Abstract
Purpose:
To determine the effect of CFH and ARMS2 risk alleles on the relationship between dietary zinc intake and incident late-stage age-related macular degeneration (AMD) in well-defined, European-derived, population-based cohorts.
Methods:
Participants from the Blue Mountains Eye Study (BMES) and Rotterdam Study (RS) had dietary zinc intake assessed by food frequency questionnaires and genotyping performed using an Illumina Human 670-Quadv1 custom array and TaqMan assays. The cumulative 15-year probability of late-stage AMD was calculated using logistic regression analysis adjusting for age and smoking and after testing for interactions. Zinc intake was assessed both as a categorical and continuous variable. Participants were divided into four genotype groups according to CFH (C) and ARMS2 (A) risk allele number (0, 1 or 2); Group 1 (C01, A0), Group 2 (C2, A0), Group 3 (C01, A12), Group 4 (C2, A12).
Results:
A total of 5246 participants were studied. The frequencies of Groups 1, 2, 3 and 4 were 53.0%, 8.7%, 32.9%, 5.4% respectively in the BMES and 55.1%, 7.7%, 32.5%, 4.8% respectively in the RS. Mean zinc intake was 12mg/day and 10.6mg/day in the two population respectively, and not significantly different among genotype groups. The overall incidence of AMD was 5.0% and varied by genotype group, 2.7%, 8.2%, 6.2%, and 19.7% in Groups 1, 2, 3 and 4 respectively in a combined two-population dataset. After adjusting for age, smoking and zinc intake, AMD was 3.0, 2.5 and 11.8 times more likely in Groups 2, 3 and 4 compared with Group 1. Zinc intake was associated with a substantially reduced AMD incidence in Group 4 (per mg/day increase, OR 0.8, 95% CI 0.7-0.9, p=0.0042). Compared with the lowest quartile of zinc intake (7.5mg/day), those consuming the highest quartile (15.3mg/day) were 81% less likely to develop AMD (OR 0.19, 95% CI 0.06-0.64, p=0.0074). No statistically significant increase or decrease in AMD risk associated with dietary zinc intake levels was found in the other genotype groups.
Conclusions:
The frequencies of CFH/ARMS2 risk genotypes vary in European-derived populations with the lowest risk genotype found in over 50% of populations. In the genotype group at highest risk of AMD, increased dietary zinc intake was associated with a substantial reduction in AMD risk, in a dose-dependent manner.