Nutrition Guide · ALA, EPA and DHA

ALA, EPA and DHA: why flaxseed oil is not the omega-3 you need

Nuts, flaxseed and rapeseed oil deliver ALA. Brain, retina and cell membranes need EPA and DHA. Between the two sits an enzyme pathway that is remarkably narrow in adult humans.

SJ
Shukri Jarmoukli · Physician, Integrative Medicine · ViveCura Berlin
Fatty acid biochemistry Nutrigenetics Toxicology 35 sources with DOI
Why I am writing this

Flaxseed oil, chia and walnuts do not deliver the omega-3 form your body prefers to build into its cell membranes. They deliver an 18 carbon precursor from which an enzyme pathway with several stations would first have to make what brain, retina and cell membranes need.

In adult humans this pathway is so narrow that for DHA often less than one percent arrives at the end. So the decisive question is rarely: are you eating enough omega-3? It is: which of the three forms are you eating?

You may know this moment. Someone asks how you cover your omega-3. You point at the dark bottle in the fridge. Flaxseed oil, a spoonful in your muesli every morning, plus walnuts. The label says, in large letters and quite correctly: rich in omega-3.

And then a lab report is in front of you, and your omega-3 index is low. Clearly low. That is almost never about discipline. It is because the word omega-3 refers to three different molecules, and your body treats them very differently.

What to expect here

  • What separates ALA, EPA and DHA chemically
  • The conversion chain station by station, with the enzyme names
  • Why the last step to DHA happens inside the peroxisome
  • Why the percentages online differ so widely
  • Why women convert differently and what estrogen has to do with it
  • The competition from linoleic acid for the same enzyme
  • What FADS1 and FADS2 say about your conversion rate
  • Why I prefer small fish and algae oil
  • What the large heart trials honestly show

Three molecules, one name, and that is where the misunderstanding starts

Omega-3 is not a substance, it is an address. The name only tells you that the first double bond sits three carbon atoms from the far end. Three different residents live at this address.

18:3n-3 ALA, alpha-linolenic acid. 18 carbon atoms, 3 double bonds. Flaxseed, chia, walnut
20:5n-3 EPA, eicosapentaenoic acid. 20 carbon atoms, 5 double bonds. Small fish, algae
22:6n-3 DHA, docosahexaenoic acid. 22 carbon atoms, 6 double bonds. Small fish, microalgae

Four carbon atoms and three double bonds of difference sound like a detail. For biology they are not. Think of a hardware store. ALA is a length of timber, EPA and DHA are a finished window frame with fittings. Both are made of wood. If the frame is what you are missing, the stack of timber helps you only as far as your workshop can build from it.

Even the European food safety authority makes this separation: ALA counts as its own essential fatty acid, and for EPA plus DHA there is a separate intake value.

Reframe

The common question is: Am I getting enough omega-3?

The more useful question is: Which of the three forms am I getting, and what does my body make of it? Nothing on the flaxseed oil bottle is false. It simply answers a different question from the one you are actually asking.

And now you know why two people can eat the same amount of omega-3 and still show completely different blood values.

The chain that thins out inside your body

The most common objection here is a fair one: the body can convert ALA. The pathway exists and it runs in the liver. The only question is how much arrives at the end. Picture a row of workshops: at two stations the same machine is standing, it has to be used twice, and the last station is not even in the same building.

The conversion chain, station by station

A schematic view. The bars show the order of magnitude of what continues onward. Numbers appear only where measurements exist.

0
The furnace: beta-oxidation
ALA 18:3n-3

Before any conversion starts, part of the ALA is burned for energy.

Around 33 percent had been breathed out by young men within 24 hours, in women about 22 percent.
1
Delta-6 desaturase, the gene is called FADS2
18:3n-3 to 18:4n-3

The first bottleneck. This enzyme also takes linoleic acid from vegetable oils. Both stand in the same queue.

Only about 7 percent of dietary ALA reached the phospholipid pool of the liver in one compartmental study.
2
Elongase and delta-5 desaturase, the gene is called FADS1
18:4n-3 to 20:4n-3 to EPA 20:5n-3

First the chain is lengthened, then EPA appears.

Related to all ingested ALA, roughly 5 percent reach the EPA stage, related to the liver pool almost 100 percent.
3
Lengthening twice: beyond DPA
EPA to DPA 22:5n-3 to 24:5n-3

In adult men the tracer study ended here: labelled DPA was detectable for up to 14 days, labelled DHA never.

DPA is not a meaningless dead end, though.
4
A second pass and a trim inside the peroxisome
24:5n-3 to 24:6n-3 to DHA 22:6n-3

Now delta-6 desaturase is needed again, the same enzyme as at station 1. The final step then takes place in the peroxisome, a different cell compartment: there the chain is shortened by two carbon atoms through beta-oxidation. Only now does DHA exist.

Related to the ALA you ate, less than 0.5 percent arrives here. A dedicated delta-4 desaturase, long assumed to exist, is not present in humans.
Human The detour via 24 carbon atoms

Howard Sprecher described the last part of this route, which is why it carries his name. DHA does not arise through one more simple desaturation, but through a detour: lengthen to 24 carbon atoms, desaturate again, then shorten inside the peroxisome.

A group around Sheila Innis confirmed this C24 route: brain and retina possess highly selective uptake routes for n-3 fatty acids and their own mechanisms for conserving DHA there.

For you this means: your body treats DHA like a scarce building material.

Sprecher H. Biochim Biophys Acta. 2000. DOI: 10.1016/s1388-1981(00)00077-9 · Innis SM et al. Lipids. 1999. PMID: 10102240
At a glance

ALA is raw material. EPA and DHA are the finished component. The factory in between exists, but in adult humans it runs largely at idle, and the final step happens in a different building.

And now you know why a spoonful of flaxseed oil moves more at the start of the chain than at its end.

What the tracer studies actually measured

Online you will find conversion rates between 21 percent and below 0.1 percent. That looks like a contradiction. But the studies are measuring different things.

Human Young men: nothing came after DPA

A group in Southampton gave six young men ALA labelled with the isotope 13C and followed for 21 days where this molecule showed up.

About 33 percent appeared as carbon dioxide within the first 24 hours. Labelled EPA was detectable for up to seven days, labelled DPA for up to 14 days. Labelled DHA was not measurable at any point.

For you this means: in adult men the chain effectively ends at DPA.

Burdge GC, Jones AE, Wootton SA. Br J Nutr. 2002;88(4):355-363. DOI: 10.1079/BJN2002662
Human Young women: same method, different numbers

The same research group repeated the experiment in women of childbearing age, who received 700 mg of labelled ALA.

Estimated net conversion was 21 percent for EPA, 6 percent for DPA and 9 percent for DHA. About 22 percent was breathed out within 24 hours, less than in the men.

For you this means: the route is more open in women. But nine percent of a tablespoon of flaxseed oil is still little DHA. The number of participants in this work is not reported.

Burdge GC, Wootton SA. Br J Nutr. 2002;88(4):411-420. DOI: 10.1079/BJN2002689
Human Maastricht: untangling the confusion of numbers

29 healthy people ate a controlled diet for 28 days and received labelled ALA as a bolus plus small follow-up doses. From this a compartmental model of liver conversion was calculated.

Just under seven percent of dietary ALA reached the plasma phospholipid pool. Out of that pool, 99.8 percent was then turned into EPA, but only one percent continued to DPA and DHA.

For you this means: there are two bottlenecks here, one behind the other. Depending on which one a study looks at, a different percentage comes out.

Goyens PLL et al. J Lipid Res. 2005;46(7):1474-1483. DOI: 10.1194/jlr.M400514-JLR200
Why the numbers online scatter so widely

Three reference quantities get mixed up:

  • Share of all ingested ALA. Smallest numbers: around 5 percent for EPA, below 0.5 percent for DHA.
  • Share of the ALA that reaches the liver pool. Largest numbers, up to nearly 100 percent for EPA.
  • Net estimate from tracer curves. Sits in between, depending on model and sex.

All three are sound, they simply answer different questions. The direction is identical everywhere: EPA goes, DHA barely.

Clinical Flaxseed oil against fish oil, side by side

20 people took either fish oil with 1296 mg EPA and 864 mg DHA per day or flaxseed oil with 3510 mg ALA, for eight weeks.

On fish oil, membrane EPA rose by 300 percent and DHA by 42 percent, and the omega-3 index averaged 7.8 percent. On flaxseed oil, EPA rose to 133 percent of baseline and DHA stayed unchanged.

For you this means: same design, same duration. Flaxseed oil moves EPA a little, DHA not at all.

Cao J et al. Clin Chem. 2006;52(12):2265-2272. DOI: 10.1373/clinchem.2006.072322
Reframe

The common reading is: The studies contradict each other, so nobody really knows.

The more precise reading is: The studies agree on the direction and disagree on the number. Disagreement about the second decimal place is not disagreement about the message.

And now you know why two serious sources can quote different percentages.

Why women convert differently, and what estrogen has to do with it

When two people eat the same and only one of them has a decent omega-3 index, that is not a measurement error. It is a pattern in the data. And there is one study on this that goes beyond mere observation.

Clinical Not only correlation: the hormone study

A Dutch group compared the DHA status of 72 men and 103 women on an identical, controlled diet. It then studied 56 people on feminising and 61 on masculinising hormone therapy, in part randomised.

The DHA share was 15 percent higher in the women. Oral ethinylestradiol raised DHA by 42 percent, transdermal 17-beta-estradiol did not. Parenteral testosterone lowered DHA by 22 percent.

For you this means: here something was not only observed, here something was changed on purpose. Estrogen appears to favour the conversion, testosterone appears to dampen it.

Giltay EJ et al. Am J Clin Nutr. 2004;80(5):1167-1174. DOI: 10.1093/ajcn/80.5.1167

A second piece fits with this: according to Burdge and Calder, women use a smaller share of ALA for beta-oxidation. I still stay careful. These data come from special groups on hormone therapy. Transferring them to the natural cycle is an interpretation, not a measured fact. More on this in the article on cycle based nutrition and on the blood sugar and hormone axis in women.

Reframe

The tempting reading is: As a woman I need to worry less about EPA and DHA.

The sturdier reading is: The sex difference mainly explains why the numbers in the literature scatter. Nine percent conversion of a small amount of ALA is still a small amount of DHA. And in pregnancy and breastfeeding, the need for DHA rises.

And now you know why a single blanket percentage for all people could never work.

Your enzyme is not yours alone: the linoleic acid competition

This part interests me most in clinical work. It is not about your omega-3 at all, but about what lies next to it on the plate.

Delta-6 desaturase, the bottleneck at station 1 and station 4, is not picky. It takes ALA from the omega-3 family and linoleic acid from the omega-6 family. Both stand at the same counter.

Linoleic acid is the most common polyunsaturated fatty acid in a western diet. Sunflower oil, safflower oil and many processed foods deliver it in amounts that never existed in evolutionary terms. The obvious conclusion would be: less omega-6, more conversion. That conclusion is too simple.

Clinical Absolute amounts beat the ratio

29 people ate a control diet for four weeks, then for six more weeks either continued the control, or a diet low in linoleic acid, or one higher in ALA. The clever part: both intervention diets had the same ratio of 1 to 7.

In the group with less linoleic acid, ALA incorporation into phospholipids rose by 3.6 percent. In the group with more ALA, percentage incorporation fell by 8.0 percent. Further processing from EPA to DPA and DHA stayed below 0.1 percent of dietary ALA throughout.

For you this means: same ratio, opposite outcome. What mattered were the absolute amounts of both fatty acids, not their quotient.

Goyens PLL et al. Am J Clin Nutr. 2006;84(1):44-53. DOI: 10.1093/ajcn/84.1.44
Reframe

The popular thinking is: I have to bring my omega-6 to omega-3 ratio down to 4 to 1.

The more precise view is: The ratio is a good memory aid and a poor lever. You can reach the same quotient with very little or very much of both, and biologically those are not the same thing.

I would also warn against over-reading this. 3.6 percent more incorporation is real, but it is not a doubling, and the exit toward DHA stays practically closed. Anyone who changes their cooking oils does so for good reasons. It just does not answer the DHA question. More on this under fries or banana.

And now you know why it also matters how long the queue in front of your enzyme is.

Genes, ancestry and why some people do better with flaxseed oil

You may know someone who swears flaxseed oil changed something for them, while you feel nothing at all. Before you start doubting yourself: there is a biological explanation for this in the genome. The two enzymes are encoded by neighbouring genes, FADS1 for delta-5 desaturase and FADS2 for delta-6 desaturase.

Human Erfurt: 727 people, 18 gene variants

A German research group genotyped 727 adults from Erfurt for 18 polymorphisms in the FADS1 and FADS2 region and compared this with their serum phospholipids.

The associations were unusually strong, with p values below 1.0 times 10 to the power of minus 13, and they also involved the omega-3 fatty acids ALA, EPA and DPA. Carriers of the rarer alleles had more starting material in their blood and less product.

For you this means: two people can eat the same flaxseed oil and show completely different blood patterns. That is not an excuse, it is an argument for measuring instead of guessing.

Schaeffer L et al. Hum Mol Genet. 2006;15(11):1745-1756. DOI: 10.1093/hmg/ddl117

A review article puts the genetically explained variance for arachidonic acid levels at 28.5 percent. That figure applies explicitly to the omega-6 branch, not to DHA. But it does show the order of magnitude: for a diet dependent blood value, that is a great deal.

A Swedish group studied 5652 people from five European cohorts and found two common FADS haplotypes that differ markedly in how well they build long chain fatty acids. The more efficient one arose after the lineages of modern humans and Neanderthals separated, and it shows signs of positive selection. So the ability to build long chain fatty acids from precursors is itself an adaptation. On current data, though, even the more efficient variant is not enough to cover the DHA requirement from flaxseed oil alone.

Please do not shorten this

An American study found that 79 to 82 percent of the African American participants carried two copies of an allele associated with stronger conversion, compared with 42 to 45 percent of the participants of European descent. These data concern the omega-6 branch, however. Ancestry shifts a probability, not the biology of the individual enzyme.

And there is one more level above this. A study of 144 human liver samples found allele specific DNA methylation between FADS1 and FADS2, clearly associated with FADS1 activity. So between your gene and your enzyme performance sits another layer of regulation. For me this is one of the most honest moments in this whole topic. That the effect is not just a snapshot is shown by a third piece of work: 122 men were followed for three years, and carriers of one particular variant ended up with lower levels of EPA and DPA.

Reframe

The discouraging reading is: My genes are bad, so I might as well give up.

The more usable reading is: Your genes are not a verdict, they are a probability. That is exactly why a measured value tells you more than any rule of thumb. Whoever converts poorly can bypass the conversion by taking in the end product.

And now you know why your neighbour's experience with flaxseed oil says little about you.

What EPA and DHA do in your body that ALA cannot

Up to here it has been about amounts, now it is about jobs. The question is not whether EPA and DHA are rarer than ALA, but whether they hold roles that ALA cannot take over.

DHA changes the physics of your cell envelopes

DHA with its six double bonds is a biophysical extreme case. A review describes what its incorporation into membrane phospholipids sets off: altered order of the acyl chains, altered fluidity, different permeability, different activity of membrane proteins. My metaphor for this: the difference between a rigid plastic window and one that can be opened. Membrane fluidity can co-determine how well receptors change their shape and how signals travel inward. Why this matters for fat metabolism is covered in the article on zone 2 training.

From EPA and DHA come the brakes on inflammation

Animal Resolution is not simply running out

Charles Serhan described a class of substances that arises from EPA and DHA: resolvins, protectins and maresins, the specialised pro-resolving mediators.

His central finding: the fading of an inflammation is not a passive running out, but an actively steered process. Synthetic versions of these mediators showed strong inflammation-resolving effects in animal experiments.

For you this means: EPA and DHA are the raw material of this resolution chemistry, ALA is not. For context: the evidence of effect comes from animal experiments, clinical endpoint studies in humans are still outstanding.

Serhan CN. Nature. 2014;510(7503):92-101. DOI: 10.1038/nature13479

This is the strongest functional difference between the three forms. An inflammation that is not actively brought to an end stays quietly in the background, a theme in the article on silent inflammation.

DHA sits where it is hardest to replace

In the retina, DHA is a main structural lipid of the outer segment membranes of the photoreceptors and can shape permeability, fluidity and thickness there, and according to a review on the retina it may thereby shape phototransduction itself. It is hard to make the difference between precursor and end product more concrete: if your body is to see, it needs DHA in the retinal membrane, not ALA in the blood.

Human What is missing on a purely plant based diet

A British group compared 23 vegans with 24 omnivores aged between 40 and 70, matched for age and BMI.

In the vegans, not only the long chain n-3 fatty acids were significantly lower, but so were the lipid mediators derived from them in plasma.

For you this means: it is not only the fatty acids that are missing, but the signalling molecules made from them. Honesty requires adding: this is a cross-sectional study with 47 people.

Pinto AM et al. Br J Nutr. 2017;117(5):669-685. DOI: 10.1017/S0007114517000629
At a glance

ALA can supply energy and it is a precursor. EPA and DHA are building material and signalling raw material at once. An 18 carbon fatty acid can take on these two roles only to the narrow degree to which the body converts it. And in adults that degree is small.

And now you know why this is not about quantity, but about function.

Where EPA and DHA sensibly come from, and why fish size matters

If the conversion is narrow, two routes remain: small fatty fish such as sardine, anchovy, herring and mackerel, or microalgae oil. Here I part ways with many guides. I do not recommend large predatory fish, no salmon, no tuna, not even in organic quality. There is a physical reason for that.

Human Mercury follows length, not the label

An international team measured mercury in 451 fish from 92 species across 40 water bodies in 26 countries, analysed with a model of length, trophic level, location and taxonomy.

Predators at higher trophic levels had the highest values on average. The relationship between mercury and body size as well as trophic level was positive and significant.

For you this means: the rule follows from accumulation along the food chain, not from a certificate. For context: this work covers mostly inland waters and supports the principle, not specific values for marine fish.

Buck DG et al. Sci Total Environ. 2019;687:956-966. DOI: 10.1016/j.scitotenv.2019.06.159

An organic label says something about feed, husbandry and stocking density. It says nothing about how far up the food web an animal sits and how long it lived. Yet both determine the mercury burden. Small schooling fish therefore answer two questions at once: plenty of EPA and DHA at a low trophic level. The European authority also weighs the benefit and risk of eating fish species by species rather than judging across the board.

For everyone eating plant based: algae oil is not a fallback

It matters to me to say this without any undertone. Plant based eating has many well documented advantages. This article names one gap, and there is an equally plant based answer to it.

Clinical 104 vegetarians, eight weeks, one blood value

A Munich university group randomised 104 healthy vegetarians double blind to microalgae oil with 0.94 g DHA per day or olive oil as placebo, over eight weeks.

The omega-3 index rose from 4.8 to 8.4 weight percent. 69 percent of the DHA group went above 8 percent, nobody in the placebo group did. EPA rose considerably less.

For you this means: eating plant based and a good omega-3 status do not exclude each other. The secondary finding speaks for checking the label for combined EPA and DHA oils.

Geppert J et al. Lipids. 2005;40(8):807-814. DOI: 10.1007/s11745-005-1442-9

This was also tested directly: 32 adults received 600 mg DHA per day for two weeks, once from algae oil capsules and once from cooked salmon. DHA rose by about 80 percent in plasma phospholipids in both groups, with no significant difference. Transparency on this: the author group had ties to an algae oil manufacturer, but the same conclusion is supported by the independent Munich group.

The omega-3 index makes the question measurable

One blood value maps exactly this: the share of EPA plus DHA in the red blood cell membrane. It shows what has landed in your membranes, regardless of how much flaxseed oil you bought.

In a dose response study over five months, 115 adults received 0, 300, 600, 900 or 1800 mg EPA plus DHA per day. The index rose with the dose, and dose alone explained 68 percent of the variability. That is the contrast with ALA: with direct intake, the response is far more predictable. Everything on measurement methods and target ranges is in the article on the omega-3 index, and a broader overview of sources under omega-3 from plants, animals or algae.

From clinical practice

What I talk about with people standing exactly here

Clinically I observe one thing above all: the disappointment does not come from too little effort, it comes from mixing up precursor and end product. The direction that follows from the data is simple: measure your status instead of guessing it, choose your source by molecular form instead of by label advertising, and with fish think about size. Which amount makes sense for you belongs in a conversation with your doctor.

If you want to work this out for yourself rather than only read about it: below this article you will find the option to book an appointment.

And now you know why I ask about body length when it comes to fish, and not about the certificate.

What ALA can still do, and what the large heart trials show

Now it gets uncomfortable, and specifically for my own argument. This article moves on the level of molecules, enzymes and blood values. Whether more EPA and DHA leads to fewer heart attacks in the end is a different question. There the data are mixed.

Clinical Cochrane: 86 studies, almost 163,000 people

A Cochrane team analysed 86 randomised studies with 162,796 participants, separately for long chain n-3 fatty acids and for ALA.

More long chain n-3 had little effect on all-cause mortality, but may have slightly lowered coronary mortality and lowered triglycerides by about 15 percent. More ALA probably slightly lowered cardiac arrhythmias.

For you this means: ALA is not worthless, quite the opposite. And capsules with EPA and DHA are not heart protection on prescription.

Abdelhamid AS et al. Cochrane Database Syst Rev. 2020;3:CD003177. DOI: 10.1002/14651858.CD003177.pub5

Three large individual trials show how differently results can turn out, depending on dose, molecular form and baseline risk.

TrialWhat was givenResult at the primary endpoint
VITAL
25,871 people
1 g per day of marine n-3 versus placebo, median 5.3 years No difference in major events. At the secondary endpoint of heart attack, an advantage appeared.
REDUCE-IT
8,179 people, high risk
4 g per day of icosapent ethyl, that is highly purified EPA, median 4.9 years Clear reduction, 17.2 versus 22.0 percent. At the same time more hospital admissions for atrial fibrillation, 3.1 versus 2.1 percent.
STRENGTH
13,078 people, high risk
4 g per day of EPA and DHA as carboxylic acid form versus corn oil, stopped early No difference, 12.0 versus 12.2 percent.

These findings cannot be fully reconciled. Differences in molecular form, comparator substance and baseline risk are being discussed. What matters to me is the safety signal: at very high doses, atrial fibrillation was observed more often in REDUCE-IT. So more is not automatically better, and the European authority has assessed the upper intake level in a dedicated opinion.

Reframe

The shortened reading is: The trials are negative, so I can skip this whole topic.

The more precise reading is: A marker is not an endpoint, and a supplement is not a diet. Whether a capsule prevents events in a broad population is something different from the question of whether your body is supplied with a building material it cannot make itself.

And now you know why I do not talk flaxseed oil down and still do not call it the omega-3 solution.

Frequently asked questions about ALA, EPA and DHA

What is the difference between ALA, EPA and DHA?

All three are omega-3 fatty acids, but they are different molecules. ALA, alpha-linolenic acid, has 18 carbon atoms and three double bonds and sits in flaxseed, chia, walnuts and rapeseed oil. EPA has 20 carbon atoms and five double bonds, DHA has 22 and six. Only EPA and DHA are built into cell membranes in meaningful amounts, and only from them do the inflammation-resolving signalling molecules arise. ALA is the plant based precursor from which both would first have to be made.

What percentage of ALA does the body convert into EPA and DHA?

The anchor values from a review article are roughly five percent for the route to EPA and less than 0.5 percent for the route to DHA. Individual studies land well above or well below that, because they use different reference quantities and very small groups. In a tracer study with six young men, labelled DHA was not measurable at all across 21 days. The direction is the same across all of this work, the exact number is not.

Is flaxseed oil enough to cover your omega-3 needs?

For your intake of ALA yes, for your supply of EPA and DHA the current data say no. In a direct comparison over eight weeks, flaxseed oil lifted EPA in the red blood cell membrane to about 133 percent of baseline, while DHA stayed unchanged. Fish oil in the same design lifted EPA by about 300 percent and DHA by 42 percent. Flaxseed oil remains a good food. It is simply not an answer to the DHA question.

Why do the figures for the conversion rate differ so much online?

Because the studies do not measure the same thing. Some report the share of all ingested ALA, others the share of the ALA that reaches the phospholipid pool of the liver at all. In one compartmental study, only about seven percent of dietary ALA reached that pool. Out of the pool, 99.8 percent then went on to EPA, but only one percent continued to DPA and DHA. Both numbers are correct, they simply have different denominators.

Do women convert ALA better than men?

The data point in that direction. In a tracer study in young women, the estimated net conversion was 21 percent for EPA and 9 percent for DHA, while in young men labelled DHA was not measurable at all. In a study using targeted hormone administration, the DHA share was 15 percent higher in women than in men, oral ethinylestradiol raised it by 42 percent and parenteral testosterone lowered it by 22 percent. So estrogen appears to favour the conversion. That is not a free pass, because nine percent of a tablespoon of flaxseed oil is still little DHA.

Does omega-6 block the conversion of omega-3?

Block is too strong a word, compete fits better. Linoleic acid from sunflower, safflower and corn oil and ALA from flaxseed oil go to the same enzyme, delta-6 desaturase. If you eat a lot of linoleic acid, you occupy that bottleneck more heavily. In a controlled diet study, the incorporation of ALA into phospholipids rose by 3.6 percent when linoleic acid intake was lowered. That is a measurable shift, but not a dramatic one.

Does the omega-6 to omega-3 ratio matter, or the absolute amounts?

According to the cleanest study on this so far, the absolute amounts. In a randomised diet study, two intervention diets had the same ratio of 1 to 7 but produced opposite results: less linoleic acid raised ALA incorporation, more ALA lowered the percentage incorporation. The popular arithmetic with the ratio is a good memory aid and a poor lever. And even under optimised conditions, further processing to DHA stayed below 0.1 percent.

Is algae oil as good as fish oil?

For DHA the data suggest yes. In a randomised study with 32 adults, both groups received 600 mg DHA per day for two weeks, once from algae oil capsules and once from cooked salmon. DHA rose by about 80 percent in plasma phospholipids in both groups, with no difference between the sources. What matters is the label: some algae oils deliver almost only DHA, while combined EPA and DHA oils cover both forms.

Which fish has plenty of omega-3 and little mercury?

The small, fatty schooling fish that sit low in the food chain: sardine, anchovy, herring and mackerel. In a worldwide analysis of 451 fish from 92 species, mercury content rose significantly with body length and with trophic level, meaning position in the food web. Small fish deliver plenty of EPA and DHA and at the same time carry a considerably lower mercury burden.

Why do you advise against salmon and tuna, even organic salmon?

Because a certification does not shift a position in the food web. Mercury accumulates along the food chain and rises with body length and trophic level, not with the label. This is a precautionary judgement, not a statement that salmon is harmful. The European authority weighs the benefit and risk of eating fish explicitly species by species rather than across the board. Anyone who weighs it differently is doing so on reasonable grounds.

How much EPA and DHA per day makes sense?

The European authority sets its own intake value for EPA plus DHA in adults, separate from ALA. In a randomised dose response study over five months, 115 adults received 0, 300, 600, 900 or 1800 mg EPA plus DHA per day, and the omega-3 index rose with the dose. Dose alone explained 68 percent of the response, together with body weight 70 percent. So one standard amount does not fit everyone, and the amount that makes sense for you belongs in a conversation with your doctor.

Can I reach a good omega-3 index as a vegan?

Yes, with microalgae oil. In a double blind randomised study, 104 vegetarians took either microalgae oil with 0.94 g DHA per day or olive oil as placebo for eight weeks. The omega-3 index rose from 4.8 to 8.4 weight percent, 69 percent of the DHA group went above 8 percent, and nobody in the placebo group did. EPA rose considerably less, which speaks for combined EPA and DHA products. Plant based eating and a good omega-3 status do not exclude each other, it simply takes the right molecular form.

Omega-3 in the bigger picture

The question of ALA, EPA and DHA hangs on three other topics: on measurement, on inflammation biology and on the toxicology of the food chain. From here, routes lead into those areas.

SJ

Shukri Jarmoukli

Physician, Integrative Medicine · ViveCura Berlin

In my private practice I work at the intersection of conventional medicine, functional medicine and Clinical Psychoneuroimmunology. With nutrients I am less interested in what the package says and more in which molecular form actually arrives in the cell membranes, and what is driving the inflammatory load in the background.

This article does not replace medical advice. It is meant to help you ask better questions before you buy or take something.

ViveCura, Privatpraxis Shukri Jarmoukli, Skalitzer Straße 137, 10999 Berlin

Scientific sources

  1. Burdge GC, Jones AE, Wootton SA. Eicosapentaenoic and docosapentaenoic acids are the principal products of alpha-linolenic acid metabolism in young men. Br J Nutr. 2002;88(4):355-363. DOI: 10.1079/BJN2002662 · PMID: 12323085 [Case Series, n=6]
  2. Burdge GC, Wootton SA. Conversion of alpha-linolenic acid to eicosapentaenoic, docosapentaenoic and docosahexaenoic acids in young women. Br J Nutr. 2002;88(4):411-420. DOI: 10.1079/BJN2002689 · PMID: 12323090 [Case Series, sample size not reported]
  3. Burdge GC, Calder PC. Conversion of alpha-linolenic acid to longer-chain polyunsaturated fatty acids in human adults. Reprod Nutr Dev. 2005;45(5):581-597. DOI: 10.1051/rnd:2005047 · PMID: 16188209 [Mechanism Review]
  4. Plourde M, Cunnane SC. Extremely limited synthesis of long chain polyunsaturates in adults. Appl Physiol Nutr Metab. 2007;32(4):619-634. DOI: 10.1139/H07-034 · PMID: 17622276 [Mechanism Review]
  5. Brenna JT, Salem N, Sinclair AJ, Cunnane SC. Alpha-linolenic acid supplementation and conversion to n-3 long-chain polyunsaturated fatty acids in humans. Prostaglandins Leukot Essent Fatty Acids. 2009;80(2-3):85-91. DOI: 10.1016/j.plefa.2009.01.004 · PMID: 19269799 [Systematic Review]
  6. Arterburn LM, Hall EB, Oken H. Distribution, interconversion, and dose response of n-3 fatty acids in humans. Am J Clin Nutr. 2006;83(6 Suppl):1467S-1476S. DOI: 10.1093/ajcn/83.6.1467S · PMID: 16841856 [Systematic Review]
  7. Goyens PLL, Spilker ME, Zock PL, Katan MB, Mensink RP. Conversion of alpha-linolenic acid in humans is influenced by the absolute amounts of alpha-linolenic acid and linoleic acid in the diet and not by their ratio. Am J Clin Nutr. 2006;84(1):44-53. DOI: 10.1093/ajcn/84.1.44 · PMID: 16825680 [RCT, n=29]
  8. Goyens PLL, Spilker ME, Zock PL, Katan MB, Mensink RP. Compartmental modeling to quantify alpha-linolenic acid conversion after longer term intake of multiple tracer boluses. J Lipid Res. 2005;46(7):1474-1483. DOI: 10.1194/jlr.M400514-JLR200 · PMID: 15834128 [Case Series, n=29]
  9. Cao J, Schwichtenberg KA, Hanson NQ, Tsai MY. Incorporation and clearance of omega-3 fatty acids in erythrocyte membranes and plasma phospholipids. Clin Chem. 2006;52(12):2265-2272. DOI: 10.1373/clinchem.2006.072322 · PMID: 17053155 [RCT, n=20]
  10. Giltay EJ, Gooren LJG, Toorians AWFT, Katan MB, Zock PL. Docosahexaenoic acid concentrations are higher in women than in men because of estrogenic effects. Am J Clin Nutr. 2004;80(5):1167-1174. DOI: 10.1093/ajcn/80.5.1167 · PMID: 15531662 [RCT, n=117 in the intervention arm]
  11. Schaeffer L, Gohlke H, Müller M, et al. Common genetic variants of the FADS1 FADS2 gene cluster and their reconstructed haplotypes are associated with the fatty acid composition in phospholipids. Hum Mol Genet. 2006;15(11):1745-1756. DOI: 10.1093/hmg/ddl117 · PMID: 16670158 [Cohort, n=727]
  12. Lattka E, Illig T, Koletzko B, Heinrich J. Genetic variants of the FADS1 FADS2 gene cluster as related to essential fatty acid metabolism. Curr Opin Lipidol. 2010;21(1):64-69. DOI: 10.1097/MOL.0b013e3283327ca8 · PMID: 19809313 [Mechanism Review]
  13. Ameur A, Enroth S, Johansson Å, et al. Genetic adaptation of fatty-acid metabolism: a human-specific haplotype increasing the biosynthesis of long-chain omega-3 and omega-6 fatty acids. Am J Hum Genet. 2012;90(5):809-820. DOI: 10.1016/j.ajhg.2012.03.014 · PMID: 22503634 [Cohort, n=5,652]
  14. Mathias RA, Sergeant S, Ruczinski I, et al. The impact of FADS genetic variants on omega-6 polyunsaturated fatty acid metabolism in African Americans. BMC Genet. 2011;12:50. DOI: 10.1186/1471-2156-12-50 · PMID: 21599946 [Cohort, n=329]
  15. Hong SH, Kwak JH, Paik JK, Chae JS, Lee JH. Association of polymorphisms in FADS gene with age-related changes in serum phospholipid polyunsaturated fatty acids and oxidative stress markers in middle-aged nonobese men. Clin Interv Aging. 2013;8:585-596. DOI: 10.2147/CIA.S42096 · PMID: 23818766 [Cohort, n=122]
  16. Howard TD, Mathias RA, Seeds MC, et al. DNA methylation in an enhancer region of the FADS cluster is associated with FADS activity in human liver. PLoS One. 2014;9(5):e97510. DOI: 10.1371/journal.pone.0097510 · PMID: 24842322 [Cohort, n=144 liver tissue samples]
  17. Sprecher H. Metabolism of highly unsaturated n-3 and n-6 fatty acids. Biochim Biophys Acta. 2000;1486(2-3):219-231. DOI: 10.1016/s1388-1981(00)00077-9 · PMID: 10903473 [Mechanism Review]
  18. Innis SM, Sprecher H, Hachey D, Edmond J, Anderson RE. Neonatal polyunsaturated fatty acid metabolism. Lipids. 1999;34(2):139-149. DOI: 10.1007/s11745-999-0348-x · PMID: 10102240 [Mechanism Review]
  19. Stillwell W, Wassall SR. Docosahexaenoic acid: membrane properties of a unique fatty acid. Chem Phys Lipids. 2003;126(1):1-27. DOI: 10.1016/s0009-3084(03)00101-4 · PMID: 14580707 [Mechanism Review]
  20. Serhan CN. Pro-resolving lipid mediators are leads for resolution physiology. Nature. 2014;510(7503):92-101. DOI: 10.1038/nature13479 · PMID: 24899309 [Mechanism Review, effect data mostly In vivo, animal]
  21. SanGiovanni JP, Chew EY. The role of omega-3 long-chain polyunsaturated fatty acids in health and disease of the retina. Prog Retin Eye Res. 2005;24(1):87-138. DOI: 10.1016/j.preteyeres.2004.06.002 · PMID: 15555528 [Mechanism Review]
  22. Harris WS, von Schacky C. The Omega-3 Index: a new risk factor for death from coronary heart disease? Prev Med. 2004;39(1):212-220. DOI: 10.1016/j.ypmed.2004.02.030 · PMID: 15208005 [Cohort, analysis of published prevention studies]
  23. Flock MR, Skulas-Ray AC, Harris WS, et al. Determinants of erythrocyte omega-3 fatty acid content in response to fish oil supplementation: a dose-response randomized controlled trial. J Am Heart Assoc. 2013;2(6):e000513. DOI: 10.1161/JAHA.113.000513 · PMID: 24252845 [RCT, n=115]
  24. Skulas-Ray AC, Flock MR, Richter CK, et al. Red blood cell docosapentaenoic acid (DPA n-3) is inversely associated with triglycerides and C-reactive protein and dose-dependently increases following n-3 fatty acid supplementation. Nutrients. 2015;7(8):6390-6404. DOI: 10.3390/nu7085291 · PMID: 26247967 [RCT, n=115 plus n=28]
  25. Arterburn LM, Oken HA, Bailey Hall E, et al. Algal-oil capsules and cooked salmon: nutritionally equivalent sources of docosahexaenoic acid. J Am Diet Assoc. 2008;108(7):1204-1209. DOI: 10.1016/j.jada.2008.04.020 · PMID: 18589030 [RCT, n=32]
  26. Geppert J, Kraft V, Demmelmair H, Koletzko B. Docosahexaenoic acid supplementation in vegetarians effectively increases omega-3 index: a randomized trial. Lipids. 2005;40(8):807-814. DOI: 10.1007/s11745-005-1442-9 · PMID: 16296399 [RCT, n=104]
  27. Pinto AM, Sanders TAB, Kendall AC, et al. A comparison of heart rate variability, n-3 PUFA status and lipid mediator profile in age- and BMI-matched middle-aged vegans and omnivores. Br J Nutr. 2017;117(5):669-685. DOI: 10.1017/S0007114517000629 · PMID: 28366178 [Cohort, n=47, cross-sectional]
  28. Buck DG, Evers DC, Adams E, et al. A global-scale assessment of fish mercury concentrations and the identification of biological hotspots. Sci Total Environ. 2019;687:956-966. DOI: 10.1016/j.scitotenv.2019.06.159 · PMID: 31412499 [Cohort, n=451 fish from 92 species]
  29. EFSA Panel on Dietetic Products, Nutrition and Allergies. Scientific Opinion on Dietary Reference Values for fats. EFSA Journal. 2010;8(3):1461. DOI: 10.2903/j.efsa.2010.1461 [Regulatory Document]
  30. EFSA Panel on Dietetic Products, Nutrition and Allergies. Scientific Opinion on the Tolerable Upper Intake Level of eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA) and docosapentaenoic acid (DPA). EFSA Journal. 2012;10(7):2815. DOI: 10.2903/j.efsa.2012.2815 [Regulatory Document]
  31. EFSA Scientific Committee. Scientific Opinion on health benefits of seafood consumption in relation to health risks associated with exposure to methylmercury. EFSA Journal. 2014;12(7):3761. DOI: 10.2903/j.efsa.2014.3761 [Regulatory Document]
  32. Abdelhamid AS, Brown TJ, Brainard JS, et al. Omega-3 fatty acids for the primary and secondary prevention of cardiovascular disease. Cochrane Database Syst Rev. 2020;3:CD003177. DOI: 10.1002/14651858.CD003177.pub5 · PMID: 32114706 [Meta-analysis, k=86, n=162,796]
  33. Manson JE, Cook NR, Lee IM, et al. Marine n-3 Fatty Acids and Prevention of Cardiovascular Disease and Cancer (VITAL). N Engl J Med. 2019;380(1):23-32. DOI: 10.1056/NEJMoa1811403 · PMID: 30415637 [RCT, n=25,871]
  34. Bhatt DL, Steg PG, Miller M, et al. Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia (REDUCE-IT). N Engl J Med. 2019;380(1):11-22. DOI: 10.1056/NEJMoa1812792 · PMID: 30415628 [RCT, n=8,179]
  35. Nicholls SJ, Lincoff AM, Garcia M, et al. Effect of High-Dose Omega-3 Fatty Acids vs Corn Oil on Major Adverse Cardiovascular Events in Patients at High Cardiovascular Risk: The STRENGTH Randomized Clinical Trial. JAMA. 2020;324(22):2268-2280. DOI: 10.1001/jama.2020.22258 · PMID: 33190147 [RCT, n=13,078]
Transparency on the evidence

The percentages for conversion scatter considerably. That comes from different reference quantities and from very small participant groups in the tracer studies. The direction is consistent across all of this work, the exact number is not. That is why I have named the reference quantity everywhere instead of picking one number.

The omega-3 index is well studied as a risk marker. But a marker is not proof of benefit in an individual case. The large endpoint trials on omega-3 supplements come out mixed, and at very high doses atrial fibrillation was observed more often in one of those trials. Both belong here honestly.

The recommendation to avoid large fish is a judgement made from exposure data and precaution. It is not a statement that salmon is harmful. Anyone who weighs this differently is doing so on reasonable grounds. The mercury study cited also draws mostly on inland waters and supports the principle, not specific values for marine fish.

Two of the cited papers on algae oil come from the orbit of an algae oil manufacturer. Their core findings are supported by independent university groups, but the note belongs here nonetheless. On the genetically explained variance of 28.5 percent: that figure applies to arachidonic acid in the omega-6 branch and cannot be transferred to DHA.

One plus point at the end: not a single core statement in this article rests on a mouse study or a cell study. The only place with animal data in the background is the evidence of effect for the synthetic resolvins and protectins, and that is stated explicitly in the text.

Have questions or want to book an appointment?

We'd be happy to advise you personally at our practice.

Book appointment