Sage Journals: Discover world-class research

Abstract

Background

Leptomeningeal enhancement (LME) is a putative magnetic resonance imaging (MRI) marker of meningeal inflammation in multiple sclerosis (MS). 3D inversion-prepared fast-spin-echo sequence with real-reconstruction inversion recovery (Real-IR) MRI at 3 tesla (T) is highly sensitive to LME.

Objectives

To assess LME prevalence across brain regions and characterize the relationship between LME and subtle cortical pathology.

Methods

LME distribution patterns across brain regions were recorded for 90 adults (Age: 51 ± 11; women: 57; MS: 78) using 3T Real-IR. A subset of 15 participants had corresponding T₁-maps at 7T. T₁ relaxation times were calculated in the normal-appearing cortex subjacent to the LME, in comparison to the adjacent and homologue cortex.

Results

243 LME foci were found across 65 participants (73%). One hundred and sixty-one (66.3%) LME foci were posterior to the central sulcus. Three of 15 7T participants had a cortical lesion nearby LME (3/49 foci). Mean T₁ times within cortex beneath LME (1734 ± 135 ms) were elevated compared to homologue (1668 ± 167 ms, p = .0052) and adjacent cortex (1651 ± 133 ms, p < .0001).

Conclusions

Regional variations in LME distribution may point to topographical differences in the blood-meningeal barrier. Alterations in T₁ relaxation time observed in the cortex adjacent to LME may signify subtle tissue changes in the absence of cortical lesions.

Keywords

Multiple sclerosis MRI leptomeningeal enhancement blood-cerebrospinal fluid barrier real inversion-recovery MRI

Introduction

In multiple sclerosis (MS), inflammation occurs beyond the characteristic white matter (WM) lesions, including in the meninges as well as normal-appearing WM and gray matter (GM).^1,2 Histopathological analyses of brain biopsies and autopsies have demonstrated diffuse³ and focal^4,5 meningeal inflammation, even in early MS. These studies have suggested a possible topographical relationship between inflammatory cell accumulations in the meninges and subpial cortical demyelination.³ Meningeal inflammation has also been associated with accelerated disease progression,⁵ inflammatory activity within chronic active white matter lesions,⁶ and cortical pathology, including demyelination, neuronal loss, and glial reactivity.^3,5,7

A magnetic resonance imaging (MRI) marker of meningeal inflammation would prove useful in better understanding meningeal involvement in MS in vivo. Focal extravasation and collection of gadolinium-based contrast agent (GBCA) in the subarachnoid space visible on MRI is referred to as leptomeningeal enhancement (LME) and may represent a compromised cerebrospinal fluid (CSF) barrier.⁸ LME may also serve as a potential indicator of ongoing or previous focal meningeal inflammation, as histopathological analysis of corresponding meningeal tissue from three LME foci demonstrated meningeal inflammation characterized by T-cells, B-cells, and macrophages, as well as fibrosis.^9–11

In MS, LME has been linked to aging, disease duration, higher Expanded Disability Status Scale (EDSS) scores,¹⁰ and paramagnetic rim lesions, an MRI correlate of chronic active white matter lesions.¹² LME is not specific to MS and can be seen in healthy aging and other neuroinflammatory conditions, such as human T-lymphotropic virus type 1 (HTLV-1)-Associated Myelopathy/Tropical Spastic Paraparesis (HAM-TSP).^9,13,14 LME foci generally persist for years¹⁰ and are conventionally detected on 3T T₂-FLAIR imaging.⁹ A recently introduced imaging technique, 3D inversion-prepared fast-spin-echo (FSE) sequence with real-reconstruction inversion recovery (Real-IR) MRI, is a more sensitive method, with a ∼4-fold higher LME detection rate due to a higher contrast-to-noise ratio, and has a high inter-rater agreement.¹²

Despite histopathological evidence of associations between meningeal inflammation and cortical demyelination in MS, studies assessing the relationship between LME and cortical lesions on MRI have conflicting results.^8,15 Cortical lesions in one MS cohort were not spatially associated with LME foci and were more prevalent in the parietal and frontal lobes.¹⁶ Beyond cortical lesions, local quantitative T₁ and T₂ alterations may also reflect cortical pathology.¹⁷ Such changes can indicate differences in tissue water content, cell density, iron concentration, and myelination¹⁸ that may arise from fibrosis, collagen deposition, or subtle demyelination. One prior study also demonstrated focal cortical thinning around LME foci.¹⁹ LME distribution in MS has been explored in two prior studies. At 7T, LME detected with T₂-FLAIR was reported to be most prevalent in the frontal lobe (52%), followed by the parietal (31%), temporal (13%), and occipital (4%) lobes.²⁰ Another study found that 10% of LME foci identified with 3T T₂-FLAIR were frontal, 48% were parietal, 22% were temporal, and 21% were occipital.²¹

In this study, we analyzed spatial distribution patterns of LME with 3T Real-IR. We further investigated the topographical relationship between LME and possible subtle cortical changes reflected by T₁ relaxation time alterations at 7T in normal-appearing cortex underneath LME as compared to adjacent and corresponding contralateral normal-appearing cortex. T₁ and T₂* MRI without GBCA injection at 7T were used to create regions of interest in normal-appearing cortex near LME foci, as it provides higher sensitivity to tissue changes and higher spatial resolution relative to 3T.¹⁷

Methods

Participants

Demographic, clinical, and imaging data were acquired from participants enrolled under the institutional review board-approved National Institute of Neurological Disorders and Stroke protocols, “Evaluation of Progression in Multiple Sclerosis by Magnetic Resonance Imaging” (NCT00001248) and “Immuno-Virological Evaluation of Human T Cell Lymphotropic Virus Infection and Associated Neurological Diseases” (NCT00001778). After clinical and radiological evaluations, participants were categorized as relapsing-remitting MS (RRMS),²² primary-progressive MS (PPMS),²² secondary progressive MS (SPMS),²² clinically isolated syndrome (CIS),²² radiologically isolated syndrome (RIS),²³ HAM-TSP,²⁴ HTLV-1 asymptomatic carriers,²⁴ or other inflammatory neurological disease (OIND). A relapsing MS (RMS) cohort was comprised of individuals categorized as CIS or RRMS. Disease modifying treatment (DMT) status (on vs off or naïve), EDSS, Symbol Digit Modalities Test (SDMT),²⁵ and Paced Auditory Serial Addition Test (PASAT)²⁶ scores were recorded when available. Imaging data was acquired more than 30 days after any prior steroid administration. Written informed consent was obtained from participants. Inclusion criteria for the present analyses are summarized in Figure 1.

Figure 1.

Flowchart of inclusion criteria for study analyses. Abbreviations: GBCA: gadolinium-based contrast agent; Real-IR: 3D real-reconstruction inversion recovery; LME: leptomeningeal enhancement; T₂-FLAIR: T₂-weighted fluid-attenuated inversion recovery; T₁-MP2RAGE: T₁ magnetization prepared 2 rapid gradient echo; C-DEF: Classification using DErivatives-based Features.

MRI acquisition

3T Real-IR (TR/TE/TI = 15,320/549/2700 ms, flip angle = 145°, resolution: 0.5 × 0.5 × 1.0 mm), T₁-weighted magnetization prepared 2 rapid gradient echo (MP2RAGE, TR/TE/TI₁/TI₂ = 5000/1.9/700/2500 ms, flip angle = 0°, resolution: 1.0 mm isotropic), and T₂-FLAIR (TR/TE/TI = 4800/352/1800 ms, flip angle = 120°, resolution: 1.0 mm isotropic) were performed on a Siemens Skyra scanner with a 20- or 32-channel head coil. A standard dose of GBCA (gadobutrol, 0.1 mmol/kg) was administered during all 3T scans, and post-GBCA images were acquired approximately 15 min after contrast administration, ensuring adequate delay.^10,12 Real-IR subtraction images were generated by first applying N4 bias correction²⁷ using ANTs²⁸ to pre- and post-GBCA images, skull-stripping using ROBEX,²⁹ median-normalizing using AFNI,³⁰ and rigidly coregistering the post-GBCA image to the pre-GBCA using ANTs to generate a transformation matrix. This transformation matrix was then applied to the non-skull-stripped images, and the final subtraction was completed with AFNI.

MS cases with available unenhanced 0.5-mm isotropic MP2RAGE acquired three or four times per scanning session (TR/TE/TI₁/TI₂ = 6000/5.14/1000/2900 ms, flip angle = 4/5°) and either 0.5-mm isotropic T₂*-weighted (T₂*w) gradient echo (GRE) scans³¹ (3‒4 motion and B₀-corrected slabs; TR/TE/TI = 72.2/12.26/15 ms, flip angle = 12°) or 0.7-mm isotropic T₂*w segmented echo planar imaging (EPI) (whole brain coverage; TR/TE = 38/23 ms, flip angle = 10°), acquired on a 7T Siemens Magnetom scanner within 18 months of 3T scan, were identified. This time interval was considered acceptable, as both LME¹⁰ and cortical lesions³² are almost always stable over time. T₁-maps were generated on the scanner, through vendor-supplied software, from the MP2RAGE acquisitions. Voxel-wise T₁-map medians were generated after coregistering T₁-map MP2RAGE repetitions, using FSL (https://fsl.fmrib.ox.ac.uk/) for voxel computation (fslmaths) and AFNI for rescaling (3dcalc). Sequence parameters for 3T and 7T scans are summarized in Table S1 in the supplementary materials.

LME assessment

For each scan, pre-GBCA images were rigidly registered to Montreal Neurological Institute (MNI) space (voxel size: 1.0 mm isotopic) using ITK-SNAP (http://www.itksnap.org/, version 4.2.0). Post-GBCA and subtraction Real-IR images were separately registered to the registered pre-GBCA images using the same method. Due to failed registration, subtraction images were not available in 14 cases. For these cases, only the pre- and post-GBCA Real-IR images were used for LME assessment. In remaining cases, subtraction images were used as an adjunct to pre- and post-contrast imaging to improve LME identification by improving LME-to-background contrast. However, as some images were affected by registration inaccuracies due to relative lack of overall tissue contrast in Real-IR images, pre- and post-GBCA imaging were the primary tool for LME identification. LME foci were defined as subarachnoid-space signal intensity visibly higher than the adjacent brain parenchyma, identified on two or more planes and seen in post-GBCA but not pre-GBCA images and in the subtraction images.¹² LME foci were identified and segmented by a stratified consensus approach by two raters: SVO (a neurologist with 6 years of experience in MS neuroimaging, trained by DSR, a neuroradiologist with 17 years of experience) and AAT (neuroimaging researcher trained by SVO and DSR). AAT identified and segmented LME using ITK-SNAP across the cohort first, then reviewed all masks with SVO, discussing each focus. Nodular LME (spherical foci) and spread/fill LME (diffusely spread along the sulci),¹⁴ demonstrated in Figure 2, were identified. Aligning with prior studies,^9,14 dural and paravascular enhancements were excluded from analysis due to the study's focus on understanding contrast agent extravasation to the subarachnoid space at the pial surface of the leptomeninges. 3T T₁-MP2RAGE images were used for anatomical guidance. After consensus LME identification, LME foci were masked manually using ITK-SNAP on the linearly registered post-GBCA Real-IR scan. Regions corresponding to LME foci in MNI space were also masked. This was done manually to ensure appropriate localization and size of the foci.

Figure 2.

Representative nodular, spread/fill, and infratentorial LME foci. (a) Representative image of a nodular LME focus visible in multiple planes on the subtraction image and post-GBCA Real-IR, but not pre-GBCA Real-IR, in a participant with relapsing-remitting MS in their late 50s. (b) Representative image of a nodular LME focus located in the cerebellar folia in a participant in their early 40s with secondary progressive MS, highlighting the sensitivity of Real-IR in infratentorial regions. (c) Representative image of a spread/fill LME focus in another individual with relapsing-remitting MS in their late 40s.

Spatial analysis of LME foci

A 3D map of LME location was generated in MIPAV (version 11.2)³³ by combining LME masks that were created in the MNI space. Cortical regions directly subjacent to each LME focus were annotated by overlaying the Yale Brain Atlas³⁴ on the 3T MNI space. The atlas has 17 cortical regions, and each region is associated with the frontal, parietal, temporal, or occipital lobe or the insula. LME identified in regions below the tentorium cerebelli was classified as “infratentorial LME.” LME foci were then classified as anterior or posterior, taking the central sulcus as an anatomical reference.

Brain segmentation and white matter lesion analysis

The Classification using DErivatives-based Features (C-DEF) machine learning algorithm^35,36 was applied to 3T T₂-FLAIR and 3T T₁-MP2RAGE images to obtain tissue and lesion segmentation. Brain masks were created with SynthStrip.³⁷ Lesion segmentations were manually confirmed and adjusted as needed. CSF, GM, normal-appearing white matter (NAWM), and white matter lesion (WML) volumes, as well as fractional (F: normalized to the total intracranial volume) GM (GM-F) and WML (WML-F) volumes, were calculated.

Cortical T₁ alteration analysis

In 7T scans from cases with at least one supratentorial LME, T₁ times in four normal-appearing gray matter regions of interest (ROI) were calculated from 7T T₁-map medians. The 7T scans were registered and resliced to the 3T images through linear, rigid registration in ITK-SNAP. The four ROI were masked semi-automatically in 3D Slicer (https://www.slicer.org/, version 5.6.2). Cortex directly subjacent to LME foci was termed the “central” ROI. These ROI were generated by overlaying the LME masks created on 3T images on registered 7T T₁-MP2RAGE uniform denoised images. Using voxel-based intensity thresholding to exclude CSF and WM on the T₁-MP2RAGE uniform denoised images, any region of the LME mask overlying the cortex was included in the “central” ROI. The threshold was determined visually and uniquely for each participant to ensure the exclusion of CSF areas for measurements. The “central” mask was expanded by a radius of 3 mm in all directions, then similarly thresholded by intensity to only include cortex, to create the “adjacent” ROI. These regions (“central” and “adjacent”) were compared to size-matched ROIs from contralateral, homologous cortex to adjust for the possible confounding effect of variation in T₁ across different regions of the brain.³⁸ The contralateral region to the “central” ROI was termed the “cont_central” ROI, while “cont_adjacent” was contralateral to the “adjacent” ROI. The contralateral ROI were created following the same procedure as the “central” and “adjacent” ROI, after the LME masks created on 3T images were flipped across the midline.

If any of the four ROI exhibited a cortical lesion, the corresponding LME was excluded from T₁ measurement. Cortical lesions were identified by SVO as hypointense cortical regions in at least two axial slices on 7T T₁-MP2RAGE uniform denoised images that were simultaneously hyperintense on corresponding motion-corrected T₂*w GRE images (6/15 scans) or non-motion-corrected T₂*w EPI images (9/15 scans).¹⁶

The automatically generated masks for each region were confirmed and refined manually using both 3D Slicer and ITK-SNAP to ensure no inclusion of CSF or WM. The final masks were compared to confirm size-matching, then overlaid on the 7T T₁-map median scans. Volumes and T₁ times in each ROI were calculated. Figure S1 in the supplementary materials demonstrates the process of cortical ROI creation.

Statistical analysis

Correlation analyses were performed using the Spearman method. Wilcoxon signed-rank (paired) and Mann‒Whitney U (unpaired) comparison tests were performed for categorical group analyses. LME distribution across cortical regions was analyzed with a chi-squared test. LME prevalence among categorical groups, calculated as the percent of the group with LME, was compared with a Fisher's exact test. LME prevalence accounting for volume was calculated as the number of LME adjacent to the cortical brain region divided by the volume of those regions in the Yale Brain Atlas.

Associations between EDSS and LME were assessed with multivariable linear regression analyses, adjusting for age. This approach was applied to assess associations between LME count and PASAT, SDMT, and brain volumetrics. Log-normal variables, as assessed by the D'Agostino & Pearson test, were transformed [Y = log(Y)] prior to use in the multivariable linear regression models.

In the cohort with 3T and 7T imaging within 18 months, mean T₁ relaxation times across the four regions of interest (“central,” “adjacent,” “cont_central,” and “cont_adjacent”) were compared collectively at the lesion-level for each LME focus with a repeated-measures ANOVA followed by a Tukey test for pairwise calculations adjusted for multiple comparisons. A sensitivity analysis with median T₁ relaxation times was run with the same approach to assess whether skewness in ROI T1 distribution influenced findings. Additionally, for a subcohort comprised of cases with 3T and 7T imaging within 3 months, mean T₁ relaxation times across the four regions of interest were compared with the same method to assess whether time between scans influenced findings. Within the cohort with 3T and 7T imaging within 18 months, model assumptions were tested by also running a linear mixed model for the mean T₁ relaxation times with the ROI as fixed effect and subject and LME focus as nested random effects with Bonferroni correction to assess T₁ time differences across four cortical ROI while accounting for psuedoreplication bias from multiple LME foci per subject.

Data availability

Deidentified data used for statistical analysis are available in the Data Supplement.

Results

Demographic and clinical characteristics

3T scans were evaluated from 90 adults (57 women). Seventy-eight individuals had MS, 5 had HAM-TSP, 1 was an asymptomatic carrier of HTLV-1, and six were classified as OIND. Age (mean ± SD) was 51 ± 11 years. EDSS (median) in the MS subgroup was 2.0 (range: 0–8.5). Table 1 summarizes study cohort characteristics.

Table 1.

Clinical and demographics cohort characteristics.

3T Cohort	Number of participants (%)	Age (mean ± SD)	EDSS (median; range)	Number of participants (%) with at least 1 LME	Number of LME (% of total LME)
Total	90 (100%) (33 M, 57 W)	51 ± 11	2; (0, 8.5)	65 (72%)	243 (100%)
Diagnosis
RIS	2 (2%)	45 ± 15	–	2 (100%)	4 (2%)
RMS	54 (60%)	48 ± 11	1.5; (0, 6)	39 (72%)	126 (52%)
CIS	3 (3%)	54 ± 7	1; (1, 2)	1 (33%)	4 (38%)
RRMS	51 (57%)	48 ± 11	1.5; (0, 6)	38 (75%)	122 (50%)
PMS	22 (24%)	55 ± 11	6, (1.5, 8.5)	15 (68%)	79 (33%)
PPMS	8 (9%)	52 ± 10	6.5; (2.5, 8.5)	5 (62%)	29 (12%)
SPMS	14 (16%)	57 ± 12	5; (1.5, 7)	10 (71%)	50 (21%)
HTLV-1 infection	6 (7%)	60 ± 8	–	5 (83%)	16 (7%)
OIND	6 (7%)	51 ± 11	–	4 (67%)	18 (7%)

7T MS Cohort	Number of participants (%)	Age (mean ± SD)	EDSS (median; range)	Time (days) between 3T and 7T scans (median; range)	Number of LME
Total	15 (17%)	52 ± 13	1.5; (0, 7)	35; (1, 376)	52

Abbreviations: RMS: relapsing multiple sclerosis; RRMS: relapsing-remitting multiple sclerosis; PMS: progressive multiple sclerosis; PPMS: primary progressive multiple sclerosis; SPMS: secondary progressive multiple sclerosis; CIS: clinically isolated syndrome; RIS: radiologically isolated syndrome; HTLV-1: human T-lymphotropic virus type 1; OIND: other inflammatory neurological diseases (e.g., undiagnosed demyelinating and inflammatory diseases, transverse myelitis).

Sixty-five individuals had at least one LME focus (LME prevalence: 72%, median per individual: 3, interquartile range (IQR): 1–5). The cohort had 243 LME foci in total. LME count per scan was correlated with age (ρ=.34, p = .001). In individuals with MS, EDSS was not associated with the number of LME after adjusting for age (p = .17). In a subcohort of patients who completed the PASAT (n = 18) and SDMT (n = 22), number of LME was not associated with either score after age adjustment (p = .14, p = .23, respectively). Individuals naïve to or off DMT did not have a difference in LME as compared to participants on DMT (p = .10). LME prevalence was not significantly different between PMS and RMS cases (p = .78); however, PMS cases had increased LME count as compared to RMS cases (p = .013). After age adjustments, LME count per scan in the total cohort was not associated with GM-F. However, LME count was associated with log-normalized WML-F after adjusting for age (r = .35, p = .0007).

LME foci are more prevalent in posterior brain regions

LME foci were distributed across 17 cortical Yale Brain Atlas regions, the insula, and the cerebellum. The top four subjacent cortical regions accounted for 56% of the 243 LME foci: occipital (17%), superior parietal (15%), supramarginal (13%), and middle frontal (11%) cortical regions. By lobe, 94 (39%) LME were localized to the parietal lobe cortex, 82 (34%) to the frontal lobe, 40 (10%) to the occipital lobe, and 25 (17%) to the temporal lobe. Additionally, 18 (7.4%) LME foci were infratentorial, all of which were within the cerebellar folia. Detailed LME distribution by region is provided in Table S2. The regional distribution of LME foci did not differ by diagnosis (MS vs non-MS, p = .96), clinical phenotype (RMS vs PMS, p = .46), sex (p = .99), disability (in MS cohort) (EDSS <5 vs. EDSS ≥5, p = .73), or hemisphere (p = .32).

Across the cohort, 79 (33%) LME foci were anterior to, 161 (66%) posterior to, and 3 (1.2%) along the central sulcus. The volume of cortex anterior to the central sulcus in the MNI space was 303 ml, while the volume of cortex posterior to the central sulcus was 371 ml. LME prevalence accounting for volume in cortex anterior to the central sulcus was 0.26. In the cortex posterior to the central sulcus, it was 0.42.

Since LME prevalence is correlated with age, anteriority and posteriority of LME were evaluated in age cohorts above and below 50 (cohort mean age: 51). Among individuals under 50, 24 of 42 (57%) had at least one LME. In these participants, 19 (31%) LME foci were anterior, and 43 (69%) posterior, to the central sulcus. In individuals 50 and older, 41 of 48 (85%) had at least one LME; 60 (34%) LME foci were anterior and 118 (66%) posterior to the central sulcus. Figure 3 demonstrates the regional predilections of LME.

Figure 3.

Spatial distribution of LME across brain regions. (a) 3-plane visualization of LME foci distribution (see Video 1 for 3D visualization). (b) LME prevalence was highest in posterior brain areas, including the occipital, superior parietal, and supramarginal regions. (c) Representative nodular LME focus in the posterior brain in the top left image, anatomically localized on the registered 3T T₁-MP2RAGE scan (top right) and MNI space (bottom left). Yale Brain Atlas cortical labels were used to record brain region in the MNI space, seen in the bottom left image.

Normal appearing cortex subjacent to LME foci exhibit higher T₁ relaxation times

Fifteen individuals [RRMS: 12, SPMS: 2, OIND (transverse myelitis): 1; women: 9; men: 6] included in the 3T analyses had a 7T scan conducted within 18 months (median, IQR, range in days: 35, 1–287, 1–376). Across these scans, there were 52 LME foci. Two infratentorial LME were excluded from T₁ measurements due to incomplete infratentorial coverage of the 7T T₁ images. One LME was excluded from analysis due to artifact and noise in the corresponding region of the 7T scan. Four LME foci were excluded from T₁-mapping analyses due to the presence of a cortical lesion in one of the four regions of interest.

Of the excluded LME foci, three were due to a cortical lesion within 3 mm of the LME, and one was due to a cortical lesion in the cortex contralateral to the LME. Thus, 6.1% of LME (3/49 foci; 3/15 individuals) evaluated had a cortical lesion nearby. T₁ relaxation times were investigated in four cortical ROI for each of the 45 remaining LME foci.

T₁ times across these regions were significantly different (p < .0001, repeated measures ANOVA). T₁ times in the “central” ROI (1734 ± 135 ms) cortex directly subjacent to LME were higher than T₁ times in the “adjacent” (1651 ± 130 ms, p < .0001) and “cont_central” (1668 ± 167 ms, p = .0052) ROI. T₁ times in the “adjacent” ROI and “cont_adjacent” ROI (1633 ± 133 ms) were not different (p = .39). T₁ times in “cont_central” were not different than “cont_adjacent” (p = .13) nor “adjacent” ROI (p = .77). Figure 4 shows cortical T₁ alterations found in the cortex subjacent to LME compared to adjacent and homologous cortex. This analysis and its findings were replicated with a sensitivity analysis using median T₁ relaxation times and within a subcohort of 10 cases with a 3-month-or-less difference between 3T and 7T imaging. Details of these findings are reported in the Supplement. A linear mixed model accounting for subject (n = 15) and LME (n = 45) as nested random effects demonstrated a similar pattern. T₁ times were prolonged in the “central” ROI (1735 ± 34 ms) as compared to the “cont_central” ROI (1667 ± 34 ms, p < .0001) and “adjacent” ROI (1653 ± 34 ms, p < .0001). T₁ times in the “cont_central” ROI were not significantly different than those in “cont_adjacent” (1633 ± 34 ms, p = .16) or “adjacent” ROI (p > .99).

Figure 4.

T₁ times in LME subjacent cortical regions were increased compared to contralateral and adjacent cortical tissue. T₁ times were higher in the (a) “central” versus “adjacent” ROI and (b) “central” versus “cont_central” ROI.

Discussion

In this study, we characterized the topographical distribution of LME, a potential MRI indicator of meningeal inflammation and a compromised blood-CSF-barrier, showing a posterior predilection in the spatial distribution of LME, whereby approximately 2/3 of identified foci were found posterior to the central sulcus. As LME may reflect ongoing or prior inflammation and fibrosis,^9–11 and also appears to be a feature of normal aging,⁸ higher LME prevalence in the posterior brain may indicate increased susceptibility of the posterior meninges to blood-CSF-barrier alteration, reflecting possible regional variability of neurovascular unit integrity after injury across the brain. Of interest, predominant posterior brain involvement is also seen in some neurological diseases that affect the neurovascular unit, such as posterior reversible encephalopathy syndrome³⁹ and cerebral amyloid angiopathy.⁴⁰ However, as we found no association of LME anteriority/posteriority with age in our MS cases, its relevance remains to be further explored.

Animal studies suggest possible regional differences of embryonic origins across anterior (neural crest, ectodermal) and posterior (mesodermal) meninges.⁴¹ In humans, the meninges are thought to be mesenchymal in origin, with limited evidence suggesting they arise from a combination of neural crest (ectodermal tissue) and mesodermal cells,⁴² and, as such, the regional distribution of these distinct cell populations remains uncertain. As cell populations induced from different embryologic layers may display different molecular profiles and gene expression,⁴³ further investigation of the topography of human meninges may provide insight into differential vulnerability of the posterior meninges to vascular injury. Aside from alterations affecting the leptomeninges (pia mater and arachnoid mater) and the subarachnoid space, factors involving the dura (including the meningeal lymphatic system and dura–skull bone marrow interactions) could potentially underlie the observed posterior predilection due to direct connections between the dura and the SAS via arachnoid cuff exit points.⁴⁴ The relationship between leptomeninges and dura may be better understood by characterization of dural and paravascular enhancements, which we did not include in this study, as our focus was on the leptomeninges. Potential differences in CSF flow and dorsal meningeal clearance in the human brain may also partly explain our findings.⁴⁵ Additional characterization of CSF dynamics may help clarify these observations further.

Our findings of spatial distribution of LME align broadly with previously reported patterns of LME distribution, though there are some differences. While one study at 7T found LME to be most prevalent in the frontal lobe (52%), we found the highest prevalence of LME in the parietal lobe (39%), followed by the frontal lobe (34%).²⁰ Another study at 3T also demonstrated high LME prevalence in the parietal lobe (48%) but reported low LME prevalence in the frontal lobe (10%).²¹ Differences in LME distribution findings are most likely related to cohort and methodological differences across studies. Specifically, our study implemented Real-IR, as compared to T₂-FLAIR, which was used in the previously mentioned studies. Real-IR has been previously demonstrated to have a higher noise-to-contrast ratio than T₂-FLAIR,¹² but this increase may be to different extents in different brain regions.

Our findings also replicated previously reported relationships between LME count and age.^10,12 Prior literature has reported conflicting data regarding the relationship between LME with EDSS,^12,46 clinical phenotype (RRMS vs PMS),^10,12,20,46 WML volume,^12,46 and cortical volume,^8,12 highlighting complexities in relating LME to other clinical findings in MS. In this study, we did not find a relationship between EDSS and LME count. This is potentially due to cohort and technical differences and inter- and intra-rater variability of EDSS scoring.⁴⁷ We found no significant difference in LME prevalence in PMS as compared to RMS but did report increased LME count in the PMS group. The PMS cohort was older, which may explain this finding. RIS and CIS participants were included in the study to capture the early stages of MS, and LME foci were detected in these subgroups. However, our sample sizes for these subgroups were small, and thus there is uncertainty regarding LME prevalence in these groups, as well as in healthy volunteers, a group not included in our study. Our findings should be interpreted with discretion, and this may apply to other studies with similarly small sample sizes for these subgroups.⁴⁸ Further exploration into LME prevalence in these cohorts, as well as in non-MS cohorts such as other inflammatory disease and healthy controls, may prove informative. We also demonstrated an association, after age-adjustment, with between LME count and WML-F. However, we did not replicate these findings with LME and GM-F, after adjusting for age, potentially due to the strong correlation between GM volume and age in MS.⁴⁹ These unclear associations and findings further highlight the need for a more nuanced study of LME and its biological relevance.

We also described increased T₁ relaxation times in normal-appearing cortex underneath LME foci as compared to adjacent and contralateral cortex. T₁ relaxation time alterations in normal-appearing cortex can be caused by changes in water content. Increased T₁ relaxation could be caused by local edema, hypercellularity via gliosis, subtle demyelination, or neuroaxonal damage.^18,50 These findings may thus point to cortical abnormalities underneath LME that are not visible as cortical lesions but are nonetheless reflected quantitatively on MRI, aligning with histopathological observations of focal meningeal inflammation and adjacent subpial demyelination.⁵ A potential relationship between LME and cortical changes may highlight the utility of LME as an indirect indicator of abnormal cortical pathology in vivo even in the absence of cortical lesions, aligning with a prior study that demonstrated cortical thinning around LME foci.¹⁹ This relationship may be particularly relevant in better monitoring and understanding MS at multiple stages, including early MS, which demonstrates concomitant meningeal inflammation and cortical pathology,^3,51,52 and in later stages like PMS, where cortical damage may contribute to progression of the disease or neurological symptoms without corresponding WML development.^53,54 Furthermore, these LME-adjacent cortical alterations may align with a surface-in pattern of damage in MS.^55–57 Paired radiological and histopathological work describing LME, meningeal inflammation, and corresponding cortical changes may yield additional evidence to support these findings and provide a deeper understanding of the underlying biological processes associated with meningeal and cortical damage in MS.

One main limitation of our study was the small cortical ROI volumes that were susceptible to partial volume effects, especially due to the downsampling of the 7T images to match the 3T images. To account for these effects, a signal intensity thresholding method with visual assessments for each scan was implemented when creating cortical regions of interest to systematically exclude CSF and WM. Furthermore, contralateral and size-matched cortical regions were created with the same method for comparison. Finally, we replicated our results using median T₁ measures rather than mean measures to account for potential skewness in our data. Another limitation of this study was the comparatively smaller sample size (n = 15) of the 7T cohort on a scan level. Our 7T cohort was largely limited by requiring both 3T and 7T scans to occur within 18 months of one another. While most 7T scans were conducted within a few days to months of the 3T scans, 5 individuals had a longer scan difference, roughly one year. We believe that including these scans is still acceptable since LME and cortical lesions are usually persistent findings,^10,32 and our sensitivity analysis of a subcohort of cases with 3T and 7T imaging within three months aligned with our findings in the larger cohort. While LME were treated as individual replicates, we demonstrated that accounting for both subject and LME as random effects in a linear mixed model still demonstrated the same findings of elevated T₁ times in the “central” ROI. However, further confirmation of these preliminary, exploratory findings in a larger sample size of patients is important.

In conclusion, our findings suggest that LME in the setting of neuroinflammation exhibits a posterior predilection, which may point to a potential heterogeneity of blood-CSF barrier, and that cortical T₁ abnormalities adjacent to the LME regions can occur even in the absence of discrete cortical lesions. Deepening our understanding of LME and its utility as a marker of meningeal inflammation in MS may enhance precision in characterizing the disease at an individual level.

Supplemental Material

sj-docx-1-mso-10.1177_20552173251408625 - Supplemental material for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T₁ alterations in the adjacent cortex

Supplemental material, sj-docx-1-mso-10.1177_20552173251408625 for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T₁ alterations in the adjacent cortex by Ashley A. Thommana, Erin S. Beck, Matthew A. Greenwald, Gina Norato and Hallie Gaitsch, Dzung L. Pham, Steven Jacobson, María I. Gaitán, Govind Nair, Daniel S. Reich, Serhat V. Okar in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Supplemental Material

sj-doc-2-mso-10.1177_20552173251408625 - Supplemental material for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T₁ alterations in the adjacent cortex

Supplemental material, sj-doc-2-mso-10.1177_20552173251408625 for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T₁ alterations in the adjacent cortex by Ashley A. Thommana, Erin S. Beck, Matthew A. Greenwald, Gina Norato and Hallie Gaitsch, Dzung L. Pham, Steven Jacobson, María I. Gaitán, Govind Nair, Daniel S. Reich, Serhat V. Okar in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Supplemental Material

sj-xlsx-3-mso-10.1177_20552173251408625 - Supplemental material for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T₁ alterations in the adjacent cortex

Supplemental material, sj-xlsx-3-mso-10.1177_20552173251408625 for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T₁ alterations in the adjacent cortex by Ashley A. Thommana, Erin S. Beck, Matthew A. Greenwald, Gina Norato and Hallie Gaitsch, Dzung L. Pham, Steven Jacobson, María I. Gaitán, Govind Nair, Daniel S. Reich, Serhat V. Okar in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Footnotes

Glossary

Acknowledgments

A Cooperative Research and Development Agreement (CRADA) between NINDS and Siemens supported MRI methodology for this study. We thank Tobias Kober for his assistance in 7T T₁-mapping and Luis Santos for his contributions to the development of algorithms that are used in this study. We thank the Neuroimmunology Clinic and the National Institute of Mental Health fMRI Facility staff for their support of this study.

Author contributions

AAT contributed to the conception, design, data acquisition, data analysis, drafting, and revision of the manuscript. ESB contributed to the design, data acquisition, data analysis, and revision of the manuscript. MAG contributed to the data acquisition, data analysis, and revision of the manuscript. GN contributed to the data analysis and revision of the manuscript. HG contributed to the data analysis and revision of the manuscript. DLP contributed to data acquisition and revision of the manuscript. SJ contributed to the data acquisition and revision of the manuscript. MIG contributed to the data acquisition, data analysis and revision of the manuscript. GN contributed to the data acquisition, data analysis, and revision of the manuscript. DSR contributed to the conception, design, data acquisition, data analysis, drafting, and revision of the manuscript and secured funding. SVO contributed to the conception, design, data acquisition, data analysis, drafting, and revision of the manuscript. DSR and SVO supervised the study.

Consent for publication

Participants or their relatives provided consent for publication during the informed consent process within the study protocol.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported in part by the Intramural Research Program of the National Institutes of Health (NIH) (Z01 NS003119) and by the Cure Alzheimer’s Fund. The contributions of the NIH authors were made as part of their official duties as NIH federal employees, are in compliance with agency policy requirements, and are considered Works of the United States Government. However, the findings and conclusions presented in this paper are those of the authors and do not necessarily reflect the views of the NIH or the U.S. Department of Health and Human Services. ESB was supported by the National Multiple Sclerosis Society (Career Transition Award, TA-2109-38412). DLP was funded by CDMRP W81XWH2010912. SVO is supported by National Multiple Sclerosis Society (Postdoctoral Fellowship Grant, FG-2208-40289).

Disclosures

AAT has nothing to disclose. ESB has nothing to disclose. MAG has nothing to disclose. GN has nothing to disclose. HG has nothing to disclose. DLP has nothing to disclose. SJ has nothing to disclose. MIG has nothing to disclose. GN has nothing to disclose. SVO has nothing to disclose. DSR has received research funding from Abata and Sanofi, unrelated to the current study.

Ethics approval and consent to participate

The data included in this research was acquired following the approval of the Institutional Review Board and after obtaining written informed consent from the participants, as part of the National Institute of Neurological Disorders and Stroke's “Evaluation of Progression in Multiple Sclerosis by Magnetic Resonance Imaging” protocol (NCT00001248) and “Immuno-Virological Evaluation of Human T Cell Lymphotropic Virus Infection and Associated Neurological Diseases” (NCT00001778).

ORCID iDs

Ashley A. Thommana

Erin S. Beck

Matthew A. Greenwald

Gina Norato

Hallie Gaitsch

Dzung L. Pham

Steven Jacobson

María I. Gaitán

Govind Nair

Daniel S. Reich

Serhat V. Okar

Supplemental material

Supplemental material for this article is available online.

References

Lassmann

. Multiple sclerosis pathology. Cold Spring Harb Perspect Med 2018; 8: 20180301.

Kuhlmann

Moccia

Coetzee

, et al. Multiple sclerosis progression: time for a new mechanism-driven framework. Lancet Neurol 2023; 22: 78–88.

Lucchinetti

Popescu

Bunyan

, et al. Inflammatory cortical demyelination in early multiple sclerosis. N Engl J Med 2011; 365: 2188–2197.

Serafini

Rosicarelli

Magliozzi

, et al. Detection of ectopic B-cell follicles with germinal centers in the meninges of patients with secondary progressive multiple sclerosis. Brain Pathol 2004; 14: 164–174.

Magliozzi

Howell

Vora

, et al. Meningeal B-cell follicles in secondary progressive multiple sclerosis associate with early onset of disease and severe cortical pathology. Brain 2007; 130: 1089–1104.

Ahmed

Fransen

Touil

, et al. Accumulation of meningeal lymphocytes correlates with white matter lesion activity in progressive multiple sclerosis. JCI Insight 2022; 7: 20220308.

Magliozzi

Howell

Reeves

, et al. A gradient of neuronal loss and meningeal inflammation in multiple sclerosis. Ann Neurol 2010; 68: 477–493.

Okar

Fagiani

Absinta

, et al. Imaging of brain barrier inflammation and brain fluid drainage in human neurological diseases. Cell Mol Life Sci 2024; 81: 31.

Absinta

Cortese

Vuolo

, et al. Leptomeningeal gadolinium enhancement across the spectrum of chronic neuroinflammatory diseases. Neurology 2017; 88: 1439–1444.

10.

Absinta

Vuolo

Rao

, et al. Gadolinium-based MRI characterization of leptomeningeal inflammation in multiple sclerosis. Neurology 2015; 85: 18–28.

11.

Vercellino

Costantini

Cogoni

, et al. Association of MRI leptomeningeal enhancement with disability worsening in progressive multiple sclerosis: a clinical and post-mortem study. Mult Scler 2023; 29: 1526–1539.

12.

Okar

Dieckhaus

Beck

, et al. Highly sensitive 3-Tesla real inversion recovery MRI detects leptomeningeal contrast enhancement in chronic active multiple sclerosis. Invest Radiol 2024; 59: 243–251.

13.

Freeze

Schnerr

Palm

, et al. Pericortical enhancement on delayed postgadolinium fluid-attenuated inversion recovery images in normal aging, mild cognitive impairment, and Alzheimer disease. AJNR Am J Neuroradiol 2017; 38: 1742–1747.

14.

Harrison

Allette

Zeng

, et al. Meningeal contrast enhancement in multiple sclerosis: assessment of field strength, acquisition delay, and clinical relevance. PLoS One 2024; 19: e0300298.

15.

Ighani

Jonas

Izbudak

, et al. No association between cortical lesions and leptomeningeal enhancement on 7-Tesla MRI in multiple sclerosis. Mult Scler 2020; 26: 165–176.

16.

Beck

Maranzano

Luciano

, et al. Cortical lesion hotspots and association of subpial lesions with disability in multiple sclerosis. Mult Scler 2022; 28: 1351–1363.

17.

Louapre

Govindarajan

Gianni

, et al. Beyond focal cortical lesions in MS: an in vivo quantitative and spatial imaging study at 7T. Neurology 2015; 85: 1702–1709.

18.

Kupeli

Kocak

Goktepeli

, et al. Role of T1 mapping to evaluate brain aging in a healthy population. Clin Imaging 2020; 59: 56–60.

19.

Bergsland

Ramasamy

Tavazzi

, et al. Leptomeningeal contrast enhancement is related to focal cortical thinning in relapsing-remitting multiple sclerosis: a cross-sectional MRI study. AJNR Am J Neuroradiol 2019; 40: 620–625.

20.

Zurawski

Tauhid

Chu

, et al. 7T MRI cerebral leptomeningeal enhancement is common in relapsing-remitting multiple sclerosis and is associated with cortical and thalamic lesions. Mult Scler 2020; 26: 177–187.

21.

Hildesheim

Ramasamy

Bergsland

, et al. Leptomeningeal, dura mater and meningeal vessel wall enhancements in multiple sclerosis. Mult Scler Relat Disord 2021; 47: 102653.

22.

Thompson

Banwell

Barkhof

, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol 2018; 17: 162–173.

23.

Okuda

Mowry

Beheshtian

, et al. Incidental MRI anomalies suggestive of multiple sclerosis: the radiologically isolated syndrome. Neurology 2009; 72: 800–805.

24.

Murphy

Figueroa

Gibbs

, et al. Sexual transmission of human T-lymphotropic virus type I (HTLV-I). Ann Intern Med 1989; 111: 555–560.

25.

Strober

DeLuca

Benedict

, et al. Symbol digit modalities test: a valid clinical trial endpoint for measuring cognition in multiple sclerosis. Mult Scler 2019; 25: 1781–1790.

26.

Rosti

Hamalainen

Koivisto

, et al. The PASAT performance among patients with multiple sclerosis: analyses of responding patterns using different scoring methods. Mult Scler 2006; 12: 586–593.

27.

Tustison

Avants

Cook

, et al. N4ITK: improved N3 bias correction. IEEE Trans Med Imaging 2010; 29: 1310–1320.

28.

Avants

Tustison

Song

, et al. A reproducible evaluation of ANTs similarity metric performance in brain image registration. Neuroimage 2011; 54: 2033–2044.

29.

Iglesias

Liu

Thompson

, et al. Robust brain extraction across datasets and comparison with publicly available methods. IEEE Trans Med Imaging 2011; 30: 1617–1634.

30.

Cox

. AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput Biomed Res 1996; 29: 162–173.

31.

Liu

Beck

Filippini

, et al. Navigator-guided motion and B0 correction of T2*-weighted magnetic resonance imaging improves multiple sclerosis cortical lesion detection. Invest Radiol 2021; 56: 409–416.

32.

La Rosa

Beck

Maranzano

, et al. Multiple sclerosis cortical lesion detection with deep learning at ultra-high-field MRI. NMR Biomed 2022; 35: e4730.

33.

McAuliffe

Lalonde

McGarry

, et al. Medical image processing, analysis and visualization in clinical research. In: Proceedings 14th IEEE symposium on computer-based medical systems. CBMS 2001, 26 July 2001, pp. 381–386. IEEE.

34.

McGrath

Zaveri

Collins

, et al. High-resolution cortical parcellation based on conserved brain landmarks for localization of multimodal data to the nearest centimeter. Sci Rep 2022; 12: 18778.

35.

Selvaganesan

Whitehead

DeAlwis

, et al. Robust, atlas-free, automatic segmentation of brain MRI in health and disease. Heliyon 2019; 5: e01226.

36.

Dieckhaus

Meijboom

Okar

, et al. Logistic regression-based model is more efficient than U-net model for reliable whole brain magnetic resonance imaging segmentation. Top Magn Reson Imaging 2022; 31: 31–39.

37.

Hoopes

Mora

Dalca

, et al. Synthstrip: skull-stripping for any brain image. Neuroimage 2022; 260: 119474.

38.

Eminian

Hajdu

Meuli

, et al. Rapid high resolution T1 mapping as a marker of brain development: normative ranges in key regions of interest. PLoS One 2018; 13: e0198250.

39.

Sudulagunta

Sodalagunta

Kumbhat

, et al. Posterior reversible encephalopathy syndrome(PRES). Oxf Med Case Reports 2017; 2017: omx011.

40.

Biffi

Greenberg

. Cerebral amyloid angiopathy: a systematic review. J Clin Neurol 2011; 7: 1–9.

41.

DeSisto

O'Rourke

Jones

, et al. Single-cell transcriptomic analyses of the developing meninges reveal meningeal fibroblast diversity and function. Dev Cell 2020; 54: 43–59 44.

42.

O'Rahilly

Muller

. The meninges in human development. J Neuropathol Exp Neurol 1986; 45: 588–608.

43.

Faal

Phan

DTT

Davtyan

, et al. Induction of mesoderm and neural crest-derived pericytes from human pluripotent stem cells to study blood-brain barrier interactions. Stem Cell Rep 2019; 12: 451–460.

44.

Smyth

LCD

Okar

, et al. Identification of direct connections between the dura and the brain. Nature 2024; 627: 165–173.

45.

Aspelund

Antila

Proulx

, et al. A dural lymphatic vascular system that drains brain interstitial fluid and macromolecules. J Exp Med 2015; 212: 991–999.

46.

Ineichen

Tsagkas

Absinta

, et al. Leptomeningeal enhancement in multiple sclerosis and other neurological diseases: a systematic review and meta-analysis. Neuroimage-Clin 2022; 33: 102939.

47.

Cohen

Bresch

Thommel Rocchi

, et al. Should we still only rely on EDSS to evaluate disability in multiple sclerosis patients? A study of inter and intra rater reliability. Mult Scler Relat Disord 2021; 54: 103144.

48.

Arnold

Okar

, et al. Sensitivity of portable low-field magnetic resonance imaging for multiple sclerosis lesions. Neuroimage Clin 2022; 35: 103101.

49.

Tokarska

Tottenham

Baaklini

, et al. How does the brain age in individuals with multiple sclerosis? A systematic review. Front Neurol 2023; 14: 1207626.

50.

Gracien

Reitz

Hof

, et al. Longitudinal quantitative MRI assessment of cortical damage in multiple sclerosis: a pilot study. J Magn Reson Imaging 2017; 46: 1485–1490.

51.

Bevan

Evans

Griffiths

, et al. Meningeal inflammation and cortical demyelination in acute multiple sclerosis. Ann Neurol 2018; 84: 829–842.

52.

Magliozzi

Reynolds

Calabrese

. MRI Of cortical lesions and its use in studying their role in MS pathogenesis and disease course. Brain Pathol 2018; 28: 735–742.

53.

Kutzelnigg

Lucchinetti

Stadelmann

, et al. Cortical demyelination and diffuse white matter injury in multiple sclerosis. Brain 2005; 128: 2705–2712.

54.

Junker

Wozniak

Voigt

, et al. Extensive subpial cortical demyelination is specific to multiple sclerosis. Brain Pathol 2020; 30: 641–652.

55.

Mainero

Louapre

Govindarajan

, et al. A gradient in cortical pathology in multiple sclerosis by in vivo quantitative 7T imaging. Brain 2015; 138: 932–945.

56.

Fadda

Brown

Magliozzi

, et al. A surface-in gradient of thalamic damage evolves in pediatric multiple sclerosis. Ann Neurol 2019; 85: 340–351.

57.

Ouellette

Treaba

Granberg

, et al. 7T Imaging reveals a gradient in spinal cord lesion distribution in multiple sclerosis. Brain 2020; 143: 2973–2987.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB

13.48 MB

0.04 MB

Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T 1 alterations in the adjacent cortex

Abstract

Background

Objectives

Methods

Results

Conclusions

Keywords

Introduction

Methods

Participants

MRI acquisition

LME assessment

Spatial analysis of LME foci

Brain segmentation and white matter lesion analysis

Cortical T1 alteration analysis

Statistical analysis

Data availability

Results

Demographic and clinical characteristics

LME foci are more prevalent in posterior brain regions

Normal appearing cortex subjacent to LME foci exhibit higher T1 relaxation times

Discussion

Supplemental Material

sj-docx-1-mso-10.1177_20552173251408625 - Supplemental material for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T1 alterations in the adjacent cortex

Supplemental Material

Supplemental Material

sj-doc-2-mso-10.1177_20552173251408625 - Supplemental material for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T1 alterations in the adjacent cortex

Supplemental Material

sj-xlsx-3-mso-10.1177_20552173251408625 - Supplemental material for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T1 alterations in the adjacent cortex

Footnotes

Glossary

Acknowledgments

Author contributions

Consent for publication

Declaration of conflicting interests

Funding

Disclosures

Ethics approval and consent to participate

ORCID iDs

Supplemental material

References

Supplementary Material

Cortical T₁ alteration analysis

Normal appearing cortex subjacent to LME foci exhibit higher T₁ relaxation times

sj-docx-1-mso-10.1177_20552173251408625 - Supplemental material for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T₁ alterations in the adjacent cortex

sj-doc-2-mso-10.1177_20552173251408625 - Supplemental material for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T₁ alterations in the adjacent cortex

sj-xlsx-3-mso-10.1177_20552173251408625 - Supplemental material for Leptomeningeal enhancement in multiple sclerosis demonstrates posterior predilection and T₁ alterations in the adjacent cortex