Relationship Between Retinal Layer Thickness and Disability Worsening in Relapsing-Remitting and Progressive Multiple Sclerosis

• Published in: Journal of neuro-ophthalmology Vol. 41; no. 3; pp. 329 - 334
• Main Authors: Cellerino, Maria, Priano, Luca, Bruschi, Nicolò, Boffa, Giacomo, Petracca, Maria, Novi, Giovanni, Lapucci, Caterina, Sbragia, Elvira, Uccelli, Antonio, Inglese, Matilde
• Format: Journal Article
• Language: English
• Published: United States 01.09.2021
• Subjects: Adolescent; Adult; Aged; Disability Evaluation; Disease Progression; Female; Follow-Up Studies; Humans; Male; Middle Aged; Multiple Sclerosis, Relapsing-Remitting - diagnosis; Multiple Sclerosis, Relapsing-Remitting - rehabilitation; Nerve Fibers - pathology; Retinal Ganglion Cells - pathology; Retrospective Studies; Tomography, Optical Coherence - methods; Young Adult
• ISSN: 1070-8022; 1536-5166
• DOI: 10.1097/WNO.0000000000001165

Data regarding the predictive value of optical coherence tomography (OCT)-derived measures are lacking, especially in progressive multiple sclerosis (PMS). Accordingly, we aimed at investigating whether a single OCT assessment can predict a disability risk in both relapsing-remitting MS (RRMS) and PMS. One hundred one patients with RRMS and 79 patients with PMS underwent Spectral-Domain OCT, including intraretinal layer segmentation. All patients had at least 1 Expanded Disability Status Scale (EDSS) measurement during the subsequent follow-up (FU). Differences in terms of OCT metrics and their association with FU disability were assessed by analysis of covariance and linear regression models, respectively. The median FU was 2 years (range 1-5.5 years). The baseline peripapillary retinal nerve fiber layer (pRNFL) and ganglion cell + inner plexiform layer (GCIPL) were thinner in PMS compared with RRMS (P = 0.02 and P = 0.003, respectively). In the RRMS population, multivariable models showed that the GCIPL significantly correlated with FU disability (0.04 increase in the EDSS for each 1-μm decrease in the baseline GCIPL, 95% confidence interval: 0.006-0.08; P = 0.02). The baseline GCIPL was thinner in patients with RRMS with FU-EDSS >4 compared with those with FU-EDSS ≤4, and individuals in the highest baseline GCIPL tertile had a significantly lower FU-EDSS score than those in the middle and lowest tertile (P = 0.01 and P = 0.001, respectively). These findings were not confirmed in analyses restricted to patients with PMS. Among OCT-derived metrics, GCIPL thickness had the strongest association with short-medium term disability in patients with RRMS. The predictive value of OCT metrics in the longer term will have to be further investigated, especially in PMS.