In the Emergency Department, chest pain triage systems are based on either clinical features or ECG recording. In this prospective, single-center, observational study, we aimed to compare the diagnostic performance of these triage systems in distinguishing acute coronary syndromes (ACS) from diseases of mild severity. Patients were sorted into the triage systems based on collected data at admission and on a systematic 12-lead ECG performed at triage. The final diagnosis was determined after a 30-day follow-up. For ACS, we determined a high-acuity triage score (Level 1 or 2) as being adequate, and for mild severity diseases a low-acuity triage score (Level 3, 4 or 5) as being adequate. The diagnostic performance of all studied systems was moderate (AUC from 0.644 to 0.694), with no statistically significant difference found between them. However, characteristics of the systems differed because the clinical-based systems had a higher sensitivity (87-91%) but lower specificity (32-39%) compared with the ECG-based system (sensitivity 62% and specificity 64%). A higher sensitivity limits the risk of a patient with acute coronary syndrome staying unsafely in the waiting room, while a higher specificity prevents overcrowding. ECG at triage also ensures that no STEMIs or high-risk NSTEMIs are missed. Based on these findings, each Emergency Depatment could more accurately select the triage system that fits their local particularities.
Keywords: Acute coronary syndrome; Chest pain; ED crowding; Emergency triage system.