Comparison of CRUSADE and ACUITY-HORIZONS Bleeding Risk Scores in Patients With Acute Coronary Syndromes.

Background: Compare the discriminative performance of two validated bleeding risk models for in-hospital bleeding events in a non-selected cohort of acute coronary syndrome (ACS) patients.

Methods: CRUSADE (Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA Guidelines) and ACUITY-HORIZONS (Acute Catheterization and Urgent Intervention Triage strategY-Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction) scores were calculated in 501 consecutive patients (median age 68 years (IQR 57-77), 31% female) admitted for ACS to the coronary care unit (CCU) of San Paolo Hospital in Milan (Italy). In-hospital haemorrhagic events and mortality were recorded and calibration and discrimination of the two risk models were evaluated using the Hosmer-Lemeshow test and the C-statistic, respectively.

Results: Overall bleeding events were observed in 32 patients and major bleedings in 11 (with an incidence of 6.4% and 2.2%, respectively). In-hospital mortality was 2.6%. Regarding major bleedings both risk scores demonstrated an adequate calibration (H-L test p>0.20) and a moderate discrimination with no significant difference in predictive accuracy between the two models (C-statistic 0.69 for CRUSADE and 0.73 for ACUITY-HORIZONS). We also tested the performance of the two risk models in predicting in-hospital mortality, showing an adequate calibration and a very good discrimination (C-statistic 0.88 and 0.89 for the CRUSADE and ACUITY-HORIZONS scores, respectively), with no significant difference in predictive accuracy.

Conclusions: In our ACS population the CRUSADE and the ACUITY-HORIZONS risk scores showed a fairly good and comparable predictive accuracy regarding in-hospital bleeding events and they appeared to be very good predictors of in-hospital mortality.