Towards a Unified Set of Diagnostic Criteria for Multiple Sclerosis

Abstract

Objective The 2017 McDonald criteria continued the separation of diagnostic criteria for relapsing–remitting multiple sclerosis (RRMS) and primary progressive MS (PPMS) for historical, rather than biological, reasons. We aimed to explore the feasibility of a single, unified set of diagnostic criteria when applied to patients with suspected PPMS. Methods We retrospectively identified patients evaluated for suspected PPMS at 5 European centers. The 2017 McDonald PPMS criteria was the gold standard against which the 2017 McDonald RRMS dissemination in space (DIS) and dissemination in time criteria were evaluated. We also investigated modified RRMS DIS criteria, including: (i) optic nerve lesions; (ii) ≥2 spinal cord lesions; and (iii) higher fulfilment of DIS criteria alone (lesions in ≥3 regions) without dissemination in time/positive cerebrospinal fluid, for a diagnosis of PPMS. Results A total of 282 patients were diagnosed with PPMS using the 2017 McDonald criteria, and 40 with alternate disorders. The 2017 McDonald RRMS DIS criteria and the modified DIS criteria including the optic nerve or ≥2 spinal cord lesions performed well in PPMS diagnosis when combined with dissemination in time/positive cerebrospinal fluid (sensitivity 92.9–95.4%, specificity 95%, accuracy 93.2–95.3%). A diagnosis of PPMS based on high fulfillment of modified RRMS DIS criteria had high specificity, but low sensitivity. A diagnostic algorithm applicable to patients evaluated for suspected MS is proposed. Interpretation The 2017 McDonald RRMS criteria and modifications to DIS criteria, currently under consideration, performed well in PPMS diagnosis. Forthcoming revisions to the McDonald criteria should consider a single, unified set of diagnostic criteria for MS. ANN NEUROL 2025;97:571–582