Abstract 17271: High-Sensitive Cardiac Troponin-I and NT-proBNP Improve Predictive Ability of Risk Stratification Model for Sudden Cardiac Death in End-Stage Renal Disease
Introduction: Sudden cardiac death (SCD) is the single most common cause of death in end-stage renal disease (ESRD) patients, and accounts for up to 30% of all deaths. However, there is currently no accurate tool for risk stratification. Although some biomarkers, including high-sensitive troponin I (hsTnI) and NT-proBNP, have been shown to be a prognostic predictor, there is no data on whether these biomarkers are independent predictors for SCD, and whether these biomarkers have incremental value for predicting future SCD.
Hypothesis: Evaluation of hsTnI and NT-proBNP would add incremental prognostic value for SCD in ESRD patients.
Methods: We measured serum hsTnI and NT-proBNP in 301 asymptomatic ambulatory ESRD patients simultaneously just before dialysis. All patients were prospectively followed-up for SCD defined by Hinkle-Thaler definition or all deaths. We used SCD predicting model recently constructed from Hemodialysis Study (HEMO model), and the predictive abilities by adding hsTnI and NT-proBNP to this model for predicting future SCD were evaluated by c-statistics in receiver operating characteristic curve analysis.
Results: The mean age of our patients was 67.4 years old, and 63.1% were male. At baseline, hsTnI was detectable in 92.0% of all patients, and the median hsTnI and NT-proBNP levels were 13 pg/ml (IQR: 6-29) and 4476 pg/ml (IQR: 2184-10666), respectively. There were 21 SCDs during 2.5 years follow-up. Logistic regression analysis indicated that both hsTnI (OR 1.12, 95% CI: 1.05-1.20, per 10 pg/ml increase) and log-transformed NT-proBNP (OR 1.73, 95% CI: 1.07-2.78) were independent predictors for SCD even after adjusted by HEMO model. C-statistics for SCD prediction by HEMO model only, HEMO model + NT-proBNP, and HEMO model + NT-proBNP + hsTnI were 0.73 (95% CI: 0.62-0.84), 0.80 (95% CI: 0.71-0.89), and 0.86 (95% CI: 0.77-0.94), respectively. With comparison by DeLong’s method, c-statistic of HEMO model + NT-proBNP + hsTnI was significantly superior to both HEMO model only and HEMO model + NT-proBNP (P=0.003 and 0.013, respectively).
Conclusion: NT-proBNP and hsTnI predict future SCD independently from existing risk factors, and evaluating these cardiac biomarkers can improve predictive ability for future SCD in ESRD patients.
- © 2013 by American Heart Association, Inc.