Outcome of deep anterior lamellar keratoplasty with intraoperative Descemet membrane perforation

Abstract

Purpose The aim of this study is to report the outcome of deep anterior lamellar keratoplasty (DALK) with intraoperative Descemet membrane (DM) perforation. Patients and methods A retrospective case series of all DALK cases performed from October 2015 to October 2021 at Alexandria Main University Hospital, Alexandria, Egypt, with and without intraoperative DM perforation was evaluated. The study included 110 eyes of 108 patients who met the inclusion criteria, had complete records, and at least 12 months of follow-up. They were further divided into four groups; DALK with intact DM, DALK with DM microperforation, DALK with DM macroperforation, and DALK converted to penetrating keratoplasty (PK). The best corrected visual acuity (BCVA), graft rejection, endothelial cell (EC) density, and complications were compared between the four groups. Results The most common surgical step during which DM perforation occurred was deep lamellar dissection (41 cases, 70.7%). At 12-month postoperatively, there was no statistically significant difference (P=0.087) in BCVA between the four studied groups (mean=0.33±0.19, 0.39±0.23, 0.45±0.39, and 0.32±0.34 LogMAR, respectively). The DALK groups had significantly less rate of graft rejection (P=0.009) and complications (P=0.006) compared to the PK group. The EC density was statistically significantly higher in the eyes with intact DM compared to the eyes with DM perforation (P<0.001) and to the eyes converted to PK (P<0.001). Interestingly, there was no statistically significant difference between the PK group and the DM macroperforation group in the EC density (P=0.243). Conclusion DALK, with or without DM perforation, has equivalent visual outcome, less graft rejection, and less complications compared to the cases converted to PK. The EC density in DALK with intact DM or DM microperforation was significantly higher than in DALK with DM macroperforation and DALK converted to PK.