Sukhum Silpa-ArchaJoan J. LeeSutasinee BoonsoponMarisol T. LizárragaJanine M. PrebleDusit SujiraratPranav PatelC. Stephen FosterMassachusetts Eye Research and Surgery InstitutionOcular Immunology and Uveitis FoundationRangsit UniversityMahidol UniversityHarvard Medical School2018-12-112019-03-142018-12-112019-03-142016-11-01Retina. Vol.36, No.11 (2016), 2220-2226153928640275004X2-s2.0-84966559292https://repository.li.mahidol.ac.th/handle/20.500.14594/41050Copyright © by Ophthalmic Communications, Society, Inc. Purpose: To identify prognostic factors for poor visual outcome in patients with birdshot retinochoroidopathy. Methods: A case-control study of 98 patients with birdshot retinochoroidopathy (196 eyes) was evaluated with a follow-up period of at least 12 months. After exclusion of glaucoma, optic atrophy, and macular scar, the remaining eligible patients were categorized into two groups: poor visual outcomes and good visual outcomes. Poor visual outcome was defined as less than -6 mean deviation score on Swedish interactive threshold algorithm (SITA) short-wavelength automated perimetry (SWAP) test and abnormality (amplitude or implicit time) of 30 Hz flicker electroretinogram at 4-year follow-up and at the most recent visit for separate analysis. Potential factors between both groups were statistically analyzed by Chi-square test and logistic regression model. Results: After the aforementioned exclusion, the remaining 77 patients with an average follow-up period of 52 ± 29 months (335 person-years, 36% with follow-up of more than 5 years) were divided into two groups. Sixteen patients were categorized as having poor visual outcome. Univariate analysis identified significant association of abnormal 30 Hz flicker electroretinogram amplitude (P 0.004), implicit time (P 0.002), and SITA SWAP mean deviation at the initial visit (P < 0.001) in the poor visual outcome group. Multivariate logistic regression analysis identified only SITA SWAP mean deviation to be associated with poor visual outcome (adjusted odds ratio, 32.50; 95% confidence interval [3.84-275.32]; P 0.001) at the initial visit. To verify the model validity, an analysis of 42 patients at 4-year follow-up was performed and the outcome was confirmed (adjusted odds ratio, 8.80; 95% confidence interval [1.58-49.16]; P 0.013). Conclusion: Worse SITA SWAP mean deviation at the initial visit is a predictor of poor visual outcome in patients with birdshot retinochoroidopathy, and may serve as a proxy marker for delayed effective steroid sparing therapy in patients with birdshot retinochoroidopathy.Mahidol UniversityMedicineArticleSCOPUS10.1097/IAE.0000000000001051