Advanced magnetic resonance imaging techniques in the diagnosis of Alzheimer's disease (AD), and evaluation of AD pathogenesis in an aging brain

Abstract

Background: The impact of Alzheimer’s disease (AD) on society’s resources and manpower has been forecasted for more than a decade, and recent statistics across the globe reveal that it is forthcoming. Pioneering work since 1980s confirmed the pathological hallmarks of the disease, such as neuritic plaques, neurofibrillary tangles, amyloid-β peptides, and microtubule-associated tau proteins. There are many gaps in the full appraisal of this complex disease, which possibly begins early in life in susceptible individuals and present at different severity and speed. In current thesis, advanced magnetic resonance imaging (MRI) techniques were evaluated for their efficacy in diagnosis, and in exploring AD pathogenesis.

Methods: In MRI volumetry/perfusion and diffusion studies, 20 and 18 AD subjects (different cohorts) were recruited respectively from the memory clinic of a University hospital, while 20 and 18 cognitively normal older adults (CN) were recruited respectively from elderly centers, community and university volunteers, as well as 17 young adults in the diffusion study. In MR Spectroscopy (MRS), 30 healthy volunteers of 3 different age ranges (20-39, 40-59, 60-89) were recruited. All studies were performed using a 3-tesla MRI scanner. For MRI volumetry, a standardized T1-weighted 3D volumetric Fast Field Echo sequence, and for pulsatile Arterial Spin Labeling (ASL), a Look-Locker-based echo-planar imaging sequence was employed. Single-shot echo-planar diffusion weighted imaging was used to examine white matter (WM) integrity; diffusion sensitizing gradients (b = 800 s/mm2) applied in 16 directions, and one additional image without diffusion gradient (b0 = 0 s/mm2). Single voxel MRS was performed in the limbic regions, with point resolved spectroscopy for volume selection and excitation for water suppression.

Results: Voxel Based Morphometry with Diffeomorphic Anatomical Registration using Exponentiated Lie algebra and standard registration has similarly high efficacies as manual hippocampal volumetry in discriminating AD from CN. Using pulsatile ASL, we found impairment of hemodynamic parameters other than cerebral blood flow (CBF) in moderate AD, indicative of underlying vascular abnormality. Combined MR hippocampal volumetry and ASL CBF was superior to single measure in AD diagnosis. Using diffusion tensor imaging (DTI) techniques, parallel evidence of anterior WM disintegrity in normal elderly, and more extensively in the AD was found. AD showed further loss of WM integrity in the posterior brain regions, and such WM disintegrity may result from demyelination. In aging, we found increased choline and creatine, and N-acetyl-aspartate in cingulate gyri, which might suggest glial proliferation and neuronal hypertrophy respectively.

Discussion: A ‘one-stop-shop’ study combining structural and perfusion MRI has improved efficacy in discriminating AD from CN. DTI showed possible WM retrogenesis in normal aging and AD, although ischemic effect on WM cannot be ruled out. Our MRS study highlighted metabolic changes with age, which could be compensatory to an increased energy demand coupled with a lower CBF. Neuroimaging is likely to have a great impact on early diagnosis of AD, which will benefit patient care, prognostication and future therapy. Hopefully, insights into the physiology of normal aging and pathophysiology of AD derived from neuroimaging can guide future basic science and clinical research.