O'Malley 2007: Vivitrol in people already abstinent before the first shot

What they tested

Extended-release naltrexone (sold as Vivitrol) is a shot given once a month. This paper looked at a subgroup from a larger trial: people who had voluntarily stopped drinking for 4 or more days before treatment began.

They received the 380 mg shot, a lower 190 mg shot, or a placebo (a dummy injection). Everyone also had counseling. The team compared a wide range of drinking outcomes between groups.

What they found

On the 380 mg shot, 32% stayed continuously abstinent through the end of the study, compared with 11% on placebo (P = 0.02). The median time to a first heavy drinking day was more than 180 days, versus 20 days on placebo.

Drinking days per month fell to a median of 0.7 versus 7.2 on placebo, 90% fewer. Heavy drinking days per month were 0.2 versus 2.9, 93% fewer.

More than twice as many people on 380 mg counted as responders, defined as no more than 2 heavy drinking days in any 28-day stretch: 70% versus 30%. A liver blood test linked to drinking (GGT) also improved more on the drug.

The 190 mg group, with only 26 people, had results that were generally in between, suggesting a higher dose had a bigger effect.

Limits

• This is a subgroup of one trial, not a new study. The site's guides count it as the same evidence as the main trial.
• The subgroup was small, and the 190 mg arm had just 26 people.
• People who can stop drinking on their own for several days before treatment may be more motivated than average, so the results may not apply to people still drinking.

What it means if you're considering naltrexone

Vivitrol is FDA-approved for people who can stop drinking before they start it. This analysis shows that, in that group, the shot was linked to much longer stretches without drinking, and the abstinence gap appeared with as few as 4 days off alcohol beforehand.

It doesn't tell us how the shot works for people who are still drinking at the first injection.