Abstract
Purpose::
To present a novel technique to deliver the endothelial graft in small incision Descemet’s Stripping Endothelial Keratoplasty (DSEK) and to compare the amount of graft trauma between forceps delivery and this novel technique in an in vitro surgical model.
Methods::
Twenty healthy human donor corneas (ten pairs) were obtained from the North Carolina Eye Bank. Each cornea was sectioned using an automated microkeratome system (Moria; Antony, France) with a 250 µm head for corneas thinner than 500 µm and a 300 µm head for all others. The recipient model was prepared by affixing a human donor corneal rim to a styrofoam block and creating a 3mm clear corneal wound with a keratome. A curved well slide was placed directly underneath the host cornea. For each pair of corneas, the left endothelial graft underwent single fold delivery with forceps, while the right endothelial graft was delivered with a bent needle from a paracentesis 180° from the main wound through a SFC-45 Visian implantable Collamer lens cartridge (STAAR Surgical; Monrovia, CA). Prior to insertion, Healon (AMO; Santa Ana, CA) was applied to the donor endothelium. Each graft was inserted via its respective method into the host model directly onto the slide. Each graft was stained with 0.25% trypan blue and 0.2% alizarin red, and digital photomicrographs at 25X were taken. A proportion of graft injury (sum of abnormally stained areas divided by total graft area) was calculated and differences were analyzed with the Wilcoxon signed rank test.
Results::
The mean graft pachymetry and post-cut cell count were 165 µm and 2648 cells/mm2 for the forceps group and 151 µm and 2731 cells/mm2 for the cartridge group. After insertion, the mean proportion of graft endothelial injury from forceps delivery through a 3mm wound was 26.02% (N=10, SD±14.85%). The mean proportion of graft endothelial injury from cartridge delivery through a 3mm wound was 9.85% (N=10, SD±4.33%). Within each pair, the median difference (cartridge - forceps) between the two methods was -13%, representing less endothelial injury with the cartridge. This difference was statistically significant (p=0.006).
Conclusions::
In our DSEK surgical model, inserting an endothelial graft through a 3mm corneal wound using a novel cartridge technique created significantly less endothelial damage than with forceps insertion. Customized adaptations of the instrumentation used in this novel method may offer better endothelial protection and could potentially result in better clinical outcomes with DSEK.
Keywords: cornea: clinical science • transplantation • cornea: endothelium