Abstract
PURPOSE: Both genetic and environmental factors contribute to the development of Fuchs' endothelial corneal dystrophy (FECD), the most common indication for corneal transplantation in the United States. Prior studies have suggested an association of height, weight, or body mass index (BMI) with FECD. We examined the association between anthropometric measures and incident FECD in the Women's Health Initiative Observational Study (WHI-OS) of postmenopausal women (n = 22,983). METHODS: Medicare Part B claims data from the WHI-OS baseline visit (1993-1998) to 2019 were used to identify incident cases of FECD. At baseline and follow-up year 3, weight, height, waist circumference (WC), and hip circumference were measured. At baseline, women were asked to recall their historic weight at ages 18, 35, and 50 years. At follow-up years 1 and 4 to 8, the women were asked to self-report their weight. Height and weight were used to calculate BMI at each time point. Adjusted hazard ratios (HRs), 95% confidence intervals (CI), and p for trend for incident FECD were estimated by measures of historic BMI, baseline anthropometrics measures, and anthropometric measures that incorporated more than one baseline visit (time-varying). Anthropometric measures were parameterized as continuous and categorical in analyses. RESULTS: There were 1399 incident FECD cases with an annualized incidence rate of 5.06 per 1000 person-years (95% CI = 4.80-5.33) over 276,443 person-years of follow-up. No statistically significant associations were observed between baseline height and risk of FECD. Women with baseline BMIs ≥35 (obesity II) compared to <25 kg/m2 (normal weight or underweight) had lower risk of incident FECD (HR = 0.68, 95% CI = 0.53-0.88) with a P value = 0.0373 for an ordinal trend analysis across BMI categories. Significant inverse associations were observed for continuous measures of WC (HR = 0.97, 95% CI = 0.95-0.99 per 5 cm increase) and waist-to-hip ratio (WHR; HR = 0.92, 95% CI = 0.86-0.99 per 0.1 unit increase). No statistically significant associations were observed for time-varying BMI, but time-varying WC and WHR has statistically significant inverse associations with risk for FECD (data not shown). CONCLUSIONS: In this cohort of postmenopausal women, BMI, WC, and WHR were inversely associated with incident FECD. These findings generally support the potential role of anthropometric measures, particularly those indicative of abdominal obesity in FECD susceptibility in women.