelines regarding the discontinuation of antithrombotics and rates of severe bleeding in coronary artery bypass grafting (CABG). Methods and results: This is an analysis of 7118 patients from the prospective multicentre European CABG (E-CABG) registry who underwent isolated CABG in 15 European centres. Preoperative pause of P2Y12 receptor antagonists shorter than that suggested by the 2017 ESC guidelines (overall 11.6%) ranged from 0.7% to 24.8% between centres (adjusted P < 0.0001) and increased the rate of severe-massive bleeding [E-CABG bleeding grades 2-3, OR 1.66, 95% confidence interval (CI) 1.27-2.17; Universal Definition of Perioperative Bleeding (UDPB) bleeding grades 3-4, OR 1.50, 95% CI 1.16-1.93]. The incidence of resternotomy for bleeding (overall 2.6%) ranged from 0% to 6.9% (adjusted P < 0.0001), and surgical site bleeding (overall 59.6%) ranged from 0% to 84.6% (adjusted P = 0.003). The rate of the UDPB bleeding grades 3-4 (overall 8.4%) ranged from 3.7% to 22.3% (P < 0.0001), and of the E-CABG bleeding grades 2-3 (overall 6.5%) ranged from 0.4% to 16.4% between centres (P < 0.0001). Resternotomy for bleeding (adjusted OR 5.04, 95% CI 2.85-8.92), UDPB bleeding grades 3-4 (adjusted OR 6.61, 95% CI 4.42-9.88), and E-CABG bleeding grades 2-3 (adjusted OR 8.71, 95% CI 5.76-13.15) were associated with an increased risk of hospital/30-day mortality. Conclusions: Adherence to the current guidelines on the early discontinuation of P2Y12 receptor antagonists is of utmost importance to reduce excessive bleeding and early mortality after CABG. Inter-institutional variation should be considered for a correct interpretation of the results in multicentre studies evaluating perioperative bleeding and use of blood products.