AUTHOR=Park Soo-Hyun , Park Choul-Yong , Shin Young Joo , Jeong Kyoung Sook , Kim Nam-Hee TITLE=Low Contrast Visual Evoked Potentials for Early Detection of Optic Neuritis JOURNAL=Frontiers in Neurology VOLUME=13 YEAR=2022 URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2022.804395 DOI=10.3389/fneur.2022.804395 ISSN=1664-2295 ABSTRACT=

Optic neuritis (ON) detection is important for the early diagnosis and management of multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD). However, the conventional high-contrast visual evoked potential (VEP) used for ON detection lacks sensitivity for identifying ON presenting as mild or unremarkable visual disturbance, which is common in first-episode ON. Therefore, this study aimed to investigate whether a change in contrast or check size improves the sensitivity of VEP to first-ever ON. In total, 60 patients with the demyelinating disease (29 MS and 31 idiopathic patients with ON) without ON or with first-ever ON at least 6 months prior and 32 healthy controls underwent neuro-ophthalmic evaluations. VEPs were induced using three pattern-reversal checkerboard stimuli having, respectively, 10% contrast with a check size of 32' (LC32 VEP), 100% contrast with a check size of 32' (HC32 VEP; conventional VEP), and 100% contrast with a check size of 16' (HC16 VEP). The receiver operating characteristic (ROC) curve analysis and area under the curve (AUC) were calculated to determine the most appropriate VEP method for detecting optic nerve involvement. The optimal cut-off point was determined using the Youden index (J-index). The McNemar test was used to determine whether dichotomous proportions were equivalent. In comparison with first-ever ON eyes (n = 39) and healthy eyes (n = 64), LC32 VEP showed the highest AUC for discriminating ON (0.750, p < 0.001; 0.730 for HC32 VEP, p < 0.001; 0.702 for HC16 VEP, p = 0.001). In the first-ever ON group, LC32 VEP and conventional HC32 VEP were abnormal in 76.9 and 43.6%, respectively (McNemar, p < 0.001), and combining these tests did not improve sensitivity. These indicate that LC32 VEP is the most sensitive method for detecting first-ever ON. Visual evoked potential with 10% contrast stimuli was superior to conventional VEP for detecting first-ever ON. Thus, adding these LC stimuli might be helpful in identifying optic nerve involvement in ON with mild or unremarkable visual impairment.