• Andrew McRae, Michelle Graham, Tasnima Abedin, Yunqi Ji, Hong Yang, Dongmei Wang, Danielle Southern, James Andruchow, Eddy Lang, Grant Innes, Isolde Seiden-Long, Lawrence DeKoning, and Peter Kavsak.
• *Departments of Emergency Medicine and Community Health Sciences,University of Calgary,Calgary, AB.
• Can J Emerg Med. 2019 Jan 1; 21 (1): 26-33.

ObjectiveSex-specific diagnostic cut-offs may improve the test characteristics of high-sensitivity troponin assays for the diagnosis of myocardial infarction (MI). The objective of this study was to quantify test characteristics of sex-specific cut-offs of a single, high-sensitivity cardiac troponin T (hs-cTnT) assay for 7-day MI in patients with chest pain.MethodsThis observational cohort study included consecutive emergency department (ED) patients with suspected cardiac chest pain from four Canadian EDs who had an hs-cTnT assay performed within 60 minutes of ED arrival. The primary outcome was MI at 7 days. We quantified test characteristics (sensitivity, negative predictive value [NPV], likelihood ratios and proportion of patients ruled out) for multiple combinations of sex-specific, rule-out cut-offs. We calculated the net reclassification index compared to universal rule-out cut-offs.ResultsIn 7,130 patients (3,931 men and 3,199 women), the 7-day MI incidence was 7.38% among men and 3.78% among women. Optimal sex-specific cut-offs (<8 ng/L for men and <7 ng/L for women) had a 98.5% sensitivity for MI and ruled out MI in 55.8% of patients. This would enable an absolute increase in the proportion of patients who were able to be ruled out with a single hs-cTnT of 13.2% to 22.2%, depending on the universal rule-out concentration used as a comparator.ConclusionsSex-specific hs-cTnT cut-offs for ruling out MI at ED arrival may improve classification performance, enabling more patients to be safely ruled out at ED arrival. However, differences between sex-specific and universal cut-off concentrations are within the variation of the assay, limiting the clinical utility of this approach. These findings should be confirmed in other data sets.

29 keywords...

hide…