Diagnostic Accuracy of High-Sensitive Troponin for the Identification of Myocardial Infarction in Patients Presenting with Acute Heart Failure

Background: The differentiation of myocardial infarction (MI) in the setting of acute heart failure (AHF) can be challenging because the majority of patients presenting with AHF show elevations of high-sensitive troponin (hs-Tn). Fast identification of MI is crucial to perform timely coronary angiography and to improve clinical outcome. Objectives: The aim of the present study was to assess the diagnostic accuracy of different levels of hs-Tn for the identification of type 1 MI in patients with AHF. Methods: This was a retrospective single-center analysis of admitted AHF patients with documentation of high-sensitive troponin T (hs-TnT). Results: A total of 649 patients were enrolled into the present study. Of them, 18% had type 1 MI, 7% had type 2 MI, 69% had myocardial injury, and 6% had no myocardial injury. The area under the curve of hs-TnT for the prediction of type 1 MI was 0.70. Sensitivity and specificity of the hs-TnT 99^th percentile upper reference limit (URL) for type 1 MI was 100% and 8%, respectively. The Youden index derived cut-off of hs-TnT was 50 ng/L, showing a sensitivity and specificity for type 1 MI of 63% and 68%, respectively. No significant difference regarding 30-day mortality was found depending on the presence of type 1 MI (odds ratio 1.86; 95% confidence interval 0.91–3.81). Conclusions: Hs-TnT-based identification of type 1 MI in patients with AHF requires higher cut-offs compared with the 99^th percentile URL used in overall acute coronary syndrome populations. However, the adjusted cut-off provided only moderate sensitivity and specificity.