Graft Rejection After Descemet's Stripping Automated Endothelial Keratoplasty: Graft Survival and Endothelial Cell Loss

Abstract

To evaluate the effects of graft rejection episodes after Descemet's stripping automated endothelial keratoplasty surgery (DSAEK) on long-term endothelial cell density (ECD) decline and graft survival. Retrospective, comparative analysis of an interventional case series. We included 615 eyes of 415 Fuchs' dystrophy patients at a single institution with ≥ 6 months follow-up and without comorbidities known to influence postoperative ECD. All patients were enrolled as part of an ongoing, institutional review board-approved clinical protocol for a long-term, prospective study of endothelial keratoplasty in patients with endothelial dysfunction. Preoperative specular microscopy of donor corneal tissue was performed. Postoperative specular microscopy measurements were recorded at 6 and 12 months, and yearly thereafter. The percentages of endothelial cell loss recorded at 1, 2, 3, and 4 years were compared with the Mann-Whitney U test. Percentage ECD declines were calculated at each time point from the results of the preoperative and postoperative specular microscopy. Patients with graft rejection episodes and late endothelial failure were identified. Graft rejection was defined as findings of keratic precipitates with or without corneal edema, or anterior chamber cell and flare with or without corneal edema after the initial resolution of perioperative inflammation. We identified 45 cases of graft rejection. The greatest number of rejections occurred between postoperative months 12 and 18. Eyes with a graft rejection episode had a higher median percentage decline in ECD at all time points compared with eyes without graft rejection episodes. This was statistically significant at 2 and 3 years postoperatively. There is a trend toward a greater percentage of ECD loss with time in eyes experiencing graft rejection after DSAEK surgery. Proprietary or commercial disclosure may be found after the references.