1.4.1   Fuel

Fats are a very good source of energy: they provide about 9 calories (37.8kJ) of energy per gram [1].

1.4.2   Vitamin Bearing

Vitamins A, D, E and K are fat-soluble. This means that they can only be digested, absorbed and transported in conjunction with fats. A decrease in absorption of fat may result in a decrease in absorption of these vitamins [2].

1.4.3   Essential Fatty Acids

Certain fatty acids are required for various biological functions besides just acting as a fuel source. Of those fatty acids, some can be synthesised by the human body from other food components and some cannot. Those that the human body cannot synthesise, and which therefore must be ingested, are called essential fatty acids (EFAs) [3].

EFAs are unsaturated fatty acids. The carbon atom at the end of the fatty acid chain next to the carboxyl group is known as α (alpha). The carbon atom at the other end of the chain is known as ω (omega). The physiological properties of EFAs are largely dependent on the position of the first double bond relative to ω. They are therefore classed according to the position of that double bond. For example, if we move along the chain from ω, counting carbon-to-carbon bonds as we go, and we find 5 single bonds and then the 6th bond is a double bond, then this fatty acid is an ω-6 fatty acid.

Within each class of fatty acid, different molecules are further specified by counting the number of carbon atoms in the chain and the number of double bonds within the chain. Consecutive double bonds are assumed to be separated by a single carbon atom unless otherwise specified. For example, stearidonic acid has a chain of 18 carbon atoms with 4 double bonds, the first of which is the third bond from the ω end of the chain. It is therefore written like this: ω-3 18:4.

stearidonic acid

There are only two EFAs:

1. α-linolenic acid (αLNA) ω-3 18:3
2. linoleic acid (LA) ω-6 18:2

However, there are a number of other fatty acids which perform vital biological functions and are therefore sometimes referred to as EFAs, even though strictly speaking they are not "essential" since the human body can synthesise them from αLNA or LA [4].

1. eicosapentaenoic acid (EPA) ω-3 20:5
2. docosahexaenoic acid (DHA) ω-3 22:6
3. γ-linolenic acid (γLNA) ω-6 18:3
4. dihomo-γ-linolenic acid (DγLNA) ω-6 20:3
5. arachidonic acid (AA) ω-6 20:4

The ease and efficiency with which these fatty acids can be synthesised from αLNA or LA determines how likely they are to be considered "essential".

EFAs appear to be important for a whole host of different biological functions.

ω-3 fatty acids are currently being investigated as preventative agents in coronary heart disease, diabetes, thyrotoxicosis, bronchial asthma, multiple sclerosis and psoriasis, primarily as a result of a study comparing the health and diets of the inhabitants of Greenland with mainland Danes [5].

Increasing the relative proportions of ω-3 fatty acids ingested compared to ω-6 fatty acids seems to moderate inflammatory reactions, which are "at the basis of a number of disease processes in virtually any systemic or organ-specific disease, ranging from classical rheumatic diseases to bronchial airway hyper-responsiveness, inflammatory bowel disease, kidney diseases, psoriasis and atopic eczema" [4]. Most of these diseases appear to share a background in inflammation or derangement of immune systems.

EFAs, especially arachidonic acid, are also involved in neurological processes. "The arachidonic acid cascade is arguably the most elaborate signaling system neurobiologists have to deal with" [6]. AA appears to be related to various mental conditions, including bipolar disorder and Alzheimer disease [7].

1.5.1   Saturated vs. Unsaturated

Although a lot of people, including the World Health Organisation, the United States Centers for Disease Control, the American Heart Association, the UK Food Standards Agency, and loads of others, recommend minimising the amount of saturated fat in your diet, I cannot actually find any particularly good reason to do so. They all seem to claim that saturated fats cause cardiovascular disease but I don't know why they say that.

So far, I have only found two meta-analyses of research into possible links between saturated fat and cardiovascular diseases [8] [9], both of which conclude that there is no statistically significant link.

The UK Food Standards Agency claims that "Eating a diet that is high in saturated fat can raise the level of cholesterol in your blood, over time. This increases your chance of developing heart disease" [10]. But they don't cite any evidence to back up that claim.

A study of two populations of Polynesians seems to support the theory that an increased level of dietary saturated fat causes an increase in cholesterol levels: "Tokelauans obtain a much higher percentage of energy from coconut than the Pukapukans, 63% compared to 34%, so their intake of saturated fat is higher. The serum cholesterol levels are 35 to 40 mg higher in Tokelauans than in Pukapukans. These major differences in serum cholesterol levels are considered to be due to the higher saturated fat intake of the Tokelauans." However: "Vascular disease is uncommon in both populations and there is no evidence of the high saturated fat intake having a harmful effect in these populations" [11].

In post-menopausal women, a diet high in saturated fat is associated with less progression of coronary atherosclerosis, whereas replacing other fats with polyunsaturated fat is associated with greater progression (monounsaturated fats and total fat intake are not associated with progression at all) [12].

Heating polyunsaturated fats (e.g. when using them for frying or other forms of cooking) may cause them to break down into toxic chemicals and trans fats [13].

1.5.2   Trans Fats

In contrast to the saturated/unsaturated confusion, everyone seems to agree that trans fats are very bad for you. There does appear to be a strong statistical link between the consumption of trans fats and cardiovascular disease [14].

One proposed mechanism for the causation of cardiovascular disease by trans fats is the observation that trans fats raise the level of low-density-lipoprotein (LDL) cholesterol in the blood while simultaneously lowering the level of high-density-lipoprotein (HDL) cholesterol.

Lipoproteins are the complex molecules that transport cholesterol through blood. The density of a lipoprotein refers to the ratio of cholesterol to protein within the molecule: the more cholesterol and the less protein it has, the less dense it is. LDL cholesterol is positively correlated with cardiovascular disease and diabetes, whereas HDL is negatively correlated with those diseases [15]. Thus simultaneously raising LDL while lowering HDL would seem to be the worst of both worlds.

(Incidentally, perhaps a failure to distinguish between different lipoprotein densities is behind the apparently unjustified condemnation of saturated fats?)

Trans fats also appear to inhibit the enzymatic desaturation of linoleic and α-linolenic acids, thus preventing the synthesis of important essential fatty acids [14]. As noted above, EFAs are linked to the prevention of coronary heart disease, so the prevention of their synthesis is evidently not a good thing.

Trans fats only occur in trace amounts in nature. They do not occur in plant matter at all and only occur in very small amounts in meat from ruminants (such as cattle and sheep) [16]. Almost all the trans fat consumed by humans today is created by the process of partial hydrogenation. This process begins with a natural unsaturated fat (which will always be a cis-isomer) and then modifies the fatty acid chains in order to saturate some of the unsaturated double bonds. However, the hydrogenation is only partial: some double bonds remain and some of those that do remain are transformed from cis-isomer bonds to trans-isomer bonds. This process is popular in the food industry because it turns fats that are liquid at room temperature into solids, which are more convenient for storage (they are less susceptible to turning rancid and so have a longer shelf-life and require less refridgeration) and can be used as low-cost replacements for the solid fats used in baking. The use of margarine to replace butter or lard is a good example of this. Trans fats are also used for deep-frying in restaurants as they can be used for longer than natural oils before they turn rancid. They are therefore very popular in the fast food industry.

So not only are trans fats very bad for your health in themselves but they are also used to replace natural fats and so reduce your consumption of essential fatty acids.

1.5.3   Essential Fatty Acids

As noted above, EFAs are vital for a wide range of biological processes. Although only two fatty acids are strictly "essential", i.e. impossible for the human body to synthesise, the ability to synthesise the other important fatty acids, especially EPA and DHA, is limited:

Two main families of long-chain polyunsaturated fatty acids exist, biologically derived from the shortest non-synthesizable precursors linoleic acid (C18:2n-6) and alpha-linolenic acid (C18:3). Linoleic acid is abundant in oils from most vegetable seeds such as corn and safflower. Alpha-linolenic acid is found in the chloroplasts of green leafy vegetables. Humans can desaturate and elongate alpha-linolenic acid to eicosapentaenoic acid (EPA) and, further, to docosahexaenoic acid (DHA). However, the elongation and desaturation processes are likely to be slow and possibly further limited with ageing and disease conditions. For these reasons, EPA and DHA are considered, to a large extent, nutritionally essential and nearly exclusively derived from fish. Fish increase their membrane content by eating the phytoplankton rich in either the precursor alpha-linolenic acid or the more elongated compounds EPA and DHA. Fatty fish living in cold seas (e.g. mackerel, salmon and herring) are particularly rich in these compounds, which may give them a selective advantage in preventing low-temperature-related loss in membrane fluidity in cell membranes. Therefore, EPA and DHA may be considered nutritionally essential for most practical purposes and largely derived from fish and marine oils. [4]

This claim is further supported by epidemiological studies that show that fish consumption is inversely related to fatal coronary heart disease [17] and ischemic stroke [18]. That this benefit is derived primarily from consumption of EPA and DHA is also quite well established:

This evidence exists. Results of case-control studies, prospective cohort studies, and randomized controlled trials each indicate that modest consumption of fish or fish oil lowers the risk of cardiac mortality, specifically CHD death and sudden cardiac death. This effect appears to be nonlinear: compared with little or no intake, modest consumption (~250 mg/d) of the marine n-3 PUFA eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) significantly lowers risk of cardiac mortality, whereas higher intakes do not substantially further lower risk, suggesting a threshold of effect. The similarity of findings between observational studies of fish consumption and randomized controlled trials of n-3 PUFA supplementation suggests that, at least for cardiac mortality, much of the benefit of fish intake is related to the n-3 PUFA content. Consistent with this, when different types of fish meals are considered, lower risk is more strongly related to intake of fatty (oily or dark meat) fish, compared with lean (white meat) fish. The quantity of fish servings needed to consume an average of 250 mg/d EPA+DHA varies depending on the particular fish species, but for fatty fish (e.g., anchovies, herring, salmon, sardines, trout, white tuna) is ~1–2 servings/week.

The strength and consistency of the evidence, and the magnitude of the benefit, for lowering of cardiac mortality by fish consumption are each notable. The risk reduction is supported by consistent evidence from experimental studies and randomized trials investigating effects of fish or fish oil consumption on cardiovascular risk factors; case-control studies evaluating fish consumption or objective biomarkers of n-3 PUFA intake and risk of cardiac outcomes; prospective studies of habitual fish consumption and cardiac outcomes that have followed hundreds of thousands of people for many years across a range of countries; and randomized controlled trials of fish or fish oil consumption that have enrolled thousands of subjects and demonstrated reductions in clinical events. The magnitude of the benefit is also considerable: pooling of the prospective studies and controlled trials indicates that, compared with no intake, modest consumption (~250 mg/day [2¼ calories/day] of EPA+DHA, equivalent to ~1–2 servings/week of fatty fish) lowers risk of CHD mortality by 36%. Analyses restricted to populations free of established heart diease (i.e., primary prevention) demonstrate similar results. Fish intake may also reduce the risk of other cardiovascular and noncardiovascular outcomes, including but not limited to nonfatal heart attacks, ischemic stroke, atrial fibrillation, cognitive decline, and depression, but the evidence for these benefits is not yet as robust as for CHD death. [19].

There is some concern that the cardiovascular benefits of eating fish may be reduced by the danger of consuming toxic mercury compounds that are known to be present in fish that swim in our polluted waters (i.e. all of them). This area has not yet been well researched but current evidence suggests that the benefits of eating one or two servings of fatty fish per week far outweigh any risks from mercury poisoning [19].

The apparent benefits of consuming long-chain ω-3 fatty acids (i.e. fatty acids that are synthesisable from αLNA) have prompted a lot of research. It is fairly well established that although EPA and DHA can be synthesised from αLNA, such synthesis does not appear to be enough to reach optimal levels in humans. Conversion from αLNA to EPA does occur but it is limited in men and further conversion to DHA is very low. More conversion occurs in women, which may be due to a regulatory effect of oestrogen [20].

Apparently, some strains of algae also contain EPA and DHA, and so would be a good alternative to fish for vegetarians, but I cannot find any freely available journal articles to back up this claim. The idea appears to be that the fish get their EPA and DHA from algae in the first place, so we ought to be able to skip the fish-eating step and just get our long-chain ω-3 FAs straight from the algae. This sounds plausible, assuming humans have the ability to digest algae efficiently. I have found many references to algal sources of EPA and DHA but so far they have all either been in closed journals that require a subscription fee to read, or they have been marketing screeds for companies selling nutritional supplements to vegetarians. The only reference in a reputable, open journal that I have found is in the midst of an article about DHA in breast milk. Apparently they used algae-derived DHA as a nutritional supplement [21]. However, since I have found so many references to this claim, including some that seem quite reputable [22], I think it is probably true.

Less research has been performed to investigate the synthesisability of long-chain ω-6 fatty acids (i.e. those that are synthesisable from LA). However, it does seem that vegetarians have a lower level of arachidonic acid (AA) in their blood than omnivores [23]. There is some evidence to suggest that this is a deficiency caused by their inability to synthesise enough AA from the ω-6 fatty acids present in a vegetarian diet, whereas omnivores consume preformed AA in animal products, but this evidence is far from conclusive and other studies have produced varying results. I haven't found a meta-analysis of this research yet.

1.5.4   Animal Fats vs Vegetable Fats

Animal fats contain long-chain EFAs that are not found in any vegetable fats [4] [20] [23]. Apart from this, I have found no evidence to suggest any difference in health benefits or detriments between animal and vegetable fats.