Vitrectomy for persistent diffuse diabetic macular edema

Purpose: To evaluate the potential benefit of vitrectomy in eyes with persistent diffuse macular edema. Design: Prospective randomized comparative clinical trial. Methods: Eyes with diffuse diabetic macular edema for 6 to 18 months, an attached posterior hyaloid, and grid laser photocoagulation performed at least 4 months before were included. Patients were randomized either to a vitrectomy group or to a control group. MAIN OUTCOME MEASURES: Evaluations of Early Treatment Diabetic Retinopathy Study (ETDRS) visual acuity, reading vision, and retinal thickness were carried out at baseline and 1, 3, and 6 months after enrollment. Results: Fifty,six eyes (100%) were enrolled in this study. Twenty-five eyes (44.6%) were randomized into Gr I (vitrectomy group) and 31 eyes (55.4%) into Gr II (controls). Both groups were comparable in mean age (62.7 years and 63.9 years) and distribution of gender (one third male, two thirds female). ETDRS visual acuity showed a statistical significance in favor of Gr I at all time points (P=.035 to .005 Fisher's exact test). With Jaeger charts a significance for Gr I was found only at the 6-month examination (P=.01). With optical coherence tomography, the different behavior of retinal thickness changes in both groups during follow-up was statistically significant; P values were <.0001 for month 1, 3, and 6, preferring Gr I. Conclusions: We provide evidence that vitrectomy with internal limiting membrane peeling is superior to observation alone in eyes with persistent diffuse diabetic macular edema for 6 to 18 months. Longer follow-up periods and larger series might be needed to confirm these results and gain additional information.