| Title: | Fundus Stretch Index: a centile-based retinal measure of myopia |
| Journal: | Ophthalmology Science |
| Published: | 1 Mar 2026 |
| DOI: | https://doi.org/10.1016/j.xops.2026.101162 |
| URL: | http://dx.doi.org/10.1016/j.xops.2026.101162 |
Publication 17569
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
Objective To propose and evaluate the risk-stratification potential of Fundus Stretch Index (FunSI), a centile-based retinal metric that quantifies the geometric deviation of the fundus from what is expected for a given spherical equivalent refraction (SER). Design Population-based cohort study. Participants Phakic eyes of 25,222 adults (SER: 0D to -12D) with adequate-quality fundus photographs at baseline (2009-2010), individually linked to routinely collected health records until 2022. Methods Quantile regression was used to estimate the 5th to 95th centiles of 10 dimensionless fundus imaging features (e.g. arterial/venous tortuosity, disc tilt), conditioning on SER. FunSI was computed as the sum of an eye's SER-specific centile position across all imaging features, normalised to 0-1, such that an eye consistently ranked in the worst centile (5th or 95th, depending on the direction of change as myopia increased) for its SER would have a value of 1. Cox regression was used to test the association between baseline FunSI and the risk of: rhegmatogenous retinal detachment (RD), adjusting for baseline SER, age, sex, ethnicity, Townsend deprivation index, diabetes, hypertension and ocular trauma; and primary open-angle glaucoma (POAG), adjusting for the same baseline covariates (excluding ocular trauma) plus intraocular pressure and corneal hysteresis. Main Outcomes and Measures Rhegmatogenous RD and POAG onset. Results A total of 25,030 and 24,835 adults aged 40-69 years without any prior history of RD/breaks and glaucoma of any subtype were analysed, respectively. The event rates (new cases per 10,000 person-years) were 4.9 for rhegmatogenous RD and 15.0 for POAG. Higher baseline FunSI was associated with an increased risk of developing rhegmatogenous RD (adjusted hazard ratio per 1 SD [0.09], 1.26; 95% confidence intervals, 1.08-1.48; P=.004) and POAG (1.11; 1.01-1.22; P=.03). The addition of FunSI led to a much clearer improvement in the concordance index for the rhegmatogenous RD model compared with the POAG model. Conclusions FunSI is an explainable approach to characterising myopia at the retinal level. Two individuals may share similar baseline risk factors such as SER, but an overall geometric difference in their fundi, as quantified by FunSI, may reveal a difference in their long-term risk of sight-threatening diseases.</p>
1 Author
- Fabian Yii
1 Application
| Application ID | Title |
|---|---|
| 90655 | Investigating retinal features and other health characteristics that are predictive of refractive error and risk of pathologic myopia. |
Enabling scientific discoveries that improve human health