We investigated and compared serum levels of cardiac troponins I(cTnI) and cardiac troponin T (cTnT) in 85 renal patients (chronic renal impairment n = 23, continuous ambulatory peritoneal dialysis n = 20, hemodialysis n = 42). Patients with the following conditions were excluded: myocardial infarction, angina pectoris, liver disease, malignant neoplasms, enforced physical activity, skeletal muscle trauma, myositis, rhabdomyolysis and seizures. Troponin T was measured by the second generation cTnT-ELISA with a cutoff value = 0.1 μg/l. Troponin I was measured by a cTnI immunoassay analyser with a cut-off value = 2.0 μg/l. Additionally, creatine kinase (CK), CK-MB activity, CKMB mass concentration and myoglobin levels were measured. Specificity was determined as the fraction of true-negative cases compared to the total number of false-positive and true-negative cases. Specificity for cTnT was 96 % [78–100] in patients with renal impairment (creatinine > 150 μmol/l), 95 % [75–100] in continuous ambulatory peritoneal dialysis patients, but in hemodialysis patients it was 75 % [53–92] for short-term hemodialysis (< 1 year) and 46 % [24–68] for long-term hemodialysis (> 1 year). There was a weak correlation between cTnT levels and duration of hemodialysis therapy (r = 0.35, n = 34, p < 0.04). Specificity for cTnI in renal impairment patients was 96 % [78–100] and 100 % [84–100] in continuous ambulatory peritoneal dialysis and all hemodialysis patients. None of the studied markers showed higher specificity than cTnI. Only myoglobin was less specific than cTnT in hemodialysis patients. Different clearances of the troponins during dialysis (investigated by pre-hemodialysis and post-hemodialysis levels) cannot explain the discordant results of cTnT and cTnI. Conclusion: Cardiac troponin I exhibits higher specificity than cardiac troponin T in hemodialysis patients. Uremic myopathy could explain falsely elevated troponin T levels in hemodialysis patients.