Erythritol and the heart: what the 2023 studies really showed
Erythritol is a near-zero-calorie sugar alcohol found in many keto, diabetic and sugar-free products. In 2023 it made headlines when a Cleveland Clinic study linked higher blood levels to heart attacks and strokes, and showed it could make platelets clot more easily. The science is real and worth understanding, but more tangled than the coverage suggested, not least because your own body makes erythritol from sugar. This guide separates what the evidence shows from what it does not.
Key facts
- Erythritol is a sugar alcohol (polyol), roughly 60 to 80 percent as sweet as table sugar, with about 0.2 kcal/g (labelled as zero in the EU) and almost no effect on blood glucose or insulin.2
- It is not really "artificial": it occurs in fruit and fermented foods, is produced commercially by fermenting glucose with food-grade yeast, and your body also makes it from glucose internally.34
- A 2023 Cleveland Clinic study (Witkowski et al., Nature Medicine) found higher blood erythritol tracked with 3-year heart attack, stroke and death, with adjusted hazard ratios of about 1.8 (US cohort) and 2.2 (European cohort) comparing the top to the bottom quartile.1
- The same paper showed erythritol made platelets clot more readily in the lab and in mice, and a single 30 g drink kept blood levels raised for more than two days in 8 volunteers.1
- This is an association plus a mechanism, not proof of cause. EFSA found no established causal link to heart disease, and erythritol remains an approved sweetener (E 968) in both the EU and the UK.3
On this page
What erythritol actually is
Erythritol belongs to a family of sweeteners called sugar alcohols, or polyols, which also includes sorbitol, xylitol and mannitol. It is about two-thirds as sweet as sugar, carries roughly 0.2 kcal/g (low enough that EU labelling rounds it to zero), and has a glycaemic index close to zero, so it has little effect on blood glucose or insulin.2 That combination is why it dominates the "sugar-free" and ketogenic aisles, often blended with stevia or monk fruit to mask its faint cooling aftertaste.
The word "artificial" in the original study title is a little misleading. Erythritol occurs naturally in fruits such as grapes, melons and pears and in fermented foods, and the commercial product is made by fermenting glucose with food-grade yeasts, not by chemical synthesis. EFSA defines it precisely as a substance "obtained by fermenting carbohydrate sources" with yeasts such as Moniliella pollinis.3 One quirk matters for what follows: about 90 percent of ingested erythritol is absorbed in the small intestine, circulated, and then excreted largely unchanged in the urine. Very little reaches the colon, which is why it is gentler on the gut than most polyols.3
The twist: your body makes it too
Here is the fact that reframes the whole debate. Humans do not only eat erythritol, we manufacture it. The body converts glucose into erythritol through a branch of the pentose phosphate pathway, a routine part of sugar metabolism. This was shown clearly in a 2017 study in PNAS that followed 264 university students across their first year. Those who went on to gain weight and abdominal fat had about 15-fold higher blood erythritol at the start of the year, and isotope experiments confirmed the erythritol was being made internally from glucose.4
The implication is important. Your blood erythritol level is partly a window onto how your body is handling sugar. People with poorer glucose control, higher body weight or diabetes may simply produce more of it themselves, regardless of what they eat. So a high blood level can be a sign of metabolic strain rather than a measure of how much sweetener has passed your lips. Hold that thought, because it is the key to reading the heart-risk study fairly.
The 2023 Cleveland Clinic study
The paper that started the headlines was published in Nature Medicine in February 2023 by Marco Witkowski, Stanley Hazen and colleagues at the Cleveland Clinic.19 It is best understood as three separate pieces of work stitched together, each answering a different question.
| Strand | What was done | What it shows | Main limitation |
|---|---|---|---|
| Observational | Blood erythritol measured in cardiac patients: discovery cohort (n=1,157) plus US (n=2,149) and European (n=833) validation cohorts, tracked for 3-year events1 | Higher erythritol tracked with more heart attacks, strokes and deaths (top vs bottom quartile adjusted hazard ratios 1.80 and 2.21) | Association, not cause; measured blood levels, not intake |
| Mechanistic | Erythritol added to human platelets in the lab, and to mice with an induced artery injury1 | Erythritol made platelets more reactive and sped up clot formation | Cell and animal models; effects at chosen concentrations |
| Interventional | 8 healthy volunteers drank 30 g of erythritol; blood levels followed for a week1 | Blood erythritol rose sharply and stayed above clot-relevant thresholds for over 2 days | Tiny sample; surrogate marker, not actual heart events |
Put together, the picture is coherent and that is why it landed: people with more erythritol in their blood had more cardiovascular events over three years, erythritol can make platelets stickier, and a realistic dose pushes blood levels into the range where that stickiness was seen. The hazard ratios were adjusted for age, diabetes, blood pressure, cholesterol and smoking.1
How strong is the evidence?
This is where a calm reading matters, because the same data carry a second, less alarming interpretation. There are three honest caveats, and the third is the big one.
First, the cohort data are observational. They show association, not causation. High blood erythritol travels in the company of diabetes, obesity and insulin resistance, which are themselves powerful drivers of heart disease, so confounding is very plausible even after statistical adjustment.
Second, and most important, reverse causation is a live possibility. Because the body makes erythritol from glucose, a high blood level may be a consequence of poor glucose handling rather than a cause of disease.46 The cohort samples were taken from fasting cardiac patients, many enrolled before erythritol was a common sweetener, so the erythritol measured almost certainly reflects mostly what their bodies produced, not what they consumed. The study measured blood erythritol, never erythritol intake. A commentary in Clinical Chemistry by Heianza, Qi and Manson put the question directly in its title: is it the sweetener, or the circulating metabolite, that carries the risk?6
Third, the human dose experiment was small and indirect: eight people, and the outcome was a laboratory marker of platelet activity, not a single heart attack. It shows a 30 g bolus raises blood levels into a flagged range. It does not show that everyday erythritol use causes clots in real people.
Evidence strength: the cardiovascular signal rests on observational cohorts (consistent across the US and Europe) plus mechanistic lab and animal work. That is a genuine, reproducible association with a plausible biological mechanism. What is missing is the link that would prove cause: a trial showing that consuming erythritol, rather than merely having high blood levels, raises cardiovascular events. The authors themselves concluded that long-term safety studies are warranted.1
The xylitol sequel and a fresh cohort
The story did not stop in 2023. In 2024 the same Cleveland Clinic group published a near-identical investigation of xylitol, another common sugar alcohol, in the European Heart Journal. In more than 3,000 patients, higher blood xylitol again tracked with three-year cardiovascular risk; the preclinical work again showed clot-promoting effects; and in 10 volunteers a 30 g xylitol drink raised plasma levels around 1,000-fold within 30 minutes, with platelet activity rising in every person.5 A consistent pattern across two polyols makes the mechanistic concern harder to dismiss.
Equally, an independent line of evidence keeps the reverse-causation door open. In 2025 a study using the long-running ARIC community cohort (4,006 older adults with no prior cardiovascular disease) found that higher erythritol, and its metabolite erythronate, tracked with worse cardiometabolic profiles and more cardiovascular events.7 Crucially, that paper framed circulating erythritol as a marker of cardiometabolic health, which sits as comfortably with the "it is a sign of metabolic trouble" reading as with the "it causes harm" one. The association is real and reproducible. Its direction of travel is still not settled.
The gut-tolerance point
One effect of erythritol is not in dispute at all. At higher single doses it causes gastrointestinal symptoms: rumbling, nausea and a laxative effect. It is better tolerated than sorbitol, mannitol or xylitol precisely because most of it is absorbed and excreted rather than fermented in the colon, but "better tolerated" is not "unlimited".3
For scale, EFSA identified the level that did not cause diarrhoea as about 0.5 g per kilogram of body weight, which works out to roughly 35 g for a 70 kg adult.3 The 30 g used in the heart studies is therefore a large single dose, close to that gut threshold, and not typical of sprinkling it on porridge. Products containing significant polyol in the EU and UK still carry the warning that excessive consumption may have a laxative effect.
Safety note: the most reliable adverse effect of erythritol is gastrointestinal, not cardiac. Large single servings, common in some "sugar-free" sweets, keto baking and protein drinks, can trigger bloating, wind and loosened stools, especially if you are not used to polyols. If sugar-free products reliably upset your stomach, the polyols are a likely culprit. Children reach the gut-tolerance threshold at much smaller amounts than adults.3
What EFSA and UK regulators say
European regulators looked at all of this directly. EFSA re-evaluated erythritol (E 968) in December 2023, after the Witkowski paper, and reached measured conclusions: erythritol is not genotoxic, does not meaningfully affect blood sugar, and "current evidence does not show a connection (cause and effect) between consuming foods containing erythritol and an increased risk of cardiovascular diseases". EFSA explicitly noted the central uncertainty, that the circulating erythritol in the observational studies may not reflect dietary intake at all.3
EFSA did make one change, but for a different reason. It set an acceptable daily intake of 0.5 g per kilogram of body weight per day, driven entirely by the laxative threshold rather than any heart concern, and kept the laxative warning label. It also flagged that many people, children especially, may already exceed that intake.3 In the UK, erythritol remains an authorised sweetener in Great Britain as well as the EU; the Food Standards Agency has not restricted it or issued a cardiovascular warning. The regulatory position on both sides is consistent: approved, with a laxative caveat, and the cardiovascular research kept under review.
An honest bottom line
For most people, occasional erythritol in foods is very unlikely to be a meaningful heart risk on current evidence, and no regulator has found a proven causal link. The cardiovascular signal is associational plus mechanistic, not established as cause, and reverse causation, your body making more erythritol when glucose control slips, remains a genuine alternative explanation rather than a fringe quibble.
That said, the mechanistic work is not nothing, and proportionate caution is reasonable for the people with the least margin: those with established cardiovascular disease, a prior clot or stroke, or diabetes. For them, it is sensible to avoid large daily intakes of erythritol-sweetened drinks and sugar-free products while better trials are done, not because harm is proven, but because the cost of moderating is trivial. You can read more in our cardiovascular and heart-risk guides in the health library.
The least sensible reaction is to swap back to sugar, whose cardiometabolic harms are far better established. Whole foods, water and unsweetened options sidestep the question entirely. And if your real interest is blood sugar and metabolic health, that is the bigger lever, which we cover in our metabolic and insulin-resistance guides.
- If you have heart disease, diabetes or a past clot, ask whether your overall cardiovascular risk is well controlled (blood pressure, cholesterol or ApoB, HbA1c). That matters far more than any single sweetener.
- Mention if you use a lot of sugar-free, keto or diabetic products daily, so your intake is on the record.
- Ask whether unexplained bloating, wind or loose stools could be down to the polyols in sugar-free foods.
- Do not stop prescribed treatment, or switch back to sugar, on the strength of a headline. Ask for the risk in context.
References
- Witkowski M, Nemet I, Alamri H, et al. The artificial sweetener erythritol and cardiovascular event risk. Nat Med. 2023;29(3):710-718. nature.com.
- EFSA Panel on Food Additives and Flavourings. Re-evaluation of erythritol (E 968) as a food additive. EFSA Journal. 2023;21(12):e8430. efsa.onlinelibrary.wiley.com.
- EFSA. Plain language summary: Re-evaluation of erythritol (E 968) as a food additive. 20 December 2023. efsa.europa.eu.
- Hootman KC, Trezzi JP, Kraemer L, et al. Erythritol is a pentose-phosphate pathway metabolite and associated with adiposity gain in young adults. PNAS. 2017;114(21):E4233-E4240. pnas.org.
- Witkowski M, Nemet I, Li XS, et al. Xylitol is prothrombotic and associated with cardiovascular risk. Eur Heart J. 2024;45(27):2439-2452. doi.org/10.1093/eurheartj/ehae244.
- Heianza Y, Qi L, Manson JE. Is the Nonnutritive Sweetener Erythritol or Its Circulating Metabolite a Risk Factor for Cardiovascular Events? Clin Chem. 2023;69(10):1098-1100. PMID 37473406.
- Erythritol, Erythronate, and Cardiovascular Outcomes in Older Adults in the ARIC Study. JACC Adv. 2025;4(3):101605. PMC11889355.
- NIH Research Matters. Erythritol and cardiovascular events. 2023. nih.gov.
- Cleveland Clinic, Lerner Research Institute. Common artificial sweetener, erythritol, associated with higher rates of heart attack, stroke. 2023. lerner.ccf.org.
This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.