Performance of High-Sensitivity Troponin T Risk Stratification Strategies for 90-day Cardiac Death or Myocardial Infarction

BACKGROUND: Thirty-day performance of the high-sensitivity troponin T (hs-cTnT) European Society of Cardiology 0/1-hour (ESC 0/1-h) and "one-and-done" (hs-cTnT<limit of quantification) strategies are established. However, 90-day performance is unclear. Our objective was to evaluate the 90-day performance of these hs-cTnT strategies in a US cohort. METHODS: A preplanned secondary analysis of a prospective multisite US cohort was conducted. Adults with chest pain were enrolled from 8 emergency departments (January 2017-September 2018). hs-cTnT measures (0- and 1-h) were used to classify patients by the ESC 0/1-h algorithm into rule-out, observation, and rule-in zones. Patients with 0-h measures <limit of quantification were considered ruled out by the one-and-done strategy. The primary outcome was adjudicated 90-day cardiac death or myocardial infarction (MI). Negative predictive value (NPV) for the primary endpoint and efficacy (proportion ruled out) were calculated for each strategy alone and in combination with the History, ECG, Age, Risk factor, and Troponin (HEART) score. RESULTS: Among 1462 patients with a mean age of 57.6 ± 12.9 years, 46.4% (678/1462) were female, and 14.0% (205/1462) had cardiac death or MI at 90 days. One-and-done strategy efficacy was 32.8% (479/1462), and NPV was 99.0% [95% confidence interval (CI), 97.6-99.7]. Adding the HEART score decreased efficacy to 20.1% (293/1462) and increased NPV to 99.7% (95% CI, 98.1-100). ESC 0/1-h efficacy was 57.8% (826/1430) and NPV was 98.3% (95% CI, 97.2-99.1). Combined with a HEART score, NPV increased to 99.3% (95% CI, 98.0-99.9), but efficacy decreased to 30.8% (95% CI, 28.3-33.2). CONCLUSIONS: The one-and-done strategy and ESC 0/1-hour algorithm had modest rates of missed 90-day cardiac death or MI. Adding a HEART score improved safety but decreased efficacy.