Effects of Postoperative Intravitreal Injections on Outcomes of Traditional Glaucoma Surgery in Patients with Preoperative Intravitreal Injections

PURPOSE To compare outcomes of glaucoma drainage device (GDD) implantation and trabeculectomies with and without postoperative intravitreal injections (IVIs) in glaucoma patients with a history of preoperative IVIs. DESIGN Retrospective cohort study. PARTICIPANTS A total of 133 eyes of 133 glaucoma patients who underwent GDD implantation or trabeculectomy with at least 1 IVI preoperatively between January 2005 and October 2020 at Massachusetts Eye and Ear. METHODS Chart review of glaucoma patients with traditional glaucoma surgery and at least 1 IVI before surgery. All statistical analyses were conducted with R statistical programming software. MAIN OUTCOME MEASURES Intraocular pressure (IOP), medication burden, best-corrected visual acuity (BCVA), Kaplan-Meier success rates, adjusted hazard ratios (HRs), and complications. RESULTS Baseline demographics were similar between the groups with and without postoperative IVIs. The group with postoperative IVIs had a higher proportion of diabetic retinopathy and retinal vascular occlusions than the group without postoperative IVIs, which had more eyes with age-related macular degeneration. Intraocular pressure, medication burden, and visual acuity were similar between groups at all time points except for IOP at 6 weeks, which was lower in the group with postoperative IVIs. The group with postoperative IVIs had significantly more preoperative IVIs than the group without postoperative IVIs (6.6 vs. 3.3, P = 0.017). For success defined as IOP reduction ≥ 20% with 5 < IOP ≤ 21 mmHg, Kaplan-Meier analyses demonstrated similar success rates between groups with and without IVIs. When stratified by the number of IVIs, success rates for the group with 7 or more IVIs were significantly higher than the success rates for the group with 0-6 IVIs (P = 0.005). Each additional postoperative IVI resulted in a 7.2% decrease in the hazard of failure to achieve our stated success criteria. With regard to late complications, the group with postoperative IVIs had a higher incidence of vitreous hemorrhage (18.5% vs. 3.2%, P = 0.039) than the group without postoperative IVIs. CONCLUSIONS A higher number of postoperative IVIs, specifically 7 or more IVIs, may be associated with improved success rates of traditional glaucoma surgery in glaucoma patients who received IVIs before surgery.