Continued drinking is the single strongest determinant of outcome in alcohol-associated cirrhosis, and there has been no approved pharmacotherapy for it in this population — largely because hepatotoxicity concerns kept naltrexone out of the very patients who need it most. NAL-CI tested that assumption. One hundred patients with compensated alcohol-associated cirrhosis and DSM-5 alcohol use disorder were randomised to naltrexone 50 mg daily or placebo for 12 weeks, with standardised psychosocial support given to both arms. Mean age was around 43, mean MELD 12.6 and 12.7, and Child-Turcotte-Pugh scores 5.9 and 6.2.

Abstinence at 12 weeks, defined as no alcohol in the preceding four weeks, was 64% (32 of 50) with naltrexone against 22% (11 of 50) with placebo, p<0.001, odds ratio 10.86 (95% CI 1.89 to 62.2). Lapses at three months fell from 54% to 28% (p=0.008). Heavy-drinking relapse showed a trend, 12% against 28% (p=0.07), and abstinence maintained at six months favoured naltrexone without reaching significance, 22% against 8% (p=0.09). Craving scores fell significantly on both subscales of the Obsessive Compulsive Drinking Scale.

The safety data are what make this actionable. No patient developed hepatic decompensation attributable to the study drug, and no AST or ALT elevation exceeded five times the upper limit of normal in either arm. Adverse events were comparable. In a compensated population with a mean MELD around 12, naltrexone did not behave as a hepatotoxin.

This is directly applicable to Indian practice, and not only because the trial was conducted here. Alcohol-associated cirrhosis presents younger in India than in Western series — the mean age of 43 in this trial reflects that — and it is a leading indication for transplant assessment in a system with very few transplants. A drug that triples short-term abstinence in exactly this group, at low cost and with a reassuring hepatic safety profile, deserves to be offered rather than withheld on a theoretical concern.

The limits are real and should be stated with the offer. One hundred patients, 12 weeks, compensated disease only, and both arms received structured psychosocial support — so this is naltrexone added to counselling, not instead of it. The six-month figures show most patients had resumed drinking by then. This starts abstinence; it does not maintain it on its own.

Offer naltrexone 50 mg daily to patients with compensated alcohol-associated cirrhosis and alcohol use disorder.
Provide structured psychosocial support alongside it — both trial arms received it.
Do not withhold it over hepatotoxicity: no decompensation and no transaminase rise above five times normal.
Restrict to compensated disease; decompensated cirrhosis was not studied.
Plan for the six-month drop-off — abstinence at 12 weeks does not persist without continued support.

The statistics, in plain English

An odds ratio of 10.86 sounds enormous, and its confidence interval of 1.89 to 62.2 shows why the number itself should not be quoted to a patient. That width is what a hundred-patient trial produces. The raw proportions are far more useful and just as convincing: 64% against 22%, meaning about two and a half patients need treating for one extra person to be abstinent at 12 weeks. The secondary endpoints that did not reach significance — heavy-drinking relapse at p=0.07, six-month abstinence at p=0.09 — are pointing the same way but are underpowered rather than negative. Note that abstinence was point-prevalence, based on the preceding four weeks, so it measures being abstinent at 12 weeks rather than having been abstinent throughout.