Purchase this article with an account.
Alan Poon, Thomas Wright, Annie Dupuis, Carol A Westall; Retinal thickness irregularities in preclinical diabetic retinopathy. Invest. Ophthalmol. Vis. Sci. 2016;57(12):4237.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Diabetic retinopathy (DR) is the most common complication of diabetes and the leading cause of vision loss. Clinical detection relies on the manifestation of sight-threatening vascular lesions in the retina. However, retinal thickness irregularities are present during preclinical DR. Currently these irregularities are poorly defined.We hypothesized that in preclinical DR, retinal thickness irregularities are localized to specific regions and layers. The purpose of this prospective, cross-sectional observational study was to identify specific regions in individual retinal layers that present with thickness irregularities.
Training data identified a subset of regions with thickness irregularities, which were verified by validation data. Participants were adolescents/young adults with Type 1 diabetes (T1D) with no/minimal DR and healthy controls.Spectral-domain optical coherence tomography was used to obtain high-resolution volumetric retinal scans centered on the fovea (Cirrus HD-OCT 5000; Carl Zeiss Meditec). Automated 3D segmentation measured thicknesses for each retinal layer. Between-group thickness differences were analyzed by subtraction and t-statistic. Linear mixed-effects models of thicknesses by diabetes status determined the significance of thickness irregularities for each region in each layer.
A total of 9 ETDRS regions in each of 10 retinal layers were analyzed. Training data consisted of 5 participants with T1D (mean age 23.2±1.8y) and 8 controls (23.5±4.8y); validation data consisted of 12 additional participants with T1D (20.6±5.5y) and 23 controls (19.8±5.1y). Each retinal layer had unique thickness distributions. In T1D, retinal layers had areas of thinner and/or thicker retina relative to control. Modelling the training data identified four regions with significant retinal thickness irregularities: outer temporal GCL (p=0.01, d=1.7), foveal INL (p=0.04, d=1.3) and inner nasal and outer superior OPL (p=0.0499, d=1.2 and p=0.03, d=1.4, respectively). These four regions did not reach significance with validation data. An exploratory analysis of validation data found significant thickness irregularities in the inner nasal GCL (p=0.03, d=-0.1).
Retinal thickness irregularities in preclinical DR are non-uniformly distributed across the retina. This study is the first to examine the retina in 9 ETDRS regions in each of 10 retinal layers in the 17 participants with T1D tested here.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only