Cardiac troponin T (cTnT) and I (cTnI) levels are considered as important diagnostic tools in acute coronary events. They could be of predictive value in hemodialysis (HD) patients. The aim of this study was to determine the prevalence of increased cTnI and cTnT in HD patients and their prognostic relevance to all-cause mortality. We measured cTnT and cTnI at baseline in 145 asymptomatic HD patients. We used three different cut-off criteria to define elevated cardiac troponin levels as follows: the 99 th percentile of a reference population, the lowest concentration to give a 10% imprecision [10% coefficient of variation (10% CV)] and the relative operating characteristic (ROC) curve-determined value optimized for diagnostic sensitivity and specificity for detection of myocardial injury (MI). These concentrations were 0.01 ng/mL, 0.03 ng/mL and 0.1 ng/mL for cTnT and 0.2 ng/mL, 0.6 ng/mL and 1 ng/mL for cTnI, respectively. Patients were followed for all-cause mortality (median follow-up 551 days). Kaplan-Meier survival curves, log-rank test and Cox models were employed to determine whether baseline cTnT and cTnI levels were predictive of mortality. Greater percentages of patients had an increased cTnT versus cTnI at each cut-off as follows: 99 th percentile, 90.3% versus 35.2%; 10% CV, 73.1% versus 2.1%; and ROC, 20.7% versus 0.7%. During follow-up, 40 patients died. Elevated cTnT levels above the ROC concentration were associated with increased mortality, although it was not significant after adjustment for other risk factors. Univariate and adjusted hazard ratios were 2.3 [confidence intervals (CI), 1.2-4.5; P = 0. 01] and 1.9 (CI, 0.9-3.9; P = 0.07). No differences were found for cTnI levels. Diabetes mellitus was also an independent predictor of mortality. There is a high prevalence of positive cTnT and cTnI in asymptomatic HD patients, with a greater number of patients having an increased cTnT. Elevated troponin T, but not cTnI, seems to be associated with poor prognosis.