The present study shows that a daily intake of a moderate portion of chicken meat for 4 weeks can appreciably increase the concentration of EPA in serum phospholipids of young healthy humans, provided that the chickens had been fed rapeseed and linseed oil instead of similar amounts of soybean oil.
The daily intake of chicken meat, about 160 g/day, is much higher than the average daily intake of chicken meat in the Norwegian population; being about 50 g/day [2, 3]. The total average intake of meat per person in Norway is estimated in two different reports to be about 130 g/day , and 200 g /day . The subjects were advised to follow their normal diet, but to eat as much as possible of the two chickens they received each week, in preference to other meats. Their average total meat intake may therefore have been somewhat higher than 160 g/day during the study. In the present study the participants were mostly students at The Norwegian University of Life Sciences, living in student accommodation houses and having a limited budget. Since meat is expensive compared to cereal based food, they are likely to have less meat in their regular diet than the average Norwegian intake. Some of the students at The Norwegian University of Life Sciences are taking a course in nutrition where they undergo a dietary assessment showing that they have a diet based on much bread and cereals, milk and milk products, some meat, some fish, margarine, vegetables and fruit.
The RLO feed resulted in a significant increase in EPA, DPA and DHA and a decrease in AA in the chicken breast fillets (Table 4). The chicken is thus a good producer of LC n-3 PUFA from ALA, and chicken meat has potential to be a good source of LC n-3 PUFA in the human diet. The concentration of LC PUFA (AA + EPA + DPA + DHA) made from LA and ALA was 112 mg/100 g in the SO chicken breast muscle and 124 mg/100 g in the RLO group. The percentages of LA + ALA in the two feeds were 40% and 34% of the total fatty acids, respectively, indicating that the synthesis of LC PUFA from LA and ALA was higher in breast muscle from the RLO group compared to the SO group. Thus a diet containing rapeseed and linseed oil appears to trigger the chicken to synthesize LC PUFA. This has also been indicated by others in chickens , pigs  and bulls .
The ability of the chickens and other domestic animals to produce EPA, DPA and DHA from ALA should be valued and given more focus seen in light of the limitations in the world supply of LC n-3 PUFAs from fish and marine sources. To replace the soybean oil (that is now the commonly used feed oil) with linseed and rapeseed oil seems to be an efficient way to increase the intake of LC n-3 PUFA for humans without having to change dietary habits or to take fish oil supplement pills.
As seen from Table 4, the daily intake of EPA + DPA + DHA when eating a portion of 175 g of breast muscle from the RLO chicken would be 142 mg. This is 57% of the proposed EFSA reference intake value of LC n-3 PUFA (250 mg/day) to reduce the risk of CVD . Chicken breast meat from the SO fed group contained 68 mg in 175 g breast meat, thus a portion of the RLO chicken breast meat contained 74 mg more LC n-3 PUFA than the traditional SO chicken. The optimal dose for LC n-3 PUFA remains to be established. The EFSA Panel in 2010  has suggested that 450 mg may be a recommended daily intake of LC n-3 PUFA. This shows that even if all meat consumed had about the same fatty acid composition as the RLO breast meat from this experiment, it would not be enough alone to cover the recommended intake of LC n-3 PUFA.
The chicken thigh meat may be about four to five times higher in fat content compared to the breast meat [4, 23], but the percentage of LC n-3 PUFA (g/100 g fatty acids) is lower in thigh meat compared to breast meat . Thus, the LC n-3 PUFA content is somewhat (about 30–50%) higher in thigh muscle compared to breast muscle , and by consuming 175 g of the RLO thigh muscle the LC n-3 PUFA intake can be estimated to be about 190 mg instead of 142 mg when consuming the breast muscle.
The concentration of AA was lower in the RLO chicken breast muscle compared to SO. This is in accordance to findings by Poureslami et al. and is shown in both breast muscle and in thigh muscle . The reduction in AA concentration in meat may have implications for the consumer given the nature of the competition between AA and EPA for binding to enzymes and cellular structures . It has been found that purified COX-1 oxygenates EPA at a rate which is only 10% of the rate for AA, while EPA significantly inhibits AA oxygenation by COX-1 . A portion of breast meat from the RLO chicken contained 75 mg AA, while the SO meat contained 128 mg, and the ratio of AA/EPA was only 2 in the RLO chicken breast meat compared to nearly 17 in the meat in the SO group. Such a big difference in AA and EPA balance could be expected to have an impact on the prostanoid synthesis both for the chicken itself and for the consumer eating the chicken.
The human intervention study
In the present study, an increase in EPA and ALA concentrations, and a decrease in the ratio AA/EPA in serum phospholipids were shown in the persons consuming the RLO chickens. This is in line with the study of Weill et al. and McAfee et al. showing that subjects consuming meat from animals offered a concentrate feed supplemented with linseed oil or a grass based diet had higher LC n-3 PUFA concentrations in erythrocytes, platelets and plasma compared to subjects consuming animal products from animals fed a standard diet [25, 26].
The fatty acid composition of serum phospholipids has become established as a valid marker for assessing the status of various fatty acids and to predict dietary fat intakes . As reviewed by Fekete et al. , four weeks intervention time and sampling of serum phospholipid fatty acids was a suitable method for studying long-term LC n-3 status in humans.
The content of EPA in 160 g breast muscle from the RLO chicken was 34 mg. In contrast, the content of EPA in the SO chicken breast muscle was only 7 mg per day. Since the participants were not eating oily fish during the study, most of their dietary EPA intake originated from the chicken meat. EPA is synthesized in the body from ALA, but there are variations in the ability to convert ALA to EPA [29–31], and it may be speculated that to some persons a dietary intake of EPA is imperative. The concentration of EPA in serum phospholipids was 1.4 mg/100 ml serum (about 1.3% of FAME) in the RLO treatment group and 0.9 mg/100 ml (about 0.8% of FAME) in the SO group. The EPA concentrations varied between the persons, and at post intervention time three persons in the SO diet group had levels lower than 0.5 mg EPA in phospholipids/100 ml serum (0.5% FAME), but in the RLO group no persons had lower levels than 0.5 mg EPA in phospholipids/100 ml serum post trial.
The sum of EPA plus DHA in serum phospholipids have in populations studies been linked to assess risk of heart disease [32, 33]. EPA + DHA levels amounting to more than 4.6% of total fatty acids in serum phospholipids have been associated with a 70% lower risk compared to those with a lower level of these fatty acids [32, 33]. In the present study, at baseline, three test subjects in the SO group had less than 4.6% of EPA + DHA (percent of total fatty acids) in serum phospholipids, and five subjects in the RLO were below 4.6%. After the intervention period, two of the three persons in the SO group had reduced their EPA + DHA sum, while all five of the persons in the RLO group improved (increased) their sum of percent EPA + DHA in serum phospholipids. Thus, in the present study, nearly 1/5 of the subjects had less than 4.6% of EPA + DHA in their serum phospholipids. Consumption of the RLO chicken gave an increase in EPA + DHA in serum phospholipids of persons already low in EPA + DHA, and this might theoretically contribute to reduce the risk of coronary heart disease [32, 33].
The DPA concentration in serum phospholipids was at about the same level as EPA, but the DHA concentration in serum phospholipids was about five times higher in both treatment groups. In the chicken meat however, it is different; DPA was the most abundant of these three fatty acids, and especially in the RLO chicken there was much DPA (39 mg/100 g meat, Table 4). DPA can be converted to both EPA and DHA . The DPA percentage in serum phospholipids of the subjects eating the RLO chicken showed a tendency to be higher compared to those eating the SO chicken, and to have a significantly lower DPA/EPA ratio (Table 5). No increase in the serum phospholipid concentration of DHA in the test subjects eating RLO chicken meat was observed, although EPA was enhanced. The reasons for this are unknown. One possible explanation might be faster removal of DHA than of EPA from the blood plasma of our test subjects.
The concentration of ALA in serum phospholipids was about 30% higher in the RLO group than the SO group. This is plausible since the amount of ALA in the RLO breast muscle was higher (three times higher) than the SO breast meat.
The ratio AA/EPA was significantly lower in serum phospholipids from the persons eating the RLO chicken. This ratio has been shown to affect the production of different types of eicosanoids and prostanoids , and the production of eicosanoids and prostanoids may be altered in a favorable direction towards lower production of thromboxanes of the 2-series which should imply reduced risk of thrombosis.
There were no differences in the concentration of AA in serum phospholipids between the subjects consuming RLO and SO chickens. In the SO group, the AA concentration was higher at post-intervention compared to baseline being 10.2% and 8.4%, (Table 5). It may be speculated that high intake of AA rich meat during the intervention period may increase AA levels in serum phospholipids. However, Kawabata et al. showed no correlations between dietary AA intake and AA in blood lipid fractions .
The difference in numbers of men and women in the two intervention groups was of concern, since women have been reported to have a more efficient synthesis of EPA, DPA and DHA from ALA . When calculating the results for men and women separately, there was, however, no difference between the sexes in this study, and the final results would not be different if we excluded the men from the study.
There were no significant differences in serum pre or post trial concentrations of total cholesterol, LDL cholesterol, HDL cholesterol or triacylglycerol in the two intervention groups. This is in accordance to the study of McAfee et al. , where the participants were consuming red meat with different amounts of LC PUFAs. Even if the intake of LC n-3 PUFA is nearly twice as high in the RLO group compared to the SO group, the intakes may be too low to significantly affect serum cholesterol or triacylglycerol [38, 39].
In the present study with young subjects having normal blood pressure there were no effects on blood pressure when eating the two different meats. Long chain n-3 PUFAs have been shown to have mild antihypertensive effect [11, 40], however, the difference in LC n-3 PUFA intake between the two groups may have been too low to reveal any effects. There were no differences in CRP between the two groups. Although research studies have suggested that LC n-3 PUFA may have anti-inflammatory effects [12, 39], this was not observed in the present study with young healthy persons.