Retinal Thinning Predicts Cognitive Decline in MS

TOPLINEThinning of the peripapillary retinal nerve fiber layer (pRNFL) and macular ganglion cell-inner plexiform layer (mGCIPL) at baseline was associated with subsequent cognitive decline in patients with multiple sclerosis (MS), independent of relapse activity. Furthermore, thicker macular inner nuclear layer (mINL) at baseline was linked to worsening in processing speed, while thinner mINL predicted decline in verbal learning and memory.METHODOLOGYResearchers conducted a prospective cohort study involving 98 patients with MS (mean age, 50.5 years; 63% women; 80.6% relapsing-remitting MS, 19.4% progressive MS) from University Hospital Basel in Switzerland, followed for a median of 5 years.Participants underwent optical coherence tomography (OCT) at baseline to measure pRNFL, mGCIPL, and mINL thickness; they also underwent cognitive testing annually using the brief international cognitive assessment for multiple sclerosis, which includes the symbol digit modalities test (SDMT), verbal learning and memory test (VLMT), and brief visuospatial memory test.Cognitive progression independent of relapse activity (PIRA) was defined as a more than 10% decrease in score compared with the previous visit (reference visit), confirmed at the following 12-monthly visit (confirmation visit) with the same or a lower score, in the absence of relapses within 90 days before or 30 days after both the event and confirmation visits. Cognitive PIRA with MRI activity (PIRMA) additionally required the absence of new or enlarging fluid-attenuated inversion recovery lesions between the reference and confirmation visits.Brain MRI at baseline measured total lesion volume and gray matter volume to evaluate inflammatory vs neurodegenerative damage.Cox regression models adjusted for age, education, baseline cognitive scores, and treatment assessed associations between baseline retinal layer thickness and risk of cognitive PIRA, cognitive PIRMA, and purely cognitive PIRA events.TAKEAWAYThinner pRNFL at baseline was associated with increased risk for cognitive PIRA in SDMT (hazard ratio [HR], 0.94; P = .027).Thinner mGCIPL at baseline was linked to higher risk for cognitive PIRA in VLMT (HR, 0.94; P = .040).Thicker mINL at baseline was associated with increased risk for cognitive PIRA in SDMT (HR, 1.50; P = .049), cognitive PIRMA in SDMT (HR, 2.64; P = .017), and purely cognitive PIRA in SDMT (HR, 1.60; P = .038).Thinner mINL at baseline was linked to higher risk for cognitive PIRA in VLMT (HR, 0.80; P = .030), cognitive PIRMA in VLMT (HR, 0.72; P = .010), and purely cognitive PIRA in VLMT (HR, 0.79; P = .031).IN PRACTICE"Our findings, which need further validation in larger cohorts, suggest that OCT can provide easily obtainable and noninvasive prognostically relevant biomarkers of progression in MS," the authors wrote.SOURCEThe study was led by Nuria Cerdá-Fuertes, Translational Imaging in Neurology (ThINK) Basel, Department of Medicine and Biomedical Engineering, University Hospital Basel and University of Basel, Basel, Switzerland. It was published online on August 6 in Journal of Neurology.LIMITATIONSThe study was limited by a relatively small number of patients with MS, which resulted in a low total number of PIRA and PIRMA events, further reduced by declining participant numbers after year 5. The researchers also did not control for neuropsychiatric comorbidities, fatigue, or sleep disturbances, all of which can affect cognitive test performance.DISCLOSURESThis study was supported by grants from the Swiss Multiple Sclerosis Society, Stiftung zur Förderung der gastroenterologischen und allgemeinenklinischen Forschung sowie der medizinischen Bildauswertung, and the Swiss National Science Foundation. Multiple authors reported speaker fees, honoraria, advisory board roles, consulting fees, research support, or travel support from various pharmaceutical and biotechnology companies; some authors also reported employment or leadership roles.This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.