Trial design for assessing analytical and clinical performance of high-sensitivity cardiac troponin I assays in the United States: The HIGH-US study

Abstract

Background: High-sensitivity cardiac troponin I (hs-cTnI) assays have been developed that quantify lower cTnI concentrations with better precision versus earlier generation assays. hs-cTnI assays allow improved clinical utility for diagnosis and risk stratification in patients presenting to the emergency department with suspected acute myocardial infarction. We describe the High-Sensitivity Cardiac Troponin I Assays in the United States (HIGH-US) study design used to conduct studies for characterizing the analytical and clinical performance of hs-cTnI assays, as required by the US Food and Drug Administration for a 510(k) clearance application. This study was non-interventional and therefore it was not registered at clinicaltrials.gov. Methods: We conducted analytic studies utilizing Clinical and Laboratory Standards Institute guidance that included limit of blank, limit of detection, limit of quantitation, linearity, within-run and between run imprecision and reproducibility as well as potential interferences and high dose hook effect. A sample set collected from healthy females and males was used to determine the overall and sex-specific cTnI 99th percentile upper reference limits (URL). The total coefficient of variation at the female 99th percentile URL and a universally available American Association for Clinical Chemistry sample set (AACC Universal Sample Bank) from healthy females and males was used to examine high-sensitivity (hs) performance of the cTnI assays. Clinical diagnosis of enrolled subjects was adjudicated by expert cardiologists and emergency medicine physicians. Assessment of temporal diagnostic accuracy including sensitivity, specificity, positive predictive value, and negative predictive value were determined at presentation and collection times thereafter. The prognostic performance at one-year after presentation to the emergency department was also performed. This design is appropriate to describe analytical characterization and clinical performance, and allows for acute myocardial infarction diagnosis and risk assessment. © 2019