Performance of the HEART, Emergency Department Assessment of Chest Pain Score (EDACS) and Vancouver Chest Pain Rule (VCPR) to identify low risk chest pain in the Emergency Department

ABSTRACT

Objective: Chest pain is a common Emergency Department (ED) presentation, but most patients will not have an Acute Coronary Syndrome (ACS). Decision tools have been developed to limit admissions to those at high risk of ACS, whilst allowing safe discharge of those at low risk. We aimed to evaluate the performance of three international decision tools (HEART, EDACS and VCPR) in the local setting. Design and Methodology: A prospective, observational cohort study was conducted to include ED patients presenting with low/intermediate risk chest pain. Participants were followed up at 6 weeks for Major Adverse Cardiac Events (MACE). Performance of clinical gestalt and the 3 scores were estimated. Results: 537 patients were enrolled, with 12 lost to followup, leaving 525 for analysis. Six patients developed MACE (1.14%). HEART had sensitivity, 16.67% (95% CI; 0.42% - 64.12%); specificity, 96.15% (95% CI; 94.11% - 97.63%); negative likelihood ratio (NLR), 0.87 (95% CI; 0.61 ­ 1.24). EDACS had sensitivity, 16.67% (95% CI; 0.42% - 64.12%); specificity, 95.38% (95% CI; 93.2% - 97.02%); NLR, 0.87 (95% CI; 0.61 ­ 1.25). VCPR had sensitivity 83.33% (95% CI; 35.88% - 99.58%); specificity, 67.05% (95% CI; 62.82% - 71.09%); NLR, 0.25 (95% CI; 0.04 ­ 1.49). Clinician gestalt had sensitivity 16.67% (95% CI; 0.42% - 64.12%); specificity, 97.5% (95% CI; 95.75% - 98.66%); NLR, 0.85 (95% CI; 0.6 ­ 1.22). Conclusion: In local patients with chest pain, the VCPR performed best for identifying patients for safe discharge, as well as with the lowest risk of MACE. The HEART, EDACS and clinical gestalt appear to be poorer decision tools.