004 Predicting early and late mortality in patients presenting for STEMI with CHADS2 score

Abstract

Based on age and simple history data of patients, the CHADS2 score provides accurate assessment of the 1 year risk of stroke in the setting of atrial fibrillation. We sought to assess the hypothesis that the CHADS2 score may be as useful in the setting of STEMI, in comparison with more complex validated scores. We assessed CHADS2, TIMI and GRACE risk scores, in 697 consecutive patients admitted for STEMI within 24 hours after the onset of the symptom. In-hospital, 30-days, 6-months and 1-year mortality rates were assessed according to the CHADS score (0 to 6). Low, intermediate and high risk patients were identified by different scores : <2, 2-3, >3 for CHADS; <4, 4-6, >6 for TIMI; <126, 126-154, > 154 for GRACE respectively. The cumulative in-hospital, 30-day, 6-month and 1-year mortality rates were 5.3%, 6.3%, 7.8% and 8.8% respectively. CHADS2 score was significantly associated with all mortality rates (p<0.0001) with ORs of 2.2 (95%CI 1.7-2.8), 2.1 (95%CI 1.6-2.7), 2 (95%CI 1.6-2.7), and 2 (95%CI 1.6-2.6) per score point respectively. The concordance between the scores in predicting 1-year mortality was fair as shown by kappa values of 0.53 and 0.50 for CHADS2 versus the 2 other and 0.53 between GRACE and TIMI scores (Figure). CHADS2 score is highly correlated to early and late mortality after STEMI. The risk prediction by this simple purely clinical and available at first medical contact score, is concordant with more sophisticated scores.