Long-chain n–3 polyunsaturated fatty acids in plasma in British meat-eating, vegetarian, and vegan men

¹ From the Cancer Research UK Epidemiology Unit, University of Oxford, Oxford, United Kingdom (MSR, PNA, NEA, and TJK), and the Nutrition, Food and Health Research Center, King’s College London, London, United Kingdom (ZL-W and TABS)

² The EPIC-Oxford study is supported by Cancer Research UK, the Medical Research Council, and the European Commission under the Europe Against Cancer Programme. MSR was supported by a grant from the Swedish Council for Working Life and Social Research.

³ Reprints not available. Address correspondence to MS Rosell, Cancer Research UK Epidemiology Unit, University of Oxford, Gibson Building, Radcliffe Infirmary, Oxford OX2 6HE, United Kingdom. E-mail: magdalena.rosell{at}cancer.org.uk.

ABSTRACT  
Background: Plasma concentrations of long-chain n–3 polyunsaturated fatty acids are lower in vegetarians and in vegans than in omnivores. No data are available on whether these concentrations differ between long- and short-term vegetarians and vegans.

Objectives: We compared plasma fatty acid composition in meat-eaters, vegetarians, and vegans and examined whether the proportions of eicosapentaenoic acid (20:5n–3; EPA), docosapentaenoic acid (22:5n–3; DPA), and docosahexaenoic acid (22:6n–3; DHA) were related to the subjects’ duration of adherence to their diets or to the proportions of plasma linoleic acid (18:2n–6; LA) and -linolenic acid (18:3n-3; ALA).

Design: The present cross-sectional study included 196 meat-eating, 231 vegetarian, and 232 vegan men in the United Kingdom. Information on anthropometry, diet, and smoking habits was obtained through a questionnaire. Total fatty acid composition in plasma was measured.

Results: The proportions of plasma EPA and DHA were lower in the vegetarians and in the vegans than in the meat-eaters, whereas only small differences were seen for DPA. Plasma EPA, DPA, and DHA proportions were not significantly associated with the duration of time since the subjects became vegetarian or vegan, which ranged from <1 y to >20 y. In the vegetarians and the vegans, plasma DHA was inversely correlated with plasma LA.

Conclusions: The proportions of plasma long-chain n–3 fatty acids were not significantly affected by the duration of adherence to a vegetarian or vegan diet. This finding suggests that when animal foods are wholly excluded from the diet, the endogenous production of EPA and DHA results in low but stable plasma concentrations of these fatty acids.

Key Words: n–3 Fatty acids • vegetarians • vegans • cross-sectional study

INTRODUCTION  
Apart from being a source of energy, fatty acids have a wide range of physiologic functions. Many fatty acids can be produced endogenously; however, n–3 and n–6 polyunsaturated fatty acids are essential fatty acids that must be provided in the diet. The long-chain metabolites of these essential fatty acids are needed for cellular membrane functions and the production of eicosanoids, which play a role in inflammatory reactions, blood pressure control, and platelet aggregation. These essential fatty acids also influence gene regulation; for example, they act as ligands of peroxisome proliferator-activated receptors that are involved in growth and development. Of particular interest are the long-chain n–3 fatty acids eicosapentaenoic acid (20:5n–3; EPA) and docosahexaenoic acid (22:6n–3; DHA), which are abundant in oily fish and therefore are also referred to as the marine fatty acids. Although their action is not fully understood, these fatty acids probably account for the inverse relation between fish consumption and the risk of developing coronary heart disease and stroke that has been observed in epidemiologic studies (1–4). Apart from its probable cardioprotective effects, DHA is, with arachidonic acid, one of the 2 most prevalent polyunsaturated fatty acids in brain and retinal phospholipids and plays a role in normal neurotransmission and visual function.

The major dietary source of EPA and DHA is seafood, but they can also be found in lean red meat, in organ meats such as liver and brain, and in eggs (5, 6); very small amounts are found in dairy products. Consequently, vegetarians, who do not eat meat or fish, and vegans, who in addition exclude eggs and dairy products from their diets, have very low or negligible intakes of EPA and DHA. However, vegetarians and vegans have a relatively high intake of linoleic acid (18:2n–6; LA) (7, 8), which is largely derived from plant sources. EPA and DHA can also be synthesized in the body from -linolenic acid (18:3n-3; ALA); however, this conversion is limited in humans (9), and possibly more in men than in women (10, 11). The conversion may be further suppressed by a high intake of LA, because the desaturation of ALA and LA involves the same rate-limiting ⁶ desaturase enzyme (12). Indeed, trials have shown that ALA supplementation increases plasma EPA and DPA but has little effect on DHA (13–15). As a result, vegetarians and vegans, who have a low dietary intake of DHA (and a correspondingly high intake of LA), have lower amounts of EPA and DHA in blood and adipose tissue than do omnivores (16–19).

Previous studies that compared the plasma fatty acid composition in different dietary groups involved a small number of subjects, and few studies included both vegetarians and vegans. The present study reports the plasma fatty acid composition in British meat-eating, vegetarian, and vegan men and explores the effect of duration of adherence to a vegetarian or vegan diet on plasma fatty acid composition.

SUBJECTS AND METHODS  
Subjects
The subjects of the study were men selected from the Oxford arm of the European Prospective Investigation into Cancer and Nutrition (EPIC-Oxford). The EPIC-Oxford cohort consists of 65500 participants aged 20 y who lived in the United Kingdom between 1993 and 1999. The aim of the EPIC-Oxford study was to recruit participants with a wide range of diets by targeting vegetarians, vegans, and the general UK population (20). The study protocol was approved by a multicenter research ethics committee. The participants were recruited through collaborating general practitioners, vegetarian and vegan societies, and vegetarian and health food magazines or were friends or relatives of other participants. The participants completed a questionnaire that included details on age, anthropometry, diet, smoking, and other lifestyle factors. Blood samples were obtained from 19700 volunteers. The present study is based on a sample of men who were originally selected to investigate the relation between diet and the hormones involved in the etiology of prostate cancer (21). The sample consisted of all the vegan men who were recruited to EPIC-Oxford between 1994 and 1997 and gave a blood sample and approximately equal numbers of meat-eaters and vegetarians who were randomly selected from the meat-eating and vegetarian men who gave blood samples. The men had no self-reported history of cancer and were not taking medication that influenced hormone concentrations. For the current study, we excluded subjects without data on plasma fatty acid composition (n = 12) and subjects reporting that they were taking fish oil supplements (n = 25), which left 659 men for the analyses (196 meat-eaters, 231 vegetarians, and 232 vegans).

Diet groups and duration of diet adherence
The classification of diet group and duration of adherence to a vegetarian or vegan diet was based on the following 4 questions: 1) "Do you eat any meat (including bacon, ham, poultry, game, meat pies, and sausages)? (Yes or No). If no, how old were you when you last ate meat?"; 2) "Do you eat any fish? (Yes or No). If no, how old were you when you last ate fish?"; 3) "Do you eat any dairy products (including milk, cheese, butter, and yogurt)? (Yes or No). If no, how old were you when you last ate dairy products?"; and 4) "Do you eat any eggs (including eggs in cakes and other baked foods)? (Yes or No). If no, how old were you when you last ate eggs?" Men who reported that they ate meat were classified as meat-eaters. Men who reported that they did not eat meat or fish but reported that they ate dairy products or eggs were classified as vegetarians; the duration of adherence to a vegetarian diet was calculated as the age at recruitment minus the age at which the respondent last ate meat or fish. Men who reported that they did not eat any foods of animal origin were classified as vegans; the duration of adherence to a vegan diet was calculated as the age at recruitment minus the age at which the respondent last ate any foods of animal origin.

Plasma fatty acid composition
Blood samples were obtained from the participants on average 5 mo after they had completed the questionnaire. The blood samples were collected into 10-mL Safety-Monovettes (Sarstedt, Nümbrecht, Germany) containing 1 mL of 3.13% trisodium citrate. The blood samples were then sent by mail to the central laboratory, where they were separated into serum, plasma, buffy coat, and red blood cell fractions. The plasma samples were frozen at –70°C and stored for 4 y until analyzed. The concentration of total plasma esterified and nonesterified fatty acids was measured by transmethylating the total lipids with methanolic HCl in the presence of a pentadecanoic acid internal standard and then analyzing the methyl esters by capillary gas-liquid chromatography, as described previously (22). The proportions of fatty acids in the plasma were calculated as the concentration of the fatty acid (in mg/L) divided by the sum of the concentrations of all the fatty acids analyzed. The stability of EPA, DPA, DHA, LA, and ALA was assessed by examining the associations between the percentages of these fatty acids in the plasma and the duration of storage at –70°C; no significant associations were found.

Fish and egg intake
Dietary intake during the 12 mo prior to recruitment was assessed with the use of a 130-item food-frequency questionnaire (FFQ) (18). Six questions related to the consumption of fish: 1) fried fish in batter, such as fish and chips; 2) fish fingers and fish cakes; 3) other fresh or frozen white fish (eg, cod, haddock, plaice, sole, and halibut); 4) fresh or canned oily fish (eg, mackerel, kippers, tuna, salmon, sardines, and herring; tuna was included because fresh tuna is classed as an oily fish, although canned tuna is not); 5) shellfish (eg, crab, prawns, and mussels); and 6) fish roe and taramasalata. One question related to the intake of eggs. Each question had 9 possible answers: never or <1/mo, 1–3/mo, 1/wk, 2–4/wk, 5–6/wk, 1/d, 2–3/d, 4–5/d, and 6/d. For each fish food question, the subjects were categorized into 3 groups of intake: never or <1/mo, 1–3/mo, and 1/wk. The range of intake of eggs was wider than the intakes of fish, and analyses were also performed with egg intake as a continuous variable. The frequency of oily fish intake was also calculated from 7-d food records that were completed by 135 of the meat-eating men.

Other variables
Information on age, height, body weight, smoking habits, and intakes of energy, macronutrients, fiber, and alcohol was obtained through the questionnaire. Body mass index (BMI) was calculated as body wt (in kg)/height² (in m).

Statistical analysis
The differences between the diet groups in age, BMI, and other characteristics were calculated with an analysis of variance or a chi-square test for categorical variables. Arithmetic means or geometric means [for 14:0, 16:1, 18:0, ratio LA:ALA, cis-9,trans-11-conjugated linoleic acid (CLA), 22:1, 20:3, and 20:5, for which the distributions were more normal when log-transformed] and 95% CIs of the proportions of the fatty acids by diet group were calculated from fitted values of multiple linear regression models that were adjusted for age (categorized as 10-y age bands), BMI (categorized by quartiles), cigarette smoking (never, former, current <10 cigarettes/d, current 10 cigarettes/d), and alcohol consumption (categorized by quartiles). The mean fatty acid proportions in each diet group pairing were compared by using standard t tests arising from the multiple regression analyses; the P values for each comparison were corrected for multiple testing with Bonferroni’s method. To investigate the effect of the duration of adherence to a vegetarian or a vegan diet on the proportions of the long-chain polyunsaturated fatty acids, we subdivided the vegetarians and the vegans into 4 categories according to the number of years they had followed their diet (0–4, 5–8, 9–15, or 16 y). The mean values and 95% CIs for the 9 diet-duration groups (meat-eaters, 4 groups of vegetarians, and 4 groups of vegans) were calculated with a multilinear regression, which was adjusted for age, BMI, smoking, and alcohol intake. P values for trend of the duration of adherence to the diet were calculated in separate subset analyses for the vegetarians and for the vegans by treating the duration of time as a continuous variable in the regression analysis. Correlations between the proportion of the long-chain polyunsaturated fatty acids and LA, ALA, or LA:ALA within each diet group were calculated with a partial Pearson’s correlation that was adjusted for age, BMI, smoking, and alcohol intake. Adjusted means of the proportions of the long-chain polyunsaturated fatty acids by categories of fish intake, together with their 95% CIs and P values for linear trend, were calculated with regression models as described above. All analyses were performed by using version 8.1 of the STATA statistical package (Stata Corp, College Station, TX).

RESULTS  
The mean age of the participants was 46.8 y (range: 20–78 y). The meat-eaters were on average 6 y older than the vegetarians and 9 y older than the vegans (Table 1). Mean BMI and the prevalence of smoking were highest in the meat-eaters and lowest in the vegans. In the vegetarians, the median duration of time since becoming vegetarian was 11 y and ranged from <1 to 76 y. The corresponding median duration for the vegans was 7 y and ranged from <1 to 36 y. There were significant differences in the intakes of energy, macronutrients, fiber, and alcohol between the diet groups. According to the FFQ, 30% of the meat-eaters reported eating 1 serving of oily fish/wk. However, of these men only 13 (6%) reported eating >1 serving of oily fish/wk (data not shown). The reported intake of fried fish in the FFQ was slightly lower and the reported intake of other white fish was somewhat higher than the intake of oily fish. Of the meat-eating men, only 7%, 4%, and 3% consumed 1 serving of fish cake or fish fingers, shellfish, and fish roe, respectively, per week (data not shown).

View this table:
TABLE 1. Characteristics of the men by diet group

 
The mean concentration of total plasma fatty acids, adjusted for age, BMI, smoking, and alcohol intake, was 15% lower in the vegetarians and 25% lower in the vegans than in the meat-eaters (Table 2). To allow for these differences, the means of the separate fatty acids in the diet groups are presented as proportions, which were calculated as the concentration of each fatty acid divided by the concentration of total fatty acids. Overall, the most prevalent fatty acids were palmitic acid, oleic acid, and LA, which together composed, on average, 79% (range: 41–89%) of the assessed fatty acids. Myristic acid, stearic acid, palmitoleic acid, ALA, dihomo--linolenic acid, arachidonic acid, and DHA each composed 1–7% of the assessed fatty acids. The proportions of arachidic acid, cis-9,trans-11-CLA, trans-10,cis-12-CLA, erucic acid, EPA, and DPA were each <1%.

View this table:
TABLE 2. Concentrations of total plasma fatty acids and proportions of plasma fatty acids by diet group¹

 
Of the saturated fatty acids, the proportions of myristic and palmitic acid were 40% and 11% lower, respectively, in the vegans than in the meat-eaters; there were no significant differences in the proportions of plasma stearic or arachidic acid between diet groups. The proportion of palmitoleic acid in the vegans was less than one-half that seen in the meat-eaters, whereas no significant differences were seen for the other monounsaturated fatty acids. Of the polyunsaturated fatty acids, the proportion of LA was 22% higher in the vegans than in the meat-eaters. The proportion of the cis-9,trans-11 isomer of CLA in the vegans was less than one-half that in the meat-eaters, whereas a smaller difference was seen for the trans-10,cis-12 isomer of CLA. The mean proportion of dihomo--linolenic acid was low in both the vegetarians and the vegans; however, compared with the meat-eaters, the difference was significant for the vegetarians only. No significant differences between the diet groups were found for the proportions of arachidonic acid. In the vegetarians, the proportions of the fatty acids were generally in between those seen in the meat-eaters and those seen in the vegans.

For each of the long-chain n–3 polyunsaturated fatty acids, the mean proportions in plasma were lower in the vegetarians and lower still in the vegans than in the meat-eaters. EPA was 28% lower in the vegetarians and 53% lower in the vegans, and DHA was 31% and 59% lower in the vegetarians and vegans, respectively, than in the meat-eaters. The differences in DPA were smaller and were significant only between the meat-eaters and the vegans.

The mean proportions of the long-chain polyunsaturated fatty acids in the vegetarians and the vegans, which were divided into groups based on how many years they had followed a vegetarian or vegan diet, are shown in Figure 1. The mean proportions in the meat-eaters are included for comparison. The proportions of plasma EPA, DPA, and DHA were not significantly different in the vegetarians or in the vegans with regard to how long they had followed their diets.

View larger version (20K):
FIGURE 1.. Mean proportions of eicosapentanoic acid (EPA), docosapentaenoic acid (DPA), and docosahexaenoic acid (DHA) in meat-eaters (n = 196), vegetarians (n = 231), and vegans (n = 232) by duration of time since becoming vegetarian or vegan. The figure shows arithmetic means, or the geometric mean (for EPA), and 95% CIs adjusted for age, BMI, smoking, and alcohol intake by diet group (, meat-eaters; , vegetarians; and •, vegans). P values for trend in the mean proportions by duration of adherence to a vegetarian or vegan diet within these subsets were obtained by treating duration of time as a continuous variable in the regression analysis.

 
To investigate the hypothesis that dietary intakes of LA and ALA are related to the endogenous production of the long-chain n–3 polyunsaturated fatty acids, we explored the correlations between the plasma compositions of these fatty acids by diet group (Table 3). The proportion of LA was inversely correlated with EPA in all 3 diet groups, inversely correlated with DPA in the vegans, and inversely correlated with DHA in the vegetarians and in the vegans. Although the correlation between LA and DHA was stronger in the vegans than in the meat-eaters, the correlation coefficients were not significantly different between the diet groups. ALA was positively correlated only with DPA in all 3 diet groups. The ratio of LA:ALA was generally inversely correlated with EPA, DPA, and DHA, most noticeably among the vegans.

View this table:
TABLE 3. Correlations between the proportions of linoleic acid (LA), -linolenic acid (ALA), and the ratio LA:ALA and the long-chain n–3 polyunsaturated fatty acids in plasma by diet group¹

 
When the analysis was restricted to the meat-eaters, no associations were found between the proportions of EPA, DPA, and DHA in plasma and the intake of oily fish that was reported in the FFQ (Table 4); nor were there any associations with the intake of other fish (fried fish, white fish, shellfish, or fish roe) or with the intake of eggs (data not shown). When fish intakes were combined as a total fish intake variable, no associations with the proportions of EPA, DPA, or DHA were seen (data not shown). In the vegetarians, no associations were found between the proportions of EPA, DPA, or DHA and the intake of eggs (data not shown). In the 135 meat-eaters who also completed a 7-d food record, the frequency of oily fish intake reported in the FFQs showed statistically significant agreement with the frequency of oily fish intake reported in the food records (weighted = 0.35). No associations were found between the frequency of oily fish intake reported in the food records and the proportions of EPA, DPA, and DHA in plasma (data not shown).

View this table:
TABLE 4. Proportions of plasma eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA), and docosahexaenoic acid (DHA) in meat-eaters by oily fish intake¹

 

DISCUSSION  
We measured the concentration of fatty acids in plasma in a large sample of British meat-eating, vegetarian, and vegan men. The plasma samples were stored at –70°C for a median of 4 y before the assays were conducted; the percentages of the major fatty acids examined in the present study appeared to be reasonably stable during storage because the percentages were not associated with storage time and because the concentrations of the fatty acids in our study were similar to those observed in freshly collected plasma from the same subjects (23). Overall, the vegetarians and the vegans had substantially lower concentrations of total plasma fatty acids than did the meat-eaters. This is consistent with the lower concentration of total blood cholesterol found in vegetarians and in vegans than in meat-eaters, which has been reported previously (21). Because the concentration of total fatty acids influences the concentrations of individual fatty acids, we compared the fatty acids as proportions of the total fatty acids. The proportions of fatty acids in plasma partly reflect the composition of fatty acids in the diet: the fatty acids found in triacylglycerols reflect the dietary intake of the past few hours, the fatty acids found in phospholipids and cholesterol esters reflect dietary intake of the past few days, and free fatty acids are those released from adipose tissue (24). Because fatty acids are converted and metabolized in the body, the composition of fatty acids in plasma is not simply a marker of dietary intake.

In the present study, large differences in the proportion of fatty acids were found between the dietary groups. In particular, the proportions of plasma myristic acid, palmitic acid, palmitoleic acid, LA, and cis-9,trans-11-CLA were significantly higher in meat-eaters than in vegetarians and vegans. The proportion of myristic acid in serum has been shown to be associated with reported intakes of milk and ruminant fat (25), and it seems likely that the lower proportion of plasma myristic acid reflects the lower intake of dairy fat of the vegetarians, and particularly of the vegans, than of the meat-eaters. Palmitic acid is a major constituent of all animal fats, but it is also abundant in palm oil. Palmitoleic acid is present in animal foods, but it is also derived from the desaturation of palmitic acid. This conversion may be inhibited by high intakes of linoleic acid (26), which may contribute to the low proportion of plasma palmitoleic acid in the vegans compared with the meat-eaters. The intake of CLA derives mainly from dairy and ruminant meat products, and the lower plasma proportions of cis-9,trans-11-CLA (which accounts for >90% of the CLA intake; 27) in the vegetarians and the vegans than in the meat-eaters largely reflect differences in dietary intake. Much smaller differences between the diet groups were seen for the trans-10,cis-12 isomer of CLA, which has gained increasing attention because of its possible body-fat-lowering effects (28) and its possible adverse effects on insulin sensitivity (29) and blood lipids (30). The increased proportions of plasma LA in the vegetarians and vegans are consistent with studies that showed that dietary intake of LA is higher in vegetarians and vegans than in omnivores (7, 8).

Of the long-chain n–3 polyunsaturated fatty acids, the proportions of plasma EPA and DHA were markedly lower in the vegetarians, and lower still in the vegans, than in the meat-eaters. The differences in DPA between the diet groups were much smaller and only significant between the meat-eaters and the vegans. Sanders et al (16) first reported the effects of vegetarian and vegan diets on plasma concentrations of long-chain n–3 fatty acids, and similar findings have also been reported in other studies (8, 17–19). However, few studies have compared fatty acid composition between vegetarians and vegans. In the present study, the proportions of plasma EPA and DHA were noticeably lower in the vegans than in the vegetarians, even though neither vegetarians nor vegans consume fish. Some DHA is present in eggs, and this may account for the higher concentrations of plasma DHA in the vegetarians than in the vegans. Another possibility is that some of the vegetarians in the current study occasionally ate fish but did not report doing so, or that the difference in plasma concentrations of EPA and DPA was related to a lower intake of LA in the vegetarians than in the vegans.

The large number of subjects allowed us to compare subgroups of vegetarians and vegans, and a main objective of the present study was to examine whether the proportions of the long-chain fatty acids vary with the duration of adherence to a vegetarian or a vegan diet. To our knowledge, no previous studies have investigated the differences in plasma fatty acids between long- and short-term vegetarians and vegans. We found no effect of the duration of time since the subjects became vegetarian or vegan, which ranged from <1 y to >20 y, on the proportions of plasma fatty acids. The similar proportions of EPA and DHA in long- and short-term vegans suggest that when foods of animal origin are wholly excluded from the diet, endogenous synthesis of EPA and DHA is sufficient to result in relatively low but stable plasma concentrations of these fatty acids.

Whether the endogenous production of DHA can be modified by altering the intakes of LA or ALA is not clear. Trials have shown that when dietary ALA is increased, plasma EPA and DPA concentrations increase, whereas DHA concentrations either do not change or increase by a small amount (13–15), which indicates that conversion beyond DPA is limited. In the present study, dietary assessments did not allow us to reliably quantify the intake of specific fatty acids. However, the inverse relations between the proportions of LA and DHA in plasma that we observed in the vegetarians and the vegans suggest that the long-term intake of LA may have an influence on the production of DHA. This finding supports the recommendation that vegetarians and vegans should increase their intake of ALA and limit their intake of LA to optimize their n–3 fatty acid status (8). However, our analyses are based on proportions of plasma fatty acids rather than absolute amounts of fatty acids, and the results should therefore be interpreted with caution.

Although clear differences in the proportions of plasma EPA and DHA were seen between the diet groups, which indicates that dietary intake influences plasma concentrations of fatty acids, we did not find any association between these fatty acids and the intake of either oily fish or other fish, which is in contrast with the findings of other studies (31–35). However, fish intake among the meat-eaters in the present study was relatively low, and the number of meat-eaters and the range of fish intakes may not have been sufficient to detect an association.

The importance of the lower plasma concentrations of DHA seen in vegetarians and in vegans than in omnivores is not known, and plasma concentrations do not necessarily reflect tissue concentrations. Low intakes or low plasma concentrations of n–3 fatty acids have been associated with cardiovascular disease, inflammatory diseases, and neurologic disorders such as schizophrenia and Alzheimer disease (9). However, overall mortality in vegetarians is low compared with that in the general population and is similar to that of health-conscious meat-eaters (36). Indeed, mortality from ischemic heart disease has been shown to be lower in vegetarians than in omnivores (37). This may be explained by favorable differences in several risk factors for ischemic heart disease; for example, plasma LDL-cholesterol concentrations were shown to be lower in vegetarians than in omnivores (38, 39). No differences in platelet function have been found between vegans and omnivores (40), but the concentrations of several clotting factors, notably factor VII, have been found to be lower in vegetarians than in omnivores (39). Few data are available on whether inflammatory or neurologic diseases are more common in vegetarians and in vegans. The importance of long-chain n–3 fatty acids in the diet needs further investigation.

In conclusion, in the present study of 659 British men, the proportions of plasma EPA and DHA were markedly lower in vegetarians and particularly in vegans than in meat-eaters. The proportions of these fatty acids were not affected by the duration of time since the subjects became vegetarian or vegan, which suggests that when animal foods are excluded from the diet, the endogenous production of EPA and DHA results in low but stable plasma concentrations of these fatty acids.

ACKNOWLEDGMENTS  
We thank all of the participants in this study and the EPIC-Oxford study staff at the Cancer Research UK Epidemiology Unit.

All of the authors were involved in interpreting the data and contributed to the writing of the manuscript. In addition, MSR performed the statistical analyses and wrote the manuscript; ZL-W performed the plasma fatty acid analyses; PNA helped with the statistical analyses and edited the manuscript; TABS advised on the analyses of the plasma fatty acids; NEA selected the study sample from the EPIC-Oxford cohort; and TJK is the principal investigator of the EPIC-Oxford study. None of the authors had any conflicts of interest.

REFERENCES  

1. Marckmann P, Gronbaek M. Fish consumption and coronary heart disease mortality. A systematic review of prospective cohort studies. Eur J Clin Nutr 1999;53:585–90.
2. Hu FB, Bronner L, Willett WC, et al. Fish and omega-3 fatty acid intake and risk of coronary heart disease in women. JAMA 2002;287:1815–21.
3. Hu FB, Eunyoung C, Rexrode KM, Albert CM, Manson JE. Fish and long-chain omega-3 fatty acid intake and risk of coronary heart disease and total mortality in diabetic women. Circulation 2003;107:1852–7.
4. He K, Rimm EB, Merchant A, et al. Fish consumption and risk of stroke in men. JAMA 2002;288:3130–6.
5. Sinclair AJ, Johnson L, O’Dea K, Holman RT. Diets rich in lean beef increase arachidonic acid and long-chain omega 3 polyunsaturated fatty acid levels in plasma phospholipids. Lipids 1994;29:337–43.
6. Meyer BJ, Mann NJ, Lewis JL, Milligan GC, Sinclair AJ, Howe PR. Dietary intakes and food sources of omega-6 and omega-3 polyunsaturated fatty acids. Lipids 2003;38:391–8.
7. Sanders TAB. Essential fatty acid requirements of vegetarians in pregnancy, lactation, and infancy. Am J Clin Nutr 1999;70(suppl):S555–9.
8. Davis BC, Kris-Etherton PM. Achieving optimal essential fatty acid status in vegetarians: current knowledge and practical implications. Am J Clin Nutr 2003;78(suppl):S640–6.
9. Muskiet FAJ, Fokkema R, Schaafsma A, Boersma ER, Crawford MA. Is docosahexaenoic acid (DHA) essential? Lessons from DHA status regulation, our ancient diet, epidemiology and randomized controlled trials. J Nutr 2004;134:183–6.
10. Burdge GC, Wootton SA. Conversion of alpha-linolenic acid to eicosapentaenoic, docosapentaenoic and docosahexaenoic acids in young women. Br J Nutr 2002;411–20.
11. Burdge GC, Jones AE, Wootton SA. Eicosapentaenoic and docosapentaenoic acids are the principal products of alpha-linolenic acid metabolism in young men. Br J Nutr 2002;88:355–63.
12. Brenna JT. Efficiency of conversion of alpha-linolenic acid to long chain n–3 fatty acids in man. Curr Opin Clin Nutr Metab Care 2002;5:127–32.
13. Li D, Sinclair A, Wilson A, et al. Effect of dietary -linolenic acid on thrombotic risk factors in vegetarian men. Am J Clin Nutr 1999;69:872–82.
14. Sanders TA, Younger KM. The effect of dietary supplements of omega 3 polyunsaturated fatty acids on the fatty acid composition of platelets and plasma choline phosphoglycerides. Br J Nutr 1981;45:613–6.
15. Fokkema MR, Brouwer DAJ, Hasperhoven MB, Martini IA, Muskiet FAJ. Short-term supplementation of low-dose gamma-linolenic acid (GLA), alpha-linolenic acid (ALA), or GLA plus ALA does not augment LCP omega 3 status of Dutch vegans to an appreciable extent. Prostaglandins Leukot Essent Fatty Acids 2004;63:287–92.
16. Sanders TAB, Ellis FR, Dickerson JWT. Studies of vegans: the fatty acid composition of plasma choline phosphoglycerides, erythrocytes, adipose tissue, and breast milk, and some indicators of susceptibility to ischemic heart disease in vegans and omnivore controls. Am J Clin Nutr 1978;31:805–13.
17. Li D, Ball M, Bartlett M, Sinclair A. Lipoprotein(a), essential fatty acid status and lipoprotein lipids in female Australian vegetarians. Clin Sci 1999;97:175–81.
18. Lee HY, Woo J, Chen ZY, Leung SF, Peng XH. Serum fatty acid, lipid profile and dietary intake of Hong Kong Chinese omnivores and vegetarians. Eur J Clin Nutr 2000;54:768–73.
19. Fokkema MR, Brouwer DAJ, Hasperhoven MB, Hettema Y, Bemelmans WJE, Muskiet FAJ. Polyunsaturated fatty acid status of Dutch vegans and omnivores. Prostaglandins Leukot Essent Fatty Acids 2000;63:279–85.
20. Davey GK, Spencer EA, Appleby PN, Allen NE, Knox KH, Key TJ. EPIC-Oxford: lifestyle characteristics and nutrient intakes in a cohort of 33 883 meat-eaters and 31 546 non meat-eaters in the UK. Publ Health Nutr 2003;6:259–69.
21. Allen NE, Appleby PN, Davey GD, Key TJ. Hormones and diet: low insulin-like growth factor-1 but normal bioavailable androgens in vegan men. Br J Cancer 2000;83:95–7.
22. Sanders TA, Oakley FR, Miller GJ, Mitropoulos KA, Crook D, Oliver MF. Influence of n–6 versus n–3 polyunsaturated fatty acids in diets low in saturated fatty acids on plasma lipoproteins and hemostatic factors. Arterioscler Thromb Vasc Biol 1997;17:3449–60.
23. Lloyd-Wright Z. The nutritional status of vegan men in the United Kingdom with particular reference to vitamin B12. PhD dissertation. University of London, London, United Kingdom, 2005.
24. Arab L. Biomarkers of fat and fatty acid intake. J Nutr 2003;133(suppl):S925–32.
25. Wolk A, Furuheim M, Vessby B. Fatty acid composition of adipose tissue and serum lipids are valid biological markers of dairy fat intake in men. J Nutr 2001;131:828–33.
26. Roshanai F, Sanders TA. Influence of different supplements of N–3 polyunsaturated fatty acids on blood and tissue lipids in rats receiving high intakes of linoleic acid. Ann Nutr Metab 1985;29:189–96.
27. Ritzenthaler Kl, McGuire MK, Falen F, Shultz TD, Dasgupta N, McGuire MA. Estimation of conjugated linoleic acid intake by written dietary assessment methodologies underestimates actual intake evaluated by food duplicate methodology. J Nutr 2001;131:1548–54.
28. Terpstra AHM. Effects of conjugated linoleic acid on body composition and plasma lipids in humans: an overview of the literature. Am J Clin Nutr 2004;79:352–61.
29. Riserus U, Vessby B, Arner P, Zethelius B. Supplemention with trans10cis12-conjugated linoleic acid induces hyperproinsulinaemia in obese men: close association with impaired insulin sensitivity. Diabetologia 2004;47:1016–9.
30. Tricon S, Burdge GC, Kew S, et al. Opposing effects of cis-9,trans-11 and trans-10,cis-12 conjugated linoleic acid on blood lipids in healthy humans. Am J Clin Nutr 2004;80:614–20.
31. Amiano P, Dorronsoro M, de Renobales M, et al. Very-long-chain omega-3 fatty acids as markers for habitual fish intake in a population consuming mainly lean fish: the EPIC cohort of Gipuzkoa. Eur J Clin Nutr 2001;55:827–32.
32. Ma J, Folsom AR, Shahar E, Eckfeldt JH. Plasma fatty acid composition as an indicator of habitual dietary fat intake in middle-aged adults. The Atherosclerosis Risk in Communities (ARIC) Study Investigators. Am J Clin Nutr 1995;62:564–71.
33. Andersen LF, Solvoll K, Johansson LR, Salminen, Aron A, Drevon CA. Evaluation of a food frequency questionnaire with weighed records, fatty acids, and alpha-tocopherol in adipose tissue and serum. Am J Epidemiol 1999;150:75–87.
34. Fusconi E, Pala V, Riboli E, et al. Relationship between plasma fatty acid composition and diet over previous years in the Italian centers of the European Prospective Investigation into Cancer and Nutrition (EPIC). Tumori 2003;89:624–35.
35. Hjartaker A, Lund E, Bjerve KS. Serum phospholipid fatty acid composition and habitual intake of marine foods registered by a semi-quantitative food frequency questionnaire. Eur J Clin Nutr 1997;51:736–42.
36. Key TJ, Appleby PN, Davey GD, Allen NE, Spencer EA, Travis RC. Mortality in British vegetarians: review and preliminary results from EPIC-Oxford. Am J Clin Nutr 2003;78(suppl):S533–8.
37. Key TJ, Fraser GE, Thorogood M, et al. Mortality in vegetarians and nonvegetarians: detailed findings from a collaborative analysis of 5 prospective studies. Am J Clin Nutr 1999;70(suppl):S516–24.
38. Beilin LJ, Burke V. Vegetarian diet components, protein and blood pressure: which nutrients are important? Clin Exp Pharmacol Physiol 1995;22:195–8.
39. Thorogood M, Carter R, Benfield L, McPherson K, Mann JI. Plasma lipids and lipoprotein cholesterol concentrations in people with different diets in Britain. Br Med J 1987;295:351–3.
40. Sanders TA, Roshanai F. Platelet phospholipid fatty acid composition and function in vegans compared with age- and sex-matched omnivore controls. Eur J Clin Nutr 1992;46:823–31.