Noninferiority of Microhook to Trabectome

Purpose To elucidate the non-inferiority of ab interno microhook trabeculotomy (μTLO) using a recently developed reusable stainless spatula-type microhook device to incise the trabecular meshwork to trabectome (TOM) in terms of the 1-year postoperative outcomes of Japanese patients with glaucoma by means of propensity score analyses. Design A multicenter retrospective cohort study. Participants We enrolled 553 and 392 patients who underwent TOM and μTLO, respectively, between January 2014 and March 2020 at 10 facilities. Methods Logistic regression analysis was conducted to calculate the propensity score, which indicates the likelihood of treatment assignment (TOM or μTLO). We set the following factors as outcome-related covariates: age, sex, facility, glaucoma disease types, preoperative intraocular pressure (IOP), glaucoma drug score, mean deviation of the Humphrey visual field test and antithrombotic drug use, the presence or absence of combined cataract surgery, and incision range of trabecular meshwork (one or two quadrants). We analyzed four different methods (matching, inverse probability of treatment weighting (IPTW), stratification, and regression adjustment) using the propensity score. We set 15% as the non-inferiority margin based on previous trabectome meta-analysis results. Main outcome measures The primary outcome was surgical success at 1 year postoperatively. We defined surgical success as satisfying all three criteria: (1) IOP within 5–21 mmHg; (2) IOP reduction of ≥20% from preoperative IOP; and (3) no additional glaucoma surgery. Results The 95% confidence interval of risk difference of surgical failure in μTLO in reference to TOM was −12.1 to +9.5% in matching, −12.7 to +11.1% in IPTW, −12.2 to +7.0 in stratification, and −9.7 to +8.1% in regression adjustment, all of which fell within the predetermined non-inferiority margin of 15%. Conclusions Surgical success of μTLO at 1-year postoperatively was not inferior to that of TOM.