Endothelial keratoplasty: prospective, randomized, masked clinical trial comparing an injector with forceps for tissue insertion.

Purpose: To compare the complications and outcomes of Descemet stripping automated endothelial keratoplasty (DSAEK) when the tissue is either folded and inserted with a forceps or inserted using a platform injector device without folding.

Design: Prospective, randomized, masked clinical trial.

Methods: DSAEK was performed in 100 eyes of 79 patients undergoing DSAEK surgery for Fuchs corneal dystrophy. Fifty eyes were randomized to have the donor tissue inserted with Charlie II insertion forceps (Bausch & Lomb Surgical) and 50 eyes were randomized to have the donor tissue inserted with the Neusidl Corneal Inserter (Fischer Surgical Inc). All other steps of the surgical procedure were exactly the same. Surgical problems, postoperative complications, and central endothelial cell density at 6 months were recorded and then measured by a masked observer. The study's main outcome measures were total central endothelial cell density and percentage of donor endothelial cell loss from before surgery to 6 months after surgery and rate of complications (graft dislocation and primary graft failure).

Results: No primary graft failures occurred in either group and only 1 dislocation occurred in the series (Neusidl group). One late failure occurred at 6 months (Neusidl group). There was no difference in the preoperative endothelial cell density between the Neusidl and forceps groups, but there was a higher percentage of cell loss with the Neusidl group (33%) than with the forceps group (25%) at 6 months (P = .017).

Conclusions: The Neusidl Corneal Inserter yielded a low immediate complication rate for DSAEK surgery for novice and experienced surgeons. Although still at an acceptable level, short-term endothelial survival was significantly worse after Neusidl tissue insertion than that after forceps tissue insertion.