Neurobehavioral outcomes of infants exposed to buprenorphine-naloxone compared with naltrexone during pregnancy

Abstract

Background:

Naltrexone is a medication used to treat both opioid and alcohol use disorder with limited experience in pregnant individuals, particularly in comparison to more commonly utilized treatments such as buprenorphine-naloxone. The long-term outcomes of infants exposed to naltrexone has not been previously examined.

Aims:

To compare the neurobehavioral outcomes of naltrexone versus buprenorphine-naloxone exposed infants.

Study design:

Multi-centered prospective cohort study.

Subjects:

Pregnant people on prescribed buprenorphine-naloxone or naltrexone were enrolled during pregnancy and the dyad followed until 12 months after delivery.

Outcome measures:

Infants were evaluated at 4–6 weeks corrected gestational age (CGA) using the NICU Neonatal Neurobehavioral Scale (NNNS) and at the 12-month CGA visit using the Ages and Stages Questionnaire, Third Edition (ASQ-3).

Results:

There were 7 dyads in the naltrexone group and 34 in the buprenorphine-naloxone group. On the NNNS, infants exposed to naltrexone had higher median scores for arousal and excitability, and lower median scores for attention and regulation at 4–6 weeks CGA compared to the buprenorphine-naloxone group. None of the infants in the naltrexone group were monitored for NOWS and had shorter length of hospital stay compared with the buprenorphine-naloxone group. Although no statistically significant differences were observed, more infants in the buprenorphine-naloxone group were identified as at risk for development delays in the communication, problem solving, and personal social domains of the ASQ-3 at 12 months CGA. Results should be interpreted with caution given this study's small sample size and lack of a prospective comparison cohort.

Conclusions:

In this small cohort, there are differences noted in infant neurobehavior by NNNS at 4–6 weeks of age when comparing the buprenorphine-naloxone and naltrexone groups. At 12 months, ASQ-3 scores were similar but with percentage differences in potential development delay risk observed between the two groups. Larger cohort studies are needed to determine the long-term child outcomes after naltrexone exposure in pregnancy.

1. Introduction

Opioid use disorder (OUD) and alcohol use disorder (AUD) during pregnancy are a significant public health concern with a risk for possible adverse infant and child outcomes. According to the 2020 National Survey on Drug Use, the national rate of birthing person opioid-related diagnoses in the United States increased from 3.5 in 1000 delivery hospitalizations in 2010 to 8.2 per 1000 in 2017 [1]. Rates of binge drinking among pregnant individuals was 5.3 % in 2022, and 1.7 % for heavy alcohol use according to the 2022 National Survey on Drug Use and Health [2,3]. Treatment of individuals in pregnancy with medication for OUD such as buprenorphine or methadone is the standard of care during pregnancy. Studies have shown less non-prescribed opioid use and better birthing person and neonatal health outcomes with opioid agonist treatment in pregnancy [4,5]. Recently, the sublingual buprenorphine-naloxone formulation has been utilized increasingly in pregnancy with no known harmful effects on the birthing person or infant [6].

Though medication treatment is considered first-line for moderate to severe AUD in non-pregnant individuals, there is limited data on pharmacotherapy for AUD during pregnancy [7]. Naltrexone is a medication used to treat both OUD and AUD in non-pregnant individuals [7-9]. It is a mu-opioid receptor antagonist blocking effects of opioids and a modulator of the hypothalamic pituitary axis reducing alcohol cravings [10]. Based on its mechanism of action, naltrexone treatment is largely used to maintain opioid abstinence [10]. Naltrexone has been demonstrated to decrease the risk of heavy drinking in patients with AUD by 83 % when compared to placebo in prior randomized trials in non-pregnant patients, and has also demonstrated benefit for high-risk individuals with OUD in increasing retention in treatment and abstinence [11,12].

There are limited studies on the efficacy of naltrexone in the treatment of AUD or OUD in pregnancy; however, small retrospective studies have demonstrated the use of naltrexone in pregnancy is associated with a significant reduction in the rate of Neonatal Opioid Withdrawal Syndrome (NOWS) when compared to buprenorphine and overall reassuring pregnancy and neonatal outcomes without reports of birth defects or neonatal complications [13-16]. Prior studies have examined the long-term outcomes of infants exposed to buprenorphine or methadone in-utero [17-20]; though these studies report variable results, some have indicated a risk for altered neurocognitive and physical development that can begin as children reach school age [19,20]. Prior studies of naltrexone in pregnancy have focused on pregnancy and immediate neonatal outcomes [13-16]; no prior studies to our knowledge have examined longer-term child health or neurodevelopmental outcomes.

Thus, despite promising preliminary data supporting the use of naltrexone for AUD and OUD in pregnancy, there remain significant gaps in knowledge, specifically about the full range of infant and childhood outcomes. Therefore, the objective of this study was to present a comparison of buprenorphine-naloxone versus naltrexone-exposed infants with an examination of child health, neurobehavioral, and developmental outcomes to 12 months of age.

2. Materials and methods

2.1. Subject enrollment

The MOM NEST Study (NCT 03718104) is a multi-site prospective cohort study conducted at two medical centers in Massachusetts and North Carolina. Both centers have a dedicated peripartum integrated care program for pregnant and postpartum individuals with substance use disorders, providing substance use disorder care, obstetrical and psychiatric care, case management services, peer support, and social work services from pregnancy through one year after delivery. This study was approved by the institutional review boards of both institutions.

Buprenorphine-naloxone was selected as the comparison group due to similar demographic characteristics of perinatal individuals on naltrexone in our prior studies, and given it was the primary formulation of buprenorphine utilized in pregnancy at both institutions [12]. To be considered eligible for the buprenorphine-naloxone group, pregnant individuals had to be stabilized on buprenorphine-naloxone for the treatment of OUD before approach for the study. To be considered eligible for the naltrexone group, pregnant individuals had to be stabilized on naltrexone for the treatment of OUD or AUD before approach for the study. For both groups, individuals had to be at least 18 years of age, English-speaking, willing to attend study visits at our institutions, and have a singleton pregnancy. Individuals were excluded from eligibility if they were incarcerated at the time of approach for consent, had a severe psychiatric illness or cognitive impairment, or were over 34 weeks gestation at the time of approach. Participants were enrolled in the study between November 2018 and May 2022.

2.2. Clinical treatment protocol

Participants in both buprenorphine-naloxone and naltrexone groups were treated per the routine treatment protocols at the two institutions during the pregnancy and postpartum period. Those in the naltrexone group could be on extended-release (380 mg intramuscular every 28 days) or oral (50 mg daily) naltrexone. All infants were treated per routine care protocols for the monitoring of substance-exposed newborns at our institutions. At the Massachusetts institution, all opioid-exposed newborns are cared for with a rooming-in approach in the postpartum unit and pediatric unit with use of a non-pharmacologic care treatment bundle, the Eat, Sleep, Console (ESC) NOWS Care Tool and as needed methadone for pharmacologic treatment [21,22]. At the North Carolina institution, infants are assessed with the ESC care approach and given as needed morphine for pharmacologic treatment. The North Carolina institution also utilizes a rooming-in approach with emphasis on non-pharmacologic interventions to support the dyad [23].

2.3. Study protocol and data collection

The study protocol included serial study visits in each trimester with urine toxicology testing, Addiction Severity Index (ASI) questionnaire for substance use assessment at 4 time points, questionnaire administration related to medical health, a delivery hospitalization visit, a 4–6 week CGA neonatal visit, every other month phone calls postpartum, and a 12-month CGA child follow-up visit.

At the 4–6 week visit, infants were assessed with the NICU Neonatal Neurobehavioral Scale (NNNS) by certified examiners. The NNNS is a comprehensive assessment of infant neurobehavior that can assess neurological integrity and behavioral function in at-risk infants within the first two months after birth and is commonly utilized as a measure after in-utero substance exposure [24]. Examiners were blinded to prenatal exposure information before and after performing the assessments.

At the 12-month visit, parents (birth parent or foster parent) were asked to complete an Ages and Stages Questionnaire, Third Edition (ASQ-3) as well as a child health questionnaire. The ASQ-3 is a parent-centric development screening test that includes questions about communication, gross motor, fine motor, problem solving, and person-social of children from one month old to 5.5 years old. It includes cut-off scores in these five domains to identify potential development-delayed children for seeking further professional assessment [25,26].

Electronic medical records were also reviewed monthly for any adverse events not reported during the study visits such as re-hospitalizations or emergency room visits. Additional electronic medical record data collected included pregnancy and infant outcomes (medical problems, medications, co-exposures, delivery outcomes, NOWS outcomes, hospitalizations, emergency room visits, ultrasounds, urine toxicology tests), child health diagnoses to 12-months of age, and child growth parameters to 12-months of age. All data were entered into a secure 21 CFR Part 11-ready electronic study database (REDCap).

2.4. Statistical methods

Given the small sample size (n = 7 the naltrexone group), differences in demographic and outcome variables were assessed using Mann-Whitney U tests. Fisher's exact test of independence was used to evaluate differences in dichotomous and categorical variables. Regression analyses were not performed due to the small sample size of the naltrexone group. All statistical tests are performed two-sided and determined at a significance level of α = 0.05 [27]. Analyses are performed using SAS v9.4 software and R-studio v2022.7.1.554 (R v4.2.1) software.

3. Results

During the enrollment period of December 2018 to December 2022, a total of 147 pregnant individuals were screened eligible at the two sites, with 10 individuals being eligible for the naltrexone group and 137 individuals for the buprenorphine-naloxone group. Among the eligible individuals, 70 % in the naltrexone group and 33.6 % in the buprenorphine-naloxone group were approached to allow for temporal spacing of participants over time in both groups. From the approached individuals, 100 % were enrolled in the naltrexone group (n = 7) and 73.9 % in the buprenorphine-naloxone group (n = 34). Percentages of completion of the 4–6 week and 12-month visits for both groups are shown in Fig. 1, with some loss to follow-up in both groups; 3 of the participants in the naltrexone and 16 in the buprenorphine-naloxone group completed the 12-month follow-up. After combining questionnaire data with electronic medical record data, 6 participants in the naltrexone group and 26 in the buprenorphine-naloxone group had 12-month health data available for review.

Fig. 1.

Enrollment and Retention Summary.

Although overall baseline demographic factors were similar between the two cohorts of pregnant individuals with no statistically significant differences, participants in the naltrexone group had more private insurance (43.0 % versus 20.6 %, p = 0.15) and higher rates of depression (86.0 % versus 50.0 %, p = 0.21), post-traumatic stress disorder (71.0 % versus 38.2 %, p = 0.21), and attention deficit hyperactivity disorder (43.0 % versus 11.8 %, p = 0.17) compared to the buprenorphine-naloxone group. There was a statistically significant difference in race and ethnicity between the two groups, however both cohorts were primarily White Non-Hispanic. In terms of other pregnancy substance exposures, there were no statistically significant differences with similar rates between the two groups (Table 1).

Table 1.

Baseline characteristics of pregnant study participants on buprenorphine-naloxone and naltrexone.

Baseline characteristic	Buprenorphine-naloxone group N = 34 N(%) or Median (IQR)	Naltrexone group N = 7 N(%) or Median (IQR)	P-value
Study site – N(%)
Boston	25 (73.5 %)	7 (26.5 %)	0.31
North Carolina	9 (100.0 %)	0 (0.0 %)
Reason for naltrexone – N(%)	N/A		N/A
Opioid use disorder		4 (57.1 %)
Alcohol use disorder		3 (42.9 %)
Naltrexone formulation – N(%)	N/A		N/A
Oral		3 (42.9 %)
Extended release		4 (57.1 %)
Gestational age at study enrollment (weeks) – Mean (SD)	23.05 (12.4)	26.4 (22.5)	0.92
Mean buprenorphine dose at delivery (mg/day) – Mean (SD)	16 (4.0)	N/A	N/A
Race and ethnicity – N(%)
Black Non-Hispanic	2 (5.9 %)	0 (0.0 %)	0.04
White Hispanic	0 (0.0 %)	0 (0.0 %)
White Non-Hispanic	32 (94.1 %)	7 (100.0 %)
Education level – N(%)
Less than high school	1 (2.9 %)	1 (14.0 %)	0.87
High school	12 (35.3 %)	3 (43.0 %)
Some College	14 (41.2 %)	2 (29.0 %)
Completed College	4 (11.8 %)	0 (0.0 %)
Other	3 (8.8 %)	0
Missing	0	1 (14.0 %)
Health insurance – N(%)
Public insurance	27 (79.4 %)	3 (43.0 %)	0.15
Private insurance	7 (20.6 %)	3 (43.0 %)
Other	0 (0.0 %)	0 (0.0 %)
ACES total score – median (IQR)	4.0 (4.0)	2.0 (7.0)	0.92
BDI score baseline – median (IQR)	12.0 (17.0)	8.0 (11.0)	0.35
BAI score baseline – median (IQR)	11.5 (20.0)	2.0 (2.0)	0.11
Gestational age at first prenatal visit (weeks) – median (IQR)	10.4 (6.3)	7.3 (23.0)	0.21
Total prenatal visits – median (IQR)	16.0 (9.0)	13.0 (13.0)	0.22
Psychiatric diagnoses - N(%)
Any	29 (85.3 %)	7 (100.0 %)	1.00
Depression	17 (50.0 %)	6 (86.0 %)	0.21
Anxiety	20 (58.8 %)	5 (57.0 %)	1.00
Bipolar disorder	8 (23.5 %)	0 (0.0 %)	0.31
Post-traumatic stress disorder	13 (38.2 %)	5 (71.0 %)	0.21
Attention deficit hyperactivity disorder	4 (11.8 %)	3 (43.0 %)	0.09
Other	9 (26.5 %)	0 (0.0 %)	0.17
Hepatitis C – N(%)	15 (44.1 %)	3 (43.0 %)	1.00
Non-prescribed substance use in pregnancy – N(%)
Any
Non-prescribed opiates	30 (88.2 %)	6 (86.0 %)	1.00
Cocaine	0 (0.0 %)	0 (0.0 %)	N/A
Amphetamines	2 (5.9 %)	0 (0.0 %)	1.00
Benzodiazepines	5 (14.7 %)	2 (29.0 %)	0.58
Cannabis	2 (5.9 %)	0 (0.0 %)	1.00
Alcohol	4 (11.8 %)	1 (14.0 %)	0.52
Nicotine	1 (2.9 %)	1 (14.0 %)	0.28
Heroin	26 (76.5 %)	5 (71.0 %)	1.00
Fentanyl	4 (11.8 %)	0 (0.0 %)	1.00
Barbituates	4 (11.8 %)	0 (0.0 %)	1.00
Non-prescribed sedatives, hypnotics,	2 (5.9 %)	0 (0.0 %)	1.00
Tranquilizers	3 (8.8 %)	0 (0.0 %)	1.00
Nicotine use in the third trimester – N(%)	27 (79.4 %)	5 (71.0 %)	0.68

Abbreviations: ACES = Adverse Childhood Experiences Score; BDI = Beck's Depression Index, BAI = Beck's Anxiety Index.

P-values calculated by Mann-Whitney U test for continuous variables; by Fisher's exact test for non-continuous variables.

Pregnancy, delivery, and neonatal outcomes are shown in Table 2. There were no meaningful differences in pregnancy complications except for a diagnosis of gestational hypertension which was higher in the naltrexone group (57.1 % versus 11.8 %, p = 0.02). There were no reported congenital anomalies noted in either cohort. Neonatal outcomes differed in that none of the infants in the naltrexone group were monitored for NOWS. Additionally, shorter median hospitalizations were found in naltrexone infants in comparison to the buprenorphine-naloxone group (2 versus 6 days, p = 0.001). There were no differences in birth weight or head circumference between the groups, after excluding a preterm infant born at gestational age of <30 weeks in the naltrexone group (Table 2). Other percentage differences in delivery and infant outcomes between the two groups were driven by the preterm infant.

Table 2.

Pregnancy and delivery outcomes buprenorphine-naloxone and naltrexone-exposed pregnancies.

Outcome	Buprenorphine-naloxone group N = 34 N(%) or Median (IQR)	Naltrexone group N = 7 N(%) and Median (IQR)	P-value
Maternal outcome
Pregnancy/Delivery complications: N(%)
Premature rupture of membranes	2 (5.9 %)	1 (14.3 %)	0.45
Placenta previa	1 (2.9 %)	0 (0.0 %)	1.00
Intrauterine growth restriction	3 (8.8 %)	1 (14.3 %)	0.55
Non-reassuring fetal heart tracing	3 (8.8 %)	0 (0.0 %)	1.00
Gestational hypertension	4 (11.8 %)	4 (57.1 %)	0.02
Pre-eclampsia	4 (11.8 %)	1 (14.3 %)	1.00
Gestational diabetes	1 (2.9 %)	1 (14.3 %)	0.32
Preterm labor	1 (2.9 %)	1 (14.3 %)	0.32
Placental abruption	1 (2.9 %)	0 (0.0 %)	1.00
Chorioamnionitis	1 (2.9 %)	0 (0.0 %)	1.00
Non-prescribed substance use in pregnancy - N(%)	10 (29.4 %)	1 (14.3 %)	0.65
Overdose in pregnancy - N(%)	1 (2.9 %)	0 (0.0 %)	1.00
Gestational age at delivery (weeks) – median (IQR)
Late preterm (>34 weeks) and full-term	39.0 (1.9)	37.8 (1.5)	0.21
<30 weeks preterm (n = 1)	N/A	28.4 (N/A)	N/A
Mode of delivery – N(%)
Vaginal	23 (67.6 %)	7 (100.0 %)	0.16
C-section	10 (29.4 %)	0 (0.0 %)
Infant outcomes
Infant gender at birth - N(%)
Male	23 (67.6 %)	3 (42.9 %)	0.21
Female	10 (29.4 %)	4 (57.1 %)
Infant monitored for NOWS -N(%)	32 (94.1 %)	0 (0 %)	N/A
ICD10 code for NOWS –N(%)	14 (41.2 %)	1 (14.3 %)	0.37
Pharmacologic treatment for NOWS-N(%)	6 (17.6 %)	N/A	N/A
NOWS medications – N(%)		N/A	N/A
Morphine	0 (0.0 %)
Methadone	6 (17.6 %)
Clonidine	0 (0.0 %)
Phenobarbital	0 (0.0 %)
Opioid treatment days for NOWS – median (IQR)	3.5 (6.0)	N/A	N/A
Birth weight (grams) – median (IQR)
Late Preterm (>34 weeks) and Full-term	3120.0 (615.0)	3225.0 (210.0)	0.73
<30 weeks preterm (n = 1)	N/A	1281.0 (0.0)	N/A
Birth weight percentile – median (IQR)
Late Preterm (>34 weeks) and Full-term	36.0 (37.0)	59.0 (50.0)	0.34
<30 weeks preterm	N/A	59.1 (0.0)	N/A
Birth head circumference (cm) – median (IQR)
Late Preterm (>34 weeks) and Full-term	34.3 (2.5)	34.3 (2.0)	1.00
<30 weeks preterm	N/A	24.8 (0.0)	N/A
Head circumference percentile – median (IQR)
Late Preterm (>34 weeks) and Full-term	47.0 (56.0)	66.5 (30.0)	0.32
<30 weeks preterm	N/A	34.0 (0.0)	N/A
APGAR 1 min – median (IQR)	8.0 (1.0)	8.0 (3.0)	0.48
APGAR 5 min – median (IQR)	9.0 (0.0)	9.0 (2.0)	0.07
Infant complications – N(%)
Respiratory distress syndrome	1 (2.9 %)	1 (14.3%)	0.34
Feeding intolerance	0 (0.0 %)	1 (14.3 %)	0.34
Sepsis	0 (0.0 %)	1 (14.3 %)	0.18
Meconium aspiration syndrome	0 (0.0 %)	0 (0.0 %)	N/A
Transient tachypnea	1 (2.9 %)	1 (14.3 %)	0.34
Jaundice requiring phototherapy	2 (5.9 %)	1 (14.3 %)	0.47
Severe Hypoglycemia	0 (0.0 %)	1 (14.3 %)	0.18
Other infection	0 (0.0 %)	0 (0.0 %)	N/A
Nasogastric feeds	3 (8.8 %)	0 (0.0 %)	1.00
Other medical condition	4 (11.8 %)	1 (14.3 %)	1.00
NICU admission – N(%)	3 (8.8 %)	2 (28.6 %)	0.22
Eligible to provide breastmilk – N(%)	29 (85.3 %)	6 (85.7 %)	1.00
Received breastmilk – N(%)	23 (67.6 %)	3 (42.9 %)	0.18
Received breastmilk 4–6 weeks after delivery- N(%)	1 (2.9 %)	1 (14.3 %)	0.34
Infant length of hospital stay (days) – median (IQR)
Late Preterm (>34 weeks) and Full-term	6.0 (4.0)	2.0 (1.0)	0.001
<30 weeks preterm	N/A	88.0 (0.0)	N/A
Infant discharge destination – N(%)
With birth parent	30 (88.2 %)	7 (100.0 %)	1.00
Other	1 (2.9 %)	0 (0.0 %)

Abbreviations: NOWS = neonatal opioid withdrawal syndrome; NICU = neonatal intensive care unit; NAS = neonatal abstinence syndrome; cm = centimeter.

Infant health outcomes to 12-month CGA were available for 26 of the buprenorphine-naloxone infants and 6 infants in the naltrexone group. Infant growth parameters were similar between the two groups and within the normal range for corrected age. There were no meaningful differences in child health diagnoses between the two groups, with overall low frequencies of additional health diagnoses, emergency room visits, and infant re-hospitalizations (Table 3).

Table 3.

12-month health outcomes of buprenorphine-naloxone versus naltrexone-exposed children.

Variable	Buprenorphine-naloxone group N = 26 N(%) or Median (IQR)	Naltrexone group N = 6 N(%) or Median (IQR)	P-value
Child's corrected age at time of 12-month visit (months) – median (IQR)
	12.30 (3.00)	12.00 (0.00)	0.26
Child's weight (kg) – median (IQR)	9.84 (1.34)	8.50 (1.50)	0.24
Child weight percentile – median (IQR)	32.00 (47.00)	18.00 (61.00)	0.90
Child's head circumference (cm) – median (IQR)	45.85 (2.00)	45.75 (2.75)	0.70
Child head circumference percentile – median (IQR)	44.00 (45.00)	65.50 (69.00)	0.37
Medical conditions in the first 12-months – N(%)
Viral infection	4 (15.4 %)	2 (33.3 %)	0.61
Ear infection	6 (23.1 %)	1 (16.7 %)	0.85
Heart problems	1 (3.8 %)	1 (16.7 %)	0.32
Seizures	0 (0.0 %)	0 (0.0 %)	N/A
Hearing problem	0 (0.0 %)	0 (0.0 %)	N/A
Allergies	5 (19.2 %)	0 (0.0 %)	0.51
Genetic disorder	1 (3.8 %)	0 (0.0 %)	0.71
Blood problem	0 (0.0 %)	0 (0.0 %)	0.65
Gastrointestinal problem	3 (11.5 %)	0 (0.0 %)	0.84
Feeding problem	4 (15.4 %)	0 (0.0 %)	0.55
Head injury	0 (0.0 %)	0 (0.0 %)	0.65
Poor weight gain	2 (7.7 %)	1 (16.7 %)	0.40
Eye problems	2 (7.7 %)	0 (0.0 %)	0.80
Gastrointestinal reflux	1 (3.8 %)	1 (16.7 %)	0.32
Fetal alcohol spectrum disorder	0 (0.0 %)	1 (16.7 %)	0.24
Emergency room visit – N(%)	6 (23.1 %)	2 (33.3 %)	0.63
Hospital re-admission – N(%)	1 (3.8 %)	1 (16.7 %)	0.35
Child in parental custody at 12-months – N(%)	20 (76.9 %)	5 (83.3 %)	1.00

The median CGA of conduct of the NNNS exams was 5.6 weeks in the buprenorphine-naloxone (n = 24) and 4.4 weeks in the naltrexone (n = 3) group. On the NNNS, infants exposed to naltrexone had higher median scores for arousal and excitability, and lower median scores for attention and regulation in comparison to the buprenorphine-naloxone group. The median CGA for the ASQ-3 questionnaire was 12.6 months in the buprenorphine-naloxone (n = 16) and 12 months in the naltrexone (n = 3) group. Although there were no statistically significant differences in median scores or percentage of potential development delay between the two groups on the ASQ-3, more infants in the buprenorphine-naloxone group scored below the cut-off in the communication (6.25 % versus 0.00 %, p = 1.00), problem solving (18.75 % versus 0.00 %, p = 1.00), and personal social (12.5 % versus 0.00 %, p = 1.00) domains (Table 4).

Table 4.

Neurodevelopmental scores of buprenorphine-naloxone and naltrexone exposed infants within the first 12-months.

Variable	Score range (NNNS) [26] or cut-off score (ASQ-3) [31]	Buprenorphine-naloxone group N = 24 for NNNS N = 16 for ASQ-3 Median (IQR) or N(%)	Naltrexone group N = 3 for both NNNS and ASQ-3 Median (IQR) or N(%)	P- value
NNNS
Corrected gestational age at the time of the NNNS exam (weeks)		5.57 (3.01)	4.43 (6.00)	0.49
Habituationa	1–9	8.17 (1.33)	N/A	N/A
Attention	1–9	5.93 (2.29)	3.50 (0.71)	0.04
Arousal	1–9	3.29 (0.64)	4.83 (1.43)	0.02
Regulation	1–9	5.71 (1.20)	4.18 (0.90)	0.02
Handling	0–1	0.00 (0.75)	0.50 (0.38)	0.25
Quality of Movements	1–9	4.90 (1.03)	4.67 (2.27)	0.79
Excitability	0–15	1.00 (2.5)	6.00 (6.00)	0.01
Lethargy	0–15	4.00 (3.00)	3.00 (4.00)	0.84
Non-optimal reflexes	0–15	4.00 (3.00)	5.00 (7.00)	1.00
Asymmetric reflexes	0–16	1.50 (3.00)	0.00 (2.00)	0.33
Hypertonicity	0–10	0.00 (0.00)	0.00 (1.00)	0.26
Hypotonicity	0–10	0.00 (0.00)	0.00 (0.00)	0.49
Stress/abstinence	0–1	0.09 (0.12)	0.22 (0.20)	0.11
ASQ-3
Corrected gestational age at time of assessment (months)		12.6 (2.9)	12.0 (5.3)	0.31
ASQ Gross Motor Total		55.0 (20.0)	60.0 (10.0)	0.44
ASQ Fine Motor Total		55.0 (20.0)	55.0 (30.0)	0.77
ASQ Communication Total		52.5 (12.5)	55.0 (5.0)	0.95
ASQ Problem Solving Total		45.0 (25.0)	55.0 (20.0)	0.52
ASQ Personal Social Total		50.0 (25.0)	55.0 (5.0)	0.37
ASQ Gross Motor Below Cut-off Score	21.49	0 (0.00 %)	0 (0.00 %)	N/A
ASQ Fine Motor Below Cut-off Score	34.50	1 (6.25 %)	1 (33.33 %)	0.30
ASQ Communication Below Cut-off Score	15.64	1 (6.25 %)	0 (0.00 %)	1.00
ASQ Problem Solving Below Cut-off Score	27.32	3 (18.75 %)	0 (0.00 %)	1.00
ASQ Personal Social Below Cut-off Score	21.73	2 (12.5 %)	0 (0.00 %)	1.00

Abbreviations: NNNS = NICU Neonatal Neurobehavioral Scale; ASQ-3 = Ages and Stages Questionnaire, Third edition.

^aFor Habituation, no scoring range is available. Many infants are not scored for habituation because they are awake during the exam.

4. Discussion

In this prospective comparative cohort study of naltrexone and buprenorphine-naloxone prenatally-exposed infants, pregnancy and neonatal outcomes were similar between the two groups. The slight differences in demographics and co-morbidities noted may be related to differences in the participants' substance use disorder diagnoses or other unmeasured variables. Differences were seen on NNNS examination at 4–6 weeks CGA between the groups, which could suggest some transient neurobehavioral differences based on prenatal exposures. Specifically, exposure to naltrexone was associated with worse attention, regulation, arousal, and excitability in the first month after birth. However, overall health outcomes and parentally assessed neurodevelopment did not differ at 12-months CGA.

Infants exposed to naltrexone had higher mean scores for arousal and excitability, and lower mean scores for attention and regulation at 4–6 weeks CGA in comparison to infants exposed to buprenorphine-naloxone on the NNNS. On the NNNS, higher scores for arousal indicate higher activity levels with less ability to control state; higher scores for excitability indicate higher levels of irritability; lower attentional scores indicate worse ability to attend to auditory and visual stimuli; and lower scores for regulation indicate less of an ability to self-regulate [28]. It is unclear why infants demonstrated such differences on the NNNS, though we hypothesize that it could be related to differences in other prenatal exposures (ex: alcohol) contributing to these different neurobehavioral profiles, as few of the infants received pharmacotherapy for NOWS. Overall scores for both groups of infants are similar to published norms at 1 month of age for non-exposed infants, with the exception of the naltrexone group having lower scores for attention and higher scores for excitability compared to the published normative data [29]. In previous studies, in comparison to non-opioid exposed infants, those exposed to opioids in-utero had poorer quality of movement and increased hypertonicity [30]. In addition, infants exposed to methadone or buprenorphine in-utero that received pharmacologic treatment for NOWS have demonstrated decreases in excitability, hypertonia, and lethargy, along with improvements in quality of movements on the NNNS when assessed at 1–5 weeks of age [31,32]. Future larger cohort studies could adjust for prenatal exposures to examine the neurobehavioral development of in-utero naltrexone and buprenorphine-naloxone exposure infants.

The ASQ-3 is a parental report questionnaire for neurodevelopment, aiming to identify high-risk children for the purposes of referrals for developmental follow-up. Though there is not a direct correlation, the problem solving and personal-social domains of the ASQ-3 overlap with some of the domains of assessment of the NNNS, specifically regarding how the child interacts with their surrounding and how they respond to stimuli. In general, ASQ-3 scores for both cohorts were similar to or higher than those previously reported in non-opioid exposed children [33]. There are no statistically significant differences between the two cohorts, however higher percentages of infants in the buprenorphine-naloxone group are below the ASQ-3 cut-off scores in the communication, problem solving, and personal-social domains, which is typically an indicator of risk for neurodevelopmental delays [34]. Results should be interpreted with caution given this study's small sample size and lack of a prospective comparison cohort.

Strengths of our study are that it is the first prospective cohort study to our knowledge to follow infants exposed to naltrexone from pregnancy through 12-months of age and to assess neurodevelopmental outcomes in this population in comparison to a cohort exposed to buprenorphine-naloxone. However, despite these strengths, our study is limited by a low enrollment rate in the naltrexone group, even with extensive outreach efforts across the study time period [35]. In addition, within the naltrexone group there was variability in the diagnosis of the birthing person (AUD vs OUD) as well as the formulation of naltrexone (oral versus extended-release). There were also in-utero co-exposures in both groups, though they did not differ statistically between the groups. Additional limitations include the lack of racial and ethnic diversity of our participants, and the loss to follow-up after delivery, which is partly related to the COVID-19 pandemic and changes in child custody limiting our ability to conduct in-person developmental formal assessments.

Given the low number of pregnant individuals on naltrexone nationally, there is a need for additional education about the safety and efficacy of this medication for pregnant individuals [30]. Furthermore, there is currently a lack of other studies examining the long-term outcomes of naltrexone-exposed children. Future studies should attempt to enroll larger cohorts of pregnant individuals on naltrexone and follow naltrexone-exposed infants beyond 12 months of age for formal neurodevelopmental assessments closer to school-age. In addition, an attempt should be made to distinguish any differences based on diagnosis of the birthing person or formulation of naltrexone, as well as to compare child health outcomes with those exposed to methadone or other formulations of buprenorphine.

5. Conclusions

In conclusion, in this prospective cohort study of naltrexone versus buprenorphine-naloxone exposed pregnancies, 12-month child health outcomes appear similar between the exposures groups though with some differences noted in infant neurobehavior at 4–6 weeks CGA and ASQ-3 results at 12 months. Future studies with larger cohorts and longer-term birthing person and child follow-up are required regarding this newer treatment option for pregnant individuals with OUD and AUD.

Data statement

Data obtained though this study are confidential and not available for sharing.