Discriminative Power of the HEART Score for Obstructive Coronary Artery Disease in Acute Chest Pain Patients Referred for CCTA.

ABSTRACT

OBJECTIVE: To investigate the ability of the HEART score to predict the presence of obstructive coronary artery disease (CAD) determined by coronary computed tomography angiography (CCTA) and its ability to predict the occurrence of major adverse cardiac events (MACE) in patients referred for CCTA after emergency department (ED) presentation. METHODS: From December 2011 to August 2014, 710 ED patients with chest pain who underwent CCTA within 30 days were included. The HEART score was retrospectively calculated and patients were followed for MACE, comprised of death, myocardial infarction, and revascularization. Association of CAD at CCTA in the different categories of the HEART score was analyzed using χ test. The performance of the HEART score in discriminating between those with and without obstructive CAD was evaluated by receiver operating characteristics. Kaplan-Meier analysis was used to assess MACE-free survival stratified by HEART-score categories. RESULTS: During median follow-up of 826 days (interquartile range 563-1056), MACE occurred in 46 (6.5%) patients; 3 (0.4%) myocardial infarction, 8 (1.1%) death, and 36 (5.1%) revascularizations. A low HEART score was a significant predictor for MACE-free survival (P = 0.010). CCTA revealed obstructive CAD in 11.7% of patients, with no significant difference between patients with a low and intermediate/high HEART score, respectively 10.7% and 13.2% (P = 0.29). The ability of the HEART score to identify obstructive CAD was poor with an AUC of the receiver operating characteristics curve of 0.53. CONCLUSION: The HEART score does not adequately identify patients with obstructive CAD at CCTA. It does however predict occurrence of MACE in medium-term follow-up. Excluding patients from additional testing based solely on a low HEART score may lead to suboptimal patient management. CCTA had important implications on patient management and may be a more appropriate tool to further stratify risk in ED chest pain patients.