The incidence and risk factors of acute kidney injury after hepatobiliary surgery: a prospective observational study

Abstract

Background: Although intraperitoneal surgery is a major operation associated with postoperative acute kidney injury (AKI), the incidence, risk factors, and long-term renal outcome are not well known. We aimed to determine the risk factors and 6 months renal outcome in patients with clinical or subclinical AKI after hepatobiliary surgery. We also assessed the validity of urine neutrophil gelatinase-associated lipocalin (NGAL) in the early detection of AKI or prediction of renal outcome.

Methods: This prospective observational study enrolled patients with normal renal function who underwent hepatobiliary surgeries. Urine and serum samples were collected for NGAL measurement.

Results: Among 131 patients, 10 (7.6%) developed postoperative AKI. Urine NGAL at 12 h postsurgery was the most predictive parameter for the diagnosis of AKI (cutoff, 92.85 ng/mL). With the cutoff value, subclinical AKI was diagnosed in 42 (32.1%) patients. Patients with clinical AKI and those with subclinical AKI were assigned to the AKI group. The AKI group had significantly higher model for end-stage liver disease and sodium (MELD-Na) score, lower albumin level, and longer hospital stay after surgery than the non-AKI group. Older age and higher MELD-Na score were independent risk factors for the development of postoperative AKI. At 6 months postsurgery, the estimated glomerular filtration rate (eGFR) in the AKI group was significantly lower than that in the non-AKI group, although the baseline eGFR was not different. In multiple linear regression analysis, the maximum urine NGAL level during 24 h postsurgery, intraoperative fluid balance, and having liver transplantation were significantly associated with a poor 6 months renal outcome.

Conclusion: Urine NGAL was useful in the early diagnosis of postoperative AKI as well as in predicting the 6 months renal outcome after hepatobiliary surgery. A considerable proportion of patients developed subclinical AKI, and these patients showed worse renal outcome compared with the non-AKI group.

Figure 1

Receiver operating characteristic (ROC) curves of urine NGAL for the diagnosis of clinical AKI. The areas under the ROC curves are as follows: urine NGAL-6 h, 0.780 (95% CI 0.659-0.901, p = 0.005); urine NGAL-12 h, 0.886 (95% CI 0.803-0.969, p <0.001); urine NGAL-24 h, 0.714 (95% CI 0.549-0.880, p = 0.033).