Glaucoma Risk Factors and Outcomes Following Boston Keratoprosthesis Type 1 Surgery

PURPOSE: To evaluate glaucoma risk factors and associated outcomes of the Boston keratoprosthesis type I (KPro). DESIGN: Clinical case-control study. METHODS: This is a single-center study of 140 eyes of 118 patients who underwent KPro surgery between October 2008 and March 2017 by a single surgeon. A total of 118 eyes of 118 patients with at least 6 months of follow-up were analyzed to account for intereye correlation. Patients without glaucoma were compared to those diagnosed with glaucoma, which included treatment with intraocular pressure (IOP)-lowering medications or glaucoma surgery. A subgroup analysis compared eyes with pre-KPro glaucoma with those with post-KPro glaucoma. Statistical analysis was performed using univariate and multivariate analyses and Kaplan-Meier survival curves. Main outcome measures were glaucoma diagnosis and progression. Other outcomes included demographics, preoperative diagnosis, best-corrected visual acuity, IOP, cup-to-disc ratio progression and postoperative complications. RESULTS: The mean age at surgery was 60.7 +/- 16.7 years, with a follow-up of 6.9 +/- 3.2 years. De novo KPro glaucoma incidence was 24% (n = 28/118), equivalent to 3.4 cases per 100 eye-years, with onset at 2.1 +/- 2.2 postoperative years. A total of 17 of 118 eyes (14%) did not have glaucoma. Multiple logistic regression showed that high preoperative IOP was a predictor of higher rates of glaucoma development (odds ratio [OR] = 1.538, 95% confidence interval [CI] = 1.030-2.297, P = .035) and progression (OR = 1.450, 95% CI = 1.084-1.937, P = .012). Stromal and endothelial disorders were protective preoperative diagnoses for glaucoma progression after KPro (OR = 0.002, 95% CI = 0.000-0.227, P = .010). A greater proportion of eyes with autoimmune and ocular surface diseases developed de novo glaucoma after KPro compared with other preoperative diagnoses (P < .05). A total of 45% of glaucomatous KPro eyes suffered postoperative glaucoma progression. The mean final best corrected visual acuity of the cohort was 1.76 +/- 1.0, with no difference between eyes with and without glaucoma (P > .05). The rate of serious vision-threatening complications was higher in KPro eyes without glaucoma (77%) than in those with glaucoma (41%, P = .006). CONCLUSIONS: High preoperative IOP signals a higher risk for glaucoma development and progression after KPro surgery. Autoimmune diseases and ocular surface diseases precipitate de novo glaucoma, whereas stromal and endothelial disorders protect against glaucoma progression after KPro. The minority of KPro eyes without glaucoma remain at high risk of complications that can hinder promising visual outcomes. Despite all available treatments and surgical interventions, a majority of eyes will suffer from glaucoma progression, even later during follow-up. ((C) 2021 Elsevier Inc. All rights reserved.)