E322 Lecithin's Real Risk Is the Choline Load, Not the Additive

Lecithin (E322) is cleared by EFSA with no numerical intake limit, and the additive itself is not where the interesting science lives. The real story is about choline, the nutrient lecithin is mostly made of, and a gut bacteria pathway that turns it into a compound cardiologists have spent over a decade arguing about. The amount of lecithin in your candy bar has almost nothing to do with that argument.

That is the actual shape of the evidence. Shoppers who stop at “lecithin is fine” or “lecithin is linked to heart disease” are both missing the more useful answer in between.

What EFSA concluded about lecithin (E322) in 2017

Start with the regulatory baseline, because it is the least controversial part of this story.

EFSA’s 2017 re-evaluation of lecithins concluded there was no safety concern at the reported uses and use levels for the general population above 1 year of age.^[1] The panel went further than most re-evaluations: it did not consider it necessary to set a numerical acceptable daily intake at all, a designation reserved for additives with a wide enough margin of safety that a numeric cap adds nothing.^[1] EFSA’s reviewers cleared it even for infants from 12 weeks of age, a population regulators are typically far more cautious with.

Lecithin is most commonly extracted from soybeans, though other natural sources such as egg yolk are also used to produce phospholipid-rich lecithin for food and pharmaceutical use.^[2] Producers purify it from the raw material using solvent extraction and chromatography rather than chemical synthesis, which is why it behaves in food the same way the phospholipids in your own cell membranes behave.^[2]

The pathway that put lecithin in the TMAO story

Lecithin is mostly phosphatidylcholine, a phospholipid built around choline.^[3] That single fact is why lecithin gets dragged into cardiovascular research that has nothing to do with food additives.

Gut bacteria can process dietary choline, including the phosphatidylcholine form found in lecithin, through the same trimethylamine pathway.^[3] Once trimethylamine reaches the liver, the enzyme flavin-containing monooxygenase-3 oxidizes it into trimethylamine-N-oxide, or TMAO, the molecule that shows up in every headline about eggs, meat and heart disease.^[4]

The foundational evidence for this pathway comes from mice. Researchers found that dietary supplementation with choline, TMAO or betaine promoted atherosclerosis in mice, and that germ-free mice, lacking the gut bacteria needed to produce TMAO, were protected from the effect.^[5] That mechanism is real and well replicated in mice. It does not, by itself, tell you what a food additive does inside a human body.

The mechanism is real, the leap to your food is where people get it wrong. A pathway that exists in mice fed choline does not automatically tell you what a fraction of a gram of lecithin in your chocolate bar does to you.

The one human cohort behind the TMAO scare, and what it doesn’t cover

The human evidence is thinner and more specific than the mouse work, and it is worth being precise about what it measured.

A prospective cohort of 4,007 patients undergoing coronary angiography found that higher fasting plasma TMAO levels, produced through gut-microbial metabolism of dietary phosphatidylcholine, were associated with an increased risk of death, heart attack or stroke over three years of follow-up.^[6] The same study showed that swallowing a phosphatidylcholine challenge (two hard-boiled eggs plus a labeled tracer dose) reliably spiked plasma TMAO in the hours afterward.^[6]

4,007 patients in the cohort linking plasma TMAO to cardiovascular risk PMID 23614584

That is an observational association between a blood marker and an outcome. It is not a controlled trial of lecithin intake, so it cannot, by itself, establish that dietary lecithin causes heart disease, because the patients were not randomized to eat more or less phosphatidylcholine. The cohort tells you TMAO tracks with risk. It does not tell you that adding lecithin to your diet raises that risk, or that removing it would lower it.

In fact, the one randomized controlled trial that comes close to testing that question found something more limited. Adults with metabolic syndrome given a soy-lecithin-enriched beverage for two weeks showed no worse endotoxemia, gut-barrier markers or cardiometabolic risk markers than a comparison group given a different phospholipid source.^[7] That is a short-term biomarker trial. It measured neither heart attacks nor strokes, which is exactly the point: the event-level evidence on lecithin and cardiovascular outcomes in humans stays observational, because nobody has run the harder trial.

The dose gap nobody selling this story mentions

Here is where the additive itself, as opposed to the nutrient inside it, matters.

We could not verify a published clinical trial that tested soy lecithin supplementation specifically against blood cholesterol at a defined gram-scale dose. If that research exists, it did not surface in a citation search built around this exact question, so it is left out rather than guessed at.

What is verifiable is simpler: the lecithin used to keep chocolate or margarine from separating is added at a fraction of a percent of the product’s weight, an amount set by what the emulsion needs rather than by any safety ceiling. The cardiovascular research discussed above used a full phosphatidylcholine challenge, two eggs’ worth in a single dose, or a supplemented beverage measured in grams. Nobody has shown that the milligram-scale amount doing emulsifying work in a cookie behaves the same way.

It is the same principle behind why an orange and a vitamin C tablet are not interchangeable in nutrition research: the nutrient can be chemically identical while the dose, the matrix and the delivery route change what your body does with it. For a deeper look at which additives do have real trial evidence behind the concern and which don’t, see food additives to avoid: the two with real trial evidence.

The soy allergy question

This is the one place a real, sourced downside might exist, and it is also the place the evidence runs out fastest.

Soy is one of the major allergen groups that most countries, including EU member states, require to be declared on food labels.^[8] That labeling requirement is what protects a soy-allergic shopper, not an assumption about how refined the lecithin is.

We looked for research measuring whether refining strips enough soy protein out of commercial lecithin to make it safe for soy-allergic people, and for trials testing soy-allergic patients against refined lecithin directly. Neither turned up a citation that would support a specific safety claim. Here is what that leaves established, and what it leaves open:

• What’s established: soy lecithin is legally required to be labeled, so avoidance is straightforward for anyone who wants it.
• What isn’t established here: a specific, cited figure for residual protein in refined lecithin, or a clinical study confirming most soy-allergic people tolerate it.
• The practical workaround: sunflower lecithin performs the same emulsifying job as soy lecithin and is increasingly used as a non-soy, non-GMO alternative, which sidesteps the question entirely for anyone who would rather not test it.

If you have a diagnosed soy allergy, don’t rely on refining as your safety margin. Check the label, and reach for a product using sunflower lecithin if soy-free matters to you.

Why the causal debate over TMAO isn’t settled

Even within the research community that built this pathway, the causal question stays open.

Researchers reviewing the field describe TMAO’s role in cardiovascular disease as still questioned: is it a mediator that drives disease, or a bystander marker of some other underlying dietary and metabolic pattern?^[9] That is not a fringe objection. It is the standard caveat in the field’s own literature.

Part of the reason the picture stays muddled is that people do not produce TMAO uniformly from the same food. A randomized crossover trial giving 37 healthy men an identical 600 mg choline dose, either as choline bitartrate or as phosphatidylcholine, found gut microbiota composition differed sharply between high-TMAO producers and low-TMAO producers, with high producers carrying more Clostridium lineages from two specific bacterial families.^[3] The same trial found something that undercuts the simple “lecithin raises TMAO” framing directly: phosphatidylcholine, the form found in lecithin, produced no significant TMAO rise compared with a no-choline control, while free choline (choline bitartrate) produced roughly three times the TMAO spike.^[3]

3x greater TMAO spike from free choline than from phosphatidylcholine, the form in lecithin, at an identical dose PMID 32722424

That single trial is the most useful fact in this entire article, because it is the one place the mechanism and the additive meet, and they don’t line up the way the scare version of this story assumes.

The bottom line

Lecithin (E322) clears every regulatory bar EFSA has set for it, with no numerical intake limit and no flagged safety concern at any age group reviewed.^[1] The additive is not the source of the TMAO conversation that keeps attaching itself to lecithin online.

That conversation is really about dietary choline broadly, tested with egg-sized doses in a cardiac cohort and gram-scale supplements in short trials, not with the fraction of a percent used to stabilize a chocolate bar. Where a real, if narrower, TMAO-lecithin connection could exist, the best available trial found phosphatidylcholine behaves differently from the free choline that drives the effect.

If you are avoiding lecithin because of the heart-disease headlines, you are avoiding the wrong ingredient for the wrong reason. If you are avoiding it because of soy allergy, check the label and reach for a sunflower-lecithin product, because that is the one piece of this story with a straightforward, verifiable fix. Scanning the label for the source rather than the E-number is where an app like EatMatrix earns its keep.

This article summarises publicly available regulatory assessments and peer-reviewed research about food ingredients. It is general information and is not intended to diagnose, treat, cure, or prevent any disease, and it is not personalised dietary or medical advice. Regulatory positions change; the EatMatrix app reflects the current assessments in our database. If you have an allergy, a medical condition, or are pregnant, talk to a qualified clinician or dietitian before changing what you eat.