PURPOSE: To determine the visual outcomes and complications after multiple vitrectomies for repeat diabetic vitreous hemorrhage. METHODS: A retrospective review during a 4-year period of patients requiring multiple vitrectomies for nonclearing vitreous hemorrhages with at least a 6-month follow-up. RESULTS: Of the 38 cases of multiple vitrectomies for diabetic vitreous hemorrhage, the initial visual acuity was 20/50 or better in 5%, between 20/60 and 20/400 in 37%, and worse than 20/400 in 58%. The final visual acuity after the last vitrectomy was 20/50 or better in 25%, between 20/60 and 20/400 in 47%, and worse than 20/400 in 28%. Patients had a mean improvement of 1.08 lines of visual acuity, and a statistically significant difference in logMAR visual acuity was noted when the last corrected visual acuity was compared with baseline acuity by way of paired t-testing. Although a trend toward visual improvement was noted in patients who underwent multiple vitrectomies, multivariate models failed to detect any association between number of surgeries or demographic variables and change in visual acuity. CONCLUSION: Multiple vitrectomies for recurrent diabetic vitreous hemorrhage can have a favorable anatomic outcome while maintaining ambulatory vision.
PURPOSE: To determine the visual outcomes and complications after multiple vitrectomies for repeat diabetic vitreous hemorrhage. METHODS: A retrospective review during a 4-year period of patients requiring multiple vitrectomies for nonclearing vitreous hemorrhages with at least a 6-month follow-up. RESULTS: Of the 38 cases of multiple vitrectomies for diabetic vitreous hemorrhage, the initial visual acuity was 20/50 or better in 5%, between 20/60 and 20/400 in 37%, and worse than 20/400 in 58%. The final visual acuity after the last vitrectomy was 20/50 or better in 25%, between 20/60 and 20/400 in 47%, and worse than 20/400 in 28%. Patients had a mean improvement of 1.08 lines of visual acuity, and a statistically significant difference in logMAR visual acuity was noted when the last corrected visual acuity was compared with baseline acuity by way of paired t-testing. Although a trend toward visual improvement was noted in patients who underwent multiple vitrectomies, multivariate models failed to detect any association between number of surgeries or demographic variables and change in visual acuity. CONCLUSION: Multiple vitrectomies for recurrent diabetic vitreous hemorrhage can have a favorable anatomic outcome while maintaining ambulatory vision.