Text

Objective: Randomized controlled trials (RCTs) form the cornerstone of evidence-based glaucoma care. However, the robustness of statistically significant results in these trials is often unclear. The fragility index (FI) quantifies how easy it is to reverse a significant result, while the fragility quotient (FQ) adjusts this number to account for trial size. This study aims to assess the statistical fragility of primary endpoints in landmark glaucoma RCTs.

Methods: Electronic databases were searched from inception until 01 September 2025. Landmark RCTs were defined as those enrolling at least 300 participants or frequently cited in clinical guidelines; only trials with statistically significant binary primary outcomes were included, resulting in 10 studies eligible for the primary analysis. The FI and FQ were calculated using methods described by Walsh et al., focusing on the most statistically significant binary endpoint per trial. Trials lacking raw binary data or containing only continuous outcomes were excluded. Secondary analyses were conducted on key binary outcomes in the nine studies where the primary endpoint was non-binary but relevant secondary binary endpoints were reported.

Results: Ten eligible RCTs were identified, encompassing trabecular bypass, canaloplasty, subconjunctival microshunt, and goniotomy-based procedures. The mean sample size was 526 (IQR: 56–1636), with a mean number lost to follow-up of 133 (IQR: 0 – 412). The average FI was 18.1 (IQR: 2–63), and the mean FQ was 0.04 (IQR: 0.002–0.245), indicating that a small proportion (≈4%) of participants could alter statistical significance. No consistent associations were observed between FI and device type, comparator, or study region.

Conclusion: Although landmark glaucoma RCTs appear robust (mean FI: 18.1), more than 60% of trials demonstrate a fragility index below 9 and a fragility quotient under 0.04. This indicates that small shifts in outcome events may be sufficient to reverse statistical significance, and potential attrition bias may further compromise reliability. These findings highlight the importance of incorporating the fragility index and fragility quotient when interpreting glaucoma RCTs to support more reliable evidence-based clinical decision-making.