Prolonged cardiopulmonary bypass (CPB) is recognized as a risk factor for acute renal failure (ARF), but the dose effect of time on bypass is unknown. We therefore examined the risk of ARF associated with increasing CPB time stratified by preoperative renal function.
A retrospective analysis was performed on 3889 patients undergoing cardiac surgery on CPB without circulatory arrest between 2011 and 2017 excluding those with a diagnosis of dialysis-dependent renal failure and those who had an intra-aortic balloon pump. Postoperative ARF was defined as a 3-fold increase in creatinine level, creatinine level > 4 mg/dL, or requirement for dialysis. A logistic regression model was built to identify predictors of ARF and to determine the probability of ARF.
Postoperative ARF occurred in 72 patients (2%) overall. Of 100 patients with an estimated glomerular filtration rate <30 mL/min/1.73 m2, 22% developed ARF, of which 16 required dialysis. Thirty-day mortality was 31% for those with ARF compared with <1% for those without ARF (P < .01). Risk factors for ARF included obesity (odds ratio, 3.03; P < .01), increasing preoperative creatinine level (odds ratio, 4.21; P < .01), CPB time scaled by a factor of 10 minutes (odds ratio, 1.06; P = .04), and postoperative transfusion (odds ratio, 11.94; P < .01). The adjusted probability of ARF as a function of CPB time was determined and stratified by preoperative glomerular filtration rate.
Increasing CPB duration is associated with postoperative ARF, particularly among those with preoperative renal impairment. For patients with an estimated glomerular filtration rate <30 mL/min/1.73 m2 the risk increases exponentially with time.