Optical coherence tomography versus visual evoked potentials for detecting visual pathway abnormalities in patients with neuromyelitis optica spectrum disorder

Background and Purpose Optical coherence tomography (OCT) and visual evoked potentials (VEPs) can be used to detect optic neuritis (ON). However, the comparative sensitivities of OCT and VEPs for detecting ON in neuromyelitis optica spectrum disorder (NMOSD) are unclear, and so we assessed these sensitivities. Methods This cross-sectional study included 73 patients with aquaporin-4 antibody-seropos-itive NMOSD, and 101 eyes with ON. The clinical characteristics, visual acuity (VA), Expanded Disability Status Scale (EDSS) scores, OCT peripapillary retinal nerve fiber layer (RNFL) thickness, and VEPs of the patients were evaluated. Results OCT and VEPs were abnormal in 68% and 73% of eyes with a history of ON, respectively, and in 2% and 9% of eyes without ON. Test sensitivities were influenced by the number of ON episodes: the OCT RNFL thickness and VEPs were abnormal in 50% and 67% of the eyes with first-ever ON episode, respectively (p=0.041), with the combination of both tests detecting abnormalities in up to 75% of the eyes. The sensitivities of the OCT RNFL thickness and VEPs increased to 95% and 83%, respectively, after the second or subsequent ON episode (p=0.06), with the combination of both tests detecting abnormalities in 95% of cases. The OCT RNFL thickness and VEP latency/amplitude were correlated with EDSS scores and VA. Conclusions VEPs were superior for detecting subclinical or first-ever ON, while OCT was better for detecting eyes with multiple ON episodes. The correlations of OCT and VEPs with clinical disability measures indicate that these tests are potential markers of the disease burden in NMOSD.