In vivo detection of connectivity between cortical and white matter lesions in early MS

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


Background: The relationship between cortical lesions (CLs) and white matter lesions (WMLs) in multiple sclerosis (MS) is poorly understood. Pathological studies support a topographical association between CLs and underlying subcortical WMLs and suggest CLs may play a role in both disease initiation and progression. We hypothesized that cortical MS lesions are physically connected to white matter MS lesions via axonal connections. Objective: To assess the presence of CL-WML connectivity utilizing novel magnetic resonance imaging (MRI) methodology. Methods: In all, 28 relapsing-remitting MS patients and 25 controls received 3 T MRI scans, including double inversion recovery (DIR) for CL detection coupled with diffusion tensor imaging (DTI). CL and WML maps were created, and DTI was used to calculate inter-lesional connectivity and volumetric connectivity indices. Results: All patients showed inter-lesional WML connectivity (median 76% of WMLs connected to another WML; interquartile range (IQR), 58%-88%). On average, 52% of detected CLs per patient were connected to at least one WML (IQR, 42%-71%). Volumetric connectivity analysis showed significantly elevated cortical lesion ratios in MS patients (median, 2.3; IQR, 1.6-3.3) compared to null MS and healthy control datasets (p < 0.001). Conclusion: These findings provide strong evidence of inter-lesional connectivity between CLs and WMLs, supporting our hypothesis of intrinsic CL-WML connectivity.

Original languageEnglish (US)
Pages (from-to)973-981
Number of pages9
JournalMultiple Sclerosis
Issue number7
StatePublished - Jun 1 2017


  • Magnetic resonance imaging
  • multiple sclerosis

ASJC Scopus subject areas

  • Neurology
  • Clinical Neurology


Dive into the research topics of 'In vivo detection of connectivity between cortical and white matter lesions in early MS'. Together they form a unique fingerprint.

Cite this