BACKGROUND: Patients with multiple sclerosis (MS) demonstrate thinning of peripapillary retinal nerve fiber layer (RNFL) and decreased macular volume as measured by optical coherence tomography (OCT). To our knowledge, there are no previous reports from a large MS OCT database with strict quality control measures that quantitate RNFL and macula in patients with relapsing-remitting multiple sclerosis. METHODS: The University of California Davis OCT Reading Center gathered OCT data at baseline as part of the North American phase 3 trial of fingolimod (Gilenya). Average RNFL thickness (RNFLT) and macular volume (TMV) were measured using time domain OCT (TD-OCT). RNFL quadrants, clock hours, and macular subfields were included. With strict quality control and accounting for signal strength differences, scans were categorized as "reduced" or "not reduced" for each field, based on being less than 5th percentile for age-matched controls derived from the normative database in the scanner software. Patients were deemed "abnormal" if at least 1 eye had reduced values for a given parameter. Patients with abnormalities in corresponding RNFL and macular subfields were compared by cross-tabulation. RESULTS: The TD-OCT data were prospectively collected from 939 of the 1,083 trial patients, 712 of whom met all final quality and data inclusion criteria. Of the final cohort, 242 (34.0%) demonstrated reduced (less than 5th percentile) average RNFLT in at least 1 eye. One hundred seventy-eight (25.0%) patients had reduced TMV. One hundred twenty-eight (18.0%) demonstrated both reduced TMV and RNFLT in the same eye, whereas 42 (5.8%) had reduced TMV and RNFLT in both eyes. Of the 242 patients with reduced average RNFL thickness, 128 (52.9%) also had reduced TMV. Fifty patients had reduced TMV in the absence of reduced RNFLT in at least 1 eye, a cohort prevalence of 7.0%. Quadrant and subfield analysis showed a predominance of temporal and inferior RNFL thinning, with inferior macular thinning corresponding best to RNFL thinning. CONCLUSION: RNFL and macular thinning/volume loss is common at baseline in relapsing-remitting multiple sclerosis, as measured by TD-OCT. When the RNFL is thin, the macular volume is reduced in more than half of the patients. There is a population of reduced TMV without any reduction in RNFLT. Documenting the prevalence and distribution of these structural abnormalities supports recent reports and suggests new retinal areas to probe for functional vision changes in MS. Copyright © 2013 North American Neuro-Ophthalmology Society.

Background

Patients with multiple sclerosis (MS) demonstrate thinning of peripapillary retinal nerve fiber layer (RNFL) and decreased macular volume as measured by optical coherence tomography (OCT). To our knowledge, there are no previous reports from a large MS OCT database with strict quality control measures that quantitate RNFL and macula in patients with relapsing-remitting multiple sclerosis.

Methods

The University of California Davis OCT Reading Center gathered OCT data at baseline as part of the North American phase 3 trial of fingolimod (Gilenya). Average RNFL thickness (RNFLT) and macular volume (TMV) were measured using time domain OCT (TD-OCT). RNFL quadrants, clock hours, and macular subfields were included. With strict quality control and accounting for signal strength differences, scans were categorized as "reduced" or "not reduced" for each field, based on being less than 5th percentile for age-matched controls derived from the normative database in the scanner software. Patients were deemed "abnormal" if at least 1 eye had reduced values for a given parameter. Patients with abnormalities in corresponding RNFL and macular subfields were compared by cross-tabulation.

Results

The TD-OCT data were prospectively collected from 939 of the 1,083 trial patients, 712 of whom met all final quality and data inclusion criteria. Of the final cohort, 242 (34.0%) demonstrated reduced (less than 5th percentile) average RNFLT in at least 1 eye. One hundred seventy-eight (25.0%) patients had reduced TMV. One hundred twenty-eight (18.0%) demonstrated both reduced TMV and RNFLT in the same eye, whereas 42 (5.8%) had reduced TMV and RNFLT in both eyes. Of the 242 patients with reduced average RNFL thickness, 128 (52.9%) also had reduced TMV. Fifty patients had reduced TMV in the absence of reduced RNFLT in at least 1 eye, a cohort prevalence of 7.0%. Quadrant and subfield analysis showed a predominance of temporal and inferior RNFL thinning, with inferior macular thinning corresponding best to RNFL thinning.

Conclusion

RNFL and macular thinning/volume loss is common at baseline in relapsing-remitting multiple sclerosis, as measured by TD-OCT. When the RNFL is thin, the macular volume is reduced in more than half of the patients. There is a population of reduced TMV without any reduction in RNFLT. Documenting the prevalence and distribution of these structural abnormalities supports recent reports and suggests new retinal areas to probe for functional vision changes in MS.

Supplemental Digital Content is Available in the Text.

Artificial intelligence (AI)-based diagnostic algorithms have achieved ambitious aims through automated image pattern recognition. For neurological disorders, this includes neurodegeneration and inflammation. Scalable imaging technology for big data in neurology is optical coherence tomography (OCT). We highlight that OCT changes observed in the retina, as a window to the brain, are small, requiring rigorous quality control pipelines. There are existing tools for this purpose. Firstly, there are human-led validated consensus quality control criteria (OSCAR-IB) for OCT. Secondly, these criteria are embedded into OCT reporting guidelines (APOSTEL). The use of the described annotation of failed OCT scans advances machine learning. This is illustrated through the present review of the advantages and disadvantages of AI-based applications to OCT data. The neurological conditions reviewed here for the use of big data include Alzheimer disease, stroke, multiple sclerosis (MS), Parkinson disease, and epilepsy. It is noted that while big data is relevant for AI, ownership is complex. For this reason, we also reached out to involve representatives from patient organizations and the public domain in addition to clinical and research centers. The evidence reviewed can be grouped in a five-point expansion of the OSCAR-IB criteria to embrace AI (OSCAR-AI). The review concludes by specific recommendations on how this can be achieved practically and in compliance with existing guidelines.