Abstract 11910: The Role of Bleeding Risk Score as a Risk Factor for In-Hospital Mortality in Acute Coronary Syndromes
Background: It is known that bleeding is a powerful risk factor for mortality in pts with acute coronary syndromes (ACS), and that bleeding score (BS) may predict the complication. We assessed the hypothesis that BS could be useful to identify patients at greater risk for in-hospital mortality (IHM).
Methods: After the publication of a new BS by Mehran et al (2010), we started to prospectively evaluate the bleeding risk of pts hospitalized with ACS. Moreover, we calculated the BS retrospectively in 1416 pts from our databank, totalizing 1665 pts. Correlation between BS and IHM were assessed by Mann-Whitney and multivariate stepwise logistic regression. Additionally, ROC curves were constructed in order to analyze the discriminatory power of BS as IHM predictor. Analyses were done for the global population and for the subsets with ST-elevation acute myocardial infarction (STEMI, N=547) and non-ST-elevation ACS (NSTEACS, N=1118).
Results: (1) For the global population, the mean BS for survivors and deceased individuals were, respectively, 17.4+7.4 and 24.7+8,2 (P<0.001); the adjusted model with 15 baseline and in-hospital variables (including surgical and catheter interventions) showed a P-value<0.001; the ROC curve showed an area under the curve (AUC) of 0.753+0.025 (P<0.001). (2) For the STEMI pts, the mean BS were 21.0+6.6 and 25.7+7.6, respectively (P<0.001); an adjusted model with the same variables previously described showed a significant and independent correlation between BS and IHM (P=0.001); including the STEMI TIMI risk score in the equation the P-value for BS was 0.034 (P<0.001 for the TIMI score); the AUC for the BS was 0.686=+0.04, being 0.798+0.032 for the TIMI score (P=0.029 for the comparison between the AUCs). (3) For the NSTEACS group, the mean BS were 15.8+7.2 and 23.8+8.7, respectively (P<0.001); in the adjusted model the correlation between BS and IHM was highly significant (P<0.001), with the same P-value when adding the NSTEACS TIMI risk score in the model; the AUC for BS was 0.769+0.036 (0.616+0,037 for the TIMI score, P=0.003). There was no difference (P=0.123) between the AUCs in pts with STEMI and NSTEACS.
In conclusion, the Mehran's bleeding risk score is an excellent tool in the evaluation of pts with ACS, especially those with NSTEACS.
- © 2011 by American Heart Association, Inc.