Meta-Analysis of Postoperative Outcome Parameters Comparing Descemet Membrane Endothelial Keratoplasty Versus Descemet Stripping Automated Endothelial Keratoplasty

Purpose: This meta-analysis compares Descemet membrane endothelial keratoplasty (DMEK) with Descemet stripping automated endothelial keratoplasty (DSAEK) to evaluate their strength and weakness profiles. Design: Meta-analysis. Methods: We performed a meta-analysis and searched the peer-reviewed literature in PubMed, Cochrane Library, Web of Science, and Embase following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Prospective and retrospective trials performing and comparing DMEK and DSAEK were included. Effects were calculated as odds ratios or standardized mean differences. Results: A total of 11 studies with a total of 723 eyes (350 DMEK and 373 DSAEK) were included. No significant difference was found regarding the total detachment rate, graft failure, and rejection. One of 340 eyes undergoing DMEK showed total detachment and 5 of 363 eyes undergoing DSAEK showed total detachment (P = 0.28). Six of 280 eyes undergoing DMEK showed graft failure; 1 of 313 eyes undergoing DSAEK developed this complication (P = 0.18). No rejection was observed in 158 eyes undergoing DMEK; 4 cases of rejection occurred in 196 eyes undergoing DSAEK (P = 0.37). No significant difference was found regarding endothelial cell loss (P = 0.48). There was a significantly higher partial detachment rate with DMEK: 88 of 340 eyes undergoing DMEK showed partial detachment; 16 of 363 eyes undergoing DSAEK showed this complication (P < 0.00001). DMEK was superior in best-corrected visual acuity after 6 months (P < 0.001), subjective evaluation of visual acuity (P = 0.001), patient satisfaction (P < 0.001), and was the method preferred by patients (P = 0.001). Conclusions: DMEK and DSAEK have a similar complication profile. However, the superiority in the visual outcome and patient satisfaction makes DMEK the preferred option for most patients.