Abstract
Purpose :
Disease activity in anterior uveitis, the most common manifestation of childhood disease, is currently assessed using the slit-lamp based SUN scale, which is open to interobserver variability, and not validated for paediatric use. We undertook a pilot study of anterior segment OCT (ASOCT) capture of active childhood anterior uveitis.
Methods :
Observational prospective study involving children with and without uveitis, with acquisition of Optovue Avanti anterior segment scans (8 line asterisk formation centred on corneal apex) per eye) following informed consent. Inter and intra-observer reliability of manual counting of acquired images (Bland-Altman limits of agreement, BA LoA), and sensitivity and specificity of ASOCT detection of active inflammation assessed.
Results :
A total of 26 children aged 3yrs to 15yrs (median 8yrs) underwent imaging, including 19 children with a known diagnosis of uveitis, of whom 12 had active chronic anterior inflammation as confirmed by two examiners at the slit lamp. Time taken to acquire images ranged from 22mins to 1.5mins per child, with patient acceptability scores for acquisition process consistently >85/100 on visual analogue scale. Mean intraobserver image count agreement -0.3(95%CI -0.2/0.2), BA LoA -1.1(-1.4/-0.8) to 1.0(0.7/1.4). Interobserver agreement -0.5(-0.9/-0.2), BA LoA -2.5(-3.1/-1.9) to 1.5(0.8/2.1). Sensitivity of ASOCT manual image cell count for diagnosis of active inflammation 92%(62%-99%), specificity 86%(58%-98%), negative predictive value 92%(65%-99%).
Conclusions :
Non-contact, high-resolution ophthalmic imaging holds the promise of transforming paediatric practice. Further work is needed to determine the analytic validity, reproducibility, and concurrent and predictive clinical validity of manual and automated anterior segment OCT quantification of active inflammation.
This abstract was presented at the 2019 ARVO Imaging in the Eye Conference, held in Vancouver, Canada, April 26-27, 2019.