Development and Validation of Prognostic Models for Bleeding and Ischemia in Elderly Patients With Comorbid Acute Coronary Syndrome and Atrial Fibrillation

Le résumé fourni par la source

Background Acute coronary syndrome and atrial fibrillation are common cardiovascular diseases in elderly individuals. Patients with comorbidities face increased risks of bleeding and ischemia; however, there is a lack of prognostic models for quantifying these risks in this special population. Methods and Results In this retrospective cohort study, 1851 patients (≥65 years old) with acute coronary syndrome and atrial fibrillation from 2 hospitals in China were included in the development cohort (1252 individuals) and 2 external validation cohorts (284 and 315 individuals). During 1‐year follow‐up, 96 Bleeding Academic Research Consortium type 3 or 5 bleeding events and 245 thromboembolic events were observed. In the development cohort, the concordance indexes for bleeding at 3, 6, and 12 mo ranged from 0.737 to 0.845 and for ischemia ranged from 0.723 to 0.777. The calibration curve and decision curve analysis indicated adequate calibration and clinical practicability. The concordance indexes varied from 0.679 to 0.809 in the validation cohorts. Subgroup analyses focusing on anticoagulant drugs and antithrombotic therapy were conducted, revealing similar discrimination and calibration. Kaplan–Meier curves demonstrated significant differences (log‐rank P <0.001). Additionally, the models outperformed conventional models in concordance indexes, integrated discrimination improvement, and net reclassification improvement. Conclusions Our study provides 2 robust prognostic models with easily available clinical factors for predicting bleeding and ischemia in elderly patients with acute coronary syndrome and atrial fibrillation. Furthermore, we provide online calculators to facilitate individualized risk evaluation and clinical decision‐making. Registration URL: www.chictr.org.cn/ . Unique Identifier: ChiCTR2200067185.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.