Background: Postoperative (postop) atrial fibrillation (AF) occurs in up to 60% of patients after cardiac surgery, leading to longer hospital stays and increased healthcare costs. Recently, B-type natriuretic peptide (BNP) has been reported to predict occurrence of nonpostoperative AF. This study evaluates whether elevated preoperative (preop) plasma BNP levels predict the occurrence of postop AF.
Methods and results: One hundred eighty-seven patients with no history of atrial arrhythmia who had a preoperative BNP level and had undergone cardiac surgery were identified. Their records were reviewed, and postoperative ECG and telemetry strips were analyzed for AF until the time of discharge. Postop AF was documented in 80 patients (42.8%). AF patients were older (68+/-11 versus 64+/-14 years, P=0.04), but there was no difference in sex distribution, hypertension, left ventricular (LV) function, LV hypertrophy (LVH), left atrial size, history of coronary artery disease (CAD), or beta-blocker use. Preop plasma BNP levels were higher in the postop AF patients (615 versus 444 pg/mL, P=0.005). After adjustment for age, sex, type of surgery, hypertension, LV function, LVH, left atrial size, CAD, and beta-blocker use, the odds ratios of postop AF according to increasing quartiles, compared with patients with lowest quartile, were 1.8, 2.5, and 3.7 (P(trend)=0.03).
Conclusions: An elevated preop plasma BNP level is a strong and independent predictor of postop AF. This finding has important implications for identifying patients at higher risk of postop AF who could be considered for prophylactic antiarrhythmic or beta-blocker therapy.