Delta High-Sensitivity Troponin Cut Points in Patients With Coronary Artery Disease

Our team has previously demonstrated that a delta high-sensitivity troponin (hs-cTnI; Beckman Coulter) of 4 ng/L may be the optimal cut-point to exclude myocardial infarction (MI) among patients with chest pain. However, data evaluating optimal delta cut-points in patients with known coronary artery disease (CAD) are limited. This study aims to evaluate the diagnostic performance of a range of delta hs-cTnI measurements for index MI within a subgroup of patients with known CAD.

We conducted an observational cohort study using the Wake Forest Chest Pain Registry. Patients ≥18 years old with chest pain, known CAD, serial hs-cTnI measures, and without STEMI on ECG were accrued from five US Emergency Departments from 11/1/2020-7/31/2022. Known CAD was defined as prior MI, coronary revascularization, or known coronary stenosis ≥70%. Patients’ 0- and 2-hour hs-cTnI (Beckman Coulter) measures were used to determine delta values. The primary outcome was MI at the index visit determined by validated ICD codes. The efficacy, negative predictive value (NPV), and negative likelihood ratio (-LR) for index MI with associated 95% confidence intervals were assessed for a range of absolute delta cut-points (1-20 ng/L). Efficacy (rule-out rate) was defined as the proportion of patients with a delta value below each cut-point. An optimal delta cut-point was defined by achieving maximal efficacy while maintaining a NPV ≥99% for index MI.

During the study period, 3,184 patients with known CAD were accrued. Among these, 39.3% (1,251/3,184) were female, 23.8% (750/3,184) were non-White, and the median age was 67 (IQR: 58-75) years. Index MI occurred in 20.7% (659/3,184). At the previously established overall delta cut-point of <4 ng/L, the efficacy was 67.8% (95%CI: 66.1-69.4%) with a NPV of 96.7% (95%CI: 95.8-97.4%) and a -LR of 0.13 (95%CI: 0.11-0.16) for index MI. No delta threshold achieved a NPV of ≥99% for index MI. However, at a delta of <2 ng/L, the -LR for index MI was 0.08 (95%CI: 0.06-0.11).

No delta value achieved a NPV ≥99% for index MI in this subgroup of patients with known CAD. However, a delta <2 ng/L may be useful in this population given its -LR.