Background: We investigated whether cardiac troponin T (cTnT) independently predicted outcome and added prognostic value over other clinical risk predictors in chronic peritoneal dialysis (PD) with end-stage renal disease.
Methods: Baseline cTnT, echocardiography, indices of dialysis adequacy, and biochemical characteristics were assessed in 238 chronic PD patients who were followed prospectively for 3 years or until death.
Results: Using multivariable Cox regression analysis, cTnT remained predictive of all-cause mortality [hazard ratio 4.43, 95% CI 1.87-10.45, P = 0.001], cardiovascular death (4.12, 1.29-13.17, P = 0.017), noncardiovascular death (8.06, 1.86-35.03, P = 0.005), and fatal and nonfatal cardiovascular events (CVEs) (3.59, 1.48-8.70, P = 0.005) independent of background coronary artery disease, inflammation, residual renal function, left ventricular hypertrophy, and systolic dysfunction. cTnT alone had better predictive value than C-reactive protein (CRP) alone for mortality [area under the ROC curve (AUC) 0.774 vs 0.691; P = 0.089] and first CVE (AUC 0.711 vs 0.593; P = 0.009) at 3 years. Survival models including age, sex, and clinical, biochemical, and echocardiographic characteristics yielded AUCs of 0.813 (95% CI, 0.748-0.877), 0.800 (95% CI, 0.726-0.874), and 0.769 (95% CI, 0.708-0.830), respectively, in relation to all-cause mortality, cardiovascular death, and fatal and nonfatal cardiovascular events. After addition of cTnT, AUCs of the above models increased significantly to 0.832 (95% CI, 0.669-0.894; P = 0.0037), 0.810 (95% CI, 0.739-0.883; P = 0.0036), and 0.780 (95% CI, 0.720-0.840; P = 0.0002), respectively; no AUCs increased when CRP was added.
Conclusions: cTnT is an independent predictor of long-term mortality, cardiovascular death and events, and noncardiovascular death in PD patients.