Naltrexone in the treatment of opioid‐dependent pregnant women: common ground

We thank Stitzer 1 for her positive assessment and value her endorsement of our call for comparative safety and efficacy trials of naltrexone. We understand her concern that ‘patient treatment matching’ presents a challenging research agenda; however, given the current progress towards the discovery of biomarkers of substance abuse 2, this research idea may be progressively less daunting. We agree with the concerns that Waal 3 has expressed regarding the use of naltrexone with opioid-dependent women, particularly: the need for, and risk associated with, medication-assisted withdrawal for opioid-dependent pregnant women prior to induction onto naltrexone, the possibility of increased morbidity/mortality risk to the dyad in the event of dropout from naltrexone treatment, and the lack of data regarding possible teratogenic effects. As we discussed, and Waal notes, none of these represent contraindications to the use of naltrexone for opioid-dependent pregnant women, but would necessitate changes in treatment service delivery for this population. There is the substantial potential benefit that the neonate will be born without opioid agonist exposure and resultant withdrawal syndrome. We agree that long-term, prospective, multi-site, longitudinal cohort studies are important to determine the impact of prenatal exposure to any substance. In the United States, the 5-year grant funding structure poses a challenge to conducting such trials. Finally, regarding patient preference trials, we endorse the SMART (Sequential Multiple Assignment Randomized Trial) design as a design choice for the study of naltrexone in the treatment of opioid dependence in pregnancy. We agree that, ethically, we must perform the necessary research to be able to provide naltrexone treatment to this population. Clark & Verster 4 suggest two studies, one comparing naltrexone to opioid-agonist treatment, and another comparing naltrexone to medication-assisted withdrawal. We prefer to approach this question from the perspective of a single patient preference trial, a complicated study requiring fewer total participants, but yielding more informative treatment comparisons. We thank Clark & Verster for suggesting that treatment of opioid-dependent pregnant women would benefit from consensus regarding protocols for medication-assisted withdrawal in pregnancy that include consideration of the fetus. We applaud the World Health Organization (WHO) for developing guidelines for treating this population. White 5 points out the inconsistencies in the literature between human and animal findings—human research focusing on birth outcomes and animal research on developmental processes. He makes an excellent suggestion that research on children exposed prenatally to naltrexone should examine many of the variables studied in the animal literature, including pain sensitivity and emotional lability. We thank McCullough and colleagues, whose ethical framework and writing (e.g. 6-10) influenced our thinking about how to conduct research with naltrexone in an opioid-dependent pregnant population. McCullough, Chervenak & Coverdale 11 offer a ‘standard’ ethical framework for research with this population. Our paper attempted to tailor this framework for research with naltrexone in an opioid-dependent pregnant population. This tailoring may result in the field choosing not to adopt the same ethical framework as might be applied to other disorders in the pregnant patient. The stimulating discussion has found common ground in three important areas: 1 research on naltrexone for the treatment of opioid-dependent pregnant women is needed; 2 such research should consider obvious and subtle short- and long-term teratogenic effects; and 3 ethical concerns should be paramount. We could not agree more. Our call is for the field to coalesce around the goal of making such research a priority, so that this highly vulnerable population can be offered every available treatment option for opioid dependence. HEJ has received reimbursement for time and travel from Reckitt Benckiser; LMJ receives medication from Reckitt Benckiser for current study-related purposes. The other authors report no conflicts of interest related to this paper.