An approach to rule-out an acute cardiovascular event or death in emergency department patients using outcome-based cutoffs for high-sensitivity cardiac troponin assays and glucose

Abstract

ObjectivesThe application of “undetectable” high-sensitivity cardiac troponin (hs-cTn) concentrations to “rule-out” myocardial infarction is appealing, but there are analytical concerns and a lack of consensus on what concentration should be used to define the lower reportable limit; i.e., limit of detection (LoD) or limit of blank. An alternative approach is to utilize a measurable hs-cTn concentration that identifies patients at low-risk for a future cardiovascular event combined with another prognostic test, such as glucose. We assessed both of these approaches in different emergency department (ED) cohorts to rule-out an event. Design and methodsWe used cohort 1 (all-comer ED population, n=4773; derivation cohort) to determine the most appropriate approach at presentation (i.e., Dual Panel test: hs-cTn/glucose vs. LoD vs. LoD/glucose) for an early rule-out of hospital death using the Abbott ARCHITECT hs-cTnI assay. We used cohort 2 (n=144) and cohort 3 (n=127), both early chest pain onset ED populations as the verification datasets (outcome: composite cardiovascular event at 72h) with three hs-cTn assays assessed (Abbott Laboratories, Beckman Coulter, Roche Diagnostics). ResultsIn cohort 1, the sensitivity was >99% for all three approaches; however the specificity (11%; 95% CI: 10–12%) was significantly higher for the Dual Panel as compared to the LoD approach (specificity=5%; 95% CI: 4–6%). Verification of the Dual Panel in cohort 2 and cohort 3 revealed 100% sensitivity and negative predictive values for all three hs-cTn assays. ConclusionsThe combination of a “healthy” hs-cTn concentration with glucose might effectively rule-out patients for an acute cardiovascular event at ED presentation.