Abstract： Objective:To analyze the prognostic factors of vision of myelin oligodendrocyte glycoprotein (MOG) antibody positive associated optic neuritis (ON) after methylprednisolone pulse therapy.Methods:A clinical observational study. A total of 32 patients (47 eyes) with MOG antibody positive ON were observed and followed up in the ophthalmology department of Beijing Tongren Hospital Affiliated to Capital Medical University and Beijing Puren Hospital from March 2019 to January 2022. Clinical data including the best corrected visual acuity (BCVA) and orbital magnetic resonance imaging were recorded. The BCVA was examined by Snellen visual acuity chart, which was finally converted into the logarithm of the minimal angle of resolution (logMAR) for statistical analysis. There were 22 case (38 eyes) with complete image data. All patients were treated with intravenous methylprednisolone pulse (IVMP) for 3-5 days. According to the intervention time (from onset to glucocorticoid treatment), the patients were divided into three groups: <7 d group, 7-14 d and> 14 d group, with 16, 13, 11 eyes, respectively. The median follow-up time was 28 months. After 1 week, 1, 3 and 6 months treatment, the same equipment and methods were used for relevant examinations to observe the changes of visual acuity and the factors influencing the prognosis of visual acuity after IVMP treatment. Logistic regression and linear regression were used to analyze the prognostic correlation factors. Receiver operating characteristic (ROC) curve was used to determine the critical cut-off point of intervention timing.Results:Among the patients, 16 were male and 16 were female. The median onset age was 26 years. The onset duration time was 5-60 days. There were 18 cases (56.3%, 18/32) with abnormal serum immune indexes. The initial symptom was decreased vision with unilateral or bilateral ON. Seventeen (53.1%, 17/32) cases had unilateral ON and 15 (46.9%, 15/32) cases with bilateral ON. Thirty-six eyes (76.6%, 36/47) got optic disc edema, 37 eyes (78.7%, 37/47) accompanied by pain of ocular movement. The nadir logMAR BCVA was mean 1.69±0.13. Long T2WI signals with segmental thickening in the orbital segment of the optic nerve were obtained in 27 eyes (71.1%, 27/38) and in 24 eyes (63.2%, 24/38) with optic nerve and sheath enhancement. During the follow-up period, there were 10 cases of relapse (31.3%, 10/32). The logMAR BCVA of attacked eyes were 0.52±0.09, 0.22±0.06, 0.12±0.06, 0.10±0.06 at 1 week and 1, 3 and 6 months after IVMP treatment, respectively. The rate of BCVA improvement was the fastest at 1 week after treatment, and BCVA returned to stability at 3 months. Logistic regression analysis showed that the timing of intervention was significantly correlated with the prognosis of vision in primary onset patients (odds ratio=12.17, P=0.006), with a negative linear regression relationship ( r=-0.48, 95% confidence interval -0.71--0.17, P=0.008). Comparing the logMAR BCVA between the intervention time >14 group with the <7 group and the 7-14 group, there were statistically significant difference ( P=0.017, 0.037), respectively. The cut-off point of ROC curve to predict the optimal intervention time was 13.5 days. Other factors such as: gender, age, predisposing factor, pain of eye motion, edema of optic disc, bilateral ON, imaging changes, abnormal autoimmune indicators were not associated with the prognosis of visual acuity. Conclusion:The timing of hormone intervention in primary onset patients is an important factor affecting the prognosis of vision and the optimal intervention time window of IVMP is two weeks.