Background: The 2010 McDonald criteria allow diagnosing multiple sclerosis (MS) with one magnetic resonance imaging (MRI) scan. Nevertheless, not all patients at risk fulfil criteria at baseline. Other predictive factors (PFs) are: age ≤40 years, positive oligoclonal bands (OBs), and ≥3 periventricular lesions.
Objective: The purpose of this study was to evaluate the 2010 McDonald criteria performance and to assess other PFs in patients without dissemination in space (DIS).
Methods: Patients with clinically isolated syndrome (CIS) underwent baseline MRI and OB determination with clinical and radiological follow-up. Adjusted hazard ratios (aHRs) for clinically definite MS were estimated for DIS, dissemination in time (DIT), and DIS+DIT. Diagnostic properties at two years were calculated. In cases without DIS, combinations of ≥2 PFs were assessed.
Results: A total of 652 patients were recruited; aHRs were 3.8 (2.5-5.8) for DIS, 4.2 (1.9-9.2) for DIT, and 8.6 (5.4-13.8) for DIS+DIT. Sensitivities were 69.6%, 42.3%, and 36.4%, and specificities were 67.3%, 87.9%, and 90.2%, respectively. In patients without DIS, aHRs varied between 2.7-5.5 and specificities ranged from 73.5-89.7% for PF combinations.
Conclusion: The high specificity of the 2010 McDonald criteria is confirmed. In patients without DIS, PF combinations could be helpful in identifying those at risk for MS.
Keywords: Multiple sclerosis; cerebrospinal fluid; clinically isolated syndrome; diagnostic; magnetic resonance imaging.
© The Author(s), 2014.