B-type natriuretic peptide following thoracic surgery: a predictor of postoperative cardiopulmonary complications

B-type natriuretic peptides (BNPs) are secreted by the human heart in response to ventricular wall stretch or myocardial ischaemia, and predict adverse cardiovascular events and death in the general population. Following non-cardiac surgical procedures, there is growing evidence supporting BNP measurement as a powerful independent predictor of death and perioperative complications. However, the clinical implication of elevated BNP measurements after pulmonary resection has not been completely defined. This study aimed to evaluate the role of BNP in predicting adverse cardiopulmonary events after thoracic surgery. A prospective, short-term, observational cohort study was conducted in a tertiary care hospital, including consecutive patients undergoing scheduled pulmonary resection between April 2012 and October 2013. Baseline clinical details were obtained; serum BNP levels were measured at baseline and on postoperative days 1 and 4. We enrolled 294 consecutive patients, median age 66 [interquartile range (IQR): 57-73], 67% male. There were 2 perioperative deaths, and 52 patients experienced one or more cardiopulmonary complications. The baseline median BNP value was normal (29.5 pg/ml, IQR: 16-57.2), and showed significant postoperative increase, peaking on day 1. Patients who developed postoperative complications had a significantly greater BNP increase (P < 0.0001) when compared with those without complications. A postoperative day 1 BNP measurement of a parts per thousand yen118.5 [receiver operating characteristic area: 0.654; 95% confidence interval (CI): 0.57-0.74; P = 0.001] was associated with a 3-fold risk of developing postoperative cardiopulmonary complications [odds ratio (OR): 2.94; 95% CI: 1.32-6.57; P = 0.008]. Logistic regression analysis showed major pulmonary resections (lobectomies or pneumonectomies), BNP a parts per thousand yen 118.5 and age a parts per thousand yen 65 to be the only independent predictive variables. In the subset of patients undergoing lobectomy or pneumonectomy (n = 226), BNP was the strongest independent predictor of complications (OR: 3.49; 95% CI: 1.51-8.04). Our results show that BNP elevation, measured in the first days after thoracic surgery, is independently associated with postoperative adverse events. In patients undergoing major pulmonary resections, a postoperative BNP elevation is the strongest independent predictor of cardiopulmonary complications.