Abstract 18733: Evaluation of a Super-Sensitivity Cardiac Troponin I Assay for Diagnosing Non ST-Segment Elevation Myocardial Infarction

Abstract

Introduction: Acute myocardial infarction (AMI) is defined by cardiac troponin (cTn) elevation. According to current guidelines the diagnosis of AMI requires a rise or fall of cTn (delta) with at least one value above the diagnostic cutoff, i.e. the 99th percentile (P99) of a reference population. Lately, cTn assays with high analytical precision have become available. Use of high-sensitivity cTn showed improved diagnostic accuracy with absolute deltas being superior compared to relative deltas. The diagnostic utility of super-sensitivity cTn assays, that are capable of cTn detection in >95% of the general population, has not been sufficiently elucidated. Hypothesis: We assessed the hypothesis that a super-sensitivity cTnI assay combined with a diagnostic algorithm that incorporates an absolute delta provides sufficient detection of non ST-segment elevation myocardial infarction (NSTEMI). Methods: The study comprised 1566 patients with new-onset chest pain (246 with NSTEMI, 222 with unstable angina pectoris, 1097 with non-coronary chest pain). On admission and 3 hours later cTnI was measured by a super-sensitivity assay (Erenna® cTnI immunoassay, Singulex, Inc., Alameda, CA, USA). Diagnostic performance was assessed via a two-step algorithm: First, 3h-cTnI had to exceed a certain cutoff concentration (P99 as well as arbitrarily chosen cTnI concentrations) and, when this criterion was fulfilled, secondly, an absolute delta criterion was applied. Results: A 3h-cTnI cutoff concentration of 40 ng/l provided a negative predictive value of 99.1% and a sensitivity of 95.5% indicating confident rule-out of NSTEMI, when this criterion was not fulfilled (n=1256). For patients with 3h-cTnI >40 ng/l (n=310) an absolute delta of ≥50 ng/l resulted in a positive predictive value, i.e. diagnostic certainty for ruling in NSTEMI, of 84.2%. Application of this algorithm to the overall study population yielded a negative predictive value of 96.1%, positive predictive value of 84.2%, sensitivity of 81.8% and a specificity of 97.1%. Conclusion: In conclusion, the super-sensitivity cTnI assay showed a clinically acceptable performance in diagnosing NSTEMI by use of an algorithm that is P99-independent and includes an absolute delta.