Clinical utility of the academic research consortium new proposed criteria for high bleeding risk definition in patients with acute coronary syndromes

Background The Academic Research Consortium High Bleeding Risk (ARC-HBR) criteria have been proposed to stratify the bleeding risk of patients undergoing percutaneous coronary intervention (PCI). While most criteria were established, 4 criteria have been proposed on a de novo basis. Purpose We assessed the prevalence and prognosis of new ARC-HBR criteria in a contemporary, prospective, multicenter, quality-improvement registry of all-comers patients with acute coronary syndromes. Methods Between 2016 and 2020, consecutive subjects were enrolled; baseline characteristics and medications were prospectively collected, and patients were followed-up at 1 year. All clinical events (including bleeding) were adjudicated by an independent committee. All 17 ARC-HBR criteria were individually evaluated by reviewing patients' charts. Results Of the 2804 patients enrolled, 782 (28.0%) met the ARC-HBR definition and 47 (6%) of them experienced a major BARC 3 or 5) bleeding at 1-year. HBR patients had a significantly higher risk of BARC 3–5 bleedings (HR for: 3.07; 95% CI: 2.02–4.67; p<0.0001; Fig. 1A), BARC 2–5 (HR: 1.845; 95% CI: 1.4–2.42; p<0.0001). Fig. 1B indicates the proportion of patients meeting each criterion. Age, (moderate or severe) chronic kidney disease, (moderate or severe) anemia and oral anticoagulant therapy included 88% of HBR patients. The 4 new ARC-HBR criteria, all together, were present in only 1.7% of our population: 1.0% was planned for major surgery while on dual antiplatelet therapy, 0.5% had a recent intracranial hemorrhage/ictus or brain arteriovenous malformations, 0.1% had hepatic cirrhosis with portal hypertension and 0.1% had a recent surgery or trauma. In a multivariable Cox regression analysis including individual ARC-HBR criteria, only CKD (major and minor criteria), anemia (major and minor criteria) and cancer were the independent predictors of BARC 3–5 events with a concordance-index for this model of 0.698 (p<0.001). In a second model including only CKD (major criterion), anemia (major criterion), age and oral anticoagulation therapy, all these criteria were independent predictors of BARC 3–5 events with a concordance index of 0.674 (pmodel<0.001 for the model) (Fig. 2). Conclusion Almost one third of contemporary ACS patients was at HBR according to the ARC-HBR definition and these patients presented a significantly higher risk of bleedings at 1-year. The most common 4 criteria (age, CKD, anemia, and oral anticoagulant therapy) allowed the identification of 88% of HBR patients. The newly proposed HBR criteria were extremely rare and therefore challenging to validate and of uncertain clinical utility. These data may inform and simplify clinical decision making and provide priority for future directions of HBR definitions. Funding Acknowledgement Type of funding sources: None. Figure 1Figure 2