Incidence, risk factors and outcomes of acute kidney injury (AKI) in patients with acute-on-chronic liver failure (ACLF) of underlying cirrhosis.Hepatol Int 2016; 10(5):807-18HI
BACKGROUND AND AIMS
Acute kidney injury (AKI) is a life-threatening complication in patients with acute-on-chronic liver failure (ACLF) of underlying cirrhosis. However, the characteristics of AKI in these patients have not been clarified. Our aim was to determine the incidence and risk factors of AKI and the association between AKI severity and 180-day transplant-free survival.
We performed a retrospective cohort analysis of patients with ACLF of underlying cirrhosis in a single center from January 2009 through December 2014. AKI was defined by the criteria proposed by International Club of Ascites (ICA). The incidence and risk factors of AKI development and its relationship to 180-day transplant-free survival rates were evaluated.
Of 1032 patients with ACLF of underlying cirrhosis, 121 (11.72 %) had AKI at admission, and 319 (30.9 %) developed AKI during hospitalization. We established a logistic regression model including four independent factors with AKI development: MELD score [odds ratio (OR), 1.1; 95 % confidence interval (CI), 1.07-1.14], presence of ascites (OR, 3.80; 95 % CI, 2.13-6.78), sepsis/infection (OR, 2.25; 95 % CI, 1.66-3.03) and acute variceal bleed (OR, 1.78; 95 % CI, 1.00-3.19). The area under receiver operating characteristics of the model in internal and external validations were 0.95 and 0.85, respectively. Patients with mild-A AKI had a higher 180-day transplant-free survival rate (23.8 %) than patients with mild-B AKI (19.0 %) or marked AKI (5.9 %) (all p < 0.001). AKI patients with a peak value of sCr <1.5 mg/dl had higher 180-day transplant-free survival rates compared to those with a peak value of sCr ≧1.5 mg/dl (23.8 % vs. 14.7 %, p < 0.001).
We developed a clinical risk model for predicting development of AKI with great accuracy. Combining the ICA-AKI criteria and the peak value of sCr with 1.5 mg/dl provides a good prognostic method for patients with ACLF of underlying cirrhosis.