A single value of high-sensitive troponin T below the limit of detection is not enough for ruling out non ST elevation myocardial infarction in the emergency department

Abstract

BackgroundRecent 2015 ESC recommendations for the management of patients with suspected acute myocardial infarction (AMI) support that a single value of high-sensitivity cardiac troponin (HS-cTn) measured at presentation could rule out AMI if below the limit of blank (LoB) or detection (LoD). ObjectivesWe aimed to evaluate whether an undetectable HS-cTnT at presentation safely rules out NSTEMI in unselected patients with chest pain. Patients and methodsThis is a post hoc analysis of two prospective cohorts with similar design that included patients suspected of AMI at three French university hospitals. Patients were followed-up during one month, before the adjudication of a final diagnosis. Results413 patients (mean age 58±17years) were analyzed; 45 (11%) had a final diagnosis of NSTEMI, and 26 (6%) had STEMI. The sensitivity of HS-cTnT value at 3ng/L (LoB) for NSTEMI was 97.8% [95% CI: 86.8–99.9], yielding a negative predictive value (NPV) of 99.3% [95% CI: 95.4–100.0]. Proportion of patients ruled out for NSTEMI was 32% when applying the LoB. The sensitivity of HS-cTnT value at 5ng/L (LoD) was 97.8% [95% CI: 86.8–99.9] yielding a NPV of 99.5% [95% CI: 96.5–100.0]. Proportion of patients ruled out for NSTEMI was 43% when applying the LoD. One patient (delay between onset of chest pain and presentation <3h) had NSTEMI and HS-cTnT <LoB at presentation. ConclusionThe NPV of a single measurement of HS-cTnT below the LoD is high in unselected patients, but not enough to rule out safely NSTEMI for very early presenters.