Abstract: 233
Type: Free Communications
Abstract Category: Pathology and pathogenesis of MS - MRI and PET
Introduction: Detection of cortical multiple sclerosis lesions using conventional MRI techniques is not reliable enough to be used routinely for disease monitoring in clinical practice, despite widespread recognition of the clinical significance of cortical pathology in multiple sclerosis.
Objective: To assess whether average magnetisation transfer ratio (MTR) of the cortex could serve as a surrogate for cortical lesion load in multiple sclerosis.
Methods: This was a cohort study comprising 56 patients with multiple sclerosis, recruited from Neurology outpatients at Nottingham University Hospitals NHS Trust.
An initial test cohort of 19 patients underwent 7 Tesla brain MRI scanning using magnetisation-prepared-rapid-acquisition-gradient-echo (MPRAGE), and MTR sequences. Cortical lesions were identified on MPRAGE and MTR images of cortical ribbons.
A subsequent validation cohort of 37 patients underwent the same MTR brain protocol, and their cortical lesions were identified using MTR images of cortical ribbons.
Primary outcome measurement was the relationship between visualised cortical lesion load and average MTR of the cortex. Regression analysis was used to determine how much of the variance in cortical MTR was explained by cortical lesion load. Pearson correlation coefficients were calculated for mean cortical MTR vs. cortical lesion counts and volumes and also disease data.
Results: Cortical lesion load was the only variable considered that made a significant contribution to the test cohort multiple linear regression model, with cortical mean MTR as the dependent variable, when also taking into account age, disease duration, expanded disability status scale, and white matter lesion volume (model summary: percentage variance accounted for = 67.6%, p = 0.041; for cortical lesion volumes: standardized coefficient (β) = -0.585, p = 0.020).
In the larger validation cohort mean cortical MTR correlated significantly with cortical lesion count (r = - 0.4308, p = 0.0039), cortical lesion volume (r = -0.3318, p = 0.0224) expanded disability status scale (r = -0.4537, p = 0.0035) and disease duration (r = -0.4118, p = 0.0078).
Conclusions: In the present study, mean MTR of the cortical ribbon correlated significantly with cortical lesion load. If replicable on clinically available scanners, cortical MTR might in future be a candidate to appraise multiple sclerosis cortical lesion load in vivo.
Disclosure: This study was funded by research grants from the UK Multiple Sclerosis Society [grant number 919]; and Medical Research Council [grant number G0700584]. Dr Mistry has participated in advisory boards for Novartis and Roche. Dr Mazibrada has participated in advisory boards for Novartis, Sanofi-Genzyme, Merck-Serono, Biogen, Teva and Roche. Dr Evangelou has participated in advisory boards for Novartis, Sanofi-Genzyme, Merck-Serono, Biogen, Teva, Bayer and Roche. In regards to potential conflicts of interest within the remit of this study none of the other authors has anything to disclose.
Abstract: 233
Type: Free Communications
Abstract Category: Pathology and pathogenesis of MS - MRI and PET
Introduction: Detection of cortical multiple sclerosis lesions using conventional MRI techniques is not reliable enough to be used routinely for disease monitoring in clinical practice, despite widespread recognition of the clinical significance of cortical pathology in multiple sclerosis.
Objective: To assess whether average magnetisation transfer ratio (MTR) of the cortex could serve as a surrogate for cortical lesion load in multiple sclerosis.
Methods: This was a cohort study comprising 56 patients with multiple sclerosis, recruited from Neurology outpatients at Nottingham University Hospitals NHS Trust.
An initial test cohort of 19 patients underwent 7 Tesla brain MRI scanning using magnetisation-prepared-rapid-acquisition-gradient-echo (MPRAGE), and MTR sequences. Cortical lesions were identified on MPRAGE and MTR images of cortical ribbons.
A subsequent validation cohort of 37 patients underwent the same MTR brain protocol, and their cortical lesions were identified using MTR images of cortical ribbons.
Primary outcome measurement was the relationship between visualised cortical lesion load and average MTR of the cortex. Regression analysis was used to determine how much of the variance in cortical MTR was explained by cortical lesion load. Pearson correlation coefficients were calculated for mean cortical MTR vs. cortical lesion counts and volumes and also disease data.
Results: Cortical lesion load was the only variable considered that made a significant contribution to the test cohort multiple linear regression model, with cortical mean MTR as the dependent variable, when also taking into account age, disease duration, expanded disability status scale, and white matter lesion volume (model summary: percentage variance accounted for = 67.6%, p = 0.041; for cortical lesion volumes: standardized coefficient (β) = -0.585, p = 0.020).
In the larger validation cohort mean cortical MTR correlated significantly with cortical lesion count (r = - 0.4308, p = 0.0039), cortical lesion volume (r = -0.3318, p = 0.0224) expanded disability status scale (r = -0.4537, p = 0.0035) and disease duration (r = -0.4118, p = 0.0078).
Conclusions: In the present study, mean MTR of the cortical ribbon correlated significantly with cortical lesion load. If replicable on clinically available scanners, cortical MTR might in future be a candidate to appraise multiple sclerosis cortical lesion load in vivo.
Disclosure: This study was funded by research grants from the UK Multiple Sclerosis Society [grant number 919]; and Medical Research Council [grant number G0700584]. Dr Mistry has participated in advisory boards for Novartis and Roche. Dr Mazibrada has participated in advisory boards for Novartis, Sanofi-Genzyme, Merck-Serono, Biogen, Teva and Roche. Dr Evangelou has participated in advisory boards for Novartis, Sanofi-Genzyme, Merck-Serono, Biogen, Teva, Bayer and Roche. In regards to potential conflicts of interest within the remit of this study none of the other authors has anything to disclose.