The predictive role of ammonia in liver-related decompensation and mortality in patients with compensated cirrhosis: A retrospective cohort study

Background: Liver cirrhosis may present in different stages with distinct prognoses. Commonly used models for predicting liver-related decompensation and mortality have several limitations. Ammonia was proposed as a possible alternative or complementary tool. Our primary objective was to explore the predictive role of plasma ammonia levels regarding liver-related outcomes. Our secondary aim was to validate a new measure that corrects baseline ammonia levels to the upper limit of normal (AMM-ULN) as a predictor of liver-related outcomes. Methods: In this retrospective single-centre study, 278 patients with cirrhosis and an ammonia measurement were included. Relevant outcomes (liver-related decompensation and mortality) were evaluated in the compensated and decompensated groups, and AMM-ULN was assessed as a predictor. Results: Plasma ammonia levels and AMM-ULN were predictors of hepatic decompensation (hazard ratio HR 1.453, 95% CI 1.085–1.945, p = 0.012) and mortality (HR 1.010, 95% CI 1.003–1.017, p < 0.001) at 12 months in the compensated cirrhosis group. These parameters were not predictors of further decompensation and mortality in already decompensated patients. AMM-ULN ≥ 1.4 identified patients at higher risk of liver-related decompensation ( p = 0.009) and mortality ( p = 0.020) at 12 months in the compensated cohort. Conclusions: Both ammonia levels and AMM-ULN were predictors of hepatic decompensation and liver-related mortality in compensated patients at 12 months of follow-up.