Multimodal neuroimaging evidence of alterations in cortical structure and function in HIV-infected older adults

Abstract

Combination antiretroviral therapy transformed human immunodeficiency virus (HIV)-infection from a terminal illness to a manageable condition, but these patients remain at a significantly elevated risk of developing cognitive impairments and the mechanisms are not understood. Some previous neuroimaging studies have found hyperactivation in frontoparietal networks of HIV-infected patients, whereas others reported aberrations restricted to sensory cortices. In this study, we utilize high-resolution structural and neurophysiological imaging to determine whether alterations in brain structure, function, or both contribute to HIV-related cognitive impairments. HIV-infected adults and individually matched controls completed 3-Tesla structural magnetic resonance imaging (sMRI) and a mechanoreception task during magnetoencephalography (MEG). MEG data were examined using advanced beamforming methods, and sMRI data were analyzed using the latest voxel-based morphometry methods with DARTEL. We found significantly reduced theta responses in the postcentral gyrus and increased alpha activity in the prefrontal cortices of HIV-infected patients compared with controls. Patients also had reduced gray matter volume in the postcentral gyrus, parahippocampal gyrus, and other regions. Importantly, reduced gray matter volume in the left postcentral gyrus was spatially coincident with abnormal MEG responses in HIV-infected patients. Finally, left prefrontal and postcentral gyrus activity was correlated with neuropsychological performance and, when used in conjunction, these two MEG findings had a sensitivity and specificity of over 87.5% for HIV-associated cognitive impairment. This study is the first to demonstrate abnormally increased activity in association cortices with simultaneously decreased activity in sensory areas. These MEG findings had excellent sensitivity and specificity for HIV-associated cognitive impairment, and may hold promise as a potential disease marker.

Keywords: AIDS; HIV-associated neurocognitive disorder; biomarker; cognitive disorders; magnetoencephalography; oscillation.

Figure 1

Tactile stimulation device. Participants were seated in a custom MEG chair with both arms resting on a tray attached to the chair body. Mechanoreceptors within the pad of the second digit of the right hand were stimulated using a small airbladder that was encased in plastic housing and clipped onto the index finger. A plastic red hose connected the airbladder to a pneumatic delivery system that was located outside the MSR. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 2

Gray matter volume reductions in HIV‐infected participants. (A) HIV‐infected patients had significantly reduced gray matter volume relative to uninfected controls in the left postcentral gyrus (green areas; P < 0.001, corrected). This area of reduced gray matter spatially overlapped with the maximal postcentral gyrus MEG response in HIV‐infected patients (Fig. 3, middle panel). (B) Significantly reduced gray matter volume was also found in the right cerebellum, left parahippocampal gyrus, bilateral lingual gyri, left middle temporal area near the occipitotemporal notch, and an area of the right postcentral gyrus. Lines in the orthogonal image (far right) indicate the placement of shown sagittal (top) and axial (bottom) slices in the volume. All images are shown in neurological convention (left = left) and at the same threshold (P < 0.001, corrected). [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 3

Significant theta (4–8 Hz) neuronal activity following tactile stimulation to the right hand. Uninfected controls (left) and HIV‐infected participants (middle) exhibited robust activity in the left postcentral gyrus in response to mechanoreceptor stimulation, with neural activity being stronger and more consistent in uninfected controls (notice different scale for P‐values). The group effect (right image; Controls > HIV‐infected participants) was restricted to a small area of the left somatosensory cortex, and no brain areas had stronger 4–8 Hz activity in HIV‐infected participants. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 4

Prefrontal Alpha (8–14 Hz) Abnormalities in HIV‐infected Participants. Uninfected controls showed a strong decrease in local alpha activity in the left prefrontal cortex (left) during the 0.01 – 0.26 s time window, whereas HIV‐infected participants exhibited a small increase in alpha activity in this same brain region (middle) and time window. Thus, these neural responses were in the opposite direction across groups, which gave rise to a strong group effect of higher alpha activity in the HIV‐infected participants (right panel). Notice the different scales for P‐values across images. These data and Figure 3 highlight the distinct nature of HIV‐associated abnormalities in the primary sensory and association cortices. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 5

Correlation of MEG metrics with neuropsychological assessments and age. In all panels, scores on neuropsychological assessments are shown in demographically‐adjusted z‐scores. The amplitude of 4–8 Hz neural activity in the left postcentral gyrus was significantly (P < 0.01) correlated with performance on the Digit Symbol task in HIV‐infected participants (top‐left panel). There was also a significant (P < 0.01) negative correlation between prefrontal alpha activity and performance on the Grooved Pegboard task, and the retention and delayed‐recall indices of HVLT‐R in HIV‐infected participants (top‐right panel). In all cases, these correlations indicate that superior neuropsychological performance was associated with brain responses that were similar to those seen in controls. Finally, there was a significant negative correlation between age and response amplitude in the left postcentral gyrus of controls (P < 0.001; bottom panel), but not HIV‐infected participants (P = 0.34). The color legend for each panel appears on the top right. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]