Abstract
Purpose :
Specular microscopy has significant limitations when it comes to evaluating the suitability of post-processed DMEKs. Peripheral damage from peeling, scoring, and s-stamping is not captured by the limited field of view of specular microscopy. Trypan blue is a non-toxic vital dye that stains dead endothelial cells. Area of cell damage (ACD) on trypan staining is a recognized alternative metric for evaluating global damage in post-processed DMEKs. Many eye banks routinely estimate ACD based on trypan staining, but no eye bank currently quantifies ACD due to how labor intensive it is. Modern segmentation software can rapidly quantify area of staining in cornea photos. However, different eye banks use different imaging setups with a variety of specifications.Therefore, ACD measured using automated software is potentially different. The purpose of this study is to determine if ACD is consistent when analyzed from images taken with different cameras that meet minimal system resolution requirements.
Methods :
We compared a basic digital microscope setup (Dinolite) and a high-end dissecting stereoscope and digital camera (Nikon). Both cameras were set up to obtain DMEK graft-sized images at a system resolution of 57 line-pairs/mm. DMEK grafts were stained with an FDA-approved 0.06% trypan solution. Grafts were then imaged consecutively by the 2 hardware setups. Each pair of photos for a given cornea were superimposed on each other, rotated and scaled to match each other in order identify the same region of interest (ROI) in both images. ACD was then calculated for the matched ROIs and compared statistically.
Results :
11 post-processed, trypan-stained DMEK images were obtained consecutively with digital scope and stereo microscope. 1 was excluded for poor image quality. Linear regression showed good agreement in ACD (R2= 0.63, residual SE = 0.88%, P=0.006) between hardware setups. Mean ACD did not differ between cameras (5.74±1.38% vs. 5.62±1.25%, P=0.67).
Conclusions :
We found it is possible to obtain comparable ACD results from different microscopes if they have matching resolution and ROIs. Small digital microscopes are a cost-effective, standardizable solution for ACD imaging in DMEK. The consistency of ACD across microscopes may facilitate broader eye bank adoption of ACD and establishment of large image datasets for future development of AI-driven ACD analysis software.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.