Summary
    Aims Several high-sensitivity cardiac troponin (hs-cTn) assays have recently been developed. It is unknown which hs-cTn provides the most accurate prognostic information and to what extent early changes in hs-cTn predict mortality. Methods and results In a prospective, international multicentre study, cTn was simultaneously measured with three novel [high-sensitivity cardiac Troponin T (hs-cTnT), Roche Diagnostics; hs-cTnI, Beckman-Coulter; hs-cTnI, Siemens] and a conventional assay (cTnT, Roche Diagnostics) in a blinded fashion in 1117 unselected patients with acute chest pain. Patients were followed up 2 years regarding mortality. Eighty-two (7.3%) patients died during the follow-up. The 2-year prognostic accuracy of hs-cTn was most accurate for hs-cTnT [area under the receivers operating characteristic curve (AUC) 0.78 (95% CI: 0.73-0.83) and outperformed both hs-cTnI (Beckman-Coulter, 0.71 (95% CI: 0.65-0.77; P = 0.001 for comparison), hs-cTnI (Siemens) 0.70 (95% CI: 0.64-0.76; P < 0.001 for comparison)] and cTnT 0.67 (95% CI: 0.61-0.74; P < 0.001 for comparison). Absolute changes of hs-cTnT were more accurate than relative changes in predicting mortality, but inferior to presentation values of hs-cTnT. Combining changes of hs-cTnT within the first 6 h with their presentation values did not further improve prognostic accuracy. Similar results were obtained for both hs-cTnI assays regarding the incremental value of changes. Hs-cTn concentrations remained predictors of death in clinically challenging subgroups such as patients with pre-existing coronary artery disease, impaired renal function, and patients older than 75 years. Conclusion High-sensitivity cardiac Troponin T is more accurate than hs-cTnI in the prediction of long-term mortality. Changes of hs-cTn do not seem to further improve risk stratification beyond initial presentation values