Abstract
AIMS: To conduct the first cross-sectional epidemiological investigation of pathologic myopia (PM) in UK adults with high myopia. METHODS: Fundus photographs of 3024 highly myopic eyes (spherical equivalent refraction (SER) ≤-5.00D) from 2000 randomly sampled adults (aged 40-70 years) in the UK Biobank were double graded by an ophthalmic reading centre using the Meta-analysis for Pathologic Myopia framework. Adjudication was performed by one of two retinal specialists. Multivariable mixed-effects logistic regression was used to explore potential risk factors and fundus biomarkers-initially adjusting for SER, age and sex, before including these and other variables with p<0.10 in a single model. RESULTS: PM was present in 1138 of 3006 gradable fundus photographs, with 41.7% (95% CI 39.5% to 43.9%) of participants affected in at least one eye graded. Most eyes with PM exhibited diffuse chorioretinal atrophy (97.4%), while the more severe stages-patchy chorioretinal atrophy and macular atrophy-were observed in only 24 and 5 eyes, respectively. 13 eyes had 'plus' lesions or suspected staphyloma. Factors independently associated with increased odds of PM (all p<0.05) included decreasing SER (adjusted OR: 0.22, 95% CI 0.15 to 0.32), older age (2.20, 1.63 to 2.97), female sex (1.87, 1.12 to 3.12), lower deprivation (0.73, 0.56 to 0.94), white ethnicity (52.3, 17.3 to 158.3), lower retinal arteriovenous ratio (0.47, 0.37 to 0.58), increased retinal vascular complexity (4.60, 3.16 to 6.70) and a relatively horizontal disc orientation (2.98, 1.88 to 4.72). None of the explored modifiable lifestyle or health-related variables were associated with PM. CONCLUSIONS: PM prevalence is high among mid-life adults with high myopia in the UK Biobank, although most cases are relatively mild (diffuse chorioretinal atrophy). The only modifiable risk factor identified is myopia severity.