August 2014

Brain Atrophy in Multiple Sclerosis

Alex Rovira

Alex Rovira

Multiple sclerosis (MS) is a chronic, persistent inflammatory-demyelinating disease of the central nervous system (CNS), characterized pathologically by areas of inflammation, demyelination, axonal loss, and gliosis scattered throughout the CNS. The disease has a predilection for the optic nerves, brain stem, spinal cord, and cerebellar and periventricular white matter, though cortical and subcortical gray matter damage is also prominent.

Although MS is typically characterized by focal areas of white matter demyelination and inflammation, significant brain atrophy, a measure of the most destructive pathological process occurring in MS, is also seen from the earliest stages of the disease and may lead to irreversible neurological and cognitive impairment. In fact, whole-brain atrophy has emerged as a clinically relevant component of disease progression. Several studies have shown that this parameter correlates better with disability and cognitive impairment than focal lesions, and it has been recognized as the most feasible and robust measure of the neurodegenerative aspect of MS and as a plausible surrogate outcome measure for disability. In addition to reducing inflammation, control of the neurodegenerative component is an important target for treatment. For this reason, brain volume measurements have been used in randomized clinical trials to monitor the effect of various disease-modifying therapies on reducing this outcome.

Progressive loss of whole or regional brain volume can be detected in vivo in a sensitive and reproducible manner by MR imaging, mainly with the use of quantitative measures acquired by automated techniques. These techniques can be classified into 2 broad categories: segmentation-based and registration-based. Segmentation-based techniques enable measurement of whole or regional (eg, gray matter) brain volume at a single time point, while registration-based techniques are able to measure changes in whole-brain volume over time by comparing 2 sets of MR imaging scans acquired at different time points.

Regional analysis of brain volume has clinical implications related to disease progression. Gray matter (GM) tissue loss is of particular importance in MS because GM makes up more than half the total brain parenchyma, and damage to this tissue is a large component of the overall MS disease burden. Moreover GM atrophy reflects the disease subtype, disability, and neuropsychological impairment to a greater extent than white matter (WM) atrophy or focal T2 lesion load, indicating that GM atrophy is a clinically relevant marker of disease progression. Unfortunately, assessment of cortical and deep GM volume, though likely to be more clinically relevant, is technically more demanding and less robust than whole-brain volume assessment, and more difficult to implement in longitudinal clinical studies.

This issue of the AJNR News Digest focuses on the work of several authors who provide data on the clinical relevance of brain atrophy in patients with MS and report various postprocessing algorithms that can be used for cross-sectional and longitudinal studies involving brain MR imaging.

In a study performed in a cohort of 54 patients with MS, Sämann et al1 demonstrated the value of whole-brain diffusivity and brain volume for predicting cross-sectional disability and the short- to medium-term clinical evolution. These authors showed that baseline brain volume is associated with disability progression regardless of intercurrent relapses, supporting the concept that these MRI-based measures are clinically valid and sensitive markers to monitor cerebral disease burden in MS during a clinically short time interval. This capability could have an impact on selection of the most appropriate treatment in individual cases.

In an elegant article, Calabrese et al2 analyzed structural alterations in the brain cortex in pediatric and adult patients with MS of short duration. In particular, the authors assessed the number and volume of focal cortical lesions and volume of cortical GM and observed that cortical lesion burden and atrophy progressed at the same rate in the 2 groups. Furthermore, they found that progression of cortical atrophy was significantly greater in both MS groups compared with the control group, thus confirming the hypothesis that GM pathology is an early phenomenon in MS. In addition, these authors showed that progression of GM pathology did not correlate with the extent of WM lesion volume, supporting the concept that GM degeneration proceeds in a manner largely independent of WM inflammation.

Vågberg et al3 proposed a novel, reproducible, and fast automatic method for calculating brain parenchymal fraction, named synthetic tissue mapping, based on absolute quantification of tissue relaxation rates (R1 and R2) and proton attenuation. This method demonstrated a strong correlation with a manual segmentation technique and a relatively low coefficient of variation (0.45%), which was, however, not too different from

the annualized brain volume change observed in patients with MS (0.5–1.3%). The study results support the use of this method for calculating whole-brain volume in cross-sectional MS studies, but it should be used with caution in longitudinal studies, particularly when analyzing changes in short time periods (less than 2 years), because of the small brain volume changes that occur in MS.

Durand-Dubief et al4 investigated the robustness of 7 postprocessing algorithms for longitudinal brain volume loss measurement applied to images acquired with different MR imaging systems in a small group of patients with MS. The algorithms tested were based on either segmentation or registration-based techniques. As could be expected, the results showed that segmentation-based techniques provide larger and more heterogeneous results than registration-based techniques, supporting the use of these algorithms for longitudinal studies.

Bergsland et al5 performed a cross-sectional analysis of regional and whole-brain volume in a relatively large number of patients with clinically isolated syndrome and early relapsing-remitting MS using different segmentation-based algorithms. The results indicate that there is significant deep GM volume loss in the early phases of MS, while cortical GM volume is preserved. These results are consistent with those from a recent meta-analysis6 that found a highly localized pattern of regional GM volume loss in relapsing-remitting MS, mainly involving the thalamus bilaterally and basal ganglia structures.

Finally, Cappellani et al7 performed a similar study in a group of patients with relapsing-remitting and progressive MS, but instead of analyzing volumetric changes in deep GM, they assessed microstructural abnormalities using different diffusion tensor imaging (DTI) metrics. DTI metrics were significantly different in patients with MS compared with healthy controls, and were associated with T2 lesion load and WM volume but not with cortical GM volume, indicating that deep GM microstructural changes in patients with MS are related to the WM burden of the disease and are independent of the diffuse cortical GM neurodegenerative component.

Brain atrophy has emerged as a clinically relevant MR imaging–based measure in MS that can be identified in a sensitive and robust manner using various automatic algorithms. There may be some technical and logistic difficulties in implementing brain volume measures as a prognostic marker in individual patients, and there is a lack of definite evidence supporting their prognostic and therapeutic impact. Nonetheless, it is likely that they will be used as predictors of treatment response and disease progression in the near future and will have a major impact on treatment decisions, mainly related to a new generation of MS drugs that seem to have a beneficial effect on preventing the neurodegenerative component of the disease.


  1. Sämann PG, Knop M, Golgor E, et al. Brain volume and diffusion markers as predictors of disability and short-term disease evolution in multiple sclerosis. AJNR Am J Neuroradiol 2012;33:1356–62, 10.3174/ajnr.A2972
  2. Calabrese M, Seppi D, Romualdi C, et al. Gray matter pathology in MS: a 3-year longitudinal study in a pediatric population. AJNR Am J Neuroradiol 2012;33:1507–11, 10.3174/ajnr.A3011
  3. Vågberg M, Lindqvist T, Ambarki K, et al. Automated determination of brain parenchymal fraction in multiple sclerosis. AJNR Am J Neuroradiol 2013;34:498–504, 10.3174/ajnr.A3262
  4. Durand-Dubief F, Belaroussi B, Armspach JP, et al. Reliability of longitudinal brain volume loss measurements between 2 sites in patients with multiple sclerosis: comparison of 7 quantification techniques. AJNR Am J Neuroradiol 2012;33:1918–24, 10.3174/ajnr.A3107
  5. Bergsland N, Horakova D, Dwyer MG, et al. Subcortical and cortical gray matter atrophy in a large sample of patients with clinically isolated syndrome and early relapsing-remitting multiple sclerosis. AJNR Am J Neuroradiol 2012;33:1573–78, 10.3174/ajnr.A3086
  6. Lansley J, Mataix-Cols D, Grau M, et al. Localized grey matter atrophy in multiple sclerosis: a meta-analysis of voxel-based morphometry studies and associations with functional disability. Neurosci Biobehav Rev 2013;37:819–30, 10.1016/j.neubiorev.2013.03.006
  7. Cappellani R, Bergsland N, Weinstock-Guttman B, et al. Subcortical deep gray matter pathology in patients with multiple sclerosis is associated with white matter lesion burden and atrophy but not with cortical atrophy: a diffusion tensor MRI studyAJNR Am J Neuroradiol 2014;35:912–19, 10.3174/ajnr.A3788


Image from: Bergsland N, Horakova D, Dwyer MG, et al. Subcortical and cortical gray matter atrophy in a large sample of patients with clinically isolated syndrome and early relapsing-remitting multiple sclerosis.