Despite the well-known fact that cervical spondylotic myelopathy (CSM) is the major cause of spinal cord dysfunction in people over 55 years old in North America, the relationship between the degree of stenosis, tissue damage, and sensorimotor deficits is not yet well understood. Conventional approaches for CSM diagnosis rely mainly on clinical indications, as well as signal hyperintensities on MRI of the spinal cord (i.e., qualitative measures). As a complementary diagnostic approach, somatosensory evoked potentials (SSEPs)¹ represent an objective measure of conduction deficits in the dorsal columns of the spinal cord. Myelin water imaging (MWI) is a validated MRI method specific to myelin which can be used to quantitatively study pathology in CSM³. The goal of our study was to investigate the relationship between myelin content, using MWI, and tibial SSEP in subjects with CSM, and to assess the sensitivity of myelin water imaging to clinical findings of demyelination.

Subjects: 14 subjects diagnosed with CSM based on clinical deficits (mean age 60 years, range 50-69 years) and 17 age-matched healthy adults (mean age 59 years, range 51-75 years) were studied.

Electrophysiology: Standard tibial SSEPs were retrieved by stimulation of the posterior tibial nerves at the medial ankle with combined recordings at the popliteal fossa (to assure appropriate stimulation and exclude impairment of peripheral nerve conduction) and cortical levels. SSEP amplitude, latency and configuration were used by a trained neurologist to score subjects as either ‘normal’ or ‘pathological’⁴.

MRI: Subjects were scanned on a 3.0T MRI system (Philips, Best, The Netherlands) with a phased array spine coil using only the first four channels. Myelin water imaging was performed by a T₂ relaxation measurement using a 3D 32-echo sequence (1st echo =10ms, echo spacing=10ms, TR=1300ms, six 5mm thick axial slices perpendicular to the spinal cord, 256×128 matrix, FOV 180mm×135mm, one acquisition)⁵. The stack was centered at the level of stenosis in CSM subjects, and at the C5 level in controls.

Data analysis: Voxel-wise T₂ decay curve analysis used in-house MATLAB software which employed the extended phase graph algorithm to estimate the refocusing flip angle in each voxel as well as correcting the T₂ decay curve for stimulated echo artifacts⁶. The myelin water fraction (MWF), which measures water trapped between myelin bilayers, was defined as the fractional signal with T₂ less than 35ms. For each subject, the average MWF in the whole cord and dorsal column were calculated by drawing regions of interest (ROI) on T₂ weighted images and combining the ROIs over all slices to yield a volume of interest. Statistics: A Mann-Whitney U test was performed to examine differences in MWF between groups. Linear regression analysis examined the correlation between MWF and SSEP.

Although CSM shows a trend towards reduced MWF, no significant differences in MWF of the whole cord or dorsal column were observed between controls and patients with CSM, (for healthy controls and CSM, respectively: whole cord mean (stdev): 0.26 (0.03) versus 0.24 (0.04); dorsal column: 0.29 (0.04) versus 0.27 (0.04), p>0.05; Figure 1). When categorizing all study participants (controls and CSM) as having either normal or pathological SSEP, individuals with pathological SSEPs had significantly reduced MWF in the dorsal column, but not the whole cord (for normal and pathological SSEP, respectively: whole cord 0.26 (0.03) versus 0.24 (0.05); dorsal column: 0.30 (0.04) versus 0.26 (0.04), p<0.05; Figure 2). A linear regression demonstrated a significant negative relationship between SSEP latency (both left and right) and MWF in the dorsal columns (Figure 3) in CSM group while no relationship was found in the healthy control group.Combining neurophysiological and neuroimaging data reveals a significant correlation between SSEP latency and MWF. There was a clear agreement between SSEP classification and MWF values. Delayed SSEP latencies were related to a reduction in MWF, especially in the CSM group dorsal column. Our findings suggest that there is focal demyelination in the dorsal column, and thus myelin water imaging may be a novel tool to assess clinical interventions aimed at treating diseases of myelin in spinal cord white matter.