Risk of Uveitis Following COVID-19 Diagnosis

Background: Several case reports/series have documented instances of uveitis occurring after COVID-19 infection. Limited conclusions can be drawn from these findings given the small study sizes and lack of robust design. Understanding the risk of uveitis after COVID-19 is important to better contextualize reports of uveitis as a potential adverse event related to ophthalmic treatments.

Objectives: To compare the relative incidence of uveitis after COVID-19 diagnosis to an unexposed control period prior to COVID-19 diagnosis within the same patient.

Methods: This self-controlled case series study used open and closed healthcare claims from HealthVerity. The study design was selected given the acute and rare onset of uveitis, and ability to control for individual level time invariant risk factors.
Data from patients ≥ 18 years old with their first COVID-19 diagnosis (index date) identified between December 1, 2019 to June 15, 2022 were included. Exposed (index date to 45 days thereafter), control (60 days prior to index date to 15 days prior to index date), and washout (365 days prior to control periods) were identified for each patient. Patients were required to have continuous enrollment throughout these periods. Only patients with incident uveitis during either the exposed or control period were included. The incidence rate ratio (IRR) of uveitis in the exposed versus control period was estimated using conditional Poisson regression. Subgroup analyses were conducted by demographic and clinical characteristics. Sensitivity analyses were performed removing day zero from the exposed time period and reducing the at-risk period to 30 days.

Results: Overall, 2,144 patients were included. Median(IQR) age was 50(34,61) years and 64.4% of patients were female. The IRR comparing uveitis risk during the exposed period versus the control period was 1.04 (95% CI: 0.96, 1.14). There was little difference in IRR estimates when stratified by age, type of uveitis, or presence of baseline systemic disorder. Females had an increased risk of uveitis during the exposed period [females: IRR =1.13 (95% CI: 1.01, 1.25); males: IRR =0.91 (95% CI: 0.79, 1.05)].
After day 0 was removed from the exposed period, the IRR was 0.99 (95% CI: 0.90, 1.07) for all patients, and 1.05 (95% CI: 0.94, 1.17) among females. A total of 84 (7.7%) of the uveitis cases during the exposed period were identified on day 0. Estimates remained similar when reducing the at-risk time to 30 days.

Conclusions: COVID-19 diagnosis was not associated with an increased risk of uveitis in the overall population. Findings from sensitivity analyses removing day 0 suggest the elevated risk observed among females is due to detection bias, although further study is warranted.