P1558Clinical utility of the PRECISE-DAPT score in the prediction of hemorrhagic events in elderly patients with acute coronary syndrome

Abstract Introduction The clinical utility and validity of the PRECISE-DAPT bleeding risk score for elderly patients with acute coronary syndrome (ACS) is unknown. We investigated the previous aspect in a contemporary population treated with percutaneous coronary intervention (PCI) and dual antiplatelet therapy (DAPT) at discharge. Methods Retrospectively, from 2010 to 2016, we studied 3,814 consecutive patients with the diagnosis of ACS. All patients were treated with in-hospital PCI and DAPT at discharge. Elderly was defined if patients aged ≥75 years. Patients were categorized into three risk strata according to their PRECISE-DAPT score (very low-low: <17 points, moderate: 18–24 points, and high risk: ≥25 points). We included the first bleeding event occurred during the first year after discharge. Bleeding events were defined according to the BARC classification system, and divided into two subgroups: BARC 2–5 and BARC 3–5. The ability to separate high bleeding risk patients from lower bleeding risk patients was checked by the cumulative incidence function curves and compared using the Fine-Gray test, thus adjusting for death (non-bleeding related) as a competing risk. Discrimination (C statistic) and calibration (Hosmer-Lemeshow test) were used to test the predictive capacity of the score in pts aged ≥75 years and <75 years. Results 25.2% (n=961/3814) were ≥75 years old, 38.4% of them were women. DAPT duration was 11.5 (interquartile range [IQR] 2.5–13.7) vs. 12.0 (RIQ 8.2–14.1) months in the elderly vs. younger; (p<0.001). 92.5% (n=889) of the elderly were at high risk of bleeding (PRECISE-DAPT≥25 points), compared to 21.3% (n=607) of the youngest. The incidence of BARC 2–5 and BARC 3–5 was 7.4% and 2.7% in the elderly compared to 5.1% and 1.4% among the younger patients (p<0.001). The figure shows the ability of the PRECISE-DAPT score at capturing the risk of BARC 2–5 bleeding (panel A and B), in both age groups. Using the cut-off point ≥25, the effect in the prediction of BARC 2–5 bleeding and BARC 3–5 did not differ significantly between the elderly and those <75 years: sHR = 1.9 (95% CI: 1.2–6.00) in the elderly vs. 1.8 (95% CI: 1.3–2.5) in the other group (p=0.99) and sHR = 3.3 (95% CI: 1.9–6.0) vs. 3.6 (95% CI: 1.9–6.7) (p=0.83), respectively. There were no significant differences between the elderly and those under 75 years in terms of statistical C values either for BARC 2–5 bleeding (0.60 vs. 0.58) or BARC 3–5 bleeding (0.64 vs. 0.67). The score performed well in term of calibration in both groups (all p-values >0.3). Conclusion Although the use of PRECISE-DAPT resulted in classifying the majority of elderly patients at high risk of bleeding and despite exhibiting modest discriminative power, it performed well at classifying patients according to their risk of 1-year out-of-hospital bleeding in both age groups. PRECISE-DAPT appears to identify the truly low risk patients among the elderly, as compared to the younger group.