Corneal Endothelial Cell Loss in Femtosecond Laser-assisted Descemet's Stripping Automated Endothelial Keratoplasty: A 12-month Follow-up Study
Yun Feng,Hong-Qiang Qu,Jing Ren,Philipp Prahs,Jing Hong
DOI: https://doi.org/10.4103/0366-6999.220320
2017-12-20
Abstract:Background: Descemet's stripping automated endothelial keratoplasty (DSAEK) surgery offers a more standardized approach and reliable method to create corneal grafts with an instrument such as a microkeratome. With the development of Descemet's membrane endothelial keratoplasty, an excellent clinical outcome is seen in the treatment of corneal endothelial dysfunctions, which indicates that thinner corneal graft results in better clinical outcome. With the recent development of the femtosecond laser, ultrathin corneal graft preparation has become possible. This study aimed to report corneal graft endothelial cell loss (ECL) in a large series of cases undergoing DSAEK with femtosecond laser-assisted corneal graft preparation within a 12-month period. Methods: This study was designed as a prospective, noncomparative, interventional case series. Totally 126 consecutive eyes with endothelial failure of 120 patients, who had corneal endothelial decompensation and underwent femtosecond-assisted DSAEK using the VisuMax femtosecond laser system, were included in the study. Central endothelial cell density (ECD) was recorded postoperatively at 2 weeks (n = 126), 1 month (n = 126), 3 months (n = 110), 6 months (n = 101), and 12 months (n = 71) and then compared with the preoperative eye bank measurements. Pre- and postoperative central ECDs were evaluated using Heidelberg retina tomography-III confocal microscopy. ECL was calculated for each postoperative time point. Graft thickness was examined using anterior segment-optical coherence tomography. Results: Mean preoperative cell count was 3383 ± 350 cells/mm2. Mean postoperative cell counts were 2382 ± 707 cells/mm2, 2179 ± 685 cells/mm2, 2074 ± 688 cells/mm2, 1884 ± 662 cells/mm2, and 1723 ± 624 cells/mm2 at 2 weeks, 1, 3, 6, and 12 months, respectively; these represented the ECL of 29.7 ± 19.7%, 35.4 ± 19.5%, 38.6 ± 19.8%, 44.3 ± 18.9%, and 48.9 ± 18.4% at the each corresponding time point. The mean corneal graft thickness after surgery was 142 ± 48 μm, 118 ± 41 μm, 108 ± 37 μm, 100 ± 32 μm, and 99 ± 32 μm at each corresponding study visit, respectively. There was no correlation between corneal graft thickness and corneal ECL (R = 0.039). Conclusions: Corneal ECL remained relatively stable up to 12 months after femtosecond laser-assisted ultrathin DSAEK in a large case series. No correlation between cell loss and corneal graft thickness was found, which indicated that corneal graft preparation by the femtosecond laser was safe. ECL was faster within the first 6 months and relatively stable thereafter.