Adherence to a Mediterranean diet and its association with age-related macular degeneration. The Coimbra Eye Study–Report 4

Nutrition 51-52 (2018) 6–12

Contents lists available at ScienceDirect

Nutrition j o u r n a l h o m e p a g e : w w w. n u t r i t i o n j r n l . c o m

Applied nutritional investigation

Adherence to a mediterranean diet and its association with age-related macular degeneration. The Coimbra Eye Study–Report 4 Sandrina Nunes Ph.D. a,*, Dalila Alves M.Sc. a, Patrícia Barreto M.Sc. a, Miguel Raimundo M.D. b, Maria da Luz Cachulo M.D., Ph.D. a,b,c, Cláudia Farinha M.D. a,b, Inês Laíns M.D. b,d, João Rodrigues M.D. e, Carlos Almeida M.D. f, Luísa Ribeiro M.D., Ph.D. a, João Figueira M.D., Ph.D. a,b,c, Lelita Santos M.D., Ph.D. c,g, Rufino Silva M.D., Ph.D. a,b,c a Association

for Innovation and Biomedical Research on Light and Image, Coimbra, Portugal Department, Centro Hospitalar e Universitário de Coimbra, Coimbra, Portugal c Faculty of Medicine, Institute for Biomedical Imaging and Life Sciences University of Coimbra, Coimbra, Portugal d Massachusetts Eye and Ear, Harvard Medical School, Boston, Massachusetts e Primary Health Care Unit of Lousã, Unidade de Saúde Familiar da Serra da Lousã e Trevim Sol, Coimbra, Portugal f Primary Health Care Center of Mira, Coimbra, Portugal g Serviço de Medicina Interna, Centro Hospitalar e Universitário de Coimbra, Coimbra, Portugal b Ophthalmology

A R T I C L E

I N F O

Article history: Received 29 September 2017 Received in revised form 19 December 2017 Accepted 29 December 2017 Keywords: Epidemiology Mediterranean diet Nutrition Nutrients AMD

A B S T R A C T

Objectives: This study aimed to characterize the association of lifestyle and nutritional risk profiles with age-related macular degeneration (AMD) in two subpopulations with differing AMD prevalence. Methods: This case-control study (n = 1992) included 768 patients with AMD and 1224 age- and sexmatched participants without AMD with a single visit at a primary health care unit. Enrolled participants completed a validated lifestyle and food frequency questionnaire. A score to measure adherence to the Mediterranean diet (mediSCORE; Range, 0–9) was constructed from individual food intakes, which were further analyzed by conversion to nutrient consumption. Results: Higher adherence to the Mediterranean diet (mediSCORE ≥6) was significantly associated with no AMD (odds ratio [OR] = 0.73; P = 0.009). The subpopulation with lower AMD prevalence presented significantly higher adherence to the Mediterranean diet in relation to all individual food groups that comprised the mediSCORE (P < 0.014) with the exception of cereals. Food group analysis showed significant associations between the increased consumption of vegetables (OR = 0.63; P < 0.001) and fruit and nuts (OR = 0.78; P = 0.010) with no AMD. Nutrient analysis revealed that an increased ingestion of water, fibers, total fat, monounsaturated and polyunsaturated fatty acids, linoleic acid, vitamins A and C, carotene, alpha-tocopherol, folate, magnesium, iron, and zinc were significantly associated with no AMD (P < 0.0013). Finally, regular physical activity was associated with no AMD (P = 0.003). Conclusions: High adherence to a Mediterranean diet and regular physical activity seem to be protective factors for AMD in a Portuguese population. The effect of the diet is likely driven by the increased consumption of vegetables, fruits, and nuts. © 2018 Elsevier Inc. All rights reserved.

Introduction

Conflicts of interest: None. Sources of support: This investigator-initiated study was financially supported by Novartis Pharma AG. * Corresponding author. Tel.: +351 239 480 137; fax: + 351 239 480 117. E-mail address: [email protected] (S. Nunes). https://doi.org/10.1016/j.nut.2017.12.010 0899-9007/© 2018 Elsevier Inc. All rights reserved.

Age-related macular degeneration (AMD) is a chronic disease of the central retina and the leading cause of irreversible blindness in the elderly population in developed countries [1–3]. There is a great need to identify preventive measures to delay or halt AMD progression, particularly in light of the current lack of effective pharmacologic options for the dry form of the disease,

S. Nunes et al. / Nutrition 51-52 (2018) 6–12

which presents in 0.61% of patients [4]. The only established intervention relates to the Age-Related Eye Disease Study (AREDS) [5], which showed that high-dose zinc/antioxidant supplements reduce progression from intermediate to late AMD. Preventive strategies through dietary modulation are attractive because they are easy to implement, relatively cheap, and could enable the development of specific dietary recommendations for people at risk of AMD. Previous studies have suggested that certain micro- and macronutrients [5–10] such as lutein and zeaxanthin [11,12] and longchain, omega-3 fatty acids [13,14] are beneficial. However, the association of single nutrients with AMD have often been inconsistent across studies and are impossible to disentangle from other aspects of diets as they do not account for the synergistic relationship of food components [15]. Therefore, an examination of the overall diet modeled on dietary guidelines/recommendations and AMD risk is important. Other modifiable risk factors for AMD have also been proposed including smoking [16–20], body mass index (BMI) outside the normal range [21], and physical activity [15,22]. Improved population health coupled with lifestyle guideline adherence will undoubtedly be beneficial to at-risk persons. The aim of this study is to characterize the association between adherence to a Mediterranean diet and AMD in two subpopulations (one from a coastal town, the other from an inland town) with differing prevalence rates of AMD because, to the best of our knowledge, the protective role of the Mediterranean diet has already been investigated but studies are scarce [15,23–25]. Methods

7

classification 0–no AMD features or only drusen <63 μm) were invited to answer a validated lifestyle and food habits questionnaire. This study was approved by the Association for Innovation and Biomedical Research on Light and Image (AIBILI) Ethics Committee and all participants provided written informed consent. We planned to enroll 1000 participants from each town over a 12-mo recruitment period. Participants who were previously graded with AMD were contacted first and the other participants (selected at random) were contacted to achieve an age- and sex-matched population. For the inland town, we contacted 1582 patients by phone and invited them to participate. Of these, 999 (63.1%) accepted and completed the questionnaire. In the coastal town, because we had the “Five-year Incidence of Age-related Macular Degeneration in the Central Region of Portugal” study (NCT02748824) ongoing at the same time, a letter was sent to 1387 patients to invite them to also answer the questionnaire. Of the contacted participants, 1008 (72.7%) agreed to participate in this study. Therefore, a total of 2007 participants were included in this analysis (Fig. 1).

Age-related macular degeneration grading and staging In the Coimbra Eye Study, all participants were subjected to a complete bilateral ophthalmologic examination with an evaluation of the best-corrected visual acuity, anterior segment biomicroscopy, tonometry, and digital mydriatic color fundus photograph (Topcon TRC-50 EX; Topcon Corp., Tokyo, Japan). Images were graded for the presence of AMD or no AMD in a centralized reading center (Coimbra Ophthalmology Reading Center, CORC–AIBILI) and a differential analysis for AMD lesions was conducted thereafter by two senior, independent, and certified ophthalmologists using the International Classification and Grading System for AMD [28]. Fifty participants who were graded with AMD were excluded from the study due to the low quality of the images (i.e., the presence of AMD lesions could not be confirmed due to other ocular pathologies such as cataract). The full study protocol of the “Epidemiologic Study of the Prevalence of Age-Related Macular Degeneration in Portugal: The Coimbra Eye Study” [4,26] has previously been described. In the coastal town, the grading results for the presence of AMD that were obtained in the incidence study were used (NCT02748824).

Study design and population

Demographic and lifestyle data

This is a case-control study (NCT01715870) nested in the “Epidemiologic Study of the Prevalence of Age-Related Macular Degeneration in Portugal: The Coimbra Eye Study” (NCT01298674)–a cross-sectional, population-based study. In the Coimbra Eye Study, we included participants over 55 y of age from primary health care units in two locations in the center of Portugal: one in the coastal area (Mira– Coastal Town) and the other 70 km away from the sea (Lousã–Inland Town) [4,26]. For the present study, a select sample of participants with AMD (Rotterdam classification [27] 1–4) and an age- and sex-matched control group (Rotterdam

Demographic data as well as information on the participants’ general and ophthalmic medical history were collected for all study participants in the original Coimbra Eye Study (NCT01298674 and NCT0274/8824). For this study, participants were invited to answer questionnaires that included information on education, smoking habits, and regular physical activity (i.e., any kind of exercise that was reported by the participant such as walking, cycling, fitness, swimming at least once a week), medical history, and food frequency. The food frequency questionnaire was adapted from the food frequency questionnaire by

Fig. 1. Study flowchart. (1) Patients included from October 2012 to January 2014 in Lousã did not have their images graded for the presence of AMD when the questionnaire was administrated. (2) Images that were not graded due to lack of photo quality or obscuring lesions. (3) AMD presence was considered on the basis of the grading that was performed in the epidemiologic study. (4) Excluded because AMD grading could not be confirmed in the incidence study (2016–2017). AMD, agerelated macular degeneration.

8

S. Nunes et al. / Nutrition 51-52 (2018) 6–12

Willett and Lenart [29] and validated for the Portuguese population by the Faculty of Medicine of the University of Porto [30]. The questionnaire includes 86 types of food that are structured into nine major food groups: Vegetables, legumes, fruits and nuts, cereals, fish, meat, dairy products, alcohol and a ratio of monounsaturated lipids (mainly olive oil) to saturated lipids. For each food, nine categories of frequencies are included that range from “never or less to once per month” to “six or more times per day.” For the analysis, food groups were obtained by merging food items according to their nutritional composition similarity. To analyze the macro—and micronutrients, the 2007 Table of Portuguese Food Composition [31] as developed by the National Health Institute Doutor Ricardo Jorge was applied. The questionnaire was completed during a single visit by specially trained interviewers. For each item, participants were asked to report the frequency of their consumption in the past year, the portion size, and whether or not this consumption was seasonal. The interviewer also measured each participant’s weight, height (for BMI calculation), and abdominal perimeter. Processing of nutritional data All questionnaires were checked for completeness. Food frequency was calculated for average daily consumption values, adjusted for the size of the portion to yield a value in g or mL per type of food. A factor of seasonal variation consumption was also included if indicated by the participant (0.25 for a period of 3 mo). The following daily food frequency factor was considered: 0 when missing; 1/30 when “never or <1 month”; 2/30 when “1 to 3 per month”; 4/30 when “1 per week”; 12/30 when “2 to 4 per week”; 20/30 when “5 to 6 per week”; 1 when “1 per day”; 2 when “2 to 3 per day”; 4 when “4 to 5 per day”; and 6 when “6 + per day”. The following portion factor was considered: 0 when missing; 0.5 when “less”; 1 when “equal”, and 2 when “more”. The average daily consumption values were calculated as: food frequency = food frequency factor × portion factor × quantity

( × 0.25 if seasonal variation consumption was indicated by thee participant ) . Adherence to the mediterranean diet We adopted a model of adherence to the Mediterranean diet that has already been assessed in other medical fields [32,33] and initially validated for a Greek population [34]. According to this model, we calculated a score (mediSCORE) from the sum of the nine food group indicator variables. Each of these indicator variables can take a value of 0 or 1, which was attributed by comparison with a cutoff that was defined by the median sex-specific food group consumption in grams. In accordance with the original model [34], consumption above this cutoff of beneficial dietary components (vegetables, legumes, fruits and nuts, cereals, and fish) was assigned a value of 1 and below a value of 0. Conversely, consumption below the cutoff of dairy products and meat components was assigned a value of 1 and above a value of 0. For alcohol consumption, a moderate consumption was considered beneficial because alcohol consumption in the Portuguese population consists mainly of wine (average weight for wine 87%, beer 12%, and spirits 1%). A value of 1 was attributed for consumption between 10 and 50 g/d of alcohol for men and between 5 and 25 g/d for women. Consumptions outside of this range were attributed a value of 0. Finally, as a proxy of overall beneficial fat intake, we considered a ratio of monounsaturated to saturated lipids and similarly defined sex-specific cutoffs using the median whereby consumption above this median was assigned a value of 1 and below a value of 0. Cutoff values are presented in Supplementary Table S1. The final score of adherence to the Mediterranean diet (mediSCORE) was obtained by adding these dummy variables, which resulted in an ordinal scale from 0 (minimum adherence) to 9 (maximum adhesion). High adherence was defined as a mediSCORE ≥6 (Range, 6–9) and deemed nutritionally relevant [34]. Statistical analysis All statistical analyses were performed with STATA, Version 12.1 (StataCorp LP; College Station, TX) and P values ≤ 0.05 were considered statistically significant. To address the multiple comparisons problem, we performed a Bonferroni correction to control the familywise error rate. Categorical variables were summarized with frequencies and percentages and numerical variables with mean and standard deviation (SD). The variables of interest were compared between the two subpopulations and between the groups with and without AMD. The independent samples t test was used for continuous variables and the Pearson’s χ2 test was used for categorical variables. A univariate logistic regression model to estimate the odds ratio (OR) and 95% confidence interval (CI) for each food group of mediSCORE and for each

possible confounder was computed as well as a multivariate logistic regression model that was adjusted for the covariates (i.e., age, sex, BMI, abdominal perimeter, physical activity, smoking status [non-smoker, ex-smoker, smoker], diabetes, and hypertension).

Results Characterization of the included study population We analyzed a total of 1992 participants: 985 (49.4%) from the inland town and 1007 (50.6%) from the coastal town. Supplementary Table S2 presents the characteristics of the included study population as well as a comparison between the two subpopulations. Among the differences, participants from the coastal town were significantly older (P < 0.001) and had a BMI that was significantly lower (P < 0.001) than that of participants from the inland town. We recruited a total of 768 patients with AMD (38.6%) and 1224 without AMD (61.4%). In the inland town, there were 434 patients with AMD (44.1%) and 551 without AMD (55.9%). In the coastal town, there were 334 patients with AMD (33.2%) and 673 without AMD (66.8%). Adherence to the mediterranean diet (mediSCORE) and specific food groups The analysis of the nine food groups that comprise the mediSCORE (Supplementary Table S3) revealed that the intake of all but the cereals group was statistically different between the two subpopulations. Vegetable, legume, fish, and dairy product consumption that was above the sex-specific median was higher in the coastal town but fruits and nuts and meat consumptions were higher in the inland town. A moderate consumption of alcohol, which was considered beneficial, was higher in the inland town but the ratio of monounsaturated lipids and saturated lipids that were above the sex-specific median was higher in the coastal town. The adherence to a Mediterranean diet (mediSCORE) for each subpopulation is presented in Supplementary Figure S1 and was higher in the coastal town. In fact, in the coastal town, only 0.5% of the total population showed extreme values (0 or 9) and 46.9% of the participants had a score between 1 and 4 and 52.6% between 5 and 8. In the inland town, 1.1% of the total population showed the extreme value (0) and 67.7% of the participants had a score between 1 and 4 and 31.2% 5 and 8. The mean diet score was 4.5 ± 1.6 in the coastal town and 3.7 ± 1.6 in the inland town and its distribution was approximately normal. Association between lifestyle and mediterranean diet profile and AMD In our study population, participants with and without AMD did not differ significantly in terms of sex (Table 1). However, despite the attempt to also select an age-matched sample, age was significantly higher in participants with AMD (P = 0.002). Other parameters (all described in Table 1) also significantly differed between patients with AMD and those with a healthy macula. As intended, we assessed the association between food habits and AMD versus no AMD. Our results revealed that intake of sandwiches was significantly more frequent in participants without AMD (P = 0.025). The number of meals/d, consumption of traditional meals, light meals, breakfast, fast-food and premade meals

S. Nunes et al. / Nutrition 51-52 (2018) 6–12

Table 1 Sample characterization for participants without AMD and patients with AMD No AMD (n = 1224) Demographic characteristics Age (y), mean (SD) Male, n (%) Biometrics Body mass index (kg/m2), mean (SD) Abdominal perimeter (cm), mean (SD) Education University, n (%) High school, n (%) Primary school, n (%) None, n (%) Smoking status Never, n (%) Ex-smoker, n (%) Current smoker, n (%) Pack-years (n), mean (SD)† Practice of physical exercise Regular physical activity,‡ n (%) Physical activity duration (h/wk), mean (SD)§ Supplement intake,¶ n (%) Medical history Diabetes, n (%) Hypertension, n (%) Dyslipidemia, n (%) Obesity, n (%) Predominant food habits Meals/d (n), mean (SD) Breakfast, n (%) Traditional meals, n (%) Fast-food, n (%) Premade meals, n (%) Sandwiches, n (%) Light meals, n (%) Energy intake (kcal/d), mean (SD)

AMD (n = 768)

Adherence to the mediterranean diet (mediSCORE) and age-related macular degeneration

71.6 (7.7) 323 (42.1)

0.002 0.141

28.4 (4.4)

28.1 (4.5)

0.073

99.8 (12.2)

98.3 (13.4)

0.019

91 (7.5) 39 (3.2) 1064 (87.7) 19 (1.6)

55 (7.3) 29 (3.8) 662 (87.1) 14 (1.8)

0.858

964 (79.4) 212 (17.5) 38 (3.1) 30.2 (33.9)

605 (78.9) 137 (17.9) 25 (3.3) 31.8 (33.5)

0.959

464 (38.0)

241 (31.5)

0.003

0.96 (1.23)

1.22 (1.38)

0.016

61 (5.0)

45 (5.9)

0.403

305 (25.0) 659 (54.1) 612 (50.2) 642 (52.8)

152 (19.8) 462 (60.4) 382 (49.8) 460 (60.1)

0.007 0.006 0.876 0.002

3.97 (0.87) 748 (97.5) 749 (97.5) 8 (1.0) 20 (2.6) 102 (13.3) 501 (65.4) 1882.7 (511.7)

were similarly distributed in both groups. Energy intake was significantly higher in participants without AMD (P = 0.003) (Table 1).

P value*

70.6 (7.0) 556 (45.4)

4.03 (0.88) 1183 (97.2) 1196 (98.0) 6 (0.5) 29 (2.4) 208 (17.1) 811 (66.5) 1953.3 (516.4)

9

0.634

0.129 0.670 0.450 0.152 0.747 0.025 0.606 0.003

AMD, age-related macular degeneration; SD, standard deviation. * Between-group comparisons using Pearson’s χ2 test for categorical variables and Student’s t test for continuous variables. † Average smoking pack-years of current smokers and ex-smokers. ‡ Principal physical activities as reported by the participants: Walking (approximately 74%), cycling (approximately 12%), fitness (approximately 7%), and swimming (approximately 4%). § Average physical activity duration of participants who reported regular physical activity. ¶ Vitamins and/or minerals intake.

A high adherence to the Mediterranean diet (cutoff mediSCORE ≥6) was associated with a decreased risk for AMD for 67.2% of participants without AMD and 32.8% with AMD (P = 0.009) and an OR of 0.73 (95% CI, 0.58–0.93). When analyzing the individual food groups that comprise the mediSCORE (Table 2), vegetables and fruit and nuts were the only groups with a statistically different consumption between participants with and without AMD. Consumption of vegetables above the sex-specific median as well as fruits and nuts were higher in the group without AMD (P < 0.001 and P = 0.005, respectively). A higher consumption of vegetables and fruits and nuts was associated with no AMD with an OR = 0.63 (95% CI, 0.52–0.76; P < 0.001) for vegetables consumption and an OR 0.78 (95% CI, 0.65–0.94; P = 0.010) for fruits and nuts consumption. To account for the influence of age, sex, BMI, abdominal perimeter, practice of regular physical exercise, smoking habits, and diabetes, an adjusted OR of 0.64 (95% CI, 0.53–0.78; P < 0.001) for vegetable consumption and 0.79 (95% CI, 0.66–0.96; P = 0.015) for fruit and nut consumption were obtained (Fig. 2, Table 2). Exploratory micronutrient analysis To analyze the impact of the macro- and micronutrient composition of the participants’ diet and AMD frequency, consumptions were compared between tertiles, below the first tertile and above the second tertile to define low and high nutrient consumption, respectively. A significantly higher consumption of water, total fat, fibers, mono- and polyunsaturated fatty acids, linoleic acid, vitamin A, carotene, alpha-tocopherol, vitamin C, folate, magnesium, iron, and zinc was found in the group without AMD (Supplementary Table S4). These findings remain significant after a Bonferroni correction was applied. Discussion We present a cross-sectional study on the association between diet as a whole and AMD in two subpopulations (inland and coastal towns) with different lifestyles and food habits. Most of

Table 2 Logistic regression model that considers AMD versus no AMD as an outcome and the nine food group consumption, adjusted for age, sex, body mass index, abdominal perimeter, regular physical activity, smoking habits, diabetes, and hypertension mediSCORE groups

No AMD (n = 1224)

AMD (n = 768)

OR

95% CI

P value

1. Vegetables 2. Legumes 3. Fruits and nuts 4. Cereals 5. Fish 6. Dairy products 7. Meat 8. Alcohol‡ 9. Ratio of monounsaturated lipids/saturated lipids

667 (54.5) 629 (51.4) 641 (52.4) 608 (49.7) 605 (49.4) 621 (50.7) 610 (49.8) 134 (11.0) 625 (51.1)

332 (43.2)* 373 (48.6) 352 (45.8)* 394 (51.3) 400 (52.1) 377 (49.1) 380 (49.5) 66 (8.6) 379 (49.4)

0.66 1.01 0.77 1.10 1.32 0.93 0.96 0.82 0.97

(0.54–0.80) (0.83–1.23) (0.64–0.93) (0.91–1.33) (0.98–1.77) (0.77–1.13) (0.79–1.16) (0.59–1.12) (0.80–1.18)

<0.001† 0.895 0.008† 0.321 0.065 0.487 0.676 0.214 0.766

AMD, age-related macular degeneration; CI, confident interval; OR, odds ratio. Data are presented as n (%). * Denotes statistical significance at P < 0.006 (after applying Bonferroni correction) for comparisons between groups using the Pearson’s χ2 test. † Denotes statistical significance at P < 0.05 for the model adjusted for age, sex, body mass index, abdominal perimeter, regular physical activity, smoking habits and diabetes. ‡ Alcohol consumption is composed of wine (87%), beer (12%), and spirits (1%).

10

S. Nunes et al. / Nutrition 51-52 (2018) 6–12

Fig. 2. Odds ratio and 95% confidence intervals for AMD for each of the nine food groups that comprise the mediSCORE. Alcohol consumption is composed of wine 87%, beer 12%, and spirits 1%. AMD, age-related macular degeneration; conf. int., confidence invterval.

the prior work that has been performed in this field evaluated the association between single nutrients, single foods or food groups, and AMD risk. The consideration of a diet as a whole is not only more realistic but also investigates the influence of some foods and nutrients that may only be beneficial in combination with others. We used a well-established and validated model of adherence to the Mediterranean diet [34] called the mediSCORE. Our results revealed that a higher adherence to this diet (mediSCORE ≥6) was significantly associated with a reduced probability of having AMD. These results are in line with the prospective longitudinal study on the AREDS cohort, which demonstrated that a high adherence to the Mediterranean diet (using the alternate Mediterranean diet score ≥6 in an American population) was associated with a reduced progression to advanced AMD [23,24]. Similarly, in a cross-sectional study using the AREDS baseline dietary data, Chiu et al. [35] described two dietary patterns (i.e., Western and Oriental) and verified that those participants who followed the Oriental pattern, which closely resembles the Mediterranean diet pattern, presented with a reduced risk of early and late AMD [35]. Our results also reinforce the results of the European Eye Study, which included participants from seven European countries and showed that an increasing Mediterranean diet score was significantly associated with reduced odds of developing exudative AMD [24,25]. When looking at each food group individually, we observed that an increased intake of vegetables, fruit, and nuts was associated with no AMD. This result is in agreement with previous publications [36] and biologically plausible [37,38] because these foods are high in antioxidants. Therefore, one might consider it likely that vegetables, fruit, and nuts drive the results that are

observed in our study and thus, the potential protective effect of the Mediterranean diet in AMD. With regard to the dietary macro- and micronutrients, our exploratory analysis revealed that the consumption of water, total fat, fibers, mono- and polyunsaturated fatty acids, linoleic acid, vitamin A, carotene, alpha-tocopherol, vitamin C, folate, magnesium, iron, and zinc was significantly higher in the population without AMD. This in agreement with the results from the AREDS study that demonstrated the efficacy of beta-carotene, vitamin C, vitamin E, and zinc as exogenous [5]. Although some studies have suggested that a small protective effect is associated with the moderate consumption of alcohol [39], we were unable to replicate these results. We also were not able to replicate the association between AMD and smoking, which is probably due to our small percentage of smokers (3.2%). In our study, the number of participants who practiced regular physical activity was significantly higher in the cohort without AMD. This might be linked to oxidative stress and inflammation, which are known to have a role in AMD pathophysiology [40–43]. Indeed, physical exercise has been shown to promote an antioxidant environment [44,45], which might prevent AMD development and progression. In addition, physical exercise might also influence AMD risk by lowering systolic blood pressure [46], which has been associated with a decreased incidence of AMD [47]. This is in agreement with the results found in this study where increasing blood pressure was found to be associated with increased odds of AMD. Despite these hypotheses, the mechanisms behind a potential association between physical activity and AMD are still only partially understood and the prior literature is controversial. Some authors have reported a positive impact with higher levels of exercise associated with a lower risk of AMD [15,48–51]. The

S. Nunes et al. / Nutrition 51-52 (2018) 6–12

Women’s Health Initiative Observational Study also suggested that physical activity might lower the risk of AMD by improving the status of the macular pigment either directly by reducing inflammation and oxidative stress or indirectly by reducing the obesity-associated phenotype of lower macular pigment and increased carotenoid turnover [15]. However, large populationbased cohorts such as the Blue Mountains Eye Study have reported a nonsignificant association [52]; therefore, further studies are required to definitively establish the relationship between physical exercise and AMD. This study has several strengths. The first relates to the welldefined strict methodology including AMD diagnosis, grading of the fundus photographs that were taken by independent graders with no knowledge of any characteristics of the participants, and standardized personal interviews of all patients rather than database records. Secondly, this study provides an overview of the individual health profiles and is not limited to isolated reports of food consumption; thus, the study reflects the concept that the magnitude of risk reduction that is associated with multiple healthy lifestyle interventions might be greater than the sum of the individual risk factors [15,48]. Third, the study presents a relatively large sample size and, to our knowledge, is the first study to assess the lifestyle and dietary risk profile of patients with AMD in a large Southern European population. The limitations of this study include the healthy user bias, which is consistent across most epidemiologic studies. This reflects a tendency for healthier individuals to be more likely to adhere to a preventive strategy or healthy behaviors [53]. In this case, it is possible that healthy users have less severe disease because they have more frequent, routine eye care, greater adherence to screening and prevention strategies, and more motivation and health consciousness, which is also reflected in their levels of physical activity. In addition, lifestyle and dietary data were based on self-reported information. Another limitation of this study is its cross-sectional nature, which did not allow us to analyze the development of AMD. In fact, due to the type of study, no causal inferences could be made with regard to the associations found. We also did not take into account the influence of genetic factors and their potential interaction with the remaining risk parameters. Conclusions Adopting a Mediterranean-type diet including a high intake of vegetables, fruit, and nuts and regular physical exercise (any kind of exercise at least once a week) is likely to be beneficial to prevent AMD. Further studies are needed to validate and extend our findings. Acknowledgments The authors gratefully acknowledge the financing by Novartis Pharma AG, which made this study possible as well as the collaboration and dedication of the personnel at the primary health care units of Mira and Lousã, namely Leonor Borralho, and the collaboration of Sónia Simões, João Almeida, Miguel Costa, Liliana Carvalho e Vanessa Santos from the AIBILI Coordinating Center. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.nut.2017.12.010.

11

References [1] Taylor HR, Keeffe JE, Vu HT, Wang JJ, Rochtchina E, Pezzullo ML, et al. Vision loss in Australia. Med J Aust 2005;182:565–8. [2] Sin HPY, Liu DTL, Lam DSC. Lifestyle modification, nutritional and vitamins supplements for age-related macular degeneration. Acta Ophthalmol 2013;91:6–11. [3] Mitchell P, Wang JJ, Foran S, Smith W. Five-year incidence of age-related maculopathy lesions: the Blue Mountains Eye Study. Ophthalmology 2002; 109:1092–7. [4] Cachulo Mda L, Lains I, Lobo C, Figueira J, Ribero L, Marques JP, et al. Agerelated macular degeneration in Portugal: prevalence and risk factors in a coastal and an inland town. The Coimbra Eye Study—report 2. Acta Ophthalmol 2016;94:1–12. [5] Age-related Eye Disease Study Research Group. A randomized, placebocontrolled, clinical trial of high-dose supplementation with vitamins C and E, beta carotene, and zinc for age-related macular degeneration and vision loss. Arch Ophthalmol 2001;119:1417. [6] Bressler NM, Bressler SB, Congdon NG, Ferris FL 3rd, Friedman DS, Klein R, et al. Potential public health impact of age-related eye disease study results. Arch Ophthalmol 2003;121:1621. [7] Clemons TE, Milton RC, Klein R, Seddon JM, Ferris FL 3rd, Age-related Eye Disease Study Research Group. Risk factors for the incidence of advanced age-related macular degeneration in the Age-related Eye Disease Study (AREDS) AREDS report no. 19. Ophthalmology 2005;112:533–9. [8] Davis MD, Gangnon RE, Lee LY, Hubbard LD, Klein BE, Klein R, et al. The Agerelated Eye Disease Study severity scale for age-related macular degeneration: AREDS report no. 17. Arch Ophthalmol 2005;123:1484– 98. [9] Ferris FL, Davis MD, Clemons TE, Lee LY, Chew EY, Lindblad AS, et al. A simplified severity scale for age-related macular degeneration: AREDS report no. 18. Arch Ophthalmol 2005;123:1570–4. [10] AREDS2 Research Group, Chew EY, Clemons T, SanGiovanni JP, Danis R, Domalpally A, et al. The Age-related Eye Disease Study 2 (AREDS2): study design and baseline characteristics. Ophthalmology 2012;119:2282–9. [11] Moeller SM, Parekh N, Tinker L, Ritenbaugh C, Blodi B, Wallace RB, et al. Associations between intermediate age-related macular degeneration and lutein and zeaxanthin in the carotenoids in Age-related Eye Disease Study (CAREDS): ancillary study of the Women’s Health Initiative. Arch Ophthalmol 2006;124:1151–62. [12] Age-related Eye Disease Study Research Group, SanGiovanni JP, Chew EY, Clemons TE, Ferris FL 3rd, Gensler G, et al. The relationship of dietary carotenoid and vitamin A, E, and C intake with age-related macular degeneration in a case-control study. Arch Ophthalmol 2007;125:1225–32. [13] SanGiovanni JP, Agron E, Meleth AD, Reed GF, Sperduto RD, Clemons TE, et al. {omega}-3 Long-chain polyunsaturated fatty acid intake and 12-y incidence of neovascular age-related macular degeneration and central geographic atrophy: AREDS report 30, a prospective cohort study from the Age-related Eye Disease Study. Am J Clin Nutr 2009;90:1601–7. [14] SanGiovanni JP, Chew EY, Agron E, Clemons TE, Ferris FL 3rd, Gensler G, et al. The relationship of dietary omega-3 long-chain polyunsaturated fatty acid intake with incident age-related macular degeneration. Arch Ophthalmol 2008;126:1274–9. [15] Mares JA, Voland RP, Sondel SA, Millen AE, Larowe T, Moeller SM, et al. Healthy lifestyles related to subsequent prevalence of age-related macular degeneration. Arch Ophthalmol 2011;129:470–80. [16] Klein R, Klein BE, Linton KL, DeMets DL. The Beaver Dam Eye Study: the relation of age-related maculopathy to smoking. Am J Epidemiol 1993;137: 190–200. [17] Smith W, Mitchell P, Leeder SR. Smoking and age-related maculopathy. The Blue Mountains Eye Study. Arch Ophthalmol 1996;114:1518–23. [18] Vingerling JR, Hofman A, Grobbee DE, de Jong PT. Age-related macular degeneration and smoking. The Rotterdam Study. Arch Ophthalmol 1996;114: 1193–6. [19] Klein R. Overview of progress in the epidemiology of age-related macular degeneration. Ophthalmic Epidemiol 2007;14:184–7. [20] Tomany SC, Wang JJ, van Leeuwen R, Klein R, Mitchell P, Vingerling JR, et al. Risk factors for incident age-related macular degeneration. Ophthalmology 2004;111:1280–7. [21] Smith W, Mitchell P, Leeder SR, Wang JJ. Plasma fibrinogen levels, other cardiovascular risk factors, and age-related maculopathy: the Blue Mountains Eye Study. Arch Ophthalmol 1998;116:583–7. [22] Parekh N, Chappell RJ, Millen AE, Albert DM, Mares JA. Association between vitamin D and age-related macular degeneration in the Third National Health and Nutrition Examination Survey, 1988 Through 1994. Arch Ophthalmol 2007;125:661–9. [23] Merle BM, Silver RE, Rosner B, Seddon JM. Adherence to a Mediterranean diet, genetic susceptibility, and progression to advanced macular degeneration: a prospective cohort study. Am J Clin Nutr 2015;102:1196– 206. [24] Hogg RE, Woodside JV, McGrath A, Young IS, Viogue JL, Chakravarthy U, et al. Mediterranean diet score and its association with age-related macular degeneration. Ophthalmology 2017;124:82–9.

12

S. Nunes et al. / Nutrition 51-52 (2018) 6–12

[25] Carneiro Â, Andrade JP. Nutritional and lifestyle interventions for agerelated macular degeneration: a review. Oxid Med Cell Longev 2017;2017: 6469138. [26] Cachulo Mda L, Lobo C, Figueira J, Ribeiro L, Lains I, Vieira A, et al. Prevalence of age-related macular degeneration in Portugal: the Coimbra Eye Study-Report 1. Ophthalmologica 2015;233:119–27. [27] Klaver CC, Assink JJ, van Leeuwen R, Wolfs RC, Vingerling JR, Stijnen T, et al. Incidence and progression rates of age-related maculopathy: the Rotterdam Study. Invest Ophthalmol Vis Sci 2001;42:2237–41. [28] Bird AC, Bressler NM, Bressler SB, Chisholm IH, Coscas G, Davis MD, et al. An international classification and grading system for age-related maculopathy and age-related macular degeneration. The International ARM Epidemiological Study Group. Surv Ophthalmol 1995;39:367–74. [29] Willett W, Lenart E. Reproducibility and validity of food-frequency questionnaires. Nutritional epidemiology. 2nd ed. New York, NY: Oxford University Press; 1998. p. 101–47. [30] Lopes C. Validação de um questionário semi-quantitativo de frequência alimentar. Universidade do Porto; 2000. [31] Ministério da Saúde. Instituto Nacional de Saúde Dr. Ricardo Jorge. Centro de Segurança Alimentar e Nutrição-Tabela da composição de alimentos. Compilação Ilda Martins; concepção, elaboração e coordenação da edição Andreia Porto e Luísa Oliveira. Lisboa: INSA; 2007. [32] Féart C, Samieri C, Rondeau V, Amieva H, Portet F, Dartiguest JF, et al. Adherence to a Mediterranean diet, cognitive decline, and risk of dementia. JAMA 2009;302:638–48. [33] Trichopoulou A, Kouris-Blazos A, Wahlqvist ML, Gnardellis C, Lagiou P, Polychronopoulos E, et al. Diet and overall survival in elderly people. BMJ 1995;311:1457–60. [34] Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med 2003;348: 2599–608. [35] Chiu CJ, Chang ML, Zhang FF, Li T, Gensler G, Schleicher M, et al. The relationship of major American dietary patterns to age-related macular degeneration. Am J Ophthalmol 2014;158:118–27, e1. [36] Amirul Islam FM, Chong EW, Hodge AM, Guymer RH, Aung KZ, Makeyeva GA, et al. Dietary patterns and their associations with age-related macular degeneration: the Melbourne collaborative cohort study. Ophthalmology 2014;121:1428–34, e2. [37] Beatty S, Koh H, Phil M, Henson D, Boulton M. The role of oxidative stress in the pathogenesis of age-related macular degeneration. Surv Ophthalmol 2000;45:115–34. [38] Mozaffarieh M, Sacu S, Wedrich A. The role of the carotenoids, lutein and zeaxanthin, in protecting against age-related macular degeneration: a review based on controversial evidence. Nutr J 2003;2:20.

[39] Vinding T, Appleyard M, Nyboe J, Jensen G. Risk factor analysis for atrophic and exudative age-related macular degeneration. An epidemiological study of 1000 aged individuals. Acta Ophthalmol 1992;70:66–72. [40] Brown JC, Winters-Stone K, Lee A, Schmitz KH. Cancer, physical activity, and exercise. Compr Physiol 2012;2:2775–809. [41] Ballard-Barbash R, Friedenreich CM, Courneya KS, Siddiqi SM, McTiernan A, Alfano CM. Physical activity, biomarkers, and disease outcomes in cancer survivors: a systematic review. J Natl Cancer Inst 2012;104:815–40. [42] Stasi R, Abriani L, Beccaglia P, Terzoli E, Amadori S. Cancer-related fatigue: evolving concepts in evaluation and treatment. Cancer 2003;98:1786– 801. [43] VanderBeek BL, Zacks DN, Talwar N, Nan B, Stein JD. Role of statins in the development and progression of age-related macular degeneration. Retina 2013;33:414–22. [44] Vincent HK, Innes KE, Vincent KR. Oxidative stress and potential interventions to reduce oxidative stress in overweight and obesity. Diabetes Obes Metab 2007;9:813–39. [45] Ji LL, Gomez-Cabrera M-C, Vina J. Role of free radicals and antioxidant signaling in skeletal muscle health and pathology. Infect Disord Drug Targets 2009;9:428–44. [46] Adams SA, Matthews CE, Ebbeling CB, Moore CG, Cunningham JE, Fulton J, et al. The effect of social desirability and social approval on self-reports of physical activity. Am J Epidemiol 2005;161:389–98. [47] Ahn GE, Chmiel JS, Dunlop DD, Helenowski IB, Semanik PA, Song J, et al. Self-reported and objectively measured physical activity in adults with systemic lupus erythematosus. Arthritis Care Res 2015;67:701–7. [48] Meyers KJ, Liu Z, Millen AE, Iyengar SK, Blodi BA, Johnson E, et al. Joint associations of diet, lifestyle, and genes with age-related macular degeneration. Ophthalmology 2015;122:2286–94. [49] Williams PT. Prospective study of incident age-related macular degeneration in relation to vigorous physical activity during a 7-year follow-up. Invest Ophthalmol Vis Sci 2009;50:101–6. [50] Knudtson MD, Klein R, Klein BE. Physical activity and the 15-year cumulative incidence of age-related macular degeneration: the Beaver Dam Eye Study. Br J Ophthalmol 2006;90:1461–3. [51] Loprinzi PD, Swenor BK, Ramulu PY. Age-related macular degeneration is associated with less physical activity among U.S. adults: cross-sectional study. PLoS ONE 2015;10:e0125394. [52] Gopinath B, Liew G, Burlutsky G, Mitchell P. Physical activity and the 15year incidence of age-related macular degeneration. Invest Ophthalmol Vis Sci 2014;55:7799–803. [53] Shrank WH, Patrick AR, Brookhart MA. Healthy user and related biases in observational studies of preventive interventions: a primer for physicians. J Gen Intern Med 2011;26:546–50.