To evaluate the safety and effect of bevacizumab pretreatment on the incidence of recurrent vitreous hemorrhage and visual acuity after vitrectomy for proliferative diabetic retinopathy.
This was a consecutive, retrospective, and comparative cohort study. Patients undergoing vitrectomy from September 2006 through November 2007 at the Emory Eye Center for complications of proliferative diabetic retinopathy were identified and reviewed. A total of 33 eyes pretreated with bevacizumab and 104 untreated eyes were observed for postoperative vitreous hemorrhage and final visual acuity.
Patients in the bevacizumab group were significantly younger than those in the untreated group (average age, 46.4 vs. 58.4 years) and were more likely to have 20-gauge instrumentation (58% vs. 36%). An average of 9.6 days passed between injection and surgery. Early (4-6 weeks) rebleed rates were 15% versus 13% in the bevacizumab and untreated groups, respectively, and not statistically different. Preoperative (7/200 vs. count finger at 4′), 1-month postoperative (20/200−3 vs. 20/150), and 3-month postoperative visual acuity (20/100−3 vs. 20/100+2) were not statistically different between groups. No statistical difference was found in rebleed rates regarding the gauge of vitrectomy.
Bevacizumab pretreatment for diabetic vitrectomy was not associated with any observed complications but did not influence rates of postoperative vitreous hemorrhage or final visual acuity in this retrospective series. The overall incidence of postoperative early vitreous hemorrhage in this series was 13% and seems lower than historically reported rates.
A consecutive, retrospective, and comparative cohort study was performed on patients undergoing vitrectomy for complications from proliferative diabetic retinopathy. Bevacizumab pretreatment for diabetic vitrectomy did not seem to affect rates of postoperative vitreous hemorrhage or final visual acuity in this series but was not associated with any observed complications. The overall incidence of postoperative vitreous hemorrhage in this series was 13%.
From the *Department of Ophthalmology, Emory University, Atlanta, Georgia; and †Department of Ophthalmology, Vanderbilt University, Nashville, Tennessee.
Supported in part by an unrestricted grant from Research to Prevent Blindness, the Heed Foundation (to S.J.K.), and the Ronald G. Michels Foundation (S.J.K.).
Dr. Hubbard is a paid consultant for Genentech Corporation and Theragenics Corporation.
The authors have not disclosed any other potential conflicts of interest.
Reprint requests: Wayne R. Lo, MD, 3600 North Interstate Avenue, Portland, OR 97227; e-mail: firstname.lastname@example.org