All seed oils safe to eat? Fat chance!

By Marika Sboros

Seeds are sometimes called vegetable oils, though not strictly correctly

Seed oils are sometimes called vegetable oils, though not strictly correctly

Globally, scientists say growing and compelling evidence reveals significant health risks associated with use of seed oils such as canola and sunflower – unless you live in South Africa. Scientists in that country say their research proves these oils are perfectly safe, will protect you from heart disease, and are free from “Frankenfood” genetic modifications to boot.

That ringing endorsement comes in a recent study published in the South African Journal of Clinical Nutrition (SAJCN). It’s enough to warm the cockles of the heart of the seed oil industry that makes billions from these products, even though concerns about serious health risks continue: 

The study, South African Seed Oils Are Safe For Human Consumptionis led by dietitian and associate professor Dr Maretha Opperman, of the Cape Peninsula University of Technology Department of Biotechnology and Consumer science Functional Foods Research Unit, and includes  Dr Carl Albrecht, past head of research of the Cancer Association of SA.

The researchers say they were motivated to perform the study out of “concerns” at a comment in The Real Meal Revolution, co-authored by world-renowned South African scientist Prof Tim Noakes, saying that the polyunsaturated fatty acid (PUFA) seed oils are toxic, high in trans fatty acids (TFAs), genetically modified, and best avoided. (Noakes didn’t make the comment. He only provided the extensive chapter on the science behind low-carb, high-fat  to treat and prevent serious disease.)  

In the study abstract, the authors say they compared three commercially bought oils – canola, sunflower and olive oil –  against “internationally accepted standards”. (Opperman told me in an email  the oils were chosen because they are “the three most used oils on the South African market”). The study aim was to determine the quality of these oils using gas chromatographic and further analyses, to determine “whether or not there was any foundation to the statement pertaining to the toxicity of South African seed oils”. Reported parameters included the oil’s fatty acid profile, TFA content and lipid peroxide levels.

VEG OILThe researchers say their analysis shows that seed oils (sometimes called vegetable oils, though not strictly correctly) are “not only a good source of fatty acid, but also provide a good source of essential polyunsaturated fatty acids in the diet”. 

They do include the caveat that consumers should ensure they don’t overuse cooking oils and should replenish them timeously”. Failure to do so will expose the oils to “chemical reactions, such hydrolysis, oxidation and polymerisation, which contribute to their degradation”.

The researchers conclude that their study data show that the oils “contain a negligible amount of TFA, adhere to international recommendations with regard to erucic acid content, and are well within the safe range for oxidation (effects of air exposure) products”. They go on to say that the oils are “of good quality, safe for human consumption, GM-free and should not be excluded from the diet”. 

Other South African nutrition academics support the findings. In an editorial in the same issue of the journal, Prof Marius Smuts and Dr Linda Malan, of Potchefstroom’s North-West University Centre of Excellence in Clinical Nutrition, say issues such as TFA content, lipid oxidation and genetic modification “apply to all edible oils”, but the natural trans-fat content of seed oils is “almost negligible”. They say Opperman and colleagues have made “a valuable contribution by measuring oxidation products and the fatty acid composition of newly purchased South African canola, olive and sunflower oils”.

Smuts and Malan say there is “no evidence to support the premise that seed oils are toxic or unhealthy in view of their fatty acid composition, as stated in The Real Meal Revolution”.  Click here for full text of Perspectives On The Use of Seed Oils In The South African Diet editorial.

International experts say the study is riddled with flaws and errors: the arguments are unsubstantiated and the findings are based on “unrepresentative real-life scenarios and “not supported by the evidence”.  They say the researchers erred in looking only at uncooked oils, in finding the oils safe because they don’t contain TFAs, and in ignoring substances in the oils that may be even more toxic than TFAs and that are linked with DNA damage.

They say the researchers have ignored the extensive body of science that contradicts their findings. They appear to be biased against saturated and monounsaturated fatty acids that are shown to be safer for cooking and for health.

James DiNicolantonio

Dr James DiNicolantonio

Dr James DiNicolantonio is a cardiovascular research scientist at Saint Luke’s Mid America Heart Institute, author or co-author of over 150 medical publications,  an Associate Editor of British Medical Journal’s (BMJ) Open Heart, and serves on the editorial advisory board of several medical journals including Progress in Cardiovascular Diseases.

DiNicolantonio says olive and canola (especially extra virgin olive oil and cold-pressed, organic versions) are on their own not problematic. “It’s the high-omega-6 (linoleic acid) seed oils, and mainly when they are cooked,” he says.

It is problematic that the study only examined uncooked seed oils: 

“When you cook with seed oils there is the formation of hydroperoxides and aldehydes which can lead to damage in the body; with aldehydes there can even be damage to the DNA. This may increase the risk of a host of chronic diseases including cancer” he says.

The study also doesn’t consider what happens to these seed oils once consumed, DiNicolantonio says: studies have found that when seed oils are ingested, the high acidity of the stomach also forms hydroperoxides and aldehydes which are then absorbed by the body.

“ Thus, this study has really nothing to do with the health effects of consuming industrial seed oils, rather than simply looking at their composition prior to consumption  – which are two different things.”

The best evidence to date regarding omega-6 seed oils and health outcomes is two meta-analyses of RCTs led by Dr Christopher Ramsden, DiNicolantonio says. Ramsden is a chief investigator in the US National Institutes of Health Section on Nutritional Neurosciences, Laboratory of Membrane Biochemistry and Biophysics. His research suggests that consuming vegetable oils is more harmful compared with animal fat plus TFAs.

Ramsden’s newest research, published in the BMJ in April 2016, concluded that while replacement of saturated fat in the diet with linoleic acid effectively lowers serum cholesterol, it “does not support the hypothesis that this translates to a lower risk of death from coronary heart disease or all causes”. It adds to “growing evidence that incomplete publication has contributed to overestimation of the benefits of replacing saturated fat with vegetable oils rich in linoleic acid”.


DiNicolantonio says it is misleading for Opperman and colleagues to conclude that seed oils are a good source of essential fatty acids and  don’t contain significant levels of TFAs, as this depends on the source and how they are cooked.

“The estimated need of linoleic acid (omega-6) intake is likely to be no more than 2 grams per day, whereas we now get 10 times this much due to a high seed oil intake.”

“Moreover, we never consumed refined seed oils during evolution but obtained our omega-6 from real foods that also bring a coating (nuts, seeds) to protect the highly susceptible PUFA oils from oxidising but also bringing fibre, phytonutrients, antioxidants, minerals and vitamins to prevent the linoleic acid from oxidising both exogenously (from outside the body) and endogenously after consumption.”

Just because some vegetable oils may not have high levels of peroxide values prior to cooking “does not equate to what we consume after cooking with these highly susceptible omega-6 rich vegetable oils”, DiNicolantonio says.  Cooking with the oil in a pan at high heat, or especially deep frying, can create “unnatural trans-fats”.

It is also misleading for the researchers to state that refined oils are more suitable than their unrefined counterparts for heat applications such as frying, he says. The smoking point has “nothing to do with oxidative rate of the oil” and high smoke point PUFA vegetable oils are still “highly susceptible to oxidation”.

When it comes heart disease, DiNicolantonio says high-heat cooking “absolutely increases oxidised linoleic acid metabolites” which is shown in research to indicate “a direct causative effect in the development of coronary heart disease”.

It is also misleading to state that seed oils are an excellent source of vitamin E, when ingestion (especially if they are cooked with) increases the oxidative stress in the body and likely depletes endogenous vitamin E. 

“When seed oils are extracted with solvents and high-heat, this can dramatically reduce the vitamin E content compared with cold-pressed seed oils and can also oxidise the oil,” he says.

To suggest that people need to consume 5-8% total energy as omega-6 for their hearts’ sake is simply c“not supported by evidence”, DiNicolantonio says.  On the contrary, the Lyon Diet Heart Study (a randomised secondary prevention study) found that dropping linoleic acid intake from 5.3% to 3.6% significantly reduced mortality and cardiovascular events by around 70%. 

Dr Martin Grootveld

Dr Martin Grootveld

British scientist Dr Martin Grootveld, professor of bioanalytical chemistry and chemical pathology at De Montfort University’s faculty Health and Life Sciences, is similarly critical of Opperman’s study. He says it contains “grave interpretational flaws and errors”.

Grootveld’s research team at De Montfort found sunflower oil and corn oil produced aldehydes at levels 20 times higher than recommended by the World Health Organisation. 

Olive oil and rapeseed oil produced far fewer aldehydes as did butter and goose fat.

Not only that, but Grootveld’s team also identified two previously unknown aldehydes in the samples of these oils – a world first.

He says the researchers have “chosen to neglect how health index parameters they have selected may change completely when the oils are exposed to high temperatures commonly associated with frying or cooking purposes”. Cooking is, after all, a primary purpose for which consumers buy these oils, he notes

Grootveld finds it  “particularly perplexing”  that the researchers failed completely to “subject the oils featured to recommended or rigorous thermal stressing episodes according to standard frying practices”. It doesn’t help, he says, that the authors’ arguments are based on poorly representative, biased  “real-life” scenarios. These combine to provide “an incomplete message” to readers and consumers.

He finds it puzzling that the researchers completely ignored important considerations regarding the generation of highly toxic lipid oxidation products (LOPs). This is especially the case when the oils are subjected to high temperatures according to high-temperature frying or cooking episodes, or alternatively  when stored for prolonged periods at room temperature and exposed to light, he says.     

“The oxidation of unsaturated fatty acids at temperatures commonly used for frying or cooking is a highly complex chemical degradation process which involves highly reactive chemical species known as free radicals,”Grootveld says. 

It primarily involves the oxidative conversion of such unsaturated fatty acids to primary “LOPs”, lipid hydroperoxides, also known as conjugated hydroperoxydienes (CHPDs). This  process is sequentially followed by  fragmentation of these primary LOPs to secondary ones that include highly toxic aldehydes in particular, he says.

Grootveld questions why the researchers failed to reflect evidence that PUFAs produce very high levels of LOPS when subjected to heat; that monounsaturated fatty acids (MUFAs) are more resistant to oxidation than polyunsaturates are, and hence generate lower levels of only particular LOPs when heated in this manner, and generally only after prolonged thermal stressing episodes; and that saturated fatty acids (SFAs) generate very little or no toxic LOPs when heated according to standard frying practices (at 180oC), as Grootveld and colleagues have shown in reported research.

His research, using nuclear magnetic resonance spectroscopy,  has also shown oxidative degradation of PUFA-rich culinary oils when subjected to standard frying practices. The degradation includes very high levels of highly toxic substances called aldehydes, in addition to their similarly toxic precursor substances.  These aldehydes and their precursors are often “readily detectable in newly purchased unheated culinary oils”, Grootveld says, and are shown to present health risks.

TWEET veg oilThe LOPs act as powerful toxins since they are “extremely chemically reactive”, Grootveld says, and cause damage to “critically important biomolecules such as DNA”.

The South African researchers also state (albeit correctly), he says, that “…animal fat is also prone to degradation during cooking since it does not consist of saturated fat only, but is rather a combination of SUFA [saturated fatty acid], MUFA and PUFA in different ratios’’. However, they once again fail to indicate that because saturated fats contain much lower PUFA contents, they are therefore much less susceptible to oxidative damage during standard frying or cooking practices.

Grootveld says PUFA-rich oils are quite simply “the worst possible choice” for cooking.

Other research  supports concerns about the health risks associated with seed oils.  A Japanese study titled Medicines and Vegetable Oils as Hidden Causes of Cardiovascular Disease and Diabetes, published in Pharmacology in June 2016, suggests that seed oils in part share  “a common mechanism” with medicines such as statin and warfarin to inhibit vitamin K2-dependent processes. The scientists interpret this as leading to increased onset of cardiovascular disease, diabetes chronic kidney disease, bone fracture and other lifestyle-related diseases. 

And of course, concerns about health risks linked to seed oils are not new. Way back in 1965, British physicians at St Mary’s Hospital London published a well-designed study in the British Medical Journal (BMJ) known as “Rose Corn Oil Trial”. The researchers concluded that “under the circumstances of this trial, corn oil cannot be recommended as a treatment of ischaemic heart disease. It is most unlikely to be beneficial, and it is possibly harmful.”

When it comes to what’s behind this study, it may just be coincidence that  the South African Journal of Clinical Nutrition (SAJCN) is, as it says on its website, the  official joint publication of the Association for Dietetics in South Africa (ADSA), the Nutrition Society of South Africa (NSSA) and the South African Society of Parenteral and Enteral Nutrition”;  and that  both ADSA and NSSA are actively involved as witnesses and consultants in the prosecution of Noakes by the Health Professions Council of South Africa (HPCSA). Noakes faces a charge of unprofessional conduct for two tweets in February 2014, in which he told a breastfeeding mother that good first foods for infant weaning are LCHF. 

Must read! Noakes exposed! The REAL beef dietitians have with him

Or just as with marrying in haste, this study could simply be a case of researching in haste, repenting at leisure. It seems as if the researchers just didn’t do their basic scientific homework properly.