Antioxidant Capacity of Diet and Risk of Cataracts
Rautiainen S, Lindblad BE, Morgenstern R, Wolk A. Total antioxidant capacity of the diet and risk
of age-related cataract: a population-based prospective cohort of women. JAMA
Ophthalmol. Published online December 26, 2013. doi:10.1001/jamaophthalmol.2013.6241.
Rautiainen S, Lindblad BE, Morgenstern R, Wolk A. Total Antioxidant Capacity of the Diet and Risk of Age-Related CataractA Population-Based Prospective Cohort of Women. JAMA Ophthalmol. 2014;132(3):247-252. doi:10.1001/jamaophthalmol.2013.6241
To our knowledge, no previous epidemiologic study has investigated the association between all antioxidants in the diet and age-related cataract. The total antioxidant capacity (TAC) concept aims to measure the capacity from all antioxidants in the diet by also taking synergistic effects into account.
To investigate the association between the TAC of the diet and the incidence of age-related cataract in a population-based prospective cohort of middle-aged and elderly women.
Design, Setting, and Participants
Questionnaire-based nutrition survey within the prospective Swedish Mammography Cohort study, which included 30 607 women (aged 49-83 years) who were observed for age-related cataract incidence for a mean of 7.7 years.
The TAC of the diet was estimated using a database of foods analyzed with the oxygen radical absorbance capacity assay.
Main Outcomes and Measures
Information on incident age-related cataract diagnosis and extraction was collected through linkage to registers in the study area.
There were 4309 incident cases of age-related cataracts during the mean 7.7 years of follow-up (234 371 person-years). The multivariable rate ratio in the highest quintile of the TAC of the diet compared with the lowest was 0.87 (95% CI, 0.79-0.96; P for trend = .03). The main contributors to dietary TAC in the study population were fruit and vegetables (44.3%), whole grains (17.0%), and coffee (15.1%).
Conclusions and Relevance
Dietary TAC was inversely associated with the risk of age-related cataract. Future studies examining all antioxidants in the diet in relation to age-related cataract are needed to confirm or refute our findings.
The prevalence of cataract is high in the elderly and is the leading cause of blindness worldwide.1 Surgical extraction, the only available treatment, is a major medical cost for several countries. Oxidative stress is thought to be involved in the development of age-related cataract by causing damage to lens proteins and lipids in the epithelium of the lens.2- 4 Several factors are associated with increased oxidation of the lens, such as aging, UV light, smoking, and corticosteroid use.5,6 Furthermore, factors such as abdominal obesity7 and hormone therapy8 generate proinflammatory components, contributing to increased production of reactive oxygen species,4,9,10 and have been associated with increased risk of cataract.
To our knowledge, no previous study has investigated the role of all antioxidants in the diet in relation to age-related cataract. Previous randomized clinical trials and observational studies have focused only on individual antioxidants such as vitamin C, vitamin E, and different carotenoids and have reported inconsistent results.11 A recent Cochrane review concluded that there is no evidence from randomized clinical trials that supplements of beta-carotene, vitamin C, or vitamin E prevent or slow the progression of age-related cataract.12 However, besides beta-carotene, vitamin C, and vitamin E, a much wider range of antioxidants is in the diet.
The total antioxidant capacity (TAC) concept aims to measure the capacity from all dietary antioxidants in a single estimate by taking into account synergistic effects between compounds. However, to our knowledge, no previous study has investigated a potential role of the TAC of the diet in age-related cataract prevention. Therefore, in this study, we aimed to prospectively study the association between all dietary antioxidants by summarizing the TAC of foods and the incidence of age-related cataract in middle-aged and elderly women from the population-based Swedish Mammography Cohort (SMC).
The SMC was established from 1987 through 1990 among women residing in the Uppsala and Västmanland counties in central Sweden. From 1987 through 1990, all women born between 1914 and 1948 received a mailed questionnaire about diet and other lifestyle factors; 73.8% completed the questionnaire. A new questionnaire was sent in 1997 to all 56 030 eligible cohort members. The extended 1997 questionnaire included questions about diet and vitamin supplement use as well as a wide range of lifestyle and clinical factors, including cigarette smoking; alcohol consumption; physical activity; use of some medications, including corticosteroids and hormone therapy; and diagnosis of hypertension and type 2 diabetes mellitus. The 1997 questionnaire was completed by 38 984 women (70%) and served as the sole source of nutrition and lifestyle risk factor data for the present age-related cataract study.
We excluded women who had missing identification number information, a diagnosis of cancer (except nonmelanoma skin cancer, identified through the National Cancer Register), and a cataract diagnosis and/or extraction before baseline in 1997. We also excluded women who moved from the study area (identified through the Swedish Population Register) before baseline. To avoid influence from changes in dietary habits due to disease diagnosis, we excluded women with a history of myocardial infarction, angina pectoris, and stroke (identified through the National Inpatient Register at the National Board of Health and Welfare) as well as a history of type 2 diabetes mellitus at baseline (self-reported and identification through the inpatient registry). We also excluded women who had high or low energy intake (±3 SD from the mean value for loge- transformed energy). After these exclusions, the study cohort included 30 607 women at the start of follow-up for cataract diagnosis and cataract extractions in September 15, 1997. The regional ethical review board at Karolinska Institutet approved this investigation, and return of the self-administered questionnaire was considered to imply informed consent to participate in the study.
Women completed a 96-item food frequency questionnaire (FFQ), in which they were asked how often, on average, they had consumed each type of food or beverage during the past year. There were 8 predefined response categories, ranging from “never/seldom” to “3 or more times per day.” Open-ended questions were used for commonly consumed foods (eg, coffee, tea, and bread). The calculation of the TAC estimates is described in detail elsewhere.13 In brief, we constructed the TAC estimates by using a database of the most common foods in the United States, analyzed with the oxygen radical absorbance capacity (ORAC) assay.14- 16 Dietary TAC was calculated by multiplying the mean frequency of consuming each food item by ORAC content (micromole Trolox equivalents [TEs]) of age-specific portion sizes. All major food sources contributing to the TAC had available ORAC values. Dietary TAC estimates were adjusted for total energy intake with the residual method.17 The validity of FFQ-based ORAC estimates compared with ORAC in plasma has been measured in 108 women from the SMC.13 Because antioxidants in coffee and tea have shown to be poorly absorbed, we took into account absorption (6% for coffee and 4% for tea)18 when calculating the TAC of the diet. The Pearson correlation between the TAC of the diet and plasma ORAC was 0.31.19 Food items contributing to the TAC of the diet compared with food records showed reasonably good validity; for example, r ranged from 0.4 to 0.7 for individual fruit and vegetable items and was 0.6 for coffee consumption (A. Wolk, unpublished data, 1995).
Between September 15, 1997, and December 31, 2005, there were 4309 incident cases of age-related cataracts (cataract diagnosis and/or cataract extraction) among women in the cohort. Cataract diagnoses and extractions were ascertained by linkage to the National Outpatient and Inpatient Registers at the National Board of Health and Welfare and to the “Swedish National Day-Surgery Register” (International Classification of Diseases, Tenth Revision, code H25 and operation codes CJC, CJD, CJE, and CJG). As a complement, women were also matched against local cataract extraction registers at both public and private clinics in the study area. Cataracts considered congenital or secondary to ocular trauma, intraocular inflammation, or previous intraocular surgery were excluded. According to the Swedish National Cataract Register, most women had a preoperative visual acuity in the cataract-containing eye of less than 20/40 Snellen equivalents, which corresponds to difficulties in driving. The mean preoperative visual acuity in the cataract-containing eye was 20/70 Snellen equivalents during the study period, which corresponds to difficulties in reading a newspaper. Approximately one-fourth of the women were legally blind (visual acuity 20/200 Snellen equivalents).9
Women in the cohort were observed until the date of cataract diagnosis, cataract extraction, death, migration out of the study area, or the end of follow-up (December 31, 2005), whichever came first. Women were categorized into quintiles of the TAC of the diet (TE/d). A Cox proportional hazards regression model was used to estimate relative risks as rate ratios (RRs), with 95% CIs,20 using the PHREG procedure in SAS (version 9.1; SAS Institute, Inc). All relative risks were adjusted for potential risk factors, including age (5-year age groups: ≤52, 53-57, 58-62, 63-67, 68-72, 73-77, and ≥78 years), smoking (never, mean lifetime ≤10, or >10 cigarettes per day), waist circumference (<80 cm or ≥80 cm), educational level (<10, 10-12, or >12 years), hypertension (yes or no), postmenopausal hormone therapy (never, past, or current), corticosteroids (yes or no), dietary supplement use (yes or no), alcohol consumption (g/d in quartiles), and energy intake (kcal/d). We tested the proportional hazard assumption by calculating scaled Schoenfeld’s residuals and did not find evidence of violating this assumption.
The RR of age-related cataract per 4000-µmol/TE/d increment of ORAC (corresponding to approximately 1 SD in the cohort) was corrected for bias due to measurement error in the TAC estimates with the regression calibration method correcting for both random and systematic error.21 Based on a validation study of 108 women from the SMC,13 the validity coefficient between dietary TAC and TAC in plasma (r = 0.3) was used.
To evaluate whether the association between the TAC of the diet and age-related cataract varied by a potentially increased presence of reactive oxygen species, we examined differences in the risk of age-related cataract within categories of age (<65 or ≥65 years), waist circumference (<80 cm or ≥80 cm), smoking (nonsmokers or current smokers), and corticosteroid use (never or ever). To assess trends across quintiles, we used the median value of each category to create a single continuous variable. Interaction tests on all categories with the likelihood test were performed. All P values shown are 2-sided. P < .05 was considered statistically significant.
During the mean 7.7 years of follow-up, we identified 4309 incident cases of age-related cataract (234 371 person-years), which corresponds to a rate of 184 cases per 10 000 person-years. The number of cataract extractions each year of follow-up ranged between 1881 and 2298 extractions per 100 000 women. Baseline characteristics of the women are presented in Table 1. Women in the highest quintile of the TAC of the diet compared with those in the lowest quintile were more likely to have more than 12 years of education, less likely to smoke, and more likely to use dietary supplements. According to dietary factors, women in the highest quintile had a higher consumption of fruits and vegetables (164%), whole grains (15.2%), and coffee (34.6%). The major contributors to the TAC of the diet were fruit and vegetables (44.3%), whole grains (17.0%), and coffee (15.1%).
The association between the TAC of the diet and risk of age-related cataract is shown in Table 2. In the multivariable analysis, women in the highest quintile of the TAC of the diet compared with the lowest had a 12.8% (95% CI, 3.9%-20.9%; P for trend = .03) lower risk of age-related cataract.
The multivariable-adjusted RR of age-related cataract for an increment of 4000 ORAC TE/d (corresponding to approximately 1 SD in the study population) was 0.96 (95% CI, 0.93-0.99); the corrected RR (taking into account measurement error in the dietary TAC estimates) was 0.87. A 6 increment of 4000 ORAC TE is equivalent to approximately 1 to 2 apples or 2 peppers.
In sensitivity analyses, we investigated the association between dietary TAC and the risk of cataract among women who reported no use of dietary supplements at baseline. We observed an RR of 0.87 (95% CI, 0.73-1.03) in the highest quintile of dietary TAC compared with the lowest.
In further analyses, we examined whether the association between the TAC of the diet and age-related cataract varied by potential risk factors related to oxidative stress such as age, smoking, waist circumference, and corticosteroid use (Table 3). The inverse association was somewhat stronger among women 65 years or younger (P for interaction = .07) and corticosteroid users (P for interaction = .27). The inverse association was similar across subgroups of smoking (P for interaction = .39) and waist circumference (P for interaction = .52).
In this population-based prospective cohort study of middle-aged and elderly women, we observed that women in the highest quintile of the TAC of the diet compared with the lowest had a 12.8% lower risk of incident age-related cataract. The association was consistent across subgroups defined by potential risk factors for age-related cataract and sources of oxidative stress. The major contributors to the TAC of the diet were fruit and vegetables (44.3%), whole grains (17.0%), and coffee (15.1%).
To our knowledge, no previous study has examined the association between the TAC of the diet and the risk of age-related cataract. The lens contains high concentrations of dietary antioxidants, such as lutein, zeaxanthin, vitamin C, and vitamin E,22,23 and therefore these compounds are hypothesized to be of major importance in preventing cataract development. However, observational studies of these specific antioxidants from the diet, supplements, and plasma concentrations, as well as intervention trials, have reported inconsistent results.11 In the diet, much wider ranges of antioxidants are present than those studied in the above-mentioned studies.
Antioxidants may prevent cataract development by inhibiting lens opacification. An excessive amount of reactive oxygen species damages the lens proteins. Exogenous (dietary) antioxidants and endogenous antioxidants (eg, antioxidant enzymes) may protect the lens against oxidative stress by several mechanisms.24 An estimate of the TAC of the diet gives a single summary measure, including the antioxidant effect of not only lutein, zeaxanthin, vitamin C, and vitamin E but also many other food constituents with antioxidant properties (eg, polyphenols) and takes synergistic effects between these antioxidants into account.25
These findings that the TAC of the diet is inversely related to age-related cataract do not apply to intakes of antioxidant supplements. We have previously observed in the same SMC cohort,26 as well as in the Cohort of Swedish Men,27 that high-dose vitamin C supplements (approximately 1000 mg) are associated with an increased risk of age-related cataract. Multivitamins, usually containing lower doses of vitamin C (approximately 60 mg), are not associated with cataract risk.
This study has several strengths, which include the population-based design, the high response rate, and nearly complete follow-up of an almost homogeneous ethnic population. The incidence rates of cataract extractions at each year of follow-up were in the same range as those reported for women in the entire Swedish population in these specific age groups.28 The large number of cases allowed us to perform subgroup analyses of women at higher risk of cataract to identify potential effect modifiers with reasonable statistical power. There are some possible limitations with this study. We were not able to use a standardized eye examination, performed preferably by 1 ophthalmologist, of all women in this large cohort and therefore cannot exclude the possibility of misclassifying the outcome. However, lens opacity is clinically relevant only if the patient’s visual function has declined substantially. We, therefore, focused on the degree of cataract severe enough to cause visual impairment (20/40 Snellen equivalents) affecting activities of daily life.29 Thus, in this study, we focused on the outcome with the greatest clinical and public health importance. We cannot exclude that women with cataract diagnosis could be more health conscious and more likely to seek medical help for visual impairment. However, in the Swedish health care system, everyone has the same access, with a patient charge of less than $50 per operation. Moreover, we had no information on cataract subtype and cannot exclude that the inverse association between the TAC of the diet and age-related cataract was restricted to a specific subtype. We were not able to control for possible effects of sunlight exposure. However, this study was carried out in northern Europe, where the risk associated with sunlight exposure is not likely to be large, although we cannot rule out its effect on our risk estimates. Dietary TAC was measured using FFQs; therefore, we cannot exclude some measurement error in the TAC of the diet. However, due to the prospective design, the misclassification of exposure is most likely to be nondifferential and therefore lead to attenuation of the observed association. In addition, when we corrected for measurement error, an even lower relative risk was observed. Moreover, we had only 1 measurement of dietary TAC at baseline and therefore cannot rule out whether women changed their TAC intake during the follow-up. We observed, however, a reasonably good correlation (r = 0.5) between dietary TAC assessed from the FFQ administrated in 1987 and dietary TAC from the 1997 FFQ, suggesting that these women may not change their dietary habits dramatically over time. As with all observational studies, we were not able to rule out bias due to residual or unmeasured confounding. We adjusted for several potential confounders such as smoking, body mass index, physical activity, and educational level, and the results remained statistically significant. The association between the TAC of the diet and age-related cataract incidence was studied in a population-based cohort of middle-aged and elderly Swedish women, and thus our results can be generalized only to this particular population. However, similar associations may be expected among other populations.
In summary, these prospective data indicate that a high TAC of the diet is associated with a lower risk of age-related cataract among middle-aged and elderly women. More studies are needed to investigate the association between all antioxidants in the diet and age-related cataract and whether a potential preventive role of the TAC of the diet is restricted to a specific cataract subtype.
Submitted for Publication: April 3, 2013; final revision received July 3, 2013; accepted July 26, 2013.
Corresponding Author: Susanne Rautiainen, PhD, Institute of Environmental Medicine, Karolinska Institutet, Nobels väg 13, 17177 Stockholm, Sweden (email@example.com).
Published Online: December 26, 2013. doi:10.1001/jamaophthalmol.2013.6241.
Author Contributions: Drs Rautiainen and Wolk had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Rautiainen, Wolk.
Acquisition of data: Rautiainen, Wolk.
Analysis and interpretation of data: All authors.
Drafting of the manuscript: Rautiainen.
Critical revision of the manuscript for important intellectual content: Lindblad, Morgenstern, Wolk.
Statistical analysis: Rautiainen.
Obtained funding: Wolk.
Administrative, technical, and material support: Wolk.
Study supervision: Morgenstern, Wolk.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported by the Swedish Research Council/Medicine and Swedish Council for Working Life and Social Research, Stockholm, Sweden.
Role of the Sponsor: The funding organizations had no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.