Diffuse nuclear Overhauser effect MRI contrast changes detected in multiple sclerosis subjects at 7T.

Multiple sclerosis is an inflammatory demyelinating condition of the central nervous system affecting approximately 1 million people in the USA. Although standard structural MRI techniques are now the main imaging modality for multiple sclerosis diagnosis and management, they are yet to provide information regarding the metabolic profile of the disease. Ultra-high field 7T MRI systems have provided gains in signal-to-noise ratio (SNR) and spatial resolution for structural MRI as well as larger chemical shifts leading to improvements in specialized imaging sequences, such as nuclear Overhauser effect (NOE) imaging, that can evaluate macromolecular metabolite composition. In this work, NOE images were acquired on a cohort of multiple sclerosis and healthy control subjects to spatially map differences in their lipid metabolites as a result of NOE effects. NOE image data were acquired on a total of 25 subjects {15 multiple sclerosis subjects [10 females, 5 males (21-70 years)] and 10 healthy controls [5 females, 5 males (23-71 years)]} on a 7T MRI system with a frequency offset range of -5 to 5 ppm. A five-pool Lorentzian line fitting model was utilized to fit and quantitatively compare direct saturation (DS), magnetization transfer (MT), amide proton transfer (APT), amine, and relayed NOE (rNOE) and used as a comparison to conventional T maps. Grey and white matter tissues were segmented using the T maps, while the lesion tissue was segmented manually. Correlations between disease duration and lesion load were performed to investigate any existing relationship to image contrast. The primary findings of this work include statistically significant decreases in the rNOE pool for the normal-appearing white matter (NAWM) (11.4% decrease) and normal-appearing grey matter (NAGM) (10.6% decrease) in multiple sclerosis subjects compared to healthy controls. Additionally, a significant decrease in the amine pool was also observed for NAWM (15.3% decrease) in multiple sclerosis subjects compared to healthy controls. Changes in multiple sclerosis lesion contrast were also observed for several pools (DS, amine, and rNOE). Decreases in both the rNOE and amine pools suggest that in multiple sclerosis, there are diffuse decreases in mobile lipids, such as those found in neuronal cell bodies, as well as a decrease in proteins with amine groups. Furthermore, these measurable contrast changes were not detected in the corresponding T maps. NOE imaging can provide complementary metabolic information to conventional MRI methods. Future studies will focus on utilizing this technique for longitudinal tracking of disease progression and investigating similar demyelinating diseases.