Browsing by Author "Holmes, Martha"

Now showing 1 - 10 of 10
  • Thumbnail Image
    Item
    Open Access
    Differences in resting state functional networks in HIV infected and uninfected children at age 7 years
    (2015) Toich, Jadrana; Meintjes, Ernesta M; Holmes, Martha; Taylor, Paul A
    Although early administration of highly active antiretroviral therapy (HAART) in infants provides the brain some protection against HIV damage, few studies have examined the long-term effects of HIV infection and HAART on neurodevelopment, and none have measured their impact on functional brain networks in young children. We use resting state functional magnetic resonance imaging (RS-fMRI) to explore differences in functional connectivity (FC) in HIV infected children stable on HAART and in HIV uninfected children. The 9 resting state networks (RSNs) identified using independent component analysis (ICA) included the visual lingual gyrus, visual occipital gyrus, salience, dorsal attention, auditory, motor, executive control, posterior default mode network (pDMN) and default mode network (DMN) . No significant group level differences were found in any RSNs using ICA. However, seed-based correlation analysis ( SCA ) revealed two regions where uninfected children had a higher FC compared to infected children (p < 0. 05 corrected for multiple comparison); specifically, between a seed in the left cingulate gyrus of the DMN and the left middle frontal gyrus, and between a seed in the right middle frontal gyrus of the executive control network and the right supramarginal gyrus. Consistent with our findings, previous RS-fMRI studies in HIV infected adults have reported reduced connectivity compared to uninfected adults in numerous DMN regions and executive control network. However, in contrast to the adult literature, in which a number of areas within the networks have been implicated, we only observed a focal effect in each of the two RSNs. Given that some of the RSNs are still undergoing major developments at age 7 years (i.e . time of scan for the children), the reduced FC may represent delayed network maturation within the infected cohort , with potential effects on cognitive functioning, information processing and memory recall abilities . Furthermore, positive associations were found between the clinical CD4/CD8 at time of enrollment and two regions within the dorsal attention and auditory networks. These results were independent of treatment arm and suggest that reduced FC in these networks at age 7 years are a result of poor immune function in early infancy (6-8 weeks of age), supporting the notion of in itiating ART immediately in HIV infected infants.
  • Thumbnail Image
    Item
    Open Access
    DTI-based tractographic analysis of white matter alterations in HIV infected children
    (2019) Madzime, Joanah; Jankiewicz, Marcin; Holmes, Martha; Meintjes, Ernesta
    Despite early combination antiretroviral therapy (cART) administration, children born with human immunodeficiency virus (HIV) continue to demonstrate neurodevelopmental abnormalities. Often, there is a link between structural and functional abnormalities. Previously, we found HIV-associated changes in white matter and functional networks in a cohort of 7-year-old HIV infected (HIV+) children who intiatied early cART compared to uninfected controls. To explore possible relationships between these alterations, we used tractography to identify HIV-related abnormalities within structural connections located in functional resting state networks. Within HIV+ children (n=61), we identified white matter (WM) tracts with lower mean fractional anisotropy (FA) and/or higher mean diffusivity (MD) located in several functional networks, including the somatosensory, auditory, salience, default mode network (DMN), motor and basal ganglia networks compared to uninfected controls (n=46). Among the uninfected controls, children born to HIV+ mothers (exposed uninfected, HEU) (n=19) showed WM alterations (higher FA) compared to HIV unexposed uninfected children (HUU) (n=27) within tracts in the posterior DMN, visual (occipital lobe and lingual gyrus), salience and motor networks. The observed WM alterations in HIV+ children point to demyelination/dysmyelination within six networks. Four of these networks – the basal ganglia, default mode, salience and somatosensory – were all found to have altered functional connectivity in a previous study; therefore, these results point to damage or developmental delay in white matter may be related to or responsible for the HIV-associated functional abnormalities. The observed WM alterations in the HEU children suggest that even exposure to HIV and/or antiretroviral therapy (ART) also has long-term effects on axonal integrity in the developing brain.
  • No Thumbnail Available
    Item
    Open Access
    Influence of maternal human immunodeficiency virus (HIV) and antiretroviral (ARV) drugs on neonate neurometabolism.
    (2023) Masina, Nosipho; Holmes, Martha
    The human immunodeficiency virus (HIV) can be vertically transmitted from a woman living with HIV to her child during pregnancy, childbirth, and breastfeeding. Antiretroviral therapy (ART) prevents vertical transmission of HIV by providing prophylaxis to the fetus or infant and lowering the mother's viral load (VL). Due to the prevention of vertical HIV transmission, there is a decrease in infants acquiring HIV vertically. Correspondingly, there is an increasing population of HIV-exposed and uninfected (HEU) infants and children, with the current global estimate of HEU infants and children at around 15 million. While results have been inconsistent, some studies show that HEU children have an increased risk of morbidity and mortality, as well as neurodevelopmental delays across language, motor and cognitive domains, when compared to their HIV-unexposed and uninfected (HUU) counterparts. Several factors have been associated with the higher morbidity and mortality rates exhibited by HEU children. These factors include perinatal and postnatal (during breastfeeding) exposure to maternal HIV and ART, a pro-inflammatory state in the mother, and a compromised maternal immune system. This study aimed to see if the metabolic brain abnormalities seen in older HEU children could be detected in neonates. Based on previous results, we hypothesize lower ratios of glutamate (Glu) to total creatine (creatine plus phosphocholine) (Glu/Cr+PCr), N-acetyl-aspartate (NAA) to Cr+PCr (NAA/Cr+PCr), and choline-containing compounds phosphocholine plus glycerophosphocholine (GPC+PCh) to Cr+PCr (GPC+PCh/Cr+PCr) in the basal ganglia (BG) of HEU neonates compared to HUU neonates, as seen previously for absolute concentrations of these metabolites in 9-year-old HEU children. Furthermore, we hypothesized lower Glu/Cr+PCr and NAA/Cr+PCr ratios in the midfrontal gray matter (MFGM) of HEU neonates compared to HUU neonates, as previously observed for absolute concentrations of Glu and NAA in 11-year-old HEU children. Using proton magnetic resonance spectroscopy (MRS), metabolite/Cr+PCr ratios were measured in the BG (83 HEU neonates and 45 HUU neonates) and MFGM (65 HEU neonates and 31 HUU neonates) of neonates at a mean gestational age (GA) equivalent of 41.56 weeks (range 39-45 weeks). Linear regression models were used to compare HIV and ART exposure group differences in metabolite/Cr+PCr ratios in HEU neonates and HUU neonates, as well as HEU neonates who have been exposed to ART since conception (pre-conception) and HEU neonates who have been exposed to ART after 5 weeks of GA to HUU neonates. NAA/Cr+PCr, GPC+PCh/Cr+PCr, Glu/Cr+PCr, and myo-inositol (Ins)/Cr+PCr were the metabolite/Cr+PCr ratios measured.
  • Thumbnail Image
    Item
    Open Access
    Longitudinal analysis of Brain Metabolite levels for HIV infected Children from ages five to eleven
    (2020) Van Biljon, Noëlle; Little, Francesca; Meintjes, Ernesta; Holmes, Martha; Robertson, Frances
    HIV infected (HIV+) children initiate antiretroviral therapy (ART) early in life and remain on it lifelong. However, the long-term impact of ART and HIV on the maturing brain is not well documented and longitudinal neuroimaging studies are rare, especially in developing countries most heavily impacted by HIV/AIDS where access to imaging resources are limited. We have examined HIV related changes in metabolite level trajectories from 5-11 years in three brain regions using Magnetic Resonance Spectroscopy (MRS). We used univariate linear mixed effect models to identify independent profiles of the metabolites measured in each region of the brain. To explore the metabolite trends in a multivariate setting we generated multilevel mixed effects models, and correlated response models. There was an element of confounding introduced through the change of MRI scanner during the follow-up period and we compare different methods to resolve this issue. Consequently, we did observe differences in metabolite profiles from HIV+ children compared to HIV uninfected (HIV-) controls. This suggests that while these children are on ART treatment, there is still some underlying effect on their neurochemistry which sets their development apart from the normal healthy profiles we expect.
  • Thumbnail Image
    Item
    Open Access
    Multimodal neuroimaging signatures of early cART-treated paediatric HIV - Distinguishing perinatally HIV-infected 7-year-old children from uninfected controls
    (2020) Khobo, Isaac Lebogang; Robertson, Frances; Jankiewicz, Marcin; Holmes, Martha
    Introduction: HIV-related brain alterations can be identified using neuroimaging modalities such as proton magnetic resonance spectroscopy (1H-MRS), structural magnetic resonance imaging (sMRI), diffusion tensor imaging (DTI), and functional MRI (fMRI). However, few studies have combined multiple MRI measures/features to identify a multivariate neuroimaging signature that typifies HIV infection. Elastic net (EN) regularisation uses penalised regression to perform variable selection, shrinking the weighting of unimportant variables to zero. We chose to use the embedded feature selection of EN logistic regression to identify a set of neuroimaging features characteristic of paediatric HIV infection. We aimed to determine 1) the most useful features across MRI modalities to separate HIV+ children from HIV- controls and 2) whether better classification performance is obtained by combining multimodal MRI features rather than using features from a single modality. Methods: The study sample comprised 72 HIV+ 7-year-old children from the Children with HIV Early Antiretroviral Therapy (CHER) trial in Cape Town, who initiated combination antiretroviral therapy (cART) in infancy and had their viral loads suppressed from a young age, and 55 HIV- control children. Neuroimaging features were extracted to generate 7 MRI-derived sets. For sMRI, 42 regional brain volumes (1st set), mean cortical thickness and gyrification in 68 brain regions (2nd and 3rd set) were used. For DTI data: radial (RD), axial (AD), mean (MD) diffusivities, and fractional anisotropy (FA) in each of 20 atlas regions were extracted for a total of 80 DTI features (4th set). For 1H-MRS, concentrations of 14 metabolites and their ratios to creatine in the basal ganglia, peritrigonal white matter, and midfrontal gray matter voxels (5th, 6th and 7th set) were considered. A logistic EN regression model with repeated 10-fold cross validation (CV) was implemented in R, initially on each feature set separately. Sex, age and total intracranial volume (TIV) were included as confounders with no shrinkage penalty. For each model, the classification performance for HIV+ vs HIV- was assessed by computing accuracy, specificity, sensitivity, and mean area under the receiver operator characteristic curve (AUC) across 10 CV folds and 100 iterations. To combine feature sets, the best performing set was concatenated with each of the other sets and further EN regressions were run. The combination giving the largest AUC was combined with each of the remaining sets until there was no further increase in AUC. Two concatenation techniques were explored: nested and non-nested modelling. All models were assessed for their goodness of fit using χ 2 likelihood ratio tests for non-nested models and Akaike information criterion (AIC) for nested models. To identify features most useful in distinguishing HIV infection, the EN model was retrained on all the data, to find features with non-zero weights. Finally, multivariate imputation using chained equations (MICE) was explored to investigate the effect of increased sample size on classification and feature selection. Results: The best performing modality in the single modality analysis was sMRI volumes
  • No Thumbnail Available
    Item
    Open Access
    Structural organization in 9-year-old HIV-infected children and HIV-exposed children
    (2023) Mberi, Farai; Holmes, Martha; Jankiewicz Marcin
    Despite early initiation of antiretroviral treatment (ART) in children, studies continue to report HIVrelated neurodevelopmental damage and delays. Neuroimaging and neuropsychological studies have identified brain regions and cognitive domains vulnerable to the effects of HIV infection and exposure in children. As most studies focus on a single time point, it is unclear if the HIV-related abnormalities reported represent damage or developmental delay. Follow up studies are needed to better understand the long-term consequences of HIV in the ART era. Further, few studies examine links between brain imaging and cognitive outcomes, which provides a better understanding of the consequences of imaging abnormalities. This thesis includes a follow up study of a Cape Town-based cohort using neuroimaging, neurocognitive and clinical data at age 9 – 10 years. The cohort includes Children Perinatally infected with HIV (CPHIV) from the Children with HIV Early antiRetroviral (CHER) trial, who initiated ART by the age of 2 years. In addition, the cohort includes a control group made up of Children who are HIV Uninfected (CHU) and Children HIV-exposed, but Uninfected (CHEU). The age group in this study represents the beginning of adolescence, a critical period for the maturing brain. In particular, white matter increases in early adolescence to later stabilize in the middle teenage years. In parallel, this growth is related to maturing cognitive abilities allowing for more complex thinking processes. Diffusion tensor imaging (DTI) measures various aspects of white matter microstructure. DTI data can be explored using different methods, such as voxel wise analysis (VWA) and tractography. Within this cohort, we've previously reported white matter alterations using VWA at ages 5 and 7 years in the inferior/superior longitudinal fasciculus (ILF/SLF), inferior fronto-occipital fasciculus (IFOF), forceps minor, superior corona radiata (SCR), anterior thalamic radiation (ATR) and corticospinal tract (CST), in CPHIV despite early ART, pointing to ongoing HIV related white matter injury associated with myelin loss and axonal degradation. We hypothesize continued diffuse white matter abnormalities in these regions, pointing to chronic damage from at age 5. Tractography quantifies tract properties within the context of gray matter regions, giving functional context to DTI measures. Tractography also allows for the quantification of structural organization measures. Graph theory measures various aspects of brain organization, such as segregation and efficiency. Network measures can identify changes in structural organization due to disease, brain disorders and age. These measures also reflect cognitive and behavioral outcomes. Studies examining structural organization in CPHIV have reported alterations in gray matter volumes affecting local and network properties, suggesting HIV may influence signal propagation. To fill the gap between imaging outcomes and functional consequences within the population of children living with or exposed to HIV, we used tractography to identify the gray matter regions most dependent on tracts affected by HIV and the influence of HIV infection and exposure on white matter organization. Further, we explored possible links between tractography and graph outcomes and neuropsychological measures. Methods: We performed DTI-based VWA and tractography in a cohort of 58 CPHIV, 24 CHEU and 18 CHUU at age 9 – 10 years. Structural measures included DTI parameters, fractional anisotropy (FA), mean diffusivity (MD), axial diffusion (AD) and radial diffusion (RD). Tractography based structural network measures included degree, strength, transitivity, nodal efficiency, and local efficiency as measures of local network properties, and network modularity, global efficiency, and global transitivity as measures global organization. Statistical analysis was performed to identify HIV infection and exposure related differences, as well as explore relationships between imaging outcomes and neuropsychological data. Results: In the DTI voxelwise analysis, we observed higher MD in the IFOF, ILF/SLF, SCR and forceps minor of CPHIV. Many changes in MD were attributed to above-normal RD and/or AD which point to demyelination and axonal degradation. Furthermore, some regional HIV mediated white matter abnormalities characterized by higher MD were associated with markers of early immune compromise, including CD4% around age 12 weeks and timing of disease severity (age at nadir CD4%). Using DTI tractography, we observed that tracts with higher mean MD, AD and/or RD included seeds with connections to the insula, putamen, and fronto-temporoparietal regions. Tracts with higher MD, AD and/or RD were within the same hemisphere, with the exception of one interhemispheric connection of higher AD between insula and right caudate. Across participants, higher MD and RD in frontotemporoparietal connections were associated with lower performance in sequential processing, which measures auditory working memory. Among CHU, we found significant negative correlations between sequential processing and MD in basal ganglia connections to limbic and fronto-parietal regions. Among CPHIV, we found significant negative correlations between sequential processing and MD in a handful of tracts connecting regions in the limbic and temporoparietal junction. Lastly, in the graph theory-based analysis we found lower degree in the inferior temporal gyrus in CPHIV compared to CHU. Further, nodal degree in the inferior temporal gyrus across study participants was associated with auditory working memory. We did not find any HIV exposure effects on DTI measures. Conclusions: The DTI results presented in this thesis indicate persistent localized white matter damage or delayed development in CPHIV, with some regions influenced by early immune health. In addition, we observe a robust global network despite disconnections in tracts connecting the inferior temporal gyrus of CPHIV. Long term axonal loss and demyelination may result in hypomyelinated axons that are loosely packed in the now widened extra-exonal space. Further, white matter connectivity in frontotemporoparietal and limbic regions as well as nodal disconnections in the inferior temporal gyrus was associated with auditory working memory among CPHIV, suggesting these abnormalities contribute to deficits in this domain. In CPHIV, we observe typical networks that are robust to the effects of altered connectivity among regions of the temporoparietal cortex, limbic and subcortical areas. Lastly, we observe no HIV exposure effects on white matter integrity and network measures suggesting differences reported at 7 years in this cohort have resolved.
  • Thumbnail Image
    Item
    Open Access
    The impact of maternal HIV infection on uninfected neonate brain structure
    (2022) Ibrahim, Abdulmumin; Holmes, Martha; Meintjes, Ernesta; Warton, Fleur
    Successful prevention of mother-to-child HIV transmission (PMTCT) programs have reduced the risk of infant HIV infection in South Africa from 8% in 2008 to an estimated 1.4% in 2015, resulting in an increasing population of HIV-exposed uninfected (HEU) children. However, the long-term effects of HIV and antiretroviral therapy (ART) exposure on the developing brain is not well understood. While HEU children perform better than their counterparts living with HIV, they continue to demonstrate greater neurodevelopmental delay than HIV-unexposed uninfected (HUU) children. As a result, neuroimaging studies have looked at the developing brain in this population, however there is little consensus about typical exposure related effects. In addition, it is unclear whether previously reported exposure-related results are directly related to in utero exposure to HIV, or indirectly via family and/or environmental factors. Research focused on newborns allows one to eliminate possible contributions from other factors, clarifying the influence of ART and HIV exposure on the developing brain. This dissertation employs neuroimaging and neurocognitive data in a well-characterized infant cohort to better understand the influence of maternal HIV infection on the uninfected brain. HEU infants were exposed to ART in utero between 3 and 9 months, allowing for the study of potential ART exposure effects of as well as HIV exposure. This dissertation will identify HIV and ART exposure effects on brain structure. In addition, the relationship between neonate brain structural outcomes and cognitive abilities at 9-12 months will be determined to identify potential functional consequences of early structural abnormalities. Chapter two presents an analysis of manually traced subcortical volumes in 120 unexposed uninfected (HUU) and exposed uninfected (HEU) neonates. HEU neonates demonstrated significantly reduced mean caudate volumes bilaterally and left mean putamen volumes relative to HUU neonates. Further analysis revealed the observed differences in basal nuclei volumes were related to duration of ART in utero. Infants exposed to ART throughout pregnancy had similar caudate and putamen volumes compared to their HU counterparts. While infants exposed to ART post conception (from 3 - 8 months in utero) had significantly smaller mean caudate volumes bilaterally, and a trending smaller left putamen volume compared to HUU infants. Chapter three examines the potential functional consequences of HIV/ART volumetric reductions. We modelled manually traced neonatal subcortical volumes with neuropsychological outcomes at 9 - 12 months. Among HUU infants, bilateral pallidum volumes predicted neuropsychological measures across all domains. All volumes, with the exception of bilateral thalamus and vermis, predicted the general quotient score in HUU infants. In contrast, among the HEU infants, volumes did not relate to neuropsychological outcomes with the exception of the caudate, putamen and vermis predicting locomotion scores in the preconception group. While no HIV exposure differences were present in neuropsychological domains, HEU infants recruit alternative subcortical structures compared to typically developing unexposed infants. Chapter four presents a DTI-tractographic analysis of white matter connections between subcortical structures manually traced. HEU demonstrate white matter alterations in two tracts - higher FA between right putamen and left thalamus and higher MD between caudate and thalamus on the right hemisphere. The WM alterations observed in HEU appear to be from roles of both HIV and ART exposure. In contrast to ART dependent subcortical grey matter reductions, the observed white matter alterations are independent of maternal treatment initiation. In addition, we also find associations between unaltered white matter connections and both maternal immune health and ART duration during pregnancy. These results suggest white matter is influenced to varying degrees by HIV and ART exposure, as well as maternal health in pregnancy. Chapter five looks at the possible functional consequences of the reported alterations in white matter integrity. We modelled white matter connections between manually traced neonatal subcortical volumes with neuropsychological outcomes at 9 - 12 months. Similar to chapter 3, within HUU infants, we observed a number of white matter connections predictive of neuropsychological outcomes across all domains. And almost no white matter tracts predicted neuropsychological measures in HEU infants. These results again point to HEU infants recruiting different pathways to perform basic tasks. In conclusion, the results documented in this thesis points to the influence of HIV exposure, ART duration and maternal immune health on fetal brain development. However, these factors impact grey and white matter differently. ART initiated pre-conception was protective of caudate volumes but did not protect two white matter connections, the WM tract between right thalamus and right caudate, and WM that between left thalamus and right putamen. Within HUU neonates, basal ganglia and cerebellar volumes and white matter connections predicted neuropsychological outcomes in late infancy. However, HEU infants did not demonstrate the same associations suggesting they utilize alternate pathways from their HUU peers. While there were no exposure related differences across neuropsychological domains, the long-term functional consequences of altered structural recruitment is unknown. Finally, this thesis adds to the body of literature that early ART in pregnancy is neuroprotective, and that HIV exposure related structural alterations are evident as early as 2 - 4 weeks after birth.
  • No Thumbnail Available
    Item
    Open Access
    The influence of HIV and ART exposure on neonate brain volumes
    (2023) Olagunju, Aishah; Holmes, Martha; Meintjes Ernesta
    Introduction: There is growing evidence that maternal health in pregnancy influences the infant neurodevelopment. However, there are limited studies including broad metrics of maternal health when studying typical and atypical infant neurodevelopment. Although the human immunodeficiency virus (HIV) is a global pandemic, South Africa has the biggest and most high-profile HIV epidemic. Over the past decade, South Africa has reduced vertical transmissions of HIV, from improvements in antiretroviral medicines and the widespread accessibility of prevention programs. An outcome of this achievement is an increasing population of HIV-exposed-uninfected (HEU) infants and children. Despite improved outcomes compared to their peers living with HIV, HEU infants and children are at risk of neurodevelopmental delays in relation to HIV–unexposed–uninfected (HUU) children. As a result, there is a need to better understand the outcome of HIV/ART exposure on the fetal/infant/child brain. This study aimed to investigate the effect(s) of HIV and duration of ART exposure and the potential impact of additional maternal health factors during pregnancy on neonate brain volumes. Methods: Using magnetic resonance imaging (MRI), T1-weighted brain images of neonates were acquired. Infants included those whose mothers initiated ART preconception (HEU-pre), those whose mothers initiated ART post-conception (HEU-post) and infants born to mothers living without HIV or HIV unexposed uninfected (HUU). The data were quality checked and volumes were determined using the infant FreeSurfer tool. Statistical analysis was done in R to identify maternal health factors related to neonatal volumes as well as volumetric group differences due to HIV and ART exposure. Results: This analysis included 151 infants (49 HEU-pre; 48 HEU-post; 54 HUU; mean age 1.8 weeks; 50.3% male). Across all newborns, maternal Harvard Trauma Questionnaire (HTQ) score during pregnancy was associated with bilateral amygdala volumes. Within HEU infants, maternal CD4 count was associated with right thalamus and caudate volumes bilaterally. Group analysis showed a significant decrease in mean caudate volume bilaterally in the HEUpost group (left hemisphere p=0.006; right hemisphere p=0.009), as well as reduced right amygdala volume after controlling for maternal HTQ assessed in pregnancy. There was also a significant increase in the left lateral ventricle (p=0.04) and a decrease in the left cerebral white matter of HEU-pre infants (p=0.03). Conclusion: This study observed volumetric differences in HEU infants dependent on timing of maternal ART initiation, maternal immune health and maternal trauma assessed during pregnancy.
  • Thumbnail Image
    Item
    Open Access
    The influence of maternal HIV and ART exposure on neonate brain white matter integrity and organisation
    (2022) Magondo, Ndivhuwo; Holmes, Martha; Jankiewicz, Marcin
    Improved legislation and widespread access to treatment has led to a new cohort of children who have in utero exposure to maternal HIV, but remain HIV negative through the use of antiretroviral therapy (ART), known as HIV exposed-uninfected (HEU) children. While HEU infants and children are healthier than their infected peers, they experience some developmental delays as compared to uninfected unexposed populations. Few brain imaging studies have been done in HEU infants and children, and so the effects of the exposure to in utero HIV and ARVs on their neurodevelopment are not well understood. As this population matures, there is a need for studies to determine how HIV as well as ART exposure in utero and at birth affect white matter (WM) structure and its connectivity. This study used diffusion tensor imaging (DTI) tractography and graph theory to examine the possible influence of HIV and ART exposure in utero on neonate WM integrity (as measured by DTI parameters) and organisation (evaluated with graph theory measures). We hypothesised that HIV exposure will alter WM integrity, however structural organisation will remain unchanged across groups. We investigated HIV exposure and ART duration group differences in fractional anisotropy (FA) and mean diffusivity (MD) of the WM connections in the brain, as well as graph measures including strength, local efficiency, nodal efficiency, global efficiency, transitivity, and modularity. To examine these differences, a linear regression analysis was performed between the groups while correcting for maternal weight gained per week of pregnancy and maternal education. The study found that, regardless of when ART is started by the mothers, certain regions and tracts in the brain are seemingly influenced by HIV exposure. Infants whose mothers have been on ART pre-conception have higher MD values than their unexposed and uninfected peers, while those who have been exposed to ART postconception were shown to have lower FA values than their unexposed and uninfected counterparts. These results imply that ART duration influences WM integrity and may be neuroprotective for FA, which is more related to WM integrity, but not for MD, which relates to WM organisation. ARV exposure duration and CD4 count are shown to be positively associated with FA tracts, while CD4 count is negatively associated with MD. This relationship highlights the potential impact of maternal immune health on fetal brain development. While there are structural differences in certain WM tracts, the overall structural organisation remains unchanged, as no graph theory measures yielded significant results besides nodal efficiency. The brain's dense connective network may bridge gaps from damage in specific tracts throughout the connectome.
  • No Thumbnail Available
    Item
    Open Access
    The longitudinal effects of HIV and ART on the developing brain: a structural MRI study using manual segmentation
    (2024) Randall, Steven; Holmes, Martha; Meintjes Ernesta
    Background: Previous neuroimaging research into the effects of HIV in paediatric populations receiving treatment has reported brain abnormalities across a variety of ages and modalities. There are limited longitudinal studies, which would clarify if the observed changes represented developmental delay or ongoing damage. In addition, most neuroimaging studies do not include links to functional, immunological, or genetic outcomes which aid in understanding the consequence, mechanisms and factors underlying the brain imaging abnormalities. There is a need for longitudinal work and interdisciplinary crosssectional imaging studies to comprehend the consequences of the current literature. To address literature gap, this thesis builds on a cross-sectional study, by the same author, of this group of 106 children who were perinatally infected with HIV (PHIV), and form part of “Children with HIV early antiretroviral”(CHER) trial who have been followed since birth (Violari et al. 2008; Laughton et al. 2012; Cotton et al. 2013). The study showed larger subcortical gray matter and smaller white matter in children with perinatal HIV infection compared to controls, at 5 years of age. This thesis presents a longitudinal follow-up to assess whether these morphometric differences persisted into later childhood, as well as assess possible early contributors that may exacerbate or predict certain highlighted disturbances observed from 5 to 10 years of age. We aim to determine if the volumetric abnormalities observed at age 5 years represent developmental delay or ongoing injury due to HIV infection. In addition, we seek to identify potential early immune markers and the 5-year-old volume changes to better understand contributing factors. Methods: MRI scans were obtained at ages 5, 7, and 9 years, on a 3 T Allegra MRI (Siemens, Erlangen, Germany), and 10 years 3 T Skyra Siemens. Images were manually traced, by the author, for volumes of basal ganglia structures and corpus callosum using MultiTracer. Twenty-seven individuals were rescanned to assess whether incorporation and integration of 10 year old scanner data with scans from previous ages on a different scanner was possible, as a result of decommissioning the 3 T Allegra scanner. Volumetric growth curves were fit for forty children using mixed effects models with subject-specific random effects, with adjustment for possible two-way interactions between age and diagnosis. We investigated linear, quadratic, cubic and logarithmic fits for each volume. A panel of 44 soluble blood biomarkers were obtained at enrolment for a small sample of infants living with HIV. Cytokine concentrations were Z-score transformed and principle component analysis (PCA) performed. Results: Manual segmentation was reproducible across the different 3 T scanners within this pediatric sample allowing for volumetric data to be combined into one model without a scanner confounder. Across volumes, logarithmic models performed best. Age-related decreases were observed across children in the bilateral caudate and globus pallidus, as well as the corpus callosum. HIV-related alterations to growth trajectories were observed in the right nucleus accumbens and bilateral putamen, driven by volume abnormalitiesreported at 5 years. HIV-related corpus callosum reductions previously reported at age 5 did not alter the growth trajectories. PCA analysis identified a component associated negatively with subcortical volumes. Association with the derived component variable suggest that during the period of initial infection, in infancy, an increase of IL-10, IP-10, LBP and IFN-α, and accompanying decrease in IL-17F and CD40L contribute to smaller basal ganglia volumes at 5 years. Conclusions: This thesis expands cross sectional volumetric work which identified basal ganglia and corpus callosum volume abnormalities in 5-year-old PHIV children. Longitudinal analysis found the 5-year-old volumetric changes were likely HIV-related developmental delays, as volumes differences did not persist. The volume increases in the putamen affected the growth trajectories in PHIV and were related to early immunological factors associated with proinflammatory effects, immune activation and immune dysfunction. The volume decreases in the corpus callosum did not affect the age-related trajectory, and no cytokines related to volume abnormalities at 5 years. While cross sectional HIVrelated results in PHIV are concerning, more work is needed to contextualize their consequences on growth and function.