Red cell distribution width and additive risk prediction for major bleeding in non-ST-segment elevation acute coronary syndrome.

INTRODUCTION AND OBJECTIVES Red cell distribution width has been linked to an increased risk for in-hospital bleeding in patients with non-ST-segment elevation acute coronary syndrome. However, its usefulness for predicting bleeding complications beyond the hospitalization period remains unknown. Our aim was to evaluate the complementary value of red cell distribution width and the CRUSADE scale to predict long-term bleeding risk in these patients. METHODS Red cell distribution width was measured at admission in 293 patients with non-ST-segment elevation acute coronary syndrome. All patients were clinically followed up and major bleeding events were recorded (defined according to Bleeding Academic Research Consortium Definition criteria). RESULTS During a follow-up of 782 days [interquartile range, 510-1112 days], events occurred in 30 (10.2%) patients. Quartile analyses showed an abrupt increase in major bleedings at the fourth red cell distribution width quartile (> 14.9%; P=.001). After multivariate adjustment, red cell distribution width >14.9% was associated with higher risk of events (hazard ratio=2.67; 95% confidence interval, 1.17-6.10; P=.02). Patients with values ≤ 14.9% and a CRUSADE score ≤ 40 had the lowest events rate, while patients with values >14.9% and a CRUSADE score >40 points (high and very high risk) had the highest rate of bleeding (log rank test, P<.001). Further, the addition of red cell distribution width to the CRUSADE score for the prediction of major bleeding had a significant integrated discrimination improvement of 5.2% (P<.001) and a net reclassification improvement of 10% (P=.001). CONCLUSIONS In non-ST-segment elevation acute coronary syndrome patients, elevated red cell distribution width is predictive of increased major bleeding risk and provides additional information to the CRUSADE scale.