Time course of compromised urea synthesis in patients with alcoholic hepatitis

Abstract

Objectives: Alcoholic hepatitis (AH) markedly decreases the urea synthesis capacity. We aimed to investigate the time course of this compromised essential liver function in patients with AH and its relation to treatment and survival.

Materials and methods: Thirty patients with AH were included in a prospective cohort study. We measured the substrate-independent urea synthesis capacity, i.e., the functional hepatic nitrogen clearance (FHNC), in the patients at study entry and again at three months (survivors/available: n = 17). Patients with severe disease (Glasgow Alcoholic Hepatitis Score ≥9, n = 17) were randomized to receive either prednisolone or pentoxifylline and were in addition examined after 14 days (n = 9).

Results: FHNC (normal range = 25–45 L/h) was markedly decreased at study entry (median = 5.6 (IQR = 3.0–9.6) L/h) and increased by three-fold in survivors at three months (15.1 (12.0–22.9) L/h; p < .001). In patients with severe AH, FHNC was also increased after 14 days of pharmacologic treatment and showed the greatest increase in the patients taking prednisolone (prednisolone 25.4 (20.6–26.2) L/h vs. pentoxifylline 12.3 (8.0–15.3) L/h; p = .05). FHNC at study entry was lower in 90-day non-survivors than in survivors (p = .04).

Conclusions: The decrease in the urea synthesis capacity in patients with AH was the most marked in short-term non-survivors and partly recovered in survivors at three months. In patients on pharmacologic treatment, recovery was observed already after 14 days, and it was nearly complete in those on prednisolone. Thus, metabolic liver failure in AH seems to be prognostically important, is potentially reversible, and may recover more rapidly following treatment with prednisolone.

Keywords:

• Alcoholic hepatitis
• urea synthesis
• prednisolone
• pentoxifylline
• mortality

Acknowledgements

We wish to thank the patients who participated in the study. We thank colleagues from the regional hospitals for their help with patient recruitment and thank Rikke Charlotte Andersen, Birgitte Sperling Wilms Nielsen, Inger Schjødt, Mette Mejlby Hansen and Janni Nielsen for their excellent technical support.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

The Aase and Ejnar Danielsen’s Foundation (EG); the Novo Nordisk Foundation (HG); the Health Research Fund of Central Denmark Region (NWO); and the Department of Clinical Medicine, Aarhus University, Denmark, generously supported this work.