💬 Editorial: A sixth-generation high-sensitivity cardiac troponin T assay could allow clinicians to reassure more emergency department patients they are not having a myocardial infarction at presentation, but further study is needed to optimize clinical application.
In 2008, Roche Diagnostics introduced a high-sensitivity version of their cardiac troponin T assay (hs-cTnT), a fifth-generation assay. Researchers quickly deduced that by using the assay’s limit of detection (LoD) of 5 ng/L (to convert to micrograms per liter, multiply by 0.001) as a cutoff, many patients could safely be classified as very low-risk for myocardial infarction (MI).1,2 Researchers gathered data across multiple institutions, and a 9241-patient meta-analysis demonstrated a pooled sensitivity for the LoD of 98.7%.3 Outside the US, on presentation (0-hour) concentrations less than LoD became guideline recommended, reassuring patients quickly and reducing time spent in busy emergency departments (EDs). Once US Food and Drug Administration (FDA) approval was obtained in the US, a similar risk-stratification approach became possible, though using a threshold of 6 ng/L because the FDA mandated that exact concentrations below the limit of quantitation (LoQ) be not reported. In the meantime, troponin I assay manufacturers brought to market high-sensitivity cardiac troponin I (hs-cTnI) assays. Low-risk thresholds were derived for these that were above the LoD and LoQ by identifying the concentration that gave a minimum prespecified statistical performance. Most often these minimums are greater than or equal to 99% sensitivity4 and greater than or equal to 99.5% negative predictive value (NPV). Roche Diagnostics has now placed in the hands of researchers a sixth-generation cTnT assay. Already this has been established as high sensitivity, with very low LoD and LoQ and well-defined sex-specific upper-reference levels.5 This will allow, for the first time with hs-cTnT, the derivation of single-sample, very low-risk thresholds likely usable across institutions. In this issue of JAMA Cardiol ogy, Thurston and colleagues6 present the first such derivation of a single-sample, very low-risk threshold for the Roche sixth-generation hs-cTnT assay.
In a prospective cohort study of 987 patients, blood was drawn at multiple time points from ED presentation. cTnT concentrations were measured on the same analyzer with both the fifth-and sixth-generation assays. This allowed derivation of a sixth-generation single-sample very low-risk threshold, a comparison of the performance of that threshold with the fifth-generation LoD, and determination of the performance of the High-Sensitivity Troponin in the Evaluation of Patients With Suspected Acute Coronary Syndrome (High-STEACS) early rule-out pathway. External validation used stored samples from the Advantageous Predictors of Acute Coronary Events (APACE) study. The primary outcome was an index or subsequent MI (types 1, 4b, or 4c) or cardiac death within 30 days. The prespecified goal was to determine the highest troponin threshold with statistical metrics NPV greater than or equal to 99.5% and sensitivity greater than or equal to 99%.