Impact of Phacoemulsification Combined with XEN Gel Stent Implantation on Corneal Endothelial Cell Density: 2-Year Results.

Corneal integrity has long been a preoccupation of glaucoma surgeons considering glaucoma drainage device surgery or antimetabolite-enhanced trabeculectomy. Despite having demonstrated a good safety profile and significant intraocular pressure-lowering capacities, the impact of XEN gel stents on endothelial density was never specifically investigated. The purpose of this study is to assess the effect of XEN gel stents on central endothelial cell density (ECD) over 24 months. To achieve this, we compared the effect on ECD of combined XEN surgery with that of a standard phacoemulsification procedure. This was an investigator-initiated, retrospective study, conducted at a single tertiary glaucoma center. Patients with primary or secondary open-angle glaucoma who underwent XEN implantation combined with phacoemulsification between January 2015 and June 2016 were retrospectively enrolled. Patients who had undergone standalone phacoemulsification over the same period of time were enrolled to form the control group of this comparative study. The primary outcome measure was the ECD. Patients who had undergone standalone XEN implantation and patients for whom both a baseline and 24-month ECD could not be obtained were excluded from the analysis. Percentages of ECD reductions were calculated for each studied eye, and the mean of ECD reductions was calculated for each group as well as for subgroups. Thirty-two eyes of 23 patients (mean age=76.0±7.9 y, 60% female) underwent standalone phacoemulsification (n=15) or combined XEN surgery (n=17) and had an ECD both at baseline and 24-month postoperatively. Mean baseline ECDs were 2568±491 versus 2379±335 cells/mm, respectively (P=0.21). In the combined XEN surgery group, 58.8% of eyes (n=10) required at least 1 mitomycin C (MMC)-enhanced needling revision to maintain their target intraocular pressure. In the standalone phacoemulsification group, ECD decreased by a mean 14.5%, from 2567.7±491.2 to 2196.1±591.9 cell/mm (P=0.072). In the combined XEN surgery group, ECD decreased by a mean 14.3%, from 2378.8±334.7 to 2039.6±451.1 cell/mm (P=0.018). The difference in percentage reduction of ECD between the 2 groups was not statistically significant (P=0.226). Within the combined XEN surgery group, the ECD decreased by a mean of 15.4% in patients who did not undergo needling revisions and by 13.1% in patients who underwent the MMC-augmented procedure (P=0.485). In the 3 patients who underwent >1 needling revision, a 21.3% reduction in ECD was observed, but the difference was not statistically significant (P=0.653). Neither the time of the first needling (P=0.452), the patients' age (P=0.285), or sex (P=0.308) was statistically associated with ECD loss. The present study demonstrated that the XEN gel implant combined with phacoemulsification produces 24-month ECD loss of a similar magnitude to that observed following standalone phacoemulsification. MMC-augmented needling revisions do not appear to have an impact on ECD.