Imaging Poster Presentation

P0591 - In vivo cortical lesions detection as an early hallmark of multiple sclerosis. (ID 901)

Speakers
  • J. Pelletier
Authors
  • P. Durozard
  • A. Maarouf
  • A. Rico
  • C. Boutière
  • S. Demortière
  • S. Confort-Gouny
  • J. Stellmann
  • W. Zaaraoui
  • M. Guye
  • J. Ranjeva
  • B. Audoin
  • J. Pelletier
Presentation Number
P0591
Presentation Topic
Imaging

Abstract

Background

Ultra-high field MRI highlighted the crucial role of cortical injury in multiple sclerosis (MS). Cortical pathology is frequent in MS, and seems to be more pronounced in progressive stages of the disease. One aspect of cortical pathology is represented by focal cortical lesions (CL).

Pioneering studies demonstrated the high frequency of CL in MS, their independence from white matter injury and their clinical relevance, particularly for cognitive impairment. Nevertheless, very few is known about their presence from the onset of the disease, their dynamics of apparition, neither the differential impact of intracortical (IC) and leukocortical (LC) lesions in early MS.

Objectives

The present study aims to assess the prevalence, the topography and the clinical counterpart of cortical lesions in patients included in the first year of MS.

Methods

16 MS patients in the first year of their disease course and 12 matched controls were included. All subjects underwent a 7T brain MRI scan designed to maximize the accuracy of CL detection (sequences: MP2RAGE, FLAIR, FLAWS with 600μm3 isotropic resolution and T2*/QSM with 600μm2 in-plane resolution and thickness of 600μm, total acquisition time 58 min). EDSS and MSFC were performed by rated physicians.

Radiological analysis: a cortical lesion was considered in case of signal change with identifiable boundaries compared to adjacent cortex on MP2RAGE, confirmed by at least one other sequence (FLAIR, FLAWS and T2*/QSM), excluding anatomical structure (eg vessels). CL were divided into two groups according to their location: leukocortical when the lesion extended across both white matter (WM) and grey matter (GM) and intracortical when the lesion is exclusively located within the GM. WM lesions were depicted on MP2RAGE and FLAIR sequences with a 3 mm minimal size.

Statistical analysis: patients and lesions descriptive data were presented with means and standard deviation (SD). Spearman correlations were performed between cortical and WM lesions count and clinical evaluations.

Results

Patients' caracteristics: 13 females, 3 males; mean age = 33 yo (SD = 9); mean disease duration = 6 months (SD = 3); mean EDSS = 0.28 (SD=0.51); mean relapse = 1.68 (SD = 1.01).

399 cortical lesions were detected in 14 patients (mean = 11.68, SD = 12.32). Among them 211 LC lesions were seen in 11 patients and 187 IC lesions in 14 patients. Mean number of WM lesions was 27 (SD = 27.45), mean volume was 2.15 mL (SD = 2.75). LC lesions were significantly correlated with WM lesions (ρ=0.91, p<.001). In opposite, IC lesions were not correlated with WM lesions (ρ=0.49, p=0.05) nor LC lesions (ρ=042, p=0.09). No lesions were seen in controls.

No correlations were found between cortical lesions and EDSS nor MSFC.

Conclusions

We evidenced that the prevalence of cortical lesions is very high at the earliest stage of MS and is not correlated with white matter impairment. This study confirms the accuracy of MP2RAGE to depicit these lesions.

Collapse