The ultimate guide to seed oils and metabolic health

If you want to start a war on social media, post something about seed oils. While most functional medicine experts recommend avoiding them at all costs, other medical experts say the supposed “risks” are overblown. And the truth is, there’s research to support both points of view.

If you care about your metabolic health, it’s worth examining all of the evidence so you can make the best decision for your health goals. Use this guide to understand better what seed oils are, the science for and against consuming them, and what appears to be the smartest path forward for your overall health. 

What are seed oils, anyway?

“Seed oils” is an umbrella term used to refer to oils that have been extracted from the seeds—rather than the fruit or pulp—of certain plants. Typically, when you hear the term, it means highly refined or “industrially processed” seed oils that are relatively new to the food supply, including:

• Canola oil
• Corn oil
• Cottonseed oil
• Soybean oil
• Safflower oil
• Sunflower oil
• Grapeseed oil
• Rice bran oil

The “seed oils” category technically encompasses traditional seed-derived oils such as sesame and flaxseed oil, too—however, these aren’t usually considered as harmful (even by some staunchly anti-seed oil experts), as they contain beneficial antioxidants and don’t undergo harsh industrial processing. That said, their relatively high concentration of fragile, polyunsaturated fatty acids—a characteristic shared by other seed oils—means they are not ideal for high-heat cooking, as they can easily oxidize (more on this later). 

Production of the first commercial seed oils, such as corn and soybean oil, began in the late 1800s and early 1900s. Intake has drastically increased since. Consumption of soybean oil alone grew 1,000-fold between 1909 and 1999 and four-fold just in the past 10 years. This increase may have initially been spurred by the American Heart Association’s 1961 recommendation to replace saturated fats with polyunsaturated vegetable oils (a.k.a. seed oils) as a means to reduce heart disease risk. Now, according to some estimates, seed oils make up 8 to 10 percent of total energy intake in the average Western diet, and the average American consumes the equivalent of three tablespoons per day.

Why is the current intake so high? While you can find bottles of some seed oils in grocery stores, they’re also used in fast-food restaurants (McDonald’s uses a blend of canola, corn, and soy oil to deep-fry its french fries) and in processed packaged foods, from chips to cookies to various condiments such as salad dressings and mayonnaise.

A quick clarification: The terms vegetable oil and seed oil are often used interchangeably since products labeled and sold as “vegetable oil” are usually made from soybean oil or a blend of seed oils like corn, soy, and cottonseed. However, the broader vegetable oil category technically encompasses all plant-based oils. Olive, avocado, and coconut oil are not considered seed oils.

How are most seed oils made?

Compared to the relatively simple production of something like extra-virgin olive oil—which involves grinding olives into a paste, separating the oil and fruit solids in a centrifuge, and filtering the finished oil, all while minimizing exposure to heat—making seed oils is a complex process that requires significant processing, refinement, and heat exposure. 

Production calls for several steps to extract as much oil as possible from the seeds and eliminate compounds that could lead to premature oxidation and spoilage: 

1. The seeds are crushed, and oil is extracted. However, some oil remains in the crushed seeds.
2. The crushed seeds are mixed with the chemical solvent hexane to create a slurry that’s heated, which aids in the release of additional oils. 
3. The resulting oil is often “washed” with a citric acid solution and then centrifuged or settled to help separate out compounds like phospholipids, which can contribute to cloudiness, unwanted colors and flavors, and premature spoilage. 
4. The oil is then heated (194°F to 230°F) and mixed with a bleaching clay, which binds to and removes other unwanted compounds, including lipid oxidation byproducts.
5. The final step is typically deodorizing: The oil is steamed to remove additional compounds, such as free fatty acids and pesticides, which vaporize.

The final oils have a high smoke point, uniform color, and flavor and are relatively shelf-stable. However, the production process also leads to the removal of a variety of beneficial compounds, such as carotenoid antioxidants, sterols, and tocopherols (e.g., vitamin E)—some of which would be beneficial to help the oils resist oxidation and degradation when they’re used for cooking. 

The arguments against seed oils

The major arguments against seed oils often relate to the composition and stability of their fatty acids. So, let’s start with a bit of food science background: 

• All culinary oils contain a combination of polyunsaturated fatty acids (PUFAs, which encompass omega-6 and omega-3 fatty acids), monounsaturated fatty acids (MUFAs), and saturated fatty acids (SFAs). The percentages of each vary by type of oil.
• These fatty acids are distinguished by the number of double bonds in their molecular structure. SFAs contain only single bonds between their carbon atoms, MUFAs contain one double bond, and PUFAs contain two or more double bonds. 
• The more double bonds present in a fatty acid, the more prone they are to oxidation. Meaning, oils that contain a high proportion of omega-6 or omega-3 PUFAs are more likely to degrade when exposed to oxygen, light, and heat compared to oils with a higher proportion of MUFAs (like olive oil or avocado oil) or SFAs (like coconut oil). 
• Many highly processed seed oils contain a relatively high concentration of omega-6 PUFAs (particularly linoleic acid), and they are exposed to significant heat during processing and again when used in high-heat cooking applications.

These are all crucial factors to remember as we explore the following points. 

Heating PUFA-rich seed oils generates potentially harmful compounds

All PUFAs are less stable (i.e., more prone to degradation) than other fatty acids. As such, when they encounter heat, they can generate potentially harmful lipid oxidation products (LOPs) such as peroxides, which further degrade into secondary LOPs such as aldehydes. 

While some LOPs generated during seed oil refinement are removed during the later processing steps, research shows that cooking with PUFA-rich seed oils also generates these harmful compounds. Meaning, some end up in your food and may, therefore, be absorbed into the bloodstream and make their way to cells throughout the body. Somewhat unsurprisingly, though, the type of cooking matters: Exposing these oils to prolonged or repeated episodes of high heat (common with deep frying) increases the concentration of these compounds more than a light saute or using them in raw applications (such as salad dressing) does. 

Even experts who are generally pro-seed-oils have stated that repeatedly frying with these oils can cause a buildup of damaging chemicals that may harm health. Illustrating this point, a 2020 study lasting four months found that rabbits fed a daily dose of vegetable oils that had been repeatedly heated experienced impaired liver function due to an accumulation of liver fat and a 40 to 60 percent increase in oxidative stress in a dose-dependent manner. Other rabbits that received vegetable oil heated only once experienced no adverse effects. 

One particularly toxic lipid aldehyde that results from the oxidation of omega-6-rich seed oils is 4-hydroxy-2-nonenal (4-HNE). This has gained attention for its potential role in promoting diabetes, Alzheimer’s, atherosclerosis, cancer, and LDL oxidation, per several lab and animal studies. In a 2020 study, considerable amounts of 4-HNE were generated in seed oils stored in an open container at 140°F for 30 days (a scenario meant to simulate improper storage) and in seed oils heated to a standard frying temperature of 356°F for three hours. The oil with the highest initial omega-6 linoleic acid content—corn oil—yielded the highest levels of 4-HNE in both scenarios—a 70- to 80-fold increase from baseline. This supports a study from 2015, which found that fast food french fries contained more 4-HNE when cooked in seed oils containing higher concentrations of omega-6 linoleic acid versus oils containing higher concentrations of oleic acid, a type of monounsaturated fat that’s more heat-stable. 

The toxic nature of 4-HNE is because it is highly reactive with proteins, DNA, and phospholipids, the latter of which make up cell membranes. This means it could potentially trigger damage throughout the body. In fact, research shows that synthetic injections of 4-HNE in monkeys contributed to the dysfunction and death of neurons in the brain, liver cells, and the pancreatic beta-cells that secrete insulin. Lab and animal studies also suggest that 4-HNE may contribute to the production of pro-inflammatory cytokines, activate pro-inflammatory signaling pathways, and impair insulin signaling and glucose uptake in muscle cells via its contribution to oxidative stress. 

But just how dangerous are these lipid oxidation products to humans? It’s hard to say, as an upper limit or “safe dose” has not been established for secondary LOPs like 4-HNE. Because of this, relatively little is known about how much seed oil—or how much seed oil that’s been used in specific cooking applications, like deep-frying—would contain enough 4-HNE to pose a legitimate health risk. Still, while researchers have called for further study in this area, some metabolic health experts, such as Levels advisor Ben Bikman, PhD, have sounded the alarm on seed oil consumption and the subsequent accumulation of 4-HNE in fat cells as potential drivers of insulin resistance. 

Additionally, the heat-induced oxidation that occurs during processing and some types of cooking with select seed oils may also promote the formation of trans fats, which are widely considered the worst type of fat, as they cause inflammation, raise LDL cholesterol, and lower HDL cholesterol. In a 2018 study, researchers heated different oils to 464°F and also in a deep fryer at 356°F for six hours. Highly refined seed oils such as grapeseed and canola oil had significantly greater increases in trans fats compared to less refined, MUFA-rich olive oil and avocado oil. 

However, the actual levels of trans fats generated during common home-cooking applications may be much lower or non-existent. A 2015 study found that stir-frying with corn oil at 338°F—which consisted of stirring the oil continuously in the pan for two minutes—generated more trans fats than other cooking methods, but quantities still fell below strict European regulatory limits of 2 grams (g) trans fat per 100 g oil or fat. 

Bottom line: Harmful compounds can be generated during processing and while cooking with seed oils—particularly those oils with the highest omega-6 or linoleic acid content—but we don’t yet know the actual risk level for humans. Thus far, the most significant potential risk appears to come when these oils are repeatedly heated to high temperatures, as with deep frying. So, what to do? We recommend you avoid cooking with seed oils on high heat; other non-seed oils, such as olive or avocado, will work.

Seed oils contribute to our excessive omega-6 fat intake when what we need is more omega-3s

A popular anti-seed-oil argument is that they contribute to an excessive intake of omega-6 fats relative to omega 3s or a high omega 6:omega 3 ratio—and this imbalance harms health. But before we unpack this claim, it’s essential to understand that omega 6s are not inherently “bad.” 

Both omega-6 linoleic acid (LA) and omega-3 alpha-linolenic acid (ALA) are considered essential polyunsaturated fatty acids because our bodies can’t produce them, so we must obtain them from food. Whole-food sources of LA include sunflower seeds, pumpkin seeds, nuts, and soy foods like edamame. ALA is predominantly found in walnuts, flax, hemp, and chia seeds. The omega 3s EPA and DHA are also found in fatty fish. 

These fats and their metabolic byproducts contribute to the fluidity of cell membranes, regulate inflammation, and promote normal growth and cellular metabolism. Deficiencies of either type can contribute to health problems. For example, deficiencies in omega 6s may contribute to poor growth, dry or scaly skin, and hair loss. In comparison, deficiencies in omega 3s have been linked to neurological problems, vision issues, and more. 

The problem: In our modern diets, we get plenty of omega 6s and few omega 3s. Experts estimate that our ancestors got equal amounts of these fats. But today, thanks to the widespread use of seed oils in processed and fried foods and low intake of omega-3-rich foods, a typical Western diet has five to 15 times more omega 6s than omega 3s. In fact, intake of omega 3s has remained relatively constant since the early 1900s, while intake of omega 6 from LA has increased three-fold, from 1 to 2 percent to more than 7 percent of daily calories. 

Disproportionately high intake of seed oils may be a concern, partly because omega 6s and omega 3s compete for the same enzymes at multiple points in their metabolism. This could potentially result in us not reaping the full benefits of the omega 3s that we do consume:

• LA and ALA compete for the same desaturase and elongase enzymes, so high-LA and low-ALA intake results in LA being the dominant substrate (or molecule upon which an enzyme acts) in this metabolic pathway. In turn, enzymes may preferentially metabolize LA into the long-chain omega 6 arachidonic acid (AA), leaving fewer enzymes left to metabolize ALA into the beneficial long-chain omega 3s EPA and DHA. ALA that’s not converted into EPA and DHA may instead be burned for energy, stored in adipose tissue, or incorporated into cell membranes. (Just keep in mind: EPA and DHA can be consumed directly from sources like fatty fish, which would bypass these desaturase enzymes.) 
• AA and EPA also compete for metabolism by COX and LOX enzymes, which transform these fatty acids into oxylipins (or oxygenated PUFAs) called eicosanoids that act as signaling molecules. AA-derived eicosanoids tend to be pro-inflammatory, while EPA-derived eicosanoids are more anti-inflammatory. So, the theory goes, high intake of omega-6-rich seed oils coupled with low omega 3s may yield excessive AA-derived eicosanoids and promote excessive inflammation, which drives chronic disease.

While some experts and institutions have said that the ideal omega 6:omega 3 ratio should be in the range of 2:1 to 4:1 for optimal health, research doesn’t support slashing omega-6 intake. Instead, several studies suggest that increasing omega-3 intake (which still lowers the omega 6:omega 3 ratio) may offer the greatest health benefits and that consuming too many omega 6s at the expense of omega 3s is a bad idea. For example:

• In a 2024 study, researchers analyzed plasma levels of omega 3s and omega 6s and how they related to risk of death. People with the highest ratio of omega 6 to omega 3 were 26 percent more likely to die of any cause, including cancer and cardiovascular disease, compared to those with the lowest ratio. But, when considered individually, both high levels of omega-6 and omega-3 fatty acids were associated with reduced risk of death. In other words, both essential fatty acids were protective, but omega 3s were more protective and should be prioritized. 
• In a 2021 study, researchers tested several diets with differing levels of omega 3s (EPA/DHA from fatty fish) and omega 6s (LA from corn oil) on migraine sufferers: high-omega-3/low-omega-6, high-omega-3/high-omega-6, and an “average U.S. diet” control that was low-omega-3/high-omega-6. After four months, both interventions lowered frequency and intensity of headaches compared to the control, but the high-omega-3/low-omega-6 group had the greatest benefit, a 30 to 40 percent reduction. Both interventions also lowered circulating AA levels, but not an AA-derived compound associated with pain. They did, however, reduce levels of other pain-related oxylipins in the blood. 
• In a 2006 study, higher plasma levels of total PUFAs (mainly omega 6 and omega 3) were associated with lower inflammatory markers and higher anti-inflammatory markers, but the relationship was stronger for omega 3s.
• In a 2003 study, omega-6 fatty acids were associated with higher levels of inflammatory markers when omega-3 intake was low. But if omega-3 intake was high, the combination of both essential fatty acids resulted in the lowest levels of inflammation. 

Bottom line: If your diet includes seed oils, which are a concentrated source of omega-6 fats but not other nutrients, understand that these could be “crowding out” more nourishing fat sources with stronger research-backed benefits. It’s vital to ensure you’re getting enough omega 3s, which seed oils don’t help with. In addition to substituting seed oils with olive or avocado oil when you cook, you can also swap them with flax or walnut oils for raw applications (e.g., drizzled over finished dishes or incorporated into dressings) to get more omega-3s in your diet. 

Why mainstream medicine thinks seed oil risks are overblown

On the other hand, most mainstream health professionals believe arguments against seed oils require too much extrapolation from preliminary research and that most randomized human trials show no adverse health outcomes (or potentially even positive effects) of consuming them. Basically, seed oils are just the latest boogeyman.

One feature of this side of the debate that’s worth noting is that several of the studies in the section below (like this one, this one, and this one)—or, in some cases, the study authors affiliated with them—received funding from companies that manufacture seed oils and the food products that contain them (e.g. Unilever and ACH Food Companies), agricultural groups that have a financial stake in producing seed oil crops, and biotech companies that help bioengineer crops used to produce seed oils or manufacture the herbicides sprayed on them (e.g., Monsanto). Of course, this doesn’t automatically mean the research should be disregarded; however, it does raise important questions about what these studies may be choosing to emphasize or downplay. (This is less notable on the anti-seed-oils side, but there are companies that have a stake in that position as well.)

It’s the foods, not the oils specifically

One argument of this crowd is that staunchly anti-seed oil advocates miss the forest for the trees. After all, these fats are commonly consumed via fast foods and highly processed foods that are also high in refined carbs, sodium, and sugar and low in fiber, antioxidants, and micronutrients. So, the health risks associated with these junk foods—e.g., cardiovascular disease, metabolic syndrome, increased belly fat, and depression—are more likely to be related to their combination of ingredients rather than seed oils alone. These foods are also more likely than unprocessed foods to contain seed oils that have been heated and partially oxidized. In other words, it’s difficult to separate the harms of the oils from their delivery vehicle. Both camps would generally agree these highly processed foods are best avoided.

The byproducts may not be that harmful in real-world applications

Plenty of studies suggest that the concerns about the byproducts of seed oil metabolism aren’t entirely justified in the real world. Many arguments against seed oils come down to their high omega-6 PUFA linoleic acid (LA) content, which can be metabolized into arachidonic acid (AA) and metabolized into eicosanoids that promote inflammation. But, it’s important to note that while some AA-derived eicosanoids promote inflammation, others have anti-inflammatory properties. 

One 2011 review with government funding found that neither an increase nor a decrease in dietary LA affected circulating AA levels, meaning it would also be unlikely to affect levels of inflammatory eicosanoids. (The study authors speculate that this lack of correlation was because study participants were already consuming a typical Western diet, and therefore, the metabolic pathway to make AA from LA may have already been saturated.)

Additionally, a 2017 study with no conflicts of interest in funding found no compelling evidence that increased intake of LA was associated with increased concentrations of the inflammatory biomarker C-reactive protein (CRP)—in fact, it found LA had an inverse relationship with CRP (other studies have found the opposite effect). And in a 2021 systematic review and meta-analysis of five studies, replacing saturated fat with PUFAs was associated with reduced cardiovascular disease occurrence.    

Furthermore, authors of a 2009 science advisory from the American Heart Association (AHA) featuring data from dozens of studies concluded that consumption of at least 5 to 10 percent of energy from omega-6 PUFAs like LA reduces risk of coronary heart disease compared to lower intakes. (Study disclosures note one author received research grant funding from Monsanto.) However, all of the randomized controlled trials featured in the advisory involved using LA to replace saturated fats—and it’s less clear if LA would still be beneficial in other dietary contexts (like replacing monounsaturated fats) or at higher intakes. The advisory also didn’t specify if the LA came from seed oils.

Saturated fats may be worse

This point about saturated fat is important because, in the real world, this is often the choice: Which do I cook with: Canola oil or butter? And this might be the strongest argument of the pro-seed oil camp.

While there is debate on the harms of saturated fats, when LA is used as a replacement for saturated fat in human feeding trials, LA has been shown to lower LDL cholesterol, triglycerides, HbA1C, and insulin resistance and elevate beneficial HDL cholesterol—all of which are associated with improvements in metabolic health. As stated, omega 6s may also help reduce inflammation, provided someone’s diet contains enough omega-3 fats.

Out of the seed-oil-specific studies that exist, they tend to mimic the findings above related to LA. A 2021 review with no conflicts of interest found that replacing saturated fats with soybean oil lowered levels of total and LDL cholesterol, and a month-long soybean oil-rich diet was not associated with increased CRP.

Should you eat or avoid seed oils? Key takeaways to guide your choice

If you find it confusing to square the two sides’ evidence, it’s understandable. They often talk past each other, framing the questions differently. For example, many of the studies used to “vindicate” seed oil consumption aren’t examining the questions that health-conscious folks are genuinely interested in. 

Such as, how would eating seed oils in place of healthier MUFA-rich fats like avocado and olive oil impact inflammation and cardiovascular risk—and might it significantly worsen these? What happens when you consume a certain percentage of your fat calories from seed oils specifically (not just from linoleic acid in general) and for more than just a few months? How does an equal amount of omega-6 linoleic acid from soybean oil versus whole nuts impact health? 

These are study topics that have the potential to provide much-needed context, balance the pool of existing research, and quantify the actual risks and benefits of seed oils. Instead, the impression is often that any positive outcomes related to linoleic acid and seed oils are used to defend their consumption, while any adverse outcomes mean they must be banished. 

More long-term, randomized controlled feeding trials in humans using different seed oils in various contexts and cooking applications would undoubtedly help prove or disprove specific claims about seed oils. We can’t simply have more studies on LA intake from which we attempt to extrapolate the results to seed oils since these fats are also found in a variety of nutritious whole foods. As with most things, it’s likely that amount and consistency of intake influence the effects of seed oils and the foods that contain them.

Also, remember that the strongest data in support of omega-6-rich seed oils appears to be their potential to improve cardiovascular risk factors, provided they are used to replace saturated fats. But they’re not unique in their ability to do this: Monounsaturated fat-rich oils such as olive oil and avocado oil have similar cardiovascular benefits when used to replace saturated fats—and they are much more stable and less likely to generate harmful lipid oxidation byproducts when heated.

What does this mean for your diet? While the presence of a teaspoon of uncooked sesame oil in your salad dressing is unlikely to inflict long-term damage (at least based on the existing research on human health outcomes), and you are certainly better off eating that salad than processed food, there’s no compelling reason to buy a bottle of canola or soybean oil at the store or seek out foods that contain seed oils. They simply don’t have any unique nutritional properties or compounds you can’t get elsewhere. 

Let’s summarize some actionable tips if you do choose to scale back on seed oils:

• Reduce your intake of fried and processed foods and prioritize whole-food sources of omega-6 fats, such as nuts and seeds, which come with beneficial nutrients and antioxidants.
• Regularly consume whole-food sources of omega 3s such as salmon, mackerel, tuna, herring, sardines, walnuts, flax oil, and flax, chia, and hemp seeds. (Just avoid heating omega-3-rich flax oil since doing so makes it unstable.)
• When cooking at home, consider using a heat-stable, MUFA-rich oil such as extra-virgin olive oil, which has a distinct flavor; or an avocado oil, which has a neutral flavor. 
• Using a bit of omega-6-rich sesame oil to enhance the flavor of dishes is okay—just avoid using it for high-heat cooking. (Remember, sesame oil contains beneficial antioxidants and doesn’t undergo harsh processing, making it a smarter seed oil pick.) 
• Swap out seed oil-containing condiments like salad dressings and mayonnaise for options made with avocado or olive oil, which are becoming increasingly common. 
• When scanning food labels, know that “high-oleic” versions of oils are still highly refined, but that they are more stable and less prone to oxidation than other seed oils.