Subclinical Optic Nerve Involvement in Radiologically Isolated Syndrome: Multimodal Detection and Diagnostic Impact

OBJECTIVES: We aimed to determine the frequency of subclinical optic nerve (ON) lesions using MRI, optical coherence tomography (OCT), and visual evoked potentials (VEP) in radiologically isolated syndrome (RIS), and to assess their diagnostic and prognostic significance. METHODS: We conducted a retrospective, multicenter study of 179 RIS individuals followed in clinical practice who met the 2023 RISC criteria. The diagnostic performance of the 2024 McDonald criteria, with and without optic nerve assessment, was evaluated and compared with that of the 2017 criteria. Associations with clinical conversion, comorbidities, and MRI disease activity were analyzed using multivariate models. RESULTS: Silent ON lesions appeared in 107/179 (59.8%) individuals, mainly detected by VEP (100/164; 61.0%) and OCT (69/118; 58.5%), with routine MRI identifying 31/166 (18.7%). During follow-up, 71 (39.7%) had a first clinical event. The presence, laterality, or number of silent ON lesions did not correlate with clinical conversion, event type, or MRI activity. Younger age and the absence of comorbidities associated with other MRI lesions, rather than with ON lesions, were associated with clinical conversion. Including ON in the 2024 McDonald criteria increased sensitivity but decreased specificity, with 17/179 (9.5%) meeting dissemination-in-space criteria solely due to ON lesions. INTERPRETATION: Subclinical ON lesions are common in RIS and are mainly found by OCT and VEP, not routine MRI. Including ON increases sensitivity but does not predict clinical conversion and may lower specificity. These findings suggest cautious interpretation of ON and support a multimodal assessment approach in RIS.