Purchase this article with an account.
Emily A Swanson, Jena Tavormina, Tara L Favazza, Anne Moskowitz, Ronald M Hansen, James D Akula, Anne Fulton; Quantification of the Structure of the Perifoveal Retina in Retinopathy of Prematurity. Invest. Ophthalmol. Vis. Sci. 2014;55(13):3381.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare features of the perifoveal retina in eyes of prematurely born subjects, with and without a history of retinopathy of prematurity (ROP), to those in eyes of term-born subjects.
Spectral-domain optical coherence tomographs (OCTs), centered horizontally on the fovea and spanning 20°, were obtained from 27 term-born subjects, 8 preterm subjects with a history of ROP and 8 preterm subjects (≤33 wk GA) who never had ROP. Subjects were 9.2-37.8 years old at test (median 20.4 years). In each image, ten distinct boundaries were traced in ImageJ to delineate: 1) the nerve fiber layer, 2) the inner plexiform layer and the ganglion cell layer, 3) the inner nuclear layer, 4) the outer plexiform layer, 5) the outer nuclear layer, 6 and 7) the respective mitochondria sparse and mitochondria dense portions of the inner segments, 8) the outer segments, and 9) the retinal pigment epithelium (including the OS tips) and choriocapillaris. The ‘rim’ of the fovea was defined by zero-crossings in the derivative of a difference of Gaussians fitted to the innermost (vitreo-retinal) boundary. All boundaries were then ensemble fit to a Gaussian on a linear slope, f(x)=a*exp(-½*((x-x0)/b)2+c*x+y0. Horizontal offset (x0) and slope (c) parameters were shared across all fits. The amplitude (a) and standard deviation (b) parameters of the Gaussian fit to the vitreo-retinal boundary served as respective measures of foveal depth and breadth. The vertical offsets (y0) served as measures of layer thickness at the rim; the values of f(x) for x=x0 served as measures of thickness below the central fovea. Parameters were analyzed by ANOVA and Tukey’s HSD post-hoc test.
There was no significant difference in foveal depth between term-born and preterm subjects, but depth was significantly lower in ROP subjects; foveal breadth did not differ between any groups. Both at x0 and at the rim (defined by the difference of Gaussians), ROP subjects’ retinae were significantly thicker than those in term-born and preterm groups.
Mathematical modeling of the perifoveal layers provides convenient, objective measures of the fovea and the surrounding retina. These parameters can be used to discriminate normal eyes and those with a history of ROP. Residual perifoveal abnormalities in ROP eyes may explain abnormalities in central retinal function and vision mediated by the central retina.
This PDF is available to Subscribers Only