Horizontal vs Vertical Scleral Tunnels in Flanged IOL Fixation (HORVER)

Background: The management of aphakia, crystalline lens luxation, and intraocular lens (IOL) luxation in the absence of adequate capsular support remains challenging. Traditional surgical options include anterior chamber IOLs, iris-fixated IOLs, and scleral-sutured posterior chamber IOLs, each carrying significant long-term risks. In 2017, Yamane and colleagues introduced a sutureless flanged intrascleral IOL fixation technique that rapidly gained worldwide acceptance. A critical yet under-investigated surgical variable is the orientation of the scleral tunnels relative to the limbus. Some surgeons create tunnels parallel to the limbus (horizontal/tangential), others perpendicular (vertical/radial), based on personal preference rather than high-level evidence.

Objective: To evaluate whether horizontal versus vertical scleral tunnel orientation is associated with differing 2-year outcomes in flanged intrascleral IOL fixation.

Study Design: This is a retrospective observational cohort study conducted at a single tertiary referral centre.

Participants: A total of 107 eyes (43 vertical tunnels, 64 horizontal tunnels) with complete 24-month follow-up are included. Surgeries are performed by two experienced vitreoretinal surgeons.

Interventions: All procedures follow the original Yamane flanged intrascleral fixation method with intentional variation in tunnel orientation only. The horizontal tunnel group receives tunnels parallel to the limbus (tangential), while the vertical tunnel group receives tunnels perpendicular to the limbus (radial). A three-piece foldable acrylic IOL (Alcon AcrySof MA60AC) is used in all eyes.

Outcome Measures: The primary outcome is the proportion of eyes achieving mean absolute error (MAE) within ±0.50D of target refraction at 2 years. Secondary outcomes include MAE at 2 years, IOL tilt measured using anterior segment optical coherence tomography (AS-OCT), iris capture, conjunctival erosion, cystoid macular oedema, epiretinal membrane, retinal detachment, vitreous haemorrhage, hypotony, corneal oedema, early IOP elevation, best-corrected visual acuity (BCVA), and vector astigmatism analysis using the Alpins method.

Statistical Analysis: Propensity score matching (1:1) will be performed to address selection bias. Multivariate logistic regression and a multilevel mixed-effects model with surgeon as random effect will be used to adjust for confounders including surgical period, age, and baseline refraction. For the primary outcome, a single pre-specified comparison will be made with a two-tailed significance level of α = 0.05. For secondary outcomes, the False Discovery Rate (FDR) using the Benjamini-Hochberg procedure (q = 0.05) will be applied to control for Type I error inflation.

Timeline: Data were collected from patients who underwent surgery between March 1, 2022, and March 1, 2025. Minimum follow-up is 24 months, with final data collection completed in March 2025.