Rim lesions in MS at 3T: clinical correlation and possible radiological alternatives for daily practice at lower field strength

Eva Vanheule; Melissa Cambron; Jan W Casselman; Adrienn Dobai

doi:10.1016/j.neurad.2023.10.010

Rim lesions in MS at 3T: clinical correlation and possible radiological alternatives for daily practice at lower field strength

J Neuroradiol. 2023 Oct 30:S0150-9861(23)00262-6. doi: 10.1016/j.neurad.2023.10.010. Online ahead of print.

Authors

Eva Vanheule¹, Melissa Cambron², Jan W Casselman³, Adrienn Dobai⁴

Affiliations

¹ Department of Radiology, AZ Sint-Jan Brugge-Oostende, Campus Brugge, Ruddershove 10, Bruges 8000, Belgium. Electronic address: evavanheule@gmail.com.
² Department of Neurology, AZ Sint-Jan Brugge-Oostende, Campus Brugge, Ruddershove 10, Bruges 8000, Belgium.
³ Department of Radiology, AZ Sint-Jan Brugge-Oostende, Campus Brugge, Ruddershove 10, Bruges 8000, Belgium; Department of Diagnostic Radiology, Dalhousie University, Halifax, Nova Scotia, Canada.
⁴ Department of Oral Diagnostics, Faculty of Dentistry, Semmelweis University, Szentkirályi u. 47, Budapest, 1088, Hungary; Department of Neuroradiology, Medical Imaging Centre, Semmelweis University, Balassa street 6, Budapest, 1083, Hungary.

PMID: 37907156
DOI: 10.1016/j.neurad.2023.10.010

Abstract

Background and purpose: Paramagnetic rim lesions (PRLs) have been described as an imaging feature specific to multiple sclerosis (MS) using high-field strength phase-sensitive MR imaging. These lesions are suggested to reflect chronic active inflammation associated with greater disease severity and a more rapid disability progression. The aim of our study is to investigate the relationship between PRLs, clinical parameters, other radiological findings and disease progression.

Material and methods: This cross-sectional study included MS patients treated with teriflunomide, fingolimod, natalizumab or ocrelizumab for at least 2 years. PRLs seen at 3T MRI were analysed and correlated with clinical data and radiological progression, defined as an increase of the T2/FLAIR-lesion load during therapy. In the search for alternatives for these PRLs, we defined two additional radiological markers: 'FLAIR-bullet lesions', and on post-contrast black-blood (BB) images, 'BB-bullet lesions'.

Results: We included 84 MS patients of whom 27 (32 %) had at least 1 PRL. PRLs were associated with radiological progression under therapy (p=0.039) and higher clinical disability scores, although only significant for 9-Hole Peg Test (p=0.023). Patients with FLAIR-bullet or BB-bullet lesions at 3T MRI had a higher chance of PRL (p<0.001) with a likelihood ratio of 13.2 for FLAIR-bullets and 12.6 for BB-bullet lesions, thanks to the high negative predictive value of respectively 83 % and 90 %.

Conclusion: PRLs are associated with an increase of T2/FLAIR-lesion load under therapy and unfavourable clinical outcome. Our newly defined 'bullet lesions' are associated with PRLs and might be an interesting MRI marker for centres without access to high-field SWI images.

Keywords: MRI; Multiple sclerosis; PRL; SWI; Slowly expanding lesions (SEL).