Use of troponins in clinical practice: Evidence in favour of use of troponins in clinical practice

All troponin assays are equal but some assays are more equal than others (with apologies to George Orwell). Measurement of cardiac troponin (cTn) using ‘high sensitivity’ cTn (hs-cTn) assays is supported by evidence-based medicine and is incorporated into the guidelines of the European Society of Cardiology1 as well as the Fourth Universal Definition of Myocardial Infarction (2018).2 The majority of diagnostic companies have launched or are in the process of launching cTn assays into routine clinical practice and for use in research trials, which meet the criteria for high sensitivity. A high sensitivity assay is defined analytically3 by the ability to measure at least 50% of a reference population and an imprecision (coefficient of variation (%CV)) <10% at the 99th percentile upper reference limit (URL). In clinical practice, the imprecision of measurement at the 99th percentile for most assays is typically much lower, <5%, and the limit of detection of the assay (the concentration which can be reliably distinguished from background noise) is very low. The advent of hs-cTn assays moving globally into routine clinical and research use is associated with a number of challenges. For front-line clinicians, there needs to be acceptance that these assays are exceptionally sensitive tools for the detection of myocardial injury but as has always been the case, myocardial injury does not always equate with myocardial infarction (MI). Indeed, with hs-cTn assays, MI subsumes only a modest percentage of the increases observed.4 For laboratory medicine, there are challenges with service delivery based on the use of diagnostic algorithms incorporating an admission measurement with repeat measurement at 1–3 hours post-admission.1 For journals, including readers, editors and their reviewers, there is also the challenge of understanding and educating hs-cTn users on the nature of the improved analytics of hs-assays, how they differ from …