HIV-exposure related disruptions in functional and structural connectivity in the central auditory system in adolescence

BackgroundChildren who are HIV-exposed but uninfected (CHEU) face elevated risks of hearing loss and language deficits compared to HIV-unexposed peers. The central auditory system (CAS) undergoes substantial maturational changes during adolescence, yet no neuroimaging study has examined its structural or functional integrity in CHEU. Prior work in this cohort identified white matter (WM) alterations in regions adjacent to the CAS at age 7, and reduced auditory working memory in CHEU relative to unexposed children (CHUU). AimTo characterise WM integrity and functional connectivity (FC) of the CAS and related regions in CHEU at age 11, to investigate structural and functional network topology, and to examine associations between imaging outcomes and neurocognitive function. MethodsForty-eight children aged 11-12 (20 CHEU, 28 CHUU) from an ongoing longitudinal neurodevelopmental cohort underwent 3T MRI including diffusion tensor imaging (DTI) and resting-state fMRI (RS-fMRI). CAS regions (cochlear nucleus/superior olivary complex, inferior colliculus [IC], medial geniculate nucleus [MGN], and primary auditory cortex [PAC]) were manually segmented and combined with an automated atlas. DTI probabilistic tractography was performed, extracting FA, MD, AD, RD, fractional number of tracts, and tract volume. FC was computed using Pearson correlations between regional time series. Graph theory measures (degree, strength, transitivity, nodal and local efficiency) were derived for structural and functional networks. RS-fMRI group comparisons used Bayesian multilevel modelling (matrix-based and region-based analyses), while DTI comparisons used linear models with FDR correction. Neurocognitive testing employed the KABC-II. ResultsNo significant group differences in DTI WM metrics (FA, MD, AD, RD) were observed after FDR correction. CHEU demonstrated higher structural nodal strength in the left IC (FDR-significant) and in the bilateral rostral middle frontal cortex (rMFC) and right cuneus. RS-fMRI revealed lower FC between the bilateral IC in CHEU, alongside reduced FC in the left caudate, left hippocampus CA3, left pericalcarine, and left lingual gyrus. CHEU showed higher FC between the left MGN and right precentral, left postcentral, and right rMFC; the right PAC also showed higher FC to the right rMFC and left postcentral gyrus. No significant group differences were observed in functional nodal measures. No significant associations were found between structural or functional imaging outcomes and neurocognitive scores after multiple comparison correction. DiscussionStructural and functional alterations within the CAS were most prominent in the IC, with increased nodal strength in CHEU potentially reflecting compensatory structural connectivity, and reduced interhemispheric FC between the bilateral IC suggesting disrupted auditory integration. Altered FC between the MGN/PAC and cortical regions, including the rMFC and sensorimotor cortices, may reflect differences in top-down auditory processing. The absence of imaging-cognition associations at age 11 suggests that these connectivity differences do not, at this stage, translate into measurable deficits in auditory or language-related neurocognitive performance. ConclusionThis is the first study to examine functional and structural connectivity of the CAS in CHEU children. HIV exposure is associated with subtle but discernible alterations in IC connectivity and in CAS links to cortical regions at age 11, without detectable neurocognitive correlates. Longitudinal follow-up and inclusion of audiological and ART exposure data are needed to clarify the developmental and functional consequences of these findings.