Vitrectomy and Internal Limiting Membrane (ILM) Removal for Diabetic Macular Edema in Eyes With Absent Vitreo–Macular Traction Fails to Improve Visual Acuity: Results of a 12 Months Prospective Randomized Controlled Clinical Trial

Abstract

Abstract: : Purpose: To report the results of the first randomized controlled trial (RCT) to assess the efficacy of pars–plana vitrectomy with ILM removal on the resolution of diabetic macular edema (DME) . Methods: Prospective RCT comparing no–intervention (Observation group) versus vitrectomy and ILM removal (Treatment group) in patients with DME non–responsive to standard laser therapy. All patients had masked refracted ETDRS logMAR visual acuity, slit–lamp biomicroscopy, LOCS III scoring and optical coherence tomography measurements (OCT) at baseline, 6 months and12 months. Eyes with ischemic maculopathy, media opacity, presence of epiretinal membrane, vitreous hemorrhage or severe non–proliferative / proliferative diabetic retinopathy were excluded. Results: There were 12 eyes of 12 patients in the Treatment group and 8 eyes of 8 patients in the Observation group. The two groups were matched for age, sex, duration of diabetes, mean duration of DME (16m in Observation group and 21m Treatment group; p=0.31), mean number of macular lasers (Observation group 2.6, Treatment group 2.4 months; p=0.73) and mean baseline logMAR acuity (Observation group 45 letters, Treatment group 52 letters; p=0.30). Baseline OCT demonstrated that all the eyes in the Treatment group had an attached posterior hyaloid with no evidence of vitreo–macular traction. At 12 months follow–up, the mean foveal thickness in the observation group was 450 ± 40µm whilst in the treatment group it was reduced to 250.6 ± 56.8µm. At 12 months follow–up, there was no significant change in log MAR acuity in the Observation group (Mean loss 3 letters; p = 0.47, paired T test). In the treatment group, 3 eyes improved greater than 10 letters. However, there was no significant improvement in log MAR acuity (p=0.8). Regression analysis demonstrated that the baseline visual acuity was the only clinical variable associated with an improvement in visual acuity (p=0.002). None of the patients in the Treatment group had significant surgical complications. One patient in each group developed significant lens opacity within 12 months of follow up. Conclusions: Surgery for refractory diabetic macular edema in the absence of vitreo–macular traction is not associated with significant improvement in vision as compared to observation alone.

Keywords: vitreoretinal surgery • diabetes • clinical (human) or epidemiologic studies: treatment/prevention assessment/controlled