Anterior chamber air bubble and graft adhesion in Descemet membrane endothelial keratoplasty (DMEK): Is bigger always better?

Abstract

Purpose: To determine the influence of the size of the air bubble which is used in DMEK surgery on the rate of post-surgical graft detachment and need for re-bubbling.

Methods: Single-center, retrospective, consecutive case series of 74 eyes of 74 patients undergoing DMEK and fulfilling the inclusion criteria concerning the size of the air bubble at the end of surgery. Based on the medical records patients were divided into two groups, of which one group (n = 38) received an air bubble with a volume of 50% of the anterior chamber volume and another group (n = 36) an 80% air bubble. Patients who did not comply with instructions to remain in supine position until complete resorption of anterior chamber air or cases in which difficulties in graft preparation (e.g. radial breaks) occurred were excluded from data analysis.

Results: There was no significant difference between both groups regarding parameters such as mean patient age (50%: 67 ± 11 yrs, 80%: 71 ± 9 yrs) or mean donor endothelial cell count (50%: 2551 ± 207 cells/mm², 80%: 2482 ± 145 cells/mm). All grafts were prepared successfully from organ-cultured corneoscleral buttons. In the 50% air bubble group, 13 of 38 eyes (34.2%) needed one re-bubbling due to graft detachment of a least 3 clock hours. In the 80% air bubble group, 7 of 36 eyes (19.4%) needed one re-bubbling. The groups did not differ significantly, using Pearson-χ²-test (p = 0.15).

Conclusions: Air or gas in the anterior chamber might potentially be damaging to the endothelium. While an increased air or gas filling of the anterior chamber at the end of DMEK surgery is currently advised by most groups, our study was set out to explore whether a reduction of anterior chamber air bubble is safe and serves the purpose of graft adhesion. Indeed, in this series we did not detect any statistically significant difference regarding re-bubbling rates between the reduced and the standard size air bubble. Other confounding factors for detachment such as extent of scroll formation should be evaluated in a larger series.