Using a Cone-Shaped Glass Funnel Adapter Reduces Endothelial Cell Loss Caused by Preloading Descemet Membrane Endothelial Keratoplasty Tissue.

Purpose: The aims of this study were (1) to compare "front" and "rear" methods for loading Descemet membrane endothelial keratoplasty (DMEK) tissue into both micro-Jones and standard-Jones tubes and (2) to evaluate the efficacy of a cone-shaped glass funnel adapter designed to make loading DMEK tissue safer for corneal endothelial cells.

Methods: The corneal endothelium was stained with 0.06% trypan blue to confirm equivalence between mate corneas. The tissues were then processed using the Iowa Lions Eye Bank standard DMEK protocol. In comparison 1, one mate was loaded into the rear of a micro-Jones or standard-Jones tube and the other was loaded into the front of the same tube. In comparison 2, one mate was loaded into the front of the micro-Jones tube and the other was loaded through the cone-shaped funnel adapter into the rear. All tissues were ejected through the front of the modified Jones tubes and assessed for endothelial cell loss (ECL) with calcein AM staining, FIJI, and Trainable Weka Segmentation; scroll widths were measured digitally.

Results: There were no statistically significant differences in ECL between front and rear loading [micro (N = 6 pairs): front 15.74% vs. rear 17.95%; standard (N = 6 pairs): front 19.58% vs. rear 19.17%; all P > 0.05]. DMEK scrolls loaded with the funnel adapter exhibited lower ECL compared with scrolls loaded through the front [micro (N = 8 pairs): front 13.53% vs. loading funnel 2.40%; P < 0.001]. Loading with the adapter was not faster (front 6.66 seconds vs. loading funnel 5.52 seconds; P = 0.24).

Conclusions: Using a cone-shaped DMEK loading funnel may reduce ECL sustained during preloading.