Sex-specifc thresholds using cardiac troponin I and T assays for the diagnosis of myocardial infarction
Z Li, Y Y Tew, A J F Thurston, D Harry, P Fineran, T Fujisawa, S Chew, R Wereski, S Jenks, A R Chapman, J Jasper Boeddinghaus, N L MillsAbstract
Background
Cardiac troponin concentrations are lower in females than males. Therefore, sex-specific 99th percentile upper reference limits are recommended to identify myocardial injury and for the diagnosis of myocardial infarction, to ensure similar performance in both sexes. We assessed equivalence of sex-specific diagnostic thresholds using high-sensitivity cardiac troponin (hs-cTn) I and T assays in an interrupted time series study cohort, where all sites initially used a hs-cTnI assay and later switched to a hs-cTnT assay.
Methods
This secondary analysis was conducted in the two periods when a hs-cTnI or a hs-cTnT assay was used to assess patients presenting with suspected acute coronary syndrome. The association between identified myocardial injury and the outcome of subsequent myocardial infarction, heart failure hospitalisation, or cardiovascular death at 1 year and was evaluated using cause-specific regression models adjusting for age, sex, history of myocardial infarction, heart failure, atrial fibrillation, ischaemic stroke, and diabetes. Index diagnoses were adjudicated according to the fourth universal definition of myocardial infarction in a subset of consecutive patients (n = 1,000 in the hs-cTnI period and n = 1,000 in the hs-cTnT period). The proportion of patients with adjudicated diagnosis was compared using chi-squared test.
Results
In the study period using the hs-cTnI assay, 21% (1,431/6,961) males (median age 69 IQR [59-79] years) and 22% (1,369/6,185) females (76 [66-84] years) were identified with myocardial injury, among whom 16% (229) and 15% (199) had previous myocardial infarction. In the period using the hs-cTnT assay, 35% (2,384/6,825) males (72 [61-81] years) and 41% (2,397/5,878) females (76 [67-83] years) were identified with myocardial injury, and 9% (211) and 7% (158) had previous myocardial infarction. Similar proportions of patients had an adjudicated diagnosis of myocardial infarction using both the hs-cTnI and the hs-cTnT assays (male: 15% versus 13%, P=0.607; female: 10% versus 12%, P=0.379; Figure 1). Those identified with myocardial injury had higher risk of subsequent myocardial infarction, heart failure hospitalisation, or cardiovascular death at 1 year using both the hs-cTnI assay (adjusted hazard ratio [aHR]: 4.28 95% confidence interval [3.43, 5.34] in males and 6.17 [4.46, 8.54] in females) and the hs-cTnT assay (aHR: 7.50 [5.32, 10.58] in males and 22.80 [9.62, 54.00] in females).
Conclusions
Sex-specific thresholds identified an equivalent proportion of males and females with myocardial injury using the hs-cTnI assay, but not the hs-cTnT assay. Similar proportion of males and females had an adjudicated diagnosis of myocardial infarction using both assays despite greater proportions were identified with myocardial injury using the hs-cTnT assay.Figure1.pngFigure2.png