Superior Predictive Value for NTproBNP Compared with High Sensitivity cTnT in Dialysis Patients: A Pilot Prospective Observational Study

Background/Aims: The clinical utility of the new biomarker, high sensitivity cardiac T troponin (hs-cTnT) is still unclear in dialysis patients. Furthermore, the prognostic value of combining N-terminal pro-B-type natriuretic peptide (NT-pro-BNP) and hs-cTnT has not been explored so far. The objective of this pilot study was to determine the utility of hs-cTnT alone versus hs-cTnT in combination with NT-proBNP for predicting death in a stable hemodialysis cohort. Methods: A prospective observational pilot study including 98 chronic asymptomatic hemodialysis patients with a follow up period of 24 months was designed. The cut-off values for NT-proBNP and hs-cTnT were calculated using receiver operating characteristic (ROC) analysis, using mortality as an end-point. Based on the cut-off values, the cohort was divided into four groups. Group 1 -NT-proBNP < 14275 pg/ml and hs-cTnT < 69.48 ng/l; group 2 -NT-proBNP < 14275 pg/ml and hs-cTnT > 69.48 ng/l; group 3 -NT-proBNP > 14275 pg/ ml and hs-cTnT < 69.48 ng/l; group 4 -NT-proBNP > 14275 pg/ml and hs-cTnT > 69.48 ng/l. Survival for each group was determined using the Kaplan-Meier method and Cox regression analysis. Results: During the follow-up period 16 patients died. According to the ROC curves analysis, the cut-off point for hs-cTnT and for NT-proBNP were 69.43 ng/l (AUC = 0.618; p = 0.04) and 14275 pg/ml (AUC = 0.722; p = 0.003), respectively. In univariate Cox analysis, both hs-cTnT (HR = 3.34; p = 0.016) and NT-proBNP (HR = 5.94; p = 0.01) were predictors of death. In the multivariable Cox proportional hazards model, only NT-pro-BNP levels above the cutoff value remained an independent predictor of all-cause mortality. The combined elevation of both biomarkers did not improve significantly the prognostic value compared with NT- proBNP alone (HR = 6.15 versus HR = 4.78; p = 0.338). Conclusion: NT-pro-BNP is a strong predictor of overall mortality in asymptomatic hemodialysis patients. The addition of hs-cTnT did not improve the prognostic accuracy compared with NT proBNP alone. Copyright (C) 2014 S. Karger AG, Basel