Brain atrophy and lesion load in a large population of patients with multiple sclerosis
Tedeschi G, Lavorgna L, Russo P, Prinster A, Dinacci D, Savettieri G, et al.;
Commented by , 22 Aug 2005
Background
Axonal degeneration is the major determinant of irreversible neurological disability in multiple sclerosis (MS). The morphological consequence is CNS atrophy, which may be measured by MRI.
Various approaches have been developed in this regard, although the differential changes in white and gray matter due to disease course, lesion load and clinical features are poorly described. The current study uses a large population of patients and an unusual MRI approach to enlighten these issues.
Aim
To measure lesion load as well as white and gray matter atrophy in MS patients using a fully automated, multiparametric segmentation technique.
Methods
The authors studied 597 MS patients and 104 controls in a cross-sectional study. The MRI machine was installed in a truck in order to perform the same MRI protocol in subjects recruited from 7 different MS clinics.
All MRI studies were segmented using a fully automated algorithm, providing values of CSF gray matter (GM), normal appearing white matter (NAWM), abnormal WM (AWM). Volumes of intracranial tissues were expressed as fractions (f) of intracranial volume.
Results
MS patients had increased AWM fraction (AWM-f) and reduced global WM-f (gWM-f), as well as reduced GM-f compared with controls.
Secondary progressive (SP, n=140) MS patients had more pronounced increase in AWM-f and reduced gWM-f as well as GM-f compared with relapsing remitting (RR, n=427) patients. Primary progressive (PP, n=30) patients did not differ from RR and SP patients in these measurements (insufficient number of PP patients?)
Intercorrelations between segmented MRI fractions showed inverse correlation between lesion load (AWM-f) and all other measurements but the strongest relationship was found with GM-f.
Multivariate correlation analysis with clinical features showed:
- Lesion load was predicted by disability, age at onset (early) and drug treatment (b-INF)
- GM-f and gWM-f were predicted by age at onset and disability
- Most important predictors of disability were disease course (SP), disease duration and age at onset
- GM-f is the most important MRI variable determining disability
Dr Blinkenberg's comments
The advantages of this study, compared with other MRI atrophy studies, are the large number of subjects and the uniformity regarding data acquisition and analysis. First of all it confirms that lesion load is a significant determinant for cerebral and cortical atrophy.
Secondly, it shows that SP patients have more pronounced GM and WM atrophy compared with RR patients, suggesting that the degree of atrophy, and the underlying axonal/neuronal loss, may be determining for the evolvement of relapsing to progressive disease course.
This finding is in accordance with the hypothesis that permanent neurological disability develops when a threshold of axonal loss is reached and CNS redundancy is exhausted.
Furthermore, the authors present a number of clinically relevant correlations, the most important of these being that disability and disease duration predict lesion load as well as WM and GM atrophy, and that GM atrophy is the most important MRI determinant for disability.
The study provides insight in the future possibilities of MRI segmentation with regard to pathophysiology and surrogate measures of disease progression. Imaging of GM fraction is theoretically one of the best parameters for quantifying the irreversible neuronal deterioration in MS and more studies on this issue are therefore of great importance.
Editor's note: For more on a related topic read Dr Massimo Filippi's and Dr Maria Rocca's expert review "Functional MRI in the Study of Brain Plasticity in MS"