Trans fatty acids

Saturated fatty acids have a chain of carbon atoms joined by single bonds, allowing for rotation about the bonds.  In unsaturated fatty acids the double bond (or bonds) restricts rotation. Therefore an unsaturated acid with one double bond can exist in two forms.  The cis or z form (z for zusammen ­ German for together) has the two parts of the carbon chain bent towards each other and the trans or e form (e from entgegen, German for opposed) has the two parts of the chain almost linear.  The trans fatty acids are therefore rather similar in conformation and behaviour to the saturated acids.

The structure of saturated and unsaturated chemical bonds is represented in the following diagram. (Source: US FDA)

Hulshof (1999) reported the results of the TRANSFAIR co-operative study in 14 Western European countries of the dietary intake of trans and other fatty acids.  For the UK the consumption estimate was based on a seven-day survey of 8000 households carried out in 1994 (MAFF, 1995).  99 representative foods were purchased in the period June 1995 to April 1996, representing 95 % of total mean fat intake, and analysed.  The mean total fat intake was 77g/day, equivalent to 35.7% of energy intake.  Mean TFA consumption was 2.8g/day (= 1.3% of energy) and 28.5g/day (= 13.2% of energy) was as saturated fatty acids.

A review by The Scientific Panel on Dietetic Products, Nutrition and Allergies (NDA) of EFSA (2004) found that the intake of TFA varies between countries, with lowest intakes found in the Mediterranean countries.  The contribution of TFAs to daily energy intake (based on estimates for 1995/1996) is approximately 0.5−2.0% in comparison with that of saturated fats which ranges from 10.5 to 18%.  Recent dietary surveys indicate that TFA intakes have decreased in a number of EU countries, mainly due to the reformulation of food products (e.g. fat spreads) to reduce the TFA content.  The panel concluded that at equivalent dietary levels, the effect of TFA on heart health may be greater than that of saturated fatty acids.  However, current intakes of TFA are generally more than 10-fold lower than those of saturated fatty acids whose intakes in many European countries exceed dietary recommendations.  The NDA panel also evaluated other health effects and concluded that scientific evidence with regards to a possible relationship of TFA intake with cancer, type 2 diabetes or allergies is weak or inconsistent.

More recent dietary surveys have taken place (including a recent study in the UK Which? Magazine, October 2004) in which 30 groups of foods were analysed but the TFA contents were expressed as “g per portion” without any indication of the varying portion sizes, and with no indication of the accompanying SFA.

In the UK, the population on average eats about 50% of the recommended limit for trans ats(2).

In most respects the digestion, absorption and metabolism of TFA are the same as that of cis isomers.  They are incorporated into lipids in the tissues, are present in human milk and are catabolised in the same way as the cis isomers.  Selective accumulation in tissues does not occur.  TFA are oxidised to provide energy.  Although there is some evidence from in vitro and animal studies that conversion of essential fatty acids is inhibited by TFA, metabolism of essential fatty acids is unlikely to be impaired by TFA when intakes of essential fatty acids meet recommended levels.

Correlations have been unscientifically drawn between the increased use of hydrogenation and the increase in coronary heart disease (CHD) and other health problems.  Such correlations have been criticised (British Nutrition Foundation 1995).  Correlation does not demonstrate causation.  However, concern has arisen from a number of investigations:

• Adipose tissue samples of those who have died from coronary heart disease (CHD) have been found to have a higher concentration of TFA than average (Thomas et al 1981).
• Rats fed partially hydrogenated fish oils showed a proliferation of peroxisomes in the liver (these are subcellular organelles which provide additional oxidation capacity and are indicative of increased free radical formation).  However, since the same effect was observed after feeding a longer chain length vegetable oil that did not contain any TFA, it has been concluded that the effect was due to the longer chain length of the fatty acid of fish oils.
• Essential fatty acids (EFA) are transformed in the body by a series of reactions into long chain polyunsaturated fatty acids essential for development of the nervous system and eyesight.  TFA compete with EFA for the enzyme systems involved in these reactions. High intakes of TFA have been shown to influence the metabolism of EFA in experimental animals when the EFA intake was low.  Frank deficiency in EFA is only found in abnormal circumstances in human adults.  However, new-born infants, and especially if premature, show borderline deficiency in EFA, and their TFA intake from the mothers milk is related to her TFA intake.  This consideration led the Danish Nutrition Council to recommend the reduction of intakes of TFA from vegetable fats to an average of 2g/day.
• Suggestions that ingestion of TFA is implicated in coronary heart disease (CHD) are based on changes induced in plasma cholesterol levels.  Within the range of intakes of 3 - 11 % of dietary energy there is a dose-response relationship; an increase of 1 % of the total energy intake (at the expense of oleic acid) increased low density lipoproteins (LDL) by 0.04 mmol/l and decreased high density lipoproteins (HDL) by 0.013 mmol/l.  This amounts to a 1 % reduction in HDL and a 1 % increase in LDL.  TFA increase LDL to the same extent as SFA, but reduce the beneficial HDL.  One study (Thomas et al, 1981) claimed that the level of TFA in adipose tissue was associated with the evidence of CHD. However, the methodology in this work has been criticised (Sanders, 1988), and Kritchevsky (1982) concluded that TFA in hydrogenated vegetable fat when fed in amounts up to 14% of energy intake were not atherogenic; nor were partially hydrogenated fish oils (Sanders, 1988).
• TFA can cause an increase in plasma lipoprotein (a) concentration (especially in individuals with a high starting level) which is considered by some workers in the field to be an independent risk factor for CHD.  Hu et al (1997) made an epidemiological assessment drawn from data in the US-based Nurses' Health Study.  The authors prospectively studied 80,082 women who were 34 to 59 years of age and had no known coronary disease, stroke, cancer, hypercholesterolemia, or diabetes in 1980.  Information on diet was obtained at base line and updated during follow-up by means of validated questionnaires.  During 14 years of follow-up, they documented 939 cases of nonfatal myocardial infarction or death from coronary heart disease.  Multivariate analyses included age, smoking status, total energy intake, dietary cholesterol intake, percentages of energy obtained from protein and specific types of fat, and other risk factors.  Total fat intake was not significantly related to the risk of coronary disease.  However they estimated that the replacement of 5 percent of energy from saturated fat with energy from unsaturated fats would reduce risk by 42 percent, and that the replacement of 2 percent of energy from trans fat with energy from unhydrogenated, unsaturated fats would reduce risk by 53 percent.

They concluded:

"Our findings suggest that replacing saturated and trans unsaturated fats with unhydrogenated, monounsaturated and polyunsaturated fats is more effective in preventing coronary heart disease in women than reducing overall fat intake."

Commenting on these findings, Ronald Krauss, M.D., chairman of the American Heart Association’s Nutrition Committee stated:

"The study supports previous evidence that trans fats, like saturated fats, should be reduced in the diet. But because Americans consume more saturated fat than trans fat, the opportunities to reduce saturated fat should still be emphasised."

The UK intakes reported above (mean total fat intake 77g/day, equivalent to 35.7% of energy intake; mean TFA consumption 2.8g/day (= 1.3% of energy); and mean saturated fatty acids intake 28.5g/day (= 13.2% of energy)) suggest that here too the scope for reducing the intake of saturated fatty acids should continue to be emphasised.

It should be noted that aspects of the Hu et al study have been called into question. Ockene and Nicolosi (1998) have pointed out some inconsistencies in data that suggest the possibility that the method of assessing either the diet or exercise may be flawed, and also the danger of drawing conclusions from a study in which there was much lower risk of CHD (27 deaths per 100,000 per year) in the group of women studied, who were all nurses, than the overall risk in the population of women in the United States (rates of death from heart disease among white women 45 to 64 years of age in the United States: 62.5 and 51.0 per 100,000 per year, respectively, in 1985 and 1989).  Hegsted (1998) points out that the range of total fat intake recorded in these studies is limited and has little relevance to the protective effect of really low fat diets. However, he suggests that, within these limits, the composition of dietary fat, rather than the level, is of primary importance.

• There is conflicting evidence concerning the possible role of TFA in breast cancer.  As part of the EURAMIC study (European Community Multicentre Study on Antioxidants, Myocardial Infarction, and Breast Cancer) Kohlmeier et al (1997) investigated the relationship between TFA and postmenopausal breast cancer in European populations differing greatly in their dietary fat intakes.  A case control study using adipose tissue stores of TFA as a biomarker of exposure was conducted.  Subjects included 698 postmenopausal incident cases of primary breast cancer and controls randomly drawn from local population and patient registries, ages 50−74.  Concentrations of individual TFA in gluteal fat biopsies were measured in these women.  The adipose concentration of TFA showed a positive association with breast cancer, not attributable to differences in age, body mass index, exogenous hormone use, or socio-economic status.  The authors conclude that these findings suggest an association of adipose stores of TFA with postmenopausal breast cancer in European women, but point out that they require confirmation in other populations, with concomitant consideration of the potential roles of dietary saturated and monounsaturated fats.
• However, Holmes et al (1999) reported on the Cohort Study (Nurses' Health Study) conducted in the United States beginning in 1976.  A total of 88,795 women free of cancer in 1980 and followed up for 14 years.  Relative risk (RR) of invasive breast cancer for an incremental increase of fat intake, was ascertained by food frequency questionnaire in 1980, 1984, 1986, and 1990.  A total of 2956 women were diagnosed as having breast cancer.  Compared with women obtaining 30.1% to 35% of energy from fat, women consuming 20% or less had a multivariate RR of breast cancer of 1.15 (95% confidence interval [CI], 0.73-1.80).  In multivariate models, the RR (95% CI) for a 5%-of-energy increase was 0.97 (0.94−1.00) for total fat, 0.98 (0.96−1.01) for animal fat, 0.97 (0.93−1.02) for vegetable fat, 0.94 (0.88−1.01) for saturated fat, 0.91 (0.79−1.04) for polyunsaturated fat, and 0.94 (0.88−1.00) for monounsaturated fat.  For a 1% increase in energy from TFA, the values were 0.92 (0.86−0.98), and for a 0.1% increase in energy from omega-3 fat from fish, the values were 1.09 (1.03−1.16). In a model including fat, protein, and energy, the RR for a 5% increase in total fat, which can be interpreted as the risk of substituting this amount of fat for an equal amount of energy from carbohydrate, was 0.96 (95% CI, 0.93−0.99).  In similar models, no significant association of risk was evident with any major types of fat. They concluded that they found no evidence that lower intake of total fat or specific major types of fat was associated with a decreased risk of breast cancer.
• There is conflicting evidence about the role of conjugated linoleic acid (CLA).  In a review and meta-analysis, in mouse feeding trials at different total fat and different linoleic acid levels, Ritskes-Hoitinga et al (1996) found that mean mammary tumour incidence was higher and mean onset time shorter in the four high-fat groups than in the low-fat groups. However, no (linear) dose-response relationship between dietary linoleic acid and mammary tumour incidence and latency period was observed.  This indicates that a higher dietary linoleic acid does not increase the incidence or shorten the latency period of breast cancer in the Balb/c-MMTV mouse strain at two different dietary fat levels.  Zock and Katan (1998) concluded that 'controlled studies of coronary artery disease in men did not, except for one study, show an increased cancer incidence after consumption of diets with a very high linoleic acid content for several years.  Animal experiments indicated that a minimum amount of linoleic acid is required to promote growth of artificially induced tumors in rodents; but above (sic - they obviously mean "below") this threshold, linoleic acid did not appear to have a specific tumor-promoting effect.  Although current evidence cannot exclude a small increase in risk, it seems unlikely that a high intake of linoleic acid substantially raises the risks of breast, colorectal, or prostate cancer in humans. Conversely, however, Cesano et al (1998) determined the effect of three different diets on the local growth and metastatic properties of DU-145 human prostatic carcinoma cells in severe combined immunodeficient (SCID) mice.  Animals were fed a standard diet or diets supplemented with 1% linoleic acid or 1% CLA for 2 weeks prior to subcutaneous inoculation of DU-145 cells and throughout the study (total of 14 weeks).  Mice receiving the linoleic acid-supplemented diet displayed significantly higher body weight, lower food intake and increased local tumour load as compared to the other two groups of mice.  Mice fed the CLA-supplemented diet displayed not only smaller local tumours than the regular diet-fed group, but also a drastic reduction in lung metastases.  It appears from Cezano et al, that one particular TFA has anti-cancer activity.  The acid concerned is CLA, which is produced by the anaerobe Butyrivibrio fibrisolvens in the cow's rumen, and hence found in milk.  The main CLA isomer is cis-9, trans 11, with lesser amounts of trans 10, cis 12 octadecadienoic acid (Parodi, 1997).  The CLA content of milk lies between 0.24 and 2.81% and is highest when the animals are pasture fed.

A very thorough consideration of the health effects and implications of TFA consumption is given in the July 2004 Opinion of the EFSA Scientific Panel on Dietetic Products, Nutrition and Allergies.

The EFSA Scientific Opinion (2004) states:

TFA may be measured in a wide range of food products by infrared spectroscopy, which estimates total non-conjugated TFA, or by gas chromatography or high-pressure liquid chromatography, which can measure individual TFA with a high degree of precision. At present, there are no methods of analysis applicable to a wide range of foods that can distinguish between TFA which are naturally present in foods (e.g. in ruminant products) and those formed during the processing of fats, oils or foods.  This is because of the overlap in TFA profiles of ruminant fats and hydrogenated oils and the varying proportions of TFA isomers among different hydrogenated fats.

According to Leatherhead Food Research:

TFA can be determined simply by infrared analysis, although this is generally not accurate below 5%, and may be subject to interferences. Leatherhead Food International uses GC analysis of fatty acid methyl esters (FAMES), which is more accurate.  Where there are high levels of trans fatty acids present, it can be difficult to quantify the trans compounds because of interferences from cis fatty acids in the sample.  In this case, silver-ion chromatography is used to separate the trans and cis isomers, and the trans isomers are then re-analysed by GC. The combination of the data from the two GC analyses is then used to provide a value for the TFA content.

Comparative studies of the two analytical methods have shown that the GC-based analysis generally provides the more accurate determination of the TFA content of a sample.