The mainstay of the management of opioid use disorder in pregnancy is with methadone or buprenorphine medication-assisted treatment. Methadone and buprenorphine are opioid agonist drugs. Naltrexone, an opioid antagonist, is also a medication-assisted treatment option; however, to date, only a few retrospective studies have reported its use in pregnancy.
Our study objective was to evaluate prospectively obstetric and newborn outcomes and the maternal/fetal effects of the use of naltrexone as a medication-assisted treatment in pregnant patients with opioid use disorder.
We performed a prospective cohort study collecting data on all pregnant women who were treated with naltrexone medication-assisted treatment compared with pregnant women who were treated with methadone or buprenorphine medication-assisted treatment. Based on a sample size calculation, it was determined that for a power of 90, a minimum of 160 study participants (80 in each group) was needed with an alpha of .01 and an expected 60% rate of newborn infants who were treated for neonatal abstinence syndrome in the methadone or buprenorphine medication-assisted treatment group compared with a 30% rate in the naltrexone medication-assisted treatment group. In a random subset of 20 maternal/newborn dyads, blood levels for naltrexone and 6-beta-naltrexol (an active metabolite) were analyzed at delivery.
A total of 230 patients were studied: 121 patients with naltrexone medication-assisted treatment compared with 109 patients with methadone or buprenorphine medication-assisted treatment. No differences between groups were seen regarding demographics, the use of comedications/drugs, or obstetric outcomes. For newborn outcomes, the rate of neonatal abstinence syndrome in neonates >34 weeks gestation was significantly lower in the naltrexone medication-assisted treatment group (10/119 [8.4%] vs 79/105 [75.2%]; P <.0001). Multivariate analysis demonstrated that the only significant factor for the rate of neonatal abstinence syndrome was the form of medication-assisted treatment. Of 87 patients who received naltrexone up to delivery, no neonates experienced symptoms of neonatal abstinence syndrome. No maternal relapses occurred in the 7-day no-treatment window before the initiation of naltrexone therapy. No cases of spontaneous abortion or stillbirth occurred in either group. In 64 patients who started naltrexone therapy at ≥24 weeks gestation, no changes were seen in the fetal heart monitor tracing with drug initiation. The incidence of birth anomalies was no different between the groups. Umbilical cord blood and maternal levels for naltrexone and 6-beta-naltrexol matched; no levels were elevated, and values were undetected if naltrexone was discontinued >60 hours before delivery.
These study data demonstrate that, in pregnant women who choose to completely detoxify off opioid drugs during gestation, naltrexone, as a continued form of medication-assisted treatment, is a viable option for some pregnant patients who experience opioid use disorder. Naltrexone crosses the placenta, and maternal and fetal levels are concordant. Because naltrexone clears quickly from the maternal circulation, this rapid clearance needs to be addressed with patients. This is important because maternal relapse could occur in a short time-period if the oral drug is discontinued without the knowledge of their healthcare providers. Nonetheless, the drug is well-tolerated by both mother and fetus, and newborn infants do not experience symptoms of neonatal abstinence syndrome if naltrexone medication-assisted treatment is maintained to delivery.
Click Supplemental Materials under article title in Contents at ajog.org
Related editorial, page 1 .
Opioid use disorder (OUD) is currently a major healthcare concern that results in increased hospitalizations, overdose, death, psychosocial issues, and healthcare costs. This has also led to an increase in OUD in pregnant women. The cornerstone of medical treatment for this condition is medication-assisted treatment (MAT). Three primary medications (methadone, buprenorphine, and naltrexone) are recommended for MAT by the Substance Abuse and Mental Health Services Administration. Methadone and buprenorphine are opioid agonist medications that continue opioid drug dependence but, if used correctly, prevent cravings for illicit opioid drugs. Naltrexone is an opioid antagonist drug that can also prevent cravings but does not produce dependence. Methadone has been the mainstay for treating OUD in pregnancy for years; however, buprenorphine recently has been added as an alternate treatment option. Naltrexone, however, has not been examined extensively in the obstetric population because it requires full detoxification before it can be used.
Why was this study conducted?
Opioid use disorder is now a major healthcare problem in the United States and has greatly increased the rate of newborn infants who are treated for neonatal abstinence syndrome. Alternative options are needed for the treatment of pregnant women with opioid use disorder.
The use of naltrexone as a form of medication-assisted treatment is a viable option for the treatment of some obstetrics patients with opioid use disorder.
What does this add to what is known?
This is the first prospective study to analyze the use of naltrexone in the treatment pregnant patients who experience opioid use disorder.
Only a few retrospective studies that have reported the use of naltrexone in human pregnancy have been published that included >80 pregnancies; the results, at a minimum, were no different than for those women who were treated with methadone. Through an extensive literature search using PubMed, Medline, Scopus, Google Scholar, and the Cochrane Library, we found no prospective studies on the use of naltrexone in pregnancy. Furthermore, no study was found that analyzed maternal and newborn blood levels of patients who were receiving naltrexone at the time of delivery. Naltrexone is similar to naloxone (another opioid antagonist drug) but differs by the replacement of an allyl group (C 2 H 3 ) on the naloxone molecule with a cyclopropyl-methyl group (C 3 H 5 ) on the naltrexone molecule. This change results in a longer duration of action for naltrexone. Naloxone and naltrexone are classified by the United States Food and Drug Administration as category C medications for use in pregnancy.
In November 2016, we created a designated obstetrics OUD clinic at our institution. This clinic uses intense behavioral health management. When patients become engaged in prenatal care and are stable on traditional opioid agonist MAT, our clinic offers the option of tapering with full detoxification, if desired. If full detoxification is chosen and successful, patients are then offered naltrexone as a continued form of MAT to maintain sobriety, decrease opioid cravings, and decrease the risk for relapse. Those who choose naltrexone and those who do not are both continued in behavioral health management.
Our study objective was to perform a prospective observational cohort study that would compare those patients who chose opioid antagonist naltrexone pharmacotherapy MAT (after full detoxification) with those who opted for continued opioid agonist pharmacotherapy (traditional) MAT. Our secondary study objective was to analyze blood levels of naltrexone in paired mother/newborn infant dyads in a subset of random patients at the time of delivery.
We performed a prospective cohort study on those patients who were treated with opioid antagonist pharmacotherapy (naltrexone) MAT after full detoxification compared with those patients who were treated with continued opioid agonist pharmacotherapy (traditional) MAT of methadone or buprenorphine. The naltrexone MAT and traditional MAT populations came from the same designated obstetrics OUD clinic and delivered at University of Tennessee Medical Center, Knoxville. Pregnancies that involved fetal aneuploidy were excluded.
All patients in the designated clinic (if not already in a methadone or buprenorphine program) are placed in traditional MAT programs of buprenorphine or methadone to avoid illicit substance use. The average range of daily buprenorphine dosage was 8–16 mg and for a daily methadone dosage was 50–120 mg. For those patients who are fully engaged in traditional MAT and prenatal care, we offer 2 pregnancy management options after an informed discussion in a nondirective method. The primary concern with opioid tapering or full detoxification is the potential for maternal relapse with the inherent risk for overdose. The primary concern with continued opioid drug maintenance is the potential for newborn dependence with the inherent risk for neonatal abstinence syndrome (NAS) and other potential newborn effects. For those patients who choose medically supervised withdrawal, there are 2 options that are either inpatient or outpatient; these have previously been described. If fully detoxified, this group is then offered naltrexone MAT. Additionally, if a patient enters our program receiving naltrexone therapy, they are given the option of continuing this form of MAT after the same discussion.
The naltrexone protocol requires a patient to be opioid drug free for at least 7 days with 2 negative drug screens. Once this is accomplished, a daily 50-mg oral dose is used that is adjusted near delivery. For pregnancies at ≥24 weeks gestation, patients have continuous fetal heart rate (FHR) monitoring for 60 minutes minimum, while the initial dose is administered. For those patients at <24 weeks gestation, the FHR is evaluated by Doppler or ultrasound imaging before the initial dose, 30 minutes after dosing, and again at 60 minutes. All patients are contacted by phone 3–4 days after naltrexone initiation and are scheduled to return for follow up in 1 week. All patients in the designated obstetrics OUD clinic have biophysical profiles performed every other week from 28–32 weeks gestation and then weekly until delivery.
Maternal and neonatal outcome data were collected prospectively by the authors. Maternal data collection included demographics, medical history, social history, and delivery data. Specific information on use of tobacco, alcohol, marijuana, illicit drugs, gabapentin, and selective serotonin reuptake inhibitors (SSRIs) was recorded. Maternal drug use was confirmed by urine toxicology studies that were performed by a major laboratory. All patients had urine drug screening performed during prenatal care and at delivery in both the traditional MAT and naltrexone MAT groups. Newborn data collection included birthweight, gender, head circumference (HC) at birth, diagnosis of NAS that led to newborn treatment, neonatal intensive care unit admission, and hospital length of stay. The criteria for making a diagnosis of NAS that resulted in newborn treatment consisted of any 2 consecutive Finnegan scores of ≥10 or a single Finnegan score of ≥12.
The best obstetric estimate, as recommended by the American College of Obstetricians and Gynecologists Committee Opinion, was used for pregnancy dating in both groups. More than 1 ultrasound scan was performed in every study patient before delivery. Birthweight and HC nomograms were from the American Academy of Pediatrics based on gestational age and sex. HC measurements were performed by a single group of pediatric nurse practitioners and physicians on all newborn infants with a uniform approach. The HC was obtained after resolution of caput or molding if present at delivery.
One of the primary study endpoints was the rate of newborn infants who were treated for symptoms of NAS that was expected to be lower in the naltrexone MAT group. For this, we performed a sample size calculation; for a power of 90, a minimum of 160 study participants (80 in each group) was needed (with the use of an alpha of .01 and an expected 60% rate of newborns who were treated for NAS in the traditional MAT group compared with a 30% rate in the naltrexone MAT group). Based on the average number of pregnant patients who were seen in our clinic per month who choose medically supervised withdrawal and full detoxification (with the assumption that 50–60% might choose naltrexone MAT), it was calculated that this investigation would need to span a minimum of 16 months. Therefore, based on a start date of July 1, 2017, the decision was made to collect data through October 31, 2018.
For the secondary study objective, maternal and newborn blood levels were analyzed in a subset of 20 random patients for naltrexone and 6-beta-naltrexol at delivery. Statistical analysis involved chi-square, Fishers exact, and Student t tests, where applicable, with significance considered at a probability value of <.05. Multivariate regression analysis was also performed where indicated. All comparisons were performed against a 2-sided alternative hypothesis. This study was reviewed and approved by the institutional review board of University of Tennessee Medical Center, Knoxville. There was no commercial support or outside sponsors for this study.
There were 363 patients treated in the dedicated obstetrics OUD clinic during the study period. Of these, 29 patients (8%) never engaged in a MAT program, had numerous absentee healthcare visits, and were excluded. The Figure shows that the study population consisted of 109 patients in the traditional buprenorphine/methadone MAT group compared with 121 patients in the naltrexone MAT group.
Table 1 shows the demographics between the 2 groups; no differences were seen. A median of 8 drug screens were performed for both groups; the numbers positive for other drugs during the gestation, including benzodiazepines, amphetamines, cocaine, marijuana, alcohol, gabapentin, and SSRIs were also not different. Tobacco usage during the pregnancy was based on patient report.
|Variable||Medication-assisted treatment group||P value|
|Naltrexone (n=121)||Traditional (n=109)|
|Maternal age, y a||28.0±5.3||28.1±5.5||.89|
|White, n (%)||115 (95)||104 (95.4)||.99|
|Multiparity, n (%)||93 (76.9)||86 (78.9)||.83|
|Chronic maternal medical disorders, n (%) b||13 (10.7)||14 (12.8)||.77|
|Diabetes mellitus (gestational/pregestational), n (%)||6 (5.0)||7 (6.4)||.85|
|Other medications/drugs used during pregnancy confirmed by urine toxicology screening, n (%)|
|Benzodiazepine||12 (9.9)||14 (12.8)||.62|
|Amphetamine||20 (16.5)||13 (11.9)||.42|
|Cocaine||5 (4.1)||4 (3.7)||.99|
|Marijuana||31 (25.6)||32 (29.4)||.63|
|Gabapentin||29 (24.0)||24 (22.0)||.85|
|Selective serotonin reuptake inhibitor||39 (32.2)||29 (26.6)||.43|
|Tobacco c||87 (71.9)||82 (75.2)||.67|
|Alcohol||2 (1.7)||3 (2.8)||.67|
|Patients who received no comedications/drugs||10 (8.3)||13 (11.9)||.48|
|Patients who received >1 comedication/drug||71 (58.7)||62 (56.9)||.89|