AIMS: To evaluate how different methods of managing multiple visits within predefined time windows affect intraocular pressure (IOP)-based success rates in glaucoma surgery studies. METHODS: We applied literature-based high IOP failure criteria to two cohorts of 934 and 1760 eyes undergoing trabeculectomy and deep sclerectomy (DS) with median follow-up of 41.4 months and 45.4 months, respectively. Failure was defined by IOP thresholds, loss of light perception, hypotony requiring revision or additional IOP-lowering surgery. Visits were grouped into guideline-based windows and six visit-managing strategies were applied to all visits, mean, lowest, highest, median and closest IOP to the recommended time point. We calculated Kaplan-Meier success rates for each visit-managing strategy. Visual field (VF) analysis was conducted on patients in the trabeculectomy cohort with ≥4 VFs in ≥2 years post-surgery. RESULTS: For the 21 mm Hg threshold, 5-year success was highest with the lowest IOP (trabeculectomy: 54.8%; DS: 74.5%), followed by closest IOP (trabeculectomy: 46.7%; DS: 67.6%), the median (trabeculectomy: 46.9%; DS: 69.1%) and mean IOP (trabeculectomy: 46.3%; DS: 68.6%). Success rates were lower with peak IOP (trabeculectomy: 39.3%; DS: 60.4%) and all visits IOP (trabeculectomy: 38.8%; DS: 61.0%). In the VF subset, eyes classified as failures demonstrated significantly faster mean deviation (MD) progression than those classified as successes although substantial overlap in the distribution of MD rates persisted between groups under every strategy. CONCLUSIONS: Visit-managing strategies influence reported success rates. None of the evaluated approaches achieved a clear separation in VF progression rates, underscoring the inherent limitations of IOP-threshold-based classifications.
Rabiolo et al. (Wed,) studied this question.