Low positive predictive value of Multiple Sclerosis diagnostic criteria in demyelinating syndromes with red flags of better explanation.

Introduction: Perivenular lesions (PVL) are a cardinal pathological feature of Multiple Sclerosis (MS), that brain MRI can non-invasively detect. In demyelinating syndromes, a frequency of PVL above 50% among all white matter lesions (50%-rule) is characteristic and specific of MS. Objectives: To evaluate the positive predictive value of current MS diagnostic criteria in a group of atypical MS patients, considering the 50%-rule as the diagnostic gold standard of true MS cases. Methods and materials: The frequency of PVL was tested in “MS- plus” patients, defined as fulfilling MS 2017 revised McDonald diagnostic criteria based on the dissemination in space (DIS) and in time (DIT) but also carrying red flags (clinical, laboratory or MRI features) suggesting the existence of a better explanation, but not formally allowing an- other diagnosis. MS-plus patients were compared with typical MS controls: Relapsing-remitting (RR) MS cases with typical demyelinating syndromes (Typical MS, n=28) and RR-MS-plus patients (MS-plus, n= 60), were included. The patients received one brain MRI scan with conventional and 3DFLAIR* scans for PVL analysis. The PVL frequency (PVL- f) and other conventional brain MRI characteristics were evaluated in each patient, who was then categorized according to the MS-specific PVL-f threshold of 50%. Results: The Typical MS patients had a median PVL-f of 91% (range 67–100%) whereas the MS-plus patients of 55% (range 8–100%; p=0.001). Patients fulfilling the 50%-rule were 28/28 (100%) and 32/60 (53%) respectively in the two groups (p< 0,00001). Considering this MS-plus enriched MS population as a whole (n= 88), the patients fulfilling the 50% rule were n= 60 (68%). These frequencies represent the positive predictive values of the 50%-rule fulfillment when the diagnostic criteria for MS based on DIS/DIT are applied in each of these patient populations. The other conventional MRI measures examined did not improve the values of the DIS/DIT criteria. Conclusions: Considering the 50% rule the “gold standard” for identification of true MS cases, the positive predictive value of the current MS diagnostic criteria in demyelinating syndromes with red flags of better explanation is low. In any MS population, this predictive value will de- pend on the prevalence of MS-plus cases.