ObjectiveTo observe and analyze the subtype-specific prognostic factors for visual recovery in patients with demyelinating optic neuritis (DON) after glucocorticoid pulse therapy. MethodsA retrospective cohort study. A total of 195 patients (249 eyes) with DON diagnosed by ophthalmology examination at Department of Ophthalmology, Xi'an People's Hospital (Xi'an Fourth Hospital) from January 2021 to December 2024 were included in the study. According to the results of serum antibody detection and clinical diagnostic criteria, the patients were divided into the neuromyelitis optica spectrum disorder (NMOSD)-associated optic neuritis (ON) (NMOSD-ON) group, the myelin oligodendrocyte glycoprotein antitide-associated ON (MOG-ON) group, and the double antibody negative ON group. They were 51 cases (58 eyes), 72 cases (103 eyes), and 72 cases (88 eyes) respectively. Baseline clinical data, imaging characteristics, and treatment protocols were collected. The primary endpoints were complete visual recovery [best-corrected visual acuity (BCVA) ≥1.0] and moderate recovery (BCVA ≥0.5) at 3 months post-onset. Multivariate logistic regression was used to identify independent prognostic factors for visual outcomes within each subtype. ResultsAt 3 months post-onset, complete recovery rates were 9 (15.5%, 9/58) in the NMOSD-ON group, 64 (62.1%, 64/103) in the MOG-ON group, and 31 (35.2%, 31/88) in the double-seronegative ON group. The results of multivariate regression analysis showed that age [odds ratio (OR) =0.901, 95% confidence interval (CI) 0.854-0.950, P＜0.001] and peak visual acuity (OR=0.311, 95%CI 0.147-0.660, P=0.002) and the involvement of optic nerve length ≥1/2 (OR=3.849, 95%CI 1.083-13.682, P=0.037) were the influencing factors for the complete recovery of visual acuity in the affected eyes of the double antibody negative ON group. Age (OR=0.958, 95%CI 0.933-0.983, P=0.001) was the only influencing factor for the complete recovery of visual acuity in the affected eyes of the MOG-ON group. Peak visual acuity (OR=0.288, 95%CI 0.090-0.927, P=0.037) and optic nerve involvement length ≥1/2 (OR=19.974, 95%CI 1.905-209.559, P=0.013) were the influencing factors for the complete recovery of visual acuity in the affected eyes of the NMOSD-ON group. Age (OR=0.936, 95%CI 0.890-0.983, P=0.009), time from onset to intravenous infusion of methylprednisolone sodium succinate intervention (OR=0.854, 95%CI 0.759-0.961, P=0.009), optic disc edema (OR=4.405, 95%CI 1.108-17.512, P=0.035) and peak visual acuity (OR=0.13, 95%CI 0.046-0.365, P＜0.001) were the influencing factors for the moderate recovery of visual acuity in the affected eyes of the double antibody negative ON group. Peak visual acuity was the only influencing factor for the moderate recovery of visual acuity in the MOG-ON group (OR=0.060, 95%CI 0.010-0.352, P=0.002) and the NMOSD-ON group (OR=0.163, 95%CI 0.053-0.500, P=0.001). ConclusionsThe prognostic factors for visual recovery in patients with DON after glucocorticoid pulse therapy are subtype-specific. Peak visual acuity is a common predictor for all subtypes. For NMOSD-ON and double antibody-negative ON, attention should be paid to the length of optic nerve lesions. MOG-ON is age-related. Early intravenous infusion of methylprednisolone sodium succinate for double antiantibody negative ON is more likely to achieve moderate vision recovery.