Introduction Patients presenting to Emergency Departments with chest pain of possible cardiac origin represent a substantial and challenging cohort to risk stratify. Scores such as HE-MACS (History and Electrocardiogram-only Manchester Acute Coronary Syndromes decision aid) and HEAR (History, ECG, Age, Risk factors) have been developed to stratify risk without the need for troponin testing. Validation of these scores remains limited at present.

Methods We performed a post-hoc analysis of the LoDED (Limit of Detection and ECG discharge strategy) randomised-controlled trial dataset. Data collected during the trial, and prior to troponin results, were used to calculate HEAR and HE-MACS scores. Previously published thresholds of <4% for HE-MACS and <2 for HEAR defined very low risk patients. The primary outcome of MACE (major adverse cardiac events) at 30 days was used to assess diagnostic accuracy.

Results 629 patients were included, with a 7% (42/629) incidence of MACE within 30 days. HE-MACS and HEAR scores identified 85/629 and 181/629 patients predicted to be at very low risk of MACE. Within these cohorts, MACE occurred in 0/85 and 1/181 patients respectively. Sensitivity of HE-MACS and HEAR were 100% (95% CI: 91.6-100%) and 97.6% (95% CI: 87.7-99.9%) respectively. Receiver operating characteristic (ROC) curves demonstrated area under curve (AUC) of 0.80 [95%CI: 0.74-0.85] for HE-MACS and 0.76 [95% CI: 0.69-0.82] for HEAR.

Conclusion HEAR and HE-MACS show potential as rule out tools for acute myocardial infarction without the need for troponin testing. However, prospective studies are required to further validate these scores for clinical implementation.

Conflict of Interest None declared