FNR 0958-3157 1360-0478 Biocontrol Science and Technology, Vol. 00, No. 00, 2008, pp. 000–000 sici xxx PUBMED Abbreviation SteenStender

Steen Stender Department of Clinical Biochemistry, Gentofte University Hospital, Niels Andersens Vej 65, opg. 382900HellerupDenmarkPhone: +45-3977-3120+45-3997-8193

ArneAstrup JørnDyerberg Department of Clinical Biochemistry Gentofte University Hospital

Denmark

Department of Human Nutrition Centre for Advanced Food Studies, Faculty of Life Sciences, The University of Copenhagen

FrederiksbergDenmark

Biocontrol Science and Technology Review Article Ruminant and industrially produced trans fatty acids Fatty acids of trans configuration in our food come from two different sources – industrially produced partially hydrogenated fat (IP-TFA) used in frying oils, margarines, spreads, and in bakery products, and ruminant fat in dairy and meat products (RP-TFA). The first source may contain up to 60% of the fatty acids in trans form compared to the content in ruminant fat which generally does not exceed 6%. In Western Europe, including Scandinavia, the average daily intake of IP-TFA has decreased during the recent decade due to societal pressure and a legislative ban, whereas the intake of RP-TFA has remained stable. In spite of this decrease we have found that in many countries consumption >20 g of IP-TFA in a one-meal menu consisting of some popular foods is possible, even though the average intake of IP-TFA in these countries is low. Subgroups of the populations may therefore, on average, consume >5 g IP-TFA per day. This level of consumption is generally not possible for RP-TFA. A daily intake of 5 g TFA (primarily IP-TFA) is associated with a 29% increased risk of coronary heart disease. Such an association is not found for RP-TFA up to a daily intake of 4 g. The high amount of IP-TFA in popular foods, the evidence of a more harmful effect on health by IP-TFA than by RP-TFA, and the feasibility of eliminating IP-TFA from foods without side effects for the population, suggest that a selective elimination of IP-TFA from our food is a ‘low hanging fruit’ in the quest for a more healthy diet for subgroups of the population. trans fatty acids ruminant industrial hydrogenated fats health aspects cardio-vascular disorders obesity Industrially produced trans fatty acids (IP-TFA) are made by partial hydrogenation of vegetable fat, and to a lesser extent, of fish oils by heating to about 400°C under high pressure and with the addition of different catalysts. IP-TFA is used in a variety of food products, especially margarines, spreads, baked goods, and fast food. Ruminant-produced trans fatty acids (RP-TFA) are made by bacterial metabolism of polyunsaturated fatty acids in the rumen of ruminants, such as cow, sheep and goat, and are consequently present in all fats from these animals. The concentration of IP-TFA in partially hydrogenated fat may be as high as 60%, whereas the maximum content of RP-TFA in ruminant fat is about 6%. In milk, RP-TFA is 4–6% of the fat. The average distribution of RP-TFA intake in Denmark is 85% from dairy products and 15% from ruminant meat 1. Consequently, the intake of RP-TFA is mainly determined by the intake of high-fat dairy products.During industrial hydrogenation, trans double bonds are formed along the fatty acid molecules from position 6 and higher. For fatty acids with 18 carbon atoms, a peak concentration of trans double bonds is found in position 9, as elaidic acid, with a Gaussian distribution of fatty acids with the trans bond in the other positions. The bacterial desaturation of polyunsaturated fat from grass and vegetables in the rumen also produces trans double bonds all over the fatty acid molecules, but with a distinct preference for the double bond in position 11 of the 18 carbon fatty acids, as vaccenic acid (Figure 1). Low amounts of trans fatty acids from linoleic and linolenic acids together with trans fatty acids with 16, 20, and 22 carbon atoms, etc., are also formed both industrially and in the rumen. Thus, ruminant fat may contain up to 20% of its trans fatty acids as 16:1, which does not occur in IP-TFA from vegetable oils. This fatty acid together with butyric acid may therefore be used as markers for RP-TFA in a mixture of RP-TFA and IP-TFA of vegetable origin. However, hydrogenated marine oils can contain trans 16:1. Trans 18:1 contributes 80–90% of total trans fatty acids in human food. There is a considerable overlap of trans fatty acids in IP-TFA and RP-TFA 2 (Figure 1).

Isomeric distribution of trans-octadecenoic acids in industrially produced and ruminant trans fat (mean wt% of trans-18:1 fatty acids) adapted from Wolff et al. 2.

Estimates of the intake of trans fatty acids usually rely on estimates of food consumption and the use of appropriate food composition tables. The estimates of food consumption need to be detailed since the content of IP-TFA in apparently similar foods may vary considerably in the same country and from time to time. We found that a serving of fast food in Hungary in 2006 contained 24 g of IP-TFA in one fast-food chain, whereas a similar serving contained <5 g when bought at another fast food chain in the same city. Even when bought at the same chain in the same country, trans fat content may vary in apparently identical servings, for instance in Germany, where we found 5 g in Hamburg, and 0.5 g in Wiesbaden 3. The amount of IP-TFA in the same brand of food may also vary from time to time, depending on market prices for edible fats, and on the willingness of the individual food producer to comply with societal pressure for reducing the content of IP-TFA in the food.Concentrations of RP-TFA in ruminant fat varies with the feed of the animals and therefore with the seasons 1. This variation in gram per serving is, however, much lower than the variation observed for IP-TFA.As a consequence, estimates of trans fatty acid intake should be interpreted with great caution and preferably be supported by analysis of duplicate portions or composite diets, together with measurements of biomarkers for intake, such as trans fatty acid in human tissue, including plasma, erythrocytes, subcutaneous fat, and human milk.The total intake of trans fatty acids in various European countries has been investigated in the ‘Transfair Study’, conducted in 1996 4. The average daily intakes varied from 5.4 g in Iceland to about 1.5 g in Greece and Italy. In Sweden and Denmark, the intake of total TFA was approximately 2.5 g, in Finland about 2.0 g, and in Norway about 4.0 g. In general, the 90th percentiles were 50–70% above the mean values. In the US, the mean intake of trans fatty acids has been estimated to be 5.3 g/day, with the 90th percentile at 9.4 g/day 5.The intake of RP-TFA in Denmark in 1996 has been investigated in detail 1. A total of 3,098 persons were asked to complete a 7-day food diary, and the intake of RP-TFA was calculated for the population, aged 1–80 years, based on food composition tables. The mean intake was 1.2 g/day, the median intake 1.7 g, the 90th percentile 2.9 g, and the 99th percentile 4.4 g.Since 1996, the intake of IP-TFA has decreased considerably in Denmark as well as in other countries in Western Europe, including Scandinavia, due to societal pressure on food producers to reduce the content of IP-TFA (Figure 2A). The steep decline in Denmark between 1990 and 1996 may be ascribed to the effect of the first report by The Danish Nutrition Council in 1994 on trans fat and human health 6. Producers of table margarine were especially responsive to pressure and removed nearly all IP-TFA from table margarines 7. In Denmark, the amount of milk fat and ruminant meat available for consumption has been relatively constant from 1970 to 2001 8. Therefore, it seems likely that the intake of RP-TFA in 2007 is similar to the intake in 1996 (Figure 2A). This is probably also the case for the intake of RP-TFA in the other Scandinavian countries.

(A) The average intake of industrially produced and ruminant trans fatty acids in Denmark from 1978 to 2005. The intake of ruminant fat was measured in 1996 1. Some studies suggest that this intake has been constant from 1972 to 2005. (B) Mortality curves for coronary heart disease in Denmark during the period with decreasing intake of industrially produced trans fatty acids, but constant intake of ruminant trans fatty acids.

Although the mean intake of IP-TFA in Denmark in 1996 was only about 1 g/day, a wide distribution of IP-TFA in some popular foods made it possible for some people to consume much more. A high intake of IP-TFA in a minority of the population concerned the Danish Nutrition Council who conducted a survey in 2001 and found that a number of popular food items contained fat with up to 60% IP-TFA 9. The popular food items consisted of fast food meals (fried meat and French fries), microwave oven popcorn, and a variety of biscuits and wafers. In Denmark in 2001 it was thus possible to have a high trans fat menu with about 30 g of IP-TFA, and about 1% of the 5.6 million Danes were estimated to have a daily intake of >5 g/day. This situation prompted the Danish parliament to pass legislation in 2003 making it illegal to use fat with >2% IP-TFA, whether the food was produced in Denmark or imported 1011. After a few years, the intake of IP-TFA was virtually eliminated in Denmark, without any nutritional or economic side effects. Initially, the legislation was challenged by the EU Commission, but was endorsed by the Commission in 2007 due to ‘increased scientific evidences on the dangers of trans fat’. Similar legislation has now been introduced for food sold in the 20,000 restaurants in New York City.In a series of countries, from 2005 to 2007 we investigated the extent to which it was possible to obtain a high intake of IP-TFA 12. The results are shown in Figure 3. It is noteworthy that a menu with a high IP-TFA content could not be obtained in Denmark in 2005. In contrast, the existence of popular foods with high amounts of IP-TFA in the various other countries suggests that certain groups within these populations daily consume 5 g of IP-TFA or more. These groups might represent <1% of the total population, however in the EU with a population of 500 million, they may still constitute 5 million people. The majority of people in the EU exposed to a high intake of IP-TFA live in Eastern Europe.

Grams of industrially produced trans fatty acids in a ‘high trans fat menu’. A large serving of fast food consists of 160 g of nuggets and 171 g of French fries. In each country, the products with the highest amount of trans fatty acids per 100 g product in each category are shown. Values in brackets are percent trans fatty acids of total fat in French fries, biscuits/cakes/wafers and microwave oven popcorn, respectively. The foods were bought between 2005 and 2008.

The possibility of consuming higher amounts of IP-TFA compared to RP-TFA is exemplified in Figure 4 by the differing amounts of trans fatty acids in ruminant meat compared to a similar sized portion of a deep fat fried chicken meat product from the same fast food chain, fried in oils with a high IP-TFA content 3.

Trans fatty acids in 160 g meat from two different sources. Values in brackets are percent trans fatty acids of total fat.

The results from observational studies concerning the intake of TFA and coronary heart disease (CHD) demonstrate that the intake of RP-TFA has either not been associated with or has been negatively associated with the risk of CHD (Table 1). By contrast, the intake of IP-TFA has consistently shown a positive association with the risk of CHD 19. TFA intake (g/day) Relative risk of CHD Study name (reference) No. of subjects No. of cases Definition of end points Industrial Ruminant Industrial Ruminant Adapted from Weggemans et al. 19. MI = myocardial infarction, CHD = coronary heart disease, W = women. M = men. *p for trend. Prospective cohort studies Nurses’ Health Study 13 85,095 W 431 Non-fatal MI or coronary death 1.44–3.42 0.96–2.28 1.78 (p=0.009)* 0.59 (p=0.23)* ATBC Study 14 21,930 M 635 Coronary death 0.1–5.1 0.6–2.5 1.23 (p=0.004)* 0.83 (p=0.0035)* Zutphen Cohort Study 15 667 M 98 Fatal and non-fatal MI 5.0 1.7 1.05 1.17 Danish ruminant trans fatty acid study 16 1,866 M 253 Fatal and non-fatal MI – 0.6–4.1 – 1.00 (0.91–1.07) 1,911 W 121 Fatal and non-fatal MI – 0.5–3.1 – 0.84 (0.70–1.01) Case-control studies Ascherio et al. 17 407 M; 114 W 239 MI 0.84–5.04 0.45–1.79 1.94 (p=0.0001)* 1.02 (p=0.57)* Table 1 also shows that the average daily intake of IP-TFA in these studies is considerably larger than that from ruminant sources. In the Danish study, the average daily intake of RP-TFA is higher than the 0.5–2.5 g in the other studies in Table 1. The Danish study did not find any positive association between intakes of RP-TFA and the risk of CHD, when the risk was adjusted for the intake of saturated fat. The intake of IP-TFA as a dietary risk factor for CHD was adjusted for by taking into account the weighted intake of foods containing high amounts of IP-TFA. The RP-TFA intake of up to 4 g/day had a slightly negative association with CHD in females, whereas there was no association in males (Figure 5). These results are consistent with the findings on RP-TFA intake in the Nurses Health Study and the Finnish ATBC study 1314; the intake of RP-TFA approaches the intake of IP-TFA in these studies (Table 1). Taken together, the findings support the hypothesis that RP-TFA intake does not increase the risk for CHD, and that IP-TFA is more harmful than RP-TFA when compared on a gram-to-gram basis. Therefore, the findings contradict the suggestion that, at equal amounts of intake, IP-TFA and RP-TFA are associated with similar risks for CHD 19.

Spline regression analysis of the log hazard ratios of coronary heart disease according to intake of ruminant trans fatty acid (RP-TFA) in an 18-year prospective study ending in 2000 of 3,686 Danish women and men aged 30–71 years. Adapted from Jakobsen et al. 16.

In a prospective, nested, case control study of trans fatty acids in erythrocytes, an objective marker of dietary intake of trans fatty acids, it was observed that higher total trans fatty acid content was associated with an elevated risk of CHD 20. For 18:1 trans isomers, CHD cases had significantly higher contents (p<0.001 and <0.01, respectively) of 18:1 n-12 and 18:1 n-9 (elaidic acid) than controls. For vaccenic acid (18:1 n-7), the difference did not reach statistical significance (p=0.05). Death due to CHD in Denmark (and in other countries in the western hemisphere) has declined since 1980. This reduction coincides with the reduction in IP-TFA intake during a period of unchanged RP-TFA consumption (Figure 2A,B). The reduction in CHD has been explained by a favorable change in a number of traditional risk factors, such as smoking, hypercholesterolemia and hypertension. On the other hand, physical inactivity and obesity rates have increased. We hypothesize that the decline in the intake of IP-TFA also plays a beneficial role in the reduction.Iceland and Norway have previously had a high intake of IP-TFA compared with the other Nordic countries, but did not have a correspondingly higher incidence of CHD. In both Iceland and Norway, the intake of omega-3 fatty acids from fish and cod liver oil has been considerably higher than in the other Nordic countries. This intake of fish oil may have counteracted the harmful effect of greater IP-TFA intake, according to the Yin-Yang hypothesis for the effect on human health of these two types of fatty acids 22 Observational studies with RP-TFA are few compared with the same type of studies with IP-TFA. Furthermore, observational studies on the association between health and RP-TFA intake are difficult to interpret because one must adjust first, for the effect of the much larger amounts of saturated fatty acids in dairy products and meat, which varies in proportion with the content of RP-TFA, and second, for the effects of fatty acids from other sources, such as omega-3 fatty acids and IP-TFA. Adjusting for the intake of the latter would not be necessary in Denmark, where the intake of IP-TFA has virtually been eliminated. Since Denmark has a wide distribution of intake of RP-TFA it may be possible to shed more light on the RP-TFA/CHD association by conducting an observational study on the Danish population, commencing with intake data after 2003, when enforcement of the legislative ban on IP-TFA went into effect. In many intervention studies, IP-TFA intake has been shown to result in lower HDL and higher LDL concentrations in plasma compared with intake of saturated fat. Whether this is the case at low intakes of IP-TFA has been challenged. Most studies suggest a linearity of this effect on plasma lipids, meaning that a gram-for-gram replacement of IP-TFA with saturated fat results in a more beneficial LDL/HDL ratio 21.The unfavorable change in blood lipids accounts, however, for only a fraction of the harmful effect of trans fatty acids found in observational studies 18. The remaining effect might be due to the effects of trans fatty acids beyond those exerted on plasma lipids. It has been suggested that pro-inflammatory and pro-arrhythmogenic actions occur 23. Except for the impact of IP-TFA on plasma lipoproteins, the effect on other risk markers of CHD has not been reproducibly demonstrated in intervention studies 24.In a study with male hamsters fed diets enriched with fatty acids for 4 weeks, it was observed that when elaidic and vaccenic acid were compared, the ratio of LDL/HDL-cholesterol in plasma was significantly higher in hamsters fed vaccenic acid 25.To date, the effect of RP-TFA on plasma lipid and lipoprotein risk-markers of CHD has been investigated in only one human, randomized, cross-over study in 22 males and 24 females, fed either 11–12 g/day of RP-TFA or IP-TFA for 3 weeks, with a 1-week wash-out period in between. There was apparently no difference in the effect of RP-TFA on LDL or HDL concentration compared with the effect of IP-TFA in males, whereas both LDL and HDL were significantly higher after RP-TFA in females 26.Unlike elaidic acid from IP-TFA, vaccenic acid from either IP-TFA or RP-TFA can be converted to rumenic acid, most notably by ruminant animals, but also in non-ruminant animals as well as in humans. Rumenic acid is a so-called conjugated linoleic acid (CLA) which may have positive metabolic effects, although the results in humans have been contradictory. It is noteworthy that a high intake of IP-TFA may provide more vaccenic acid than an average intake of RP-TFA. If vaccenic acid has a beneficial effect on coronary risk, this must be more than counteracted by the harmful effect of the other IP-TFA to explain the negative association between the intake of IP-TFA and coronary risk. There are biochemical reasons to believe that IP-TFA and RP-TFA compete with long chain fatty acids during their incorporation in the developing brain. The observational evidence for a harmful effect of IP-TFA on the development of the fetus and newborn is weak. The possibility of avoiding the introduction of a new type of fatty acid in the brain early in life, however, seems preferable. It should be noted, though, that a high intake of RP-TFA by the newborn in infant formulas based on cow milk may also be considered as a new type of fatty acid in the diet compared with the fatty acids from mother's milk. Due to ethical concerns, we will probably never see randomized studies in humans with hard end points, such as coronary events or death, concerning the health effects of IP-TFA and RP-TFA.In a study of 32 male green monkeys given a diet with a high content of IP-TFA fat compared with a diet in which IP-TFA was replaced by oleic acid for 6 years, the IP-TFA intake promoted abdominal fat deposition and increased postprandial insulin levels in plasma 27. This study thus suggests that IP-TFA may interfere with biological processes in the body other than those directly linked to the development of CHD, even if caution should be exerted in extrapolation from monkeys to man. However, the observational epidemiology also supports the association between IP-TFA intake and abdominal obesity 28.If an obesity promoting effect of IP-TFA is making a contribution to the current obesity epidemic, one would assume that obesity prevalence in Denmark might be related to the rapid decline and elimination of IP-TFA over the last 20 years. This is indeed the case. Results from the Danish Army draft boards, which is mandatory and where height and weight are measured, clearly demonstrate that the increasing obesity rates seem to have leveled off (Figure 6).

Prevalence of obesity among young Danish males attending mandatory draft boards (% with BMI >20 kg/m²). Data up to 2000 were modified from Sorensen and Price 29 and Sørensen et al. 30, data after 2000 were kindly provided by the Danish Ministry of Interior and Health.

Intake of IP-TFA is associated with the occurrence of CHD, and there are some plausible mechanisms to explain at least some of this association.For RP-TFA intake, no association with increased CHD has been found. By contrast, even negative associations have been seen in some studies after adjustment for the simultaneous intake of saturated fat.For both IP-TFA and RP-TFA there is a lack of randomized intervention studies with hard end points. Such studies could be helpful in establishing the extent to which intake of these fatty acids affect human health. However, intervention studies will probably never be carried out because it is unethical to verify suspicions of harmful effects when the suspected exposure can be removed without negative effects on health.In the opinion of the authors, the strong suspicion of a harmful effect of IP-TFA intake, even more harmful than that of saturated fats, and the possibility of removing IP-TFA from food, places the burden of proving that IP-TFA in human foods is safe on those who would like to add IP-TFA to our food supply.On the other hand, the lack of evidence for a harmful effect of RP-TFA intake, and the fact that these fatty acids cannot be removed from ruminant fat and have been part of human food in millennia, places the burden of proving that these fats are more harmful than saturated fats on those who would like to reduce the intake of RP-TFA beyond levels produced by a reduction in saturated fats generally. The intake of ruminant trans fatty acids is maximally 4–5 g/day, whereas the intake of industrially produced trans fat in some population groups far exceeds 5 g/day. Since there is evidence for harmful effects of industrially produced trans fatty acids at this level, but not for ruminant trans fatty acids, and since it is feasible to eliminate industrially produced trans fatty acids from the food supply without side effects in the population, this elimination can be considered a ‘low hanging fruit’ in the quest for a more healthy diet for the entire population in any given country. How to do it is a political issue. The paper is based on a presentation at a conference on health aspects of milk fat, 30–31 August 2007, Stockholm, Sweden, arranged by the Swedish Dairy Association, who also sponsored the fee due to the publication of the article. Arne Astrup is a member of Global Dairy Platform, Chicago and member of Arla Amba Nutrition Advisory Board. JakobsenMUBystedAAndersenNLHeitmannBLHartkoppHBLethT Intake of ruminant trans fatty acids in the Danish population aged 1–80 years Eur J Clin Nutr 2006603 3128 WolffRLPrechtDMolkentinJ. Occurrence and distribution profiles of trans-18:1 acids in edible fats of natural origin Trans fatty acids in human nutrition SebedioJL ChristieWW The Oily PressDundeeUK1998 133 StenderSDyerbergJAstrupA. High levels of industrially produced trans fat in popular fast foods N Engl J Med 2006354 16502 HulshofKFvan Erp-BaartMAAnttolainenMBeckerWChruchMCouetC Intake of fatty acids in western Europe with emphasis on trans fatty acids: the TRANSFAIR Study Eur J Clin Nutr 1999532 14357 AllisonDBEganSKBarrajLMCaughmanCInfanteMHeimbachJT. Estimated intakes of trans fatty and other fatty acids in the US population J Am Diet Assoc 1999992 16674 Stender S, Dyerberg J, Hølmer G, Ovesen L, Sandstrøm B. Transfedtsyrers betydning for sundheden. A report from the Danish Nutrition Council, Publication No 2. Copenhagen; 1994. LethT.BystedAHansenKOvesenL. Trans fat content in Danish margarines and shortenings marketed in Denmark JAOCS 199875 107983 Fagt S , Trolle E . Forsyning af fødevarer 1955–1999. Udvikling i danskernes kost-forbrug, indkøb og vaner. National Food Agency of Denmark, Søborg (in Danish); 2001. Stender S , Dyerberg J . Transfedtsyrers betydning for sundheden. Opdatering år 2001 A report from the Danish Nutrition Council, Publication No 23. Copenhagen; 2001. Stender S , Dyerberg J . The influence of trans fatty acids on health (4th edn.) A report from the Danish Nutrition Council. Publication No 34, Copenhagen; 2003. StenderSDyerbergJAstrupA. Consumer protection through a legislative ban on industrially produced trans fatty acids in foods in Denmark Scand J Food Nutr 2006504 15560 StenderSDyerbergJBystedALethTAstrupA. A trans world journey Atheroscler Suppl 200672 4752 OhKHuFBMansonJEStampferMJWillettWC. Dietary fat intake and risk of coronary heart disease in women: 20 years of follow-up of the nurses’ health study Am J Epidemiol 20051617 6729 PietinenPAscherioAKorhonenPHartmanAMWillettWCAlbanesD Intake of fatty acids and risk of coronary heart disease in a cohort of Finnish men. The Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study Am J Epidemiol 199714510 87687 OomenCMOckéMCFeskensEJvan Erp-BaartMAKokFJKromhoutD. Association between trans fatty acid intake and 10-year risk of coronary heart disease in the Zutphen Elderly Study: a prospective population-based study Lancet 20013579258 74651 JakobsenMUOvervadKDyerbergJHeitmannBL. Intake of ruminant trans fatty acids and risk of coronary heart disease Int J Epidemiol 2008371 17382 AscherioARimmEBGiovannucciELSpiegelmanDStampferMWillettWC. Dietary fat and risk of coronary heart disease in men: cohort follow up study in the United States BMJ 19963137049 8490 MozaffarianDKatanMBAscherioAStampferMJWillettWC. Trans fatty acids and cardiovascular disease N Engl J Med 200635415 160113 WeggemansRMRudrumMTrautweinEA. Intake of ruminant versus industrial trans fatty acids and risk of coronary heart disease – what is the evidence? Eur J Lipid Sci Technol 2004106 3907 SunQMaJCamposHHankinsonSEMansonJEStampferMJ A prospective study of trans fatty acids in erythrocytes and risk of coronary heart disease Circulation 2007115 185865 AscherioAKatanMBZockPLStampferMJWillettWC. Trans fatty acids and coronary heart disease N Engl J Med 199934025 19948 DyerbergJChristensenJHEskesenDAstrupAStenderS. Trans, and n-3 polyunsaturated fatty acids and vascular function – a yin yang situation Atheroscler Suppl 200672 3335 MozaffarianD. Trans fatty acids – effects on systemic inflammation and endothelial function Atheroscler Suppl 200672 2932 DyerbergJEskesenDCAndersenPWAstrupABuemannBChristensenJH Effects of trans and n-3 unsaturated fatty acids on cardiovascular risk markers in healthy males. An 8 weeks dietary intervention study Eur J Clin Nutr 200458 106270 MeijerGWvan TolAvan BerkelTJCWeststrateJA. Effect of dietary elaidic versus vaccenic acid on blood and liver lipids in the hamster Atherosclerosis 2001157 3140 ChardignyJMMalpuech-BrugèreCDionisiFBaumanDEGermanBMensinkRP Rationale and design of the TRANSFACT project phase I: a study to assess the effect of the two different dietary sources of trans fatty acids on cardiovascular risk factors in humans Contemp Clin Trials 2006274 36473 KavanaghKJonesKLSawyerJKelleyKCarrJJWagnerJD Trans fat diet induces abdominal obesity and changes in insulin sensitivity in monkeys Obesity 200715 167584 StenderSDyerbergJAstrupA. Fast food: unfriendly and unhealthy Int J Obes 2007316 88790 SorensenTIPriceRA. Secular trends in body mass index among Danish young men Int J Obes 1990145 4119 SørensenHTSabroeSGillmanMRothmanKJMadsenKMFischerP Continued increase in prevalence of obesity in Danish young men Int J Obes Relat Metab Disord 1997218 7124