Correlation Between Bilirubinemia and Acute Kidney Injury in Patients with Cirrhosis

Document Type

Conference Proceeding

Publication Date


Publication Title

J Gen Intern Med


Background: Acute kidney injury (AKI) in the setting of cirrhosis is associated with significant morbidity. However, data is limited regarding the role that hyperbilirubinemia plays in renal injury. One proposed mechanism is direct renal tubular injury from bilirubin and bile salts, but this is not well elucidated. Using a tertiary care population, we aimed to evaluate whether serum bilirubin levels predict the likelihood of AKI in patients with cirrhosis, as well as predict the probability of renal recovery in this population. Methods: Hospitalized patients with cirrhosis, with and without AKI, were retrospectively identified from January 2012-December 2015. AKI was defined per KDIGO criteria and classified into HRS type 1, HRS type 2, and/or acute tubular necrosis (ATN). Other etiologies of AKI were excluded. Data concerning patient demographics, MELD-Na score, peak total serum bilirubin during hospi-talization, change in total serum bilirubin from baseline, peak serum creatinine (SCr) during hospitalization, change in SCr from baseline, and nadir estimated glomerular filtration rate (eGFR) were analyzed. In patients with AKI, renal recovery was evaluated after 6 months of follow up and was defined as a return of SCr towards baseline or a lack of dialysis dependence. Results: A total of 200 cirrhotic patients were included. The majority were Caucasian (70%), male (62%), and had alcoholic cirrhosis (61%) as the etiology of their liver disease. Of these, 57% had AKI, while 43% did not. The average peak total bilirubin was significantly higher in patients with AKI compared to patients without AKI (13.6 mg/dL vs. 5.0 mg/dL respectively; p< 0.001). The change in bilirubin from baseline was significantly higher in patients with AKI compared to patients without AKI (8.7 mg/dL vs. 2.5 mg/dL respectively; p< 0.001). Patients' peak bilirubin level also correlated with their peak SCr (p< 0.001), change in SCr (p=0.004), and nadir eGFR (p=0.004) during hospitalization. There was no significant difference in peak bilirbuin levels between HRS type 1, HRS type 2, and ATN. While peak bilirubin level was unable to predict the likelihood of renal recovery among patients with AKI, a lower MELD-Na score was predictive of renal recovery (p=0.015). Conclusions: Higher bilirubin levels predicted both an increased likelihood of and severity of AKI, manifested as both higher SCr and lower nadir eGFR. Elevated bilirubin also correlated with the development of both HRS and ATN, suggesting that hyperbilirubinemia may play a role in the pathogenesis of AKI, irrespective of etiology. Measured bilirubin did not predict renal recovery in cirrhotic patients with AKI. Further data on the association between hyperbilirubinemia and AKI can guide clinicians to risk stratify and plan initiation of dialysis. Identification of the pathogenic mechanisms may also lead to directed therapy.





First Page


This document is currently not available here.