DMEK grafts prepared from corneas stored in TISSUE-C and CARRY-C (deswelling medium) show similar viable endothelial cell density

The potential cytotoxicity of dextran in deswelling media for DMEK graft preparation remains controversial. This study compared endothelial viability and viable endothelial cell density (vECD) in paired deswelled and control (non-deswelled) corneas. Eight pairs of human corneas were preserved in TISSUE-C (standard organ culture medium). Baseline endothelial cell densities (ECD1, ECD2) at two time points were equivalent. Two days before DMEK preparation, one cornea from each pair was transferred to CARRY-C (6% dextran deswelling medium), while its pair remained in TISSUE-C (control). On the preparation day, ECD3 was measured. Viability (percentage of viable endothelial areas) and vECD were determined by Hoechst-33,342/Calcein-AM staining, and cell morphology was assessed with CD166 immunofluorescence. ECD3 was higher in control corneas than in deswelled corneas (2150 ± 737 vs. 1708 ± 745 cells/mm², p = 0.0176), confirmed by Hoechst-based ECD (2025 ± 723 vs. 1775 ± 706, p = 0.0167). Viability was significantly higher in deswelled corneas (78% vs. 96%, p = 0.0013), yielding comparable vECD between groups (1628 ± 715 vs. 1712 ± 723, p = 0.3733). Non-viable areas in control corneas corresponded to endothelial folds from corneal edema. Deswelling markedly reduced folds, and cell migration from viable to non-viable areas was observed, explaining the lower ECD after deswelling. In conclusion, short-term deswelling lowers the measured ECD but does not reduce the total number of viable endothelial cells. The apparent decrease reflects cell migration from viable areas to cover the entire endothelial surface. These findings support the safety of dextran-based deswelling media for DMEK graft preparation and likely apply to penetrating keratoplasty as well.