N-terminal pro-B-type natriuretic peptide is a specific predictor of appropriate device therapies in patients with primary prevention implantable cardioverter-defibrillators.

ABSTRACT

BACKGROUND: Sudden death risk stratification of patients with left ventricular systolic dysfunction remains challenging. Retrospective studies have suggested N-terminal pro-B-type natriuretic peptide (NT-proBNP) may be a useful risk stratification tool. OBJECTIVE: The purpose of this study was to ascertain the utility of NT-proBNP as a predictor of appropriate implantable cardioverter-defibrillator (ICD) therapies in primary prevention ICD recipients. METHODS: This was a prospective study of 342 stable patients with left ventricular ejection fraction ≤40% who received a primary prevention ICD. NT-proBNP assay was performed at the time of device implant and used as a dichotomized variable (1st-3rd NT-proBNP quartiles vs 4th NT-proBNP quartile) to predict primary (appropriate ICD therapies) and secondary (death, ICD-deactivation, chronic inotropic support, transplant) outcomes. RESULTS: Median follow-up was 35.0 months (interquartile range 15.2-55.3). In unadjusted analyses, NT-proBNP predicted both primary (hazard ratio [HR] 1.89; 95% confidence interval [CI] 1.00-3.56); P = .049) and secondary outcomes (HR 2.13; 95% CI 1.18-3.85; P =.012). Multivariable analysis reaffirmed NT-proBNP as a primary outcome predictor (HR 4.31; 95% CI 1.92-9.70; P <.001) but not as a secondary outcome predictor (HR 1.23; 95% CI 0.61-2.50; P = .564). Instead, secondary outcome was predicted by patient age and renal function. In an unadjusted subanalysis limited to patients with blood urea nitrogen <30 mg/dL, NT-proBNP remained a primary endpoint predictor (HR 2.51; 95% CI 1.25-5.05; P = .010) but not a secondary endpoint predictor (HR 1.34; 95% CI 0.52-3.44; P = .541). Receiver operating analyses at 2- and 3-year follow-up timepoints confirmed that NT-proBNP significantly improved the performance of multivariable models designed to predict future appropriate ICD therapies. CONCLUSION: In multivariable analysis, NT-proBNP is a reasonable and specific predictor of future appropriate device therapies in primary prevention ICD recipients. In contrast, adjusted NT-proBNP does not predict all-cause mortality.