Abstract
This cross-sectional study identifies structural differences of the precentral gyrus among children with reported prenatal opioid exposure compared with children with no reported exposure, controlling for present social factors.
Prenatal opioid exposure is associated with delayed locomotor performance at multiple stages of early child development1,2 and with smaller neuroanatomical structures, such as the basal ganglia.3 The motor cortex, which controls speech and motor skills, may also be vulnerable to drug exposure, but to our knowledge, this has yet to be assessed. Identifying risks of exposure associated with brain structures is critical for prevention and intervention strategies for cognitive effects that can last long after conception.3,4
Most neuroanatomical opioid-related research has focused on heroin, used smaller samples, and failed to control for social factors that are associated with development. Using data from the Adolescent Brain Cognitive Development (ABCD) study,5 we aimed to identify structural differences of the precentral gyrus (motor cortex) among children with reported prenatal opioid exposure compared with children with no reported exposure, controlling for present social factors.
Methods
Using magnetic resonance imaging, we compared cortical volumes, cortical areas, and thicknesses of the precentral gyrus between children prenatally exposed to opioids and nonexposed children. The data were collected through the ABCD study, which also includes developmental questionnaires, reports of parents’ and children’s medical histories, and current sociodemographic information. Parent-reported (or guardian-reported) prenatal opioid exposure was classified as a binary indicator through standardized survey questions about the biological mothers’ prescription medication and drug usage during pregnancy (n = 11 530). Mean differences in the cortical volumes, cortical areas, and thicknesses of the precentral gyri, using Destrieux parcellations from the ABCD study,6 were assessed between groups using appropriate weightings and were calculated with appropriate weighting and statistically tested using multilevel, bivariate regression analyses, with the level of significance set at a 2-tailed P value less than .05. Written informed consent from guardians and written assent from children were collected at study sites, under single institutional review board approval from the University of California, San Diego. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Multivariable regression models controlled for maternal tobacco and alcohol use after learning of pregnancy and current sociodemographic factors (financial risk, race/ethnicity, sex, and single-adult household). Sample size was reduced (6%) for this analysis (n = 10 840) due to missing data. Statistical analyses were performed applying the ABCD sampling design, clustering siblings/twins within families and by study site, with bootstrapping, using Stata software version 14.2 (StataCorp).
Results
Of the 11 530 child participants, 6042.87 (52.4%; 95% CI, 51.4-53.4) were male, and the mean (SD) age was 9.91 (0.63) years. The sample included 6030 children (52.4%; 95% CI, 51.4-53.3) who were white, 2344 (20.4%; 95% CI, 19.7-21.1) who were Hispanic, and 1702 (14.8%; 95% CI, 14.1-15.5) who were black, which was not statistically different between those with and without prenatal opioid exposure (adjusted χ2 = 1.94; P = .10). After assessing maternal medication and drug use during pregnancy, 150 children were identified as having prenatal prescription opioid exposure with 144 having magnetic resonance imaging scans. The mean (SE) cortical volume of the left precentral gyrus was smaller among children with prenatal opioid exposure compared with those without (7329.02 [88.51] mm3 vs 7566.00 [9.85] mm3; difference, 236.98 [89.82] mm3; Cohen d = 0.223) (Table 1). Similarly, the mean (SE) volume of the right precentral gyrus was smaller among those with prenatal opioid exposure compared with those without (7195.65 [88.32] mm3 vs 7481.74 [9.50] mm3; difference, 286.09 [85.84] mm3; Cohen d = 0.271). The surface areas of the precentral gyri were also smaller among children exposed to prescription opioids (Table 1). Total cortical area and volume and thickness of the precentral gyrus were not statistically significantly different between groups. The differences in cortical volumes and cortical areas of gyri remained statistically significant when controlling for sociodemographic factors in the regression models (Table 2) as well prenatal alcohol and tobacco exposures, which were also statistically significant.
Table 1. Precentral Gyrus Cortical Volume, Cortical Surface Area, and Thickness Among Children Prenatally Exposed to Opioids and Nonexposed Children Clustered Among Families by Site From the Adolescent Brain Cognitive Development (ABCD) Studya.
| Measure | ABCD variable name | Mean (SE) | Wald χ2 | P value | Cohen db | |
|---|---|---|---|---|---|---|
| Exposed children (n = 144) | Nonexposed children (n = 11 389) | |||||
| Area, mm2 | ||||||
| Left | mrisdp_331 | 1816.75 (24.06) | 1877.84 (2.82) | 6.45 | .01 | 0.213 |
| Right | mrisdp_405 | 1812.48 (21.41) | 1874.78 (2.81) | 8.47 | .005 | 0.211 |
| Volume, mm3 | ||||||
| Left | mrisdp_482 | 7329.02 (88.51) | 7566.00 (9.85) | 7.17 | .008 | 0.223 |
| Right | mrisdp_556 | 7195.65 (88.32) | 7481.74 (9.50) | 10.49 | .001 | 0.271 |
| Thickness, mm | ||||||
| Left | mrisdp_29 | 2.99 (0.02) | 2.99 (0) | 0.04 | .84 | 0.039 |
| Right | mrisdp_103 | 2.95 (0.02) | 2.97 (0) | 0.73 | .39 | 0.053 |
| Total brain | ||||||
| Total area, mm2 | mrisdp_453 | 184 095.13 (1675.37) | 186 155.14 (166.10) | 1.51 | .22 | 0.112 |
| Total volume, mm3 | mrisdp_604 | 586 655.76 (4959.52) | 594 227.26 (518.8) | 2.33 | .13 | 0.088 |
Bootstrapping was used to calculate SEs (replications = 500) due to unbalanced sample size.
Effect sizes were calculated from means and SDs with applied sampling design and group sample sizes (α = .05).
Table 2. Final Regression Models Showing Differences in the Hemispheric Precentral Gyrus Volume and Area Among 10 833 Children From the Adolescent Brain Cognitive Development Study.
| Factor | Left precentral gyrus | Right precentral gyrus | ||||||
|---|---|---|---|---|---|---|---|---|
| Coefficient (SE) | Standardized coefficient | t Value | P value | Coefficient (SE) | Standardized coefficient | t Value | P value | |
| Area | ||||||||
| Prenatal exposure | ||||||||
| Opioid | −83.37 (21.79) | −0.29 | −3.83 | <.001 | −77.05 (22.82) | −0.26 | −3.38 | <.001 |
| Alcohol | 33.87 (16.97) | 0.12 | 2.00 | .046 | 14.17 (17.05) | 0.05 | 0.83 | .41 |
| Tobacco | −33.58 (12.53) | −0.12 | −2.68 | .007 | −32.60 (14.07) | −0.11 | −2.32 | .02 |
| Black | −151.10 (8.61) | −0.53 | −17.55 | <.001 | −148.10 (8.45) | −0.50 | −17.52 | <.001 |
| Hispanic | −62.64 (6.74) | −0.22 | −9.30 | <.001 | −58.56 (7.26) | −0.20 | −8.07 | <.001 |
| Asian | 8.55 (20.75) | 0.03 | 0.41 | .68 | −21.93 (22.38) | −0.07 | −0.98 | .33 |
| Other race/ethnicity | −33.13 (8.57) | −0.12 | −3.86 | <.001 | −24.92 (9.45) | −0.08 | −2.64 | .008 |
| Financial risk | −10.83 (2.39) | −0.04 | −4.53 | <.001 | −5.82 (2.63) | −0.02 | −2.21 | .03 |
| Single-adult household | 8.61 (6.90) | −0.03 | 1.25 | .21 | 16.33 (6.98) | −0.06 | 2.34 | .02 |
| Sex | 184.98 (5.10) | 0.65 | 36.24 | <.001 | 175.13 (5.14) | 0.60 | 34.06 | <.001 |
| Age | −0.62 (0.34) | 0 | −1.80 | .07 | −0.73 (0.36) | 0 | −2.01 | .045 |
| Constant | 1899.67 (40.01) | −0.29 | 47.48 | <.001 | 1914.97 (43.56) | −0.26 | 43.96 | <.001 |
| Volume | ||||||||
| Prenatal exposure | ||||||||
| Opioid | −211.82 (90.70) | −0.20 | −2.34 | .02 | −239.15 (86.67) | −0.23 | −2.67 | .008 |
| Alcohol | 125.65 (64.38) | 0.19 | 1.95 | .05 | 119.93 (64.41) | 0.11 | 1.86 | .06 |
| Tobacco | −254.80 (46.50) | −0.24 | −5.48 | <.001 | −260.43 (47.72) | −0.25 | −5.96 | <.001 |
| Black | −653.58 (29.51) | −0.61 | −22.15 | <.001 | −650.00 (29.89) | −0.62 | −21.75 | <.001 |
| Hispanic | −318.51 (24.94) | −0.30 | −12.77 | <.001 | −295.79 (25.85) | −0.28 | −11.44 | <.001 |
| Asian | −428.65 (70.84) | −0.40 | −6.05 | <.001 | −550.54 (70.14) | −0.52 | −7.85 | <.001 |
| Other race/ethnicity | −318.40 (33.01) | −0.30 | −9.64 | <.001 | −288.91 (32.86) | −0.27 | −8.79 | <.001 |
| Financial risk | −47.20 (9.01) | −0.04 | −5.24 | <.001 | −33.27 (8.76) | −0.03 | −3.80 | <.001 |
| Single-adult household | 70.67 (24.69) | −0.07 | 2.86 | .004 | 100.84 (25.14) | −0.10 | 4.01 | <.001 |
| Sex | 568.82 (19.40) | 0.53 | 29.32 | <.001 | 534.71 (18.83) | 0.51 | 28.39 | <.001 |
| Age | 3.37 (1.28) | 0 | 2.62 | .009 | 3.80 (1.28) | 0 | 2.96 | .003 |
| Constant | 7122.56 (153.15) | −0.20 | 46.51 | <.001 | 6997.69 (153.22) | −0.23 | 45.67 | <.001 |
Discussion
To our knowledge, this article is among the first to find an association of prenatal prescription opioid exposure with neuroanatomical differences in the brains of children. Analysis of other variables within the ABCD study data set may elucidate the functional, cognitive, and behavioral results of these neuroanatomical differences and may inform future therapeutic models for improving outcomes for these children. With the longitudinal ABCD study, researchers may explore whether these structural differences persist over time.
Future research may investigate whether medications for the treatment of opioid use disorder, like buprenorphine and methadone, result in similar brain changes and assess whether opioid exposure has a dose-dependent effect on gyri size, which was a limitation of our study. In summary, these results suggest a need for clinical screening for prenatal opioid exposure during pregnancy and during the life span as well as reevaluation of risks and benefits of prescription opioids during pregnancy.
References
- 1.Bunikowski R, Grimmer I, Heiser A, Metze B, Schäfer A, Obladen M. Neurodevelopmental outcome after prenatal exposure to opiates. Eur J Pediatr. 1998;157(9):724-730. doi: 10.1007/s004310050923 [DOI] [PubMed] [Google Scholar]
- 2.Hans SL, Jeremy RJ. Postneonatal mental and motor development of infants exposed in utero to opioid drugs. Infant Ment Health J. 2001;22(3):300-315. doi: 10.1002/imhj.1003 [DOI] [Google Scholar]
- 3.Nygaard E, Slinning K, Moe V, Due-Tønnessen P, Fjell A, Walhovd KB. Neuroanatomical characteristics of youths with prenatal opioid and poly-drug exposure. Neurotoxicol Teratol. 2018;68:13-26. doi: 10.1016/j.ntt.2018.04.004 [DOI] [PubMed] [Google Scholar]
- 4.Walhovd KB, Moe V, Slinning K, et al. . Volumetric cerebral characteristics of children exposed to opiates and other substances in utero. Neuroimage. 2007;36(4):1331-1344. doi: 10.1016/j.neuroimage.2007.03.070 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Compton WM, Dowling GJ, Garavan H. Ensuring the best use of data: the Adolescent Brain Cognitive Development study. JAMA Pediatr. 2019;173(9):809-810. doi: 10.1001/jamapediatrics.2019.2081 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hagler DJ Jr, Hatton S, Cornejo MD, et al. . Image processing and analysis methods for the Adolescent Brain Cognitive Development study. Neuroimage. 2019;202:116091. doi: 10.1016/j.neuroimage.2019.116091 [DOI] [PMC free article] [PubMed] [Google Scholar]