RESUMO
OBJECTIVE: The study sought to compare the clinical performance of two more sensitive cardiac troponin (cTn) assays, a novel high-sensitivity (hs) troponin T assay and a contemporary cTnI assay. METHODS: We measured hs-cTnT (Roche TnThs) and cTnI (Siemens Centaur Ultra) on presentation in 1,384 patients with suspected acute coronary syndrome (ACS) who underwent early invasive strategy within 24 h after presentation. Kaplan-Meier, Cox proportional hazards, and receiver-operating characteristic (ROC) analysis was used to compare their prognostic performance for the prediction of all-cause death and death/MI (myocardial infarction) after a median of 271 days. We also compared the diagnostic performance of these assays on presentation for early diagnosis of non-STEMI. RESULTS: Both hs-cTnT and cTnI were independently predictive of long-term death (OR 3.51 vs. 2.19) and the composite of death/MI (OR 9.24 vs. 3.61), across the spectrum of ACS and in patients without ACS. When used as a continuous variable, ROC analysis demonstrated significantly higher areas under the curve (AUC) for hs-cTnT as compared to cTnI for the prediction of death/MI (0.721 vs. 0.672, P = 0.024), a trend to better prediction of all-cause death (0.721 vs. 0.672, P = 0.093) and significantly higher AUC for early diagnosis of non-STEMI (0.965 vs. 0.901, P < 0.001). CONCLUSION: Using the 99th percentile cutoff for hs-cTnT and cTnI, both assays enable prediction of adverse long-term outcomes and earlier diagnosis of non-STEMI. Used as a continuous variable, the hs-cTnT assay showed superior performance compared to the cTnI assay, especially in regard to prognosis.
