Descemet stripping automated endothelial keratoplasty after failed penetrating keratoplasty

PURPOSE: To report the rate of graft dislocation, surgical anatomic success, and postsurgical complications associated with Descemet stripping automated endothelial keratoplasty (DSAEK) after previous primary failed penetrating keratoplasty (PK). METHODS: Institutional review board-approved, single-center, retrospective chart review study of 30 eyes of 30 patients with prior failed PK who underwent DSAEK with a minimum of 3 months follow-up. Primary outcomes measured included rates of anatomic success and failure, postoperative complications, lenticle size, visual acuity, intraocular pressure change, and a report of external factors that may affect success. RESULTS: Thirty eyes of 30 patients were identified. The primary dislocation rate was 16.7%. Five primary DSAEKs detached; 1 was successfully rebubbled in the office, 2 had repeat successful DSAEKs, and 2 failed on a second attempt and had a subsequent PK. Despite successful anatomic attachment, 1 eye had primary graft failure, 3 that cleared initially failed within 6 months, and 1 that successfully cleared had subsequent graft rejection resulting in failure at 6 months. The 2 eyes requiring PK were excluded from the visual outcomes analysis, leaving 28 eyes with successfully attached lenticles. At 3 months postoperatively, of the 28 eyes, 19 showed an improvement in visual acuity, 6 had no change in vision from preoperative data, and 1 had worsening of vision (anatomic attachment but endothelial failure). Two of the 28 eyes had no data at that time point. CONCLUSIONS: DSAEK eyes after failed PK demonstrated improved vision with a low complication rate in a majority of patients. The graft dislocation rate and postoperative complications rates are comparable with the primary DSAEK dislocation rates in our own published series and in the literature.