Leptomeningeal Enhancement on 3D-FLAIR MRI in Multiple Sclerosis: Systematic Observations in Clinical Practice

• Published in: Journal of neuroimaging Vol. 30; no. 6; p. 917
• Main Authors: Titelbaum, David S, Engisch, Renate, Schwartz, Eric D, Napoli, Salvatore Q, Sloane, Jacob A, Samaan, Soleil, Katz, Joshua D, Lathi, Ellen S
• Format: Journal Article
• Language: English
• Published: United States 01.11.2020
• Subjects: B-lymphocytes; 3D-FLAIR; leptomeningeal; inflammation; multiple sclerosis
• ISSN: 1552-6569
• DOI: 10.1111/jon.12774

Meningeal inflammation is implicated in cortical demyelination and disability progression in multiple sclerosis (MS). Gadolinium (Gd)-enhanced 3-dimensional (3D) FLAIR (fluid-attenuated inversion recovery) magnetic resonance imaging (MRI) can identify leptomeningeal enhancement (LME) in MS. Further characterization is needed to determine if LME is an imaging biomarker for meningeal inflammation. We sought to characterize the natural history of LME in the community setting, including persistence/resolution, effect of disease-modifying therapy, scanner variability, timing of acquisition, and imaging pitfalls that may lead to misinterpretation. A total of 341 MRI exams with Gd-enhanced 3D-FLAIR were reviewed in MS and non-MS patients to determine frequency of enhancement by MS subtype and association with therapy. A phantom was used to assess scanner variability. Two MS patients with seven LME were imaged at four postinjection time points to generate time-intensity curves. Imaging pitfalls were compiled. A total of 16.6% (40/241) of MS patients revealed LME compared to 8% (8/100) in non-MS patients (P = .04). There was no association with MS subtype, therapy, or disease activity. Detection using General Electric's version of 3D-FLAIR (29%) was greater than with Siemen's 3D-FLAIR (12%) at 1.5T (Tesla) (P < .001). Lesions were generally stable but resolved in 2 patients following high-dose steroids. LME kinetics were heterogeneous, even within patients, without uniform optimal time for acquisition. Enhancement curves exhibited three different variations, similar to the two-compartment model. Imaging pitfalls included enhancements of uncertain biologic significance, cortical veins and anatomic structures, and imaging artifacts. Awareness of LME characteristics, variability with imaging parameters, and imaging pitfalls will facilitate determining the potential role as an imaging biomarker for meningeal inflammation.