Anton 2006 (COMBINE): naltrexone, acamprosate and counselling

What they tested

COMBINE enrolled 1,383 people with alcohol dependence who had recently stopped drinking. Eight groups got medical management, short visits with a health professional, plus 16 weeks of naltrexone (100 mg a day), acamprosate (3 g a day), both, or placebos, with or without a combined behavioral intervention (CBI), a more intensive counselling program. A ninth group got CBI only, with no pills.

The main outcomes were the percentage of days without alcohol and the time to the first heavy drinking day. People were followed for up to 1 year after treatment.

What they found

Every group cut its drinking substantially. During treatment, percent days abstinent was 80.6 with naltrexone plus medical management, 79.2 with CBI plus medical management and placebos, and 77.1 with naltrexone, CBI and medical management, compared with 75.1 on placebos and medical management alone.

Naltrexone lowered the risk of a heavy drinking day over time (hazard ratio 0.72; 97.5% CI, 0.53-0.98). This was most clear in people who had medical management without CBI.

Acamprosate had no significant effect on drinking compared with placebo, alone or combined with naltrexone, CBI or both.

Placebo pills and check-ins had an effect of their own. People given CBI with no pills or medical management had 66.6 percent days abstinent, compared with 73.8 for placebo plus medical management and 79.8 for placebo, medical management and CBI. One year after treatment ended, the differences between groups were similar but no longer significant.

Limits

• COMBINE used naltrexone 100 mg a day, twice the usual 50 mg label dose for alcohol.
• Everyone had recently stopped drinking, so it doesn't speak directly to people who start naltrexone while still drinking.
• The benefits faded to non-significant differences in the year after treatment.

What it means if you're considering naltrexone

COMBINE showed naltrexone can work with light support: brief check-ins with a clinician, not intensive therapy. The authors said this means naltrexone could be offered in ordinary health care settings to people who might otherwise not get treatment.

It also found acamprosate didn't help in this US population, and that taking a pill and seeing a clinician helped even when the pill was a placebo.