Advertisement

Erythrocyte membrane n-3 fatty acid levels and carotid atherosclerosis in Chinese men and women

  • Xiao-wei Dai
    Affiliations
    Guangdong Provincial Key Laboratory of Food, Nutrition and Health, School of Public Health, Sun Yat-sen University, Guangzhou 510080, People's Republic of China
    Search for articles by this author
  • Bo Zhang
    Affiliations
    Guangdong Provincial Key Laboratory of Food, Nutrition and Health, School of Public Health, Sun Yat-sen University, Guangzhou 510080, People's Republic of China
    Search for articles by this author
  • Ping Wang
    Affiliations
    Guangdong Provincial Key Laboratory of Food, Nutrition and Health, School of Public Health, Sun Yat-sen University, Guangzhou 510080, People's Republic of China
    Search for articles by this author
  • Chao-gang Chen
    Affiliations
    The Second Affiliated Hospital of Sun Yat-sen University, Guangzhou 510080, People's Republic of China
    Search for articles by this author
  • Yu-ming Chen
    Correspondence
    Corresponding author. Department of Medical Statistics & Epidemiology, School of Public Health, Sun Yat-sen University, Guangzhou 510080, People's Republic of China. Tel.: +86 20 87330605; fax: +86 20 87330446.
    Affiliations
    Guangdong Provincial Key Laboratory of Food, Nutrition and Health, School of Public Health, Sun Yat-sen University, Guangzhou 510080, People's Republic of China
    Search for articles by this author
  • Yi-xiang Su
    Correspondence
    Corresponding author. Department of Nutrition, School of Public Health, Sun Yat-sen University, Guangzhou 510080, People's Republic of China. Tel./fax: +86 20 87333166.
    Affiliations
    Guangdong Provincial Key Laboratory of Food, Nutrition and Health, School of Public Health, Sun Yat-sen University, Guangzhou 510080, People's Republic of China
    Search for articles by this author

      Highlights

      • The beneficial effects of n-3 fatty acid on atherosclerosis remains uncertain.
      • We tested the association of n-3 fatty acids in erythrocytes with atherosclerosis.
      • DHA and ALA, but not EPA inversely associated with carotid atherosclerosis.

      Abstract

      Objective

      Prospective studies have supported the beneficial effects of n-3 fatty acid consumption on cardiac deaths, but limited data focused on atherosclerosis. We investigated the associations between n-3 fatty acids in erythrocytes and atherosclerosis in middle-aged and older Chinese.

      Methods

      847 subjects (285 men and 562 women), aged 40–65 years, from Guangzhou, China were included in this community-based cross-sectional study between December 2005 and January 2008. The levels of α-linolenic acid (ALA), eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in erythrocytes were measured by gas chromatography. Carotid ultrasound examination was conducted to obtain intima–media thickness of the common carotid artery and the carotid bifurcation. Dietary data and other covariates were collected using interviewer-administered questionnaires.

      Results

      After adjustment for age, sex, and other confounders, negative dose–response associations between the contents of individual n-3 polyunsaturated fatty acids in the erythrocyte membrane and the prevalence of carotid artery wall thickening and plaque were observed. A comparison in the highest and lowest tertiles gave odds ratios (95% confidence interval) for thickening in the walls of the common carotid artery of 0.58 (0. 34–0.97; P-trend = 0. 037) for DHA, and 0.39 (0.23–0.67; P-trend < 0.001) for ALA. However, EPA was not significantly associated with carotid atherosclerosis. Similar results were found for thickening at the carotid bifurcation and the occurrence of carotid artery plaque.

      Conclusions

      Higher levels of DHA and ALA in the erythrocyte membrane were significantly associated with a lower burden of subclinical atherosclerosis.

      Keywords

      1. Introduction

      Numerous large prospective cohort studies [
      • Hu F.B.
      • Bronner L.
      • Willett W.C.
      • et al.
      Fish and omega-3 fatty acid intake and risk of coronary heart disease in women.
      ,
      • Yuan J.M.
      • Ross R.K.
      • Gao Y.T.
      • et al.
      Fish and shellfish consumption in relation to death from myocardial infarction among men in Shanghai, China.
      ,
      • de Goede J.
      • Geleijnse J.M.
      • Boer J.M.
      • et al.
      Marine (n-3) fatty acids, fish consumption, and the 10-year risk of fatal and nonfatal coronary heart disease in a large population of Dutch adults with low fish intake.
      ], randomized controlled trials [
      • Marchioli R.
      • Barzi F.
      • Bomba E.
      • et al.
      Early protection against sudden death by n-3 polyunsaturated fatty acids after myocardial infarction: time-course analysis of the results of the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardico (GISSI)-Prevenzione.
      ,
      • Tavazzi L.
      • Maggioni A.P.
      • Marchioli R.
      • et al.
      Effect of n-3 polyunsaturated fatty acids in patients with chronic heart failure (the GISSI-HF trial): a randomised, double-blind, placebo-controlled trial.
      ] and meta-analyses/reviews [
      • He K.
      • Song Y.
      • Daviglus M.L.
      • et al.
      Accumulated evidence on fish consumption and coronary heart disease mortality: a meta-analysis of cohort studies.
      ,
      • Leon H.
      • Shibata M.C.
      • Sivakumaran S.
      • et al.
      Effect of fish oil on arrhythmias and mortality: systematic review.
      ,
      • Mozaffarian D.
      Fish and n-3 fatty acids for the prevention of fatal coronary heart disease and sudden cardiac death.
      ] have shown concordant evidence that the consumption of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) from fish or fish oil is associated with a decreased risk of cardiac death. Results from intervention studies suggested that supplementation with marine-derived n-3 fatty acids might protect against the progression of atherosclerosis [
      • Baldassarre D.
      • Amato M.
      • Eligini S.
      • et al.
      Effect of n-3 fatty acids on carotid atherosclerosis and haemostasis in patients with combined hyperlipoproteinemia: a double-blind pilot study in primary prevention.
      ,
      • Mita T.
      • Watada H.
      • Ogihara T.
      • et al.
      Eicosapentaenoic acid reduces the progression of carotid intima-media thickness in patients with type 2 diabetes.
      ,
      • von Schacky C.
      • Angerer P.
      • Kothny W.
      • et al.
      The effect of dietary omega-3 fatty acids on coronary atherosclerosis. A randomized, double-blind, placebo-controlled trial.
      ]. In observational studies, the inverse association between EPA/DHA and atherosclerosis was more frequently reported in populations with greater seafood intakes, such as in the Japanese (73–82 g/day) [
      • Hino A.
      • Adachi H.
      • Toyomasu K.
      • et al.
      Very long chain N-3 fatty acids intake and carotid atherosclerosis: an epidemiological study evaluated by ultrasonography.
      ,
      • Sekikawa A.
      • Curb J.D.
      • Ueshima H.
      • et al.
      Marine-derived n-3 fatty acids and atherosclerosis in Japanese, Japanese-American, and white men: a cross-sectional study.
      ,
      • Iso H.
      • Kobayashi M.
      • Ishihara J.
      • et al.
      Intake of fish and n3 fatty acids and risk of coronary heart disease among Japanese: the Japan Public Health Center-Based (JPHC) Study Cohort I.
      ] and in the Spanish (76 g/day) [
      • Sala-Vila A.
      • Cofan M.
      • Perez-Heras A.
      • et al.
      Fatty acids in serum phospholipids and carotid intima-media thickness in Spanish subjects with primary dyslipidemia.
      ], but few studies among those with low intakes [
      • He K.
      • Liu K.
      • Daviglus M.L.
      • et al.
      Intakes of long-chain n-3 polyunsaturated fatty acids and fish in relation to measurements of subclinical atherosclerosis.
      ]. The levels of fish consumption in Mainland Chinese are far lower than those of the Japanese and Europeans, and the median consumption of fish and EPA + DHA was reported to be approximately 32 g/day and 70 mg/day in men and women in Shanghai, respectively [
      • Villegas R.
      • Xiang Y.B.
      • Elasy T.
      • et al.
      Fish, shellfish, and long-chain n-3 fatty acid consumption and risk of incident type 2 diabetes in middle-aged Chinese men and women.
      ]. It is uncertain whether the favorable associations of EPA and DHA, or total long-chain n-3 fatty acids with atherosclerosis persist in populations with such a low fish intake.
      The consumption of marine-derived n-3 fatty acids is limited in Mainland Chinese. The plant-derived α-linolenic acid (ALA) rich in oils of rapeseed, walnuts, soybeans and other plant oils, was sufficient in the habitual Chinese diet. Although most studies have focused on the effects of EPA and DHA, but not those of ALA, emerging evidence has suggested that ALA might also be beneficial for the prevention of atherosclerosis [
      • Mozaffarian D.
      • Ascherio A.
      • Hu F.B.
      • et al.
      Interplay between different polyunsaturated fatty acids and risk of coronary heart disease in men.
      ,
      • Campos H.
      • Baylin A.
      • Willett W.C.
      Alpha-linolenic acid and risk of nonfatal acute myocardial infarction.
      ,
      • Djousse L.
      • Folsom A.R.
      • Province M.A.
      • et al.
      Dietary linolenic acid and carotid atherosclerosis: the National Heart, Lung, and Blood Institute Family Heart Study.
      ,
      • Djousse L.
      • Arnett D.K.
      • Carr J.J.
      • et al.
      Dietary linolenic acid is inversely associated with calcified atherosclerotic plaque in the coronary arteries: the National Heart, Lung, and Blood Institute Family Heart Study.
      ]. The majority of these studies evaluated the effects of dietary ALA estimated from questionnaires. However, because of the high percentage of β-oxidation for energy supply (≥15–35%) of ALA after digestion [
      • Burdge G.C.
      • Finnegan Y.E.
      • Minihane A.M.
      • et al.
      Effect of altered dietary n-3 fatty acid intake upon plasma lipid fatty acid composition, conversion of [13C]alpha-linolenic acid to longer-chain fatty acids and partitioning towards beta-oxidation in older men.
      ], low conversion rate of ALA to EPA/DHA (0.2%∼8% to EPA, and 0%∼4% to DHA) [
      • Burdge G.
      Alpha-linolenic acid metabolism in men and women: nutritional and biological implications.
      ], assessment of the level of ALA in blood or tissue might provide more accurate evaluation of ALA-atherosclerosis association. The intima–media thickness (IMT) of the carotid artery is a well-established surrogate measure for the detection of early atherosclerosis [
      • Lorenz M.W.
      • Markus H.S.
      • Bots M.L.
      • et al.
      Prediction of clinical cardiovascular events with carotid intima-media thickness: a systematic review and meta-analysis.
      ]. The present study aimed to investigate whether total and various types of n-3 fatty acids (ALA, EPA and DHA) in the erythrocyte membrane (biomarkers as the internal dose of dietary intakes) were inversely associated with the prevalence of carotid atherosclerosis among middle-aged and older Chinese.

      2. Methods

      2.1 Study participants

      A total of 1181 subjects, aged 40–65 years, were enrolled in this community-based cross-sectional study in Guangzhou, Guangdong Province, China. Participants were required to have lived in urban Guangzhou for more than 5 years, and were recruited between December 2005 and January 2008 through community advertisements and subject referral. Of the 1181 subjects, 334 were excluded according to the following criteria: 1) had a self-reported history of confirmed diseases, such as diabetes, cardiovascular disease (CVD), dyslipidemia, cancers, or took any medicines known to affect plasma lipids (n = 116); 2) were missing data on the levels of fatty acids (n = 188); and 3) were missing data on covariates (n = 30); thus, 847 subjects (285 men, 562 women) remained in the final analyses. The study protocol was approved by the Medical Ethics Committee of Sun Yat-Sen University and written informed consent was obtained from each participant at the time of enrollment.

      2.2 Data collection

      A structured questionnaire was used to collect data on dietary nutrient intakes, demographic variables, years of education, and general risk factors for CVD, such as smoking habits and alcohol drinking, physical activity, a history of disease and menopausal status (yes or no, for women only). Questionnaires were administered by trained medical interviewers during face-to-face interviews and conducted prior to the laboratory assays and carotid ultrasound examination.
      Dietary data were assessed by an interviewer-administered 119-item food-frequency questionnaire (FFQ) designed to cover the whole diet and the use of supplements (such as fish oil capsules) during the past year. None of the participants reported consumption of fish-oil supplements. The relative validity of the FFQ (intake assessed by the FFQ compared with that assessed by four 3-day 24-h dietary records) had been confirmed in a previous study [
      • Zhang B.
      • Wang P.
      • Chen C.G.
      • et al.
      Validation of an FFQ to estimate the intake of fatty acids using erythrocyte membrane fatty acids and multiple 3d dietary records.
      ]. Participants were asked to report the usual frequency of consumption of each food item (“never”, “per year”, “per month”, “per week”, or “per day”) and the mean portion size, which was estimated using color photographs. Average daily intakes of nutrients were calculated based on the Chinese Food Composition Table 2002 [
      ].
      Body weight and height were measured while subjects were wearing light clothing and no shoes. Body mass index was calculated as the body weight in kilograms divided by the body height in meters squared. The waist circumference was measured at the mid-line between the costal margin and iliac crest, while the hip circumference was measured at the point of maximum girth around the buttocks. The waist-to-hip ratio was calculated as the ratio of the two circumferences. Subjects were categorized as never smokers and as current or ex-smokers when they reported having smoked ≥1 cigarettes/d and having smoked ≥100 cigarettes or quit smoking ≥3 months earlier, respectively. Participants reported their usual alcohol consumption per week of wine, beer, and distilled spirit (in grams), and the amount of alcohol consumption was calculated. We categorized daily ethanol intake in grams into 3 categories: none, 0.1–12 (light drinkers), >12 g or more (modest drinkers or more) Daily physical activity was estimated using a 24-h physical activity questionnaire that comprised 19 items [
      • Wang P.
      • Chen Y.M.
      • He L.P.
      • et al.
      Association of natural intake of dietary plant sterols with carotid intima-media thickness and blood lipids in Chinese adults: a cross-section study.
      ], and the metabolic equivalent intensity (excluding sleeping and sitting) was calculated.

      2.3 Laboratory assay

      Venous blood samples were drawn after of 12-h fasting for lipid analysis and red blood cell membrane extraction, using vacuum tubes containing ethylenediaminetetraacetic acid. After centrifugation at 1500× g for 15 min at 4 °C, plasma and red blood cells (RBCs) were separated within 2 h. All samples were stored at −80 °C until the tests were performed. Plasma levels of total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDLc) and low-density lipoprotein cholesterol (LDLc) were measured by colorimetric methods using commercial kits (Biosino Biotechnology Company Ltd, Beijing, China) in a Hitachi 7600-010 automated analyzer. The coefficient of variation for lipid measurements was 2.3% (at 174.0 mg/dL total cholesterol), 5.8% (at 156.7 mg/dL triglycerides), 4.3% (at 51.2 mg/dL HDLc), and 3.1% (at 131.6 mg/dL LDLc). Fasting glucose concentration was measured by using the glucose oxidase method (Roche, Basel, Swiss), and the coefficient of variation for fasting glucose measurement was 4.94% (at 82.1 mg/dL).
      To determine the percentages of fatty acids in the erythrocyte membrane, lipids were extracted from the stored RBCs. Fatty acid methyl esters from the erythrocyte membrane were obtained as described previously [
      • Zhang B.
      • Wang P.
      • Chen C.G.
      • et al.
      Validation of an FFQ to estimate the intake of fatty acids using erythrocyte membrane fatty acids and multiple 3d dietary records.
      ]. Briefly, RBC samples were supplemented with hypotonic Tris–HCl buffer (10 mmol/L, pH = 7.4) at 4 °C for 2 h. The RBC membrane fragments were obtained by ultracentrifugation (12,000 rpm for 30 min at 4 °C) and were extracted with chloroform/methanol (2:1, v/v) supplemented with 0.005% butylated hydroxytoluene. The extract was dried under nitrogen and was resuspended in 14% boron-trifluoride ether/methanol (1:3, v/v) solution at 100 °C for 5 min. Samples were extracted after cooling with hexane, dried under nitrogen and redissolved in 50 μL hexane for gas chromatographic analysis. Fatty acid methyl esters were separated using an Agilent 6890 gas chromatograph (Agilent, Palo Alto, CA, USA) equipped with a 50-m WCOT-fused silica CP-SIL 88 fame column (Varian, Bergen op Zoom, Netherlands).
      Individual fatty acids were identified by the comparison of retention time with known standards (Sigma Aldrich Inc., St. Louis, MO, USA) and expressed as a percentage of the total fatty acids quantified from peak areas. The parallel duplicate samples (n = 35) were analyzed throughout the study. Coefficients of variation for EPA, DHA and ALA were 3.38%, 0.39% and 9.51%, respectively.

      2.4 Measurement of the carotid IMT

      The carotid IMT was measured bilaterally between the lumen-intima and media-adventitia interfaces of the innermost walls of the common carotid artery (CCA; the 1-cm segment proximal to the bifurcation) and the carotid bifurcation (BIF; the 1-cm segment proximal to the flow divider) using a high-resolution, 10.0-MHz linear-array transducer (Technos MPX DU8, Esaote) in the supine position after the subjects had rested for 10 min. All B-mode images were recorded at the diastolic phase to minimize variability during the cardiac cycle. A single certified sonographer from the Ultrasound Department of the First Affiliated Hospital of Sun Yat-Sen University performed the ultrasound examination (including the image analysis) and was blinded to the health status of each subject. Plaques (defined as focal thickening of >1.5 mm) [
      • Touboul P.J.
      • Hennerici M.G.
      • Meairs S.
      • et al.
      Mannheim carotid intima-media thickness consensus (2004–2006). An update on behalf of the Advisory Board of the 3rd and 4th Watching the Risk Symposium, 13th and 15th European Stroke Conferences, Mannheim, Germany, 2004, and Brussels, Belgium, 2006.
      ] were avoided in the measurement of IMT. The wall thickness was measured manually using the computerized assistance of an electronic caliper. The IMT of the BIF and the CCA were calculated in the analyses as the average of bilateral values. Intra-sonographer reproducibility of the IMT measurement was assessed by using a new image in 81 randomly selected subjects on the same day; the correlation coefficient was 0.78 and 0.69 (P < 0.001) for the mean carotid far wall IMT at the common carotid arteries and carotid bifurcation, respectively.

      2.5 Statistical analysis

      All statistics were analyzed using SPSS 16.0 for Windows (SPSS Inc., Chicago, IL, USA) and P < 0.05 (two sided) was considered to be statistically significant. Data were presented as means (standard deviations [SD]) for continuous variables or proportions (%) for categorical variables. Participants were categorized into tertiles by each fatty acid of interest (total n-3 fatty acids, ALA, EPA, DHA, and EPA + DHA) in the erythrocyte membrane and in their diet. Dietary intakes of n-3 fatty acids were the energy adjusted values using the residual method [
      • Willett W.
      • Stampfer M.J.
      Total energy intake: implications for epidemiologic analyses.
      ].
      Multivariable logistical regression models were applied to estimate the odds ratios [95% confidence interval (CI)] for the prevalence of carotid intima–media thickening (defined as IMT ≥1 mm) [
      • Greenland P.
      • Alpert J.S.
      • Beller G.A.
      • et al.
      ACCF/AHA guideline for assessment of cardiovascular risk in asymptomatic adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.
      ] and carotid artery plaque by comparing the tertiles of n-3 fatty acids in diet and red blood cell. Analysis of covariance (ANCOVA) was used to compare the mean carotid IMT (at the CCA and BIF) among the tertiles of n-3 fatty acids. Model 1 adjusted for age and sex. Model 2 included additional adjustment for waist circumference, smoking (never, past, and current smoker), alcohol drinking (non-drinkers, light drinkers, modest drinkers and heavy drinkers), physical activity (in metabolic equivalent), education (secondary school or below, high school, and college or above), systolic blood pressure (SBP), diastolic blood pressure (DBP), fasting plasma lipids including TC, TG, HDLc and LDLc and fasting plasma glucose. To determine the independent association of each fatty acid (ALA, EPA, DHA), we further adjusted for the other two of the three fatty acids. In the analyses of dietary intakes of n-3 fatty acids, we further adjusted for total energy, fiber and saturated fat. P values for linear trend across increasing tertiles were determined by assuming the median values of tertiles as continuous variables. All independent variables were entered into the models.
      We performed stratification analyses to investigate whether the above-mentioned associations differed across various subgroups classified by gender (male versus female) and body mass index (<25 kg/m2 versus ≥25 kg/m2). For the multivariate analysis with women only, we further adjusted for menopausal status (yes or no). Interactions were tested by adding cross-product terms between n-3 fatty acids and gender or body mass index into the full multivariate model.

      3. Results

      3.1 Characteristics of study participants

      Table 1 summarizes the characteristics of the study participants by gender. The 285 men and 562 women had a mean age of 54.0 ± 6.4 years and 52.3 ± 5.8 years, respectively. The mean daily intakes of ALA, EPA, and DHA were 0.95 g, 11.21 mg, and 44.35 mg for men, and 0.90 g, 10.16 mg, and 43.75 mg for women, respectively; the proportions of corresponding fatty acids in the erythrocyte membrane were 0.10%, 0.50%, and 7.77% for men, and 0.10%, 0.49%, and 7.74% for women, respectively. The percentages of subjects with carotid artery wall thickening at the CCA and the BIF were 15.4% and 58.6% for men and 13.3% and 46.8% for women, respectively.
      Table 1Characteristics of study participants.
      MaleFemale
      Mean ± SD or n (%)nMean ± SD or n (%)n
      Age, year54.0 ± 6.428552.3 ± 5.8562
      BMI, kg/m223.5 ± 2.828523.4 ± 3.3562
      Waist circumference, cm85.1 ± 8.428580.3 ± 8.9562
      WHR0.89 ± 0.052850.85 ± 0.06562
      Smoking, n (%)
      Non-smoker102 (35.8)555 (98.8)
      Ex-smoker35 (12.3)3 (0.5)
      Current-smoker148 (51.9)4 (0.7)
      Alcohol, n (%)
      Non-drinkers140 (49.1)430 (76.5)
      Light drinkers107 (37.5)126 (22.4)
      Modest drinkers or more38 (13.4)6 (1.1)
      Education, n (%)
      Secondary or below124 (43.3)219 (38.9)
      High school121 (42.6)282 (50.3)
      College or above40 (14.1)61 (10.8)
      Physical activity
      Physical activity included occupational, leisure-time, and household-chores, presented as metabolic equivalent (MET) hours per day (excluding sleeping and sitting time).
      , MET
      26.1 ± 6.728525.8 ± 5.3562
      Systolic blood pressure, mm Hg124.2 ± 18.3285123.4 ± 18.7562
      Diastolic blood pressure, mm Hg82.2 ± 10.928580.5 ± 10.4562
      Total cholesterol, mg/dL209.7 ± 36.3285220.1 ± 37.8562
      Triglycerides, mg/dL203.2 ± 165.3285168.7 ± 107.2562
      HDL cholesterol, mg/dL57.1 ± 15.628565.7 ± 13.6562
      LDL cholesterol, mg/dL143.8 ± 39.7285146.7 ± 39.3562
      Fasting glucose, mg/dL90.2 ± 18.928588.0 ± 23.8562
      n-3 fatty acid daily intakes
      Total n-3 PUFA, g/d1.01 ± 0.882850.96 ± 0.80562
      ALA, g/d0.95 ± 0.872850.90 ± 0.79562
      EPA, mg/d11.21 ± 13.4428510.16 ± 10.70562
      DHA, mg/d44.35 ± 33.7628543.75 ± 31.83562
      EPA + DHA, mg/d55.60 ± 45.0528553.90 ± 39.99562
      Fatty acid composition in erythrocyte membrane, % of total fatty acids
      Total n-3 PUFA, %10.04 ± 2.9428510.03 ± 3.10562
      ALA, %0.10 ± 0.052850.10 ± 0.06562
      EPA, %0.50 ± 0.242850.49 ± 0.26562
      DHA, %7.77 ± 2.892857.74 ± 2.94562
      Carotid intima-media thickness
      Carotid bifurcation, mm1.11 ± 0.362851.00 ± 0.28562
      Common carotid artery, mm0.73 ± 0.382850.67 ± 0.27562
      Plaque, n (%)44 (15.4)52 (9.3)
      BMI, body mass index; WHR, waist to hip ratio; n-3 PUFA, n-3 polyunsaturated fatty acid; ALA, α-linolenic acid (18:3 n-3); EPA, eicosapentaenoic acid (20:5 n-3); DHA, docosahexaenoic acid (22:6 n-3).
      a Physical activity included occupational, leisure-time, and household-chores, presented as metabolic equivalent (MET) hours per day (excluding sleeping and sitting time).

      3.2 Associations of the levels of n-3 fatty acids with carotid IMT

      Overall, erythrocyte ALA and DHA levels showed significant inverse associations with the prevalence of carotid atherosclerosis as determined by carotid artery wall thickening and carotid plaques after adjustment for age, sex, waist circumference, smoking, alcohol intake, physical activity, education, SBP, DBP, TC, TG, HDLc, LDLc and fasting glucose (Tables 2 and 4). The odds ratios (95% CI) of having the IMT thickening at the CCA in the highest (versus the lowest) tertile were 0.39 (0.23–0.67) (P trend < 0.001) for ALA, and 0.58 (0.34–0.97) (P trend = 0.037) for DHA. The corresponding odds ratios (95% CI) of the IMT thickening at the BIF for ALA, DHA, and EPA + DHA, were 0.58 (0.40–0.83), 0.60 (0.41–0.87), and 0.65 (0.45–0.94), respectively (all P-trend < 0.05). Erythrocyte ALA, DHA, and EPA + DHA levels were also associated with a lower prevalence of carotid plaques. The odds ratios (95% CI) of carotid plaques comparing extreme tertiles were 0.55 (0.31–0.97) (ALA), 0.52 (0.29–0.94) (DHA), 0.48 (0.27–0.87) (DHA + EPA), respectively. Similar associations were obtained when carotid IMT was presented as a continuous variable (Supplemental Table 1). However, these associations were slightly attenuated after further adjusted for other fatty acids of interest, of which remained significant were ALA and carotid IMT at both CCA and BIF, DHA and carotid IMT at BIF. Unexpectedly, EPA was not significantly associated with carotid atherosclerosis. We did not observe any significant associations between dietary intakes of n-3 fatty acids estimated by FFQ and carotid IMT value or thickening at the CCA and BIF in the multivariate analyses (all P-trend > 0.05) (Supplemental Table 2 and Supplemental Table 3).
      Table 2Odds ratios and 95% CI for carotid intima-media thickening by tertiles of erythrocyte n-3 PUFA concentrations.
      Odds ratios (95% CI) for CCA thickeningOdds ratios (95% CI) for BIF thickening
      Tertile 1Tertile 2Tertile 3P-trendTertile 1Tertile 2Tertile 3P-trend
      n-3 PUFA
      Cases/n48/28235/28336/282157/282143/283130/282
      Model 11.000.69 (0.43–1.10)0.71 (0.45–1.14)0.1451.000.87 (0.62–1.23)0.68 (0.48–0.96)0.028
      Model 21.000.74 (0.46–1.21)0.81 (0.50–1.32)0.3781.000.86 (0.60–1.23)0.71 (0.49–1.02)0.064
      ALA
      Cases/n58/28237/28024/285163/282141/280126/285
      Model 11.000.60 (0.38–0.94)0.36 (0.21–0.59)<0.0011.000.79 (0.56–1.12)0.56 (0.39–0.79)0.001
      Model 21.000.62 (0.39–0.99)0.39 (0.23–0.67)<0.0011.000.80 (0.55–1.14)0.58 (0.40–0.83)0.004
      Model 31.000.64 (0.40–1.03)0.43 (0.25–0.74)0.0021.000.83 (0.57–1.19)0.62 (0.43–0.91)0.014
      EPA
      Cases/n30/28246/28343/282142/282139/283149/282
      Model 11.001.61 (0.98–2.64)1.51 (0.92–2.49)0.1161.000.90 (0.64–1.27)1.16 (0.82–1.64)0.395
      Model 21.001.52 (0.92–2.52)1.44 (0.85–2.43)0.1791.000.88 (0.61–1.26)1.13 (0.78–1.63)0.541
      Model 31.001.61 (0.97–2.70)1.47 (0.86–2.50)0.1691.000.92 (0.64–1.33)1.14 (0.78–1.65)0.506
      DHA
      Cases/n51/28239/28329/282162/282140/283128/282
      Model 11.000.73 (0.46–1.15)0.52 (0.32–0.85)0.0081.000.76 (0.54–1.07)0.60 (0.42–0.84)0.003
      Model 21.000.79 (0.49–1.28)0.58 (0.34–0.97)0.0371.000.73 (0.51–1.06)0.60 (0.41–0.87)0.007
      Model 31.000.77 (0.47–1.25)0.64 (0.38–1.10)0.1001.000.73 (0.50–1.06)0.64 (0.43–0.93)0.021
      EPA + DHA
      Cases/n51/28235/28333/282160/282141/283129/282
      Model 11.000.64 (0.40–1.03)0.60 (0.37–0.96)0.0301.000.81 (0.58–1.15)0.64 (0.45–0.90)0.010
      Model 21.000.68 (0.42–1.11)0.67 (0.41–1.10)0.1051.000.78 (0.54–1.12)0.65 (0.45–0.94)0.021
      The median levels of erythrocyte membrane n-3 PUFA in each tertile were: 7.27%, 9.59% and 13.04% for total n-3 fatty acids; 0.06%, 0.10%, and 0.15% for ALA; 0.25%, 0.45%, and 0.74% for EPA; 5.11%, 7.15%, and 10.84% for DHA.
      Model 1: adjusted for age and sex. Model 2: further adjusted for waist circumference, alcohol consumption, smoking status, education, physical activity, SBP, DBP, TC, TG, HDL, LDL and fasting glucose. Model 3: further adjusted for the other two of three (ALA, DHA and EPA) fatty acids.
      Table 3Multivariate-adjusted odds ratios (95% CI) for carotid intima-media thickening by tertiles of erythrocyte n-3 PUFA concentrations stratified by gender, body mass index
      Covariates adjusted for: see Model 2 in Table 2, except for menopause status (yes or no) was further adjusted for in female in the sex-stratified analyses.
      .
      nOdds ratios (95% CI) for CCA thickeningP-value
      P-value for trend and P-value for interaction (italics).
      Odds ratios (95% CI) for BIF thickeningP-value
      P-value for trend and P-value for interaction (italics).
      Tertile 1Tertile 2Tertile 3Tertile 1Tertile 2Tertile 3
      ALA
      Gender0.1550.885
      Male2851.001.42 (0.65–3.11)0.59 (0.23–1.48)0.3291.000.56 (0.29–1.07)0.63 (0.32–1.23)0.158
      Female5621.000.40 (0.22–0.73)0.35 (0.18–0.68)0.0011.000.88 (0.56–1.36)0.58 (0.36–0.91)0.018
      BMI, kg/m20.7260.175
      <256101.000.59 (0.33–1.03)0.47 (0.25–0.87)0.0121.000.73 (0.48–1.13)0.47 (0.30–0.73)0.001
      ≥252371.000.80 (0.34–1.86)0.23 (0.07–0.73)0.0161.001.12 (0.56–2.24)1.09 (0.53–2.23)0.818
      DHA
      Gender0.6710.940
      Male2851.000.64 (0.28–1.48)0.52 (0.22–1.23)0.1301.000.57 (0.29–1.11)0.64 (0.32–1.27)0.210
      Female5621.000.88 (0.48–1.59)0.68 (0.35–1.30)0.2441.000.76 (0.49–1.19)0.58 (0.37–0.92)0.020
      BMI, kg/m20.9700.178
      <256101.000.85 (0.47–1.52)0.68 (0.37–1.26)0.2231.000.88 (0.57–1.37)0.67 (0.43–1.05)0.079
      ≥252371.000.54 (0.21–1.38)0.45 (0.17–1.17)0.0931.000.50 (0.24–1.01)0.50 (0.24–1.03)0.061
      EPA + DHA
      Gender0.3310.816
      Male2851.000.70 (0.31–1.59)0.48 (0.20–1.16)0.1011.000.66 (0.34–1.29)0.66 (0.33–1.29)0.228
      Female5621.000.78 (0.42–1.44)0.88 (0.47–1.64)0.6611.000.89 (0.57–1.39)0.63 (0.40–0.99)0.044
      BMI, kg/m20.7900.192
      <256101.000.81 (0.46–1.45)0.69 (0.37–1.27)0.2251.000.94 (0.61–1.46)0.70 (0.45–1.10)0.121
      ≥252371.000.46 (0.18–1.18)0.53 (0.21–1.32)0.1521.000.52 (0.25–1.05)0.48 (0.24–0.98)0.041
      a Covariates adjusted for: see Model 2 in Table 2, except for menopause status (yes or no) was further adjusted for in female in the sex-stratified analyses.
      b P-value for trend and P-value for interaction (italics).
      Table 4Odds ratios and 95% CI for the prevalence of carotid artery plaque by tertiles of erythrocyte n-3 PUFA concentrations.
      Odds ratios (95% CI) for carotid artery plaque
      Tertile 1Tertile 2Tertile 3P-trend
      n-3 PUFA
      Cases/n41/28230/28325/282
      Model 11.000.73 (0.44–1.23)0.57 (0.33–0.97)0.037
      Model 21.000.68 (0.39–1.18)0.54 (0.30–0.95)0.031
      ALA
      Cases/n41/28227/28028/285
      Model 11.000.67 (0.40–1.14)0.63 (0.37–1.06)0.076
      Model 21.000.62 (0.36–1.08)0.55 (0.31–0.97)0.034
      Model 31.000.66 (0.38–1.16)0.61 (0.34–1.09)0.083
      EPA
      Cases/n34/28233/28329/282
      Model 11.000.92 (0.55–1.56)0.90 (0.53–1.54)0.701
      Model 21.000.84 (0.48–1.45)0.92 (0.51–1.64)0.745
      Model 31.000.91 (0.51–1.59)0.96 (0.53–1.73)0.870
      DHA
      Cases/n40/28231/28325/282
      Model 11.000.77 (0.46–1.29)0.57 (0.33–0.98)0.041
      Model 21.000.69 (0.39–1.20)0.52 (0.29–0.94)0.031
      Model 31.000.71 (0.40–1.25)0.59 (0.32–1.08)0.082
      EPA + DHA
      Cases/n41/28231/28324/282
      Model 11.000.78 (0.47–1.29)0.54 (0.31–0.93)0.026
      Model 21.000.68 (0.39–1.18)0.48 (0.27–0.87)0.014
      Covariates adjusted for: see Model 1, Model 2 and Model 3 in Table 2.
      In stratified analyses (Table 3), the associations between the levels of ALA, DHA and EPA + DHA in erythrocyte and carotid atherosclerosis were not modified by sex or body mass index (all P interaction > 0.05).

      4. Discussion

      In this community-based study, higher levels of ALA, and DHA in the erythrocyte membrane were significantly and dose-dependently associated with a lower burden of carotid atherosclerosis in apparently healthy men and women with a low habitual n-3 fatty acid intake. These associations persisted after controlling for conventional CVD risk factors. No significant associations between EPA and the extent of carotid atherosclerosis were observed.
      In this study, we observed a significant inverse association between the levels of RBC DHA and the combination of DHA and EPA and the prevalence of carotid IMT thickening in our population with a relatively low intake of EPA + DHA (mean 54.5 mg/day). Our findings were consistent with a few studies including cross-sectional studies [
      • He K.
      • Liu K.
      • Daviglus M.L.
      • et al.
      Intakes of long-chain n-3 polyunsaturated fatty acids and fish in relation to measurements of subclinical atherosclerosis.
      ], prospective studies [
      • Hu F.B.
      • Bronner L.
      • Willett W.C.
      • et al.
      Fish and omega-3 fatty acid intake and risk of coronary heart disease in women.
      ,
      • Sun Q.
      • Ma J.
      • Campos H.
      • et al.
      Blood concentrations of individual long-chain n-3 fatty acids and risk of nonfatal myocardial infarction.
      ] and clinical trials [
      • Baldassarre D.
      • Amato M.
      • Eligini S.
      • et al.
      Effect of n-3 fatty acids on carotid atherosclerosis and haemostasis in patients with combined hyperlipoproteinemia: a double-blind pilot study in primary prevention.
      ,
      • von Schacky C.
      • Angerer P.
      • Kothny W.
      • et al.
      The effect of dietary omega-3 fatty acids on coronary atherosclerosis. A randomized, double-blind, placebo-controlled trial.
      ]. A population-based cross-sectional study showed an inverse association between fish and marine-derived n-3 fatty acids and carotid atherosclerosis among 5488 Americans [
      • He K.
      • Liu K.
      • Daviglus M.L.
      • et al.
      Intakes of long-chain n-3 polyunsaturated fatty acids and fish in relation to measurements of subclinical atherosclerosis.
      ]. A randomized, placebo-controlled trial further demonstrated a slower progression of atherosclerosis at the coronary [
      • von Schacky C.
      • Angerer P.
      • Kothny W.
      • et al.
      The effect of dietary omega-3 fatty acids on coronary atherosclerosis. A randomized, double-blind, placebo-controlled trial.
      ] (but not carotid [
      • Angerer P.
      • Kothny W.
      • Stork S.
      • et al.
      Effect of dietary supplementation with omega-3 fatty acids on progression of atherosclerosis in carotid arteries.
      ]) artery in 223 patients with established coronary artery disease who supplemented with a 1.65 g/day of EPA + DHA for 2 years. The significant benefit was also observed on the carotid artery when the dose was increased to 6 g/day [
      • Baldassarre D.
      • Amato M.
      • Eligini S.
      • et al.
      Effect of n-3 fatty acids on carotid atherosclerosis and haemostasis in patients with combined hyperlipoproteinemia: a double-blind pilot study in primary prevention.
      ]. The Nurses' Health Study, which included 84,688 women, showed that those with the dietary intakes of EPA and DHA in the highest (versus the lowest) quintile had a 27% (95% CI, 7%–43%) lower risk of non-fatal myocardial infarction (MI) during the 16 years of follow-up [
      • Hu F.B.
      • Bronner L.
      • Willett W.C.
      • et al.
      Fish and omega-3 fatty acid intake and risk of coronary heart disease in women.
      ]. Similar results were observed using plasma biomarkers of n-3 fatty acids in the Nurses' Health Study [
      • Sun Q.
      • Ma J.
      • Campos H.
      • et al.
      Blood concentrations of individual long-chain n-3 fatty acids and risk of nonfatal myocardial infarction.
      ]. However, many other studies did not observe a beneficial effect of DHA alone or the combination with EPA [
      • de Goede J.
      • Geleijnse J.M.
      • Boer J.M.
      • et al.
      Marine (n-3) fatty acids, fish consumption, and the 10-year risk of fatal and nonfatal coronary heart disease in a large population of Dutch adults with low fish intake.
      ,
      • Lemaitre R.N.
      • King I.B.
      • Mozaffarian D.
      • et al.
      n-3 Polyunsaturated fatty acids, fatal ischemic heart disease, and nonfatal myocardial infarction in older adults: the Cardiovascular Health Study.
      ]. The discrepancy in the results of these studies might be partly due to the differences in endpoints, exposure variables, exposure dosage or study design. The beneficial effect tended to be more significant in the studies with greater doses, or with the endpoints of atherosclerosis or fatal CHD (vs. non-fatal CHD).
      In the current analysis, ALA in erythrocyte membrane was strongly and inversely associated with carotid wall thickening and carotid plaque after adjustment for potential confounders. Similar inverse associations were observed for atherosclerosis in cross-sectional studies [
      • Djousse L.
      • Folsom A.R.
      • Province M.A.
      • et al.
      Dietary linolenic acid and carotid atherosclerosis: the National Heart, Lung, and Blood Institute Family Heart Study.
      ,
      • Djousse L.
      • Arnett D.K.
      • Carr J.J.
      • et al.
      Dietary linolenic acid is inversely associated with calcified atherosclerotic plaque in the coronary arteries: the National Heart, Lung, and Blood Institute Family Heart Study.
      ] and for non-fatal MI and total CHD in the Health Professionals Follow-up Study [
      • Mozaffarian D.
      • Ascherio A.
      • Hu F.B.
      • et al.
      Interplay between different polyunsaturated fatty acids and risk of coronary heart disease in men.
      ]. As reviewed by Vedtofte et al. [
      • Vedtofte M.S.
      • Jakobsen M.U.
      • Lauritzen L.
      • et al.
      The role of essential fatty acids in the control of coronary heart disease.
      ], many other human studies found a null or marginally significant benefit of ALA on CVD risk. However, a recent meta-analysis of 27 observational studies indicated that greater level of dietary ALA (RR: 0.90, 95% CI: 0.81–0.99) and the combination of dietary ALA and ALA biomarkers (RR: 0.86, 95% CI: 0.77–0.97), were significantly associated with a lower risk of total CVD events [
      • Pan A.
      • Chen M.
      • Chowdhury R.
      • et al.
      alpha-Linolenic acid and risk of cardiovascular disease: a systematic review and meta-analysis.
      ]. A borderline significant association of ALA biomarker with total CVD risk (RR: 0.80, 95% CI: 0.63–1.03) was observed in the meta-analysis possibly due to the relatively smaller study size and greater between-study heterogeneity [
      • Pan A.
      • Chen M.
      • Chowdhury R.
      • et al.
      alpha-Linolenic acid and risk of cardiovascular disease: a systematic review and meta-analysis.
      ].
      Few RCTs examined the effects of ALA intervention on CVD risks. A 6-week treatment with a moderate-dose of ALA (4.4 g/d) significantly decreased fasting serum triacylglycerol concentrations in 74 normolipidemic men and women [
      • Egert S.
      • Kannenberg F.
      • Somoza V.
      • et al.
      Dietary alpha-linolenic acid, EPA, and DHA have differential effects on LDL fatty acid composition but similar effects on serum lipid profiles in normolipidemic humans.
      ]. Nonetheless, data from RCTs has not shown convincing beneficial effects of ALA on the risk of major cardiovascular events in 4837 patients who had a myocardial infarction (The Alpha Omega Trial) [
      • Kromhout D.
      • Giltay E.J.
      • Geleijnse J.M.
      n-3 fatty acids and cardiovascular events after myocardial infarction.
      ] or the 2-year changes in carotid IMT in 110 patients at high risk for coronary heart disease (The MARGARIN study) [
      • Bemelmans W.J.
      • Lefrandt J.D.
      • Feskens E.J.
      • et al.
      Increased alpha-linolenic acid intake lowers C-reactive protein, but has no effect on markers of atherosclerosis.
      ]. In the Alpha Omega Trial, using the combination groups of EPA-DHA supplementation and placebo as the comparison group and limited case number might partially explain the null effect; moreover, statin treatment in the majority of participants in the study might also mask the effect of ALA [
      • Eussen S.R.
      • Geleijnse J.M.
      • Giltay E.J.
      • et al.
      Effects of n-3 fatty acids on major cardiovascular events in statin users and non-users with a history of myocardial infarction.
      ]. Small study size (55 subjects in ALA group), no placebo (or blank) control and large between-individual variation in the 2-y changes in IMT might limit the power to detect a significant benefit in the latter RCT [
      • Bemelmans W.J.
      • Lefrandt J.D.
      • Feskens E.J.
      • et al.
      Increased alpha-linolenic acid intake lowers C-reactive protein, but has no effect on markers of atherosclerosis.
      ].
      In summary, the health benefit of ALA has not been consistently observed in the previous studies. Several reasons might contribute to the study heterogeneity: (1) possibly a small effect size of ALA on CVD risks; (2) varied intake in different study populations; (3) the ALA effects might be modified by the intake of n-6 fatty acids or other n-3 long-chain fatty acids in different populations [
      • Vedtofte M.S.
      • Jakobsen M.U.
      • Lauritzen L.
      • et al.
      The role of essential fatty acids in the control of coronary heart disease.
      ]. Previous study showed that the benefits of ALA on CVD events were seen when EPA + DHA intake was very low (<100 mg/d) in the Health Professionals Study [
      • Mozaffarian D.
      • Ascherio A.
      • Hu F.B.
      • et al.
      Interplay between different polyunsaturated fatty acids and risk of coronary heart disease in men.
      ]. Due to the inherent limitations in the previous studies, further well-designed RCTs are needed to verify the ALA effect on cardiovascular health.
      Several physiologic effects of DHA and ALA could potentially explain our findings. As summarized by Siegel et al. [
      • Siegel G.
      • Ermilov E.
      Omega-3 fatty acids: benefits for cardio-cerebro-vascular diseases.
      ], n-3 fatty acids might prevent atherosclerosis by reducing the quotient oxLDL/LDL [
      • Koppe C.
      • Rodriguez M.
      • Winkler K.
      • et al.
      Reduction of arteriosclerotic nanoplaque formation and size by n-3 fatty acids in patients after valvular defect operation.
      ], decreasing inflammation [
      • Zhao G.
      • Etherton T.D.
      • Martin K.R.
      • et al.
      Anti-inflammatory effects of polyunsaturated fatty acids in THP-1 cells.
      ], improving blood lipids [
      • Egert S.
      • Kannenberg F.
      • Somoza V.
      • et al.
      Dietary alpha-linolenic acid, EPA, and DHA have differential effects on LDL fatty acid composition but similar effects on serum lipid profiles in normolipidemic humans.
      ] or endothelial function [
      • Thies F.
      • Miles E.A.
      • Nebe-von-Caron G.
      • et al.
      Influence of dietary supplementation with long-chain n-3 or n-6 polyunsaturated fatty acids on blood inflammatory cell populations and functions and on plasma soluble adhesion molecules in healthy adults.
      ]. An in-vitro study reported that both ALA and DHA suppressed the expression of pro-inflammatory cytokines, e.g., interleukin-6, interleukin-1, and tumor necrosis factor-α, via a reduction in nuclear factor κB (NF-κB)–DNA binding activity in THP-1 cells [
      • Zhao G.
      • Etherton T.D.
      • Martin K.R.
      • et al.
      Anti-inflammatory effects of polyunsaturated fatty acids in THP-1 cells.
      ]. Supplementation with moderate levels of ALA (2 g/day) and EPA + DHA (720 mg EPA + 280 mg DHA/day) for 12 weeks improved endothelial function in healthy subjects, by decreasing the plasma concentrations of soluble vascular cell adhesion molecule-1 [
      • Thies F.
      • Miles E.A.
      • Nebe-von-Caron G.
      • et al.
      Influence of dietary supplementation with long-chain n-3 or n-6 polyunsaturated fatty acids on blood inflammatory cell populations and functions and on plasma soluble adhesion molecules in healthy adults.
      ]. The favorable effects of ALA and DHA on decreasing lipids in intervention studies also supported our results [
      • Egert S.
      • Kannenberg F.
      • Somoza V.
      • et al.
      Dietary alpha-linolenic acid, EPA, and DHA have differential effects on LDL fatty acid composition but similar effects on serum lipid profiles in normolipidemic humans.
      ]. In addition, DHA might also prevent atherosclerosis by reducing the quotient oxLDL/LDL [
      • Siegel G.
      • Ermilov E.
      Omega-3 fatty acids: benefits for cardio-cerebro-vascular diseases.
      ,
      • Koppe C.
      • Rodriguez M.
      • Winkler K.
      • et al.
      Reduction of arteriosclerotic nanoplaque formation and size by n-3 fatty acids in patients after valvular defect operation.
      ], the latter may initiate the development of atherosclerosis.
      We found no significant inverse associations between EPA and the prevalence of carotid atherosclerosis. In consistent with our results, a meta-analysis including case–control and prospective cohort studies reported that the levels of DHA, but not EPA, in tissue/blood/plasma were significantly lower in individuals who experienced CHD events than in the controls [
      • Harris W.S.
      • Poston W.C.
      • Haddock C.K.
      Tissue n-3 and n-6 fatty acids and risk for coronary heart disease events.
      ]. A review that summarized the effects of supplementation with either purified EPA, DHA or both on cardiovascular risk factors demonstrated that DHA may have a greater effect than EPA on most of the risk factors by reducing blood pressure, increasing high-density lipoprotein cholesterol, and attenuating platelet function [
      • Mori T.A.
      • Woodman R.J.
      The independent effects of eicosapentaenoic acid and docosahexaenoic acid on cardiovascular risk factors in humans.
      ]. A randomized controlled trial has shown that DHA rather than EPA increased HDLc and LDL particle size [
      • Mori T.A.
      • Burke V.
      • Puddey I.B.
      • et al.
      Purified eicosapentaenoic and docosahexaenoic acids have differential effects on serum lipids and lipoproteins, LDL particle size, glucose, and insulin in mildly hyperlipidemic men.
      ]. Therefore, DHA might be more relevant to cardiovascular benefits than EPA.
      Our findings showed that n-3 fatty acids estimated by RBC biomarkers but not dietary intakes by FFQ were associated with a decreased prevalence of atherosclerosis. In consideration of the limitation in food composition database for assessing fatty acid contents, the uncertainty of cooking oil contents (particularly when eat out), the difficulty in reporting oil/fat intake in Chinese, and great variations in the biologic processes such as absorption, incorporation, and metabolism of dietary n-3 fatty acids, the use of objective RBC biomarkers might provide much more accurate and precise estimates of the internal dose of n-3 fatty acids than those of dietary assessments by FFQ, and would thus improve the validity of the associations between n-3 fatty acids and carotid atherosclerosis in our study.
      This study had some limitations. The development of carotid artery atherosclerosis is a long-term process, while n-3 fatty acid content in the erythrocyte membrane reflects the intake status within recent months. Therefore, a causal relationship cannot be inferred. However, we excluded all subjects who had previously confirmed conditions that might have changed their dietary habits, including dyslipidemia, diabetes, CVD and cancers, to avoid a reverse causal relationship. This study was limited to healthy adults and moderate reproducibility of carotid IMT measurement (especially at the carotid bifurcation), which might have underestimated the associations. Thus, the power to detect an interaction between n-3 fatty acids and gender, body mass index or other factors on the risk of carotid atherosclerosis may have been insufficient.
      In conclusion, this study indicated that the levels of ALA and DHA in the erythrocyte membrane were significantly and inversely associated with the prevalence of carotid atherosclerosis in a middle-aged and older Chinese population with a low intake of n-3 fatty acids.

      Conflicts of interest

      None of the authors had any conflict of interest.

      Acknowledgments

      Source of funding: The present study was supported by the National Natural Science Foundation of China ( 30872102 ). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. We are grateful for the help of other staff, and the doctors in the First Affiliated Hospital of Sun Yat-Sen University in data collection in this study.

      Appendix A. Supplementary data

      The following is the supplementary data related to this article:

      References

        • Hu F.B.
        • Bronner L.
        • Willett W.C.
        • et al.
        Fish and omega-3 fatty acid intake and risk of coronary heart disease in women.
        JAMA. 2002; 287: 1815-1821
        • Yuan J.M.
        • Ross R.K.
        • Gao Y.T.
        • et al.
        Fish and shellfish consumption in relation to death from myocardial infarction among men in Shanghai, China.
        Am J Epidemiol. 2001; 154: 809-816
        • de Goede J.
        • Geleijnse J.M.
        • Boer J.M.
        • et al.
        Marine (n-3) fatty acids, fish consumption, and the 10-year risk of fatal and nonfatal coronary heart disease in a large population of Dutch adults with low fish intake.
        J Nutr. 2010; 140: 1023-1028
        • Marchioli R.
        • Barzi F.
        • Bomba E.
        • et al.
        Early protection against sudden death by n-3 polyunsaturated fatty acids after myocardial infarction: time-course analysis of the results of the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardico (GISSI)-Prevenzione.
        Circulation. 2002; 105: 1897-1903
        • Tavazzi L.
        • Maggioni A.P.
        • Marchioli R.
        • et al.
        Effect of n-3 polyunsaturated fatty acids in patients with chronic heart failure (the GISSI-HF trial): a randomised, double-blind, placebo-controlled trial.
        Lancet. 2008; 372: 1223-1230
        • He K.
        • Song Y.
        • Daviglus M.L.
        • et al.
        Accumulated evidence on fish consumption and coronary heart disease mortality: a meta-analysis of cohort studies.
        Circulation. 2004; 109: 2705-2711
        • Leon H.
        • Shibata M.C.
        • Sivakumaran S.
        • et al.
        Effect of fish oil on arrhythmias and mortality: systematic review.
        BMJ. 2008; 337: a2931
        • Mozaffarian D.
        Fish and n-3 fatty acids for the prevention of fatal coronary heart disease and sudden cardiac death.
        Am J Clin Nutr. 2008; 87: 1991S-1996S
        • Baldassarre D.
        • Amato M.
        • Eligini S.
        • et al.
        Effect of n-3 fatty acids on carotid atherosclerosis and haemostasis in patients with combined hyperlipoproteinemia: a double-blind pilot study in primary prevention.
        Ann Med. 2006; 38: 367-375
        • Mita T.
        • Watada H.
        • Ogihara T.
        • et al.
        Eicosapentaenoic acid reduces the progression of carotid intima-media thickness in patients with type 2 diabetes.
        Atherosclerosis. 2007; 191: 162-167
        • von Schacky C.
        • Angerer P.
        • Kothny W.
        • et al.
        The effect of dietary omega-3 fatty acids on coronary atherosclerosis. A randomized, double-blind, placebo-controlled trial.
        Ann Intern Med. 1999; 130: 554-562
        • Hino A.
        • Adachi H.
        • Toyomasu K.
        • et al.
        Very long chain N-3 fatty acids intake and carotid atherosclerosis: an epidemiological study evaluated by ultrasonography.
        Atherosclerosis. 2004; 176: 145-149
        • Sekikawa A.
        • Curb J.D.
        • Ueshima H.
        • et al.
        Marine-derived n-3 fatty acids and atherosclerosis in Japanese, Japanese-American, and white men: a cross-sectional study.
        J Am Coll Cardiol. 2008; 52: 417-424
        • Iso H.
        • Kobayashi M.
        • Ishihara J.
        • et al.
        Intake of fish and n3 fatty acids and risk of coronary heart disease among Japanese: the Japan Public Health Center-Based (JPHC) Study Cohort I.
        Circulation. 2006; 113: 195-202
        • Sala-Vila A.
        • Cofan M.
        • Perez-Heras A.
        • et al.
        Fatty acids in serum phospholipids and carotid intima-media thickness in Spanish subjects with primary dyslipidemia.
        Am J Clin Nutr. 2010; 92: 186-193
        • He K.
        • Liu K.
        • Daviglus M.L.
        • et al.
        Intakes of long-chain n-3 polyunsaturated fatty acids and fish in relation to measurements of subclinical atherosclerosis.
        Am J Clin Nutr. 2008; 88: 1111-1118
        • Villegas R.
        • Xiang Y.B.
        • Elasy T.
        • et al.
        Fish, shellfish, and long-chain n-3 fatty acid consumption and risk of incident type 2 diabetes in middle-aged Chinese men and women.
        Am J Clin Nutr. 2011; 94: 543-551
        • Mozaffarian D.
        • Ascherio A.
        • Hu F.B.
        • et al.
        Interplay between different polyunsaturated fatty acids and risk of coronary heart disease in men.
        Circulation. 2005; 111: 157-164
        • Campos H.
        • Baylin A.
        • Willett W.C.
        Alpha-linolenic acid and risk of nonfatal acute myocardial infarction.
        Circulation. 2008; 118: 339-345
        • Djousse L.
        • Folsom A.R.
        • Province M.A.
        • et al.
        Dietary linolenic acid and carotid atherosclerosis: the National Heart, Lung, and Blood Institute Family Heart Study.
        Am J Clin Nutr. 2003; 77: 819-825
        • Djousse L.
        • Arnett D.K.
        • Carr J.J.
        • et al.
        Dietary linolenic acid is inversely associated with calcified atherosclerotic plaque in the coronary arteries: the National Heart, Lung, and Blood Institute Family Heart Study.
        Circulation. 2005; 111: 2921-2926
        • Burdge G.C.
        • Finnegan Y.E.
        • Minihane A.M.
        • et al.
        Effect of altered dietary n-3 fatty acid intake upon plasma lipid fatty acid composition, conversion of [13C]alpha-linolenic acid to longer-chain fatty acids and partitioning towards beta-oxidation in older men.
        Br J Nutr. 2003; 90: 311-321
        • Burdge G.
        Alpha-linolenic acid metabolism in men and women: nutritional and biological implications.
        Curr Opin Clin Nutr Metab Care. 2004; 7: 137-144
        • Lorenz M.W.
        • Markus H.S.
        • Bots M.L.
        • et al.
        Prediction of clinical cardiovascular events with carotid intima-media thickness: a systematic review and meta-analysis.
        Circulation. 2007; 115: 459-467
        • Zhang B.
        • Wang P.
        • Chen C.G.
        • et al.
        Validation of an FFQ to estimate the intake of fatty acids using erythrocyte membrane fatty acids and multiple 3d dietary records.
        Public Health Nutr. 2010; 13: 1546-1552
      1. Yang Y.X. Wang G.Y. Pan X.C. China food composition table. Peking University Medical Press, Beijing2002
        • Wang P.
        • Chen Y.M.
        • He L.P.
        • et al.
        Association of natural intake of dietary plant sterols with carotid intima-media thickness and blood lipids in Chinese adults: a cross-section study.
        PLoS One. 2012; 7: e32736
        • Touboul P.J.
        • Hennerici M.G.
        • Meairs S.
        • et al.
        Mannheim carotid intima-media thickness consensus (2004–2006). An update on behalf of the Advisory Board of the 3rd and 4th Watching the Risk Symposium, 13th and 15th European Stroke Conferences, Mannheim, Germany, 2004, and Brussels, Belgium, 2006.
        Cerebrovasc Dis. 2007; 23: 75-80
        • Willett W.
        • Stampfer M.J.
        Total energy intake: implications for epidemiologic analyses.
        Am J Epidemiol. 1986; 124: 17-27
        • Greenland P.
        • Alpert J.S.
        • Beller G.A.
        • et al.
        ACCF/AHA guideline for assessment of cardiovascular risk in asymptomatic adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.
        Circulation. 2010; 122: e584-e636
        • Sun Q.
        • Ma J.
        • Campos H.
        • et al.
        Blood concentrations of individual long-chain n-3 fatty acids and risk of nonfatal myocardial infarction.
        Am J Clin Nutr. 2008; 88: 216-223
        • Angerer P.
        • Kothny W.
        • Stork S.
        • et al.
        Effect of dietary supplementation with omega-3 fatty acids on progression of atherosclerosis in carotid arteries.
        Cardiovasc Res. 2002; 54: 183-190
        • Lemaitre R.N.
        • King I.B.
        • Mozaffarian D.
        • et al.
        n-3 Polyunsaturated fatty acids, fatal ischemic heart disease, and nonfatal myocardial infarction in older adults: the Cardiovascular Health Study.
        Am J Clin Nutr. 2003; 77: 319-325
        • Vedtofte M.S.
        • Jakobsen M.U.
        • Lauritzen L.
        • et al.
        The role of essential fatty acids in the control of coronary heart disease.
        Curr Opin Clin Nutr Metab Care. 2012; 15: 592-596
        • Pan A.
        • Chen M.
        • Chowdhury R.
        • et al.
        alpha-Linolenic acid and risk of cardiovascular disease: a systematic review and meta-analysis.
        Am J Clin Nutr. 2012; 96: 1262-1273
        • Egert S.
        • Kannenberg F.
        • Somoza V.
        • et al.
        Dietary alpha-linolenic acid, EPA, and DHA have differential effects on LDL fatty acid composition but similar effects on serum lipid profiles in normolipidemic humans.
        J Nutr. 2009; 139: 861-868
        • Kromhout D.
        • Giltay E.J.
        • Geleijnse J.M.
        n-3 fatty acids and cardiovascular events after myocardial infarction.
        N Engl J Med. 2010; 363: 2015-2026
        • Bemelmans W.J.
        • Lefrandt J.D.
        • Feskens E.J.
        • et al.
        Increased alpha-linolenic acid intake lowers C-reactive protein, but has no effect on markers of atherosclerosis.
        Eur J Clin Nutr. 2004; 58: 1083-1089
        • Eussen S.R.
        • Geleijnse J.M.
        • Giltay E.J.
        • et al.
        Effects of n-3 fatty acids on major cardiovascular events in statin users and non-users with a history of myocardial infarction.
        Eur Heart J. 2012; 33: 1582-1588
        • Siegel G.
        • Ermilov E.
        Omega-3 fatty acids: benefits for cardio-cerebro-vascular diseases.
        Atherosclerosis. 2012; 225: 291-295
        • Koppe C.
        • Rodriguez M.
        • Winkler K.
        • et al.
        Reduction of arteriosclerotic nanoplaque formation and size by n-3 fatty acids in patients after valvular defect operation.
        Forsch Komplementmed. 2009; 16: 237-245
        • Zhao G.
        • Etherton T.D.
        • Martin K.R.
        • et al.
        Anti-inflammatory effects of polyunsaturated fatty acids in THP-1 cells.
        Biochem Biophys Res Commun. 2005; 336: 909-917
        • Thies F.
        • Miles E.A.
        • Nebe-von-Caron G.
        • et al.
        Influence of dietary supplementation with long-chain n-3 or n-6 polyunsaturated fatty acids on blood inflammatory cell populations and functions and on plasma soluble adhesion molecules in healthy adults.
        Lipids. 2001; 36: 1183-1193
        • Harris W.S.
        • Poston W.C.
        • Haddock C.K.
        Tissue n-3 and n-6 fatty acids and risk for coronary heart disease events.
        Atherosclerosis. 2007; 193: 1-10
        • Mori T.A.
        • Woodman R.J.
        The independent effects of eicosapentaenoic acid and docosahexaenoic acid on cardiovascular risk factors in humans.
        Curr Opin Clin Nutr Metab Care. 2006; 9: 95-104
        • Mori T.A.
        • Burke V.
        • Puddey I.B.
        • et al.
        Purified eicosapentaenoic and docosahexaenoic acids have differential effects on serum lipids and lipoproteins, LDL particle size, glucose, and insulin in mildly hyperlipidemic men.
        Am J Clin Nutr. 2000; 71: 1085-1094