Features of Age-Related Macular Degeneration in the General Adults and Their Dependency on Age, Sex, and Smoking: Results from the German KORA Study

Abstract

Age-related macular degeneration (AMD) is a vision impairing disease of the central retina characterized by early and late forms in individuals older than 50 years of age. However, there is little knowledge to what extent also younger adults are affected. We have thus set out to estimate the prevalence of early AMD features and late AMD in a general adult population by acquiring color fundus images in 2,840 individuals aged 25 to 74 years of the Cooperative Health Research in the Region of Augsburg project (KORA) in South Germany. Among the 2,546 participants with gradable images for each eye, 10.9% (n = 277) had early AMD features (applying the 9-step Age-Related Eye Disease Study Severity Scale), 0.2% (n = 6) had late AMD. Prevalence increased with age, reaching 26.3% for early AMD features and 1.9% for late AMD at the age 70+. However, signs of early AMD were found in subjects as young as 25 years, with the risk for early AMD features increasing linearly by years of age in men, and, less consistent with a linear increase, in women. Risk for early AMD features increased linearly by pack years of smoking in men, not in women, nor was there any association with other lifestyle or metabolic factors. By providing much sought-after prevalence estimates for AMD from Central Europe, our data underscores a substantial proportion of the adult population with signs of early AMD, including individuals younger than 50 years. This supports the notion that early AMD features in the young might be under-acknowledged.

Fig 1. The age-trend of early AMD risk modelled separately for men and women.

Shown is the modelled trend of early AMD relative risk by years of age utilizing a thin plate regression spline for A) men (n early AMD = 134, n AMD-free = 1,139) or B) women (n early AMD = 143, n AMD-free = 1,124). Relative risk is given as log odds ratio with persons at the age of 50 as reference and 95% confidence interval shaded in grey. The modelled trend was in line with a linear relationship for men (degrees of freedom = 1.001), but not for women (degrees of freedom = 3.283).