Conventional MRI has a major role in the recently developed diagnostic criteria for MS, because of its exquisite sensitivity for detecting MS lesions and their changes over time. Conventional MRI lesions are seen as brain (and spinal cord) multiple foci of various size, irregular shape, and asymmetric distribution of white matter hyperintensity on T2-weighted
images. Abnormalities seen on T2-weighted images may reflect edema, demyelination, remyelination, gliosis, or axonal loss, with a lack of pathological specificity. A subset of these lesions appear as hypointense on T1-weighted images, probably more specifically representing areas of axonal loss and severe matrix destruction. On post-gadolinium T1-weighted images, some MS lesions can appear hyperintense, reflecting intense inflammatory activity and mononuclear cell infiltration.
Despite the sensitivity of conventional MRI for detecting MS lesions, it does have some important limitations. First, there is low pathological specificity of the abnormalities seen on conventional MRI scans.
Second, there is the inability of conventional MRI metrics to detect and quantify the extent of damage in normal-appearing brain tissues, which are known to be involved in the pathological process.

These limitations probably result in the limited correlation that is found to exist between the conventional MRI metrics and patients’ clinical status in MS.
These inherent limitations of conventional MRI have prompted the development and application of modern quantitative MR techniques such as 1H-MRS, magnetization transfer (MT) MRI, diffusion-weighted and functional MRI (fMRI) to the study of MS.

1H-MRS

In the last decade, a great number of 1H-MRS studies have provided in vivo accurate chemical–pathological characterization of MR-visible lesions and normal appearing brain tissues in MS brains. In demyelinating lesions large enough to allow spectra to be acquired without substantial partial volume effects, 1H-MRS at both short and long echo times reveals increases in Cho and sometimes lactate, resonance intensities from the early phases of the pathological process. Changes in the resonance intensity of Cho can be interpreted as a measure of increases in the steady-state levels of membrane phospholipids released during active myelin breakdown. Increases in Lac may reflect primarily the metabolism of inflammatory cells. In large, acute demyelinating lesions decreases of Cr can also be seen.
Short echo time spectra give evidence for transient increases in visible lipids (released during myelin breakdown), and more stable increases in mI. These changes are consistently accompanied by substantial decreases in NAA, interpreted as a measure of axonal injury reflecting metabolic or structural changes. Recently, glutamate levels were found to be elevated in acute lesions suggesting a link between axonal injury in active lesions and glutamate excitotoxicity.
After the acute phase and over a period of days to weeks, there is a progressive return of raised Lac resonance intensities to normal levels in focal lesions. Cr also returns to normal within a few days, or may show small residual increases, presumably related to gliosis. Persistent increases in mI signals in chronic lesions may be related to microglial proliferation. Resonance intensities of Cho and lipids typically return to normal over months.
The signal intensity of NAA may remain decreased or show partial recovery, starting soon after the acute phase and lasting for several months. The recovery of NAA may be related in various proportions to reversible metabolic changes in neuronal mitochondria, the resolution of edema, or changes in the relative partial volume of neuronal processes.
Initial 1H-MRS studies were focused mainly on MRI-defined lesions. However, more recent studies exploiting the greater coverage and resolution of 1H-MRSI have shown that metabolic abnormalities in MS patients are not restricted to lesions, but are present both adjacent to and distant from the lesions.
The NAA decreases found in the normal-appearing white matter are usually attributed to axonal damage, and, although they can be present at early disease stages, are more pronounced in advanced disease stages. The extent of this NAA reduction decreases with the distance from the core of a lesion, consistent with the notion that the diffuse changes are at least in part related to dying back of axons transected within plaques. However, decreased levels of NAA also occur without obvious relation to T2-visible lesions.
Recent 1H-MRS studies have focused on gray matter metabolic changes in MS patients, supporting the notion that the contribution of gray matter pathology is substantial in MS. It has been found that cortical decreases in NAA might be small or absent early, but seem to be considerable in patients with progressive disease. In contrast, subcortical gray matter decreases in NAA seem to be more consistently found from early stages. In some studies, 1H-MRS and histopathological methods have been used in parallel and the amount of ex vivo total loss of thalamic neurons was comparable to the in vivo NAA decrease.
A number of spectroscopic studies have demonstrated highly significant correlations between NAA/Cr and clinical disability in patients with isolated acute demyelinating lesions, and in patients with established MS followed through periods of relapse and remission.
Consistent with other evidence of widespread pathology in MS, a strong correlation also has been found between NAA/Cr decreases and increases in clinical disability in normal-appearing WM. Since changes in Cr could contribute to any changes in NAA/Cr, it has been suggested that it would be more accurate to interpret decreases of brain NAA/Cr as markers of a less specific disturbance in the “cerebral tissue integrity”.
Despite its potential to monitor the temporal evolution of metabolite changes reflecting tissue integrity in demyelinating lesions and normal-appearing brain tissue, the use of 1H-MRS in longitudinal studies to monitoring the response to drug therapies are uncommon, and its large-scale use as a primary or secondary endpoint in clinical trials has not been attempted. However, recently recommendations for a standardized use 1H-MRS protocol in MS multicenter clinical studies have been provided.

A
B
C
D
	Figure-1: A young patient with second recent attack of multiple sclerosis, the first was involving the pons and hemispheres and the second recent involving the spinal cord at the level of D7-8. Spectroscopic studies were performed in the old brain lesions with short and medium TE. Choline, NAA, Cr and Lac were studied in details. The conclusion from this case was that NAA levels (C) were distributed evenly in the normal brain and the MS plaques. Ch (A) was increasing dramatically over the plaques and still having elevation of the level outside the plaques. Lactate (D) was decreased all over. The Ch/NAA level (B) showed elevation in the plaque secondary to the above obtained results. The studies were performed in Skyra 3 tesla magnetom. Notice the spinal cord recent lesion, which at the present time is difficult to perform the spectroscopic studies at this area, because of technical limitations (E).
E