Preliminary evidence that prenatal maternal choline supplementation protects the white matter microstructure of infants born to heavy-drinking women

Maiphetlho, Oreneile

Preliminary evidence that prenatal maternal choline supplementation protects the white matter microstructure of infants born to heavy-drinking women

Thesis / Dissertation

2024

Publisher

University of Cape Town

Department

Department of Human Biology

Faculty

Faculty of Health Sciences

Abstract

Prenatal alcohol exposure (PAE) is the most widespread preventable cause of neurocognitive and neurodevelopmental disabilities worldwide, known collectively as fetal alcohol spectrum disorders. It has been shown in a small double-blind, randomized choline supplementation trial that prenatal maternal choline supplementation reduces PAE-related postnatal growth restrictions, cognitive deficits, and regional brain volume reductions in infants born to heavydrinking women. The impact of prenatal maternal choline supplementation on white matter (WM) microstructure – an area particularly vulnerable to PAE – has not been assessed. This project used diffusion tensor imaging (DTI) in 43 neonates born to heavy-drinking mothers enrolled in that choline supplementation trial to examine the effects of prenatal maternal choline supplementation on the microstructure of WM connecting select subcortical regions. Given that pre-myelination (at 30-40 weeks gestational age (GA)) and maturation of the axonal cytoskeleton are categorised by stable fractional anisotropy (FA), decreasing axial diffusivity (AD) and radial diffusivity (RD), we hypothesised that neonates with PAE whose mothers had received choline would demonstrate lower AD and RD than those in the placebo arm, and that lower AD and RD would be associated with better cognitive performance. T1-weighted structural magnetic resonance (sMR) and diffusion tensor (DT) imaging (DTI) data were acquired for each neonate between 1-7 weeks postpartum on a 3T Siemens Allegra. Previously, 17 subcortical brain regions (ROIs) had been manually traced on sMR images in Freeview. After DTI pre-processing, the manually traced ROIs were transformed into the DTI space and used as seeds for full probabilistic tractography. FA, AD and RD were extracted for each identified connection. Visual recognition memory scores on the Fagen test of Infant Intelligence (FTII) assessed at 12 months were also available for 34 participants. We examined differences in the DTI measures between the choline and placebo groups, as well as associations of DTI measures with choline dose and FTII scores using robust linear regression. Choline dose was approximated in 2 ways: (1) using maternal treatment adherence (in % packets consumed), with choline dose set to zero for infants in the placebo arm, and (2) computing the total choline (in grams) consumed by the mothers throughout participation in the study (enrolment through to delivery). A total of 22 neonates (14 choline; 11 boys; mean GA (SD) = 41.8 (2.0) weeks) provided usable, non-biased DTI data and FTII scores were available for 14 of these neonates (8 choline). We examined treatment effects in 15 WM connections that were present in a least 85 % of the neonates. Three neonates in the choline arm whose mothers had low treatment adherence (< 50 %) were excluded from the group difference analyses. Additionally, one choline-treated participant whose average AD value introduced bias was excluded from all analyses that included the DTI measures. After controlling for confounders, FA in the WM connection between the caudate and pallidum on the left was lower in the choline group (p = 0.023) and on the right was inversely associated with choline dose – both in % packets consumed (β (ε) = -0.55 (0.20); p = 0.016) and in grams (β (ε) = -0.51 (0.20); p = 0.024). Moreover, increasing choline dose – in % packets consumed and in grams – was associated with lower AD in 2 WM connections: between the right caudate and right putamen, and right putamen and right pallidum (|β|'s > 0.40; p's < 0.05). Finally, better visual recognition memory on the FTII was associated with lower RD (β (ε) = -0.62 (0.27); p = 0.038) in the connection between the left and right thalami, and with higher RD (β (ε) = 0.54 (0.18); p = 0.009) in the WM connection between the right caudate and right putamen. Although choline-related decreases in FA were observed in the connection between the right caudate and right pallidum, the choline-related decreases in AD in the caudate, putamen and pallidum on the right aligned with our hypothesis. Studies of participants with Williams syndrome and attention-deficit/hyperactivity disorder have speculated that the lower FA observed in their control groups may be accredited to higher levels of axonal branching. Given that choline promotes axonal growth and branching in rodent models, the lower FA observed here may be indicative of choline-related increases in axonal branching in the caudate to pallidum connections. Studies of postnatal WM development have found that AD and RD decrease during the first postnatal year. Given that AD provides information on axonal integrity and organization, the age-related decreases for AD have been accredited to increases in axonal growth. Human studies have demonstrated increases in fibre cross-sections in the first 6 months of life and animal studies have shown the importance of choline in axonal growth and elongation. Thus, the choline-related reductions in AD in connections between the caudate, putamen and pallidum on the right suggest accelerated brain development through the promotion of axonal growth in these WM connections. Axonal pruning may also contribute to decreases in AD. Growth-related increases in axonal diameter may explain the association of higher RD seen in the same connection, right caudate to right putamen, with better visual recognition memory. The negative association between visual recognition memory and RD in the connection between the left and right thalami aligns with our hypothesis and several studies conducted in infants, children and adults. These results provide preliminary evidence that choline is associated with the microstructural properties of subcortical WM connections, specifically between the caudate, putamen and pallidum, which are regions known to be affected by PAE. These findings need to be replication in a larger sample and with unexposed controls. Further research is required to examine the impact of prenatal maternal choline supplementation on cortical WM.

Keywords

Human Biology

Reference:

Collections

Masters

Full item page