Krupitsky 2012: naltrexone implant vs pills vs placebo for opioids

What they tested

306 people who had recently detoxed from opioids were split into three groups of 102. One group got a 1000 mg naltrexone implant and placebo pills, one got a placebo implant and 50 mg naltrexone pills, and one got a placebo implant and placebo pills. Because everyone had both an implant and pills, no one could tell who was on the real drug (a double-dummy design).

Everyone also had counseling every two weeks for the 24 weeks. The main measure was the share of people still in treatment without relapse.

What they found

54 of 102 people with the implant (52.9%) were still in treatment without relapse at month 6, against 16 of 102 on pills (15.7%) and 11 of 102 on placebo (10.8%). Pills did slightly better than placebo, but that difference wasn't statistically significant (P = .07).

Counting missed urine tests as positive, 63.6% of tests were free of opiates in the implant group, 42.7% in the pill group and 34.1% in the placebo group.

There were 12 wound infections after 244 implant procedures (4.9%) in the implant group, against 2 of 181 (1.1%) and 1 of 148 (0.7%) in the other groups. All came in the first 2 weeks and cleared with antibiotics. Four people in the implant group had redness and swelling at the site in the second month, which cleared with anti-allergy medicine.

Other side effects were rare and none were serious. The researchers found no sign of more overdose deaths after naltrexone treatment ended.

Limits

• All patients came from addiction programs in one city, St Petersburg.
• It covered 24 weeks of treatment, so it doesn't show how people did over longer periods.
• It compared the implant with naltrexone pills and placebo, not with methadone or buprenorphine.

What it means if you're considering naltrexone

This trial is one of the main sources of evidence for naltrexone implants. It shows that a form of naltrexone that doesn't rely on a daily pill kept far more people in treatment and off opioids than the pills did, at the cost of more wound infections. The site's implant guide notes that no implant is FDA-approved in the US; the approved long-acting form is the monthly Vivitrol shot.