Objectives: The influence of the cardiac surgical procedure on B-type natriuretic peptide (BNP) for the identification of high-risk patients has not been evaluated. This study aimed to compare the prognostic utility of pre- and postoperative BNP in predicting adverse long-term outcome after coronary artery bypass graft (CABG) surgery and aortic valve replacement (AVR).
Design: A retrospective study.
Setting: A university teaching hospital.
Participants: One hundred eighty-nine patients.
Measurements and main results: Preoperative, early postoperative (24 hours), and late postoperative (day 5) BNP levels were measured. Major adverse cardiac events (MACEs) within 12 months after surgery were chosen as study endpoints. The predictive abilities of BNP measurements were compared using receiver operating characteristic (ROC) curves. Patients were stratified by CABG surgery (n = 100) and AVR (n = 89). Thirty-four (18%) patients experienced 44 MACEs over the study period. Preoperative BNP values were significantly different between groups. Postoperative BNP gradually increased by 431% on day 5 after CABG surgery and by 100% after AVR (both p < 0.001 v preoperative values). Pre- and early postoperative BNP values were accurate in predicting MACEs after AVR (areas under the ROC curves: 0.78 [95% confidence interval, 0.66-0.90] and 0.76 [95% confidence interval, 0.62-0.89], respectively) and inaccurate after CABG surgery (0.54 [95% confidence interval, 0.38-0.70] and 0.54 [95% confidence interval, 0.36-0.73], respectively). The late postoperative BNP value was of limited value.
Conclusions: BNP measurements should take into account the type of cardiac surgery. Whatever the time of measurement, BNP accurately predicts long-term adverse outcome in valve surgery patients. A late postoperative BNP measurement is useless after cardiac surgery.
Copyright © 2012 Elsevier Inc. All rights reserved.