Relating neocortical pathology to disability progression in multiple sclerosis using MRI
Introduction
MR imaging and histopathology have both shown that the cortical grey matter (cGM) develops a substantial burden of pathology in patients with multiple sclerosis (MS). Postmortem histopathology of the brains of MS patients over a large range of disease duration shows lesions in the cGM (Brownell and Hughes, 1962, Kidd et al., 1999, Lumsden, 1970, Peterson et al., 2001). Proton MR spectroscopy has shown that the concentration of N-acetyl-aspartate (NAA), a biomarker of neuronal integrity, is low in the cGM of patients with early MS (Chard et al., 2002, Kapeller et al., 2001) and secondary-progressive MS (DiMaio et al., 2003). Conventional MRI is insensitive to cortical lesions per se, but image processing approaches confirm the presence of cortical pathology. For example, hypointensity on T2-weighted images in Rolandic cortex has been noted and suggested to be related to pathological iron deposition (Bakshi et al., 2000). A cross-sectional study of T1-weighted MRI scans from early MS patients found lower cGM volume compared to healthy controls (De Stefano et al., 2003). Using high-resolution MRI, a cross-sectional study found reduced thickness globally in MS patients compared to healthy controls, as well as regional decreases in frontal and temporal regions, even in patients with early stage MS (Sailer et al., 2003). In later disease stages, focal loss of cortical thickness was also observed the precentral region (Sailer et al., 2003).
In this paper, we investigated how global and regional measures of cGM pathology in MS differed between MS patients with a stable disease course compared to patients with a progressive disease course. To do this, we characterized the cGM with two measures derived from conventional T1-weighted MRI. The first is a novel method to automatically measure the apparent cGM thickness. The second is a measure related to the integrity of the interface between the cortical grey and subcortical white matter (GM/WM) to quantify pathology affecting this interface. These measures were performed globally and within specified cortical regions both cross-sectionally at baseline and longitudinally. For each subject, we used MRI obtained at two time points on average 1 year apart, as a measurable amount of global atrophy can occur over this time interval, and compared patients who were stable over the interscan interval to patients who worsened over the interscan interval.
Section snippets
Patients
Patients were selected from the Multiple Sclerosis Clinic of the Montreal Neurological Hospital. All met the following criteria: (1) clinically definite MS with a RR or SP course; (2) expanded disability status scale (Kurtzke, 1983) EDSS ≤ 5; (3) at least 1-month interval between the scan and the last exacerbation; and (4) availability of two MRI scans with identical acquisition parameters performed approximately 1 year apart. To investigate changes on MRI during disability progression, we
Results
The subject demographics are shown in Table 1. The groups do not significantly differ from each other on any baseline characteristic.
Discussion
The objective of this work was to investigate how automatically obtained global and regional characteristics of cGM relate to disease course in MS. This allowed us to test directly whether patients that progress in disability also progress with respect to MRI-visible cortical pathology. By investigating whether cross-sectional variables were significantly different between the groups at baseline, the possibility that progression could be predicted was tested as well.
Summary
Detecting cortical pathology on MRI in MS is difficult. Here we have presented an alternative strategy by which the evolution of pathology can be inferred from clinically practical scanning using an image processing methodology that does not require user intervention and is sensitive to changes at a subvoxel resolution. We have applied this to the important problem of distinguishing pathological characteristics associated with disability progression. In previous studies, we have shown that
References (34)
- et al.
A global optimisation method for robust affine registration of brain images
Med. Image Anal.
(2001) - et al.
Automated 3-D extraction of inner and outer surfaces of cerebral cortex from MRI
NeuroImage
(2000) - et al.
Evidence for axonal pathology and adaptive cortical reorganization in patients at presentation with clinically isolated syndromes suggestive of multiple sclerosis
NeuroImage
(2003) - et al.
Accurate, robust, and automated longitudinal and cross-sectional brain change analysis
NeuroImage
(2002) - et al.
MRI T2 shortening (‘black T2’) in multiple sclerosis: frequency, location, and clinical correlation
NeuroReport
(2000) - et al.
The Human Brain in Figures and Tables
(1968) - et al.
Distribution of plaques in the cerebrum in multiple sclerosis
J. Neurol. Neurosurg. Psychiatry
(1962) - et al.
Brain metabolite changes in cortical grey and normal-appearing white matter in clinically early relapsing–remitting multiple sclerosis
Brain
(2002) - et al.
In vivo evidence of cortical neuronal injury in secondary progressive multiple sclerosis
Proc. ISMRM
(2003) - et al.
Axonal damage correlates with disability in patients with relapsing–remitting multiple sclerosis. Results of a longitudinal magnetic resonance spectroscopy study
Brain
(1998)
Evidence of early cortical atrophy in MS: relevance to white matter changes and disability
Neurology
Administration and Scoring Manual for the Multiple Sclerosis Functional Composite Measure (MSFC)
Imaging axonal damage of normal-appearing white matter in multiple sclerosis
Brain
Brain segmentation and the generation of cortical surfaces
NeuroImage
Preliminary evidence for neuronal damage in cortical grey matter and normal appearing white matter in short duration relapsing–remitting multiple sclerosis: a quantitative MR spectroscopic imaging study
J. Neurol.
Cortical lesions in multiple sclerosis
Brain
Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS)
Neurology
Cited by (140)
Age-related changes in multiple sclerosis and experimental autoimmune encephalomyelitis
2022, Seminars in ImmunologyMultivariate fuzzy analysis of brain tissue volumes and relaxation rates for supporting the diagnosis of relapsing-remitting multiple sclerosis
2019, Biomedical Signal Processing and ControlCitation Excerpt :In the validation of this method, which shows a background of less than 0.6 ml of lesion load in normal subjects, results of the fully automated segmentation program were compared with manual selection of abnormal WM reporting a mean sensitivity in the detection of MS lesions of 87.3% [11], without lesion-related bias in the segmentation of GM [39]. These features, and in particular operator independence and lack of AWM-related bias [40–44], render the selected method ideal for large studies such as the present one, eliminating operator-related bias and avoiding to introduce potential artificial cross-correlation in the data. In Fig. 1, an example of QMCI of axial slices in correspondence of different levels in a RR-MS patient is reported along with the corresponding segmented image obtained with the described method.
How changes in brain activity and connectivity are associated with motor performance in people with MS
2018, NeuroImage: ClinicalCortical Gray Matter MR Imaging in Multiple Sclerosis
2017, Neuroimaging Clinics of North America