Abstract
Purpose :
Specular microscopy (SM) is widely used to analyze endothelium after processing and preservation of donor corneas. However, corneal tissue is manipulated and submitted to different temperatures during this process. Some artifacts may arise and induce misleading information on SM data. Artifact means any change in the structure or condition of the cells produced by artificial means, such as laboratory manipulation. This study proposes a classification for the different types of artifacts found in SM of donor corneas and quantifies them in a small sample.
Methods :
A trained eye bank technician performed specular microscopy (KeratoAnalyzer EKA-10 KONAN©) in 7 donor corneal tissues (mean age of 57.14 ± 6.12 years old) on the day of their arrival in the eye bank (group 1) and the day of their distribution (group 2). All tissues were preserved in Eusol-C at 4°C and heated to 27°C before examination. Four endothelial images were obtained from the central donor corneal area. The images were semi-automatically analyzed by the manufacturer’s software (KONAN KSS-EB10) and evaluated by a trained physician. Artifacts (Figure 1) were classified in: 1- Unicelullar: endothelial edema (EE) smaller or the same size as 1 endothelial cell (EC); 2- Paucicellular: EE between 2 and 10 EC in size (in this case, the approximate amount of EC fitting in the edematous area were counted) or guttae; 3 – Multicellular (also called striae): EE greater than 10 EC in size (irregular outline) or a manipulation trail (regular outline). The gradient of colors represented the damage severity. In addition, the quality of each image was classified according to McCarey’s review of corneal SM for FDA clinical trials into good, fair, poor or impossible. Descriptive statistics and Mann-Whitney U test were used for statistical analysis (p<0.05).
Results :
The mean corneal storage time was 1.86 ± 1.21 days. The number of artifacts per image is shown in Tables 1 and 2 (plotted side by side for better comparison). No guttae were found. The number of striae was significantly higher in the poor + impossible images compared to the good + fair images in both groups 1 and 2 (p<0.01).
Conclusions :
The new classification presented in this study can easily encompass and categorize all possible artifacts in donor corneas’ specular microscopy, possibly providing a universal language on this subject. As expected, striae were more prevalent in the worst quality images.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.