For the descriptive and statistical analyses, patients were divided into 3 major groups. The first group consisted of patients with HLA-B27(−)B51(+) Behçet uveitis. The second group comprised patients with HLA-B27(+)B51(+) Behçet uveitis, and the third group had HLA-B27(+)B51(−) non-Behçet uveitis. Demographic data, including age at onset of uveitis and sex, were noted, and uveitis characteristics, including anatomical location of uveitis, clinical course, bilaterality, and associated ocular findings (hypopyon, papillitis, retinal hemorrhage, retinal vasculitis, vitritis), according to the criteria of the Standardization of Uveitis Nomenclature (SUN) Working Group grading scheme,10 were recorded at the initial visit or during follow-up. We reviewed ocular complications, including significant cataract and secondary glaucoma, extensive (≥180°) posterior synechiae, clinically significant macular edema, optic atrophy, retinal vein occlusion, formation of retinal new vessels, and retinal detachment. All patients were treated according to anatomical location of the uveitis, degree of intraocular inflammation, and etiology of the uveitis. Most patients initially were treated with topical steroids, and periocular steroids were used for severe recurrent attacks with involvement of posterior segments. Systemic steroids were given if the inflammation could not be controlled with previous therapy, and systemic immunosuppressive agents were used for patients in whom previous therapeutic strategies had failed. Because the nature of Behçet uveitis was usually chronic and severely recurrent, we tried to treat Behçet uveitis with posterior segment involvement by immunosuppressive agents (azathioprine and cyclosporin A) combined with prednisone at the initial visit or early during follow-up. Colchicine treatment was administered by the patients' rheumatologists. The appropriate surgical interventions were performed for patients with visually significant cataract and medically intractable glaucoma. Pars plana vitrectomy was performed in cases of nonclearing vitreous opacity, epiretinal membranes, persistent intraocular inflammation irrespective of the use of immunosuppressive agents, and retinal detachment. Initial visual acuity was defined as acuity obtained at the initial visit. Because visual acuity is usually severely affected during inflammatory episodes, we defined the final visual acuity as the visual acuity during remission at the last follow-up. The distribution of counts was analyzed using a Fisher exact test, correcting for multiple comparisons by requiring a P value less than .05 on individual tests for a net confidence greater than 95%. Parametric data were compared using a Mann-Whitney U test.