Introduction

HIV+ individuals have increased life expectancy in the era of combined antiretroviral therapy (cART), but also face challenges such as chronic HIV-associated neurocognitive disorder (HAND). The underlying causes of cognitive impairment in aging HIV+ subjects remain to be elucidated, but may be related to brain structure. The purpose of this study was therefore to measure multiple brain volumes in elderly HIV+ and HIV- subjects using high resolution 7T MRI. Regional brain volumes were compared by age group, HIV serostatus and HAND classification.

Methods

Forty HIV+ and 22 HIV- individuals over 50 years of age were enrolled in this study. All subjects underwent structural brain MRI including a 3D T1-weighted MPRAGE scan with 1.2 mm isotropic voxel size using a 7T scanner (Philips ‘Achieva’, Best, The Netherlands) equipped with a 32-channel receive head coil and quadrature transmit coil (Nova Medical Inc., Wilmington, MA, USA). Subjects were stratified by neurocognitive disorder status using the revised American Academy of Neurology (AAN) ‘Frascati’ criteria ¹. T1-weighted images were segmented using the online atlas-based volumetric resource MriCloud ^2-5. An adult atlas for 50 years of age and above was used for segmentation. The whole brain was segmented into 289 structures which were grouped at five levels of granularity ². In this study, structural volumes were extracted from 54 brain regions with high segmentation reproducibility ² and included the gray and white matter brain regions commonly affected in HIV and aging individuals; namely, the basal ganglia, frontal white matter, precuneus, posterior cingulate, centrum semiovale, hippocampus, caudate, putamen, thalamus, prefrontal, superior frontal, and parietal regions, (Figure 1). Group comparisons were done using the Kruskal-Wallis and Mann-Whitney U tests.

Results

Mean age for HIV+ and HIV- were 59.0 years ± 4.5 SD and 64.8 ± 7.4 years respectively; 49 (79%) were male. When all participants were stratified by decade, 26 were in their 50’s (21 of them were HIV+), 29 were in their 60’s (19 of them were HIV+) and 7 were in their 70’s (1 of them was HIV+). HAND stage for the HIV+ individuals was as follows: normal cognition (n=10), asymptomatic neurocognitive impairment (ANI) (n=10), mild neurocognitive disorder (MND) (n=10), and HIV dementia (HAD) (n=10). There were no significant group differences in age, education, gender, race, CD4 count or plasma HIV RNA between groups except for the estimated IQ which was significantly lower in the HIV+ group (P=0.04) and the Center for Epidemiological Studies Depression (CES-D) score which was higher in the HIV+ group (P= 0.03). Brain segmentation and volumetric analysis showed no significant differences between HIV+ and HIV- median brain volumes (in mm3) except in the globus pallidus (regardless of age or cognitive status) (3352 versus 3183 mm3, p=0.04). Among HIV+ participants only, the median brain volumes were larger in individuals in their 50s versus those in their 60s in the basal ganglia (BG) (P=0.03) and precuneus (P=0.05). The median volumes were significantly higher among those with normal cognition compared to those with HAND (ANI, MND, or HAD) in the precuneus (P=0.04), and posterior cingulate (P=0.03) (Figure 2). There were no significant differences in other brain ROIs.

Discussion and Conclusion

This study leveraged the high SNR and contrast of 7T MRI to measure regional brain volumes in older subjects with and without HIV infection. The main finding is that brain volumes were similar between HIV+ and HIV- groups, however selected brain volumes were lower in the HAND group compared to HIV+ without cognitive impairment. Regions found to be involved were the precuneus and posterior cingulate, both of which are heavily implicated in the pathology of Alzheimer’s disease. Age related changes were also observed in the HIV+ participants, with older subjects showing lower volumes in BG and precuneus. Future longitudinal studies of regional brain atrophy in older subjects, and its relation to cognitive decline, are needed in order to establish the predictive value of neuroimaging measurements and underlying mechanisms associated with HAND in the older HIV+ population.

Acknowledgements

This study was supported by NS 081196.