© The Author(s) 2016. Published by Oxford University Press and Sixth Affiliated Hospital of Sun Yat-Sen University. Objective. Alcoholic hepatitis (AH), a unique clinical syndrome among patients with chronic and active alcohol use, is associated with high short-term mortality. An elevated ammonia level is associated with mortality in patients with acute liver failure; however, its impact in AH has not been well-studied. Methods. A retrospective study was performed on patients admitted to a tertiary-care hospital with the discharge diagnosis of AH. Patients meeting criteria for AH were included in the final data analysis. Multivariate logistic regression models were built to examine the impact of serum ammonia in predicting in-hospital mortality (IHM) and 30-day mortality (TDM). Subgroup analysis was also performed, which was limited to patients who had hepatic encephalopathy. Results. Of the 105 AH patients included, 26 (25%) died during the initial hospitalization. Among the 79 patients who survived initial hospitalization, 30 (39%) died within 30 days. Information about ammonia levels at admission was available for 82 patients. Of these, 25 patients had IHM and significantly higher ammonia level 97 vs. 69 μmol/L, P < 0.01). Among the 57 who survived hospitalization, ammonia levels were not significantly different (71 vs. 67 lmol/L, P = 0.69) in patients with and without TDM. The addition of ammonia to the multivariate regression models including age, sex, cirrhosis, treatment and model for end-stage liver disease (MELD) score improved the C statistics for IHM from 0.708 to 0.801 and for TDM from 0.756 to 0.766, respectively. These results were identical, even when limited to patients with hepatic encephalopathy. Conclusion. AH patients with elevated ammonia levels at admission have higher IHM; however, they do not seem to play a significant role in 30-day mortality for patients who survived hospitalization.