Brain microstructural changes support cognitive deficits in HIV uninfected children born to HIV infected mothers

Abstract

Introduction: Antiretroviral therapy (ART) is considered the most effective way to prevent perinatal transmission of human immunodeficiency virus (HIV). However, there is little knowledge about the effect of ART on the brain of HIV uninfected children born to HIV infected mothers (HUC). The current study evaluated the brain's microstructural integrity, and cognitive function in HUC compared to healthy children born to normal mothers (CHNM) and HIV infected children born to HIV infected mothers (HIC) to investigate the effect of in-utero exposure of ART on cerebral gray and white matter.

Materials and methods: Forty nine HIC, 12 HUC and 18 CHNM underwent neuropsychological (NP) assessment and a brain MRI. Diffusion tensor imaging (DTI) data was used to generate fractional anisotropy (FA) and mean diffusivity (MD) maps. Voxel wise comparison for FA and MD was performed between three groups using an analysis of covariance (ANCOVA) including age and sex as covariates, and correction for multiple comparisons (false discovery rate (FDR), p ​< ​0.05 with minimum extended cluster size, 150 voxels). NP test scores were also compared between three groups using ANOVA with Post Hoc Bonferroni multiple comparison corrections (p ​< ​0.05). Significantly changed FA and MD values in different brain regions in HIC and HUC compared to CHNM were used for correlation analysis with NP test scores using Pearson's correlation.

Results: HIC and HUC groups showed significantly decreased NP test scores in various domain compared to CHNM. Significantly lower NP test scores was observed in HIC than those of HUC. HIC showed decreased FA and increased MD in multiple brain sites as compared to both CHNM and HUC. Decreased FA along with both increased and decreased MD in different brain regions was present in HUC compared to CHNM. Both positive and negative correlation of altered FA and MD values from different brain regions in HIC and HUC with NP test scores was observed.

Conclusion: The presence of brain tissue changes and neurocognitive function deficit in absence of HIV infection in HUC indicates that ART may have a detrimental impact on the developing brain. The findings of the current study underscore the need for screening of ART exposed children for neurodevelopment and cognitive abnormalities at an early stage and call for access to early interventions, and nutritional and care programs.

Keywords: Antiretroviral therapies; Brain; Diffusion tensor imaging; Human immunodeficiency virus; Neuropsychological test.

Fig. 1

Effect of human immunodeficiency virus (HIV) and antiretroviral therapy (ART) on blood-brain barrier (BBB), neuron and brain tissue. The brain of healthy children shows intact BBB, and neuron (A). The BBB compromises in HIV infected and ART exposed children, which subsequently leads to brain tissue damage by upregulating the cytokines, pro-inflammatory markers and reactive oxygen species (ROS) (B). BBB is also compromised in children exposed to ART causing neuroinflammation and brain tissue damage (C). More brain tissue damage is obvious in HIV infected and ART exposed children than ART exposed children (B and C).

Fig. 2

Bar plots are showing the neuropsychological test (NP) scores (mean ​± ​SD) in healthy children born to normal mothers (CHNM), HIV uninfected children born to HIV infected mothers (HUC) and HIV infected children born to HIV infected mothers (HIC). ANOVA with Post Hoc Bonferroni multiple comparison corrections (p ​< ​0.05) were used for comparison.

Fig. 3

Voxel-wise one-way ANCOVA findings for fractional anisotropy (FA) between three groups. F-test maps are showing significant difference in FA between three groups (A). T score maps are showing significantly decreased FA value in different brain regions of HIV infected children born to HIV infected mothers (HIC) compared to children born to normal mothers (CHNM) and HIV uninfected children born to HIV infected mothers (HUC) (B, C). T score maps shows decreased FA value in multiple brain sites in HUC children compared to CHNM children (D). ANCOVA (age and gender as covariates with correction for multiple comparisons (false discovery rate, p ​< ​0.05 with minimum extended cluster size, 150 voxels).

Fig. 4

Voxel-wise one-way ANCOVA findings for mean diffusivity (FA) between three groups. F-test maps are showing significant difference in MD between three groups (A). T score maps are showing significantly increased MD value in different brain regions of HIV infected children born to HIV infected mothers (HIC) compared to children born to normal mothers (CHNM) and HIV uninfected children born to HIV infected mothers (HUC) (B and C). T score maps depicts both increased and decreased MD values in HUC compared to CHNM (D). ANCOVA (age and gender as covariates with correction for multiple comparisons (false discovery rate, p ​< ​0.05 with minimum extended cluster size, 150 voxels).