Validity of a PCI Bleeding Risk Score in patient subsets stratified for body mass index

OBJECTIVE An accurate tool with good discriminative for bleeding would be useful to clinicians for improved management of all their patients. Bleeding risk models have been published but not externally validated in independent clinical data set. We chose the National Cardiovascular Data Registry (NCDR) percutaneous coronary intervention (PCI) score to validate within a large, multisite community data set. The aim of the study was validation of this Bleeding Risk Score (BRS) tool among a subgroup of patients based on body mass index. METHODS This is a large-scale retrospective analysis of a current registry utilising data from a 37-hospital health system. The central repository of patients with coronary heart disease undergoing PCI between 1 June 2009 and 30 June 2012 was utilised to validate the NCDR PCI BRS among 4693 patients. The primary end point was major bleeding. Validation analysis calculating the receiver operating characteristic curve was performed. RESULTS There were 143 (3%) major bleeds. Mean BRS was 14.7 (range 3-42). Incidence of bleeding by risk category: low (0.5%), intermediate (1.7%) and high risk (7.6%). Tool accuracy was poor to fair (area-under-the curve (AUC) 0.78 heparin, 0.65 bivalirudin). Overall accuracy was 0.71 (CI 0.66 to 0.76). Accuracy did not improve when confined to just the intermediate risk group (AUC 0.58; CI 0.55 to 0.67). Tool accuracy was the lowest among the low BMI group (AUC 0.62) though they are at increased risk of bleeding following PCI. CONCLUSIONS Bleeding risk tools have low predictive value even among subgroups of patients at higher risk. Adjustment for anticoagulation use resulted in poor discrimination because bivalirudin differentially biases outcomes toward no bleeding. The current state of bleeding risk tools provide little support for diagnostic utility in regards to major bleeding and therefore have limited clinical applicability.