HIV-induced immune activity in the central nervous system (CNS) is believed to lead to permanent brain changes, neurocognitive dysfunction and functional impairment. MR Spectroscopy (MRS) is a powerful non- invasive tool for assessing neurochemical changes in multiple brain locations of HIV+ patients. In contrast to diffusivity of water recorded using diffusion weighted imaging (DWI), diffusion-weighted (DW)-MR spectroscopy can detect the diffusivity of intracellular metabolites such as, N-acetylaspartate, creatine, and choline, which are exclusively located in the intracellular space, with a slow exchange between intra- and extra-cellular compartments; therefore, the detected apparent diffusion coefficients (ADCs) of cerebral metabolites can only be attributed to diffusion in the intracellular space. The outcome would provide more information (diffusivity of metabolites) than averaged water diffusivity from DTI. Earlier attempts have investigated the ADCs of three metabolites using single-voxel localized MRS in mostly healthy human subjects. MR imaging using radial sampling has been shown to be less sensitive to motion and off-resonance effects. Earlier work on DW line scan echo planar spectroscopic imaging (DW-LSEPSI) for motion robustness suffered from reduced SNR. Extending radial sampling to hyperpolarized carbon-13 (13C) MRSI has been demonstrated; however, implementation in 1H MRSI has not been demonstrated. We propose to develop volumetric radial-based echo-planar diffusion- weighted spectroscopic imaging (r-DW-EPSI), to evaluate diffusion of intracellular metabolites such as NAA, Cr, Cho, and Glu/glutamine (Glx). Alterations in metabolite ADC values may generate additional information about mechanisms underlying HIV, even after anti-retroviral therapy (ART). By exploiting strategies of acquisition acceleration of spatial encoding using FISTA with variable acceleration reconstruction, the radial EPI readout in the DW-EPSI sequence will enable recording of multi-voxel DW-spectra an order of magnitude faster than when using conventional phase-encoding.
Specific aims of this study are: (1) Develop accelerated r-DW-EPSI using semi-LASER localization, and optimize it in brain phantoms and 10 healthy adults. (2) Determine ADCs of Cr, NAA, Cho, mI and Glx in 25 HIV+ subjects, and assess their differences compared to 25 age-/sex-matched healthy adults. The outcome will be correlated with DTI metrics, neuropsychological test results, and other disease variables. We will test the following hypotheses: 1) The accelerated 3D r-DW-EPSI acquisition will be less sensitive to motion and chemical shift off-resonance effects due to oversampling in the central k-space. Altered ADCs of non-water molecules will be measured in multiple brain regions as markers of microstructural abnormalities as well as relative metabolite concentrations. 2) Metabolite ADCs and relative concentrations will support DTI findings correlating with neurocognitive cognitive impairments reflecting neuroinflammation and neuronal injury in the brains of HIV-infected subjects. This is unexplored territory in HIV research. Successful implementation of the r-DW-EPSI has the potential to be extended to other organs where motion is a prime concern.
HIV infection (HIV+) is a pressing health concern in the United States, and the prevalence of HIV-associated neurocognitive disorders (HAND) remains substantial even following combination antiretroviral therapy (cART). There is a need for novel non-invasive neuroimaging to identify the progressive changes in cerebral function parameters in HIV-infected subjects. We will evaluate a novel noninvasive functional imaging tool, 3D radial diffusion-weighted spectroscopic imaging, to assess diffusivity of non-water molecules in the brains of two cohorts: first, 10 healthy subjects to evaluate test/retest reproducibility of the technique, and second, 25 HIV+ subjects and 25 age-matched healthy (HIV-) subjects.
