Simultaneous Brain and Cervical Resting State Functional MRI in Patients with Cervical Spondylotic Myelopathy

Introduction: Cervical Spondylotic Myelopathy (CSM) disrupts the neural pathways throughout both the brain and body, resulting in permanent neurological disability. The pathological processes underlying CSM are particularly unique since increasing evidence suggests an interplay between distinct circuitry in the brain and the cervical spinal cord (cSC). The aim of this study is to determine whether simultaneous collection of brain and cSC functional connectivity (FC) altered in people with CSM.

Methods: Joint brain/cSC functional (rs-fMRI) and structural imaging (T1w) data were acquired from N=19 (34.68 yrs old±1.83 (SD)) controls and N=11 (62.45 yrs old±5.68 (SD)) patients with CSM using a Siemens 3T Prisma scanner. Imaging data were processed using a custom pre-processing pipeline implemented in Python and MATLAB. Functional data were registered to the brain and spine template after default preprocessing. The mean time series for 166 brain regions from the AAL3 atlas and 32 regions from the PAM50 atlas (corresponding to spinal levels C1-8, left/right, dorsal/ventral horns) were extracted from the processed brain/spinal cord rs-fMRI data, and a Pearson’s correlation matrix was generated for all pairs of regions, followed by Fisher’s r-to-z transformation. Average ROI-to-ROI z-scores were computed for the control/CSM groups and compared using analysis of covariance (ANCOVA) with age and sex as covariates. A group effect threshold of p < 0.005 was used to limit ROI-ROI pairs for further analysis. Estimated marginal means for each ROI-ROI comparison removing the effects of the covariates were compared using pairwise t-tests. Processing pipeline has been shown in figure 1.

Results: Patients with CSM demonstrated altered FC between the cSC and multiple cortical–subcortical networks (Figure 2). At the spinal level, reduced coupling was observed between the left and right C6 ventral horns, alongside disrupted connectivity between the C4–C5 ventral horns and superior parietal regions, indicating impaired integration within the cervical sensorimotor network. CSM patients also showed aberrant connectivity between the C5 dorsal horn and medial temporal regions (parahippocampal gyrus) as well as inferior temporal cortex, reflecting abnormal spino-limbic and visuo-motor interactions. Beyond the spinal cord, differences extended to associative subcortical loops, with altered connectivity between the right anteroventral thalamus and left caudate, and between the thalamic reuniens nucleus and cerebellar vermis lobule X. Additional abnormalities were found between the parahippocampal gyrus and vermis lobule X, and between right superior parietal and temporal pole regions.

Conclusion: These findings suggest that CSM is associated with widespread disruptions across cervical sensorimotor, visuo-motor, and medial temporal–diencephalic networks, highlighting abnormal integration of sensory, motor, and cognitive systems in the context of spinal cord pathology.