Purchase this article with an account.
Luisa F Obregon Jacome, Cristina Modak, Jianyan Huang, Srinivas R Sadda, Vikas Chopra, Olivia L Lee, Doheny Image Reading Center; Consistency in Endothelial Cell Density Values from Different Central Locations as Graded by Specular Microscopy. Invest. Ophthalmol. Vis. Sci. 2014;55(13):2065.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To determine if endothelial cell density (ECD) values from different central locations are comparable enough to one another to allow longitudinal evaluations from single central ECD readings in both normal and diseased eyes.
Multiple central images per eye from 34 control eyes and 22 glaucoma eyes, of which 5 had central corneal guttata, were captured using the non-contact Konan Specular microscope NP-9900. Three good quality images were selected for each eye included in the study. All images were manually graded by trained graders using Konan’s CellChek Analysis Software to obtain ECD measurements. The relative standard deviation (RSD), representing the percent difference from the mean, was used to assess ECD variation across images of the same cornea. Less than 2.5% RSD was considered acceptable agreement. For corneas with guttata, a previously described conversion formula was used to extrapolate an “effective” ECD from “local” ECD values, where percent areas were determined using the ImageJ image analysis software.
168 total images were analyzed. ECD values obtained from different images of the same cornea showed excellent agreement for controls (RSD=1.1%) and slightly more variability in patients without guttata (RSD=2.8%). On the other hand, ECD variation in patients with guttata (RSD=7.7%) was much greater. This variation could however be drastically reduced (RSD=0.6%) when “effective” ECD values were substituted.
ECD values from slightly different central locations show little variation in normal eyes and diseased eyes without guttata, and the same result can also be achieved in eyes with guttata after conversion to an “effective” ECD, suggesting that, with trained graders, single readings may be sufficiently effective in yielding accurate ECD values for the central cornea. These findings, which may allow single ECD readings from both normal and diseased eyes even in the presence of guttata, are very important for longitudinal studies requiring ECD values.
This PDF is available to Subscribers Only