(P-060) Post-keratoplasty Fungal Keratitis in the Sentinel Distributed Database: Prevalence and Time Trends

Background: The Eye Bank Association of America reported a rise in incidence of post-keratoplasty keratitis between 2007 and 2014, particularly following endothelial versus penetrating keratoplasty. This report was based on a small number of events reported via a passive surveillance system.

Objectives: To compare risk of fungal keratitis following endothelial, penetrating, and anterior lamellar keratoplasty among patients in the Sentinel Distributed Database (SDD).

Methods: We conducted a population-based, retrospective cohort study. Using SDD claims data, we identified patients who had undergone endothelial, penetrating, or anterior lamellar keratoplasty (based on Current Procedural Terminology, Fourth Edition [CPT-4] codes) between January 1, 2008 and September 30, 2020. Patients were required to have continuous health plan enrollment and no fungal keratitis ICD-9/10-CM diagnosis codes in the 90 days prior to keratoplasty. We followed patients for 200 days post- keratoplasty, unless a subsequent keratoplasty occurred, to evaluate evidence of fungal keratitis either as an ICD-9/10-CM diagnosis code or via proxy of observed antifungal treatment of ≥15 days. We assessed the prevalence of fungal keratitis across the three keratoplasty types, overall and by calendar year, and calculated the relative risk of fungal keratitis across keratoplasty types, with endothelial keratoplasty as the referent.

Results: We identified 83,781 endothelial (mean age ± SD: 75 ± 9 years; 62% female), 52,773 penetrating (mean age ± SD: 66 ± 16 years; 53% female), and 6,455 anterior lamellar (mean age ± SD: 62 ± 17 years; 53% female) keratoplasty patients. There was a steady increase in endothelial keratoplasty use from 2009 to 2019 (from 0.38 to 1.81 procedures / 10,000 eligible persons). Use of penetrating and anterior lamellar keratoplasty procedures declined during the same period. Per 1,000 keratoplasty procedures, we observed 2.7, 16.1, and 10.6 fungal keratitis infections following endothelial, penetrating, and anterior lamellar keratoplasty, respectively. Compared to patients with an endothelial keratoplasty, without adjustment for potential confounding, patients with a penetrating keratoplasty had 5.9 (95% CI: 5.2 to 6.6) times the risk of fungal keratitis and patients with an anterior lamellar keratoplasty had 3.9 (95% CI: 3.0 to 5.0) times the risk of fungal keratitis.

Conclusions: We observed an increase in use of endothelial keratoplasty over time, and a lower unadjusted risk of post-keratoplasty fungal keratitis following this procedure compared to penetrating or anterior lamellar keratoplasty.