Endothelial Cell Loss 1 Year After Successful DMEK in the Diabetes Endothelial Keratoplasty Study: A Randomized...

The effect of cornea donor diabetes on endothelial cell density (ECD) and morphometry after Descemet membrane endothelial keratoplasty (DMEK) is not known.

To determine whether endothelial cell loss (ECL) and morphometric changes 1 year after successful DMEK are related to cornea donor's diabetes status.

This was a multicenter, double-masked, randomized clinical trial conducted from February 2022 to July 2024 at 28 US clinical sites (46 surgeons) and 13 eye banks. Included in the trial were the eyes of recipients (some of whom received tissue from donors without diabetes and others who received tissue from donors with diabetes) who underwent successful DMEK, primarily for Fuchs endothelial corneal dystrophy, and had at least 1 analyzable postoperative endothelial image.

DMEK performed with a cornea from a donor without or with diabetes, assigned using a minimization procedure to achieve an approximate 2:1 distribution.

ECD, ECL, coefficient of variation in cell area (CV), and percentage of hexagonal cells (HEX) at 1 year from eye bank and postoperative specular central endothelial images.

A total 1274 eyes of 982 recipients (mean [SD] age, 70 [8] years; 569 female [57.9%]; 816 [64.1%] with tissue from donors without diabetes and 458 [35.9%] with tissue from donors with diabetes) were included in the study. Preoperatively, mean (SD) central ECD in tissue from donors without diabetes and with diabetes were 2676 (290) cells/mm2 and 2671 (286) cells/mm2, respectively. At 1 year, mean (SD) ECL was 28.3% (16.1%) and 28.0% (17.0%), in the donor groups without and with diabetes, respectively (adjusted mean difference = -0.4%; 95% CI, -2.3% to 1.4%), resulting in a mean (SD) 1-year ECD of 1927 (498) cells/mm2 and 1920 (496) cells/mm2, respectively (adjusted mean difference = 10 cells/mm2; 95% CI, -36 to 56; P = .95). No difference in ECD at 1 year associated with diabetes severity was noted (P = .97). Mean (SD) CV did not differ at 1 year between the 2 groups of eyes (31.5% [4.1%] vs 31.4% [4.1%]; adjusted mean difference = -0.4%; 95% CI, -0.9% to 0.1%; P = .51), and mean (SD) HEX did not differ at 1 year between the 2 groups of eyes (57.7% [5.8%] vs 57.2% [5.8%]; adjusted mean difference = 0.1%, 95% CI, -0.8% to 0.9%; P = .33).

This randomized clinical trial found that ECL and morphometry 1 year after DMEK were not affected by cornea donor diabetes status. With comparable 1-year graft success with tissue from donors with and without diabetes demonstrated in this trial, these findings support the use of corneas from donors with diabetes for endothelial keratoplasty procedures.

ClinicalTrials.gov Identifier: NCT05134480.