Silent Retinal Neurodegeneration in Multiple Sclerosis: Structural Evidence from Clinically Unaffected Eyes Using Swept-Source OCT and OCT Angiography—A Cross-Sectional, Observational Study
Katarina Katanić Pasovski, Ivana Todorović, Viktor Pasovski, Dragana Ristić, Zorana Pavlović, Miloš Danilović, Nemanja Rančić, Tatjana Bošković Matić, Ranko Raičević, Evica DinčićBackground/Objectives: Retinal optical coherence tomography (OCT) has emerged as a sensitive biomarker of neurodegeneration in multiple sclerosis (MS), yet eyes without overt optic neuritis (ON) are routinely pooled as “clinically unaffected” despite their heterogeneous histories. We evaluated whether never-ON eyes and fellow eyes after unilateral ON differ structurally and microvascularly using swept-source OCT (SS-OCT) and OCT angiography (OCTA). Methods: In this cross-sectional, single-center study, 126 clinically unaffected MS eyes—96 never-ON eyes and 30 fellow eyes after unilateral ON—were compared with 118 healthy control eyes. SS-OCT quantified ganglion cell–inner plexiform layer (GCIPL), peripapillary retinal nerve fiber layer (pRNFL), and macular RNFL (mRNFL) thickness, while OCTA measured superficial vascular plexus (SVP) vessel density. Between-group differences were assessed using generalized estimating equations with participant-level clustering, empirical (sandwich) standard errors, adjustment for age and sex, and false discovery rate correction. Results: Despite preserved visual acuity, both never-ON and fellow eyes showed structural retinal thinning relative to controls. GCIPL thickness followed a stepwise gradient—66.14 ± 4.31, 62.08 ± 7.03, and 58.03 ± 7.71 µm in controls, never-ON eyes, and fellow eyes, respectively (FDR-adjusted q = 0.020 for fellow vs. never-ON eyes)—and pRNFL and mRNFL showed a similar overall pattern. After false discovery rate correction, OCTA parameters did not differ significantly between groups. Conclusions: Clinically unaffected eyes in MS are not structurally normal, and fellow eyes after unilateral ON carry a greater burden of silent retinal damage than never-ON eyes. These two phenotypes should be analyzed separately in MS imaging research. Structural OCT measures, particularly GCIPL thickness, appear more sensitive than microvascular indices for detecting subclinical retinal involvement.