Kelty 2017: pregnancy outcomes with naltrexone implants

What they tested

Researchers took women treated with naltrexone implants, methadone or buprenorphine between 2001 and 2010, plus a group of age-matched women without opioid dependence. They linked them to midwife, hospital, emergency department and death records to find pregnancies and health problems during pregnancy and after birth.

What they found

Rates of pregnancy loss needing hospital or emergency care were significantly higher in women on naltrexone implants than in women on buprenorphine (p = 0.018) and in controls (p < 0.001). They were not significantly different from women on methadone (p = 0.210).

Birth rates in the naltrexone implant group were significantly higher than in all three other groups.

Hospital and emergency visits during pregnancy, and overall complications during pregnancy and labor, were not significantly different between the naltrexone group and the methadone or buprenorphine groups. Overall complications during pregnancy were significantly higher in the naltrexone group than in controls.

Limits

• It looked back at records, so it can show links but not prove that the implant caused any outcome.
• Only pregnancy losses that led to hospital or emergency care were counted.
• The abstract gives p-values but not the actual rates or group sizes.
• It studied implants, which the site's implant guide notes are not FDA-approved in the US, rather than tablets or the Vivitrol shot.

What it means if you're considering naltrexone

The authors note that very little is known about naltrexone's safety in pregnancy. This study found more pregnancy losses with naltrexone implants than with buprenorphine, but results similar to methadone on most measures. It can't show whether naltrexone itself caused any of these differences, and it doesn't cover the tablets or the monthly shot.