Efficacy, Safety, and Risk Factors for Failure of Standalone Ab Interno Gelatin Microstent Implantation versus Standalone Trabeculectomy

Purpose To compare the efficacy, safety, and risk factors for failure of standalone ab interno gelatin microstent implantation with mitomycin C (MMC) versus trabeculectomy with MMC. Design International, multicenter, retrospective interventional cohort study. Participants Three hundred fifty-four eyes of 293 patients (185 microstent and 169 trabeculectomy) with no prior incisional surgery. Methods Consecutive eyes with uncontrolled glaucoma underwent microstent or trabeculectomy surgery from January 1, 2011 through July 31, 2015 at 4 academic ophthalmology centers: Toronto, Canada; Frankfurt, Germany; Salzburg, Austria; and Leuven, Belgium. Main Outcome Measures Primary outcome measure was hazard ratio (HR) of failure, with failure defined as 2 consecutive intraocular pressure (IOP) readings of 17 mmHg without glaucoma medications (complete success) at least 1 month after surgery despite in-clinic interventions (including needling). Secondary outcome measures included IOP thresholds of 6 to 14 mmHg and 6 to 21 mmHg and same thresholds allowing for medications (qualified success), interventions, complications, and reoperations. Results Baseline characteristics were similar, except more men (56% vs. 43%), younger patients (average, by 3 years), better preoperative visual acuity (22% vs. 32% with 0.4 logarithm of the minimum angle of resolution vision or worse), and more trabeculoplasty (52% vs. 30%) among microstent eyes. The adjusted HR of failure of the microstent relative to trabeculectomy was 1.2 (95% confidence interval [CI], 0.7–2.0) for complete success and 1.3 (95% CI, 0.6–2.8) for qualified success, and similar for other outcomes. Time to 25% failure was 11.2 months (95% CI, 6.9–16.1 months) and 10.6 months (95% CI, 6.8–16.2 months) for complete success and 30.3 months (95% CI, 19.0–∞ months) and 33.3 months (95% CI, 25.7–46.2 months) for qualified success. Overall, white ethnicity was associated with decreased risk of failure (adjusted HR, 0.49; 95% CI, 0.25–0.96), and diabetes was associated with increased risk of failure (adjusted HR, 4.21; 95% CI, 2.10–8.45). There were 117 and 165 distinct interventions: 43% and 31% underwent needling, respectively, and 50% of trabeculectomy eyes underwent laser suture lysis. There were 22 and 30 distinct complications, although most were transient. Ten percent and 5% underwent reoperation ( P  = 0.11). Conclusions There was no detectable difference in risk of failure and safety profiles between standalone ab interno microstent with MMC and trabeculectomy with MMC.