Abstract
Purpose :
Descemet Membrane Endothelial Keratoplasty (DMEK) provides excellent visual acuity and decreased tissue rejection compared to other forms of corneal transplantation. However, the thinness and fragility of DMEK tissue makes it more technically challenging with increased risks of failure with poor handling. This study was performed to determine the effect the learning curve has on DMEK outcomes.
Methods :
A single-center, retrospective review of 232 DMEK surgeries for Fuchs’ Dystrophy completed from start to finish by three corneal fellows was conducted. Using a standardized surgical technique and pre-loaded DMEK tissue, the learning curve for each individual fellow was analyzed by comparing the surgical outcomes from the beginning and the end of fellowship. Consecutive surgeries in groups of 10 were compared to the last 20 consecutive surgeries of the fellowship for tissue unscroll time, rebubble rate, and 6-month endothelial cell loss (ECL). A novice DMEK surgeon was considered proficient when the average outcomes for a group of 10 consecutive surgeries was not statistically different than the last 20 surgeries. Students t-test, Mann-Whitney U test, and a Fisher’s exact test was used for statistical analysis.
Results :
Fellow 1 (n=83 total DMEKs) had no statistical difference between the first 10 DMEK surgeries and the last 20 DMEK surgeries for rebubble rate (0% vs 15%, P=0.53) tissue unscroll time (5.5 vs 4.5 min, P=0.067), and 6-month ECL (32.2% vs 32.2%, P=0.53). Fellow 2 (n=65) did not have a statistically different rebubble rate (20% vs 10%, P=0.85) or 6-month ECL (22.9% vs 18.4%, P=0.31), however the unscroll time was significantly longer in the first 10 (4.7 vs 2.7 min, P=0.035). The difference resolved by the second 10 DMEKs. Fellow 3 (n=84) had similar 6-month ECLs (29.3% vs 28.1%, P=0.66) in the first 10 DMEKs, however significantly higher rebubble rates (30% vs 0%, P=0.030) and unscroll time (4.1 vs 1.9 min P=0.004). This difference was no longer significant by the second 10 DMEKs in all 3 categories.
Conclusions :
By the second 10 surgeries, all three fellows achieved a similar rebubble rate, 6 month ECL, and tissue unscroll time as at the end of their 1 year training. 6 month ECL was the same in all three fellows at the beginning and the end of their training. With adequate supervision and training, DMEK surgery can be taught safely to fellows without significant compromise in patient outcomes.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.