Abstract 12373: Differences in Predictive Ability of HAS-BLED and HEMORR2HAGES in Patients With Atrial Fibrillation According to Age
Introduction/Hypothesis: In stroke prevention of patients with atrial fibrillation (AF), major bleeding is the main concern which can impact decision for thromboembolic prophylaxis, especially among the elderly. HAS-BLED and HEMORR2HAGES are scores created to evaluate the bleeding risk. The objective of this study was to estimate and compare predictive ability of these scores in elderly (≥80 years) and non-elderly (<80 years) patients.
Methods: This French cohort study was based on the national hospitalization database (PMSI) covering the entire French population. Data from all patients with an AF diagnosis in 2012 were extracted. Demographic and comorbidity data were used to calculate the bleeding risk scores for each patient. Patients hospitalized with a principal diagnosis of major bleeding were identified. Odd ratios were calculated and predictive ability of the scores was compared using the c-statistic with 95% CI (i.e. HAS-BLED: 0-2 vs. ≥3 and HEMORR2HAGES: 0-3 vs. ≥4).
Results: 533,044 AF patients were identified, including 53.2% elderly patients. Bleeding hospitalization involved 7,013 patients (1.3%) among whom 1,785 had intracranial hemorrhages. Bleeding occurred more commonly in patients with higher HAS-BLED and HEMORR2HAGES scores (cf. Figure1). In elderly patients, the c-statistic was not significantly different between HAS-BLED and HEMORR2HAGES: 0.54 (0.528-0.543) vs. 0.53 (0.526-0.539) (p=0.27). In non-elderly patients, HAS-BLED had significantly higher c-statistic than HEMORR2HAGES: 0.59 (0.578-0.598) vs. 0.56 (0.553-0.570) (p<0.0001).
Conclusions: In a large National survey, the predictive value of HAS-BLED score appeared better in non-elderly AF patients. HAS-BLED and HEMORR2HAGES bleeding risk scores performed equally for bleeding events among elderly patients. Given its simplicity, the HAS-BLED score may be an attractive method for the estimation of oral anticoagulant-related bleeding risk in clinical practice.
Author Disclosures: L. Fauchier: Honoraria; Modest; Bristol-Myers Squibb / Pfizer. A. Samson: Employment; Significant; Bristol-Myers Squibb. G. Chaize: Consultant/Advisory Board; Modest; Bristol-Myers Squibb / Pfizer. A. Gaudin: Employment; Significant; Bristol-Myers Squibb. A. Vainchtock: Consultant/Advisory Board; Modest; Bristol-Myers Squibb / Pfizer. C. Bailly: Employment; Significant; Bristol-Myers Squibb. F. Cotté: Employment; Significant; Bristol-Myers Squibb.
- © 2014 by American Heart Association, Inc.