Premium
IC‐P‐129: Magnetic Resonance Spectroscopy (MRS) analysis in Alzheimer's disease: A novel approach for integrating metabolite spectra, tissue segmentation and perfusion data
Author(s) -
Szilagyi Gregory,
Scott Christopher J.M.,
Chavez Sofia,
Black Sandra E.
Publication year - 2009
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2009.05.101
Subject(s) - creatine , nuclear medicine , perfusion , white matter , in vivo magnetic resonance spectroscopy , nuclear magnetic resonance , magnetic resonance imaging , voxel , choline , medicine , chemistry , radiology , physics
Background: Short echo-time 1H magnetic resonance spectroscopy is commonly used to measure metabolite levels in the human brain. Suppression of metabolite signals by inversion-nulling methods may be used to measure the contribution of broader macromolecule resonances. The hippocampus is known to be involved early in the pathophysiology of Alzheimer disease. Objective: The purpose of this study was to compare the macromolecule content in the hippocampus in subjects with Alzheimer disease (AD) and mild cognitive impairment (MCI) to normal elderly controls (NEC). Methods: Short echo-time 1H spectra were acquired from the right hippocampus of 23 subjects with probably AD (mean 6 standard deviation; age 7467 years; MMSE 2364), 12 subjects with MCI (age 72610 years; MMSE 2862), and 15 NEC (age 7866 years; MMSE 2961) using a 4 Tesla MRI. Spectra were localized by adiabatic selective refocusing (LASER, TE/TR 1⁄4 46/4200 ms, 128 averages) preceded by two inversion pulses (TI11⁄4 2.2s, TI21⁄4 0.7s) for metabolite nulling. Voxel dimensions were chosen to maximize voxel volume while staying largely within the hippocampus of each subject and ranged from 1.8 cc to 5.1 cc (mean 6 SD, 3.7 6 0.8 cc). Spectra were line shape corrected using an unsuppressed water spectrum (8 averages, TR1⁄42.2s) and summed for each diagnostic group following amplitude normalization. The average group spectra were subtracted to reveal group differences. Results: Macromolecule spectra were successfully scaled and summed. The difference between the average macromolecule spectra from the Alzheimer patients and NEC subjects showed notable peaks at 4.0 ppm (increased in AD), 1.87 ppm (decreased in AD), and 1.57 ppm (decreased in AD). Differences between the average AD spectrum and average MCI spectrum were less remarkable. Conclusions: The macromolecule baseline in short echo-time 1H spectra is normally not carefully scrutinized. This preliminary study shows potential differences between macromolecule signals in AD subjects compared to normal elderly controls in the right hippocampus. Further study is required to identify the biological significance of these resonances.