Advertisement

Total Antioxidant Capacity from Diet and Risk of Myocardial Infarction: A Prospective Cohort of Women

      Abstract

      Background

      There are no previous studies investigating the effect of all dietary antioxidants in relation to myocardial infarction. The total antioxidant capacity of diet takes into account all antioxidants and synergistic effects between them. The aim of this study was to examine how total antioxidant capacity of diet and antioxidant-containing foods were associated with incident myocardial infarction among middle-aged and elderly women.

      Methods

      In the population-based prospective Swedish Mammography Cohort of 49-83-year-old women, 32,561 were cardiovascular disease-free at baseline. Women completed a food-frequency questionnaire, and dietary total antioxidant capacity was calculated using oxygen radical absorbance capacity values. Information on myocardial infarction was identified from the Swedish Hospital Discharge and the Cause of Death registries. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated using Cox proportional hazard models.

      Results

      During the follow-up (September 1997-December 2007), we identified 1114 incident cases of myocardial infarction (321,434 person-years). In multivariable-adjusted analysis, the HR for women comparing the highest quintile of dietary total antioxidant capacity to the lowest was 0.80 (95% CI, 0.67-0.97; P for trend = 0.02). Servings of fruit and vegetables and whole grains were nonsignificantly inversely associated with myocardial infarction.

      Conclusions

      These data suggest that dietary total antioxidant capacity, based on fruits, vegetables, coffee, and whole grains, is of importance in the prevention of myocardial infarction.

      Keywords

      SEE RELATED EDITORIAL p. 947
      Coronary heart disease is a major cause of death in women.
      • Thomson M.J.
      • Puntmann V.
      • Kaski J.C.
      Atherosclerosis and oxidant stress: the end of the road for antioxidant vitamin treatment?.
      High levels of reactive oxygen species and reactive nitrogen species have been implicated in the initiation and progression of atherosclerosis,
      • Thomson M.J.
      • Puntmann V.
      • Kaski J.C.
      Atherosclerosis and oxidant stress: the end of the road for antioxidant vitamin treatment?.
      • Patel R.P.
      • Moellering D.
      • Murphy-Ullrich J.
      • et al.
      Cell signaling by reactive nitrogen and oxygen species in atherosclerosis.
      • Darley-Usmar V.
      • Halliwell B.
      Blood radicals: reactive nitrogen species, reactive oxygen species, transition metal ions, and the vascular system.
      the underlying cause of coronary heart disease. Well-established coronary heart disease risk factors such as smoking
      • Ambrose J.A.
      • Barua R.S.
      The pathophysiology of cigarette smoking and cardiovascular disease: an update.
      and aging
      • Csiszar A.
      • Wang M.
      • Lakatta E.G.
      • Ungvari Z.
      Inflammation and endothelial dysfunction during aging: role of NF-kappaB.
      are associated with increased reactive oxygen species and reactive nitrogen species production, as are obesity,
      • de Ferranti S.
      • Mozaffarian D.
      The perfect storm: obesity, adipocyte dysfunction, and metabolic consequences.
      hypertension,
      • Touyz R.M.
      Reactive oxygen species, vascular oxidative stress, and redox signaling in hypertension: what is the clinical significance?.
      and excessive alcohol consumption.
      • Das S.K.
      • Vasudevan D.M.
      Alcohol-induced oxidative stress.
      Antioxidants are proposed to play a key role in mitigating the atherosclerotic process by scavenging reactive oxygen species and reactive nitrogen species.
      • Thomson M.J.
      • Puntmann V.
      • Kaski J.C.
      Atherosclerosis and oxidant stress: the end of the road for antioxidant vitamin treatment?.
      Accordingly, consumption of fruits and vegetables, major sources of antioxidants, have been inversely related to coronary heart disease.
      • Dauchet L.
      • Amouyel P.
      • Hercberg S.
      • Dallongeville J.
      Fruit and vegetable consumption and risk of coronary heart disease: a meta-analysis of cohort studies.
      By contrast, the use of high-dose single-antioxidant supplements does not protect against coronary heart disease and may even increase mortality, as shown by randomized controlled trials.
      • Sesso H.D.
      • Buring J.E.
      • Christen W.G.
      • et al.
      Vitamins E and C in the prevention of cardiovascular disease in men: the Physicians' Health Study II randomized controlled trial.
      • Lee I.M.
      • Cook N.R.
      • Gaziano J.M.
      • et al.
      Vitamin E in the primary prevention of cardiovascular disease and cancer: the Women's Health Study: a randomized controlled trial.
      • Lonn E.
      • Bosch J.
      • Yusuf S.
      • et al.
      Effects of long-term vitamin E supplementation on cardiovascular events and cancer: a randomized controlled trial.
      • Cook N.R.
      • Albert C.M.
      • Gaziano J.M.
      • et al.
      A randomized factorial trial of vitamins C and E and beta carotene in the secondary prevention of cardiovascular events in women: results from the Women's Antioxidant Cardiovascular Study.
      • Yusuf S.
      • Dagenais G.
      • Pogue J.
      • et al.
      Vitamin E supplementation and cardiovascular events in high-risk patients The Heart Outcomes Prevention Evaluation Study Investigators.
      • Rapola J.M.
      • Virtamo J.
      • Ripatti S.
      • et al.
      Randomised trial of alpha-tocopherol and beta-carotene supplements on incidence of major coronary events in men with previous myocardial infarction.
      Thus, high-dose single-antioxidant supplements are not a good substitute for the very complex antioxidant network of thousands of compounds in foods, present at concentrations far below those used in most randomized controlled trials.
      • Dietary antioxidants are hypothesized to protect against coronary heart disease; however, no previous study has investigated the association between all antioxidants present in the diet and myocardial infarction.
      • Total antioxidant capacity measures in a single value all antioxidants present in diet and the synergistic effects between them.
      • A diet high in total antioxidant capacity, based on fruits, vegetables, coffee, and whole grains, was associated with lower myocardial infarction incidence.
      To our knowledge, no study has investigated the overall effect of the complex antioxidant network in diet in relation to coronary heart disease. The total antioxidant capacity measures, in one single value, the free-radical-reducing capacity of all antioxidants present in foods and the synergistic effects between these substances. The aim of the present study was to examine how total antioxidant capacity of diet and antioxidant-containing foods were associated with incident myocardial infarction among women from the population-based, prospective Swedish Mammography Cohort.

      Methods

       Ethics Statement

      The Regional Ethical Review Board at Karolinska Institutet (Stockholm, Sweden) approved this investigation, and return of the self-administrated questionnaire was considered to imply informed consent to participate in the study.

       Study Population

      The Swedish Mammography Cohort was established between 1987 and 1990 among women residing in the Uppsala and Västmanland counties in central Sweden. All women born 1914-1948 were sent a questionnaire concerning diet, educational level, weight and height, and reproductive factors; 74% completed the questionnaire. To expand exposure data, a new questionnaire was sent in 1997 to all 56,030 eligible cohort members. The expanded 1997 questionnaire included questions on diet and questions on all major lifestyle factors, history of diseases, and use of some medications. Because more information on potential risk factors for myocardial infarction was collected in the 1997 questionnaire, completed by 38,984 women (70%), this questionnaire served as the baseline in the present study.
      We excluded women having history of cancer (except nonmelanoma skin cancer) and women reporting extreme total energy intake (±3 SD from the mean value for loge-transformed energy). We also excluded women with cardiovascular diseases (identified through the Swedish Hospital Discharge Registry, International Statistical Classification of Disease, 10th Revision (ICD-10); code I11.0, I20-25, I50, and I60-69)
      Socialstyrelsen: Svensk version av ICD-10 (Swedish board of health and welfare: Swedish version of ICD-10).
      and diabetes (self-reported or recorded in the Swedish Hospital Discharge Registry) because these diagnoses may lead to change in dietary habits. The remaining cohort of 32,561 women was followed from September 1997 through December 2007.

       Assessment of Food-frequency Questionnaire-based Total Antioxidant Capacity Estimates and Other Exposures

      Women completed a 96-item food-frequency questionnaire on which they were asked how often, on average, they consumed each type of food or beverage during the last year. There were 8 predefined response categories, ranging from “never/seldom” to “3 or more times per day.” Open-ended questions were used for foods and beverages commonly consumed (eg, bread, coffee and tea). The calculation of total antioxidant capacity estimates is described in detail elsewhere.
      • Rautiainen S.
      • Serafini M.
      • Morgenstern R.
      • et al.
      The validity and reproducibility of food-frequency questionnaire-based total antioxidant capacity estimates in Swedish women.
      Briefly, we calculated estimates of total antioxidant capacity from diet using a database of the most common foods in the United States analyzed with the oxygen radical absorbance capacity (ORAC) assay.
      • Wu X.
      • Beecher G.R.
      • Holden J.M.
      • et al.
      Lipophilic and hydrophilic antioxidant capacities of common foods in the United States.
      • Prior R.L.
      • Hoang H.
      • Gu L.
      • et al.
      Assays for hydrophilic and lipophilic antioxidant capacity (oxygen radical absorbance capacity (ORAC(FL))) of plasma and other biological and food samples.
      • Sanchez-Moreno C.
      • Cao G.
      • Ou B.
      • Prior R.L.
      Anthocyanin and proanthocyanidin content in selected white and red wines Oxygen radical absorbance capacity comparison with nontraditional wines obtained from highbush blueberry.
      ORAC measures the antioxidant capacity of diet to reduce free radicals, taking into account the synergism between compounds. In the 96-item food-frequency questionnaire, there were 31 items (including 17 fruit and vegetable items) with available ORAC values. Total antioxidant capacity of diet and nutrient intakes (saturated fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids) were calculated by multiplying the average frequency of consumption of each food by ORAC (μmol Trolox [Hoffman-LaRoche, Basel, Switzerland] equivalents (TE)/100 g) or nutrient content of age-specific portion sizes. Because antioxidants in coffee and tea have been shown to be poorly absorbed, we took into account absorption (6% for coffee and 4% for tea)
      • Natella F.
      • Nardini M.
      • Giannetti I.
      • et al.
      Coffee drinking influences plasma antioxidant capacity in humans.
      when calculating total antioxidant capacity of diet. The correlation between total antioxidant capacity of diet and plasma ORAC was 0.31. Food items contributing to total antioxidant capacity of diet, as compared with food records, showed reasonable good validity; for example, for fruit and vegetable consumption, r ranged from 0.4 to 0.7 for individual fruit and vegetable items; for tea consumption, r was 0.8; and for dietary fiber, r was 0.5 (Wolk A, unpublished data). The total antioxidant capacity from diet was adjusted for total energy intake using the residual method.
      • Willett W.C.
      Nutritional Epidemiology.
      Body mass index was calculated by dividing reported weight (kg) by reported height (m2). Self-reported weight and height is highly correlated with measured values in Swedish women (r = 0.9 and r = 1.0, respectively).
      • Kuskowska-Wolk A.
      • Bergstrom R.
      • Bostrom G.
      Relationship between questionnaire data and medical records of height, weight and body mass index.
      Physical activity levels were estimated by multiplying the reported duration of 5 predefined activities (occupation, housework, walking or cycling, leisure-time exercise, and inactive leisure time) by the intensity of these activities and expressed as multiples of the metabolic equivalent per day (kcal kg−1 × h−1) of sitting quietly for 1 hour. The validity of the reported total physical activity against activity records in these women were satisfactory (r = 0.6).
      • Orsini N.
      • Bellocco R.
      • Bottai M.
      • et al.
      Validity of self-reported total physical activity questionnaire among older women.

       Identification of Myocardial Infarction and Follow-up of the Cohort

      All women were followed from September 1997 (baseline) through December 2007. Cases of nonfatal and fatal myocardial infarction (I21) were ascertained through linkage via the national registration number to the Swedish Hospital Discharge Registry and the Cause of Death Registry, which are considered nearly complete. The registries for 1987 and 1995 were thoroughly validated and revealed high sensitivity (94%) and positive predictive value (86%) for myocardial infarction.
      • Hammar N.
      • Alfredsson L.
      • Rosen M.
      • et al.
      A national record linkage to study acute myocardial infarction incidence and case fatality in Sweden.

       Statistical Analysis

      All women were followed until the date of myocardial infarction, death, or the end of follow-up (December 31, 2007), whichever came first. Women were categorized into 4 categories of fruit and vegetable consumption, whole grain consumption, and coffee consumption, as well as quintiles of total antioxidant capacity of diet. Cox proportional hazards models with age as the time-scale were used to estimate hazard ratios (HR), with 95% confidence intervals (CI)
      • Cox D.
      Regression models and life-tables.
      using the PHREG command in SAS (version 9.2; SAS Institute, Inc., Cary, NC). All HRs were adjusted for potential risk factors, including body mass index (<18.5, 18.5-24.9, 25-29.9, 30+ kg/m2), smoking (never, past, current [≤10, >10 cigarettes/day]), alcohol consumption (gram/day, continuous), energy intake (calories/day, continuous), physical activity (metabolic equivalent-hours in quartiles), educational level (<10, 10-12, >12 years), hormone replacement therapy use (yes/no), aspirin use (yes/no), hypertension (yes/no), hypercholesterolemia (yes/no), family history of myocardial infarction (yes/no), and dietary supplement use (multivitamin use, use of other supplements than multivitamin, no supplements use). In additional analyses, we further adjusted for fruit and vegetable consumption (gram/day, continuous) as well as intakes of saturated fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids (all continuous, gram/day). Missing data on a covariate was treated as a separate category. To assess trends across quintiles, we used the median value of each category to create a single continuous variable. A P-value for the linearity assumption between total antioxidant capacity and myocardial infarction was obtained by testing whether the quadratic term was equal to zero. The proportional hazards assumption was assessed by calculating scaled Schoenfeld residuals, and we did not find evidence of violation of this assumption.
      We also corrected the HR of myocardial infarction per 4000 μmol/TE/day increment of ORAC (corresponding to approximately 1 standard deviation [SD] in the cohort) for bias due to dietary measurement error with the regression calibration method correcting for both random and systematic error.
      • Spiegelman D.
      • McDermott A.
      • Rosner B.
      Regression calibration method for correcting measurement-error bias in nutritional epidemiology.
      Based on a validation study of 108 women from the Swedish Mammography Cohort,
      • Rautiainen S.
      • Serafini M.
      • Morgenstern R.
      • et al.
      The validity and reproducibility of food-frequency questionnaire-based total antioxidant capacity estimates in Swedish women.
      we used the validity coefficient between total antioxidant capacity from diet and plasma (r = 0.3). All P values shown are 2-sided. P-values <.05 were considered statistically significant.
      To evaluate whether the effect of total antioxidant capacity from diet varied by risk factors for myocardial infarction, we performed subgroup analyses by age (<65 years/≥65 years), body mass index (≤25/>25), smoking (nonsmokers/current smokers), and multivitamin supplement use (non-supplement users/multivitamin users). The likelihood ratio test was used to perform interaction tests. All P values shown are 2-sided. P-values <.05 were considered statistically significant.

      Results

      During the average 9.9 years of follow-up (321,434 person-years), we identified 1114 cases of incident myocardial infarction (the average age of first myocardial infarction was 75.7 years). Baseline characteristics of the women are presented in Table 1. Women with higher total antioxidant capacity of diet were more likely to be nonsmokers, have ≥12 years of education, and to have hypercholesterolemia. Regarding dietary characteristics, women in the highest quintile of total antioxidant capacity of diet, as compared with the lowest quintile, had higher consumption of fruit and vegetables (3-fold), whole grains (15%), coffee (34%), and chocolate (38%), as well as 27% lower intake of saturated fatty acids and 19% lower intake of monounsaturated fatty acids. The major contributors to dietary total antioxidant capacity were fruit and vegetables (44%). The Pearson correlation coefficient between dietary total antioxidant capacity and fruit and vegetable consumption was 0.55. Other contributors were whole grains (18%), coffee (14%), and chocolate (4%).
      Table 1Age-standardized Background Characteristics of Women in the Swedish Mammography Cohort (n = 32,561)
      All variables except age are standardized to the age distribution of the cohort.
      CharacteristicsQuintiles of Total Antioxidant Capacity of Diet
      μmol Trolox equivalents per day, as assessed with oxygen radical absorbance capacity assay.
      Q1 (n = 6512)Q2 (n = 6512)Q3 (n = 6512)Q4 (n = 6512)Q5 (n = 6513)
      Median total antioxidant capacity of diet853710,77912,50214,49518,021
      Non-dietary factors
       Age, mean (SD)61.5 (9.5)61.0 (9.0)61.0 (8.9)60.8 (8.8)60.4 (8.7)
       > 12 years of education, %15.018.120.121.723.4
       Current smokers, %28.725.423.221.420.0
       Body mass index, mean (SD)25.1 (4.1)24.9 (3.8)24.9 (3.8)25.7 (3.7)25.6 (3.7)
       Total physical activity score,
      Expressed as multiples of the metabolic equivalent per day (MET, kcal kg−1×h−1) of sitting quietly for 1 hour.
      mean (SD)
      42.0 (5.0)42.4 (4.7)42.6 (4.7)42.6 (4.6)43.0 (4.7)
       Hypertension, %19.018.418.117.718.9
       Hypercholesterolemia, %6.16.47.47.38.2
       Family history of myocardial infarction, %12.613.112.620.926.5
       Aspirin use, %43.144.943.541.341.9
       Current and past users of hormone replacement therapy, %61.458.158.361.459.4
       Supplement users, %56.757.960.060.162.3
       Total energy, kcal/day (SD)1770 (585)1749 (510)1746 (499)1720 (490)1738 (516)
       Alcohol, gram/day (SD)5.3 (12.1)5.8 (10.2)5.9 (7.7)5.7 (7.8)5.4 (9.1)
      Foods, mean (SD)
       Fruits and vegetables, servings/day2.4 (1.4)3.4 (1.5)4.0 (1.7)4.9 (1.9)6.8 (3.0)
       Whole grains, servings/day3.3 (2.0)3.7 (2.0)3.8 (2.0)3.8 (2.0)3.8 (2.0)
       Coffee, servings/day2.6 (1.4)3.0 (1.6)3.2 (1.7)3.3 (1.8)3.5 (2.0)
       Tea, servings/week5.1 (8.1)5.0 (8.1)4.6 (7.3)4.5 (7.8)4.5 (8.1)
       Chocolate, servings/week0.8 (0.9)0.9 (1.1)1.0 (1.2)1.1 (1.5)1.1 (1.8)
      Nutrients, mean (SD)
       Saturated fatty acids, gram/day31.4 (6.6)29.2 (5.7)27.5 (5.4)25.8 (5.2)23.0 (5.3)
       Monounsaturated fatty acids, gram/day21.8 (3.5)20.0 (3.1)20.2 (2.9)19.3 (3.0)17.6 (3.1)
       Polyunsaturated fatty acids, gram/day7.9 (2.1)8.1 (1.8)8.2 (1.9)8.1 (1.8)8.0 (1.8)
      SD = standard deviation; ORAC = oxygen radical capacity absorbance.
      low asterisk All variables except age are standardized to the age distribution of the cohort.
      μmol Trolox equivalents per day, as assessed with oxygen radical absorbance capacity assay.
      Expressed as multiples of the metabolic equivalent per day (MET, kcal kg−1 × h−1) of sitting quietly for 1 hour.
      The association between total antioxidant capacity of diet and incident myocardial infarction is presented in Table 2. In the multivariable-adjusted model, women in the highest quintile of total antioxidant capacity of diet, compared with the lowest quintile, had a 20% (95% CI, 3%-33%, P for trend = .02) lower risk of myocardial infarction. In sensitivity analyses, we evaluated whether the apparent inverse association with total antioxidant capacity of diet can be explained by consumption of fruit and vegetables by adding this variable (continuous, servings/day) to the model, and results did not substantially change (HR for women in the top quintile was 0.81; 95% CI, 0.64-1.02). When adjusting for intakes of saturated fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids (all continuous, gram/day), we observed a HR of 0.71 (95% CI, 0.58-0.87). The association was similar for nonfatal myocardial infarction (HR 0.78; 95% CI, 0.65-0.95) and fatal myocardial infarction (HR 0.76; 95% CI, 0.48-1.20).
      Table 2Antioxidant-containing Foods
      Foods contributing ≥10% of total antioxidant capacity of diet.
      and Total Antioxidant Capacity of Diet
      μmol Trolox equivalents per day.
      in Relation to Risk of Myocardial Infarction among 32,561 Women
      CategoriesP for Trend
      Fruit and vegetables, servings/day≤234≥5
       No. of cases244224227419
       Person-years44,22557,06765,022155,121
       Age-adjusted HR1.000.90 (0.75-1.08)0.88 (0.73-1.05)0.75 (0.64-0.88)<.001
       Multivariable HR
      Adjusted for age, education, smoking, body mass index, physical activity, hypertension, hypercholesterolemia, family history of myocardial infarction, aspirin use, hormone replacement therapy use, dietary supplement use, and intakes of total energy and alcohol.
      1.000.95 (0.79-1.15)0.95 (0.79-1.14)0.86 (0.73-1.03).09
      Whole grains, servings/day≤2.32.3-3.43.4-4.7≥4.7
       No. of cases298260273283
       Person-years83,26977,99581,40178,769
       Age-adjusted HR1.000.90 (0.76-1.06)0.82 (0.69-0.97)0.81 (0.68-0.95).008
       Multivariable HR
      Adjusted for age, education, smoking, body mass index, physical activity, hypertension, hypercholesterolemia, family history of myocardial infarction, aspirin use, hormone replacement therapy use, dietary supplement use, and intakes of total energy and alcohol.
      1.000.95 (0.81-1.13)0.88 (0.74-1.04)0.89 (0.74-1.07).18
      Coffee, servings/day0<22-4>4
       No. of cases72408455179
       Person-years18,257113,592133,55356,032
       Age-adjusted HR1.000.87 (0.68-1.12)0.93 (0.72-1.19)1.13 (0.86-1.48).05
       Multivariable HR
      Adjusted for age, education, smoking, body mass index, physical activity, hypertension, hypercholesterolemia, family history of myocardial infarction, aspirin use, hormone replacement therapy use, dietary supplement use, and intakes of total energy and alcohol.
      1.000.87 (0.68-1.12)0.88 (0.69-1.13)0.96 (0.72-1.26).83
      Total antioxidant capacity of diet
      Foods contributing ≥10% of total antioxidant capacity of diet.
      , quintiles
      12345
       Median ORAC, μmol Trolox equivalents/day853710,77912,50214,49518,021
        No. of cases288231203201191
        Person-years63,25964,36664,38564,53464,890
        Age-adjusted HR1.000.85 (0.71-1.01)0.75 (0.63-0.90)0.76 (0.63-0.91)0.74 (0.62-0.89).001
        Multivariable HR
      Adjusted for age, education, smoking, body mass index, physical activity, hypertension, hypercholesterolemia, family history of myocardial infarction, aspirin use, hormone replacement therapy use, dietary supplement use, and intakes of total energy and alcohol.
      1.000.88 (0.74-1.05)0.80 (0.66-0.95)0.83 (0.69-0.99)0.80 (0.67-0.97).02
      Hazard ratios (HR) with 95% confidence intervals were obtained from Cox proportional hazards models.
      low asterisk Foods contributing ≥10% of total antioxidant capacity of diet.
      μmol Trolox equivalents per day.
      Adjusted for age, education, smoking, body mass index, physical activity, hypertension, hypercholesterolemia, family history of myocardial infarction, aspirin use, hormone replacement therapy use, dietary supplement use, and intakes of total energy and alcohol.
      Linear and quadratic relationships between total antioxidant capacity and myocardial infarction are presented in Figure. We found no significant evidence of departure from a constant change in the rate of myocardial infarction associated with every unit increase in total antioxidant capacity (P for nonlinearity = .25). The multivariable-adjusted HR of myocardial infarction for an increment of 4000 ORAC units/day (corresponding to approximately 1 SD in the study population) was 0.94 (95% CI, 0.88-1.00); the corrected HR (taking into account measurement error in the dietary total antioxidant capacity estimates) was 0.79. An increment of 4000 ORAC units is equivalent to approximately 1-2 apples or 2 peppers.
      Figure thumbnail gr1
      FigureMultivariable-adjusted hazard ratios for incident myocardial infarction according to dietary total antioxidant capacity, based on the oxygen radical absorbance capacity assay (μmol Trolox equivalents). Hazard ratios were adjusted for age, education, smoking, body mass index, physical activity, hypertension, hypercholesterolemia, family history of myocardial infarction, aspirin use, hormone replacement therapy use, dietary supplement use, and intakes of total energy and alcohol. Solid curve represents point estimates of the linear trend. Dotted line represents point estimates of the quadratic trend. Dashed lines represent 95% confidence intervals for the linear trend. The distribution of total antioxidant capacity is presented at the bottom of the figure as a histogram.
      To investigate whether the observed association could be due to reversed causality, as cardiovascular risk factors may be associated with changes in dietary habits, we excluded cases that occurred in the first 3 years of follow-up. This sensitivity analysis did not indicate influence of reversed causality on the observed estimate; women in the highest as compared with the lowest quintile had HR = 0.81 (95% CI, 0.68-0.98).
      In subgroup analysis, we examined whether the inverse association between dietary total antioxidant capacity and myocardial infarction varied by potential risk factors such as age, body mass index, smoking, saturated fatty acid intake, and multivitamin supplement use. The association was somewhat stronger among women aged >65 years (HR 0.74; 95% CI, 0.59-0.94), women with body mass index >25 (HR 0.76; 95% CI, 0.58-0.99), and current smokers (HR 0.66; 95% CI, 0.46-0.94), comparing the highest quintile with the lowest. However, there were no statistically significant interactions observed between dietary total antioxidant capacity and the above risk factors (all P for interaction >.16).

      Discussion

      In this large prospective population-based cohort study, we observed that higher total antioxidant capacity of diet was statistically significantly associated with lower risk of incident myocardial infarction in a dose-response manner. Consumption of antioxidant-containing foods such as fruits and vegetables and whole grains were nonsignificantly inversely associated with the risk of myocardial infarction.
      To our knowledge, this is the first study investigating dietary total antioxidant capacity in relation to myocardial infarction. The single total antioxidant capacity estimate is assumed to give a better measure of all antioxidants than sum of individual antioxidants, because it also reflects the synergism between the compounds. Moreover, we have previously observed in our study population that total antioxidant capacity of diet was inversely associated with incidence of total stroke.
      • Rautiainen S.
      • Larsson S.
      • Virtamo J.
      • Wolk A.
      Total antioxidant capacity of diet and risk of stroke: a population-based prospective cohort of women.
      Several foods that are major contributors to antioxidant intake in our study population have been linked to a decreased risk of coronary heart disease. In particular, high intake of fruit and vegetables, which contributed 44% of the dietary total antioxidant capacity in our study, have been inversely related to coronary heart disease in many studies.
      • Dauchet L.
      • Amouyel P.
      • Hercberg S.
      • Dallongeville J.
      Fruit and vegetable consumption and risk of coronary heart disease: a meta-analysis of cohort studies.
      Whole grains (18% of total antioxidant capacity) also are suggested to lower coronary heart disease risk.
      • Kelly S.A.
      • Summerbell C.D.
      • Brynes A.
      • et al.
      Wholegrain cereals for coronary heart disease.
      • Seal C.J.
      Whole grains and CVD risk.
      Coffee consumption (14% of total antioxidant capacity) has been inversely related to coronary heart disease in some but not in all studies.
      • Cornelis M.C.
      • El-Sohemy A.
      Coffee, caffeine, and coronary heart disease.
      Chocolate consumption (4% of total antioxidant capacity) has been shown to have favorable effects on cardiovascular risk biomarkers such as flow-mediated dilation and diastolic blood pressure, as shown by a meta-analysis of randomized controlled trials.
      • Hooper L.
      • Kroon P.A.
      • Rimm E.B.
      • et al.
      Flavonoids, flavonoid-rich foods, and cardiovascular risk: a meta-analysis of randomized controlled trials.
      The pathogenesis of atherosclerosis involves oxidation of low-density lipoproteins caused by reactive oxygen species.
      • Thomson M.J.
      • Puntmann V.
      • Kaski J.C.
      Atherosclerosis and oxidant stress: the end of the road for antioxidant vitamin treatment?.
      Antioxidants found in diet are thought to protect against coronary heart disease by scavenging reactive oxygen species.
      • Thomson M.J.
      • Puntmann V.
      • Kaski J.C.
      Atherosclerosis and oxidant stress: the end of the road for antioxidant vitamin treatment?.
      In addition to antioxidant effects, flavonoids also may inhibit the atherosclerotic process through other pathways. Flavonoids have been shown to improve endothelial function, to decrease blood pressure, and to have antiplatelet and anti-inflammatory effects.
      • Grassi D.
      • Desideri G.
      • Croce G.
      • et al.
      Flavonoids, vascular function and cardiovascular protection.
      Previous randomized controlled trials testing high doses of antioxidant supplements containing 1 to 3 compounds have failed to see any benefit on coronary heart disease.
      • Sesso H.D.
      • Buring J.E.
      • Christen W.G.
      • et al.
      Vitamins E and C in the prevention of cardiovascular disease in men: the Physicians' Health Study II randomized controlled trial.
      • Lee I.M.
      • Cook N.R.
      • Gaziano J.M.
      • et al.
      Vitamin E in the primary prevention of cardiovascular disease and cancer: the Women's Health Study: a randomized controlled trial.
      • Lonn E.
      • Bosch J.
      • Yusuf S.
      • et al.
      Effects of long-term vitamin E supplementation on cardiovascular events and cancer: a randomized controlled trial.
      • Cook N.R.
      • Albert C.M.
      • Gaziano J.M.
      • et al.
      A randomized factorial trial of vitamins C and E and beta carotene in the secondary prevention of cardiovascular events in women: results from the Women's Antioxidant Cardiovascular Study.
      • Yusuf S.
      • Dagenais G.
      • Pogue J.
      • et al.
      Vitamin E supplementation and cardiovascular events in high-risk patients The Heart Outcomes Prevention Evaluation Study Investigators.
      • Rapola J.M.
      • Virtamo J.
      • Ripatti S.
      • et al.
      Randomised trial of alpha-tocopherol and beta-carotene supplements on incidence of major coronary events in men with previous myocardial infarction.
      One randomized controlled trial that studied the effect of a low-dose mixture of 5 antioxidant supplements (including 120 mg ascorbic acid, 30 mg vitamin E, 6 mg beta carotene, 100 μg selenium, and 20 mg zinc) did not observe any association with ischemic cardiovascular diseases.
      • Hercberg S.
      • Galan P.
      • Preziosi P.
      • et al.
      The SU.VI.MAX Study: a randomized, placebo-controlled trial of the health effects of antioxidant vitamins and minerals.
      Notably, in a meta-analysis of high doses and very high doses of single supplements of vitamin A, β-carotene, or vitamin E tested in several randomized controlled trials, higher all-cause mortality was reported.
      • Bjelakovic G.
      • Nikolova D.
      • Gluud L.L.
      • et al.
      Antioxidant supplements for prevention of mortality in healthy participants and patients with various diseases.
      In contrast to supplements of single antioxidants, the dietary total antioxidant capacity reflects all present antioxidants, including thousands of compounds, all of them in doses present in our usual diet, and even takes into account their synergistic effects. Moreover, this study included women who were healthy at study entry, while the majority of the randomized controlled trials were performed among participants with established atherosclerosis who may not benefit from antioxidant supplementation.
      Our study has several strengths, including the prospective population-based design, detailed data on diet and potential risk factors, and the practically complete follow-up of the study cohort.
      There are, however, some potential limitations with our study. The total antioxidant capacity has not been measured for Swedish foods; we used American ORAC values. Antioxidant content may vary with geographic location and growing conditions. The observed association could be explained by consumption of fruit and vegetables; however, after this adjustment the association between total antioxidant capacity of diet and myocardial infarction remained statistically significant. We did not have ORAC values for dietary supplements, thus, our results can only be translated to food items and not to antioxidant supplements. Measurement error in self-reported dietary intake may lead to misclassification of exposure. Because of the prospective design, the misclassification is most likely to be nondifferential and therefore lead to attenuation of the observed association. Moreover, when we corrected for measurement error, an even stronger association was observed. We only estimated total antioxidant capacity of diet, which does not include antioxidant capacity of endogenous antioxidants. In our validation study, we observed a somewhat weak correlation between total antioxidant capacity from diet and plasma,
      • Rautiainen S.
      • Serafini M.
      • Morgenstern R.
      • et al.
      The validity and reproducibility of food-frequency questionnaire-based total antioxidant capacity estimates in Swedish women.
      which can be partly explained by the fact that plasma total antioxidant capacity values are influenced by many factors, such as endogenous antioxidants, homeostatic control mechanisms of plasma antioxidants, absorption, and the extent of the metabolism of dietary antioxidants.
      • Niki E.
      Assessment of antioxidant capacity in vitro and in vivo.
      The study was performed among women and cannot, therefore, be generalized to men. Furthermore, the observed inverse association could be due to a phenomenon in which women who eat an antioxidant-rich diet are more health conscious and have other healthy behaviors. When adjusting for several potential confounders such as smoking, physical activity, and educational level, the results remained statistically significant; however, we cannot rule out residual confounding by healthy lifestyle factors. Intervention studies would be needed to definitively determine the health effects of total antioxidant capacity.
      In conclusion, our data suggest that a diet high in total antioxidant capacity is associated with lower risk of incident myocardial infarction. The major contributors to dietary total antioxidant capacity were fruits, vegetables, coffee, and whole grains.

      References

        • Thomson M.J.
        • Puntmann V.
        • Kaski J.C.
        Atherosclerosis and oxidant stress: the end of the road for antioxidant vitamin treatment?.
        Cardiovasc Drugs Ther. 2007; 21: 195-210
        • Patel R.P.
        • Moellering D.
        • Murphy-Ullrich J.
        • et al.
        Cell signaling by reactive nitrogen and oxygen species in atherosclerosis.
        Free Radic Biol Med. 2000; 28: 1780-1794
        • Darley-Usmar V.
        • Halliwell B.
        Blood radicals: reactive nitrogen species, reactive oxygen species, transition metal ions, and the vascular system.
        Pharm Res. 1996; 13: 649-662
        • Ambrose J.A.
        • Barua R.S.
        The pathophysiology of cigarette smoking and cardiovascular disease: an update.
        J Am Coll Cardiol. 2004; 43: 1731-1737
        • Csiszar A.
        • Wang M.
        • Lakatta E.G.
        • Ungvari Z.
        Inflammation and endothelial dysfunction during aging: role of NF-kappaB.
        J Appl Physiol. 2008; 105: 1333-1341
        • de Ferranti S.
        • Mozaffarian D.
        The perfect storm: obesity, adipocyte dysfunction, and metabolic consequences.
        Clin Chem. 2008; 54: 945-955
        • Touyz R.M.
        Reactive oxygen species, vascular oxidative stress, and redox signaling in hypertension: what is the clinical significance?.
        Hypertension. 2004; 44: 248-252
        • Das S.K.
        • Vasudevan D.M.
        Alcohol-induced oxidative stress.
        Life Sci. 2007; 81: 177-187
        • Dauchet L.
        • Amouyel P.
        • Hercberg S.
        • Dallongeville J.
        Fruit and vegetable consumption and risk of coronary heart disease: a meta-analysis of cohort studies.
        J Nutr. 2006; 136: 2588-2593
        • Sesso H.D.
        • Buring J.E.
        • Christen W.G.
        • et al.
        Vitamins E and C in the prevention of cardiovascular disease in men: the Physicians' Health Study II randomized controlled trial.
        JAMA. 2008; 300: 2123-2133
        • Lee I.M.
        • Cook N.R.
        • Gaziano J.M.
        • et al.
        Vitamin E in the primary prevention of cardiovascular disease and cancer: the Women's Health Study: a randomized controlled trial.
        JAMA. 2005; 294: 56-65
        • Lonn E.
        • Bosch J.
        • Yusuf S.
        • et al.
        Effects of long-term vitamin E supplementation on cardiovascular events and cancer: a randomized controlled trial.
        JAMA. 2005; 293: 1338-1347
        • Cook N.R.
        • Albert C.M.
        • Gaziano J.M.
        • et al.
        A randomized factorial trial of vitamins C and E and beta carotene in the secondary prevention of cardiovascular events in women: results from the Women's Antioxidant Cardiovascular Study.
        Arch Intern Med. 2007; 167: 1610-1618
        • Yusuf S.
        • Dagenais G.
        • Pogue J.
        • et al.
        Vitamin E supplementation and cardiovascular events in high-risk patients.
        N Engl J Med. 2000; 342: 154-160
        • Rapola J.M.
        • Virtamo J.
        • Ripatti S.
        • et al.
        Randomised trial of alpha-tocopherol and beta-carotene supplements on incidence of major coronary events in men with previous myocardial infarction.
        Lancet. 1997; 349: 1715-1720
      1. Socialstyrelsen: Svensk version av ICD-10 (Swedish board of health and welfare: Swedish version of ICD-10).
        (Accessed September 16, 2010)
        • Rautiainen S.
        • Serafini M.
        • Morgenstern R.
        • et al.
        The validity and reproducibility of food-frequency questionnaire-based total antioxidant capacity estimates in Swedish women.
        Am J Clin Nutr. 2008; 87: 1247-1253
        • Wu X.
        • Beecher G.R.
        • Holden J.M.
        • et al.
        Lipophilic and hydrophilic antioxidant capacities of common foods in the United States.
        J Agric Food Chem. 2004; 52: 4026-4037
        • Prior R.L.
        • Hoang H.
        • Gu L.
        • et al.
        Assays for hydrophilic and lipophilic antioxidant capacity (oxygen radical absorbance capacity (ORAC(FL))) of plasma and other biological and food samples.
        J Agric Food Chem. 2003; 51: 3273-3279
        • Sanchez-Moreno C.
        • Cao G.
        • Ou B.
        • Prior R.L.
        Anthocyanin and proanthocyanidin content in selected white and red wines.
        J Agric Food Chem. 2003; 51: 4889-4896
        • Natella F.
        • Nardini M.
        • Giannetti I.
        • et al.
        Coffee drinking influences plasma antioxidant capacity in humans.
        J Agric Food Chem. 2002; 50: 6211-6216
        • Willett W.C.
        Nutritional Epidemiology.
        2nd ed. Oxford University Press, New York1998
        • Kuskowska-Wolk A.
        • Bergstrom R.
        • Bostrom G.
        Relationship between questionnaire data and medical records of height, weight and body mass index.
        Int J Obes Relat Metab Disord. 1992; 16: 1-9
        • Orsini N.
        • Bellocco R.
        • Bottai M.
        • et al.
        Validity of self-reported total physical activity questionnaire among older women.
        Eur J Epidemiol. 2008; 23: 661-667
        • Hammar N.
        • Alfredsson L.
        • Rosen M.
        • et al.
        A national record linkage to study acute myocardial infarction incidence and case fatality in Sweden.
        Int J Epidemiol. 2001; 30: S30-S34
        • Cox D.
        Regression models and life-tables.
        J R Stat Soc Series B Stat Methodol. 1972; 34: 187-220
        • Spiegelman D.
        • McDermott A.
        • Rosner B.
        Regression calibration method for correcting measurement-error bias in nutritional epidemiology.
        Am J Clin Nutr. 1997; 65: 1179S-1186S
        • Rautiainen S.
        • Larsson S.
        • Virtamo J.
        • Wolk A.
        Total antioxidant capacity of diet and risk of stroke: a population-based prospective cohort of women.
        Stroke. 2012; 43: 335-340
        • Kelly S.A.
        • Summerbell C.D.
        • Brynes A.
        • et al.
        Wholegrain cereals for coronary heart disease.
        Cochrane Database Syst Rev. 2007; (CD005051)
        • Seal C.J.
        Whole grains and CVD risk.
        Proc Nutr Soc. 2006; 65: 24-34
        • Cornelis M.C.
        • El-Sohemy A.
        Coffee, caffeine, and coronary heart disease.
        Curr Opin Lipidol. 2007; 18: 13-19
        • Hooper L.
        • Kroon P.A.
        • Rimm E.B.
        • et al.
        Flavonoids, flavonoid-rich foods, and cardiovascular risk: a meta-analysis of randomized controlled trials.
        Am J Clin Nutr. 2008; 88: 38-50
        • Grassi D.
        • Desideri G.
        • Croce G.
        • et al.
        Flavonoids, vascular function and cardiovascular protection.
        Curr Pharm Des. 2009; 15: 1072-1084
        • Hercberg S.
        • Galan P.
        • Preziosi P.
        • et al.
        The SU.VI.MAX Study: a randomized, placebo-controlled trial of the health effects of antioxidant vitamins and minerals.
        Arch Intern Med. 2004; 164: 2335-2342
        • Bjelakovic G.
        • Nikolova D.
        • Gluud L.L.
        • et al.
        Antioxidant supplements for prevention of mortality in healthy participants and patients with various diseases.
        Cochrane Database Syst Rev. 2008; (CD007176)
        • Niki E.
        Assessment of antioxidant capacity in vitro and in vivo.
        Free Radic Biol Med. 2010; 49: 503-515