Outcomes of femtosecond laser-assisted Descemet membrane endothelial keratoplasty for failed penetrating keratoplasty

Objective: To analyze the outcomes of femtosecond laser-enabled Descemet membrane endothelial keratoplasty (FE-DMEK) in treatment of failed penetrating keratoplasty (PK) grafts. Study Design: Retrospective, interventional case series. Participants: Patients with a failed PK graft who underwent FE-DMEK at Toronto Western Hospital, Canada, between 2014 and 2016. Methods: Outcome measures were best spectacle-corrected visual acuity (BSCVA), endothelial cell density (ECD), rates of graft detachment, rebubbling, rejection, and failure. Results: Eight eyes of 8 patients were included. Mean age was 64.7 ± 14.5 years. Average follow-up time was 27.5 ± 8.6 months (range 15–36 months). There were no intraoperative complications and no issues with the creation of the descemetorhexis—all descemetorhexis cuts were complete. There were no significant graft detachments and no need for rebubbling. There were no primary or secondary graft failures and all grafts were viable at the final follow-up. BSCVA worsened from 0.41 ± 0.33 logMAR (Snellen equivalent ∼20/50) to 1.37 ± 0.91 logMAR (Snellen equivalent ∼20/460) after PK failure (p = 0.012), and improved significantly after FE-DMEK to 0.34 ± 0.14 logMAR (Snellen equivalent ∼20/45), 0.42 ± 0.12 logMAR (Snellen equivalent ∼20/50), 0.27 ± 0.14 logMAR (Snellen equivalent ∼20/35), and 0.25 ± 0.16 logMAR (Snellen equivalent ∼20/35) at 6 months, 12 months, 24 months, and at final follow-up, respectively (p = 0.013, p = 0.027, p = 0.022, and p = 0.008, respectively). ECD decreased from 2837 ± 229 cells/mm² preoperatively to 1069 ± 413 cells/mm² (61.4% cell-loss rate) and 974 ± 344 cells/mm² (64.8% cell-loss rate) at 12 months and 24 months, respectively (p < 0.001). Cell loss was higher than in historical controls. Conclusions: FE-DMEK was effective in the management of PK graft failure, showing very low detachment and rebubble rates.