Abstract Purpose To compare one‑year postoperative outcomes between PreserFlo MicroShunt (PMS) and trabeculectomy (TLE) as primary filtration surgery in Japanese patients with open‑angle glaucoma using propensity score matching. Study design Retrospective, single‑center observational cohort study. Methods Patients with open‑angle glaucoma who underwent stand‑alone PMS implantation or TLE at Kobe University Hospital were included. After applying exclusion criteria, 102 eyes in each of 2 groups were matched 1:1 using propensity scores based on age, sex, lens status, glaucoma subtype, preoperative intraocular pressure (IOP), number of glaucoma medications, and best‑corrected visual acuity (BCVA). The primary outcome was IOP at 12 months postoperatively. Secondary outcomes included longitudinal IOP changes, medication use, BCVA, corneal endothelial cell density (ECD), surgical success rates, postoperative complications, and additional surgery. Mixed‑effects models and Kaplan–Meier analyses were performed. Results At 12 months, mean IOP was significantly lower in the TLE group than in the PMS group (12.3 vs 14.1 mmHg, P = 0.042). Both groups showed significant IOP reduction from baseline. The proportion of medication‑free eyes did not differ between groups. BCVA worsened transiently after surgery and partially recovered but remained worse than baseline at 12 months, with no between‑group differences. ECD decreased modestly in both groups without significant differences. Surgical success rates were comparable across all criteria. Postoperative complications, particularly hypotony‑related events, were more frequent in the TLE group, and persistent hypotony was observed only after TLE. Conclusions PMS demonstrated a more favorable safety profile with fewer hypotony‑related complications, whereas TLE achieved greater IOP reduction at 12 months. Surgical success rates were comparable. PMS may represent a less invasive option, while TLE may be preferable when maximal IOP lowering is required.
Katakami et al. (Sat,) studied this question.