Inner retinal thickness after intravitreal dexamethasone implant in non-infectious uveitic macular edema assessed by Spectral Domain-Optical Coherence Tomography

Abstract

Purpose: To identify the retinal and choroidal changes in non-infectious uveitic macular edema (UME) after treatment with intravitreal dexamethasone implant (IDI, Ozurdex) assessed by advanced analysis of Spectral-Domain Optical Coherence Tomography (SD-OCT) images.

Methods: Single centre consecutive case series. Five eyes of five patients with UME underwent SD-OCT 1 week pre and 1 month post IDI injection using the HD OCT Cirrus device (Carl Zeiss Meditec, Dublin, CA, USA). Raw OCT images were exported and analysed using custom-made advanced OCT analysis software (OCTOR, Doheny Eye Institute Reading Center, Los Angeles). Thirty-two B-scans were segmented covering a 6x6 mm area using the macular cube protocol (1/4 scans analysed) at each timepoint. Compartments of interest were defined as: vitreous, inner retina (retinal nerve fiber layer, ganglion cell layer, inner plexiform layer, inner nuclear layer), outer retina (outer plexiform layer, outer nuclear layer, external limiting membrane, photoreceptor layer), retinal pigment epithelium, Sattler's layer and Haller’s layer. Measurements obtained included: mean thickness, volume and intensity of each stratus analysed.

Results: Mean pre-treatment retinal thickness and volume was 306,2±56,3 microns and 11±2 mm³, and post-treatment was 258,2±47 microns (p<0,01) and 9,2±1,6 (p<0,01) respectively. Differences were obtained for inner retina layer compartment thickness (195,6±41,1 versus 166,3±34,2 microns, p<0,01) and volume (7,1±1,4 versus 5,9±1,2 mm³, p<0,01) but not for outer retinal layer compartment thickness (108±23,6 versus 90,1±18,9 microns, p=0,13) or volume (3,8±0,8 versus 3,2±0,6 mm³, p=0,13). No differences were observed in thickness or volume in any of the choroidal layers.

Conclusions: The IDI injection obtains significant decreases in retinal thickness and volume 1 month after treatment in UME eyes. More concisely, the inner retinal layer compartment appears to be responsible of this thinning, whereas the outer retinal layer compartment is less affected by the anti inflammatory effect of the IDI implant. Advanced analysis of retinal and choroidal layers in SD-OCT may provide new insights to ascertain the pathophysiology of UME and potentially monitor its response to current intravitreal treatments.