OBJECTIVE: To determine the relative prognostic merits of C-reactive protein (CRP), cardiac troponin T (cTnT) and N-terminal pro-brain natriuretic peptide (NT-pro-BNP) for prediction of all-cause death in patients with end-stage renal disease (ESRD) receiving haemodialysis. MATERIAL AND METHODS: This prospective, controlled cohort study included 109 patients. Biomarkers were sampled at inclusion and considered as categorical and continuous variables in Cox proportional hazard models. RESULTS: Mean follow-up ± SD was 926 ± 385 days, during which 52 patients (48%) died. All three markers were predictive of death in univariate analysis. In multivariable analysis, elevated cTnT (> 0.01 μg/l) and CRP (> 1.0 mg/dl) remained significantly associated with mortality [hazard ratio (95% confidence interval), 3.2 (1.2-8.5), p = 0.017 for cTnT; 2.0 (1.0-3.8), p = 0.032 for CRP], while NT-pro-BNP lost independent prognostic power. Addition of cTnT and CRP to established risk factors significantly improved the global fit of the model (p < 0.001), increased the c statistic from 0.726 to 0.758 and significantly increased the integrated discrimination improvement (p < 0.001). CONCLUSION: The results suggest that cTnT and CRP can be used in combination for risk stratification in patients with ESRD and highlight the additive effect they confer in this regard.
OBJECTIVE: To determine the relative prognostic merits of C-reactive protein (CRP), cardiac troponin T (cTnT) and N-terminal pro-brain natriuretic peptide (NT-pro-BNP) for prediction of all-cause death in patients with end-stage renal disease (ESRD) receiving haemodialysis. MATERIAL AND METHODS: This prospective, controlled cohort study included 109 patients. Biomarkers were sampled at inclusion and considered as categorical and continuous variables in Cox proportional hazard models. RESULTS: Mean follow-up ± SD was 926 ± 385 days, during which 52 patients (48%) died. All three markers were predictive of death in univariate analysis. In multivariable analysis, elevated cTnT (> 0.01 μg/l) and CRP (> 1.0 mg/dl) remained significantly associated with mortality [hazard ratio (95% confidence interval), 3.2 (1.2-8.5), p = 0.017 for cTnT; 2.0 (1.0-3.8), p = 0.032 for CRP], while NT-pro-BNP lost independent prognostic power. Addition of cTnT and CRP to established risk factors significantly improved the global fit of the model (p < 0.001), increased the c statistic from 0.726 to 0.758 and significantly increased the integrated discrimination improvement (p < 0.001). CONCLUSION: The results suggest that cTnT and CRP can be used in combination for risk stratification in patients with ESRD and highlight the additive effect they confer in this regard.
Authors: Ruth F Dubin; Alexis L Beatty; John R Teerlink; Nelson B Schiller; Dean Alokozai; Kirsten L Johansen Journal: Echocardiography Date: 2014-11 Impact factor: 1.724