Diagnostic ability of a computer algorithm to identify prehospital STEMI

Abstract

Background: Acute myocardial infarction (AMI) accounts for 43% of deaths related to ischaemic heart disease, with ST-segment elevation myocardial infarction (STEMI) accounting for 25%–40% of all AMI presentations. Given the impact of these diseases, there is a strong prehospital focus on early identification, treatment and transport of patients with acute coronary syndrome. The main aim of the STEMI system of care is to reduce the time to reperfusion of the myocardium, thereby improving morbidity and mortality rates. Therefore, the identification of STEMI by paramedics can have a dramatic effect on patients' long-term health outcomes. Ambulance Victoria paramedics play a crucial role in the care provided to AMI patients across the state, with the assistance of a computer-automated interpretation of 12-lead electrocardiograms (ECGs) to aid STEMI identification. Objectives: This study's objective is to analyse the diagnostic capability of the computer-automated interpretation to diagnose STEMI in the out-of-hospital setting. Methods: Quantitative data from January 2018 to December 2019 was sourced from the Victorian Ambulance STEMI Quality Initiative. These data were periodically matched with hospital outcome and diagnosis data from the Victorian Cardiac Outcomes Registry to compare provisional paramedic diagnoses with the final hospital diagnoses. Results: Of the 5269 cases of suspected STEMI, 765 (14.5%) could be matched with outcome data. Of these 765 cases, 88.9% were correctly identified as STEMI. The remaining 10% were categorised as either non-STEMI or unstable angina. No data were available for 1.1%. Conclusions: The diagnostic capability of the Zoll Inovise 12L interpretive algorithm to diagnose STEMI in the out-of-hospital setting appears safe and feasible. However, because of limited data matching paramedic findings with patient outcomes in hospital, no hard conclusions can be drawn. Furthermore, there is no way to ascertain how many false positives the Zoll monitor is interpreting. Further investigation is required to assess the true diagnostic capability of the Zoll Inovise 12L interpretive algorithm.