Neuroimaging in Multiple Sclerosis

Computerized tomography (CT) neuroimaging for the first time revealed areas of decreased radiodensity in the brain as well as occasional enhancing brain and spinal cord lesions in MS. Interestingly, increasing brain atrophy, although reported early, was largely ignored by the MS community [43-45]. Comparative studies of CT and MRI revealed the relative strength of MRI in visualizing plaques as well as brain

Table 2.3 Prognostic indicators in MS [41, 42]






Age at onset

Young (< 35 years)

Older (>35 years)




Tobacco abuse



First attack characteristics

Optic neuritis, sensory, unifocal

Motor, cerebellar, sphincter, multifocal incomplete

MRI lesion location


Spinal cord

Brain lesion burden



Lesion enhancement on MRI



Recovery after relapse



Attack rate


High (>2 in 1 year)

MS subtype



Disability at 5 years



atrophy in MS [46-48]. In contrast to the limitations encountered with the use of CT, MRI has had an important impact on both the diagnosis and subsequent management of MS because of the relative ease which it can detect white matter lesions in the brain and spinal cord.

Investigators have sought brain MRI correlations with clinical symptoms of MS, prognosis of the illness, other laboratory findings, as well as with central nervous system pathology. Increased T2 signal, reflecting increases in water content of lesions in hemispheric white matter, was emphasized in earlier studies, but their presence correlates poorly with symptoms and neurological findings (Fig. 2.1a). In our initial experience with this imaging modality, we found that very early in the course of clinical disease, only half of patients with clinically definite MS did have cerebral white matter lesions [47, 49]. However, almost half of those that did not have plaques in their brains exhibited spinal cord lesions that were clearly evident [50]. While, not all cerebrospinal fluids (CSF) had “diagnostic” abnormalities, only 5% of patients did not have either brain MRI abnormality or significant CSF abnormality. In part, the difficulty with the MRI findings in these early studies was related to technical issues such as image slice thickness, noncontiguous sections, etc. Use of fluid-attenuated inversion recovery (FLAIR) sequences, which are easier to visualize, has been made practicable by advances in the hardware and software (Fig. 2.1b). Newer acquisition paradigms and the use of gadolinium to identify “active” inflammatory lesions, in particular, as well as continued hardware improvements have remarkably improved the quality and utility of MRI. However, not all patients with MS, particularly those with PPMS, exhibit white matter lesions in their cerebral hemispheres. The absence of MRI abnormality does not negate the diagnosis of MS [9]. We found that after 9-12 years, the same proportion of MS patients will have white matter lesions evidence by MRI and by pathology, however [47, 49]. In a recent presentation from the Cleveland Clinic, Dr. Robert Fox revealed that

MRI scans of the brain of a 19-year-old woman with relapsing-remitting multiple sclerosis

Fig. 2.1 MRI scans of the brain of a 19-year-old woman with relapsing-remitting multiple sclerosis. Axial T2-weighted (a) and fluid-attenuated inversion recovery (b) views show hyperintense lesions in subcortical white matter. Axial T1-weighted postcontrast (c) of the same patient reveals an enhancing lesion, indicating the breakdown of the blood-brain barrier

approximately 20% of their well-documented patients with progressive MS did not have hemispheric white matter lesions at necropsy [50]. They do, however, have cortical as well as spinal cord, i.e., “corticospinal” involvement. Cortical involvement in MS is rarely evident with standard imaging parameters. Double inversion recovery is capable of documenting about 40% of the cortical lesions found in pathological study [51].

A strong correlation between increased volume of cerebral MRI T2 signal and long-term disability in MS has been reported in patients followed for 5 years after the onset of a clinically isolated syndrome. However, further follow-up of this cohort of patients has shown only a moderate correlation at 10 years [52]. A number of short-term correlations between stabilization, or reduction, of T2 volumes and clinical stabilization in patients treated with each of the immunomodulatory drugs are currently approved. After the initial 5 years of illness, with some notable exceptions, changes from 1 year to the next are difficult to see in brain MRI scans. Clearly, there must be some reservation about the use of T2 lesion volumes for assessment of longer-term treatment of any kind.

Gadolinium enhancement of white matter lesions is an accepted indicator of active disease, but enhancing lesions are seen several times more often than acute exacerbations of illness in multiple sclerosis (Fig. 2.1c). This surrogate measure of disease activity has been used effectively in preliminary drug efficacy studies to detect a treatment effect. Despite the earlier negative reports, Leist et al. reported a correlation between gadolinium-enhancing lesions and the subsequent appearance of cerebral atrophy [53]. Unlike the earlier studies reporting on correlation, this NIH study was based on frequent (monthly) gadolinium-enhanced brain MRI studies.

Although T1 hypointensities have been reported to correlate with cerebral atrophy, other studies have shown that this type of MRI lesion does not correlate well with either the amount of demyelination or gliosis in tissue lesions. The lack of correlation with tissue changes makes it difficult to understand and accept these observations at face value [54, 55]. Importantly, De Stefano et al. have reported data supporting a role between early axonal damage and subsequent development of disability in multiple sclerosis [66].

Brain atrophy progresses at a rate of 0.5-1.0% per year in patients with MS, considerably higher than the typical rate seen with normal aging at 0.1-0.3% per year. Once thought to be largely a disease of white matter, MS is now recognized to have significant manifestations in the gray matter [56]. The volumetric changes seen on MRI during the course of MS have been correlated with disability progression and cognitive impairment; however, the quantitative cutoffs to determine physiologic versus pathological brain atrophy in MS remain to be determined.

No evidence of disease activity (NEDA) has been proposed as a potential treatment goal for treatment trials in MS. Elimination of relapses and prevention of disease progression, including cognitive loss and impaired ambulation, are the clinical goals (Fig. 2.2).

NEDA-3 includes (1) no sustained increase in disability lasting 3 months, (2) no relapses, and (3) no MRI activity, defined as no new or enlarging T2 and Gad+ lesions. NEDA-4 includes similar parameters, with the addition of no annual brain volume loss >0.4%. NEDA-3 status appears to correlate with subsequent relapse and focal inflammatory MRI activity. NEDA-4, in utilizing measures for tissue destruction at both the focal inflammatory and diffuse level, may be a more comprehensive predictor for subsequent disability-related outcomes. NEDA-4 data has been collected using post hoc analyses of the FREEDOMS and FREEDOMS-II trials [57, 58].

More advanced imaging methods continue to be explored. Double inversion recovery (DIR) can be used to demonstrate cortical inflammatory lesions, although its use is limited by inadequate resolution and inability to identify purely intracortical, versus juxtacortical or leukocortical, lesions [51]. Diffusion tensor imaging

Disease-free concept

Fig. 2.2 Disease-free concept: NEDA-4

(DTI) is used to evaluate the structural integrity of the white matter tracts. DTI can be used for diffusivity measures including mean diffusivity and fractional anisotropy, which may provide even closer evaluation of tissue integrity and axonal damage [56, 59-61]. The value of proton magnetic resonance spectroscopy continues to be investigated and has resulted in many claims that are not entirely consistent. The advent of higher Tesla field strengths, up to ultrahigh-field 7-8 Tesla, has improved characterization of cortical demyelination, with good pathologic correlation but is restricted to research studies for safety reasons [62].

It is obvious that MRI is especially helpful in the evaluation of patients early in the course of their illness. Unfortunately, the question as to the utility of using MRI or other surrogate measures to evaluate the long-term response to treatment remains essentially unanswered. Cerebral atrophy may very well be the most valuable measure.

< Prev   CONTENTS   Source   Next >