Background:  Significant atrophy in the frontal regions of the corpus callosum has been observed in HIV-infected patients associated with CD4 cell decline. This study examines the integrity of the corpus callosum in a cohort of HIV-infected patients with chronic disease and on stable ART compared to controls as well as the effects of CD4 count and plasma viral load on the size of CC regions. 

Methods:  Participants for this study included 200 HIV-infected patients (84% male, median CD4 count 304, 62% undetectable plasma viral load, age 46 years) on stable ART (median duration of HIV 12 years) participating in a multicenter prospective HIV neuroimaging study and 139 HIV^– control participants (48% male, age 37 years). The corpus callosum was manually outlined on a single mid-sagittal slice on the baseline T1-weighted images and then automatically parcellated into 5 segments utilizing a modified Witelson scheme in Matlab. The area of each segment of the corpus callosum was normalized by head size (total intracranial volume). The cross-sectional areas of each segment were used as dependent measures in the study. HIV⁺ patients were divided into 2 groups based on ADC stage (168 <0.5; 32 >1) and then gross and regional areas were compared for each of the HIV⁺ groups and controls using MANCOVA controlling for age and duration of illness.

Results:  Of 5 parcellated segments (genu, p <0.02; isthmus, p <0.01; splenium, p <0.01) and gross measurements (p <0.009), 3 were significantly reduced in later ADC stages even when controlling for age and duration of illness. Gross and regional corpus callosum segments were not significantly associated with CD4 cell counts or plasma viral load (p >0.11) though nadir CD4 levels were related to 3 of 5 parcellations and the gross corpus callosum area in this cross-sectional analysis (p <0.05).

Conclusions:  This study expands previous research demonstrating anterior corpus callosum atrophy while also providing evidence for reductions in the posterior regions of the corpus callosum in the setting of stable ART with adequate viral suppression. The lack of associations with CD4 cell counts or plasma viral load levels suggests that other factors may contribute to the abnormalities in the corpus callosum. The absence of abnormalities in the neuro-asymptomatic group suggests that the reduction in corpus callosum areas may provide a biomarker of HIV cognitive decline in the setting of ART.