Abstract
Purpose :
MTD has been shown to decrease the incidence and progression of AMD. Furthermore, MTD and AMD have each been associated with all-cause mortality, though the underlying mechanisms remain unclear. To assess for possible shared pathways, we examined associations among MTD, AMD, and all-cause mortality.
Methods :
MTD was calculated using the alternate Mediterranean diet score, based on baseline food-frequency questionnaires. AMD status (baseline AMD severity) was determined using centrally graded stereoscopic fundus photographs using the AREDS AMD severity scale with 11-steps. Cox proportional-hazards regression was used to analyze associations among MTD, AMD status, and all-cause mortality, adjusting for covariates including age, sex, smoking, as well as history of diabetes mellitus, hypertension, and angina. To evaluate effect modification, all models were stratified by smoking status and genetic polymorphisms (7 variants at 4 loci associated with both AMD and longevity). Mediation analyses was used to estimate whether the effect of MTD on mortality could be explained in part by the effect of MTD on AMD status.
Results :
The analytic sample included 8,414 individuals (4,417/4,757 of AREDS and 3,997/4,203 of AREDS2 participants). Median follow-up time was 10.1 years, and there were 1,408 reported deaths. MTD was inversely associated with both AMD (Tertile 3 vs. 1, OR=0.69, 95%CI 0.61-0.78, p<0.001) and mortality risk (Tertile 3 vs. 1, HR=0.70, 95%CI 0.61-0.80, p<0.001). AMD was also associated with increased mortality risk (AMD severity score >=7 vs. <7, HR=1.20, 95% CI 1.07-1.35, p=0.002). There was no significant interaction between MTD and AMD on mortality risk (p=0.56). Smoking status and genetic polymorphisms did not modify the associations among MTD, AMD, and mortality. Only 5.9% of the effect of MTD on mortality was statistically attributable to its effect on AMD status.
Conclusions :
MTD was associated with protection against both AMD and all-cause mortality in this large sample of older adults, and these associations were not modified by smoking status or the genetic polymorphisms studied. Interestingly, mediation analysis demonstrated that the effect of MTD on AMD did not mediate its effect on mortality, suggesting that these relationships may involve partially distinct biological pathways.
This is a 2020 ARVO Annual Meeting abstract.