Neuroimaging of motor neuron diseases

Abstract

It is agreed that conventional magnetic resonance imaging (MRI) of the brain and spine is one of the core elements in the differential diagnostic work up of patients with clinical signs of motor neuron diseases (MNDs), for example amyotrophic lateral sclerosis (ALS), to exclude MND mimics. However, the sensitivity and specificity of MRI signs in these disorders are moderate to low and do not have an evidence level higher than class IV (good clinical practice point). Currently computerized MRI analyses in ALS and other MNDs are not techniques used for individual diagnosis. However, they have improved the anatomical understanding of pathomorphological alterations in gray and white matter in various MNDs and the changes in functional networks by quantitative comparisons between patients with MND and controls at group level. For multiparametric MRI protocols, including T1-weighted three-dimensional datasets, diffusion-weighted imaging and functional MRI, the potential as a 'dry' surrogate marker is a subject of investigation in natural history studies with well defined patients. The additional value of MRI with respect to early diagnosis at an individual level and for future disease-modifying multicentre trials remains to be defined. There is still the need for more longitudinal studies in the very early stages of disease or when there is clinical uncertainty and for better standardization in the acquisition and postprocessing of computer-based MRI data. These requirements are to be addressed by establishing quality-controlled multicentre neuroimaging databases.

Keywords: DTI; MRI; T1-weighted imaging; amyotrophic lateral sclerosis; motor neuron diseases.

Figure 1.

Examples of magnetic resonance imaging (MRI) findings in individual patients with motor neuron disease. (A) T2-weighted MRI hyperintensity along the corticospinal tract (axial view, at the level of the posterior limb of the internal capsule) in a 62-year-old patient with clinically definite amyotrophic lateral sclerosis (ALS). (B) Regional motor cortex atrophy and signal alterations in the adjacent centrum semiovale in a coronal fluid-attenuated inversion recovery (FLAIR) image in a 50-year-old patient with primary lateral sclerosis (PLS). (C) Symmetrical hyperintensities along the pyramidal tracts in a coronal FLAIR image of a 64-year-old patient with PLS. (D) Regional atrophy of the motor segment of the corpus callosum in T1-weighted MRI (sagittal slice) in a 53-year-old patient with ALS. (E, F) Marked frontotemporal atrophy (T1-weighted MRI in axial and sagittal view) in a 54-year-old patient with advanced ALS frontotemporal lobar degeneration complex.

Figure 2.

Comparison of fractional anisotropy (FA) maps based on diffusion tensor imaging (DTI) data of 20 patients with amyotrophic lateral sclerosis (ALS) and 20 age- and gender-matched controls. Upper panel: group averaged FA maps of controls (left) and patients with ALS (right) in coronal (large) and axial/sagittal view. FA display threshold is 0.2. Lower panel: left: comparison between the ALS group and the controls by whole brain based statistical voxelwise comparison at group level, at p < 0.05 after correction for multiple comparisons. The areas with decreased FA in ALS are displayed, with the significance of the alterations coded by temperature of the color bar. Right: fiber tracking of the corticospinal tract (CST) in group-averaged DTI datasets. The underlying FA values were averaged and statistically compared. Differences between group-averaged ALS FA maps and group-averaged control FA maps were highly significant as indicated.