Purpose
To evaluate the effect of the vitreomacular interface on treatment efficacy of intravitreal therapy of uveitis-associated cystoid macular edema (CME)
Methods
Retrospective analysis of CME resolution, CME recurrence rate, monthly course of visual acuity (VA), and central retinal thickness (CRT) after therapeutic intravitreal injection with respect to configuration of the vitreomacular interface (VMI) on OCT. Non-parametric data were compared with Chi-Quadrat test. Differences in distance VA and CRT were evaluated with one-way ANOVA.
Results
59 eyes of 55 patients (47.4±17.7years) were included. 29% had anterior uveitis, 40% intermediate uveitis, 11% posterior uveitis and 20% panuveitis. Mean CME duration was 33±31months. 66% received Intravitreal triamcinolone acetonide, 5% methotrexate, 17% dexamethasone bioerodible implant, and 12% bevacizumab. 39% had posterior vitreous detachment (PVD), 46% vitreomacular adhesion (VMA), 2% vitreomacular traction (VMT), and 13% posterior vitreous attachment (PVA). 64% had epiretinal membranes, however the presence of this confounding factor was equally distributed within groups. 38% showed persistent CME after injection, and 22% had CME relapse within the first 4 months and were retreated. Improvement of vision did not differ among groups (p=0.98) at 1, 2, and 3 months post-injection. The total central retinal thickness (CRT) decrease also did not differ among groups (p=0.29). However, the percentage of patients experiencing a ≥20% CRT-thickness decrease differed according to vitreomacular interface group (p=0.027): 83% of the PVD patients had a more than 20% CRT decrease, whereas only 64% and 16% of the VMA and the PVA groups experienced a more than 20% CRT decrease after intravitreal injection, respectively. The percentage of patients showing persistent CME did not differ among the groups (p=0.18), nor did the relapse rate (p=0.21) nor time until CME relapse (p=0.18).
Conclusions
The configuration of the VMI seems to be a factor contributing to treatment efficacy in uveitis-associated CME; nevertheless the functional VA outcome parameters did not differ according to VMI status. This is the first study showing an effect of the VMI on treatment response in uveitis, but further studies with larger patient populations are warranted to investigate this effect.
Keywords: 746 uveitis-clinical/animal model •
561 injection •
763 vitreous