The Center for Disease Control and Prevention estimates that 1,148,200 Americans aged 13 years and older are living with HIV infection, including 207,600 (18.1%) who are unaware of their infection. According to pathological data, central nervous system (CNS) involvement is commonly found during the early phase of infection. In vivo proton magnetic resonance spectroscopy studies of HIV-infected humans have demonstrated significant changes of metabolites observed in the brain N-acetylaspartate, creatine, choline, glutamate, glutamine, myo-inositol with varying changes in different brain regions. Diffusion tensor imaging (DTI) can be used to derive quantitative in vivo measurements of region-specific and diffuse brain alterations. DTI studies have demonstrated changes of mean diffusivity (MD) and fractional anisotropy (FA) in the various parts of brain. Diffusion abnormalities involving the frontal white matter and the corpus callosum have also been observed in patients infected with HIV. Single-voxel or two-dimensional (2D) magnetic resonance spectroscopic imaging (MRSI) technique with long and short echo time (TE) has been used for many years to study HIV. A drawback of the three-dimensional (3D) MRSI technique using conventional phase encoding in three directions to traverse k-space is the long time required to acquire large-volume 3D datasets. Hence, the scan time necessary for the acquisition of 3D high- resolution MRSI data with adequate spatial coverage may be prohibitively long for clinical exams. Thus, new imaging and bio-chemical characterization techniques are needed to allow repeated, non-invasive assessment of these processes in vivo. There is a need for whole brain 3D-EPSI using short TE to get 3D maps of brain metabolite distributions even though the current version of the 3D EPSI uses long TE and the total duration is more than 20 minutes. Novel acquisition using non- uniformly under sampled (NUS) 3D EPSI with compressed sensing (CS) reconstruction can further accelerate the MRSI acquisition by approximately 3-4 times. Hence the specific aims of the study are as follows: (1) to implement and optimize a volumetric 3D NUS based EPSI sequence on a Siemens 3T Skyra MRI scanner, and to reconstruct the NUS data using CS reconstruction with excellent signal fidelity;(2) to record NUS-based 3D EPSI data and DTI in 30 HIV patients and to compare the outcome with 30 healthy controls (3) to combine DTI derived FA and MD and to compare the findings with metabolite concentrations and nadir CD4 cell count. We hypothesize that the NUS based 3D EPSI acquisition will be 4 times faster than that of the current fully encoded version. An expected outcome of this study is the development of a volumetric 3D EPSI scan in HIV patients that can select metabolites from multiple brain regions in less than 10 minutes. The whole brain neurochemical changes measured by the 3D EPSI technique will be correlated with that of DTI and nadir CD4 cell count to better understand the role of CNS involvement in HIV pathology.

Public Health Relevance

Approximately 1.1 million people in the U.S. are diagnosed with HIV and there is a need for a reliable, non- invasive and optimal brain MRI protocol combined with MR Spectroscopic imaging to monitor for cognitive deficits and dementia, which are prevalent consequences of living longer with HIV infection. The imaging technique we describe in this proposal will be a powerful tool to monitor the course of progression of neuroAIDS and to evaluate the efficacy of specific drug treatments in patients.

National Institute of Health (NIH)
National Institute of Neurological Disorders and Stroke (NINDS)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
NeuroAIDS and other End-Organ Diseases Study Section (NAED)
Program Officer
Wong, May
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of California Los Angeles
Schools of Medicine
Los Angeles
United States
Zip Code