High-Sensitivity Troponin and the Application of Risk Stratification Thresholds in Patients With Suspected Acute Coronary Syndrome

Background: Guidelines acknowledge the emerging role of high-sensitivity cardiac troponin (hs-cTnl) for risk stratification and the early rule-out of myocardial infarction, but multiple thresholds have been described. We evaluate the safety and effectiveness of risk stratification thresholds in patients with suspected acute coronary syndrome. Methods: Consecutive patients with suspected acute coronary syndrome (n=48 282) were enrolled in a multicenter trial across 10 hospitals in Scotland. In a prespecified secondary and observational analysis, we compared the performance of the limit of detection (<2 ng/L) and an optimized risk stratification threshold (<5 ng/L) using the Abbott high-sensitivity troponin I assay. Patients with myocardial injury at presentation, with <= 2 hours of symptoms or with ST-segment elevation myocardial infarction were excluded. The negative predictive value was determined in all patients and in subgroups for a primary outcome of myocardial infarction or cardiac death within 30 days. The secondary outcome was myocardial infarction or cardiac death at 12 months, with risk modeled using logistic regression adjusted for age and sex. Results: In total, 32 837 consecutive patients (61 +/- 17 years, 47% female) were included, of whom 23 260 (71%) and 12,716 (39%) had hs-cTnl concentrations of <5 ng/L and <2 ng/L at presentation. The negative predictive value for the primary outcome was 99.8% (95% CI, 99.7%-99.8%) and 99.9% (95% CI, 99.8%-99.9%) in those with hs-cTnl concentrations of <5 ng/L and <2 ng/L, respectively. At both thresholds, the negative predictive value was consistent in men and women and across all age groups, although the proportion of patients identified as low risk fell with increasing age. Compared with patients with hs-cTnl concentrations of >= 5 ng/L but <99th centile, the risk of myocardial infarction or cardiac death at 12 months was 77% lower in those <5 ng/L (5.3% vs 0.7%; adjusted odds ratio, 0.23 [95% CI, 0.19-0.28]) and 80% lower in those <2 ng/L (5.3% vs 0.3%; adjusted odds ratio, 0.20 [95% CI, 0.14-0.29]). Conclusions: Use of risk stratification thresholds for hs-cTnl identify patients with suspected acute coronary syndrome and at least 2 hours of symptoms as low risk at presentation irrespective of age and sex.