Article Text
Abstract
Objectives and Backgrounds High sensitivity troponin (hs-Tn) has been shown to improve early diagnosis of acute myocardial infarction (AMI). We recently reported that levels below the limit of blank (LoB, <3 ng/l) of the Roche assay had 100.0% negative predictive value (NPV). Now, we aim to validate that finding in (i) a cohort study and (ii) routine clinical use following implementation.
Methods Initially, we included adult emergency department (ED) patients with suspected cardiac chest pain in a prospective diagnostic cohort study. Standard troponin T (TnT) and hs-TnT (Roche) were tested at presentation. The primary outcome was a diagnosis of AMI, adjudicated by two independent investigators based on reference standard ≥12 h TnT testing. Secondary outcomes included AMI defined by ≥12 h hs-TnT. After clinically implementing hs-TnT at a different institution, we identified all patients undergoing ≥2 hs-TnT tests within 24 h and evaluated initial hs-TnT level for predicting any subsequent rise.
Results 477 patients were included in the cohort study. 475 (99.6%) underwent reference standard testing. 82 (17.3%) had AMI. The area under the ROC curve was 0.95 (95% CI 0.92 to 0.97) for hs-TnT compared to 0.84 (0.78 to 0.90) for TnT (p=0.001). At the 99th percentile, sensitivity was 93.9% (86.3% to 98.0%) for hs-TnT and 70.7% (59.7% to 80.3%) for TnT (p<0.0001; Abstract 001 table 1). At the LoB (<3 ng/l) cut-off, sensitivity of hs-TnT was 98.8% (93.4% to 100.0%) with NPV 99.2% (95.5% to 100.0%). Using ≥12 h hs-TnT to define AMI, sensitivity of initial hs-TnT<3 ng/l was 99.4% (96.5% to 100.0%). On subsequently appraising hs-TnT in 915 patients in clinical practice, sensitivity (at the LoB) was 99.8% (99.1% to 100.0%) with NPV 99.4% (96.6% to 100.0%).
Conclusions This work again demonstrates that hs-TnT improves sensitivity for AMI in the ED. Undetectable hs-TnT (<3 ng/l) levels have very high NPV, a strategy that has now been prospectively validated in both research and clinical practice populations. A randomised controlled trial of early discharge using hs-TnT is being planned.