Comparative Outcomes Of Single Vs. Multiple Cyanoacrylate Glue Applications In The Management Of Corneal Thinning And Perforations

Cyanoacrylate glue is commonly used to treat corneal thinning and perforations; however, some cases necessitate multiple applications due to persistent leakage, displaced glue, or progressive thinning. This study aims to compare outcomes between patients treated with a single corneal glue patch and those requiring multiple glue applications.

We conducted a single-center, retrospective review of patients with corneal thinning and perforations from 2006 to 2024 at the Toronto Western Hospital.

Eligible patients were treated with at least one corneal glue patch for thinning or perforation during the follow-up period. The main outcome measures were the need for tectonic penetrating keratoplasty (PKP) and final best corrected visual acuity (BCVA). Associations between the number of glue applications with baseline characteristics (age, sex, initial BCVA), perforation characteristics (indication, cause, location, and area), and final outcomes were analyzed using univariable and multivariable logistic regression models, adjusted for age, sex, and eye laterality on Stata 17.0.

Overall, 189 patients (median age 69.0 years; 42% female) were included, with 116 (61%) requiring a single glue patch and 73 (39%) receiving multiple (mean 2.3 ± 0.6), over a median follow-up of 4.4 months. Baseline characteristics were similar between both groups. Central perforations were more likely to require multiple glue applications compared to peripheral perforations (OR=2.59, 95%CI=[1.11-6.02], p=0.027). No difference was observed in final BCVA between groups (median: 2.3 logMAR [single] vs. 2.0 logMAR [multiple]; p=0.995). Patients receiving multiple glue applications (47%, n=34/73) were more likely to require PKP (OR=2.70, 95%CI=1.42-5.15, p=0.003) compared to those treated with a single patch (26%, n=30/116). 

Multiple glue applications were more frequently required in patients with central perforations and were associated with a greater likelihood of needing a tectonic PKP. Given that there was no significant difference in final BCVA between groups, patients should be appropriately counselled about their guarded prognosis.