Contributions
Abstract: P842
Type: Poster
Abstract Category: Clinical aspects of MS - 1 Diagnosis and differential diagnosis
Introduction: Differentiating multiple sclerosis (MS) from other disorders that cause white matter abnormalities relies on radiographic assessments of imperfect specificity. Studies using FLAIR* MRI have demonstrated that detection of a “central vein sign” (CVS) in MRI white matter lesions may be highly specific for MS. These studies have primarily evaluated the proportion of central veins from the total number of MRI lesions on a scan, a method impractical for clinical application. This study evaluated the specificity and sensitivity of a simplified CVS algorithm using only three lesions for a diagnosis of MS.
Methods: 40 participants were studied: 10 with MS without additional comorbidities for white matter abnormalities; 10 with MS with additional comorbidities for white matter abnormalities; 10 with migraine, MRI white matter abnormalities, and no additional comorbidities for white matter abnormalities; and 10 who had previously been erroneously diagnosed with MS. 3T MRI FLAIR and T2* sequences were acquired for each participant, and FLAIR* images were created. MRIs were de-identified, randomly ordered, and provided to three MS physician raters, at three different institutions, who were blinded to diagnosis. MRIs with at least three or more ovoid lesions >3mm in at least one plane, restricted to the juxtacortical, subcortical, or deep white matter regions, were included for CVS evaluation. The three raters determined if any three of these lesions in each participant demonstrated CVS on FLAIR*. The specificity and sensitivity for a diagnosis of MS if three lesions with CVS were identified was calculated for each rater individually, and then averaged between raters.
Results: Average specificity for a diagnosis of MS for three lesions with CVS was 0.81 (0.68-0.95). Average sensitivity for three lesions was 0.83 (0.76-0.94). A correct prediction of diagnosis demonstrated kappa=0.31 between the raters.
Conclusion: A simplified determination of CVS in three white matter lesions on FLAIR* MRI demonstrated good specificity and sensitivity and acceptable inter-rater reliability for a diagnosis of MS. A simplified assessment for CVS may allow its implementation in clinical practice.
Disclosure: Funded by the University of Vermont College of Medicine, Department of Neurological Sciences
Andrew J Solomon: consulting and advisory boards for Biogen, EMD Serono, and Teva. Richard Watts: nothing to disclose. Daniel Ontaneda: Grant support from Genzyme, Genentech, Novartis, consulting for Novartis, Genentech, Biogen. Martina Absinta: nothing to disclose. Pascal Sati: nothing to disclose. Daniel S Reich: research funding from Vertex Pharmaceuticals and the Myelin Repair Foundation.
Abstract: P842
Type: Poster
Abstract Category: Clinical aspects of MS - 1 Diagnosis and differential diagnosis
Introduction: Differentiating multiple sclerosis (MS) from other disorders that cause white matter abnormalities relies on radiographic assessments of imperfect specificity. Studies using FLAIR* MRI have demonstrated that detection of a “central vein sign” (CVS) in MRI white matter lesions may be highly specific for MS. These studies have primarily evaluated the proportion of central veins from the total number of MRI lesions on a scan, a method impractical for clinical application. This study evaluated the specificity and sensitivity of a simplified CVS algorithm using only three lesions for a diagnosis of MS.
Methods: 40 participants were studied: 10 with MS without additional comorbidities for white matter abnormalities; 10 with MS with additional comorbidities for white matter abnormalities; 10 with migraine, MRI white matter abnormalities, and no additional comorbidities for white matter abnormalities; and 10 who had previously been erroneously diagnosed with MS. 3T MRI FLAIR and T2* sequences were acquired for each participant, and FLAIR* images were created. MRIs were de-identified, randomly ordered, and provided to three MS physician raters, at three different institutions, who were blinded to diagnosis. MRIs with at least three or more ovoid lesions >3mm in at least one plane, restricted to the juxtacortical, subcortical, or deep white matter regions, were included for CVS evaluation. The three raters determined if any three of these lesions in each participant demonstrated CVS on FLAIR*. The specificity and sensitivity for a diagnosis of MS if three lesions with CVS were identified was calculated for each rater individually, and then averaged between raters.
Results: Average specificity for a diagnosis of MS for three lesions with CVS was 0.81 (0.68-0.95). Average sensitivity for three lesions was 0.83 (0.76-0.94). A correct prediction of diagnosis demonstrated kappa=0.31 between the raters.
Conclusion: A simplified determination of CVS in three white matter lesions on FLAIR* MRI demonstrated good specificity and sensitivity and acceptable inter-rater reliability for a diagnosis of MS. A simplified assessment for CVS may allow its implementation in clinical practice.
Disclosure: Funded by the University of Vermont College of Medicine, Department of Neurological Sciences
Andrew J Solomon: consulting and advisory boards for Biogen, EMD Serono, and Teva. Richard Watts: nothing to disclose. Daniel Ontaneda: Grant support from Genzyme, Genentech, Novartis, consulting for Novartis, Genentech, Biogen. Martina Absinta: nothing to disclose. Pascal Sati: nothing to disclose. Daniel S Reich: research funding from Vertex Pharmaceuticals and the Myelin Repair Foundation.