• Kai M Eggers, Sally Aldous, Jaimi H Greenslade, Nina Johnston, Bertil Lindahl, William A Parsonage, John W Pickering, Martin Than, and Louise Cullen.
• Department of Medical Sciences, Uppsala University, Uppsala, Sweden. Electronic address: kai.eggers@ucr.uu.se.
• Clin. Chim. Acta. 2015 May 20;445:19-24.

AimsAssessment of patients with suspected non-ST elevation myocardial infarction (NSTEMI) is based on cardiac troponin (cTn) levels with the 99th percentile as cut-off. However, cardiovascular risk starts already at lower troponin concentrations. We therefore, aimed to investigate the utility of 2-hour algorithms using the high-sensitivity cardiac troponin I (hs-cTnI) 97.5th percentile as cut-off which corresponds to the standard URL for most biomarkers.MethodsHs-cTnI was measured at presentation and 2h in 1624 chest pain patients. Diagnostic algorithms were developed applying hs-cTnI levels dichotomized at the 99th and 97.5th percentiles combined with hs-cTnI changes and/or ECG findings.ResultsThe prevalence of NSTEMI was 13.9%. The adjusted odds ratios for 1-year mortality were 2.7 (95% CI 1.4-5.1) for the 99th percentile and 3.1 (95% CI 1.6-5.9) for the 97.5th percentile. The best-performing 99th percentile-based algorithms provided a positive predictive value (PPV) of 86.3% and a negative predictive value (NPV) of 99.3%. Using 97.5th percentile-based algorithms to define NSTEMI resulted in few reclassifications and yielded similar diagnostic estimates (PPV 85.4%, NPV 99.4%).ConclusionThe hs-cTnI 97.5th percentile integrated into 2-hour algorithms provided high diagnostic estimates and could, due to better prognostic properties serve as an alternative to the 99th percentile.Copyright © 2015 Elsevier B.V. All rights reserved.

26 keywords...

hide…