NEURORADIOLOGY

Imaging Features  
The axial T2WI shows peri-ventricular flame-shaped hyperintense areas. Pons and corpus callosum involvement is noted on the sagittal T2WI.

Diagnosis
Multiple sclerosis

Discussion
MRI has fundamentally changed the clinical evaluation of patients with MS. It has become generally accepted that the sensitivity of MRI to MS lesions far exceeds that of the clinical examination as well as any other imaging modality, i.e., CT scanning. This basic fact is in agreement with studies that have shown significant pathologic evidence of MS in optic nerves of asymptomatic patients and reports of unexpectedly discovered MS lesions on necropsy examination of brains and spinal cords. On the other hand, while MRI has increased the sensitivity to lesions of MS, the diagnosis may still be missed in up to 25% of patients with a proven clinical diagnosis. It is widely accepted that MRI is unfortunately not specific for the diagnosis of MS, in that many white matter lesions that mimic those of MS may be detected in both normal volunteers and patients harboring other pathologic conditions, although some MRI studies report very high degrees of specificity. For these reasons, MRI data cannot be the sole criterion for the diagnosis of MS, but must be included with clinical and laboratory findings to establish the diagnosis.
Prior to the development of MRI scanning, criteria based purely on the clinical picture were employed, but have largely been replaced by others that include the information from imaging studies and CSF analysis. 

The Bartel criteria establishes the diagnosis as "possible," "probable," or "definite". Using the Bartel method, three criteria must be met before the diagnosis of multiple sclerosis can be considered "definite":

(i)                  History of neurologic symptoms with relapse and remission;

(ii)                Evidence of two or more anatomically separate lesions in the central nervous system obtained by clinical examination, electrophysiologic tests, or imaging techniques; and

(iii)               Evidence of immunologic disturbance involving the central nervous system revealed by a demyelinative spinal fluid profile. 

Diagnosis of multiple sclerosis may be considered "probable" when there is evidence of two separate lesions in the central nervous system and when the patient satisfies only one of the two remaining essential criteria. Finally, patients with evidence of a single lesion or clinical deficit, but satisfying one or both of the remaining essential criteria, would be diagnosed as "possible" multiple sclerosis.

MRI Findings in MS
The MR appearance of MS generally reflects the histologic findings. Acute inflammation appears as a rounded area of high signal intensity on T2-weighted sequences (where the repetition time, TR, is long, i.e., greater than 2 seconds). Gadolinium enhancement and even significant mass effect may be seen during the acute phase.

Acute MS (Marburg type) may also present as areas of clearly defined rings within or surrounding plaques of demyelination. These rings have signal characteristics on T1-weighted images consistent with the presence of paramagnetic material, with slight increase in signal intensity. Enhancement is typically seen in the region of these rings. This appearance most likely represents the presence of free radicals in the macrophage layer forming the margin of an acute plaque. Plaques of Balo concentric sclerosis are striking in their unique concentric rings of alternating destroyed and intact myelin.

In classic MS, the area of inflammation decreases in size with time and leaves a smaller residual plaque of high intensity, often more linear or punctate in appearance than the initial lesion. Treatment with steroids may also be associated with a marked reduction in lesion enhancement and morphology. High intensity lesions in MS do not necessarily indicate demyelination, but rather might merely reflect transient inflammation. Occasionally the plaque will contain a cavity large enough to present as a fluid-containing cyst. With progression of disease, atrophy is apparent, and increased iron deposition is concomitantly found in the basal ganglia. Plaques are commonly situated in the periventricular location, internal capsule, corpus callosum, pons, and brachium pontis but may appear throughout the myelinated white matter and not uncommonly in the gray matter. When the plaques are located in the immediate periventricular region they may not be apparent on long TR/TE images (where CSF is high intensity) and "proton-density-weighted" (long TR/short TE) images will better define the MS lesions as bright signal adjacent to darker CSF. Note that the anatomic distribution of the lesions should not be considered key to the diagnosis, since "exceptional" locations are in fact quite commonly encountered. 

The corpus callosum is a region that is especially sensitive to demyelination with multiple sclerosis, possibly due to its intimate neuroanatomic relationship to the lateral ventricular roofs and its relationship to small penetrating vessels. Sagittal MR imaging has been advocated for the depiction of the majority of the corpus callosal lesions, although many can be clearly identified on conventional axial images, particularly using proton-density-weighted parameters. In up to 93% of MS patients, focal lesions can be identified in the inferior aspect of the corpus callosum on sagittal views. Long TR images typically show focal corpus callosal lesions to best advantage, but the anatomic distortion with focal thinning in the inferior aspect of the corpus callosum may be easily appreciated on T1-weighted lesions in many cases. 

Intracranial involvement with MS may appear quite similar to many other white matter diseases on MRI, with scattered foci of high intensity in the white matter on T2-weighted spin-echo images. Contrast enhancement may be used to add specificity to the finding of multiple hyperintensities on T2-weighted images, since the finding of enhancing along with nonenhancing lesions is quite common in MS but makes many other diagnoses unlikely. Similarly, the temporal changes in enhancing and nonenhancing lesions common in MS cases is very different from other entities. Periventricular lesions of MS commonly appear as linear abnormalities oriented perpendicular to the lateral ventricle, due to the propensity of MS plaques to occur in a periventricular location. Although this is common, the appearance of an MS lesion on MRI is highly variable and certainly not specific. MS can also appear as very subtle, diffuse hyperintensity in the white matter. Moreover, it has been shown by several studies that quantitative analysis of relaxation times in "normal appearing" white matter in MS patients differs from that of normal volunteers.