B-type natriuretic peptide predicts left atrial appendage thrombus in patients with nonvalvular atrial fibrillation.

ABSTRACT

PURPOSE: To investigate whether plasma B-type Natriuretic peptide (BNP), a surrogate of left ventricular filling pressure (LVFP), is predictive of left atrial appendage thrombus (LAAT) in patients with nonvalvular atrial fibrillation (AF) independent of known clinical risk predictors. METHODS: We conducted a retrospective cohort study of 297 consecutive subjects with AF who underwent a clinically indicated transesophageal echocardiogram (TEE) to evaluate for LAAT and spontaneous echo contrast (SEC). Among those, 136 had a clinically indicated BNP level. Using multivariate logistic regression analysis models, we determined factors independently predictive of the primary endpoint of LAAT and the secondary endpoint of either LAAT or SEC. RESULTS: Nineteen subjects (6.4%) had LAAT and they were found to have a higher mean CHADS2 score (2.53 vs 1.76, P = 0.01) and mean BNP level [1949 vs. 819 pg/mL, P = 0.001] than those without LAAT. None of the patients with a BNP level ≤500 pg/mL had LAAT. Multivariate logistic regression analysis demonstrated that BNP was predictive of LAAT and the composite of LAAT/SEC independent of the CHADS2 score and warfarin therapy [OR = 1.23 and 1.6 per 500 pg/mL increment in BNP, P-values = 0.03 and 0.001; respectively]. Moreover, adding BNP to the predictive model negated the influence of the CHADS2 score. CONCLUSION: These data indicate that BNP is an independent predictor of LAAT in AF and may complement the role of the CHADS2 score in predicting stroke risk.