Is Erythritol Safe? What the 2024 Cleveland Clinic Study Actually Found

Erythritol has been the sweetener of choice for keto bakers, stevia-blend manufacturers, and protein-bar companies for most of the past decade. Then, in February 2024, a study from the Cleveland Clinic published in Nature Medicine set off a wave of alarmed coverage: "Erythritol linked to heart attack and stroke." The story spread fast.

The science is more complicated than those headlines made it sound. This page goes through what the study found, where its limits are, what regulators say, and why the answer to "should I worry about erythritol" is genuinely nuanced rather than alarming or dismissible.

What is erythritol?

Erythritol is a sugar alcohol (a polyol, a class of compounds that are structurally similar to sugars but behave differently in the body). Its IUPAC name is (2R,3S)-butane-1,2,3,4-tetraol. The four-carbon structure is why it is sometimes called a 4-carbon polyol, which distinguishes it from longer-chain polyols like xylitol (5 carbons) and sorbitol (6 carbons).

It occurs naturally at low concentrations in grapes, pears, melons, mushrooms, and fermented foods including wine, beer, and soy sauce. The human body also produces erythritol endogenously through the pentose phosphate pathway, a metabolic route that converts glucose into erythritol in red blood cells and other tissues. This last point matters a great deal for interpreting the 2024 study.

Commercial erythritol is made through fermentation: glucose, typically derived from corn starch, is fermented by yeast or certain fungi (Moniliella pollinis and Torula corallina are common industrial strains), and the resulting erythritol is extracted, purified, and crystallized. The fermentation origin is why you will see erythritol marketed as a "natural sweetener": the process uses a living organism, not synthetic chemistry, though the end product is far more concentrated than what appears naturally in any food.

Erythritol is approximately 60-80 percent as sweet as sugar, with a slight cooling sensation that becomes noticeable at higher concentrations. It has essentially no calories, because the body absorbs it but lacks the enzymes to metabolize it for energy. Most of what you eat is excreted unchanged in urine within 24 hours.

Its European additive number is E968. In the US it carries GRAS (Generally Recognized As Safe) status, which it received via FDA Notice of No Objection in 2001.

Is erythritol safe? What regulators say

FDA

Erythritol received GRAS status in the US through the FDA's self-affirmation process, with a notice of no objection issued in 2001 (GRAS Notice GRN 000076). It is not governed by a numeric acceptable daily intake, which in regulatory language signals a high margin of safety: the agency has concluded that determining a specific upper limit is unnecessary. The FDA has issued no advisories, warnings, or re-evaluations following the 2024 Cleveland Clinic study as of May 2026.

Health Canada

Health Canada permits erythritol as a food additive under the Food and Drug Regulations. It is approved as a sweetener, bulking agent, and humectant across food categories including confectionery, baked goods, and table-top sweeteners. No ADI restriction and no consumer advisory.

EFSA

The European Food Safety Authority first approved erythritol in 2003. In its 2003 opinion, EFSA established an ADI of "not specified," the same language the FDA's GRAS designation implies, and a signal that normal dietary exposure does not raise safety concerns. EFSA conducted a reassessment of erythritol as part of its broader re-evaluation program for food additives. The 2022 reassessment (EFSA Journal 2022) maintained approval, noting that the evidence base did not support concern at typical food use levels.

WHO/JECFA

The Joint FAO/WHO Expert Committee on Food Additives has also set an ADI of "not specified" for erythritol. The 2023 WHO guideline on non-sugar sweeteners, which recommended against using them for weight control, did not specifically call out erythritol; the guidance focused primarily on non-nutritive sweeteners like aspartame, sucralose, stevia, and saccharin. Erythritol is a different class of compound (a polyol, providing minimal calories rather than zero) and was not the primary focus of that recommendation.

The February 2024 Cleveland Clinic study: what it found

This is the part of the erythritol story that changed the conversation.

Witkowski et al., published in Nature Medicine in February 2024 (DOI: 10.1038/s41591-023-02765-z), examined erythritol's relationship to cardiovascular risk across three components:

The observational cohort. The researchers analyzed plasma samples from 1,157 people undergoing cardiac risk assessment at the Cleveland Clinic. They measured metabolites in the blood and compared levels against cardiovascular outcomes over a three-year follow-up period. People in the highest quartile of plasma erythritol had roughly twice the rate of major adverse cardiovascular events (MACE: heart attack, stroke, or death) compared to those in the lowest quartile.

The validation cohorts. To test whether this association held outside the original group, the researchers replicated the finding in two independent European cohorts: one from the US (an additional Cleveland Clinic group) and one from Germany. The cardiovascular association appeared in both.

The mechanistic experiments. The team then tested erythritol's direct effect on platelets in lab conditions. They found that erythritol promoted platelet aggregation (the clumping of platelets that initiates clot formation) at concentrations similar to those seen in blood after consuming erythritol-sweetened foods. They also tested in mouse models, finding that erythritol accelerated clot formation in vivo.

Taken together, the study proposed a mechanism: erythritol in the bloodstream activates platelets, increasing clotting tendency, which raises cardiovascular event risk.

This is a meaningful finding. The consistency across cohorts and the mechanistic support are stronger than a purely observational correlation. It is not a study to dismiss.

Why the 2024 study has critics

The coverage that followed the Witkowski paper often skipped the methodological questions that scientists raised immediately on publication. Several are substantial.

The confounding problem. The Cleveland Clinic cohort was drawn from people undergoing cardiac risk assessment. They were not a representative sample of the general population: they were already suspected of elevated cardiovascular risk. The high-erythritol group within this population was also more likely to have obesity, diabetes, and other metabolic conditions. These factors are themselves cardiovascular risk factors. The authors attempted statistical adjustment, but critics have argued that the adjustments could not fully account for the degree of pre-existing metabolic dysfunction in the high-erythritol group.

Correlation is not causation. The observational component cannot establish that erythritol caused the cardiovascular events. People who consumed more erythritol may have done so precisely because they had conditions (diabetes, obesity, metabolic syndrome) that were already increasing their cardiovascular risk, and which also raise endogenous erythritol production.

The dietary data gap. The study measured plasma erythritol, not dietary erythritol intake. Participants were not asked how much erythritol they consumed from food. This means the study cannot confirm that the high-erythritol group was actually eating more erythritol than the low-erythritol group. High serum erythritol could reflect dietary intake, internal production, or both.

Platelet aggregation context. The platelet experiments were conducted in plasma samples and mouse models. Whether the platelet-activation effect translates to clinically significant clotting risk in healthy humans eating normal amounts of erythritol from food has not been established in a clinical trial.

Several academic responses published in the months following the paper, including a critique in the European Heart Journal (2024), raised these points. The authors responded by acknowledging the limitations while maintaining that the consistency of findings across cohorts warranted further investigation.

The honest read: the 2024 study raised a legitimate flag. It is not proof that eating erythritol gives healthy people heart attacks. It is evidence that warrants follow-up, particularly for people with pre-existing metabolic risk.

Endogenous erythritol: the complication nobody talks about

Here is the factor that makes erythritol research uniquely difficult to interpret: your body makes it.

The pentose phosphate pathway, a metabolic process that runs in red blood cells, liver cells, and other tissues, produces erythritol as a byproduct of glucose metabolism. Erythritol is not merely something you ingest. It is something your body generates, and the rate of generation goes up when glucose metabolism is dysregulated, as in insulin resistance and type 2 diabetes.

This has a direct implication for the 2024 study: participants who had high serum erythritol may have had high levels partly because their glucose metabolism was disordered, producing more erythritol internally. Those same metabolic conditions (insulin resistance, elevated blood glucose, obesity) are major drivers of cardiovascular risk.

In other words, elevated serum erythritol might be a consequence of metabolic dysfunction, not a cause of cardiovascular damage. It may be a biomarker that tracks alongside cardiovascular risk without causing it.

Distinguishing dietary erythritol from endogenous erythritol would require isotope labeling studies. The Witkowski paper did not include these. Until research makes that distinction cleanly, the causal interpretation of the 2024 findings remains uncertain.

GI tolerance: where erythritol actually differs from other polyols

One thing the 2024 study did not change is erythritol's GI tolerance profile, which is genuinely better than most other sugar alcohols.

Xylitol is absorbed at roughly 20 percent efficiency in the small intestine. Sorbitol is absorbed at roughly 35 percent. The remaining fraction reaches the large intestine intact, where gut bacteria ferment it, producing gas, bloating, and at higher doses, osmotic diarrhea. This is why xylitol and sorbitol carry mandatory laxative warnings in Europe when present above a certain threshold in food.

Erythritol is absorbed at roughly 90 percent efficiency in the small intestine and excreted in urine largely unchanged. Very little reaches the colon to be fermented. Studies suggest that most people can tolerate up to about 50 grams of erythritol in a single serving without significant GI symptoms, compared to roughly 15-20 grams for xylitol and 20-30 grams for sorbitol.

At the doses found in a typical keto bar (5-15g of erythritol) or a cup of tea sweetened with a monkfruit-erythritol blend (2-4g), GI symptoms are uncommon. For people who have found xylitol or sorbitol difficult to tolerate, erythritol is typically better.

The keto and low-carb context

Erythritol became the dominant sweetener in keto products for three reasons: it has zero net carbohydrates, it does not raise blood glucose or insulin, and it blends well with high-intensity sweeteners. Monkfruit extract is intensely sweet but has a noticeable aftertaste at high concentrations; combining it with erythritol produces a blend that tastes closer to sugar than either alone. Truvia uses the same approach: primarily erythritol by weight, stevia extract providing most of the sweetness.

It is also heat-stable in ways stevia is not, which makes it viable for baking.

The 2024 cardiovascular findings matter more to this audience than to casual users. A keto brownie recipe might call for a full cup (200g) of erythritol, far more than a packet in coffee. Whether the platelet effects observed in lab conditions are dose-dependent in humans has not been answered.

Where you will find erythritol

Erythritol turns up in table-top sweetener blends (Truvia, Swerve, Lakanto, all primarily erythritol by weight), keto baked goods and protein bars, sugar-free chocolates like Lily's and ChocZero, zero-calorie energy drinks, and some protein powders. It is almost always paired with a high-intensity sweetener because at 60-80% the sweetness of sugar it takes large amounts to match sugar volume, so pairing with monk fruit or stevia lets manufacturers use less of both.

On ingredient labels it reads "erythritol" in the US and Canada, or "erythritol (E968)" or "sweetener (E968)" in EU-regulated markets.

Products containing erythritol (E968)

These five products are verified against live Open Food Facts data as containing E968. Barcodes confirmed via direct OFF product API lookups on 2026-05-21.

Brand	Product	Barcode	Nutri-Score	NOVA	Notes
Quest	Protein Bar	0888849000005	N/A	4	One of the highest-volume keto protein bars in North America. Erythritol is a primary sweetener alongside sucralose. Representative of the protein-bar category.
Monster	Monster Energy Ultra	0070847012474	C	4	Zero-sugar energy drink. Erythritol provides bulk sweetness; the Ultra line is explicitly marketed to low-carb consumers.
Lily's	Dark Chocolate Style Baking Chips	0856481003180	D	4	The market-leader in keto baking chocolate. Sweetened with erythritol and stevia extract. Common in keto recipes at relatively high erythritol concentrations per serving.
Kirkland Signature	Protein Bar Cookies and Cream	0096619193738	N/A	4	Costco house-brand protein bar. Erythritol in the chocolate compound. Illustrates how far erythritol has penetrated mainstream retail, not just specialty keto aisles.
Bake Believe	Dark Chocolate Baking Chips	0884983000666	N/A	4	Sugar-free baking product using erythritol as primary sweetener. Designed for high-dose keto baking applications where erythritol consumption per use can be significant.

OFF community-submitted data, verified 2026-05-21. Product formulations change; scan with NoJunk or check the current ingredient panel on the product in hand.

FAQ

Is erythritol safe?

All major regulatory bodies (FDA, Health Canada, EFSA, WHO/JECFA) have approved erythritol and set no numeric acceptable daily intake, which reflects a high confidence in its safety at normal dietary levels. A 2024 observational study found an association between high blood erythritol and cardiovascular events, but it could not establish causation, and the study population was already at elevated metabolic risk. The safety assessment has not changed at the regulatory level, though some researchers have called for follow-up studies, particularly in populations with pre-existing cardiovascular risk factors.

Does erythritol cause heart attacks?

The 2024 Witkowski et al. study found that people with higher serum erythritol had higher rates of heart attacks and strokes over three years, and that erythritol promoted platelet clumping in lab conditions. This is worth taking seriously. It is not proof that eating erythritol causes heart attacks in otherwise healthy people. The study was observational, the participants were already at cardiac risk, and high blood erythritol may reflect the body's own production (elevated in metabolic dysfunction) rather than dietary intake. Follow-up clinical trials have not yet been published.

What are erythritol's side effects?

GI symptoms (bloating, gas, loose stools) can occur at high single doses, typically above 50 grams. At doses common in packaged foods (5-15g), most people tolerate erythritol without symptoms. Erythritol causes fewer GI issues than xylitol or sorbitol because it is mostly absorbed before reaching the large intestine. The 2024 study raised the possibility of platelet-activation effects at higher blood concentrations, though this has not been confirmed in a clinical trial in healthy adults.

Is erythritol keto?

Yes. Erythritol has zero net carbs, does not raise blood glucose, and does not trigger insulin release. It is the most common sweetener in keto baked goods and low-carb protein bars, often combined with monk fruit or stevia to improve taste profile. At high keto-baking doses, the 2024 cardiovascular data is relevant context: doses in a full keto recipe can be significantly higher than typical food-use levels.

Erythritol vs xylitol: which is better?

For GI tolerance: erythritol, by a clear margin. Xylitol ferments in the large intestine; erythritol mostly does not. For dental health: xylitol has a stronger evidence base for inhibiting Streptococcus mutans. For keto use: erythritol, because it has zero glycemic impact; xylitol has a modest glycemic index. For cardiovascular concern: no equivalent study to the 2024 Witkowski paper has been published for xylitol, so the evidence profile is currently different.

Is erythritol natural?

It occurs naturally in fruits and fermented foods and is produced endogenously by the human body. Commercial erythritol is made by fermenting glucose, using yeast, a biological process. The "natural" claim is technically supportable but the commercial product is far more concentrated than any naturally occurring source. This is the same argument that applies to stevia: natural origin, concentrated form.

Is erythritol safe for people with diabetes?

Erythritol does not affect blood glucose or insulin, making it a frequent recommendation for diabetes management. The 2024 study adds nuance: the cohort included people with diabetes, who showed higher baseline erythritol levels, possibly because metabolic dysfunction increases endogenous erythritol production. Whether this means people with diabetes should limit erythritol-sweetened foods, or whether elevated erythritol is simply a consequence of their underlying condition, has not been resolved. Discussing erythritol use with a healthcare provider is reasonable for anyone with existing cardiovascular risk factors.

Sources

1. Witkowski M, Nemet I, Alamri H, et al. "The artificial sweetener erythritol and cardiovascular event risk." Nature Medicine. 2024; 30:648-657. DOI: 10.1038/s41591-023-02765-z

2. US Food and Drug Administration. GRAS Notice No. GRN 000076: Erythritol. FDA Center for Food Safety and Applied Nutrition. 2001. https://www.fda.gov/food/generally-recognized-safe-gras/gras-notice-inventory

3. Health Canada. "List of Permitted Sweeteners." Food and Drug Regulations, Division 16. Government of Canada. https://www.canada.ca/en/health-canada/services/food-nutrition/food-safety/food-additives/lists-permitted/8-sweeteners.html

4. EFSA ANS Panel. "Re-evaluation of erythritol (E 968) as a food additive in foods for the general population and in foods for specific groups." EFSA Journal. 2022. https://efsa.onlinelibrary.wiley.com/doi/10.2903/j.efsa.2022.7069

5. European Parliament and Council. Regulation (EC) No 1333/2008 on food additives. E968 erythritol, permitted in specified categories. OJ L 354, 31.12.2008.

6. Joint FAO/WHO Expert Committee on Food Additives. Erythritol. ADI "not specified." JECFA Monograph. https://www.fao.org/food/food-safety-quality/scientific-advice/jecfa/en/

7. Hiele M, Ghoos Y, Rutgeerts P, Vantrappen G. "Metabolism of erythritol in humans: comparison with glucose and lactitol." British Journal of Nutrition. 1993; 69(1):169-176. DOI: 10.1079/BJN19930019 (Small-intestine absorption rate, urinary excretion data.)

8. Noda K, Nakayama K, Oku T. "Serum glucose and insulin levels and erythritol balance after oral administration of erythritol in healthy subjects." European Journal of Clinical Nutrition. 1994; 48(4):286-292. (Glucose and insulin response data.)

9. Storey D, Lee A, Bornet F, Brouns F. "Gastrointestinal tolerance of erythritol and xylitol ingested in a liquid." European Journal of Clinical Nutrition. 2007; 61(3):349-354. DOI: 10.1038/sj.ejcn.1602532 (GI tolerance threshold data, 50g single-dose finding.)

10. World Health Organization. "Use of non-sugar sweeteners: WHO guideline." Geneva: WHO; 2023. ISBN 978-92-4-007361-8. https://www.who.int/publications/i/item/9789240073616

11. Livesey G. "Health potential of polyols as sugar replacers, with emphasis on low glycaemic properties." Nutrition Research Reviews. 2003; 16(2):163-191. DOI: 10.1079/NRR200371 (Polyol comparative chemistry and metabolic properties.)

12. Islam MS, Indrajit M. "Effects of xylitol as a sugar substitute on diabetes-related parameters in nondiabetic rats." Journal of Dietary Supplements. 2012; 9(4):247-257. (Comparative context for polyol glycemic profiles.)

13. van Laar ADE, Grootaert C, Van Camp J. "Rare mono- and disaccharides as healthy alternative for traditional sugars and sweeteners?" Critical Reviews in Food Science and Nutrition. 2021; 61(5):713-741. DOI: 10.1080/10408398.2020.1743966 (Natural occurrence, fermentation production, regulatory overview.)

14. Open Food Facts database. Product entries referenced in product table above. Verified 2026-05-21. https://world.openfoodfacts.org

Frequently asked questions

Common questions about this ingredient.

Is erythritol safe?

Regulatory agencies in the US, Canada, Europe, and globally have approved erythritol and set no numeric acceptable daily intake limit, indicating confidence in its safety at normal dietary levels. A February 2024 observational study published in Nature Medicine found an association between high serum erythritol and cardiovascular events, which generated significant media coverage. However, the study could not establish causation, did not control for pre-existing metabolic risk factors adequately, and did not separate dietary erythritol from the erythritol the body produces naturally through its own metabolism. The regulatory safety picture has not changed as a result of that study, though some researchers have called for further investigation.

Does erythritol cause heart attacks?

The February 2024 Witkowski et al. study in Nature Medicine found that people with higher blood levels of erythritol had a higher rate of major adverse cardiovascular events (MACE) over a three-year follow-up. The study also showed, in lab conditions, that erythritol can promote platelet aggregation. These are real findings worth taking seriously. However, the study was observational: it found a correlation, not proof that erythritol caused the events. Crucially, the participants who had high erythritol levels were already metabolically unwell; high serum erythritol may be a marker of metabolic dysfunction rather than the cause of cardiovascular risk. Additionally, the body produces erythritol endogenously from glucose, so high blood levels do not necessarily mean high dietary intake. No regulatory body has issued a safety warning as a result of this study.

What are erythritol side effects?

Gastrointestinal effects are the most commonly reported issue, though erythritol causes fewer GI symptoms than other sugar alcohols like xylitol or sorbitol. This is because roughly 90 percent of ingested erythritol is absorbed in the small intestine and excreted unchanged in urine, leaving less to ferment in the large intestine. At high doses (studies suggest roughly 50 grams or more in a single sitting), bloating, gas, or loose stools can occur in some individuals. At typical food use levels, most people tolerate erythritol well.

Is erythritol keto?

Yes, erythritol is considered keto-compatible. It has zero net carbohydrates because it is not metabolized to a meaningful degree: the body absorbs it and excretes it in urine without breaking it down for energy. It does not raise blood glucose or insulin levels in standard human studies. This combination makes it the most common sweetener in keto baking products, protein bars, and monkfruit-erythritol blend sweeteners marketed to low-carb consumers.

Erythritol vs xylitol: which is better?

For gastrointestinal tolerance, erythritol generally wins. Xylitol is only about 20 percent absorbed in the small intestine; the rest reaches the large intestine where bacteria ferment it, causing gas and diarrhea at relatively modest doses (15-20g in sensitive individuals). Erythritol's roughly 90 percent small-intestine absorption means far less fermentation and fewer GI symptoms at equivalent doses. Xylitol has a well-established benefit for dental health (it inhibits Streptococcus mutans) and may reduce ear infections in children. Erythritol also shows some dental benefit, but the evidence base is smaller. For the 2024 cardiovascular concerns: to date, no equivalent study has found the same cardiovascular signal for xylitol, though the research landscape for all sugar alcohols is evolving.

Is erythritol natural?

Erythritol occurs naturally in small amounts in fruits (grapes, pears, melons), mushrooms, fermented foods (wine, beer, soy sauce), and is produced endogenously by the human body through the pentose phosphate pathway. Commercial erythritol is made by fermenting glucose (typically from corn starch) using yeast. The fermentation process is natural in the sense that it uses a biological organism rather than chemical synthesis, but the final product is far more concentrated than anything found in food. Products labeled 'natural sweetener' with erythritol are technically defensible but can be misleading about concentration.

Is erythritol safe for people with diabetes?

Erythritol does not raise blood glucose or trigger an insulin response in controlled human studies, which makes it often recommended over sugar for people managing diabetes. The 2024 Witkowski et al. study is particularly relevant here, however: the study population included people with type 2 diabetes and obesity, and participants with these conditions had higher baseline erythritol levels. This raises a question about whether people with diabetes are more likely to have elevated serum erythritol to begin with (whether or not they consume it as a sweetener), and whether the cardiovascular association found in the study is specific to this population. Consulting a healthcare provider before using erythritol as a primary sweetener, particularly for people with existing cardiovascular risk factors, is reasonable in light of the 2024 data.