Background: Debate surrounds the interpretation of troponin assays for the diagnosis and prognosis of cardiac disease in patients with renal failure.
Objectives: To systematically review the diagnostic and prognostic test characteristics of quantitative serum cardiac troponin I (cTnI) and T (cTnT) in renal failure patients without acute coronary syndrome (ACS) symptoms.
Methods: English-language literature was identified through searching MEDLINE from 1966 to August 2003 and reviewing reference lists. Studies were excluded if they did not meet research objectives, had fewer than 10 patients or focused primarily on nonrenal patients. Of 119 potential studies, 39 articles with over 349 patients with chronic kidney disease (CKD) and 3899 hemodialysis patients were selected for abstraction.
Results: Among CKD and hemodialysis patients without ACS symptoms, cTnI had a mean specificity of 97% (95% CI 93% to 99%) and 96% (95% CI 94% to 98%), respectively, using the myocardial infarction cut-off threshold. The mean specificity of cTnT compared less favourably at 85% (95% CI 75% to 93%) and 71% (95% CI 64% to 77%) for CKD and hemodialysis patients, respectively. In hemodialysis patients without ACS symptoms, positive and negative likelihood ratios for all-cause mortality over 12 to 24 months for cTnT were 4.5 (95% CI 2.9 to 7.1) and 0.6 (95% CI 0.4 to 0.8), and for cTnI were 1.6 (95% CI 0.9 to 2.9) and 1.0 (95% CI 0.9 to 1.1), respectively.
Conclusions: In CKD and hemodialysis patients without ACS symptoms, troponin I, at the myocardial infarction cut-off threshold, is unlikely to be falsely elevated. Among hemodialysis patients without ACS symptoms, a positive troponin T helps predict all-cause mortality.