Comparison of original and modified Academic Research Consortium for High Bleeding Risk definitions in real-world practice

BACKGROUND: The Academic Research Consortium for High Bleeding Risk (ARC-HBR) defined high bleeding risk (HBR) in patients undergoing percutaneous coronary intervention. We have reported a simplified HBR (S-HBR), excluding six items with prevalences under 1% from ARC-HBR. The Japanese Circulation Society developed an HBR specific to Japanese (J-HBR), adding three items to ARC-HBR in consideration of ethnicity. Data comparing each HBR are scarce. METHODS: Patients treated with second-generation drug-eluting stents between January 2010 and December 2013 were enrolled, in whom all items of ARC-HBR, and the incidences of major bleeding and ischemic events were examined. Also, the diagnostic values of ARC-HBR, S-HBR, and J-HBR at 1 and 7 years post procedure were compared by using receiver-operating characteristic curves. RESULTS: The study sample consisted of 3430 patients. Mean follow-up period was 2299 ± 904 days. The incidence of major bleeding at 1 and 7 years in each definition was as follows: ARC-HBC, 3.3% and 10.6%; S-HBR, 3.3% and 10.7%; and J-HBR, 2.9% and 10.0%. The diagnostic value of J-HBR for major bleeding at 1 year was lower than that of ARC-HBR (C statistics 0.64 vs. 0.68, p < 0.001). Other diagnostic values of S-HBR and J-HBR were comparable to those of ARC-HBR. CONCLUSIONS: S-HBR was as useful as ARC-HBR in predicting both short- and long-term HBR, and J-HBR is useful for predicting long-term HBR.