S1161 Relationship of Ammonia Level With Mortality in Patients With Cirrhosis

Abstract

INTRODUCTION: Elevated ammonia level is a frequent lab finding in patients with liver cirrhosis. The clinical utility of measuring ammonia levels is controversial. Recent studies have shown the prognostic role of ammonia level in cirrhotic patients, suggesting an association with multi-organ failure, and increased mortality. This study aimed to evaluate the predictive value of ammonia level in overall all-cause mortality. METHODS: This is a retrospective single-center cohort study at our tertiary care hospital, which included patients with end-stage liver disease (ESLD) admitted between January 2016 and December 2019. The study included patients between the ages of 20–90 years who were admitted with cirrhosis complications. A venous ammonia level recorded at the time of admission. Patients with acute liver failure, advanced hepatocellular carcinoma, ischemic hepatitis, or other malignancies were excluded. We identified 281 patients who met the inclusion criteria. Data collection included etiology of cirrhosis, Intensive Care Unit (ICU) admission status, laboratory studies, a model for end-stage liver disease (MELD) score, length of hospital stay, and date of death. Grade of Hepatic Encephalopathy (HE) was defined by West Haven Criteria. Logistic regression with odds ratio (ORs) and 95% confidence intervals (CIs) were used to investigate the association between admission ammonia level to the 28-day all-cause mortality. RESULTS: There were 118 (42%) females and 163 (58%) male patients with a mean age of 58.2 ± 11.6. In this cohort, the most common etiology of ESLD was secondary to alcohol (33%), followed by non-alcoholic fatty liver disease (21%). The mean ammonia level on admission was 95.6 µ/dL ± 99.8 µ/dL. At the time of hospital admission, 57 patients (20%) got admitted to the ICU. At the initial evaluation, 141(50%) patients had Grade 1 HE, 93(33%) patients had Grade 2 HE, and 47(17%) patients had Grade 3 or 4 HE. The mean MELD-Na at the time of admission was 21.5 ± 7.0. Logistic regression analysis showed that plasma ammonia did not correlate with the overall all-cause mortality at 28-days (Odds Ratio 1.00[0.999–1.005], P = 0.07) [Tables 1 and 2]. CONCLUSION: Despite recent literature about ammonia association with increased mortality, our single-center retrospective study did not find any clear association between high serum ammonia levels and mortality. Further large-scale multicenter prospective studies are needed to confirm the validity.Table 1.: Demographics and laboratory findings of our patient population