High-Sensitivity Cardiac Troponin T-gen6 Assay in Suspected Myocardial Infarction: Diagnostic Accuracy, Cutoffs, and Clinical Implications

Abstract

Background: The new high-sensitivity cardiac troponin (hs-cTn) T-generation (gen) 6 assay has higher analytical sensitivity than the current hs-cTnT-gen5 assay, but its clinical performance is unknown.

Objectives: The authors aimed to assess the clinical performance and use-optimized cutoffs of the hs-cTnT-gen6 assay.

Methods: In this secondary analysis of an international, prospective, diagnostic study with central adjudication, patients presenting with suspected myocardial infarction to the emergency department were enrolled. hs-cTnT-gen6 and hs-cTnT -gen5 were measured at the same time points. The analytical and diagnostic performance of hs-cTnT-gen6 were compared to hs-cTnT-gen5, and assay-specific cutoffs for the European Society of Cardiology 0/1- and 0/2-hour algorithms were derived and internally and externally validated.

Results: Among 3,346 patients, myocardial infarction was the final diagnosis in 616 (18.4%). The proportion of patients with concentrations above the upper reference limit and therefore having myocardial injury was lower for hs-cTnT-gen6 (35.9%) than for hs-cTnT-gen5 (43.5%; P < 0.001). Diagnostic accuracy at emergency department presentation, expressed as area under the curve, was 0.927 (95% CI: 0.918-0.937) for hs-cTnT-gen6 and 0.931 (95% CI: 0.921-0.94) for hs-cTnT-gen5. A gen6 cutoff of <6 ng/L ruled out non-ST-segment elevated myocardial infarction in 30.0% of patients with 100% sensitivity (95% CI: 99.4%-100%) irrespective of chest pain onset. An hs-cTnT-gen6 cutoff of <8 ng/L at presentation (if chest pain onset was >3 hours) or <18 ng/L together with a 0/1-hour delta of <2 ng/L ruled out 56.3% with a sensitivity of 99.7% (95% CI: 98.3%-99.9%). An hs-cTnT-gen6 concentration ≥112 ng/L or a 0/1-hour delta ≥10 ng/L ruled-in 20.0% with a specificity of 93.4% (95% CI: 92.0%-94.6%). An hs-cTnT-gen6 cutoff <8 ng/L at presentation (if chest pain onset was >3 hours) or <18 ng/L together with a 0/2-hour delta of <4 ng/L ruled out 51.1% with a sensitivity of 99.7% (95% CI: 98.1%-99.9%). An hs-cTnT-gen6 concentration ≥112 ng/L or a 0/2-hour delta ≥15 ng/L ruled-in 23.9% with a specificity 92.7% (95% CI: 90.8%-94.2%). Findings were consistent in sensitivity analyses as well as in the external validation cohort.

Conclusions: The novel hs-cTnT-gen6 assay demonstrated excellent and comparable diagnostic performance to the established hs-cTnT-gen5 assay with fewer patients identified as having myocardial injury. Use-optimized, assay-specific cutoffs for the European Society of Cardiology 0/1- and 0/2-hour algorithms provided very high sensitivity and high specificity. (Advantageous Predictors of Acute Coronary Syndromes Evaluation Study [APACE]; NCT00470587).

Keywords: cardiac troponin; myocardial infarction; triage algorithm.