Central vein sign on 3T MRI may be an accurate diagnostic biomarker for MS
FROM JAMA NEUROLOGY
The central vein sign detected on 3T MRI had a high sensitivity and moderate specificity in identifying cases of multiple sclerosis and differentiating them from other, similar-looking conditions, according to a recent study in JAMA Neurology.
Jens Wuerfel, MD, of the Medical Image Analysis Center in Basel, Switzerland, told the Multiple Sclerosis Journey in an interview that the central vein sign (CVS) could be used as a biomarker in cases where multiple sclerosis (MS) is difficult to distinguish from other conditions.
“The CVS will facilitate diagnoses in unclear cases, both clinically (e.g., radiologically isolated or clinically isolated syndromes) and radiologically (e.g., tumefactive or otherwise noncharacteristic lesions), adding diagnostic specificity,” said Dr. Wuerfel, who is also an author in the study.
Dr. Wuerfel and colleagues used data from the MAGNIMS (Magnetic Resonance Imaging in MS) Study Group [www.magnims.eu] to perform an international, cross-sectional study of 606 patients with and without MS who already were enrolled in ongoing observational studies and patients with information in neuroimaging databases at eight different European centers during January 2010–November 2016. Patients were included in the study if they had received a 3T MRI scan as well as 3T MRI with susceptibility-weighted imaging sequence or T2*-weighted and three-dimensional fluid-attenuated inversion recovery (FLAIR) sequence.
There were 236 patients (38.9%) with relapsing-remitting MS (RRMS). Patients without MS had diagnoses of cerebral small vessel disease (23.4%), clinically isolated syndrome (19.3%), neuromyelitis optica spectrum disorder (5.3%), migraine (4.8%), systemic lupus erythematosus (4.1%), and cluster headaches (0.8%), or were in a control group consisting of older patients (mean age, 60.2 years) and patients with diabetes (3.3%). A majority of patients in the study were women (413 patients; 68.2%); 66.6% of patients in the RRMS group were women, and 70.4% of patients without MS were women.
Representative image of the central vein sign.
Credit: Dr. Jens Wuerfel
In total, 4,447 lesions in 487 patients were analyzed, which included 2,815 lesions in 225 patients with RRMS, 690 lesions in 98 patients with clinically isolated syndrome, 473 lesions in 88 patients with small-vessel disease, 240 lesions in 20 patients with diabetes, 121 lesions in 29 patients with migraine and cluster headaches, and 54 lesions in 13 patients with neuromyelitis optica spectrum disorder. Using a 35% CVS proportion threshold, the sensitivity was 68.1% and the specificity was 82.9% for the CVS as a biomarker for differentiating MS from non-MS conditions.
When the criteria of three positive CVS lesions were met, the sensitivity was 61.9% and the specificity was 89.0%, and the sensitivity increased when patients received a highly optimized T2*-weighted sequence for detecting the CVS, researchers said. “Additional features such as hypointense lesion rims or lesion distribution (e.g., Dawson’s fingers) add to the specificity of the diagnosis,” said Dr. Wuerfel.
From a clinical standpoint, T2-weighted (T2w) and susceptibility-weighted sequences in addition to T2-weighted FLAIR, as well as the post hoc fusion of the datasets could extend scanning time and add a post-processing step, said Dr. Wuerfel. “The interpretation of the finding (e.g., the centrality of the vein) as well as the exclusion of other sources of susceptibility artifacts may require some training, but is absolutely feasible in a routine clinical setting,” he said.
Dr. Wuerfel noted that there are ongoing prospective cohort trials in Switzerland through the Swiss Multiple Sclerosis Cohort, through National Institute of Arthritis and Musculoskeletal and Skin Diseases in the United States, and internationally in the United Kingdom and Australia.
The authors reported personal and institutional relationships in the form of grants, personal fees, advisory board and steering committee memberships, research and educational support, paid employment, and consultancies for a variety of pharmaceutical companies, agencies, societies, and other organizations.
SOURCE: Sinnecker T et al. JAMA Neurol. 2019 Aug 19. doi: 10.1001/jamaneurol.2019.2478.