Minimizing Graft Preparation Failure in Descemet Membrane Endothelial Keratoplasty.

PURPOSE: To report the failure rate of 2 graft preparation techniques for Descemet membrane endothelial keratoplasty (DMEK) and to evaluate how to minimize graft preparation failure. METHODS: Retrospective, nonrandomized study at an eye bank specialized in graft preparation for lamellar keratoplasty. For 1416 donor corneas, the DMEK graft preparation failure rate was evaluated for 2 different techniques, technique I: "Standardized traditional technique" (n = 341) and technique II: "Standardized no-touch technique" (n = 933), and for grafts that were converted from technique II to technique I during preparation (n = 142). RESULTS: The overall failure rate averaged 3.9% (55/1416): 7.0% (24/341) for technique I and 2.9% (31/1075) for technique II (P < 0.05). Tissue preparations which were converted from technique II to technique I failed in 13.4% (19/142), whereas for grafts that were entirely prepared by technique II, the failure rate was only 1.3% (12/933). The endothelial cell density decrease (before compared with after preparation) did not differ for both techniques (1.1% vs. 0.2%, P > 0.05). CONCLUSIONS: Various DMEK graft preparation techniques may provide failure rates of <4%. A "no-touch preparation" approach (technique II) may combine good graft quality (completely intact endothelial cell layer, ie, negligible preparation-induced endothelial cell density decrease) with low risk of dissection failure, leaving the possibility of conversion to "traditional preparation" (technique I) as a backup method.

The influence of reactive oxygen species on the adhesion of pancreatic carcinoma cells to the peritoneum.

Postoperative peritoneal carcinomatosis is a significant clinical problem after "curative" resection of pancreatic carcinoma. Preoperative surgical trauma activates a cascade of peritoneal defense mechanisms responsible for postoperative intra-abdominal tumor recurrence. Reactive oxygen species (ROS) play a pivotal role in this postoperative inflammatory reaction. This study explores the influence of ROS on adhesion of human pancreatic carcinoma cells to human mesothelial cells. Furthermore this study explores the influence of ROS on the presentation of adhesion molecules on Panc-1 and mesothelial cells. ROS were produced using the enzymatic reaction of xanthine with xanthine oxidase (X/XO). A reproducible in vitro assay to study adhesion of human Panc-1 carcinoma tumor cells to a mesothelial cell monolayer of primary human mesothelial cells was used. Mesothelial monolayers were incubated with ROS produced prior to adhesion of the tumor cells. Incubation of the mesothelial cells with X/XO resulted in a significant increase (69.5%) in adhesion of Panc-1 in all patients. SOD/catalase, anti-oxidants, could reduce this increase by 56.7%. ROS significantly influenced the expression of the adhesion molecules ICAM-1, VCAM-1 and CD44h on mesothelial cells, but did not influence adhesion molecule expression on Panc-1. The ROS released during the post-operative inflammatory reaction may play an important role in the adhesion of pancreatic tumor cells to the mesothelium-possibly by influencing adhesion molecule expression on mesothelial cells. Therefore ROS can partly be responsible for the enhanced post-operative intra-abdominal tumor recurrence.