Author Affiliation: Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario, Canada
Reichlin and colleagues1 have developed an algorithm for ruling-in or ruling-out myocardial infarction (MI) within 1 hour after presentation using high-sensitivity cardiac troponin T (hs-cTnT) for patients within the Advantageous Predictors of Acute Coronary Syndrome Evaluation study. The data analyses using classification and regression tree analysis for ruling-in via a training and validation set are powerful for supporting a 1-hour sampling/testing algorithm, albeit there is a major limitation in that both groups are from the same study population.2 Most strikingly, however, is the analytical criteria that was developed in this study to rule-in an MI, specifically, that either a hs-cTnT concentration of 52 ng/L or greater (to convert to micrograms per liter, multiply by 0.001) or a change (delta) in hs-cTnT concentrations of 5 ng/L or greater within 1 hour was used to rule-in.1 The absolute concentration of 52 ng/L is not the published 99th percentile for this assay3 and is in fact similar to the fourth generation's troponin T 10% coefficient of variation (CV) concentration of 0.03 ng/L, which has been shown in a large prospective study to predict death within 30 days.4 More intriguing is the 5-ng/L absolute change in hs-cTnT concentrations. On the basis of reported precision data, the 10% CV for the hs-cTnT is 13 ng/L1; using established laboratory medicine practices, acceptable variation would permit concentrations ranging from 15.6 ng/L to 10.4 ng/L around this point estimate. Thus, the imprecision of the assay alone could result in a change of 5 ng/L or greater in absence of any clinical change.
Kavsak PA. High-Five for High-Sensitivity Cardiac Troponin T: Depends on the Precision and Analytical Platform. JAMA Intern Med. 2013;173(6):477. doi:10.1001/jamainternmed.2013.2270
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: