Background: B-type natriuretic peptide (BNP) may be a predictor of stroke risk in patients with nonvalvular atrial fibrillation (NVAF); because heart failure is associated with the incidence of stroke in AF patients. However, limited data exist regarding the association between BNP at baseline and risks of thromboembolic events (TE) and death in NVAF patients.
Methods and results: We prospectively studied 1,013 NVAF patients (725 men, 72.8±9.7 years old) from the Hokuriku-plus AF Registry to determine the relationship between BNP at baseline and prognosis among Japanese NVAF patients. During the follow-up period (median, 751 days); 31 patients experienced TE and there were 81 cases of TE/all-cause death. For each endpoint we constructed receiver-operating characteristic curves that gave cutoff points of BNP for TE (170 pg/mL) and TE/all-cause death (147 pg/mL). Multivariate analysis with the Cox-proportional hazards model indicated that high BNP was significantly associated with risks of TE (hazard ratio [HR] 3.86; 95% confidence interval [CI] 1.83-8.67; P=0.0003) and TE/all-cause death (HR 2.27; 95% CI 1.45-3.56; P=0.0003). Based on the C-index and net reclassification improvement, the addition of BNP to CHA2DS2-VASc statistically improved the prediction of TE.
Conclusions: In a real-world cohort of Japanese NVAF patients, high BNP was significantly associated with TE and death. Plasma BNP might be a useful biomarker for these adverse clinical events.
Keywords: Atrial fibrillation; B-type natriuretic peptide; Thromboembolism.