Colchicine therapy for alcoholic hepatitis
ACP J Club. 1991 Jan-Feb;114:12. doi:10.7326/ACPJC-1991-114-1-012
Akriviadis EA, Steindel H, Pinto PC, et al. Failure of colchicine to improve short-term survival in patients with alcoholic hepatitis. Gastroenterology. 1990;99:811-8.
To evaluate colchicine as an anti-inflammatory agent in patients with acute alcoholic hepatitis.
Randomized, double-blind, placebo-controlled trial.
University hospital liver unit in California, between November 1986 and April 1988.
The participants were hospitalized patients (49 men and 23 women) with a history of heavy ethanol abuse and diagnosis at admission of acute alcoholic hepatitis. Inclusion criteria were palpable hepatomegaly, serum bilirubin of 5 mg/dL or over, and one or more of hepatic tenderness, temperature above 100°F, and leukocytosis above 12 000/mm3. Patients with life-threatening infection, massive gastrointestinal hemorrhage, renal insufficiency, or hepatitis B surface antigen positivity, rapidly improving liver tests, or advanced alcoholic cirrhosis were excluded. Of 74 patients enrolled, 2 dropped out.
Patients were treated for 30 days with either colchicine, 1 mg in lactose-filled capsules (n = 36), or identical placebo (n = 36). Medical management was otherwise individualized according to the patient's condition.
Main outcome measures
Predetermined outcomes included mortality and major morbidity including gastrointestinal bleeding, renal impairment, hepatic encephalopathy, and the effect on biochemical test results.
7 of the colchicine-treated and 6 control patients died in hospital (difference 2.8%; 95% CI, −15% to 20.5%), after a mean of 17.4 and 17.8 days, respectively. 4-month follow-up of 21 of 29 colchicine-treated and 19 of 30 control patients showed an additional 2 deaths in each group. Morbidity rates and laboratory test values were similar in the two groups. Nonsignificant trends favored the control groups for all outcomes tested. However, the study had low power to detect effects on major outcomes. For example, over 300 patients per group would have been required to rule out a 50% reduction in the 18% mortality rate that was observed, with a power of 80% and a 2-sided significance level of 5%.
The use of colchicine does not appear to result in short-term benefit in acute alcoholic hepatitis. A much larger study would be required to have adequate power to rule out the possibility of a benefit.
Source of funding: Division of Research Resources of the National Institutes of Health.
Article reprints not available.
This is the first clinical trial of colchicine therapy for acute alcoholic hepatitis. The authors' rationales for using colchicine are that it has anti-inflammatory properties and that inflammation in acute alcoholic hepatitis is a precursor of necrosis and fibrosis leading to cirrhoris. However, the study design does not allow a direct comparison with the original observations on which the authors base their hypothesis, that is, colchicine's anti-inflammatory effect in carbon tetrachloride toxicity.
This study shows that colchicine does not affect mortality or morbidity in acute alcoholic hepatitis following alcohol consumption. Work by Mourelle and Meza (1) showed a protective effect when colchicine was used as a prophylactic treatment. The current study has two major variations that make it difficult to compare the results to previous clinical trials: gender composition and severity classification. Previous trials comprised mainly male patients. A third of the patients in this trial, however, were female, and results are not presented separately for the two sexes. Second, their severity classification uses only the total bilirubin level and does not include the prothrombin time.
This study properly concludes that colchicine is not effective in treating acute alcoholic hepatitis. However, the author's conclusion that colchicine cannot be recommended for treating alcoholic hepatitis goes beyond the presented evidence. The findings reported are for acute alcoholic hepatitis and apply only to short-term morbidity and mortality. Prophylactic treatment and long-term effects deserve more study.
Carlo H. Tamburro, MD
University of Louisville Louisville, Kentucky, USA