Renal Function Association With In-Hospital and 30-Day Mortality in Patients With Alcoholic Hepatitis

Abstract

Introduction: Renal function, a component of model for end-stage disease (MELD) score is known to impact the outcome of patients with cirrhosis. Modification of diet in renal disease (MDRD)-6 using age, gender, race, serum creatinine, albumin, and blood urea nitrogen (BUN) is superior to MDRD-4 (all variables as in MDRD-6 except albumin and BUN) and creatinine in estimating renal function in cirrhosis. However, data are lacking comparing accuracy of serum creatinine, MDRD-4, and MDRD-6 in predicting severity and outcome of patients with alcoholic hepatitis (AH). Methods: Medical charts were reviewed for patients admitted to 1 tertiary care hospital (2004-2013) with a discharge diagnosis of AH (ICD-9 code 571.1). Patients meeting criteria for AH diagnosis and with total bilirubin >5 mg/dL formed the study population. Chi-squared and t-tests were used for comparing categorical and continuous variables respectively. Multivariate logistic regression was used to compare MDRD-4, MDRD-6, and serum creatinine as predictors of in-hospital (IHM) and 30-day mortality. Results: A total of 105 AH patients (mean age 48±10 years, 66% males, 81% white, 56% underlying cirrhosis) were reviewed. About 79% had severe AH with discriminant function (DF) ≥32 and/or hepatic encephalopathy (HE). About 50% patients received treatment with steroids or pentoxyfylline or both, and 16% had infections. Overall in-hospital mortality and 30-day mortality was observed in 26 (25%) and 51 (49%) patients, respectively. Non-survivors compared to 79 survivors during the average hospital stay of 15 days, differed for age, HE (100 vs. 32%, P<0.0001), underlying cirrhosis (64 vs. 32%, p=0.02), white cell count (16±10 vs. 12±8, P=0.02), DF (68±27 vs. 50±33, P=0.01), and MELD (30±8 vs. 24±10, P=0.006). Renal function at admission was lower among non-survivors but not statistically different for creatinine (2.3±1.9 vs. 1.6±1.8, p=0.12), MDRD-4 (59±42 vs. 80±52, p=0.06), and MDRD-6 (56±49 vs. 75±52, p=0.1). However, acute kidney injury (AKI) during hospitalization occurred in 52 (50%) patients. On multivariate logistic regression adjusting for age, sex, cirrhosis, DF, and treatment; creatinine, MDRD-4, or MDRD-6 were neither predictors nor improved the model accuracy for IHM with c-statistics of 0.733, 0.732, and 0.729, respectively. Data were similar for 30-day mortality with respective c-statistics of 0.693, 0.675, and 0.688. However, adjusting for admission creatinine, development of AKI was associated with 9-fold risk of IHM with OR 9.0 (95% CI 1.6-50.4). Conclusion: AKI during hospitalization and not admission renal function predicts outcome of AH patients. Studies are suggested to develop strategies to prevent AKI among hospitalized patients with AH.