The essentiality of the n-6 fatty acid, linoleic acid, was established about 80 years ago⁽Reference Burr and Burr^1, Reference Burr and Burr²⁾; the essentiality of the n-3 analogue, α-linolenic acid, became accepted rather later⁽Reference Pudelkewicz, Seufert and Holman^3, Reference Holman, Johnson and Hatch⁴⁾. The intakes of these fatty acids necessary to avoid deficiency are quite low (estimated to be 1 and 0·2 % of energy intake, respectively, corresponding to intakes in adults of about 2 and 0·4 g/d), and all diets should obviously provide at least these amounts. Classic studies by Keys, Hegsted, Katan and others have established that several SFA raise blood cholesterol and LDL concentrations, while linoleic acid lowers these concentrations⁽Reference Katan, Zock and Mensink⁵⁾. The differential effects of saturated and linoleic acids on the blood cholesterol concentration are expected to affect the risk of CVD and have been reflected in dietary recommendations. In the UK, the relevant recommendations are ‘saturated fatty acids should provide less than 10 % of total dietary energy’, ‘cis-PUFAs should continue to provide an average of 6 % of total dietary energy and be derived from a mixture of n-6 and n-3 PUFAs’ and ‘dietary intake of PUFAs by individuals should not exceed 10 % of total energy’⁽⁶⁾. The realisation of the effect of linoleic acid on blood cholesterol concentration and the anticipation that an increased intake of linoleic acid at the expense of SFA would benefit cardiovascular health paved the way for the introduction of margarines and linoleic acid-rich vegetable oils into the human food chain with associated heavy marketing. The result was a large increase in the intake of linoleic acid in the population over the period from the 1960s to 1990. In the USA, the average daily intake of linoleic acid among adults increased from about 5 to about 8 % of energy between 1960 and 1990, and the current average intake in adults is about 15 g/d⁽Reference Mosfegh, Goldman and Cleveland⁷⁾. The average intake of linoleic acid among adults in the UK is currently about 5·4 % of energy or 13 g/d in men and 9·5 g/d in women⁽Reference Henderson, Gregory and Irving⁸⁾.

Despite the enthusiasm for increasing linoleic acid intake, some doubts remained when the recommendations for linoleic acid intake were established. First an increased linoleic acid content of LDL increases its sensitivity to oxidation⁽Reference Tsimikas, Philis-Tsimikas and Alexopoulos⁹⁾, and it is the oxidised LDL that is involved in atherosclerotic plaque growth. Hence, an increase in linoleic acid intake could actually promote plaque growth and CVD. Second, there was some evidence that linoleic acid could promote certain cancers, especially in animal models⁽Reference Welsch¹⁰⁾. These doubts were captured in an additional, often ignored, recommendation related to PUFA intake in the UK: ‘there is reason to be cautious about high intakes of n-6 PUFAs, and we recommend that the proportion of the population consuming in excess of about 10 % of energy [as n-6 PUFAs] should not increase’⁽¹¹⁾.

Two other scientific developments have had an impact on the general view that n-6 fatty acids, in general, and linoleic acid, in particular, are, almost by definition, associated with health. The first scientific development was the discovery of the actions of eicosanoid mediators derived from arachidonic acid. Arachidonic acid is produced by the metabolism of linoleic acid, and the arachidonic acid-derived eicosanoids are involved in many physiological and pathophysiological responses including inflammation and thrombosis; arachidonic acid metabolism is the target of several pharmaceutical agents. Since both inflammation and thrombosis are involved in atherosclerotic plaque growth and rupture, once again, this suggests the possibility of an increase in cardiovascular risk with increased n-6 fatty acid exposure. The second scientific development was the discovery of the protective effects of marine n-3 fatty acids towards CVD and the improved understanding of their mechanisms of action⁽Reference Calder^12, Reference Saravanan, Davidson and Schmidt¹³⁾. An increased intake of marine n-3 fatty acids lowers arachidonic acid status and interferes with arachidonic acid metabolism, decreasing the production of pro-inflammatory and pro-thrombotic mediators⁽Reference Calder and Yaqoob¹⁴⁾. Hence, the recognised health benefits of marine n-3 fatty acids and the realisation that they often work by ‘antagonising’ n-6 fatty acids generated a perception of ‘n-3 good, n-6 bad’⁽Reference Simopoulos^15, Reference Lands¹⁶⁾.

This status quo was perturbed last year when the American Heart Association (AHA) published an advisory on n-6 fatty acids and cardiovascular risk⁽Reference Harris, Mozaffarian and Rimm¹⁷⁾. The advisory dismissed concerns about inflammation, thrombosis and LDL oxidation. Each of these aspects requires a deeper debate given the increased evidence base that currently exists. However, the main component of the advisory was devoted to the evaluation of observational studies and randomised controlled trials (RCT) of n-6 fatty acids and CHD events. It is generally accepted that RCT provide the highest level of evidence of cause and effect. The AHA advisory evaluated the findings of nine RCT published from 1965 to 1989 and noted that at least five of these trials had design limitations; the latter limitations included the simultaneous use of plant or marine n-3 fatty acids in some studies. Although these limitations appear not to have been considered in developing the advisory, they may have influenced its major scientific conclusion (‘replacing saturated fatty acids with PUFAs lowered CHD events’), which is clearly not an n-6 fatty acid (or linoleic acid)-specific summary statement, and so somewhat blurs the boundaries between PUFA, n-6 PUFA and linoleic acid. The advisory states that ‘consumption of at least 5 % to 10 % of energy from n-6 PUFAs reduces risk of CHD relative to lower intakes’ and that ‘to reduce n-6 PUFA intakes from their current levels would be more likely to increase than to decrease risk for CHD’.

An article in this issue of the BJN addresses the question of linoleic acid and CHD mortality risk, in the light of the AHA advisory⁽Reference Ramsden, Hibbeln and Majchrzak¹⁸⁾. The initiation of the work of Ramsden et al. is their recognition that the n-6 PUFA advisory was largely based on the evidence from RCT that often did not address the question of n-6 PUFA or linoleic acid in isolation, but rather included mixtures of n-6 and n-3 PUFA. Ramsden et al. have evaluated the findings of studies specifically addressing the impact of increased linoleic acid intake separately from those that included linoleic acid in combination with n-3 fatty acids. Their efforts involved extensive detective work that included identification and use of food composition data from the locations and periods when several of the studies were performed, often over 40 years ago, and contacting original researchers of some of the older studies in order to clarify uncertain points. This scholarly approach, combined with the appreciations that (i) the terms PUFA and n-6 PUFA mean different things and (ii) linoleic acid alone and linoleic acid in combination with n-3 fatty acids may produce different findings, has yielded a different conclusion from the AHA advisory: ‘advice to specifically increase n-6 PUFA intake, based on mixed data, is unlikely to provide the intended benefits, and may actually increase the risks of CHD and death’. This piece of work by Ramsden et al. is to be applauded and will, it is hoped, open a healthy scientific debate on this important matter.