Sex-Specific Alterations of White Matter Developmental Trajectories in Infants With Prenatal Exposure to Methamphetamine and Tobacco

Abstract

Importance: Methamphetamine is a common illicit drug used worldwide. Methamphetamine and/or tobacco use by pregnant women remains prevalent. However, little is known about the effect of comorbid methamphetamine and tobacco use on human fetal brain development.

Objective: To investigate whether microstructural brain abnormalities reported in children with prenatal methamphetamine and/or tobacco exposure are present at birth before childhood environmental influences.

Design, setting, and participants: A prospective, longitudinal study was conducted between September 17, 2008, and February 28, 2015, at an ambulatory academic medical center. A total of 752 infant-mother dyads were screened and 139 of 195 qualified neonates were evaluated (36 methamphetamine/tobacco exposed, 32 tobacco exposed, and 71 unexposed controls). They were recruited consecutively from the community.

Exposures: Prenatal methamphetamine and/or tobacco exposure.

Main outcomes and measures: Quantitative neurologic examination and diffusion tensor imaging performed 1 to 3 times through age 4 months; diffusivities and fractional anisotropy (FA) assessed in 7 white matter tracts and 4 subcortical brain regions using an automated atlas-based method.

Results: Of the 139 infants evaluated, 72 were female (51.8%); the mean (SE) postmenstrual age at baseline was 41.5 (0.27) weeks. Methamphetamine/tobacco-exposed infants showed delayed developmental trajectories on active muscle tone (group × age, P < .001) and total neurologic scores (group × age, P = .01) that normalized by ages 3 to 4 months. Only methamphetamine/tobacco-exposed boys had lower FA (group × age, P = .02) and higher diffusivities in superior (SCR) and posterior corona radiatae (PCR) (group × age × sex, P = .002; group × age × sex, P = .01) at baseline that normalized by age 3 months. Only methamphetamine/tobacco- and tobacco-exposed girls showed persistently lower FA in anterior corona radiata (ACR) (group, P = .04; group × age × sex, P = .01). Tobacco-exposed infants showed persistently lower axial diffusion in the thalamus and internal capsule across groups (P = .02).

Conclusions and relevance: Prenatal methamphetamine/tobacco exposure may lead to delays in motor development, with less coherent fibers and less myelination in SCR and PCR only in male infants, but these abnormalities may normalize by ages 3 to 4 months after cessation of stimulant exposure. In contrast, persistently less coherent ACR fibers were observed in methamphetamine/tobacco- and tobacco-exposed girls, possibly from increased dendritic branching or spine density due to epigenetic influences. Persistently lower diffusivity in the thalamus and internal capsule of all tobacco-exposed infants suggests aberrant axonal development. Collectively, prenatal methamphetamine and/or tobacco exposure may lead to delayed motor development and white matter maturation in sex- and regional-specific manners.

Figure 1.. Automated Atlas-Based Analyses for Diffusion Tensor Imaging (DTI) From Infants

A, The steps involved in matching the template to the final atlas are illustrated. After the affine transformation, the large deformation diffeomorphic metric mapping (LDDMM)^, was performed to match the new DTI to the neonatal atlas developed for infants,^, which is available at http://www.mristudio.org. The atlas parcellation map (PM) automatically segmented 122 brain regions, yielding diffusivity (mean, radial, and axial) and fractional anisotropy in each region. (Modified with permission from Deshpande et al.) B, Seven major fiber tracts and 4 subcortical regions were selected for the current analyses; the superior longitudinal fasciculus was not included owing to its slow development at this age. JHU indicates Johns Hopkins University.

Figure 2.. Developmental Trajectories of the Amiel-Tison Neurological Assessment at Term (ATNAT) and Diffusion Tensor Imaging Metrics in Superior Corona Radiata

A, ATNAT showing delayed active muscle tone and total scores in methamphetamine/tobacco-exposed infants compared with the other groups. B, Fiber tracts in bilateral superior corona radiata from a 1-month-old infant are shown in the axial and coronal views. C, Fractional anisotropy (FA) increases with age while diffusivities decrease with age in all groups. However, the developmental trajectory in the FA of methamphetamine/tobacco-exposed boys showed a slightly steeper trajectory than the other 2 groups at an earlier age but normalized at later postmenstrual age (PMA), with no group differences in the age-dependent changes among the female infants. Similarly, diffusivities in both the axial and radial directions, and hence mean diffusivity values, were also higher at baseline in the methamphetamine/tobacco-exposed boys compared with the other 2 groups of boys, but no group differences in diffusivities were observed in either direction in the girls. L indicates left; R, right.

Figure 3.. Developmental Trajectories of Diffusivities and Fractional Anisotropy (FA) in Anterior Corona Radiata (ACR) and Posterior Corona Radiata (PCR) Across Groups

Diffusivities decreased with age and FA increased with age across all groups in both fiber tracts. A, Fiber tracts in the PCR from a 1-month-old infant are shown in the axial and coronal views. B, In the PCR, the developmental trajectories of both the mean diffusivity and radial diffusivity in the methamphetamine/tobacco-exposed boys declined slower than those in the tobacco-exposed and unexposed boys, but the age-dependent changes in diffusivities are not significantly different across the groups in the girls. The male methamphetamine/tobacco-exposed infants also had higher mean and radial diffusivities than did male unexposed infants in the PCR (post hoc analysis of covariance, P = .05 for both measures). C, Fiber tracts in the ACR from a 1-month-old infant are shown in the axial and coronal views. D, In the ACR, the developmental trajectories of the FA in the 2 female stimulant-exposed groups remained lower than the FA in the unexposed girls across the age span, but the age-dependent changes in FA were not significantly different across the groups in the boys. PMA indicates postmenstrual age.