Omega-6 Polyunsaturated Fatty Acids: From Molecular Essentials to Scientific Controversies

Last updated: May 21, 2026 Русский

For decades, dietary fats have been at the center of nutritional science, evolving from a simple narrative of “good” versus “bad” to a much more complex understanding of individual fatty acids and their metabolic fates. Among these, omega-6 (n-6) polyunsaturated fatty acids (PUFAs) — particularly linoleic acid (LA), gamma-linolenic acid (GLA), and arachidonic acid (AA) — occupy a unique and often controversial position. Despite being essential nutrients, they have been simultaneously praised for their cholesterol-lowering properties and blamed for promoting chronic inflammation (Maki et al., 2018; Innes & Calder, 2018).

The story of omega-6 fatty acids is a compelling case study of how scientific evidence evolves, sometimes leading to contradictory public health messages. LA, the most abundant dietary PUFA, cannot be synthesized by the human body and must be obtained from plant oils, nuts, and seeds (Jackson et al., 2024). Once consumed, LA can be metabolized through a series of desaturation and elongation steps — involving enzymes such as delta-6-desaturase — into GLA, dihomo-gamma-linolenic acid (DGLA), and ultimately arachidonic acid (AA), a key structural component of cell membranes (Djuricic & Calder, 2021; Gutierres et al., 2025). This metabolic pathway, shared with omega-3 fatty acids, lies at the heart of many scientific debates.

Historically, the 1950s and 1960s saw strong enthusiasm for replacing saturated fats with vegetable oils rich in LA, based on the pioneering work of Ancel Keys and others who linked saturated fat intake to elevated serum cholesterol and coronary heart disease risk (Kris-Etherton & Fleming, 2015; Lenighan et al., 2019). This led to a dramatic increase in LA consumption in Western countries, particularly from soybean, corn, and sunflower oils (Blasbalg et al., 2011, cited in Jackson, 2024). However, the narrative shifted in the late 20th century when concerns emerged that a high dietary ratio of omega-6 to omega-3 fatty acids could promote a pro-inflammatory state, largely because AA is a precursor to eicosanoids (prostaglandins, leukotrienes) involved in inflammatory processes (Innes & Calder, 2018; Berkowitz et al., 2026).

This controversy has been fueled by conflicting results from clinical trials. While some large meta-analyses of randomized controlled trials have demonstrated a significant reduction in coronary heart disease events when saturated fats are replaced with PUFAs (Mozaffarian et al., 2010, cited in Maki, 2018), re-analyses of older studies such as the Sydney Diet Heart Study and the Minnesota Coronary Experiment raised questions about potential harm, though these re-analyses have been criticized for methodological flaws, including the presence of industrial trans fats in the intervention margarines (Ramsden et al., 2013, cited in Jackson, 2024; Maki et al., 2018).

In recent years, large-scale pooled analyses of prospective cohort studies have provided more clarity. The Fatty Acids and Outcomes Research Consortium (FORCE) demonstrated that higher circulating levels of LA are associated with a significantly lower risk of total cardiovascular disease, cardiovascular mortality, and incident type 2 diabetes, while AA levels showed no adverse associations (Marklund et al., 2019; Wu et al., 2017). These findings challenge the notion that dietary omega-6 PUFAs are inherently harmful. Nevertheless, heterogeneity remains: some meta-analyses report higher omega-6 intake linked to increased risk of certain cancers (e.g., endometrial, ovarian), while others suggest protective effects against breast and lung cancers (Sadeghi et al., 2025; Retterstøl & Rosqvist, 2024).

This article aims to provide an educational overview of the science surrounding omega-6 fatty acids, focusing on their biochemistry, dietary sources, historical shifts in dietary guidelines, and the key epidemiological and clinical evidence that has shaped current understanding. We will also examine persistent questions — including the role of AA in inflammation, the utility of the n-6/n-3 ratio, and the influence of genetic variation in the FADS gene cluster (Wu et al., 2017; Berkowitz et al., 2026) — to illustrate the complexity of translating fatty acid research into public health advice. The goal is not to offer medical recommendations but to equip readers with a nuanced, evidence-informed perspective on this fascinating and often misunderstood class of nutrients.

1. Chemical Nature and Biochemistry of Omega-6 Fatty Acids

Omega-6 polyunsaturated fatty acids (n-6 PUFAs) are defined by the position of their first carbon-carbon double bond, which is located on the sixth carbon atom counting from the methyl (omega) end of the hydrocarbon chain (Gutierres et al., 2025). This structural feature distinguishes them from omega-3 fatty acids, where the first double bond is at the third carbon, and from monounsaturated or saturated fatty acids, which lack such multiple unsaturations. The most abundant and metabolically central omega-6 PUFA in the human diet is linoleic acid (LA; 18:2n-6), an 18-carbon chain with two double bonds.

1.1. Essentiality and Dietary Necessity

LA is classified as an essential fatty acid because humans, like all mammals, lack the enzymes (specifically Δ12- and Δ15-desaturases) required to introduce double bonds at the n-6 or n-3 positions of the carbon chain (Djuricic & Calder, 2021; Gutierres et al., 2025). Consequently, LA cannot be synthesised endogenously and must be obtained from dietary sources. The physiological importance of LA was first recognised in the mid-20th century when studies on infants fed low-fat diets revealed characteristic deficiency symptoms, including rough, dry, and scaly skin, dermatitis, and impaired growth (Hansen et al., 1958, cited in Jackson, 2024; Innes & Calder, 2018). These symptoms are reversible upon LA repletion, underscoring its non-negotiable role in maintaining the structural integrity of the epidermal barrier, where LA is an essential constituent of ceramides (Djuricic & Calder, 2021). Beyond the skin, severe LA deficiency can also affect hepatic function and reproductive health, though such deficiencies are extremely rare in healthy populations consuming varied diets (Retterstøl & Rosqvist, 2024).

1.2. Metabolism: From Linoleic Acid to Arachidonic Acid

Metabolic pathways of omega-6 and omega-3 fatty acids, showing conversion of linoleic acid to arachidonic acid and alpha-linolenic acid to EPA and DHA, as well as the formation of oxylipins, prostaglandins, leukotrienes and resolvins.

Metabolic pathways of omega-6 and omega-3 polyunsaturated fatty acids

Simplified scheme of the desaturation and elongation pathway shared by omega-6 and omega-3 fatty acids, and the major classes of bioactive lipid mediators derived from each family. Adapted from Gutierres et al. (2025), Figure 1. CC BY 4.0

Once ingested, LA undergoes a series of enzymatic transformations, primarily in the liver, to generate other biologically active omega-6 PUFAs. This pathway, shared with omega-3 fatty acids (which compete for the same enzymes), involves alternating desaturation (addition of a double bond) and elongation (addition of two carbon atoms) steps (Djuricic & Calder, 2021; Gutierres et al., 2025).

The first step is the conversion of LA (18:2n-6) to gamma-linolenic acid (GLA; 18:3n-6) via the enzyme Δ6-desaturase (encoded by the FADS2 gene). GLA is then elongated by elongase 5 (ELOVL5) to form dihomo-gamma-linolenic acid (DGLA; 20:3n-6). Finally, DGLA is desaturated by Δ5-desaturase (encoded by FADS1) to produce arachidonic acid (AA; 20:4n-6) — a 20-carbon chain with four double bonds (Berkowitz et al., 2026; Innes & Calder, 2018).

It is important to note that the conversion efficiency from LA to AA is relatively low in humans, and the rate-limiting step is the Δ6-desaturation (Djuricic & Calder, 2021). Furthermore, this conversion can be influenced by multiple factors, including the dietary ratio of LA to alpha-linolenic acid (ALA, the plant-derived omega-3), the presence of certain micronutrients (iron, zinc, magnesium), and genetic variations in the FADS gene cluster (see Section 8) (Wu et al., 2017; Berkowitz et al., 2026).

1.3. Arachidonic Acid: A Structural and Signalling Hub

Arachidonic acid is not merely an end-product of LA metabolism; it is a critical component of cell membrane phospholipids, particularly in cells of the immune, nervous, and cardiovascular systems (Innes & Calder, 2018). In humans, AA can constitute as much as 15–20% of the total fatty acids in peripheral blood mononuclear cells and up to 25% in skeletal muscle, brain, liver, and platelet phospholipids (Djuricic & Calder, 2021; Innes & Calder, 2018). The incorporation of AA into membranes confers fluidity and influences the function of membrane-bound proteins, receptors, and ion channels (Calder, 2015, cited in Djuricic & Calder, 2021).

Upon cellular activation by mechanical, hormonal, or inflammatory stimuli, AA is released from membrane phospholipids by the action of phospholipase A2 enzymes (cPLA2) (Innes & Calder, 2018). Free AA serves as the primary substrate for three major enzyme families:

  • Cyclooxygenases (COX-1 and COX-2), leading to the synthesis of prostaglandins (e.g., PGE2, PGI₂) and thromboxanes (e.g., TXA2).
  • Lipoxygenases (5-LOX, 12-LOX, 15-LOX), leading to the synthesis of leukotrienes (e.g., LTB4, LTC4), lipoxins, and hydroxyeicosatetraenoic acids (HETEs).
  • Cytochrome P450 (CYP) enzymes, producing epoxyeicosatrienoic acids (EETs) and dihydroxyeicosatrienoic acids (DHETs) (Djuricic & Calder, 2021; Innes & Calder, 2018).

These oxygenated derivatives, collectively known as eicosanoids, act as potent local signalling molecules with a wide range of physiological effects, including regulation of vascular tone, platelet aggregation, pain perception, and — most controversially — inflammatory responses (Innes & Calder, 2018; Berkowitz et al., 2026). While many AA-derived eicosanoids (e.g., PGE2, LTB4) are classically described as pro-inflammatory, it is now recognised that AA also gives rise to pro-resolving mediators such as lipoxin A4 (LXA4), which actively contribute to the termination and resolution of inflammation (Djuricic & Calder, 2021; Innes & Calder, 2018). This duality is central to the ongoing scientific debate over the health effects of omega-6 fatty acids.

1.4. Other Omega-6 Fatty Acids: GLA and DGLA

Although present in much smaller amounts in typical diets, GLA and DGLA also possess biological activities distinct from LA and AA. GLA is found in certain seed oils (evening primrose, borage, blackcurrant) and is efficiently converted to DGLA in human tissues (Innes & Calder, 2018). DGLA can itself be metabolised by COX and LOX enzymes to produce eicosanoids with generally anti-inflammatory and anti-aggregatory properties, such as prostaglandin E₁ (PGE₁) and 15-hydroxy-eicosatrienoic acid (15-HETrE) (Djuricic & Calder, 2021; Sergeant et al., 2016, cited in Innes, 2018). Unlike AA, DGLA does not give rise to leukotrienes, and some studies suggest that increasing DGLA levels may competitively inhibit the formation of AA-derived pro-inflammatory mediators (Innes & Calder, 2018).

Thus, the metabolic pathway from LA to AA is not a simple linear cascade but a complex network with multiple bioactive intermediates, each with potential physiological and pathophysiological roles. Understanding this biochemistry is essential for interpreting the often conflicting results of epidemiological and clinical studies on dietary omega-6 PUFAs and human health.

2. Major Dietary Sources and Global Consumption Patterns

The omega-6 fatty acid content of the human diet is overwhelmingly determined by linoleic acid (LA), which typically accounts for 80–90% of total dietary PUFA intake (Maki et al., 2018). Arachidonic acid (AA) and gamma-linolenic acid (GLA) contribute only minor fractions under most dietary patterns, though their physiological significance is disproportionately high relative to their dietary abundance.

2.1. Plant Sources: Vegetable Oils, Nuts, and Seeds

The richest dietary sources of LA are liquid vegetable oils derived from seeds and grains. These oils are widely used in cooking, food manufacturing, and as ingredients in salad dressings, margarines, and processed foods. Table 1 presents the fatty acid composition of commonly consumed oils, adapted from the USDA FoodData Central database (Petersen et al., 2024). The data illustrate the wide variability in LA content, from very high levels in grapeseed and soybean oils to moderate levels in olive and avocado oils, and very low levels in tropical oils such as palm and coconut.

Table. Fatty acid content of selected oils (g/100 g), listed in descending order of total PUFA content.

Oil Total PUFA Linoleic acid (LA, 18:2n-6) Alpha-linolenic acid (ALA, 18:3n-3) MUFA SFA
Grapeseed 69.9 69.6 0.1 16.1 9.6
Soybean 57.7 51.0 6.8 22.8 15.6
Corn 54.7 53.5 1.2 27.6 12.9
Cottonseed 51.9 55.5 0.2 17.8 25.9
Sesame 41.7 41.3 0.3 39.7 14.2
Rice bran 35.0 33.4 1.6 39.3 19.7
Sunflower (mid-oleic) 29.0 28.9 < 0.1 57.3 9.0
Canola 28.1 19.0 9.1 63.3 7.4
Peanut 19.9 19.6 0.0 57.1 16.2
Avocado 13.5 12.5 1.0 70.6 11.6
Safflower (high oleic) 12.8 12.7 0.1 75.2 7.5
Olive 10.5 9.8 0.7 73.0 13.8
Palm 9.3 9.1 0.2 37.0 49.3
Coconut 1.8 1.6 0.0 6.3 82.5

Source: USDA FoodData Central, as compiled in Petersen et al. (2024).

Beyond oils, important dietary sources of LA include nuts (particularly walnuts, almonds, and hazelnuts), seeds (sunflower, pumpkin, flax, and chia seeds), and whole grains (wheat germ, oats). These whole-food sources also provide fibre, protein, minerals, and phytochemicals, which may modulate the biological effects of LA (Gutierres et al., 2025).

2.2. Animal Sources: Arachidonic Acid and Meat Products

Arachidonic acid (AA) is found almost exclusively in animal-derived foods. The highest concentrations are present in organ meats (liver, brain, kidney), egg yolks, and fatty meats (pork, beef, poultry) (Djuricic & Calder, 2021). In contrast to LA, AA is not produced by plants, and its levels in animal tissues are influenced by the animal’s diet — for instance, grain-fed animals tend to have higher AA content in their fat depots compared to grass-fed animals, though the absolute differences are modest (Gutierres et al., 2025).

Estimates from dietary surveys indicate that the average daily intake of AA in developed countries ranges from 100 to 350 mg per day, contributing less than 0.1% of total energy intake (Djuricic & Calder, 2021; Calder et al., 2019, cited in Innes, 2018). This is orders of magnitude lower than typical LA intakes (see Section 2.3). GLA is even more scarce in the average diet, appearing only in small amounts in certain seed oils (evening primrose, borage, blackcurrant) and in some nuts and legumes (Innes & Calder, 2018).

2.3. Historical Trends in LA Consumption

The 20th century witnessed a dramatic transformation in the fatty acid profile of Western diets, primarily driven by the industrial production of vegetable oils. Between 1909 and 1999, per-capita vegetable oil consumption in the United States increased from approximately 0.7 kg to 14.7 kg per person per year — a more than 20‑fold increase (Blasbalg et al., 2011, cited in Jackson, 2024; Berkowitz et al., 2026). Soybean oil, which was virtually absent from the food supply in the early 1900s, became the dominant vegetable oil by the end of the century, rising from 0.01 to 11.6 kg per person per year. As a direct consequence, the percentage of dietary energy derived from LA increased from roughly 2–3% in the early 20th century to approximately 7–8% by the 1990s (Blasbalg et al., 2011, cited in Berkowitz, 2026).

This shift was partly intentional, driven by public health recommendations to replace saturated fats (from butter, lard, and tallow) with polyunsaturated vegetable oils to lower serum cholesterol and reduce cardiovascular disease risk (Kris-Etherton & Fleming, 2015; Lenighan et al., 2019). However, the simultaneous increase in LA consumption also raised concerns about a potential pro-inflammatory imbalance relative to omega-3 fatty acids, as discussed in later sections.

2.4. Current Population Intakes in Comparison with Recommendations

Current dietary guidelines for LA vary among authoritative bodies, reflecting the ongoing scientific debate. In the United States, the Institute of Medicine (now the National Academy of Medicine) has set Adequate Intakes (AIs) for LA at 17 g/day for men aged 19–50 years and 12 g/day for women of the same age, which correspond to approximately 5–6% of total energy intake on a 2000–2500 kcal diet (Jackson et al., 2024). The Acceptable Macronutrient Distribution Range (AMDR) for n-6 PUFAs is set at 5–10% of energy, while the American Heart Association recommends a similar range of 5–10% of energy from n-6 PUFAs (Harris et al., 2009, cited in Jackson, 2024; Petersen et al., 2024).

The Food and Agriculture Organization (FAO) and World Health Organization (WHO) recommend that LA intake be between 2.5% and 9% of total energy, with a suggested minimum of 2.5% to prevent deficiency (FAO/WHO, 2010, cited in Jackson, 2024; Gutierres et al., 2025). The European Food Safety Authority (EFSA) has set an Adequate Intake of 4% of total energy for LA (EFSA, 2010, cited in Petersen, 2024).

In practice, population surveys indicate that most adults in Western countries consume LA within or slightly above these recommended ranges. For example, mean LA intake in the United States is approximately 6–7% of energy (Raatz et al., 2018, cited in Jackson, 2024), while in the Nordic and Baltic countries, average PUFA intake (predominantly LA) ranges from 5.5% to 6.9% of energy across different nations (Retterstøl & Rosqvist, 2024; Lemming & Pitsi, 2022, cited in NNR2023). However, considerable inter-individual variability exists, and some population subgroups — particularly those with very low vegetable oil or nut consumption — may have intakes below the recommended minimum.

A noteworthy recent trend is the replacement of traditional LA-rich seed oils with high-oleic varieties (e.g., high-oleic sunflower, canola, and soybean oils). These modified oils have a higher monounsaturated fatty acid (MUFA) content and a correspondingly lower PUFA content. Modelling studies suggest that widespread adoption of high-oleic oils could reduce LA intake in the US population from approximately 20 g/day to 14 g/day, a decrease of about 30% (Raatz et al., 2018, cited in Jackson, 2024). The long-term health implications of such a shift remain unknown and are an active area of research.

3. Evolution of Scientific Views: From the ‘Cholesterol Hypothesis’ to Contemporary Debates

The history of omega-6 fatty acids in nutritional science is not a linear progression from ignorance to enlightenment, but rather a winding path marked by enthusiastic adoption, sharp criticism, and gradual re-evaluation. Understanding this evolution is essential for interpreting the conflicting messages that have reached both clinicians and the public. This section traces the key turning points in the scientific debate over LA and its metabolites.

3.1. Early Decades (1950s–1970s): The Cholesterol Hypothesis and the Rise of Vegetable Oils

The modern story of dietary fats and heart disease began in the 1950s with the pioneering work of Ancel Keys and his colleagues. The Seven Countries Study, which compared dietary patterns and coronary heart disease (CHD) rates across diverse populations, revealed a strong positive correlation between total saturated fat intake and CHD mortality (Keys, 1970, cited in Maki et al., 2018; Lenighan et al., 2019). This observation gave rise to the “diet–heart hypothesis”: that reducing dietary saturated fat and cholesterol would lower serum cholesterol and, consequently, reduce CHD risk.

At the same time, controlled feeding studies demonstrated that replacing saturated fatty acids (SFAs) with polyunsaturated fatty acids (PUFAs) — primarily linoleic acid from vegetable oils — produced a significant reduction in total and LDL-cholesterol (Mensink et al., 2003, cited in Kris-Etherton & Fleming, 2015). This cholesterol-lowering effect was seen as direct evidence for the benefit of PUFA-rich oils. As a result, public health authorities, including the American Heart Association (AHA), began issuing recommendations to replace solid fats (butter, lard, tallow) with liquid vegetable oils (Page et al., 1957, 1961, cited in Jackson, 2024).

This message was widely adopted. The food industry responded by promoting margarines and shortenings made from partially hydrogenated vegetable oils, which were rich in LA. Between 1909 and 1999, LA intake in the United States increased more than threefold, driven largely by soybean oil consumption (Blasbalg et al., 2011, cited in Berkowitz, 2026). The prevailing view during this era was that LA was unambiguously beneficial.

3.2. The Emergence of the Pro-Inflammatory Hypothesis (1980s–2000s)

By the 1980s and 1990s, a counter-narrative began to take shape. Researchers noted that arachidonic acid (AA), derived from LA, is the primary precursor for a family of potent signalling molecules — eicosanoids — many of which have pro-inflammatory properties (e.g., prostaglandin E2, leukotriene B4). In contrast, eicosanoids derived from the omega-3 fatty acids EPA and DHA (e.g., resolvins, protectins) are predominantly anti-inflammatory (Innes & Calder, 2018; Djuricic & Calder, 2021).

This biochemical observation led to a simple and compelling hypothesis: a high dietary intake of LA, by increasing tissue AA levels, would promote chronic, low-grade inflammation and thereby increase the risk of inflammatory diseases, including cardiovascular disease, arthritis, and even cancer. This idea was popularised by several researchers, most notably Artemis Simopoulos, who argued that modern Western diets have an excessive and harmful omega-6 to omega-3 ratio (Simopoulos, 1999, 2008, cited in Berkowitz, 2026). She and others recommended not only increasing omega-3 intake but also actively reducing LA consumption, with some suggesting an optimal ratio as low as 1:1 or 2:1 (Simopoulos et al., 1999, cited in Berkowitz, 2026; Innes & Calder, 2018).

This hypothesis resonated with both the scientific community and the public because it offered a unified mechanism linking diet, inflammation, and chronic disease. It also gained support from observational studies that reported associations between high LA intake and certain cancers, though these findings were inconsistent (Retterstøl & Rosqvist, 2024; Sadeghi et al., 2025). During this period, French national guidelines explicitly recommended limiting LA intake to no more than 4% of total energy (Legrand et al., 2011, cited in Wu et al., 2017; Petersen et al., 2024), diverging from AHA and WHO recommendations.

3.3. Conflicting Evidence from Randomised Controlled Trials: The Sydney and Minnesota Reanalyses

The debate intensified in the 2010s following the publication of two highly controversial re-analyses of decades-old randomised controlled trials.

The Sydney Diet Heart Study (1966–1973) was a secondary prevention trial in which men with recent coronary events were randomly assigned to either a control diet or a diet rich in LA (from safflower oil and margarine) (Woodhill et al., 1978, cited in Ramsden et al., 2013). The original published results showed no significant benefit. However, in 2013, Ramsden and colleagues recovered unpublished data and re-analysed the trial. They reported that the LA-rich diet was associated with a higher risk of all-cause mortality, cardiovascular mortality, and CHD mortality (Ramsden et al., 2013, cited in Jackson, 2024; Maki et al., 2018). A similar conclusion emerged from a re-analysis of the Minnesota Coronary Survey (1968–1973), which found no evidence of benefit and a trend toward increased mortality in the group receiving LA-rich vegetable oils (Ramsden et al., 2016, cited in Retterstøl, 2024).

These findings were widely publicised and fuelled the narrative that LA may be harmful. However, subsequent critical evaluations pointed to major methodological limitations in both re-analyses. Most importantly, the intervention margarines used in the Sydney trial were high in industrial trans fatty acids, which are now known to be powerfully atherogenic and pro-inflammatory (Calder, 2013, cited in Maki, 2018; Virtanen, 2018, cited in Jackson, 2024). The Minnesota trial also had serious design flaws, including short duration, high drop-out rates, and intermittent treatment administration (Sacks et al., 2017, cited in Maki, 2018). Furthermore, both trials were conducted in an era when background medical therapy (statins, antihypertensives) was minimal or absent, limiting their relevance to modern clinical practice (Maki et al., 2018).

3.4. Recent Large-Scale Syntheses: Meta-Analyses and Pooled Analyses

In response to the confusion created by the re-analyses of old trials, several large-scale meta-analyses and individual-level pooled analyses have been conducted using contemporary data and rigorous methodology.

A 2014 meta-analysis by Farvid et al., which included 13 prospective cohort studies with over 310,000 participants, found that the highest category of LA intake was associated with a 15% lower risk of CHD events and a 21% lower risk of CHD death, compared with the lowest category (Farvid et al., 2014, cited in Jackson, 2024; Maki, 2018). A 2017 Cochrane review concluded that reducing SFA intake by replacing it with PUFA (predominantly LA) reduced the risk of cardiovascular events by 17% (Hooper et al., 2017, cited in Lenighan, 2019).

The most definitive evidence comes from the Fatty Acids and Outcomes Research Consortium (FORCE), an international collaboration that harmonised individual-level data from 30 prospective cohort studies across 13 countries. In this analysis, higher circulating levels of LA (reflecting dietary intake) were associated with a 7% lower risk of total CVD, a 22% lower risk of CVD mortality, and a 12% lower risk of ischemic stroke, comparing the 90th versus the 10th percentile of LA levels (Marklund et al., 2019, cited in Jackson, 2024; Maki, 2018). A separate FORCE analysis of 20 cohorts found that higher LA levels were associated with a 35% lower risk of incident type 2 diabetes (Wu et al., 2017, cited in Jackson, 2024).

Taken together, these contemporary syntheses have largely (though not entirely) settled the debate over the association of LA with cardiovascular and metabolic outcomes: higher LA intake and status are consistently associated with lower risk. The question of whether LA influences inflammation — the original biochemical concern — has been addressed by systematic reviews of RCTs showing no adverse effect of LA on inflammatory markers such as C-reactive protein, IL-6, or TNF-α (Johnson & Fritsche, 2012, cited in Innes, 2018; Su et al., 2017, cited in Petersen, 2024; Maki et al., 2018). Nevertheless, lingering controversies remain, particularly regarding the relationship between omega-6 PUFA and certain cancers (see Section 4.4) and the potential for high LA intake to suppress endogenous EPA synthesis (see Section 7). These unresolved questions continue to shape the research agenda and fuel ongoing scientific debate.

4. Contemporary Observational Evidence: Omega-6 Fatty Acids and Cardiometabolic Health

Observational studies — particularly large prospective cohorts and nested case-control studies — have played a central role in shaping our understanding of the relationship between omega-6 fatty acids and chronic disease risk. Unlike randomised controlled trials (RCTs), observational studies cannot prove causation, but they can detect associations that persist after adjusting for potential confounders, generate hypotheses, and inform public health guidelines when consistent across diverse populations (Maki et al., 2018). Importantly, for fatty acids like linoleic acid (LA) that cannot be synthesised endogenously, circulating and tissue biomarkers (e.g., in plasma phospholipids, cholesterol esters, erythrocytes, or adipose tissue) provide objective measures of dietary exposure, free from the recall and reporting biases that affect food-frequency questionnaires (Wu et al., 2017; Marklund et al., 2019).

This section summarises key findings from meta-analyses and individual-level pooled analyses of prospective cohort studies, focusing on cardiovascular disease (CVD), type 2 diabetes (T2DM), and mortality outcomes. Findings related to cancer are briefly addressed, given the heterogeneity of results.

4.1. Linoleic Acid and Cardiovascular Disease Risk

Forest plot showing relative risk of CVD for highest vs lowest intake of omega-6 PUFAs, with pooled RR 0.92, 95% CI 0.89–0.96

Meta-analysis of prospective cohort studies: dietary omega-6 intake and risk of cardiovascular disease

Higher dietary intake of omega-6 polyunsaturated fatty acids is associated with a modest but statistically significant reduction in the risk of incident cardiovascular disease. Data from 93 distinct cohorts. Adapted from Sadeghi et al. (2025), Figure 2c. CC BY 4.0

A substantial body of evidence from prospective cohort studies supports an inverse association between LA intake or status and CVD risk.

A seminal meta-analysis by Farvid et al. (2014), which included 13 prospective cohort studies with over 310,000 participants and nearly 13,000 coronary heart disease (CHD) events, found that the highest category of dietary LA intake was associated with a 15% lower risk of CHD events (RR: 0.85; 95% CI: 0.78, 0.92) and a 21% lower risk of CHD death (RR: 0.79; 95% CI: 0.71, 0.89), compared with the lowest intake category (Farvid et al., 2014, cited in Maki, 2018; Jackson, 2024). A dose-response analysis indicated that each 5% of energy increase in LA (replacing saturated fatty acids) was associated with a 9% lower risk of CHD events and a 13% lower risk of CHD death (Farvid et al., 2014).

More recently, the Fatty Acids and Outcomes Research Consortium (FORCE) conducted an individual-level pooled analysis of 30 prospective cohort studies from 13 countries, including 68,659 participants and 15,198 incident CVD events (Marklund et al., 2019). Using harmonised biomarker measurements of LA (in phospholipids, total plasma, cholesterol esters, or adipose tissue), the investigators reported that a higher LA level was significantly associated with a lower risk of total CVD (hazard ratio [HR] per interquintile range: 0.93; 95% CI: 0.88, 0.99), CVD mortality (HR: 0.78; 95% CI: 0.70, 0.85), and ischemic stroke (HR: 0.88; 95% CI: 0.79, 0.98). The association with total CHD was of similar magnitude but did not reach statistical significance (HR: 0.94; 95% CI: 0.88, 1.00). These findings were consistent across different lipid compartments and in sensitivity analyses, supporting the robustness of the inverse association (Marklund et al., 2019; Jackson, 2024).

In a separate meta-analysis of 18 prospective cohort studies, Li et al. (2020) found that higher dietary LA intake was associated with an 8% lower risk of CVD mortality (RR: 0.92; 95% CI: 0.88, 0.96), and that each 5% increase in energy from LA was associated with a 7% lower risk (RR: 0.93; 95% CI: 0.91, 0.95). Similar inverse associations were observed for circulating and tissue LA biomarkers (Li et al., 2020, cited in Jackson, 2024; Sadeghi et al., 2025).

These results are consistent with the earlier pooled analysis by Jakobsen et al. (2009) of 11 American and European cohorts, which demonstrated that replacing 5% of energy from saturated fatty acids with PUFA (predominantly LA) was associated with a 13% reduction in coronary events and a 26% reduction in coronary death (Jakobsen et al., 2009, cited in Maki, 2018; Lenighan, 2019).

4.2. Arachidonic Acid: No Evidence of Harm

Despite theoretical concerns that arachidonic acid (AA) — a direct metabolite of LA and the precursor of pro-inflammatory eicosanoids — might increase CVD risk, observational studies have consistently failed to support such an association.

In the FORCE analysis, higher circulating AA levels were not significantly associated with total CVD (HR: 0.95; 95% CI: 0.90, 1.01), CVD mortality (HR: 0.94; 95% CI: 0.86, 1.02), total CHD (HR: 0.99; 95% CI: 0.94, 1.04), or ischemic stroke (HR: 0.99; 95% CI: 0.90, 1.10) (Marklund et al., 2019). Similarly, Li et al. (2020) reported a null association between AA biomarkers and CVD mortality (RR: 0.94; 95% CI: 0.84, 1.06) (Li et al., 2020, cited in Sadeghi, 2025). A systematic review by Chowdhury et al. (2014) also found no significant association between circulating AA levels and coronary risk (Chowdhury et al., 2014, cited in Retterstøl, 2024).

Thus, despite AA’s role as a substrate for pro-inflammatory eicosanoids, population-level data do not support the hypothesis that higher AA status is detrimental for cardiovascular health. Some analyses have even suggested a modest inverse association (e.g., for total CVD, comparing extreme quintiles: HR: 0.92; 95% CI: 0.86, 0.99), but the overall evidence indicates that AA is at worst neutral (Marklund et al., 2019).

4.3. Linoleic Acid and Type 2 Diabetes Risk

The relationship between LA and incident type 2 diabetes (T2DM) has been examined in multiple prospective cohorts, with consistent evidence of a protective effect.

A large pooled analysis by Wu et al. (2017) within the FORCE consortium included 39,740 adults from 20 prospective cohort studies across 10 countries, with 4347 incident T2DM cases during 366,073 person-years of follow-up. After multivariable adjustment, higher LA levels (measured as a percentage of total fatty acids in various lipid compartments) were associated with a substantially lower risk of T2DM. The relative risk (RR) per interquintile range was 0.65 (95% CI: 0.60, 0.72; P < 0.0001). The association was linear and consistent across different compartments (phospholipids, plasma, cholesterol esters, and adipose tissue), and no significant heterogeneity was observed by age, sex, BMI, race, or omega-3 PUFA levels (Wu et al., 2017; Jackson, 2024).

In contrast, AA levels were not significantly associated with T2DM risk overall (RR per interquintile range: 0.96; 95% CI: 0.88, 1.05; P = 0.38) (Wu et al., 2017). A subsequent systematic review and dose-response meta-analysis by Mousavi et al. (2021) confirmed that each 5% increment in energy from LA was associated with a 10% lower risk of T2DM, and each standard deviation increase in LA biomarkers corresponded to a 15% lower risk (Mousavi et al., 2021, cited in Retterstøl, 2024).

These findings from observational studies align with results from RCTs showing beneficial effects of PUFA on glucose-insulin homeostasis (Imamura et al., 2016, cited in Maki, 2018), providing biological plausibility for a causal relationship.

4.4. Mortality Outcomes and Heterogeneity by Cancer Site

The relationship between omega-6 PUFA and mortality has been examined in several large meta-analyses.

A comprehensive meta-analysis by Li et al. (2020), which included 38 studies with 44 prospective cohorts (811,069 participants), reported that higher dietary LA intake was associated with a modestly lower risk of all-cause mortality (RR: 0.88; 95% CI: 0.83, 0.94 for highest vs. lowest intake), CVD mortality (RR: 0.87; 95% CI: 0.82, 0.92), and cancer mortality (RR: 0.89; 95% CI: 0.85, 0.94) (Li et al., 2020, cited in Sadeghi, 2025; Jackson, 2024). Similarly, a meta-analysis by Sadeghi et al. (2025), which included 150 publications (230 studies, 93 distinct cohorts), found that higher dietary intake and circulating levels of omega-6 PUFAs were associated with lower risks of CVD, all-cause mortality, and cancer incidence overall, though with notable exceptions by cancer site.

Specifically, while higher omega-6 intake was associated with lower risk of lung cancer (RR: 0.84; 95% CI: 0.71, 0.99) and prostate cancer (RR: 0.86; 95% CI: 0.76, 0.96), it was associated with a higher risk of endometrial cancer (RR: 1.15; 95% CI: 1.01, 1.30) and ovarian cancer (RR: 1.15; 95% CI: 1.01, 1.30) (Sadeghi et al., 2025). These site-specific differences suggest that the relationship between omega-6 PUFA and cancer is not uniform and may depend on the hormonal sensitivity of the tissue, the balance of other dietary components, or the specific omega-6 fatty acid involved (e.g., LA vs. AA). The certainty of evidence for cancer outcomes was graded as low to very low in many analyses, highlighting the need for further research (Sadeghi et al., 2025; Retterstøl & Rosqvist, 2024).

4.5. Subgroup Analyses and Effect Modification

One of the strengths of large pooled analyses is the ability to explore whether associations differ by population characteristics. In the FORCE studies, the inverse associations between LA and CVD or T2DM were not significantly modified by age, sex, race, BMI, omega-3 PUFA levels, statin use, aspirin use, or common genetic variants in the FADS gene (Marklund et al., 2019; Wu et al., 2017). This suggests that the beneficial association of LA is broadly applicable across diverse demographic and clinical subgroups.

However, Sadeghi et al. (2025) noted that protective associations were more pronounced in cohort studies and in individuals without pre-existing health conditions (i.e., primary prevention settings), whereas associations were weaker or null in populations with established disease. This could reflect reverse causation, residual confounding, or true effect modification by baseline health status.

In summary, large-scale observational evidence consistently demonstrates that higher LA intake and status are associated with lower risks of CVD, CVD mortality, and T2DM, with neutral or weakly protective associations for AA. The relationship with cancer appears more complex and site-specific, underscoring the need for caution in making generalised statements about omega-6 PUFA and cancer risk. As with all observational data, residual confounding cannot be entirely excluded, but the consistency across cohorts, dose-response relationships, and concordance with mechanistic and RCT evidence strengthens the case for a causal role of LA in cardiometabolic protection.

5. Impact on Risk Markers: Evidence from Randomised Controlled Trials

Randomised controlled trials (RCTs) provide the highest level of evidence for causality in nutrition research. While large RCTs with hard clinical endpoints (e.g., myocardial infarction, stroke, death) are resource-intensive and often logistically challenging for dietary fat modification, a substantial number of controlled feeding studies have examined the effects of omega-6 PUFAs — particularly linoleic acid (LA) — on established surrogate markers of cardiometabolic risk. These markers include atherogenic lipoproteins (LDL-cholesterol, non-HDL-cholesterol, apolipoprotein B), blood pressure, glycaemic control (fasting glucose, insulin, HbA1c, HOMA-IR), and inflammatory markers (C-reactive protein, cytokines). This section summarises the key findings from meta-analyses of RCTs, with an emphasis on studies that replaced saturated fatty acids (SFAs) or carbohydrates with PUFA-rich oils.

5.1. Lipid and Lipoprotein Profile

The most robust and consistent effect of dietary LA is on circulating lipid and lipoprotein concentrations. A large body of evidence from controlled feeding trials shows that replacing SFAs with PUFAs (predominantly LA) reduces total cholesterol, LDL-cholesterol, and the total cholesterol to HDL-cholesterol ratio in a dose-dependent manner.

A landmark systematic review and meta-regression analysis by Mensink (2016), commissioned by the World Health Organization, included 84 controlled trials with 2,353 participants. The analysis estimated the changes in lipoprotein lipids for each 1% of dietary energy exchanged between macronutrients. When 1% of energy from SFAs was isocalorically replaced with PUFAs, the following mean changes were observed:

  • Total cholesterol: −0.064 mmol/L (95% CI: −0.070, −0.058)
  • LDL-cholesterol: −0.055 mmol/L (95% CI: −0.061, −0.050)
  • Total cholesterol to HDL-cholesterol ratio: −0.034 (95% CI: −0.040, −0.028)
  • Triglycerides: −0.010 mmol/L (95% CI: −0.014, −0.007)

Replacing SFAs with MUFAs produced similar but slightly smaller improvements, whereas replacing SFAs with carbohydrates lowered total and LDL-cholesterol but also lowered HDL-cholesterol and increased triglycerides (Mensink, 2016, cited in Maki et al., 2018; Retterstøl & Rosqvist, 2024).

A Cochrane systematic review by Hooper et al. (2020) confirmed that reducing SFA intake for at least two years reduced the risk of combined cardiovascular events by 17%, and that the effect was most pronounced in trials where SFAs were replaced with PUFAs (Hooper et al., 2020, cited in Retterstøl, 2024; Lenighan, 2019). The lipid-lowering effects of PUFA were also evident in a meta-analysis of 49 RCTs of pharmacological and dietary interventions, which demonstrated that each 1 mmol/L reduction in LDL-cholesterol was associated with a 23% reduction in major vascular events, regardless of whether the reduction was achieved by statins or diet (Silverman et al., 2016, cited in Petersen, 2024).

Specific RCTs using individual LA-rich oils have consistently shown lipid improvements. For example, Maki et al. (2015) compared corn oil (rich in LA) with extra-virgin olive oil in adults with elevated LDL-cholesterol. After 21 days of consuming 4 tablespoons per day, LDL-cholesterol decreased by 10.9% in the corn oil group compared with 3.5% in the olive oil group (Maki et al., 2015, cited in Maki, 2018). Similarly, a network meta-analysis of 13 oils and solid fats by Schwingshackl et al. (2018) concluded that replacing 10% of energy from butter with safflower, sunflower, canola, olive, flaxseed, corn, or soybean oil lowered LDL-cholesterol by 10–16 mg/dL (Schwingshackl et al., 2018, cited in Petersen, 2024).

Importantly, the LDL-cholesterol lowering effect of LA is not merely a consequence of reducing SFA intake; LA itself has independent cholesterol-lowering properties, mediated in part by upregulation of hepatic LDL receptor expression and increased bile acid synthesis via activation of liver X receptor alpha (LXRa) and cholesterol 7α-hydroxylase (CYP7A1) (Djuricic & Calder, 2021; Froyen & Burns-Whitmore, 2020, cited in Djuricic, 2021).

5.2. Blood Pressure

The effects of dietary fatty acids on blood pressure are more modest and less consistent than their effects on blood lipids. Most RCTs have found that replacing SFAs with MUFAs or PUFAs has little to no impact on systolic or diastolic blood pressure in healthy normotensive individuals.

A Cochrane review by Hooper et al. (2018) of RCTs comparing higher versus lower intake of n-6 PUFAs (minimum duration 12 months) found no clinically relevant effects on systolic or diastolic blood pressure, though this conclusion was based on only two studies (Hooper et al., 2018, cited in Petersen, 2024). In the Dietary Intervention and VAScular function (DIVAS) trial, replacing approximately 9.5% of energy from SFAs with MUFAs or n-6 PUFAs for 16 weeks did not affect 24-hour systolic or diastolic blood pressure in individuals at moderate CVD risk. However, replacing SFAs with MUFAs did reduce nighttime systolic blood pressure by 4.9 mm Hg (P = 0.019) (Vafeiadou et al., 2015, cited in Maki, 2018; Petersen, 2024).

In the OmniHeart randomised controlled feeding trial, a higher-fat, MUFA-rich diet (48% carbohydrate, 37% fat, 21% MUFA, 10% PUFA, 6% SFA) modestly lowered systolic blood pressure by 1.3 mm Hg (P = 0.005) and diastolic pressure by 0.8 mm Hg (P = 0.02) after 6 weeks compared with a carbohydrate-rich diet (58% carbohydrate, 27% fat) (Appel et al., 2005, cited in Maki, 2018; Petersen, 2024). These effects were attributed primarily to MUFA rather than PUFA.

Overall, the evidence suggests that while MUFAs may have a small blood pressure-lowering effect when replacing carbohydrates, n-6 PUFAs have a neutral effect on blood pressure (Maki et al., 2018; Retterstøl & Rosqvist, 2024).

5.3. Glycaemic Control and Insulin Sensitivity

The relationship between dietary PUFAs and glucose-insulin homeostasis has been examined in several meta-analyses of RCTs, with generally favourable findings.

A systematic review and meta-analysis by Imamura et al. (2016), which included 102 controlled feeding trials, assessed the effects of replacing SFAs or carbohydrates with PUFAs (predominantly LA) on markers of glucose metabolism. The authors reported that substituting 5% of energy from SFAs with PUFAs decreased fasting glucose by a mean of 0.04 mmol/L (95% CI: −0.07, −0.01), lowered HbA1c by 0.15% (95% CI: −0.23%, −0.06%), and reduced HOMA-IR (a measure of insulin resistance) by 4.1% (95% CI: −6.4%, −1.6%) (Imamura et al., 2016, cited in Maki, 2018; Retterstøl, 2024). PUFA substitution also improved insulin secretion capacity.

A subsequent meta-analysis by Wanders et al. (2019) of 13 RCTs confirmed that plant-derived PUFAs (mainly LA) lowered fasting insulin and HOMA-IR, though the effect on fasting glucose was not statistically significant (Wanders et al., 2019, cited in Retterstøl, 2024). In contrast, a more recent meta-analysis by Lytrivi et al. (2023) failed to find significant effects of replacing SFAs with MUFAs or PUFAs on insulin sensitivity, but the authors noted that many of the included trials were short-term and that longer studies are needed (Lytrivi et al., 2023, cited in Petersen, 2024).

Despite some inconsistencies, the preponderance of evidence from RCTs indicates that replacing SFAs or refined carbohydrates with PUFA improves glycaemic control and reduces insulin resistance. These findings are mechanistically supported by studies showing that LA and its metabolites can activate peroxisome proliferator-activated receptor gamma (PPARγ), a nuclear receptor that enhances insulin sensitivity and glucose uptake in adipose tissue and skeletal muscle (Berkowitz et al., 2026; Belury, 2023, cited in Jackson, 2024).

5.4. Inflammatory Markers

One of the most persistent concerns about dietary LA is its potential to promote systemic inflammation due to its role as a precursor to arachidonic acid (AA) and subsequent pro-inflammatory eicosanoids (e.g., PGE2, LTB4). However, RCT evidence has consistently failed to support this concern.

A systematic review and meta-analysis by Su et al. (2017) included 30 RCTs (duration 4 to 24 weeks) and found that higher intake of LA did not increase tumour necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), C-reactive protein (CRP), or adiponectin (Su et al., 2017, cited in Petersen, 2024). Similarly, an earlier systematic review by Johnson and Fritsche (2012) examined 15 RCTs in healthy adults and concluded that there was “virtually no evidence” to support the claim that increased consumption of LA promotes inflammation (Johnson & Fritsche, 2012, cited in Innes, 2018; Maki, 2018).

In controlled feeding studies using LA-rich oils, no adverse effects on inflammatory markers have been observed. For example, a study by Bjermo et al. (2012) compared a diet high in n-6 PUFAs (mainly LA) with a diet high in SFAs in individuals with abdominal obesity. Despite significant differences in LA intake, there were no differences in markers of inflammation (CRP, IL-6, TNF-α) between the groups (Bjermo et al., 2012, cited in Retterstøl, 2024). Epidemiological studies have even shown inverse associations between LA levels and inflammatory markers, with higher LA being associated with lower CRP and IL-6 concentrations (Ferrucci et al., 2006, cited in Innes, 2018).

The lack of a pro-inflammatory effect of LA in human RCTs can be explained by several factors. First, the conversion of LA to AA is very limited in humans (< 0.2% of dietary LA is converted to AA) and is not increased by higher LA intakes (Hussein et al., 2005, cited in Innes, 2018; Rett & Whelan, 2011, cited in Jackson, 2024). Second, AA itself gives rise to both pro-inflammatory and anti-inflammatory/pro-resolving mediators (e.g., lipoxin A4), and the net effect in healthy individuals is not pro-inflammatory (Innes & Calder, 2018). Third, the endogenous antioxidant defence systems are generally sufficient to prevent significant lipid peroxidation from moderate LA intakes (Birben et al., 2012, cited in Petersen, 2024). Thus, the available RCT evidence does not support the hypothesis that dietary LA increases inflammation.

5.5. Markers of Oxidative Stress

A related concern is that PUFA, because of their multiple double bonds, are more susceptible to oxidation, potentially increasing oxidative stress and the formation of oxidised LDL particles. However, RCTs have generally not found that increased LA intake raises markers of oxidative stress.

A systematic review by da Silva et al. (2017) of clinical trials examining the effect of dietary fats on F₂-isoprostanes (a validated marker of lipid peroxidation) concluded that replacing SFAs with PUFAs does not increase F₂-isoprostane levels (da Silva et al., 2017, cited in Petersen, 2024). In RCTs specifically using soybean oil (∼51% LA), most studies found no increase in oxidised LDL or other oxidative markers (e.g., thiobarbituric acid reactive substances, malondialdehyde), and one study even reported an increase in trolox equivalent antioxidant capacity (Utarwuthipong et al., 2009; Karupaiah et al., 2016; Wu et al., 2022, cited in Petersen, 2024).

The lack of adverse oxidative effects is likely due to the presence of endogenous antioxidants (e.g., vitamin E, coenzyme Q10) in whole foods and oils, and the fact that the body’s antioxidant defences are generally sufficient to handle moderate increases in PUFA intake (Petersen et al., 2024). However, caution is warranted with improper storage or repeated deep-frying of oils, which can generate harmful oxidation products (see Section 9 in the original NNR context).

5.6. Summary of RCT Evidence

In summary, RCTs provide strong and consistent evidence that replacing SFAs with LA-rich PUFAs improves the atherogenic lipid profile (lower LDL-cholesterol, total cholesterol, and total/HDL ratio). The effects on blood pressure are neutral for LA, though MUFAs may have a small benefit. Glycaemic control and insulin sensitivity are modestly improved by PUFA replacement. Importantly, neither inflammatory markers nor markers of oxidative stress are increased by LA intakes within the recommended range (5–10% of energy). These findings support the use of LA-rich vegetable oils as part of a cardioprotective dietary pattern, particularly when they replace SFAs or refined carbohydrates.

6. The Central Question: Omega-6 Fatty Acids and Inflammation

Perhaps no single issue has generated more controversy around omega-6 polyunsaturated fatty acids than their alleged pro-inflammatory effects. This concern stems from a seemingly straightforward biochemical narrative: dietary linoleic acid (LA) is converted to arachidonic acid (AA), which is then metabolised by cyclooxygenase (COX) and lipoxygenase (LOX) enzymes into eicosanoids — prostaglandins, leukotrienes, and thromboxanes — many of which are potent mediators of inflammation, pain, and platelet aggregation (Innes & Calder, 2018; Djuricic & Calder, 2021). From this perspective, a high intake of LA would be expected to increase tissue AA levels, enhance the production of pro-inflammatory eicosanoids, and thereby promote chronic inflammatory diseases such as atherosclerosis, rheumatoid arthritis, asthma, and inflammatory bowel disease.

However, as is often the case in nutrition science, the simple biochemical narrative does not align with the more complex reality revealed by human studies. This section examines the evidence for and against the pro-inflammatory hypothesis of omega-6 fatty acids, drawing on mechanistic studies, observational data, and — most importantly — randomised controlled trials (RCTs) in humans.

6.1. The Biochemical Basis of Concern: Eicosanoid Pathways

Biochemical cascade showing conversion of arachidonic acid to prostaglandins, thromboxanes, leukotrienes, lipoxins and other eicosanoids via COX, LOX and CYP enzymes

Overview of the cyclooxygenase, lipoxygenase and cytochrome P450 pathways of arachidonic acid metabolism

Arachidonic acid serves as the precursor for a diverse family of eicosanoids with both pro-inflammatory (e.g., PGE2, LTB4) and pro-resolving (e.g., lipoxin A4) functions. Adapted from Djuricic & Calder (2021), Figure 4. CC BY 4.0

The arachidonic acid cascade is one of the most extensively studied signalling systems in human physiology. When cells are activated by inflammatory stimuli (e.g., bacterial lipopolysaccharide, cytokines, or physical injury), phospholipase A2 (cPLA2) is recruited to the cell membrane and cleaves AA from the sn-2 position of membrane phospholipids (Innes & Calder, 2018). Free AA is then rapidly oxygenated by three main enzyme families:

  • Cyclooxygenases (COX-1 and COX-2): produce prostaglandins (e.g., PGE2, PGD2, PGF₂α, PGI₂) and thromboxane A2 (TXA2). PGE2 is a classic pro-inflammatory mediator causing vasodilation, increased vascular permeability, fever, and pain sensitisation. TXA2 promotes platelet aggregation and vasoconstriction.
  • 5-Lipoxygenase (5-LOX): produces leukotriene A4 (LTA4), which is further converted to LTB4 (a potent chemoattractant for neutrophils) and the cysteinyl leukotrienes LTC4, LTD4, and LTE4 (which cause bronchoconstriction and increased vascular permeability).
  • 12- and 15-Lipoxygenases: produce a range of hydroxy-eicosatetraenoic acids (HETEs) and, in conjunction with 5-LOX, contribute to the synthesis of lipoxins (LXA4, LXB4) — which, in contrast to the other AA-derived mediators, possess anti-inflammatory and pro-resolving properties (Innes & Calder, 2018; Djuricic & Calder, 2021).

The pro-inflammatory potential of the AA-derived eicosanoid pathway is indisputable. Drugs that inhibit COX (non-steroidal anti-inflammatory drugs, NSAIDs) or block leukotriene receptors are widely and effectively used to treat inflammatory conditions. Therefore, it is not unreasonable to hypothesise that increasing the availability of the substrate (AA) might increase the production of pro-inflammatory eicosanoids.

6.2. What Actually Happens in Humans? Evidence from Supplementation Trials

The critical question is whether increasing LA intake — and by extension, AA status — actually leads to increased eicosanoid production and elevated inflammatory markers in healthy humans. The available evidence from RCTs provides a clear and consistent answer: no.

A systematic review and meta-analysis by Johnson and Fritsche (2012) examined 15 RCTs in healthy adults that tested the effect of increased dietary LA on a wide range of inflammatory markers, including C-reactive protein (CRP), fibrinogen, plasminogen activator inhibitor-1 (PAI-1), cytokines (TNF-α, IL-6, IL-1β), and soluble adhesion molecules (sICAM-1, sVCAM-1, sE-selectin). The authors concluded that there was “virtually no evidence” from RCTs to support the claim that increased consumption of LA promotes inflammation in healthy adults (Johnson & Fritsche, 2012, cited in Innes, 2018; Maki, 2018).

A more recent meta-analysis by Su et al. (2017), which included 30 RCTs with durations ranging from 4 to 24 weeks, confirmed these findings. Higher LA intake did not significantly increase TNF-α (standardised mean difference [SMD]: −0.01; 95% CI: −0.19, 0.17), IL-6 (SMD: 0.11; 95% CI: −0.07, 0.29), CRP (SMD: 0.09; 95% CI: −0.05, 0.24), or adiponectin (SMD: 0.17; 95% CI: −0.17, 0.50) (Su et al., 2017, cited in Petersen, 2024). Similarly, a systematic review by Rett and Whelan (2011) found that increasing dietary LA intake (from approximately 2–3% to 7–8% of energy) did not increase AA levels in plasma phospholipids or erythrocytes, suggesting that the conversion of LA to AA is tightly regulated and saturable (Rett & Whelan, 2011, cited in Jackson, 2024; Innes, 2018).

Even studies that directly supplemented AA (rather than LA) at doses up to 1,500 mg/day have failed to demonstrate consistent pro-inflammatory effects. In a systematic review of AA supplementation trials, Calder et al. (2019) concluded that increasing AA intake had no adverse effects on platelet aggregation, immune function, or inflammatory markers (Calder et al., 2019, cited in Djuricic, 2021). In some studies, AA supplementation was associated with improved cognitive performance and increased muscle strength, with no increase in inflammatory cytokines (Djuricic & Calder, 2021).

6.3. Observational Evidence: Higher LA, Lower Inflammation

Observational studies have provided an additional layer of evidence that runs counter to the pro-inflammatory hypothesis. Rather than finding positive associations between LA status and inflammation, several large cross-sectional studies have reported inverse associations.

In an analysis of 405 healthy men and 454 healthy women from the Health Professionals Follow-Up Study and the Nurses’ Health Study, Pischon et al. (2003) found that dietary LA intake was not significantly associated with CRP, IL-6, or soluble TNF receptors. However, the lowest levels of inflammation were observed in subjects with the highest intakes of both omega-3 and omega-6 fatty acids (Pischon et al., 2003, cited in Innes, 2018; Maki, 2018).

A study of 1,123 Italian adults by Ferrucci et al. (2006) examined the relationship between plasma PUFA concentrations and eight markers of inflammation. Subjects in the lowest quartile of plasma LA had the highest concentrations of the pro-inflammatory markers IL-6 and CRP and the lowest concentrations of the anti-inflammatory cytokines IL-10 and TGF-β. Plasma LA was inversely associated with IL-6 and CRP, and plasma AA was inversely associated with IL-6 and IL-1 receptor antagonist (Ferrucci et al., 2006, cited in Innes, 2018; Maki, 2018).

In the Kuopio Ischaemic Heart Disease Risk Factor Study, Virtanen et al. (2018) reported that higher serum LA concentrations were associated with lower hs-CRP concentrations in a dose-responsive manner. The odds ratio for elevated hs-CRP in the highest compared with the lowest LA quartile was 0.47 (95% CI: 0.25, 0.87; P-trend = 0.01) (Virtanen et al., 2018, cited in Maki, 2018).

Thus, far from promoting inflammation, higher LA status is consistently associated with lower levels of systemic inflammatory markers in population studies. This does not prove causation, but it is difficult to reconcile with the hypothesis that LA is pro-inflammatory.

6.4. Reconciling the Paradox: Why the Simple Model Fails

Several mechanisms explain why the simple biochemical model of LA → AA → pro-inflammatory eicosanoids does not translate into clinical inflammation in humans.

First, the conversion of LA to AA is very limited and not dose-dependent. Stable isotope tracer studies in humans have shown that less than 0.2% of dietary LA is converted to AA (Hussein et al., 2005, cited in Innes, 2018). Moreover, increasing LA intake over a wide range (from 2% to 12% of energy) does not increase AA levels in blood or tissue lipids (Rett & Whelan, 2011, cited in Jackson, 2024). The endogenous synthesis pathway is saturated at low LA intakes, and AA levels are primarily determined by direct dietary AA intake, not by LA intake (Calder et al., 2019, cited in Djuricic, 2021).

Second, AA is not merely a precursor to pro-inflammatory mediators; it also gives rise to potent anti-inflammatory and pro-resolving mediators. The lipoxins (LXA4, LXB4) are derived from AA via the action of 15-LOX (in epithelial cells) followed by 5-LOX (in leukocytes), or vice versa. Lipoxins actively suppress neutrophil recruitment, promote macrophage clearance of apoptotic cells, and stimulate the resolution of inflammation (Serhan et al., 2008, cited in Innes, 2018; Djuricic & Calder, 2021). Thus, AA can be considered a “double-edged sword” — its metabolites can both initiate and resolve inflammation depending on the temporal and cellular context.

Third, the presence of omega-3 fatty acids (EPA and DHA) modulates eicosanoid metabolism. EPA and DHA compete with AA for incorporation into membrane phospholipids and for metabolism by COX and LOX enzymes. When EPA and DHA are present in sufficient amounts, they give rise to less potent pro-inflammatory eicosanoids (3-series prostaglandins and 5-series leukotrienes) and to specialised pro-resolving mediators (resolvins, protectins, maresins) that actively resolve inflammation (Calder, 2017, cited in Innes, 2018). The LA:ALA ratio may influence this balance, but the absolute intakes of both families are more important than the ratio (Jackson et al., 2024).

Fourth, LA itself may have direct anti-inflammatory effects through activation of PPARγ. As noted in Section 5.3, LA and its oxidised metabolites (e.g., 13-HODE) can activate PPARγ, a nuclear receptor that suppresses NF-κB-mediated transcription of pro-inflammatory cytokines and promotes the expression of anti-inflammatory genes (Berkowitz et al., 2026; Belury, 2023, cited in Jackson, 2024). This pathway may contribute to the lower inflammatory marker levels observed in individuals with higher LA status.

6.5. The Omega-6 to Omega-3 Ratio: An Unhelpful Distraction

The concept of an optimal dietary or tissue ratio of omega-6 to omega-3 fatty acids has been widely popularised, with recommendations ranging from 4:1 down to 1:1 (Simopoulos, 2008, cited in Berkowitz, 2026). However, this ratio has been severely criticised by many experts for several reasons.

First, the ratio is non-specific — it can be calculated using different sets of fatty acids (e.g., LA alone vs. total n-6 PUFA vs. including AA; ALA alone vs. EPA+DHA vs. total n-3 PUFA). Each choice yields a different number, and there is no standardisation.

Second, the same ratio can be achieved by very different absolute intakes. For example, a ratio of 10:1 could result from 20 g/day LA and 2 g/day ALA, or from 5 g/day LA and 0.5 g/day ALA. The absolute intakes of essential fatty acids are physiologically far more important than their ratio (Jackson et al., 2024).

Third, intervention studies have shown that changing the ratio while keeping absolute intakes constant produces inconsistent effects, whereas changing absolute intakes while keeping the ratio constant produces different outcomes (Goyens et al., 2006, cited in Jackson, 2024). Therefore, the ratio is neither clinically nor biologically informative.

Major health authorities, including the American Heart Association and the UK Food Standards Agency, have concluded that the omega-6 to omega-3 ratio should not be used as a dietary guidance metric (Harris, 2006, cited in Jackson, 2024; Stanley et al., 2007, cited in Jackson, 2024). Recommendations should focus on absolute intakes of LA (5–10% of energy), ALA (≥0.5% of energy), and preformed EPA and DHA (≥250–500 mg/day) rather than on a ratio.

6.6. Summary: Inflammation as a Non-Issue for Dietary LA

In conclusion, the long-standing concern that dietary LA promotes inflammation is not supported by human evidence. The biochemical pathway linking LA to AA and pro-inflammatory eicosanoids exists, but in healthy humans consuming typical Western diets, the conversion of LA to AA is minimal and not increased by higher LA intake. RCTs consistently show no increase in inflammatory markers when LA intake is increased within or even somewhat above the recommended range. Observational studies, if anything, suggest an inverse association between LA status and systemic inflammation. Arachidonic acid itself is not a simple “pro-inflammatory” fatty acid; it also produces pro-resolving mediators, and its net effect depends on the context of other dietary fatty acids (particularly EPA and DHA).

Thus, for the purpose of public health recommendations, concerns about inflammation should not be a barrier to encouraging the replacement of saturated fats with LA-rich vegetable oils as part of a heart-healthy dietary pattern (Maki et al., 2018; Petersen et al., 2024).

7. The n-6/n-3 Ratio Debate: Metabolic Competition and Its Clinical Relevance

Few concepts in fatty acid nutrition have been as widely adopted — and as fiercely criticised — as the dietary ratio of omega-6 to omega-3 polyunsaturated fatty acids (n-6/n-3 ratio). From popular health books to scientific reviews, the idea that a “balanced” ratio (often cited as 4:1, 2:1, or even 1:1) is essential for optimal health, and that the modern Western diet’s ratio of 15:1 or 20:1 is inherently harmful, has become almost an article of faith for many (Simopoulos, 2008, cited in Berkowitz, 2026; Innes & Calder, 2018). This belief is rooted in the well-established fact that linoleic acid (LA, n-6) and alpha-linolenic acid (ALA, n-3) compete for the same desaturation and elongation enzymes in the endogenous synthesis of long-chain polyunsaturated fatty acids (LC-PUFAs). However, as this section will demonstrate, the ratio has serious conceptual and practical limitations that severely limit its utility as a dietary guidance tool.

7.1. The Biochemical Basis of Competition: Shared Enzymatic Machinery

The metabolic pathways of n-6 and n-3 fatty acids converge at the level of desaturases. LA (18:2n-6) and ALA (18:3n-3) are both substrates for Δ6-desaturase (encoded by the FADS2 gene), the rate-limiting enzyme in the conversion of essential 18-carbon PUFAs to their longer-chain, more unsaturated derivatives (Djuricic & Calder, 2021; Gutierres et al., 2025). After initial desaturation, both families undergo elongation (by elongase 5, ELOVL5) and further desaturation (by Δ5-desaturase, FADS1). The products of these pathways are arachidonic acid (AA, 20:4n-6) from LA, and eicosapentaenoic acid (EPA, 20:5n-3) and docosahexaenoic acid (DHA, 22:6n-3) from ALA.

Because the desaturases have a higher affinity for ALA than for LA (some studies suggest a 2–3 fold preference), a high intake of LA could, in theory, competitively inhibit the conversion of ALA to EPA and DHA (Berkowitz et al., 2026; Jackson et al., 2024). This competition is the biochemical foundation for the argument that the n-6/n-3 ratio matters: a high ratio (i.e., much more LA than ALA) would suppress endogenous EPA synthesis, potentially leading to lower tissue levels of these cardioprotective LC-n-3 PUFAs.

7.2. Empirical Evidence: Does High LA Intake Reduce EPA/DHA Status?

Several controlled human studies have examined the effect of varying LA intake (while keeping ALA intake constant) on blood levels of EPA and DHA. The results confirm that high LA intake does reduce the conversion of ALA to EPA, but the magnitude of the effect is modest and the clinical significance is debated.

A well-known study by Liou et al. (2007) compared a low-LA diet (approximately 4% of energy from LA) with a high-LA diet (approximately 11% of energy from LA), while keeping ALA intake constant at about 1% of energy. After several weeks, the high-LA diet reduced plasma phospholipid EPA levels by approximately 50% (from about 1% to 0.5% of total fatty acids) compared with the low-LA diet (Liou et al., 2007, cited in Jackson, 2024). However, the absolute reduction was small (0.5 percentage points), and there was no significant effect on DHA levels.

A study by Goyens et al. (2006) compared two diets with the same LA:ALA ratio (7:1) but different absolute intakes: a low-PUFA diet (3% LA, 0.4% ALA) and a high-PUFA diet (7% LA, 1.1% ALA). Plasma phospholipid EPA increased by over 50% on the low-PUFA diet but by only 25% on the high-PUFA diet, demonstrating that the absolute amounts of these PUFAs are more important than their ratio (Goyens et al., 2006, cited in Jackson, 2024).

Ramsden et al. (2013) conducted a 12-week trial in which participants reduced LA intake from 7.4% to 2.4% of energy while keeping ALA constant. Erythrocyte EPA+DHA increased from 3.7% to 4.1% of total fatty acids — a modest absolute increase of 0.4 percentage points (Ramsden et al., 2013, cited in Jackson, 2024). Again, no effect on DHA was observed.

Thus, while high LA intake does modestly suppress EPA levels, the effect is small in absolute terms (typically 0.2–0.5 percentage points), and DHA levels are largely unaffected. Moreover, these changes are dwarfed by the increases in EPA and DHA achieved by direct consumption of preformed LC-n-3 PUFAs from fish, seafood, or supplements. For example, a modest intake of 250–500 mg/day of EPA+DHA can raise the omega-3 index (erythrocyte EPA+DHA) by 2–3 percentage points or more (Jackson et al., 2024). Therefore, from a public health perspective, the most effective strategy to improve n-3 PUFA status is to increase intake of preformed EPA and DHA, not to restrict LA.

7.3. Critical Flaws in the Ratio Concept

Despite its intuitive appeal, the n-6/n-3 ratio has been severely criticised by many experts for several fundamental flaws (Harris, 2006; Stanley et al., 2007; Jackson et al., 2024).

Flaw 1: Non-specificity and lack of standardisation. The ratio can be calculated using different sets of fatty acids. Should it include only the essential precursors (LA and ALA)? Or all n-6 and n-3 PUFAs, including AA, EPA, and DHA? Should it be calculated from diet, plasma, erythrocytes, or adipose tissue? Each choice yields a different numerical value, and there is no consensus on which metric is most meaningful (Jackson et al., 2024).

Flaw 2: Identical ratios can reflect vastly different absolute intakes. As noted by Goyens et al. (2006), a ratio of 10:1 could result from 20 g/day LA and 2 g/day ALA, or from 5 g/day LA and 0.5 g/day ALA. The absolute intakes of essential fatty acids are physiologically far more important than their ratio. A diet providing adequate ALA (≥0.5% of energy) but also high LA would still support reasonable EPA synthesis, whereas a diet with very low absolute intakes of both (e.g., 2 g LA and 0.2 g ALA) would be deficient regardless of the ratio (Jackson et al., 2024).

Flaw 3: Not all ways of changing the ratio are equivalent. Changing the ratio by increasing n-3 intake while keeping n-6 constant is not the same as changing it by decreasing n-6 intake while keeping n-3 constant. The former increases absolute n-3 intake and is almost certainly beneficial; the latter may have unintended consequences (e.g., reducing LA intake below its essential requirement). Yet the ratio alone cannot distinguish between these scenarios (Jackson et al., 2024).

Flaw 4: The assumption that all n-6 PUFAs are harmful and all n-3 PUFAs are beneficial is incorrect. As discussed in Section 6, higher LA status is associated with lower, not higher, inflammation and cardiovascular risk. AA itself produces pro-resolving mediators (lipoxins) and is not simply “pro-inflammatory”. Conversely, high doses of EPA can increase bleeding risk, and very high intakes of fish oil may have adverse effects (e.g., atrial fibrillation). The dichotomy of “good n-3” vs “bad n-6” is an oversimplification (Jackson et al., 2024; Innes & Calder, 2018).

Flaw 5: The ratio is not consistently associated with clinical outcomes. A 2024 analysis from the UK Biobank (Zhang et al., 2024) reported that a higher plasma n-6/n-3 ratio was associated with greater risk of all-cause, cancer, and cardiovascular mortality. However, in the same study, both n-6 and n-3 PUFAs individually were inversely associated with mortality. This suggests that the ratio effect is driven primarily by low n-3 levels, not by high n-6 levels — a distinction the ratio obscures (Zhang et al., 2024, cited in Jackson, 2024; Sadeghi et al., 2025).

7.4. What Do Official Guidelines Say?

Major health authorities have moved away from recommending a specific n-6/n-3 ratio.

The American Heart Association’s 2009 Science Advisory on omega-6 fatty acids concluded that “there is no evidence that a specific ratio of n-6 to n-3 PUFA is important; rather, the absolute intakes of both classes are what matter” (Harris et al., 2009, cited in Jackson, 2024). The UK Food Standards Agency convened a workshop in 2007 that reached a similar conclusion: “the use of the n-6:n-3 ratio should be abandoned, and recommendations should focus on absolute intakes of individual fatty acids” (Stanley et al., 2007, cited in Jackson, 2024).

The Nordic Nutrition Recommendations 2023 (NNR2023) also state: “The evidence for defining an optimal ratio between n-6 and n-3 PUFA is insufficient and no recommendation for the ratio of n-6 to n-3 can be set. If the recommendation for intake of n-3 PUFA is met, the ratio is likely of minor importance” (Retterstøl & Rosqvist, 2024, citing NNR2023).

The European Food Safety Authority (EFSA) has not set a target ratio, instead providing Adequate Intakes for LA (4% of energy) and ALA (0.5% of energy) separately (Petersen et al., 2024).

Thus, the international consensus is clear: dietary guidance should be based on absolute intakes, not on a ratio.

7.5. The Special Case of ALA-Only Diets (e.g., Vegans)

One context where the LA:ALA ratio may have some relevance is in individuals who consume little or no preformed EPA and DHA — for example, vegans, vegetarians, or those who avoid fish. In such populations, endogenous conversion of ALA to EPA (and to a much lesser extent DHA) is the only source of these LC-n-3 PUFAs. Here, a very high LA intake could theoretically impair EPA synthesis and lead to suboptimal EPA status.

Observational studies have reported that vegans and strict vegetarians often have lower blood levels of EPA and DHA than omnivores, even when their ALA intake is adequate (Baker et al., 2016, cited in Lenighan, 2019). This has led some researchers to suggest that plant-based individuals might benefit from reducing their LA intake (e.g., by limiting high-LA oils such as sunflower, corn, and soybean oils) to improve conversion of ALA to EPA (Jackson et al., 2024). However, the magnitude of the effect is modest, and direct supplementation with microalgae-derived DHA is a more reliable strategy to ensure adequate LC-n-3 status in plant-based populations.

For the general population that consumes fish or fish oil at least occasionally, the LA:ALA ratio is not a meaningful concern, as preformed EPA and DHA easily override any competitive inhibition (Jackson et al., 2024).

7.6. Practical Implications: Focus on Absolute Intakes

Given the limitations of the ratio concept, what practical guidance can be offered?

First, ensure adequate intake of ALA. The recommended intake is at least 0.5% of total energy (approximately 1.1–1.6 g/day for adults). Good sources include ground flaxseed, walnuts, canola oil, and soybean oil (Gutierres et al., 2025; Retterstøl & Rosqvist, 2024).

Second, consume preformed EPA and DHA. The recommended intake is 250–500 mg/day of EPA+DHA, achievable by eating 1–2 servings of oily fish (salmon, mackerel, herring, sardines) per week. For those who do not eat fish, microalgae-derived supplements are an option.

Third, LA intake should fall within the recommended range of 5–10% of total energy (approximately 12–17 g/day for adults). Most Western populations are within this range, though some individuals with very high intakes (e.g., >15% of energy) might consider moderate reductions, particularly if they also have low n-3 intake.

Fourth, do not obsess over the ratio. A person consuming adequate n-3 (both ALA and preformed EPA/DHA) and moderate n-6 (within the 5–10% range) is unlikely to derive additional benefit from manipulating the ratio per se. Conversely, a person with very low n-3 intake and very high n-6 intake would benefit far more from increasing n-3 than from reducing n-6.

7.7. Summary

The n-6/n-3 ratio is an oversimplified and potentially misleading concept. While LA and ALA do compete for desaturase enzymes, the effect of this competition on tissue EPA and DHA levels is modest in absolute terms and is easily overcome by direct intake of preformed LC-n-3 PUFAs. The ratio suffers from multiple conceptual flaws: lack of standardisation, confusion between absolute and relative intakes, and the false assumption that all n-6 PUFAs are harmful. Leading health authorities have therefore recommended abandoning the ratio in favour of absolute intake targets for individual fatty acids. For the general population, ensuring adequate ALA (≥0.5% energy) and preformed EPA+DHA (250–500 mg/day), while keeping LA within the 5–10% energy range, is a simpler, more evidence-based approach than trying to achieve a specific n-6/n-3 ratio.

8. Genetics and Individual Variability: The FADS Gene Polymorphism

Not all individuals respond to dietary fatty acid intake in the same way. This inter-individual variability has long been recognised in nutrition research, but only in recent decades have the genetic determinants of fatty acid metabolism begun to be elucidated. One of the most important genetic factors influencing omega-6 and omega-3 PUFA metabolism is the fatty acid desaturase (FADS) gene cluster, located on chromosome 11 (11q12–q13). This section summarises current knowledge about FADS polymorphisms, their effects on circulating and tissue levels of polyunsaturated fatty acids, their interactions with dietary intake, and their potential modification of disease risk.

8.1. The FADS Gene Cluster: Enzymes of Endogenous PUFA Synthesis

The FADS cluster comprises three genes: FADS1 (encoding Δ5-desaturase, D5D), FADS2 (encoding Δ6-desaturase, D6D), and FADS3 (whose function is less well characterised). As described in Section 1.2, D6D catalyses the first, rate-limiting step in the conversion of linoleic acid (LA, 18:2n-6) to gamma-linolenic acid (GLA, 18:3n-6) and of alpha-linolenic acid (ALA, 18:3n-3) to stearidonic acid (SDA, 18:4n-3). D5D catalyses the final desaturation step in the synthesis of arachidonic acid (AA, 20:4n-6) from dihomo-gamma-linolenic acid (DGLA, 20:3n-6) and of eicosapentaenoic acid (EPA, 20:5n-3) from eicosatetraenoic acid (ETA, 20:4n-3). Thus, variations in the activity of these enzymes have profound effects on the balance of n-6 and n-3 PUFAs in blood and tissues (Berkowitz et al., 2026; Koletzko et al., 2019, cited in Retterstøl, 2024).

8.2. Single Nucleotide Polymorphisms (SNPs) in FADS1 and FADS2

Genome-wide association studies (GWAS) have identified multiple single nucleotide polymorphisms (SNPs) within the FADS1/2 region that are strongly associated with circulating levels of PUFAs. The most studied SNP is rs174547, a common variant located in FADS1 (though it is in strong linkage disequilibrium with variants in FADS2 as well). The minor T-allele of rs174547 (frequency approximately 30–40% in European populations) is associated with reduced D5D and D6D activity, leading to characteristic changes in the fatty acid profile (Guan et al., 2014, cited in Marklund, 2019; Wu et al., 2017).

Specifically, carriers of the T-allele (CT or TT genotypes) typically have:

  • Higher levels of the precursor fatty acids LA (n-6) and ALA (n-3) in blood and tissues, because their conversion to longer-chain products is less efficient.
  • Lower levels of the downstream products AA (from LA) and EPA/DHA (from ALA).
  • A lower LA-to-AA ratio (reflecting reduced D5D activity) and a lower ALA-to-EPA ratio (Wanders et al., 2017, cited in Berkowitz, 2026).

These genetic effects are large — the FADS locus explains up to 25–30% of the inter-individual variation in circulating AA levels, making it one of the strongest known genetic determinants of PUFA metabolism (Guan et al., 2014, cited in Wu, 2017).

8.3. FADS Genotype and Disease Risk: Observational Findings

Given the strong associations between FADS variants and PUFA levels, and the established links between PUFA levels and cardiometabolic disease, it is plausible that FADS genotype modifies disease risk. However, the evidence is complex and sometimes counterintuitive.

In the pooled analysis by Wu et al. (2017) of 20 cohorts (39,740 participants, 4,347 incident T2DM cases), the association between LA biomarkers and T2DM risk was not significantly modified by rs174547 genotype (P-interaction = 0.13). This suggests that the strong inverse association between LA and T2DM is consistent across genotypes.

Similarly, in the FORCE analysis of CVD outcomes (Marklund et al., 2019; 30 cohorts, 68,659 participants), there was no significant interaction between LA or AA levels and rs174547 genotype for total CVD, CVD mortality, or total CHD (P-interaction > 0.13 for most outcomes). However, a notable exception was observed for ischemic stroke. In the 14 studies with genotype data available, the inverse association between LA levels and ischemic stroke risk appeared significantly stronger in carriers of the T-allele (the “low desaturase activity” allele). The hazard ratio per interquintile range of LA was 0.85 (95% CI: 0.74, 0.98) for T-allele carriers compared with 1.01 (95% CI: 0.86, 1.18) for non-carriers (P-interaction = 0.002) (Marklund et al., 2019). This suggests that the protective effect of LA against stroke may be genetically modulated, though the mechanism is not yet understood.

8.4. Gene-Diet Interactions: Do Genotypes Modify Responses to Dietary LA?

Beyond disease risk, there is growing interest in whether FADS genotype influences the metabolic response to changes in dietary PUFA intake. This has implications for the concept of “precision nutrition”.

Several small controlled feeding studies have reported that individuals with different FADS genotypes differ in their lipid and inflammatory responses to LA-rich diets. For example, Lankinen et al. (2019) found that the inflammatory response to a high-LA diet (from plant oils) differed depending on FADS1 genotype. Carriers of the minor allele (associated with lower desaturase activity) showed a greater anti-inflammatory response (lower IL-6, higher adiponectin) compared with major allele homozygotes (Lankinen et al., 2019, cited in Berkowitz, 2026; Jackson, 2024). Similarly, Vaittinen et al. (2022) reported that FADS1 genotype modified the effect of an LA-enriched diet on adipose tissue inflammation, with carriers of the minor allele experiencing greater reductions in pro-inflammatory eicosanoids (Vaittinen et al., 2022, cited in Berkowitz, 2026).

These findings suggest that individuals with genetically lower desaturase activity (who tend to have higher LA and lower AA levels) might derive greater metabolic benefits from increasing LA intake. However, the evidence is still preliminary, and most large-scale pooling projects (FORCE) have not found significant genotype × LA interactions for the major clinical outcomes of CVD and T2DM (Marklund et al., 2019; Wu et al., 2017). Therefore, while genotype may influence intermediate biomarkers (e.g., inflammatory markers), it does not appear to fundamentally alter the direction of association between LA intake and hard disease endpoints.

8.5. Evolutionary Considerations: Adaptation to Dietary Changes

The high frequency of FADS variants in different populations has been linked to evolutionary adaptation to dietary changes. Ancestral human diets likely had a much lower LA:ALA ratio and contained more preformed LC-PUFAs from wild game, fish, and foraged foods. The agricultural revolution and, more recently, the industrialisation of vegetable oil production have dramatically increased LA intake while reducing the intake of preformed EPA and DHA (Blasbalg et al., 2011, cited in Berkowitz, 2026).

It has been hypothesised that genetic variants associated with lower desaturase activity (and thus lower endogenous conversion of LA to AA) might have been selected for in populations that adopted high-LA diets, as they would reduce the accumulation of AA and potentially lower inflammation (Koletzko et al., 2019, cited in Retterstøl, 2024). This hypothesis remains speculative but illustrates the dynamic interplay between human genetics and dietary environment.

8.6. Clinical and Public Health Implications

What do these genetic findings mean for dietary recommendations?

First, at the population level, the consistent inverse associations between LA intake and cardiometabolic disease across different genotypes indicate that current recommendations (5–10% of energy from LA) are broadly applicable. Individuals should not be alarmed by genetic testing results suggesting they are “poor converters” of LA to AA — the conversion is already very low in all individuals, and higher LA intake is still associated with lower CVD risk regardless of genotype (Marklund et al., 2019; Wu et al., 2017).

Second, for individuals with a strong family history of ischemic stroke, the potential gene–LA interaction observed in FORCE (Marklund et al., 2019) might be of interest, but it is not yet sufficient to justify genotype-specific dietary recommendations. Further research is needed to replicate this finding and to understand its mechanism.

Third, for individuals who do not consume preformed EPA and DHA (e.g., strict vegans), FADS genotype could theoretically influence the efficiency of ALA conversion to EPA. Carriers of the minor allele (lower desaturase activity) might have lower endogenous EPA synthesis and could benefit more from direct LC-n-3 supplementation. However, even in these individuals, increasing ALA intake (e.g., from flaxseed, walnuts, canola oil) can raise EPA levels, though the effect is modest (Baker et al., 2016, cited in Lenighan, 2019).

Fourth, the emerging evidence on genotype-dependent inflammatory responses to LA (Lankinen et al., 2019) suggests that some individuals might experience greater anti-inflammatory benefits from increasing LA intake. However, given that no adverse effects have been observed in any genotype, there is no rationale for recommending restriction of LA based on FADS genotype.

8.7. Summary

Genetic variation in the FADS1/2 gene cluster is a major determinant of circulating and tissue levels of n-6 and n-3 PUFAs. The common rs174547 T-allele is associated with reduced desaturase activity, higher LA and ALA levels, and lower AA and EPA levels. Despite these genetic effects, the inverse associations between LA intake and risk of CVD, CVD mortality, and T2DM are generally consistent across genotypes, with a possible stronger protective effect for ischemic stroke in T-allele carriers. Gene–diet interaction studies suggest that some individuals may have enhanced anti-inflammatory responses to LA, but no adverse effects have been identified in any genotype. Therefore, current population-wide recommendations to consume LA within the 5–10% of energy range remain appropriate for all individuals, regardless of FADS genotype. Genetic testing for PUFA metabolism is not currently indicated for routine dietary guidance but may become part of personalised nutrition strategies in the future as the evidence base develops.

9. Conclusion

The scientific journey of omega-6 polyunsaturated fatty acids — from linoleic acid (LA) to arachidonic acid (AA) and their myriad metabolites — is a powerful illustration of how nutritional science evolves, often challenging simple dichotomies of “good” and “bad”. The evidence accumulated over more than six decades, synthesised in numerous systematic reviews, meta-analyses, and individual-level pooled analyses, supports several robust conclusions while also highlighting important gaps and nuances.

9.1. Key Takeaways

1. Linoleic acid is an essential nutrient with cardiometabolic benefits. LA cannot be synthesised by the human body and must be obtained from the diet. Severe deficiency causes characteristic skin lesions, but such deficiency is extremely rare in healthy populations consuming varied diets (Innes & Calder, 2018; Jackson et al., 2024). Across dozens of prospective cohort studies, higher LA intake and higher circulating LA levels are consistently associated with lower risks of cardiovascular disease (CVD), CVD mortality, ischaemic stroke, and incident type 2 diabetes (T2DM) (Marklund et al., 2019; Wu et al., 2017; Farvid et al., 2014). The magnitude of the association is substantial — a 35% lower risk of T2DM per interquintile range of LA, and a 22% lower risk of CVD mortality (Marklund et al., 2019; Wu et al., 2017).

2. Arachidonic acid is not harmful at typical intakes. Despite its role as a precursor to pro-inflammatory eicosanoids, AA levels are not associated with increased risk of CVD, T2DM, or all-cause mortality in prospective studies (Marklund et al., 2019; Wu et al., 2017). AA also gives rise to pro-resolving mediators (lipoxins) and is essential for normal brain development and function (Djuricic & Calder, 2021; Innes & Calder, 2018). Supplementation studies with AA up to 1,500 mg/day have not demonstrated consistent adverse effects on inflammation, platelet aggregation, or immune function (Calder et al., 2019, cited in Djuricic, 2021).

3. Randomised controlled trials confirm benefits for lipid profile and glycaemic control. Replacing saturated fatty acids (SFAs) with LA-rich PUFAs reduces total cholesterol, LDL-cholesterol, and the total:HDL-cholesterol ratio in a dose-dependent manner (Mensink, 2016, cited in Maki, 2018). PUFA substitution also modestly improves fasting glucose, HbA1c, and HOMA-IR (Imamura et al., 2016, cited in Maki, 2018). The effects on blood pressure are neutral for LA, though MUFAs may have a small benefit (Maki et al., 2018; Petersen et al., 2024).

4. The pro-inflammatory hypothesis for LA is not supported by human evidence. Although LA is the precursor of AA, the conversion of LA to AA in humans is very limited (< 0.2%) and does not increase with higher LA intake (Rett & Whelan, 2011, cited in Jackson, 2024). Systematic reviews of RCTs have found no increase in inflammatory markers (CRP, IL-6, TNF-α) when LA intake is increased (Johnson & Fritsche, 2012; Su et al., 2017). Observational studies have even reported inverse associations between LA status and systemic inflammation (Ferrucci et al., 2006; Virtanen et al., 2018). Thus, concerns about LA promoting inflammation should not be a barrier to its use in place of SFAs.

5. The n-6/n-3 ratio is an unhelpful concept. The ratio suffers from multiple fundamental flaws: lack of standardisation, confusion between absolute and relative intakes, and the false equivalence of all n-6 and n-3 PUFAs (Jackson et al., 2024; Harris, 2006). Major health authorities have recommended abandoning the ratio in favour of absolute intake targets for individual fatty acids. Practical guidance should focus on ensuring adequate ALA (≥0.5% energy), preformed EPA+DHA (250–500 mg/day), and LA within 5–10% of energy, rather than achieving a specific ratio.

6. Genetic variation in FADS genes modifies PUFA metabolism but does not alter population-level recommendations. Carriers of the minor T-allele of rs174547 have reduced desaturase activity, leading to higher LA and lower AA levels (Guan et al., 2014, cited in Marklund, 2019). However, the inverse associations between LA and CVD or T2DM are consistent across genotypes, with a possible stronger protective effect for ischaemic stroke in T-allele carriers (Marklund et al., 2019). Current recommendations apply to all individuals regardless of genotype.

9.2. Remaining Controversies and Research Gaps

Despite the substantial body of evidence, several important questions remain unresolved.

Cancer associations are heterogeneous and of low certainty. While higher LA intake is associated with lower risks of lung and prostate cancers, it is associated with higher risks of endometrial and ovarian cancers (Sadeghi et al., 2025). The mechanisms for these site-specific differences are not understood, and the certainty of evidence for cancer outcomes is generally low to very low. Further research is needed, including studies that distinguish between different n-6 PUFAs (LA vs. AA) and account for potential effect modification by hormone status, genetics, and other dietary factors.

The effect of very high LA intakes (>10–12% of energy) is poorly studied. Most observational studies and RCTs have examined LA intakes within the range of approximately 4–10% of energy. The health effects of chronically consuming LA at levels above 12–15% of energy (e.g., through excessive use of LA-rich oils in the absence of other dietary fats) are not well defined. Some authorities suggest that such high intakes might reduce EPA synthesis and potentially displace other beneficial fatty acids, but direct evidence is limited (Jackson et al., 2024).

The impact of replacing traditional LA-rich oils with high-oleic oils is unknown. Food manufacturers are increasingly replacing soybean, sunflower, and canola oils with high-oleic versions (lower in LA, higher in MUFA). Modelling studies suggest this could reduce LA intake in the US population by approximately 30% (Raatz et al., 2018, cited in Jackson, 2024). Whether this shift will have unintended consequences for cardiometabolic health has not been tested in long-term trials.

The role of LA in primary prevention in non-Western populations requires further study. Most of the large cohort studies have been conducted in North America and Europe. Data from Asia, Africa, and Latin America are more limited, though the PURE study (Dehghan et al., 2017, cited in Retterstøl, 2024) and the FORCE consortium (Marklund et al., 2019) have included some non-Western populations. The consistency of findings across diverse settings strengthens generalisability, but region-specific factors (baseline LA intake, background diet, genetic background) may modulate effects.

The optimal LA:ALA ratio in the absence of preformed EPA and DHA remains uncertain. For individuals who consume little or no fish or seafood (e.g., vegans, vegetarians, populations with limited access to marine foods), endogenous conversion of ALA to EPA is the only source of LC-n-3 PUFAs. In this context, very high LA intake could theoretically impair EPA synthesis (Liou et al., 2007). The threshold at which LA begins to significantly inhibit ALA conversion, and whether this has clinical consequences, deserves further investigation.

9.3. Implications for Dietary Guidance

Based on the totality of evidence, current dietary recommendations to replace saturated fats with unsaturated fats — particularly PUFA-rich vegetable oils — are well justified (Maki et al., 2018; Petersen et al., 2024; NNR2023). For the general adult population, the following evidence-based guidance can be offered:

  • Consume LA within the range of 5–10% of total energy. For a 2000 kcal diet, this corresponds to approximately 11–22 g of LA per day. Most Western populations are already within or near this range.
  • Replace sources of saturated fats (butter, lard, tallow, palm oil, coconut oil) with vegetable oils rich in unsaturated fats (soybean, canola, olive, corn, sunflower oils) and with whole food sources of unsaturated fats (nuts, seeds, avocados).
  • Ensure adequate intake of ALA (≥0.5% of energy, approximately 1.1–1.6 g/day) from sources such as ground flaxseed, walnuts, canola oil, and soybean oil.
  • Consume preformed EPA and DHA (250–500 mg/day) from 1–2 servings of oily fish per week or, for those who do not eat fish, from microalgae-derived supplements.
  • Do not focus on the n-6/n-3 ratio. Instead, focus on meeting absolute intake targets for LA, ALA, and EPA+DHA.
  • Store vegetable oils properly (cool, dark place, tightly sealed) and avoid reusing frying oils to minimise the formation of harmful oxidation products.

9.4. Final Remarks: Beyond Simple Narratives

The story of omega-6 fatty acids is a cautionary tale against reductive thinking in nutrition science. The initial enthusiasm for LA as a cholesterol-lowering agent was followed by a backlash based on a plausible but ultimately unsupported pro-inflammatory hypothesis. The truth, as revealed by decades of cumulative evidence, is more nuanced: LA is neither a panacea nor a poison. At intakes within the recommended range, and especially when it replaces saturated fats, LA contributes meaningfully to the prevention of cardiovascular disease and type 2 diabetes. Concerns about inflammation, the n-6/n-3 ratio, and arachidonic acid toxicity have largely been refuted by human studies, though lingering uncertainties remain for specific cancer outcomes and for individuals with very low intakes of marine n-3 PUFAs.

For the educated public and health professionals alike, the key message is one of balance and context: dietary patterns matter more than single nutrients, and replacing less healthful fats (saturated and trans fats) with more healthful ones (unsaturated fats from plants and fish) is a well-supported strategy for improving cardiometabolic health. Omega-6 fatty acids, as part of such patterns, play a beneficial role — not a harmful one.

References

  1. Berkowitz, L., Araneda, P., Cofré, G., Sara, D., Olsen, M., Urzúa, I., Pérez, D. and Rigotti, A. (2026) 'The controversial role of linoleic acid in cardiometabolic health: from molecular pathways to human studies', Frontiers in Nutrition, 12, 1728865. DOI: 10.3389/fnut.2025.1728865 PubMed
  2. Blomhoff, R., Andersen, R., Arnesen, E.K., Christensen, J.J., Eneroth, H., Erkkola, M., Gudanaviciene, I., Halldórsson, T.I., Høyer-Lund, A., Lemming, E.W., Meltzer, H.M., Pitsi, T., Schwab, U., Siksna, I., Thorsdottir, I. and Trolle, E. (2023) Nordic Nutrition Recommendations 2023. Copenhagen: Nordic Council of Ministers. DOI: 10.6027/nord2023-003
  3. Djuricic, I. and Calder, P.C. (2021) 'Beneficial outcomes of omega-6 and omega-3 polyunsaturated fatty acids on human health: an update for 2021', Nutrients, 13(7), 2421. DOI: 10.3390/nu13072421 PubMed
  4. Gutierres, D., Pacheco, R., et al. (2025) 'The Role of Omega-3 and Omega-6 Polyunsaturated Fatty Acid Supplementation in Human Health', Foods, 14(19), 3299. DOI: 10.3390/foods14193299 PubMed
  5. Innes, J.K. and Calder, P.C. (2018) 'Omega-6 fatty acids and inflammation', Prostaglandins, Leukotrienes and Essential Fatty Acids, 132, pp. 41–48. DOI: 10.1016/j.plefa.2018.03.004 PubMed
  6. Jackson, K.H., Harris, W.S., Belury, M.A., Kris-Etherton, P.M. and Calder, P.C. (2024) 'Beneficial effects of linoleic acid on cardiometabolic health: an update', Lipids in Health and Disease, 23, 296. DOI: 10.1186/s12944-024-02246-2 PubMed
  7. Kris-Etherton, P.M. and Fleming, J.A. (2015) 'Emerging nutrition science on fatty acids and cardiovascular disease: nutritionists' perspectives', Advances in Nutrition, 6(3), pp. 326S–337S. DOI: 10.3945/an.114.006981 PubMed
  8. Lenighan, Y.M., McNulty, B.A. and Roche, H.M. (2019) 'Dietary fat composition: replacement of saturated fatty acids with PUFA as a public health strategy, with an emphasis on α-linolenic acid', Proceedings of the Nutrition Society, 78(2), pp. 234–245. DOI: 10.1017/S0029665118002793 PubMed
  9. Maki, K.C., Eren, F., Cassens, M.E., Dicklin, M.R. and Davidson, M.H. (2018) 'Omega-6 polyunsaturated fatty acids and cardiometabolic health: current evidence, controversies, and research gaps', Advances in Nutrition, 9(6), pp. 688–700. DOI: 10.1093/advances/nmy038 PubMed
  10. Marklund, M., Wu, J.H.Y., Imamura, F., Del Gobbo, L.C., Fretts, A., de Goede, J., Shi, P., Tintle, N., Wennberg, M., Aslibekyan, S., et al. (2019) 'Biomarkers of Dietary Omega-6 Fatty Acids and Incident Cardiovascular Disease and Mortality', Circulation, 139(21), pp. 2422–2436. DOI: 10.1161/CIRCULATIONAHA.118.038908 PubMed
  11. Retterstøl, K. and Rosqvist, F. (2024) 'Fat and fatty acids - a scoping review for Nordic Nutrition Recommendations 2023', Food & Nutrition Research, 68, 9980. DOI: 10.29219/fnr.v68.9980 PubMed
  12. Sadeghi, R., Norouzzadeh, M., HasanRashedi, M., Jamshidi, S., Ahmadirad, H., Alemrajabi, M., Vafa, M. and Teymoori, F. (2025) 'Dietary and circulating omega-6 fatty acids and their impact on cardiovascular disease, cancer risk, and mortality: a global meta-analysis of 150 cohorts and meta-regression', Journal of Translational Medicine, 23, 314. DOI: 10.1186/s12967-025-06336-2 PubMed
  13. Wu, J.H.Y., Marklund, M., Imamura, F., Tintle, N., Ardisson Korat, A.V., de Goede, J., Zhou, X., Yang, W.-S., de Oliveira Otto, M.C., Kröger, J., et al. (2017) 'Omega-6 fatty acid biomarkers and incident type 2 diabetes: pooled analysis of individual-level data for 39 740 adults from 20 prospective cohort studies', The Lancet Diabetes & Endocrinology, 5(12), pp. 965–974. DOI: 10.1016/S2213-8587(17)30307-8 PubMed