Background: Previous studies exploring the anatomical correlates of pain in multiple sclerosis (MS) have predominantly relied on structural MRI and descriptive methodologies. While neuroimaging provides a valuable approach to investigating pain-related brain changes in MS, research in this field remains constrained by methodological variability, small sample sizes, and inconsistencies in imaging protocols.
Objective: This study aims to establish radiological correlates of neuropathic pain in MS patients through the objective segmentation and analysis of brain MRI. By employing advanced imaging techniques, this research seeks to contribute to a deeper understanding of the neuroanatomical basis of neuropathic pain in MS.
Methods: The study included three distinct groups: MS patients with neuropathic pain (MS-pain, n=8), MS patients without pain (MS no-pain, n=11), and individuals with pain-related small fiber neuropathy (SFN, n=6). Neuropathic pain was confirmed using laser-evoked potentials (LEPs), ensuring an objective assessment of pain response. All participants underwent brain MRI, with MS patients additionally undergoing spinal MRI. Brain region segmentation was conducted using two advanced automated tools: SAMSEG (Sequence Adaptive Multimodal SEGmentation) and SynthSEG. SAMSEG leverages multimodal MRI data for precise, probabilistic segmentation, while SynthSEG applies deep learning to achieve robust segmentation across diverse imaging datasets. Pain-related brain regions, including the thalamus, brainstem, basal ganglia, prefrontal cortex, and somatosensory cortex, were analyzed based on prior evidence linking these areas to neuropathic pain mechanisms.
Results: This study found that the right pallidum volume is significantly smaller in MS patients with pain compared to those without pain when measured with SynthSeg. No other brain regions showed significant differences between groups. However, multiple brain areas exhibited volume differences between all MS patients compared to SFN. Quantitative spinal cord lesion analysis showed no significant differences between groups.
Conclusion: This study identifies the right pallidum as a potential anatomical correlate of neuropathic pain in MS. Further research is required to refine imaging techniques and identify precise neuroanatomical targets for effective therapeutic interventions in a larger cohort.