Impact of Intraoperative Descemet Membrane Perforations on Deep Anterior Lamellar Keratoplasty Outcomes

Purpose: To analyse the outcomes of deep anterior lamellar keratoplasty (DALK) in cases with intraoperative Descemet membrane (DM) perforation. Methods: This is a literature review reporting outcomes of DALK with DM perforation. Studies where DALK was performed in the event of intraoperative DM perforation were included. Studies that did not separate analysis between those with and without DM perforation were excluded. The primary outcome was best-corrected distance visual acuity (BCVA). Secondary outcomes were endothelial cell density (ECD), graft survival, rejection rates, double anterior chamber and conversion to penetrating keratoplasty (PK). Data from the included studies were collated to compare the outcomes of DALK with intraoperative DM perforation vs. DALK without DM perforation. Results: Eleven retrospective case series (357 eyes) were included. DM perforations were classified as micro- (n = 236) or macroperforations (n = 106). Mean weighted preoperative BCVA was 1.11 ± 0.36 logMAR and 1.13 ± 0.52 logMAR in perforation and nonperforation groups, respectively (p=0.53), improving to 0.35 ± 0.37 logMAR and 0.39 ± 0.07 logMAR at 12 months postoperatively (p=0.02). Graft rejection rates were 1.25% and 1.6% in the perforated and nonperforated groups, respectively, and primary graft failure rates were 4% and 3.74%, respectively. The mean postoperative ECD was 1662.41 ± 319.16 cells/mm² in the perforation group. Amongst those cases with DM perforation, double anterior chamber requiring rebubbling occurred in 22.4% of cases, and conversion to PK was 4.23%. Conclusion: DALK can achieve comparable long-term outcomes in the presence of DM perforation. Micro- and some macroperforations can often be managed without conversion to PK, with good long-term outcomes.