结果：我们通过病例记录回顾，确定了1361例中度的和1118例进展的AMD视力在20/30甚至更差的病例（主要是新生血管性AMD）。比较预估的血浆叶黄素/玉米黄素评分的extreme quintiles，我们发现在男性和女性中，对于进展的AMD都有约40%的导致风险增加的降低，混合的极端比较相关风险=0.59，95%CI：0.48-0.73；P<0.001。对于其它类胡萝卜的预测血浆类胡萝卜素评分，包括β-隐黄素、α-胡萝卜素和β-胡萝卜素，与进展的AMD25%和35%的风险降低有关，当进行extreme quintiles比较。预估的血浆总类胡萝卜素指数的极端比较相关风险是0.65（95% CI：0.53-0.80; P<0.001）。我们没有发现任何类胡萝卜素，不管是预测血浆评分还是计算出的摄入量，与中度AMD有任何关系。结论和相关性：具有生物活性的叶黄素/玉米黄素的高摄取与进展的AMD 长期风险降低有关。给出其它一些类胡萝卜素同样和长期风险降低有关，为了增加各种各样的富含类胡萝卜素的水果蔬菜食用的公共卫生政策可能会降低进展期AMD的风险。
Importance: Despite strong biological plausibility， evidence from epidemiologic studies and clinical trials on the relations between intakes of lutein and zeaxanthin and age-related macular degeneration (AMD) has been inconsistent. The roles of other carotenoids are less thoroughly investigated.
Objective: To investigate the associations between intakes of carotenoids and AMD. Design， Setting， and Participants: Prospective cohort study， with cohorts from the Nurses’ Health Study and the Health Professionals Follow-up Study in the United States. A total of 63443 women and 38603 men were followed up， from 1984 until May 31， 2010， in the Nurses’ Health Study and from 1986 until January 31， 2010， in the Health Professionals Follow-up Study. All participants were aged 50 years or older and were free of diagnosed AMD， diabetes mellitus， cardiovascular disease， and cancer at baseline.
Main Outcomes and Measures: Predicted plasma carotenoid scores were computed directly from food intake， assessed by repeated food frequency questionnaires at baseline and follow-up， using validated regression models to account for bioavailability and reporting validity of different foods， and associations between predicted plasma carotenoid scores and AMD were determined.
Results: We confirmed 1361 incident intermediate and 1118 advanced AMD cases (primarily neovascular AMD) with a visual acuity of 20/30 or worse by medical record review. Comparing extreme quintiles of predicted plasma lutein/zeaxanthin score， we found a risk reduction for advanced AMD of about 40% in both women and men (pooled relative risk comparing extreme quintiles = 0.59; 95% CI， 0.48-0.73; P for trend < .001). Predicted plasma carotenoid scores for other carotenoids， including beta-cryptoxanthin， alpha-carotene， and beta-carotene， were associated with a 25% to 35% lower risk of advanced AMD when comparing extreme quintiles. The relative risk comparing extreme quintiles for the predicted plasma total carotenoid index was 0.65 (95% CI， 0.53-0.80; P for trend < .001). We did not identify any associations of carotenoids， either as predicted plasma score or calculated intake， with intermediate AMD.
Conclusions and Relevance: Higher intake of bioavailable lutein/zeaxanthin is associated with a long-term reduced risk of advanced AMD. Given that some other carotenoids are also associated with a lower risk， a public health strategy aimed at increasing dietary consumption of a wide variety of fruits and vegetables rich in carotenoids may reduce the incidence of advanced AMD.