Examining Brain Function Changes in HIV-Infected Patients With Asymptomatic Neurocognitive Impairment: A Longitudinal Study

Abstract

Background: HIV-associated neurocognitive disorders (HAND), especially asymptomatic neurocognitive impairment (ANI), the initial stage of HAND, persist among a substantial proportion of individuals living with HIV despite the introduction of combination antiretroviral therapy (cART). Resting-state functional magnetic resonance imaging (rs-fMRI) focusing on ANI among HIV-related patients is rarely reported.

Methods: 60 right-handed Chinese male patients with HIV-associated ANI underwent baseline and follow-up neurocognitive examination, and rs-fMRI scans over an average interval of 1.68 years. Brain function alterations were evaluated through amplitude of low-frequency fluctuation (ALFF/fALFF), regional homogeneity (ReHo), and functional connectivity (FC) analyses.

Results: In this study, significant reductions in ALFF were observed in the MOG, cuneus, superior frontal gyrus, and supplementary motor area in the follow-up group compared to baseline. This was accompanied by increased ALFF in the right insula. ReHo analysis revealed decreased values in the left median cingulate, right calcarine fissure, MOG, and left precentral gyrus, alongside increased ReHo in the supramarginal gyrus, postcentral gyrus, parahippocampal gyrus, and calcarine fissure. FC analysis demonstrated decreased connectivity between the precentral gyrus and calcarine cortex and between the MOG and calcarine cortex. Correlation analysis indicated that these imaging changes were correlated with declines in specific neurocognitive domains, including memory, speed of information processing, and executive function.

Conclusion: Our research demonstrates a gradual deterioration in brain function in HIV-positive individuals with ANI despite receiving cART. This decline correlates with worsening neurocognitive abilities, specifically visual processing and executive function.

Keywords: HIV‐associated neurocognitive disorders; amplitude of low‐frequency fluctuation; asymptomatic neurocognitive impairment; functional connectivity; regional homogeneity.

FIGURE 1

The differences of clinical characteristics and neurocognitive performance of the participants. Significance was set as p < 0.05. * p< 0.05; ** p< 0.01; *** p< 0.001.

FIGURE 2

Brain regions with different ALFF. MOG.L, left middle occipital gyrus; MOG.R, right middle occipital gyrus; CUN.R, right cuneus; SFGdor.L, left superior frontal gyrus, dorsolateral; SMA.R, right supplementary motor area; INS.R, right insula; The color bar indicates a scale of T values.

FIGURE 3

Brain regions with different fALFF. LING.R, right lingual gyrus; AMYG.R, right amygdala; SFGdor.R, right superior frontal gyrus, dorsolateral; SMA.R, right supplementary motor area; The color bar indicates a scale of T values.

FIGURE 4

Brain regions with different ReHo maps. SMG.L, left supramarginal gyrus; DCG.L, left median cingulate and paracingulate gyri; PoCG.R, right postcentral gyrus; PHG.R right parahippocampal gyrus; CAL.R, right calcarine fissure and surrounding cortex; CAL.L, left calcarine fissure and surrounding cortex; MOG.R, right middle occipital gyrus; PreCG.L, left precental gyrus; The color bar indicates a scale of T values.

FIGURE 5

Differences in ROI region functional connectivity between the two groups. Compared with the baseline group, patients showed decreased FC between the CAL during the follow‐up group. L and PreCG. L, while the FC between the right prefrontal cortex CAL. R and the MOG. R increased. The results were corrected for multiple comparisons (FDR correction, p<0.05). The line represents the FC, with significant differences between each pair of defined ROI regions in the two groups. Red represents a positive t‐value indicating an increase in FC during the follow‐up period compared to baseline, while blue represents a negative t‐value indicating a decrease in FC during the follow‐up period.

FIGURE 6

Correlation between abnormal ReHo/FC and clinical variables in baseline group with baseline and follow‐up group. The threshold was set at p < 0.05 without multiple comparison corrections.