Association between Iris Surface Features and Iris Thickness in Asian Eyes

Abstract

Purpose: To describe a grading system to assess iris surface features in Asian eyes and to examine the association between iris surface features and iris thickness measured by anterior segment optical coherence tomography (AS-OCT).

Methods: We recruited 50 subjects from the Singapore Malay Eye Study (SiMES), a population-based survey of Singaporean Malays aged 40-80 years. Iris surface features including iris crypts, contraction furrows, and iris color were graded from photographs taken using slit lamp digital camera with standard protocol. We proposed a grading system to assess iris surface features including iris crypts (grade 1-5, by their number and size), furrows (grade 1-3, by their number and circumferential extent), and iris color (grade 1-5, comparing with standard photos, with higher grade denoting darker iris). Intra- and inter-grader agreements between two graders were assessed by weighted kappa (Κ_w). Vertical and horizontal cross-sections of anterior chamber were imaged using AS-OCT. Iris thickness at 750µm (IT₇₅₀) and 2000µm (IT₂₀₀₀) from the scleral spur and maximum central iris thickness (IT_CM) were measured using Zhongshan Angle Assessment Program. Associations between iris features and average thickness from the four quadrants were assessed by Spearman’s rank correlation (ρ) and generalized estimating equations to control for inter-eye correlation.

Results: Of the 50 subjects, 75 eyes had complete AS-OCT data and were gradable for iris crypts and color; 64 eyes were graded for furrows. The grading scheme for iris surface features showed good to optimal inter-grader (Crypt Κ_w=0.775, Furrow Κ_w=0.836, Color Κ_w=0.718) and intra-grader (Crypt Κ_w=0.919, Furrow Κ_w=0.901, Color Κ_w=0.925) agreements. Higher crypt grade was significantly correlated with thinner IT₂₀₀₀ (ρ=-0.342, p=0.022). More extensive furrows were correlated with thicker IT₇₅₀ (ρ=0.453, p=0.008) and IT_CM (ρ=0.250, p=0.031). We did not observe any significant correlations between iris color and iris thickness (all p>0.05).

Conclusions: Iris surface features are clinically observable. Our results show that several iris surface features can be graded from slit lamp photographs and are associated with iris thickness. Irises with larger and more crypts are thinner centrally, and irises with more extensive furrows are thicker peripherally. These findings may have important implications in assessing the risk of angle closure based on iris surface features.

Keywords: 571 iris • 550 imaging/image analysis: clinical • 731 texture