Abstract
Background Acute kidney injury (AKI) is associated with poor outcomes after non-cardiac surgery. Whether preoperative N-terminal pro-B-type natriuretic peptide (NT-proBNP) predicts AKI after non-cardiac surgery is unclear. Methods We conducted a retrospective study on patients whose NT-proBNP concentrations were measured before non-cardiac surgery at a tertiary academic hospital between 2008 and 2018. Multivariable logistic regression was used to investigate the predictive role of preoperative NT-proBNP on postoperative AKI defined by the Kidney Disease: Improving Global Outcomes creatinine criteria. Results Some 6.1% (444 of 7248) of patients developed AKI within 1 week after surgery. Preoperative NT-proBNP was an independent predictor of AKI after adjustment for clinical variables (odd ratio comparing top to bottom quintiles 2.29, 95% confidence interval [CI] 1.47-3.65, p<0.001 for trend; odd ratio per 1-unit increment in natural log transformed NT-proBNP 1.27, 95% CI 1.16-1.39). Compared with clinical variables alone, the addition of NT-proBNP modestly improved the discrimination (change in area under the curve from 0.764 to 0.773, p=0.005) and reclassification (continuous net reclassification improvement 0.210, 95% CI 0.111-0.308; integrated discrimination improvement 0.0044, 95% CI 0.0016-0.0072) of AKI and non-AKI cases. Conclusions Preoperative NT-proBNP concentrations provided predictive information for AKI in a cohort of patients undergoing non-cardiac surgery, independent of and incremental to conventional risk factors. Prospective studies are required to confirm this finding and examine its clinical impact.