Limbic System Microstructure in Neonates With Antenatal Opioid Exposure

María Guadalupe Mora Álvarez; Josepheen De Asis-Cruz; Kushal Kapse; Yao Wu; Stephanie L Merhar; Carla M Bann; Jamie E Newman; Nicole Mack; Sara B DeMauro; Namasivayam Ambalavanan; Scott A Lorch; Deanne Wilson-Costello; Brenda B Poindexter; Myriam Peralta-Carcelen; Jonathan M Davis; Catherine Limperopoulos

doi:10.1001/jamanetworkopen.2026.21603

. 2026 Jul 2;9(7):e2621603. doi: 10.1001/jamanetworkopen.2026.21603

Limbic System Microstructure in Neonates With Antenatal Opioid Exposure

María Guadalupe Mora Álvarez ¹, Josepheen De Asis-Cruz ¹, Kushal Kapse ¹, Yao Wu ¹, Stephanie L Merhar ^2,³, Carla M Bann ⁴, Jamie E Newman ⁴, Nicole Mack ⁴, Sara B DeMauro ^5,⁶, Namasivayam Ambalavanan ⁷, Scott A Lorch ^5,⁶, Deanne Wilson-Costello ⁸, Brenda B Poindexter ⁹, Myriam Peralta-Carcelen ⁷, Jonathan M Davis ¹⁰, Catherine Limperopoulos ^1,^✉

¹Developing Brain Institute, Children’s National Hospital, Washington, District of Columbia

²Perinatal Institute, Division of Neonatology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio

³Department of Pediatrics, University of Cincinnati, Cincinnati, Ohio

⁴Analytics Division, RTI International, Research Triangle Park, North Carolina

⁵Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania

⁶University of Pennsylvania Perelman School of Medicine, Philadelphia

⁷University of Alabama at Birmingham

⁸Case Western Reserve University, Cleveland, Ohio

⁹Emory University, Atlanta, Georgia

¹⁰Tufts University, Boston, Massachusetts

Accepted for Publication: May 12, 2026.

Published: July 2, 2026. doi:10.1001/jamanetworkopen.2026.21603

Open Access: This is an open access article distributed under the terms of the CC-BY-NC-ND License, which does not permit alteration or commercial use, including those for text and data mining, AI training, and similar technologies. © 2026 Mora Álvarez MG et al. JAMA Network Open.

^✉

Corresponding Author: Catherine Limperopoulos, PhD, Developing Brain Institute, Children’s National Hospital, 111 Michigan Ave NW, Washington, DC 20010 (climpero@childrensnational.org).

Author Contributions: Dr Mora Ávarez had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Mora Ávarez, Merhar, Bann, Newman, Ambalavanan, Wilson-Costello, Poindexter, Limperopoulos.

Acquisition, analysis, or interpretation of data: Mora Ávarez, De Asis-Cruz, Kapse, Wu, Merhar, Bann, Mack, DeMauro, Ambalavanan, Lorch, Wilson-Costello, Peralta-Carcelen, Davis, Limperopoulos.

Drafting of the manuscript: Mora Ávarez, Limperopoulos.

Critical review of the manuscript for important intellectual content: All authors.

Statistical analysis: Mora Ávarez, Mack, Wilson-Costello, Limperopoulos.

Obtained funding: Merhar, Bann, Newman, DeMauro, Ambalavanan, Lorch, Poindexter, Peralta-Carcelen, Davis, Limperopoulos.

Administrative, technical, or material support: De Asis-Cruz, Kapse, Wu, Bann, Newman, Ambalavanan, Wilson-Costello, Davis, Limperopoulos.

Supervision: De Asis-Cruz, Merhar, Poindexter, Limperopoulos.

Conflict of Interest Disclosures: Dr DeMauro reported receiving grants from the National Institutes of Health (NIH) outside the submitted work. Dr Ambalavanan reported having a patent pending for vital sign detection of opioid exposure in neonates. Dr Davis reported receiving personal fees from Inflection Medicine outside the submitted work. No other disclosures were reported.

Funding/Support: This study was funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development through the US NIH Helping to End Addiction Long-Term (NIH HEAL) Initiative via grants 1PL1HD101059-01 to RTI International for Dr Mora Álvarez, Dr De Asis-Cruz, Mr Kapse, Dr Wu, Dr Bann, Dr Newman, Ms Mack, Dr Davis, and Dr Limperopoulos; 1RL1HD104251-01 to Drs Ambalavanan and Peralta-Carcelen at the University of Alabama at Birmingham; 1RL1HD104252-01 to Drs DeMauro and Lorch at Children’s Hospital of Philadelphia; 1RL1HD104253-01 to Dr Wilson-Costello at Case Western Reserve University; 1RL1HD104254-01 to Drs Merhar and Poindexter at Cincinnati Children’s Hospital Medical Center; and 3PL1HD101059-01(S1-4) to RTI International.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or its NIH HEAL Initiative.

Data Sharing Statement: See Supplement 2.

Additional Contributions: We are deeply grateful to the infants and their families who participated in this study; our medical and nursing colleagues at Case Western Reserve University, Cincinnati Children’s Hospital Medical Center, the University of Alabama at Birmingham, and Children’s Hospital of Philadelphia; our neuroimaging core colleagues at Children’s National Hospital; and the data coordinating center staff at RTI International. These contributors were not compensated for this work.

^✉

Corresponding author.

PMCID: PMC13329709 PMID: 42390863

This case-control study investigates compares limbic system microstructure in neonates with vs without opioid exposure.

Key Points

Question

Does limbic system microstructure differ in neonates who are opioid exposed vs unexposed?

Findings

In this case-control study of 201 neonates, antenatal opioid exposure was associated with lower mean and radial diffusivity in the fornix compared with an unexposed control group. Secondary, exploratory analyses revealed diffusion differences in limbic gray matter regions (thalamus, cingulum, parahippocampal gyrus, hippocampus, and amygdala), but these were not corrected for multiple comparisons.

Meaning

This study found that antenatal opioid exposure in neonates was associated with disrupted limbic system microstructure, especially in the fornix.

Abstract

Importance

There is limited research examining the association of antenatal opioid exposure with neonatal brain microstructure, particularly within the limbic system, which contains high concentrations of opioid receptors. Prior studies investigating microstructural changes in major white matter tracts were underpowered and lacked matched controls.

Objective

To compare limbic system microstructure between neonates who were opioid exposed and unexposed.

Design, Setting, and Participants

This case-control study assessed diffusion tensor imaging of neonates enrolled in the Outcomes of Babies With Opioid Exposure (OBOE) study and scanned between August 2020 and December 2023 at 4 US sites. Participants were term neonates with antenatal exposure and unexposed controls. Data were analyzed from July 2024 to April 2026.

Exposure

Neonates were antenatally exposed to methadone, buprenorphine, oxycodone, hydromorphone, fentanyl, hydrocodone, heroin, morphine, or some combination of these opioids.

Main Outcomes and Measures

The primary outcome was limbic system diffusion metric comparisons between neonates who were exposed and unexposed. Diffusion tensor imaging data in term neonates were acquired using harmonized protocols on Siemens and Philips 3T magnetic resonance imaging scanners, preprocessed using a neonatal-optimized pipeline, and analyzed. Using analysis of covariance, fractional anisotropy, mean diffusivity, radial diffusivity, and axial diffusivity were compared in the limbic system of neonates who were unexposed vs opioid exposed. In secondary, exploratory analyses, diffusion metrics in the limbic system were compared between subgroups of opioid exposure (methadone, buprenorphine, and other opioids) and controls. False discovery rate (FDR) was used to correct for multiple comparisons.

Results

Among 201 enrolled neonates, 75 were unexposed (mean [SD] postmenstrual age, 42.76 [2.05] weeks; 45 male [60.0%]) and 126 were antenatally opioid exposed (mean [SD] postmenstrual age, 42.72 [2.06] weeks; 65 male [51.6%]), of which 79 were exposed to buprenorphine, 30 to methadone, and 14 to other opioids. Neonates who were opioid exposed showed a significantly lower mean diffusivity (mean difference [β] = −0.06 × 10⁻³ mm²/s [95% CI, −0.11 to −0.01 × 10⁻³ mm²/s]; FDR P = .04) and radial diffusivity (β = −0.07 × 10⁻³ mm²/s [95% CI, −0.12 to −0.02 × 10⁻³ mm²/s]; FDR P = .04) in the fornix. Secondary, exploratory analyses revealed buprenorphine-specific associations (eg, higher mean diffusivity in the left parahippocampal gyrus vs no opioid exposure: β = 0.02 × 10⁻³ mm²/s [95% CI, 0.00 to 0.04 × 10⁻³ mm²/s]) and other opioid–specific associations (eg, higher fractional anisotropy in the left thalamus for other opioids vs unexposed: β = 0.01 [95% CI, 0.01 to 0.02]) in GM limbic system structures.

Conclusions and Relevance

This study found altered limbic system microstructure, particularly in the fornix, in neonates with antenatal opioid exposure, and exploratory analyses showed different patterns of limbic system microstructure across opioid exposure subgroups. Clinical correlation is needed.

Introduction

Antenatal opioid exposure in the US is a major public health problem.¹ Between 2016 and 2020, rates of antenatal substance exposure increased by 3.6%.² According to a 2019 survey,³ almost 7% of women reported prescription opioid use during pregnancy. Despite these increasing numbers, research examining the association of antenatal opioid exposure with changes in the developing human brain is limited.

Previous studies have reported disrupted oligodendrocyte lineage and myelination processes in antenatally opioid-exposed rat pups.^4,5 Opioids can cross the placenta and blood-brain barrier and affect the fetal brain due to the presence of opioid receptors in neurons, astroglia, and oligodendrocytes.^6,7,8 Notably, important populations of opioid receptors, particularly μ-opioid receptors and nociceptin or orphanin FQ peptide receptors, are located in the limbic system.^9,10,11

There is strong evidence that opioids are associated with limbic system changes in the mature brain, with studies in adults with opioid use disorder showing altered brain volumes, white matter (WM) microstructure, and functional connectivity in important limbic structures.^12,13,14 The limbic system plays a vital role in regulating appetite, sleep patterns, and emotional responses.¹⁵ Irritability, excessive crying, and poor feeding are common findings in babies born with neonatal opioid-withdrawal syndrome,^16,17 which suggests potential dysfunction of the limbic system in this high-risk population.

Moreover, a 2022 magnetic resonance study¹⁸ combined diffusion tensor imaging (DTI) and neurite orientation dispersion and density imaging on a translational-relevant animal model of prenatal methadone exposure, revealing altered diffusion metrics denoting microstructural changes (ie, changes in neurite morphology) in the limbic system. This included the hippocampus, dorsal amygdala, thalamus, septal nuclei, and nucleus accumbens of methadone-exposed mice. A small 2023 DTI study¹⁹ of 11 fetuses that were opioid exposed reported higher fractional anisotropy (FA) in WM tracts of the limbic system, including the right fornix stria terminalis. Previous studies using DTI in neonates who were antenatally opioid exposed have also reported disrupted WM microstructure in various brain areas. For example, neonates who were methadone exposed showed lower FA in the internal capsule and inferior longitudinal fasciculi,²⁰ as well as higher mean diffusivity (MD) in the superior longitudinal fasciculi.²¹ A study on older children exposed prenatally to opioids and other psychoactive substances found lower FA in regions of early myelination, such as the inferior longitudinal fasciculi.^22,23 While these prior studies reported WM microstructural changes in children who were opioid exposed, they were underpowered due to small sample sizes and a lack of matched controls. In addition, most of these studies relied on tract-based spatial statistics, which may be limited in anatomical accuracy; this is particularly true in the developing brain, where FA values are intrinsically low and may affect skeleton projection. Moreover, tract-based spatial statistics do not allow for a targeted region-of-interest analysis of limbic system structures.²⁴

The objective of our study was to use a region of interest–based DTI approach to compare brain microstructural organization in the limbic system in neonates who were opioid exposed and unexposed. While DTI is mainly sensitive to important WM brain changes, such as myelination processes and fiber integrity,^25,26 a 2022 DTI study¹⁸ showed altered gray matter (GM) structure in the limbic system of an animal model of prenatal methadone exposure. Therefore, we included WM and GM limbic system structures in our analyses. In a secondary, exploratory analysis, we investigated changes in diffusion metrics in the limbic system in subgroups of neonates who were opioid exposed (ie, methadone only, buprenorphine only, and other opioids) compared with unexposed controls.

Methods

Study Design

The Advancing Clinical Trials in Neonatal Opioid Withdrawal (ACT-NOW) Outcomes of Babies With Opioid Exposure (OBOE) study is a multisite, prospective, observational case-control study of neonates with antenatal opioid exposure and controls recruited from 4 sites in the US. The OBOE study protocol has previously been described in detail.^27,28 Newborns who were opioid exposed were eligible if they were born at 37 weeks’ gestation or later and had second- or third-trimester opioid exposure. Control newborns who were unexposed were recruited from the same birth hospitals as newborns who were opioid exposed and were born at 37 weeks’ gestation or later with no known or reported opioid exposure. Newborns from both groups were excluded if they had known chromosomal or congenital anomalies with the potential to affect the central nervous system, a 5-minute Apgar score less than 5, any requirement for positive pressure ventilation in the neonatal intensive care unit, intrauterine growth restriction less than the third percentile, exposure to heavy alcohol use during pregnancy (≥8 alcoholic drinks per week), or inability to return for an outpatient magnetic resonance imaging (MRI) or neurodevelopmental follow-up. Mothers completed detailed surveys on prenatal exposures, and maternal and newborn medical records were reviewed to assess exposure history and toxicology findings. For this analysis, exposure status was defined based on maternal self-report and was confirmed with the toxicology report, when available. For generalizability of the data, maternal race and ethnicity were extracted from the electronic health record by RTI International Data Coordinating Center staff and based on participant self-report. Race and ethnicity categories were predefined by the health care system standard electronic health record fields. Race and ethnicity were categorized as Hispanic, non-Hispanic Black, non-Hispanic White, other (including American Indian or Alaska Native, Asian, and Native Hawaiian or Other Pacific Islander), or unknown; these data were included to assess the demographic diversity of participants. Through a single institutional review board at Cincinnati Children’s Hospital Medical Center, the 4 OBOE study clinical sites, the neuroimaging core, and the data coordinating center received approval for human participant research activities. Parents or guardians provided written informed consent. Data were acquired between August 2020 and December 2023, and DTI images were preprocessed and analyzed from July 2024 to December 2025. This study is reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

MRI Acquisition Protocol

MRIs were performed using a validated, non-sedated neonatal protocol.^27,28 Three-dimensional T2-weighted and DTI images were acquired on Siemens or Philips 3T scanners with protocols harmonized across the 4 clinical sites. DTI was collected with reversed phase-encode blips, resulting in pairs of images with distortions going in opposite directions. Site 1 used a Philips 3T scanner with a 32-channel head coil ([1] T2 weighted: turbo pin-echo sequence; repetition time/echo time [TR/TE] = 2500/270 ms; flip angle = 90°; voxel size = 0.982 × 0.982 × 1 mm³; acquisition time = 3:00 minutes; [2] DTI: multiband 2D spin-echo echo-planar imaging [SE-EPI]; 48 directions; b = 1000 seconds/mm²; TR/TE = 2718/116.22 ms; flip angle = 90°; voxel size = 2.292 × 2.292 × 2.5 mm³; acquisition time = 1:58 minutes). Site 2 used a Philips 3T scanner with 32-channel head coil ([1] T2 weighted: turbo spin-echo sequence; TR/TE = 2500/251.65 ms; flip angle = 90°; image resolution = 0.982 × 0.982 × 1 mm³; acquisition time = 3:00 minutes; [2] DTI: multiband SE-EPI; 48 directions; b = 1000 seconds/mm²; TR/TE = 2330/101.88 ms; flip angle = 90°; voxel size = 2.292 × 2.292 × 2.5 mm³; acquisition time = 1:58 minutes). Site 3 used a Siemens 3T scanner with 32-channel head coil ([1] T2 weighted: SPACE sequence; TR/TE = 3200/492 ms; flip angle = 90°; image resolution = 0.982 × 0.982 × 1 mm³; acquisition time = 2:45 minutes; [2] DTI: multiband SE-EPI; TR/TE = 2000/98 ms; flip angle = 90°; voxel size = 2.5 × 2.5 × 2.5 mm³; acquisition time = 3:06 minutes). Site 4 used a Siemens 3T scanner with 32-channel coil ([1] T2 weighted: SPACE sequence; TR/TE = 3200/492 ms; flip angle = 90°; image resolution = 1 × 1 × 1 mm³; acquisition time = 2:45 minutes; [2] DTI: multiband SE-EPI; 48 directions; b = 1000 seconds/mm²; TR/TE = 3440/98.4 ms; flip angle = 90°; voxel size = 2.5 × 2.5 × 2.5 mm³; acquisition time = 3:06 minutes).

MRI Preprocessing

MR images were preprocessed using a neonate-optimized pipeline (eFigure 1 in Supplement 1). DWI data were first denoised and unringed using MRtrix3 version 3.0.4.²⁹ EPI distortions were corrected using Topup in FSL version 6.0.0 (Oxford Centre for Functional MRI of the Brain) when reversed B0 acquisition was available.^30,31 For missing reversed B0 acquisitions, nonlinear registration to the T2 scan using Advanced Normalization Tools (ANTs) version 2.2.0 was used to correct for EPI distortions.³² To evaluate whether this approach influenced findings, a sensitivity analysis was performed in which EPI distortions were corrected using only T2-weighted ANTs registration (eFigures 7-10 in Supplement 1).

For motion and eddy current corrections, we used eddy_openmp in FSL with replacement of outliers.^33,34 Final DTI images were obtained using dtifit from FMRIB Software Library version 6.0 (FSL; Oxford Centre for Functional MRI of the Brain). Preprocessed images were normalized to the Edinburgh Newborn Atlas template using ANTs.³⁵

DTI Analysis

Diffusion was measured in WM tracts and GM structures of the limbic system (Figure 1). Regions of interest were defined using the Edinburgh Neonatal Atlas³⁵ with Johns Hopkins University Atlas labels³⁶ and eroded by 1 voxel using the FSL erode function (eFigure 3 in Supplement 1). For the primary analysis comparing neonates who were opioid exposed and unexposed, FA, MD, axial diffusivity (AD), and radial diffusivity (RD) were extracted bilaterally from the following limbic system structures: cingulum (cingulate part), cingulum (hippocampal part), fornix, stria terminalis, amygdala, cingular gyrus, entorhinal cortex, hippocampus, insular cortex, parahippocampal gyrus, and thalamus. For the secondary analysis examining differences within subgroups of neonates who were opioid exposed and unexposed, FA, MD, AD, and RD values were extracted separately from left and right hemispheres.

Statistical Analysis

Using analysis of covariance (ANCOVA) per region of interest, we compared FA, MD, AD, and RD in bilateral WM tracts and GM structures of the limbic system in neonates who were antenatally opioid exposed vs unexposed; these were the primary outcomes of our study. As a secondary, exploratory analysis, we examined diffusion differences in left and right WM tracts and left and right GM limbic system structures between neonates who were unexposed vs neonates in 3 subgroups of opioid exposure: methadone only, buprenorphine only, and other opioids. Diffusion metrics were harmonized across sites using ComBat (eFigure 4-6 in Supplement 1).³⁷ Covariates in ANCOVA included type of distortion correction used, postmenstrual age at MRI, sex, birth weight, maternal age, and maternal smoking during pregnancy. The unexposed group was coded as 0 and served as the reference (baseline) category. Statistical significance was set at P < .05. All statistical tests were 2-sided. False discovery rate (FDR) was used to correct for multiple comparisons within each primary analysis set (eg, 1 metric × 4 WM tracts = 4 comparisons, and 1 metric × 7 GM structures = 7 comparisons). Influential observations were identified using studentized residuals and Cook distance, and sensitivity analyses excluding these observations yielded consistent results (eTable in Supplement 1). Secondary analysis results were not corrected for multiple comparisons given their exploratory nature. Adjusted diffusion metrics were derived using a residualization approach, in which the respective diffusion metric was regressed on covariates, and resulting residuals were added to the overall mean FA. Statistical analyses and data visualization were conducted in Jupyter Notebook using Python version 3.11.9 libraries (Python Software Foundation).³⁸

Results

Study Population

Of 201 neonates enrolled in our study, 75 were unexposed (mean [SD] postmenstrual age [PMA], 42.76 [2.05] weeks; mean [SD] birth weight, 3.45 [0.46] kg; 45 male [60.0%]; 3 with Hispanic [4.0%], 13 with non-Hispanic Black [17.3%], and 56 with non-Hispanic White [74.7%] mothers) and 126 were antenatally opioid exposed (mean [SD] PMA, 42.72 [2.06] weeks; mean [SD] birth weight, 3.15[0.44] kg; 65 male [51.6%]; 21 with non-Hispanic Black [16.7%] and 105 with non-Hispanic White [83.3%] mothers). Among neonates who were exposed, 79 were exposed to buprenorphine (mean [SD] PMA, 42.82 [2.03] weeks; mean [SD] birth weight, 3.19 [0.46] kg; 43 male [51.6%]), 30 to methadone (mean [SD] PMA, 42.99 [2.10] weeks; mean [SD] birth weight, 3.09 [0.34] kg; 16 male [53.3%]), and 14 to other opioids (mean [SD] PMA, 41.45 [1.55] weeks; mean [SD] birth weight, 3.07 [0.30] kg; 5 male [35.7%]). See eFigure 2 in Supplement 1 for a comprehensive flowchart of exclusion of study participants. Neonates who were opioid exposed had significantly lower birth weight (mean difference, −0.30 kg [95% CI, −0.43 to −0.17 kg]; P < .001) and head circumference at birth (mean difference, −0.74 cm [95% CI, −1.13 to −0.34 cm]; P < .001) compared with neonates who were unexposed. Maternal age (mean difference, 1.61 years [95% CI, 0.09 to 3.13 years]; P = .04) and maternal smoking during pregnancy (mean difference, 50.1 percentage points [95% CI, 37.7 to 62.5 percentage points]; P < .001) were significantly higher in neonates who were opioid exposed vs unexposed. The proportion of mothers with less than a high school education was higher in neonates who were opioid exposed than unexposed (mean difference, 16.3 percentage points [95% CI, 7.1 to 25.6 percentage points; P < .001). The demographics of the final sample are presented in Table 1. Table 2 shows a detailed description of self-reported maternal medication use during pregnancy. Among neonates who were opioid exposed, the most common self-reported maternal medications included Buprenorphine (81 neonates [64.3%]), methadone (32 neonates [25.4%], and oxycodone [23 neonates [18.3%]) (Table 2).

Table 1. Neonate Demographics.

Characteristic	Neonates, No. (%) (N = 201)						P value^a
	Unexposed (n = 75)		Opioid exposed
	Unexposed (n = 75)		Total (n = 126)	Buprenorphine only (n = 79)	Methadone only (n = 30)	Other opioids (n = 14)
Site
1	9 (12.00)	19 (15.08)		11 (13.92)	5 (16.67)	1 (14.67)	NA
2	53 (70.67)	78 (61.90)		54 (68.35)	21 (70.00)	2 (17.46)
3	2 (2.67)	7 (5.56)		4 (5.06)	3 (10.00)	0
4	11 (14.67)	22 (17.46)		10 (12.66)	1 (3.33)	11 (78.57)
Sex
Male	45 (60.0)	65 (51.6)		43 (54.4)	16 (53.3)	5 (35.7)	.31
Female	30 (40.0)	61 (48.4)		36 (45.6)	14 (46.7)	9 (64.3)	.31
Maternal smoking during pregnancy
Yes	19 (25.3)	95 (75.4)		63 (79.7)	24 (80.0)	5 (35.7)	<.001
No	56 (74.7)	31 (24.6)		16 (20.3)	6 (20.0)	9 (64.3)	<.001
Maternal age, mean (SD), y	28.7 (5.7	30.4 (4.5		30.4 (4.3	29.4 (4.5	32.5 (5.1	.04
Birth weight, mean (SD), kg	3.45 (0.46	3.15 (0.42)		3.19 (0.46)	3.09 (0.34)	3.07 (0.30)	<.001
Gestational age at birth, mean (SD), wk	39.20 (0.98)	39.02 (1.05)		39.10 (1.05)	38.81 (0.94)	39.13 (1.26)	.23
Head circumference at birth, mean (SD), cm	34.71 (1.36)	33.97 (1.38)		34.17 (1.40)	33.51 (1.34)	34.11 (1.15)	<.001
Maternal education
<High school	5 (6.7)	29 (23.0)		17 (21.5)	11 (36.7)	1 (7.1)	<.001
High school	24 (32.0)	50 (39.7)		33 (41.8)	11 (36.7)	6 (42.9)
<College	23 (31.7)	35 (27.8)		22 (27.8)	7 (23.3)	3 (21.4)
≥College	21 (28.0)	7 (5.6)		3 (3.8)	1 (3.3)	3 (21.4)
Unknown	2 (2.7)	5 (4.0)		4 (5.1)	0 (0.0)	1 (7.1)
Annual household income, $
<15 000	15 (20.00)	39 (30.95)		9 (11.39)	26 (86.67)	3 (21.43)	<.001
15 000-24 999	11 (14.67)	24 (19.05)		6 (7.59)	15 (50.00)	3 (21.43)
25 000-49 999	16 (21.33)	24 (19.05)		4 (5.06)	17 (56.67)	3 (21.43)
50 000-99 999	16 (21.33)	12 (9.52)		1 (1.27)	9 (30.00)	1 (7.14)
100 000-149 999	5 (6.67)	7 (5.56)		3 (3.80)	3 (10.00)	1 (7.14)
150 000-199 999	3 (4.00)	2 (1.59)		1 (1.27)	1 (3.33)	0
≥200 000	2 (2.67)	3 (2.38)		1 (1.27)	0	2 (14.29)
Refused to disclose	1 (1.33)	3 (2.38)		0	2 (6.67)	1 (7.14)
Unknown	5 (6.67)	9 (7.14)		5 (6.33)	3 (10.00)	0
Race and ethnicity
Asian, American Indian or Alaskan Native, or Native Hawaiian or Other Pacific Islander	1 (1.33)	0		0	0	0	.03
Hispanic	3 (4.00)	0		0	0	0
Non-Hispanic Black	13 (17.33)	21 (16.67		9 (11.39)	2 (6.67)	10 (71.43)
Non-Hispanic White	56 (74.67)	105 (83.33)		70 (88.61)	28 (93.33)	4 (28.57)
Unknown	2 (2.67)	0		0	0	0
Apgar score, mean (SD), min
1	7.96 (0.78)	7.98 (0.91)		8.03 (0.75)	7.87 (1.41)	8.00 (0.39)	.84
5	8.83 (0.55)	8.90 (0.46)		8.91 (0.51)	8.87 (0.43)	8.93 (0.27)	.31
Mode of delivery
Vaginal	47 (62.7)	84 (66.7)		56 (70.9)	15 (50.0)	12 (85.7)	.67
Cesarean	28 (37.3)	42 (33.3)		23 (29.1)	15 (50.0)	2 (14.3)	.67
Postmenstrual age at MRI, mean (SD), wk	42.76 (2.05)	42.72 (2.06)		42.82 (2.03)	42.99 (2.10)	41.45 (1.55)	.77

Open in a new tab

Abbreviations: MRI, magnetic resonance imaging; NA, not applicable.

^{^a}

Differences between neonates who were unexposed and opioid exposed were evaluated based on t tests for continuous variables and χ² tests for categorical variables.

Table 2. Neonate Opioid Exposure by Self-Reported Maternal Medication Use During Pregnancy.

Medication	Neonates, No. (%) (N = 201)
Medication	Opioid exposed (n = 126)	Unexposed (n = 75)
Buprenorphine	81 (64.3)	0
Methadone	32 (25.4)	0
Oxycodone^a	23 (18.3)	1 (1.3)^b
Gabapentin	20 (15.9)	0
Fentanyl	15 (11.9)	0
Selective serotonin reuptake inhibitor	13 (10.3)	11 (14.7)
Benzodiazepine	12 (9.5)	1 (1.3)
Amphetamine	8 (6.3)	0
Psychostimulant^c	7 (5.5)	0
Hydrocodone	5 (3.9)	0
Hydromorphone	4 (3.2)	0
Morphine	3 (2.4)	0
Heroin	2 (1.6)	0
Cocaine	2 (1.6)	0
Muscle relaxant	1 (0.8)	3 (4.0)
Other	15 (11.9)	2 (2.7)

Open in a new tab

^{^a}

One mother of a neonate in the control group reported minimal oxycodone use after an injury in the second trimester. Maternal urine toxicology screen at delivery and neonatal umbilical cord toxicology were negative for all substances.

^{^b}

Muscle relaxants included cyclobenzaprine.

^{^c}

Psychostimulants included lisdexamfetamine (Vyvanse; Takeda Pharmaceuticals), dextroamphetamine and amphetamine (Adderall; Teva Pharmaceuticals), and methylphenidate (Ritalin; Novartis).

Primary Analysis

Limbic System WM DTI Metrics

Results from the ANCOVA revealed significant differences in WM microstructure of the limbic system between neonates who were unexposed and opioid exposed, particularly for lower MD (β = −0.06 × 10⁻³ mm²/s [95% CI, −0.11 to −0.01 × 10⁻³ mm²/s]; FDR P = .04) and RD (β = −0.07 × 10⁻³ mm²/s [95% CI, −0.12 to −0.02 × 10⁻³ mm²/s]; FDR P = .04) in the fornix of neonates who were opioid exposed compared with unexposed (Figure 2A; Table 3). DTI differences between neonates who were unexposed and exposed are displayed in Table 3. See eFigures 7 to 10 in Supplement 1 for box plots of adjusted FA, MD, RD, and AD metrics of WM limbic system tracts in neonates who were opioid exposed vs unexposed.

Figure 2. — Limbic areas displayed in red represent lower diffusion metrics in neonates who were opioid exposed compared with unexposed. Only significant false discovery rate–corrected differences are shown.

Table 3. DTI Differences Between Neonates Who Were Unexposed and Exposed.

Region of interest	DTI metric	Mean (SD), × 10⁻³ mm²/s^a		DTI difference, β (95% CI), × 10⁻³ mm²/s^a	ANCOVA uncorrected P value
Region of interest	DTI metric	Unexposed	Opioid exposed	DTI difference, β (95% CI), × 10⁻³ mm²/s^a	ANCOVA uncorrected P value
WM
Cingulum (cingulate part)	FA^b	0.16 (0.02)	0.17 (0.03)	0.00 (−0.01 to 0.01)	.59
	MD	1.22 (0.06)	1.22 (0.06)	0.00 (−0.03 to 0.02)	.63
	AD	1.09 (0.07)	1.09 (0.07)	−0.01 (−0.03 to 0.02)	.58
	RD	1.46 (0.07)	1.47 (0.10)	0.00 (−0.03 to 0.03)	.97
Cingulum (hippocampal part)	FA^b	0.16 (0.03)	0.16 (0.03)	0.00 (−0.01 to 0.01)	.88
	MD	1.20 (0.06)	1.20 (0.07)	0.00 (−0.03 to 0.02)	.76
	AD	1.08 (0.07)	1.08 (0.08)	0.00 (−0.03 to 0.03)	.90
	RD	1.43 (0.06)	1.42 (0.08)	−0.01 (−0.04 to 0.01)	.34
Fornix	FA^b	0.26 (0.02)	0.26 (0.03)	0.00 (−0.01 to 0.01)	.56
	MD	1.61 (0.16)	1.57 (0.13)	−0.06 (−0.11 to −0.01)	.01
	AD	1.39 (0.16)	1.35 (0.13)	−0.05 (−0.10 to 0.00)	.05
	RD	2.02 (0.16)	1.98 (0.14)	−0.07 (−0.12 to −0.02)	.009
Stria terminalis	FA^b	0.26 (0.03)	0.26 (0.02)	0.00 (−0.01 to 0.01)	.52
	MD	1.18 (0.07)	1.17 (0.08)	−0.01 (−0.04 to 0.02)	.48
	AD	0.98 (0.07)	0.98 (0.08)	0.00 (−0.03 to 0.02)	.85
	RD	1.56 (0.12)	1.55 (0.11)	−0.02 (−0.06 to 0.02)	.32
GM
Amygdala	FA^b	0.16 (0.02)	0.16 (0.02)	0.00 (−0.01 to 0.01)	.66
	MD	1.17 (0.09)	1.18 (0.11)	0.02 (−0.02 to 0.06)	.34
	AD	1.08 (0.09)	1.09 (0.12)	0.02 (−0.02 to 0.06)	.37
	RD	1.38 (0.11)	1.38 (0.15)	0.00 (−0.05 to 0.05)	.90
Cingular gyrus	FA^b	0.11 (0.01)	0.11 (0.01)	0.00 (0.00 to 0.01)	.17
	MD	1.26 (0.05)	1.26 (0.05)	0.00 (−0.02 to 0.02)	.92
	AD	1.18 (0.05)	1.18 (0.05)	0.00 (−0.02 to 0.02)	.85
	RD	1.40 (0.06)	1.41 (0.06)	0.00 (−0.02 to 0.02)	.90
Entorhinal cortex	FA^b	0.11 (0.02)	0.11 (0.02)	0.00 (−0.01 to 0.00)	.39
	MD	1.70 (0.27)	1.73 (0.25)	0.04 (−0.05 to 0.13)	.44
	AD	1.60 (0.28)	1.62 (0.25)	0.04 (−0.05 to 0.13)	.42
	RD	1.88 (0.28)	1.9 (0.26)	0.04 (−0.06 to 0.13)	.45
Hippocampus	FA^b	0.13 (0.02)	0.13 (0.02)	0.01 (0.00 to 0.01)	.15
	MD	1.23 (0.07)	1.23 (0.07)	−0.01 (−0.03 to 0.02)	.62
	AD	1.15 (0.06)	1.14 (0.08)	−0.01 (−0.04 to 0.01)	.37
	RD	1.40 (0.09)	1.4 (0.08)	−0.01 (−0.04 to 0.02)	.69
Insular cortex	FA^b	0.09 (0.01)	0.09 (0.01)	0.00 (0.00 to 0.00)	.79
	MD	1.53 (0.13)	1.51 (0.13)	−0.03 (−0.07 to 0.02)	.26
	AD	1.44 (0.12)	1.43 (0.13)	−0.02 (−0.06 to 0.03)	.41
	RD	1.70 (0.15)	1.67 (0.14)	−0.03 (−0.08 to 0.02)	.19
Parahippocampal gyrus	FA^b	0.10 (0.02)	0.11 (0.02)	0.01 (0.00 to 0.01)	.05
	MD	1.22 (0.05)	1.23 (0.06)	0.02 (−0.01 to 0.04)	.14
	AD	1.14 (0.05)	1.15 (0.07)	0.01 (−0.01 to 0.03)	.23
	RD	1.38 (0.07)	1.39 (0.08)	0.02 (−0.01 to 0.04)	.24
Thalamus	FA^b	0.17 (0.01)	0.18 (0.02)	0.01 (0.00 to 0.01)	.04
	MD	1.05 (0.04)	1.05 (0.07)	0.00 (−0.02 to 0.02)	.87
	AD	0.94 (0.04)	0.94 (0.07)	0.00 (−0.03 to 0.02)	.70
	RD	1.27 (0.06)	1.27 (0.09)	0.00 (−0.02 to 0.03)	.78

Open in a new tab

Abbreviations: AD, axial diffusivity; ANCOVA, analysis of covariance; DTI, diffusion tensor imaging; FA, fractional anisotropy; GM, gray matter; MD, mean diffusivity; RD, radial diffusivity; WM, white matter.

^{^a}

Adjusted DTI measures in neonates who were opioid exposed and unexposed are displayed. The unexposed group was coded as 0 and served as the reference (baseline) category.

^{^b}

FA is dimensionless.

Limbic System GM DTI Metrics

Results from the ANCOVA revealed higher FA (β = 0.01 [95% CI, 0.00-0.01]; P = .04) in the thalamus of neonates who were opioid exposed compared with unexposed (Figure 2A; Table 3). This finding did not survive FDR correction. See eFigures 11 to 14 in Supplement 1 for box plots of adjusted FA, MD, RD, and AD metrics of GM limbic system structures in neonates who were opioid exposed vs unexposed.

Secondary Analysis

Limbic System WM and GM DTI Metrics in Subgroups

ANCOVA results of the opioid-exposed subgroup analysis demonstrated significant differences in the FA in the left thalamus and bilaterally in the hippocampal part of the cingulum. MD and RD were significantly different in the right hippocampus and right amygdala and bilaterally in the parahippocampal gyrus, and AD showed significant differences only in the parahippocampal gyrus, bilaterally (eFigure 15 in Supplement 1). For example, neonates with buprenorphine had higher MD in the left parahippocampal gyrus compared with neonates who were not exposed to opioids (β = 0.02 × 10⁻³ mm²/s [95% CI, 0.00 to 0.04 × 10⁻³ mm²/s]), and neonates with exposure to other opioids had higher FA in the left thalamus compared with those who were unexposed (β = 0.01 [95% CI, 0.01 to 0.02). These subgroup findings are exploratory and should be interpreted cautiously given that no correction for multiple comparisons was applied. Results of post hoc analyses are described in the eResults in Supplement 1.

Discussion

In this large prospective case-control study of neonates who were antenatally opioid exposed and controls who were unexposed, we investigated limbic system microstructure due to the known association of opioid exposure with the limbic system in adults with opioid use disorder and in animal models. Microstructural alterations within the limbic system of neonates who were antenatally opioid exposed compared with unexposed were found in the fornix. Specifically, we found significantly lower MD and RD in the fornix of neonates who were antenatally opioid exposed. No other limbic system structure differences remained significant after FDR correction. Subgroup exploratory analyses of neonates who were opioid exposed revealed sensitivity to type of opioid exposure in the right thalamus, bilateral hippocampal part of the cingulum, right hippocampus, bilateral parahippocampal gyrus, and right amygdala.

Our findings highlight that the fornix is a limbic WM structure at risk in the developing brain exposed to opioids. The fornix is a major WM tract connecting various nodes of the limbic system and is among the first WM tracts to develop in utero, identifiable in the fetal brain as early as at 13 weeks of gestation.^39,40 These findings confirm the susceptibility of this WM tract to opioid exposure during pregnancy and reflect disrupted maturation of the fornix. Changes in RD have been associated with altered membrane permeability and myelination densities, while lower MD in the developing tracts likely reflects less extracellular space, possibly due to cellular swelling and an accelerated myelination process.^41,42,43,44 Disruption in the timing of myelination and neuroinflammation has been previously reported in rat pups with antenatal opioid exposure.^4,45,46

The risk posed to the fornix in the neonatal brain by opioid exposure may stem from the high density of μ-opioid receptors in myelin-forming oligodendrocytes.⁴⁷ Animal studies investigating the effects of antenatal buprenorphine and methadone exposure have shown alterations in myelination processes.^4,46 Methadone and low doses of buprenorphine act similarly given that they activate μ-opioid receptors, which stimulate oligodendrocyte maturation.⁴ In contrast, high doses of buprenorphine are more likely to activate nociceptin and orphanin FQ peptide receptors, which in turn inhibit oligodendrocyte maturation.⁴⁸ Disruption in the precise timing of oligodendrocyte maturation can result in precocious or delayed myelination, altering development in neuronal connectivity.⁴⁷ A disrupted maturation process in the fornix of neonates who were opioid exposed may be associated with increased myelinated axons with thinner myelin sheaths, which has been reported in antenatally buprenorphine–exposed rat pups.⁴⁶ Pharmacokinetic differences between buprenorphine and methadone may underlie the distinct diffusion patterns observed in limbic structures across opioid-specific cohorts.

We also found that GM structures within the limbic system, such as the left parahippocampal gyrus and right amygdala, showed selective risk to buprenorphine exposure compared with methadone. The altered microstructure of the GM of the limbic system of neonates who were buprenorphine exposed may be due to the high presence of nociceptin and orphanin FQ peptide receptors in limbic GM.¹¹ In contrast, previous studies in infants exposed to methadone prenatally showed microstructural changes mainly in major WM tracts, such as the internal capsule, inferior longitudinal fasciculi, and superior longitudinal fasciculi, suggesting a higher sensitivity in major WM tracts to methadone.^20,21 Ongoing studies are needed to further investigate the role of specific medications for opioid use disorder on brain microstructure to better inform medication choice during pregnancy and its association with the developing brain.

Limitations

While this large observational study advances our understanding of the association of antenatal opioid exposure with brain microstructure, it has some limitations. First, the DTI voxel resolution was not isotropic for all scans, which may have caused loss of sensitivity or bias in our diffusion metrics.⁴⁹ However, because the same voxel resolution was used at each site for neonates who were opioid exposed and unexposed, observed group differences were likely robust to this source of error. Second, due to the unavailability of all B0 images with reversed phase encoding to correct for distortion, a nonlinear registration to high-resolution anatomical images was used to correct for EPI artifacts. To account for these variations, the type of EPI distortion correction was added as a covariate in our ANCOVA analysis. Sensitivity analyses demonstrated that primary findings were robust and replicated when using data corrected with only T2-based ANTs registration. Third, findings suggest limited sensitivity of conventional DWI metrics to detect antenatal opioid exposure–associated outcomes. Future studies will explore more advanced diffusion models that better depict tissue microstructure and complex fiber architecture (ie, crossing fibers), such as connectome-based analysis and tractography.⁵⁰ Fourth, long-term functional outcomes of reported microstructural changes are still unclear. Longitudinal measures of limbic system maturational trajectories in children who were opioid exposed are currently underway as part of the OBOE study.

Conclusions

In this case-control study of neonates who were opioid exposed and unexposed, we found altered WM microstructure in the fornix of the opioid-exposed group. Subgroup exploratory analyses revealed that limbic system microstructural patterns differed by the type of opioid exposure. Differential associations of buprenorphine vs methadone warrant further investigation due to potential implications on the choice of medication for opioid use disorder during pregnancy. These findings will be validated through longitudinal MRI studies and neurobehavioral follow-up.

Supplement 1.

eTable. ANCOVA Results With and Without Influential Outliers for Limbic System White Matter Tracts

eFigure 1. Newborn Optimized Preprocessing Pipeline

eFigure 2. Flowchart of Study Participants

eFigure 3. Limbic System ENA Atlas Eroded ROIs.

eFigure 4. Site Differences Before and After Harmonization

eFigure 5. Site Density Plots Before and After Harmonization

eFigure 6. Group Density Plots Before and After Harmonization

eFigure 7. Box Plots of Adjusted FA of WM Limbic System Tracts in Neonates Who Were Opioid Exposed vs Unexposed

eFigure 8. Box Plots of Adjusted MD of WM Limbic System Tracts in Neonates Who Were Opioid Exposed vs Unexposed

eFigure 9. Box Plots of Adjusted AD of WM Limbic System Tracts in Neonates Who Were Opioid Exposed vs Unexposed

eFigure 10. Box Plots of Adjusted RD of WM Limbic System Tracts in Neonates Who Were Opioid Exposed vs Unexposed

eFigure 11. Box Plots of Adjusted FA of GM Limbic System Tracts in Neonates Who Were Opioid Exposed vs Unexposed

eFigure 12. Box Plots of Adjusted MD of GM Limbic System Tracts in Neonates Who Were Opioid Exposed vs Unexposed

eFigure 13. Box Plots of Adjusted AD of GM Limbic System Tracts in Neonates Who Were Opioid Exposed vs Unexposed

eFigure 14. Box Plots of Adjusted RD of GM Limbic System Tracts in Neonates Who Were Opioid Exposed vs Unexposed

eFigure 15. DTI Differences Between Subgroups of Neonates Who Were Opioid Exposed vs Unexposed

eResults.

jamanetwopen-e2621603-s001.pdf^{(2.1MB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2621603-s002.pdf^{(44.7KB, pdf)}

References

1.Judd D, King CR, Galke C. The opioid epidemic: a review of the contributing factors, negative consequences, and best practices. Cureus. 2023;15(7):e41621. doi: 10.7759/cureus.41621 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.West KD, Ali MM, Blanco M, Natzke B, Nguyen L. Prenatal substance exposure and neonatal abstinence syndrome: state estimates from the 2016-2020 Transformed Medicaid Statistical Information System. Matern Child Health J. 2023;27(1)(suppl 1):14-22. doi: 10.1007/s10995-023-03670-z [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Ko JY, D’Angelo DV, Haight SC, et al. Vital signs: prescription opioid pain reliever use during pregnancy—34 U.S. jurisdictions, 2019. MMWR Morb Mortal Wkly Rep. 2020;69(28):897-903. doi: 10.15585/mmwr.mm6928a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Vestal-Laborde AA, Eschenroeder AC, Bigbee JW, Robinson SE, Sato-Bigbee C. The opioid system and brain development: effects of methadone on the oligodendrocyte lineage and the early stages of myelination. Dev Neurosci. 2014;36(5):409-421. doi: 10.1159/000365074 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Gibson JM, Chu T, Zeng W, et al. Perinatal methadone exposure attenuates myelination and induces oligodendrocyte apoptosis in neonatal rat brain. Exp Biol Med (Maywood). 2022;247(12):1067-1079. doi: 10.1177/15353702221090457 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Rosenfeld CS. The placenta as a target of opioid drugs†. Biol Reprod. 2022;106(4):676-686. doi: 10.1093/biolre/ioac003 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Schaefer CP, Tome ME, Davis TP. The opioid epidemic: a central role for the blood brain barrier in opioid analgesia and abuse. Fluids Barriers CNS. 2017;14(1):32. doi: 10.1186/s12987-017-0080-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Caritis SN, Panigrahy A. Opioids affect the fetal brain: reframing the detoxification debate. Am J Obstet Gynecol. 2019;221(6):602-608. doi: 10.1016/j.ajog.2019.07.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Le Merrer J, Becker JA, Befort K, Kieffer BL. Reward processing by the opioid system in the brain. Physiol Rev. 2009;89(4):1379-1412. doi: 10.1152/physrev.00005.2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Neugebauer V, Presto P, Yakhnitsa V, Antenucci N, Mendoza B, Ji G. Pain-related cortico-limbic plasticity and opioid signaling. Neuropharmacology. 2023;231:109510. doi: 10.1016/j.neuropharm.2023.109510 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Fulford AJ, Keskes S. Limbic circuit connectivity and the stress response: new insights into the mammalian nociceptin peptide system. Vitam Horm. 2019;111:131-145. doi: 10.1016/bs.vh.2019.05.002 [DOI] [PubMed] [Google Scholar]
12.Liu J, Qin W, Yuan K, et al. Interaction between dysfunctional connectivity at rest and heroin cues-induced brain responses in male abstinent heroin-dependent individuals. PLoS One. 2011;6(10):e23098. doi: 10.1371/journal.pone.0023098 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Lin JC, Chu LF, Stringer EA, et al. One month of oral morphine decreases gray matter volume in the right amygdala of individuals with low back pain: confirmation of previously reported magnetic resonance imaging results. Pain Med. 2016;17(8):1497-1504. doi: 10.1093/pm/pnv047 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Upadhyay J, Maleki N, Potter J, et al. Alterations in brain structure and functional connectivity in prescription opioid-dependent patients. Brain. 2010;133(Pt 7):2098-2114. doi: 10.1093/brain/awq138 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Rajmohan V, Mohandas E. The limbic system. Indian J Psychiatry. 2007;49(2):132-139. doi: 10.4103/0019-5545.33264 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Finnegan LP, Connaughton JF Jr, Kron RE, Emich JP. Neonatal abstinence syndrome: assessment and management. Addict Dis. 1975;2(1-2):141-158. [PubMed] [Google Scholar]
17.Kocherlakota P. Neonatal abstinence syndrome. Pediatrics. 2014;134(2):e547-e561. doi: 10.1542/peds.2013-3524 [DOI] [PubMed] [Google Scholar]
18.Grecco GG, Shahid SS, Atwood BK, Wu YC. Alterations of brain microstructures in a mouse model of prenatal opioid exposure detected by diffusion MRI. Sci Rep. 2022;12(1):17085. doi: 10.1038/s41598-022-21416-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Dudley JA, Nagaraj UD, Merhar S, et al. DTI of opioid-exposed fetuses using combat harmonization: a bi-institutional study. AJNR Am J Neuroradiol. 2023;44(9):1084-1089. doi: 10.3174/ajnr.A7951 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Monnelly VJ, Anblagan D, Quigley A, et al. Prenatal methadone exposure is associated with altered neonatal brain development. Neuroimage Clin. 2017;18:9-14. doi: 10.1016/j.nicl.2017.12.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Walhovd KB, Watts R, Amlien I, Woodward LJ. Neural tract development of infants born to methadone-maintained mothers. Pediatr Neurol. 2012;47(1):1-6. doi: 10.1016/j.pediatrneurol.2012.04.008 [DOI] [PubMed] [Google Scholar]
22.Walhovd KB, Westlye LT, Moe V, et al. White matter characteristics and cognition in prenatally opiate- and polysubstance-exposed children: a diffusion tensor imaging study. AJNR Am J Neuroradiol. 2010;31(5):894-900. doi: 10.3174/ajnr.A1957 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Reynolds JE, Grohs MN, Dewey D, Lebel C. Global and regional white matter development in early childhood. Neuroimage. 2019;196:49-58. doi: 10.1016/j.neuroimage.2019.04.004 [DOI] [PubMed] [Google Scholar]
24.Bach M, Laun FB, Leemans A, et al. Methodological considerations on tract-based spatial statistics (TBSS). Neuroimage. 2014;100:358-369. doi: 10.1016/j.neuroimage.2014.06.021 [DOI] [PubMed] [Google Scholar]
25.Yoshida S, Oishi K, Faria AV, Mori S. Diffusion tensor imaging of normal brain development. Pediatr Radiol. 2013;43(1):15-27. doi: 10.1007/s00247-012-2496-x [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Le Bihan D. Looking into the functional architecture of the brain with diffusion MRI. Nat Rev Neurosci. 2003;4(6):469-480. doi: 10.1038/nrn1119 [DOI] [PubMed] [Google Scholar]
27.Bann CM, Newman JE, Poindexter B, et al. Outcomes of babies with opioid exposure (OBOE): protocol of a prospective longitudinal cohort study. Pediatr Res. 2023;93(6):1772-1779. doi: 10.1038/s41390-022-02279-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Bann CM, Newman JE, Poindexter B, et al. Correction to: outcomes of babies with opioid exposure (OBOE): protocol of a prospective longitudinal cohort study. Pediatr Res. 2023;94(4):1581. doi: 10.1038/s41390-023-02662-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Tournier JD, Smith R, Raffelt D, et al. MRtrix3: a fast, flexible and open software framework for medical image processing and visualisation. Neuroimage. 2019;202:116137. doi: 10.1016/j.neuroimage.2019.116137 [DOI] [PubMed] [Google Scholar]
30.Andersson JL, Skare S, Ashburner J. How to correct susceptibility distortions in spin-echo echo-planar images: application to diffusion tensor imaging. Neuroimage. 2003;20(2):870-888. doi: 10.1016/S1053-8119(03)00336-7 [DOI] [PubMed] [Google Scholar]
31.Smith SM, Jenkinson M, Woolrich MW, et al. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage. 2004;23(suppl 1):S208-S219. doi: 10.1016/j.neuroimage.2004.07.051 [DOI] [PubMed] [Google Scholar]
32.Huntenburg JM. Evaluating Nonlinear Coregistration of BOLD EPI and T1w Images. Freie Universität Berlin; 2014. [Google Scholar]
33.Andersson JLR, Graham MS, Zsoldos E, Sotiropoulos SN. Incorporating outlier detection and replacement into a non-parametric framework for movement and distortion correction of diffusion MR images. Neuroimage. 2016;141:556-572. doi: 10.1016/j.neuroimage.2016.06.058 [DOI] [PubMed] [Google Scholar]
34.Andersson JLR, Sotiropoulos SN. An integrated approach to correction for off-resonance effects and subject movement in diffusion MR imaging. Neuroimage. 2016;125:1063-1078. doi: 10.1016/j.neuroimage.2015.10.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Blesa M, Galdi P, Sullivan G, et al. Peak width of skeletonized water diffusion MRI in the neonatal brain. Front Neurol. 2020;11:235. doi: 10.3389/fneur.2020.00235 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Oishi K, Mori S, Donohue PK, et al. Multi-contrast human neonatal brain atlas: application to normal neonate development analysis. Neuroimage. 2011;56(1):8-20. doi: 10.1016/j.neuroimage.2011.01.051 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Fortin JP, Parker D, Tunç B, et al. Harmonization of multi-site diffusion tensor imaging data. Neuroimage. 2017;161:149-170. doi: 10.1016/j.neuroimage.2017.08.047 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Kluyver T, Ragan-Kelley B, Perez F, et al. Jupyter Notebooks—a publishing format for reproducible computational workflows. In: Loizides F, Schmidt B, eds. Positioning and Power in Academic Publishing: Players, Agents and Agendas: Proceedings of the 20th International Conference on Electronic Publishing. IOS Press; 2016:87-90. doi: 10.3233/978-1-61499-649-1-87 [DOI] [Google Scholar]
39.Huang H, Zhang J, Wakana S, et al. White and gray matter development in human fetal, newborn and pediatric brains. Neuroimage. 2006;33(1):27-38. doi: 10.1016/j.neuroimage.2006.06.009 [DOI] [PubMed] [Google Scholar]
40.Huang H, Xue R, Zhang J, et al. Anatomical characterization of human fetal brain development with diffusion tensor magnetic resonance imaging. J Neurosci. 2009;29(13):4263-4273. doi: 10.1523/JNEUROSCI.2769-08.2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Stanisz GJ. Diffusion MR in biological systems: tissue compartments and exchange. Isr J Chem. 2003;43(1-2):33-44. doi: 10.1560/E0WU-7FFH-31M6-VLYT [DOI] [Google Scholar]
42.Concha L. A macroscopic view of microstructure: using diffusion-weighted images to infer damage, repair, and plasticity of white matter. Neuroscience. 2014;276:14-28. doi: 10.1016/j.neuroscience.2013.09.004 [DOI] [PubMed] [Google Scholar]
43.Beaulieu C. The basis of anisotropic water diffusion in the nervous system—a technical review. NMR Biomed. 2002;15(7-8):435-455. doi: 10.1002/nbm.782 [DOI] [PubMed] [Google Scholar]
44.Stanisz G, Henkelman RM. Effects of cellular swelling on diffusion in white matter. Proc Intl Soc Magn Reson Med. 2001:350. [Google Scholar]
45.Jantzie LL, Maxwell JR, Newville JC, et al. Prenatal opioid exposure: the next neonatal neuroinflammatory disease. Brain Behav Immun. 2020;84:45-58. doi: 10.1016/j.bbi.2019.11.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Sanchez ES, Bigbee JW, Fobbs W, Robinson SE, Sato-Bigbee C. Opioid addiction and pregnancy: perinatal exposure to buprenorphine affects myelination in the developing brain. Glia. 2008;56(9):1017-1027. doi: 10.1002/glia.20675 [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Velasco B, Mohamed E, Sato-Bigbee C. Endogenous and exogenous opioid effects on oligodendrocyte biology and developmental brain myelination. Neurotoxicol Teratol. 2021;86:107002. doi: 10.1016/j.ntt.2021.107002 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Lutfy K, Cowan A. Buprenorphine: a unique drug with complex pharmacology. Curr Neuropharmacol. 2004;2(4):395-402. doi: 10.2174/1570159043359477 [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Soares JM, Marques P, Alves V, Sousa N. A hitchhiker’s guide to diffusion tensor imaging. Front Neurosci. 2013;7:31. doi: 10.3389/fnins.2013.00031 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Verschuur AS, Tax CMW, Boomsma MF, et al. Feasibility study to unveil the potential: considerations of constrained spherical deconvolution tractography with unsedated neonatal diffusion brain MRI data. Front Radiol. 2024;4:1416672. doi: 10.3389/fradi.2024.1416672 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials