Risk stratification of patients with chest pain or anginal equivalents in the emergency department.

Abstract

We studied whether previously developed cardiac risk scores-including history, ECG, age, risk factors, and troponin (HEART); Thrombolysis in Myocardial Infarction (TIMI); Global Registry of Acute Coronary Events (GRACE); and Emergency Department Assessment of Chest Pain (EDACS)-could be applied to predict major adverse cardiac events (MACE) in patients with possible coronary artery disease, including anginal equivalents. Patients with chest pain or anginal equivalents who underwent coronary computed tomographic angiography were included. The primary outcome was 30-day MACE. We compared the cardiac risk scores by the area under the receiver-operating characteristic curves (AUC). The primary outcome occurred in 200 patients (16.0%) of the 1247 patients included. For the prediction of MACE, the AUC of the HEART score (0.765) was superior to those of the TIMI (0.726), GRACE (0.612), and EDACS (0.631) scores. Among patients identified by each score as being at low risk, the MACE rate was the lowest for the HEART score (5.7%), followed by the TIMI (8.8%), EDACS (11.2%), and GRACE (12.2%) scores. At a sensitivity level of a < 2% rate of misses, the negative predictive value of the HEART score (1.0) outperformed those of the GRACE (0.932) and EDACS (0.964). The HEART score appeared to be more predictive of MACEs than the TIMI, GRACE, and EDACS in patients with chest pain or anginal equivalents. However, previously suggested cutoff could not safely identify low-risk patients for early discharge because of the unacceptably high rate of missed MACEs.