Abstract
Purpose::
To test the feasibility and reproducibility of routine screening of eye bank globes by rasterstereography corneal topography.
Methods::
A retrospective analysis was performed on computerized rasterstereography images collected from 285 pairs of intact donor globes with no significant surface defects by Intraoperative fluorescein-based Rasterstereography Corneal Topography over six years in the Central Ohio Lions Eye Bank (Columbus, Ohio). The eye bank technicians were similarly instructed how to acquire images and there was no monitoring of the data collected. Donor eyes were applied with 1 drop of sterile fluorescein and imaged in their open storage cup. Technicians controlled the alignment, fluorescein coverage, projected grid focus and total number of images collected. The raw rasterstereography images were archived with unique identifiers. All eyes without at least 2 scans were included, and 2 scans were randomly chosen from those with greater than 2 scans. The best fit sphere over 8-mm and the simulated steep and flat K meridians were recorded.
Results::
The data acquired in this study provides a comprehensive set of topography observed in normal transplantable and nontransplantable donor eye bank eyes, including eyes with surface defects difficult to measure by Placido-based computerized videokeratography. The normal population distribution of topography by eye and scan number were determined and their values reported here:. BFS (OD, Scan1) = 41.99 ± 1.89 D, K1 = 42.19 ± 2.16 D, K2 = 43.81 ± 2.03 D, mean K = 43.00 ± 2.16 D; and BFS (OD, Scan2) = 42.06 ± 1.88 D, K1 = 42.19 ± 2.15 D, K2 = 43.82 ± 2.04 D, mean K = 43.01 ± 2.15 D; BFS (OS, Scan1) = 42.08 ± 1.91 D, K1 = 42.14 ± 2.32 D, K2 = 43.87 ± 2.27 D; mean K 43.01 ± 2.32 D; BFS (OS, Scan2) = 42.12 ± 1.89 D, K1 = 42.16 ± 2.34 D, K2 = 43.86 ± 2.27 D, mean K = 43.01 ± 2.34 D. Repeated measures ANOVA analysis (SAS) showed that there was no clinically significant difference between sequential scans or left versus right eyes for all but one condition (P>0.05). There was a statistically significant effect between the calculated BFS and scan number (P=0.004), however the overlap was clinically insignificant (≤0.05 D).
Conclusions::
Routine collection of donor eye bank topography is feasible using Rasterstereography Corneal Topopgraphy. Identification of outliers with abnormal surfaces on the normative curves supports the potential to screen eye bank eyes based on Rasterstereography Corneal Topopgraphy parameters.
Keywords: topography • cornea: clinical science • imaging/image analysis: clinical