Resting-state functional magnetic resonance imaging in clade C HIV: within-group association with neurocognitive function

Abstract

Neuroimaging abnormalities are common in chronically infected HIV-positive individuals. The majority of studies have focused on structural or functional brain outcomes in samples infected with clade B HIV. While preliminary work reveals a similar structural imaging phenotype in patients infected with clade C HIV, no study has examined functional connectivity (FC) using resting-state functional magnetic resonance imaging (rs-fMRI) in clade C HIV. In particular, we were interested to explore HIV-only effects on neurocognitive function using associations with rs-fMRI. In the present study, 56 treatment-naïve, clade C HIV-infected participants (age 32.27 ± 5.53 years, education 10.02 ± 1.72 years, 46 female) underwent rs-fMRI and cognitive testing. Individual resting-state networks were correlated with global deficit scores (GDS) in order to explore associations between them within an HIV-positive sample. Results revealed ten regions in six resting-state networks where FC inversely correlated with GDS scores (worse performance). The networks affected included three independent attention networks: the default mode network (DMN), sensorimotor network, and basal ganglia. Connectivity in these regions did not correlate with plasma viral load or CD4 cell count. The design of this study is unique and has not been previously reported in clade B. The abnormalities related to neurocognitive performance reported in this study of clade C may reflect late disease stage and/or unique host/viral dynamics. Longitudinal studies will help to clarify the clinical significance of resting-state alterations in clade C HIV.

Keywords: Global deficit scores (GDS); Human immunodeficiency virus (HIV); Neurocognitive deficits; Resting-state functional connectivity (RSFC); Resting-state functional magnetic resonance imaging.

Fig. 1

Z-score maps (thresholded Z>4) representing the 14 RSNs identified by the ICA for 56 HIV+ subjects: visual occipital lobe (Vis1), dorsal attention (dAtt), right executive (R-Exec), default mode network (DMN), dorsal DMN (dDMN), salience (Sal), visual lingual gyrus (Vis2), sensorimotor (SenMot), ventral attention (vAtt), left executive (L-Exec), attention (Att), auditory (Aud), cerebellum (Cer) and basal ganglia (Bas).

Fig. 2

(Left) Brain maps of clusters that exhibited significant associations between resting state connectivity (blue) in the dorsal attention network (hotter colours) for 56 HIV+ subjects. Peak is situated in (a) the right precentral gyrus of the dAtt network, (b) the right superior frontal gyrus of the DMN, (c – e) the left postcentral gyrus, right precentral gyrus and right medial frontal gyrus of the SenMot network. (Right) Correlation graphs show relationships between average Z-scores and GDS with r and corrected p-values

Fig. 3

(Left) Brain maps of clusters that exhibited significant associations between resting state connectivity (blue) in the dorsal attention network (hotter colours) for 56 HIV+ subjects. Peak is situated in (a – b) the right inferior temporal gyrus and right superior frontal gyrus in the vAtt network, (c – d) the right inferior frontal gyrus and right inferior parietal lobule in the Att network and (e) the right thalamus in the Bas. (Right) Correlation graphs show relationships between average Z-scores and GDS with r and corrected p-values