Quick Answer: Two large observational studies link higher blood erythritol levels to roughly doubled risk of major cardiovascular events over three years. A 2025 lab study found erythritol concentrations from a single sweetened drink disrupted blood-vessel cell function within hours. These findings are preliminary and do not establish cause and effect.
Last reviewed: 2026-06-08
Erythritol shows up in hundreds of products sold as "healthy" or "keto-friendly": protein bars, low-carb chocolates, sugar-free ice cream, and packets of monk-fruit blend at the coffee counter. It earns good press because it has almost zero calories, does not spike blood sugar, and holds GRAS ("Generally Recognized as Safe") status with the FDA.
That status has not changed. But starting in 2023, a cluster of studies began raising questions worth knowing about, especially if you eat these products regularly or have cardiovascular risk factors.
Here is what the studies found, what they did not prove, and how to make sense of the conflicting expert positions.
What Does the Research Actually Say About Erythritol and Cardiovascular Risk?
The signal started with a 2023 Cleveland Clinic study published in Nature Medicine. Researchers measured plasma erythritol levels in thousands of adults being evaluated for cardiovascular risk, then tracked their health outcomes for three years. People in the highest quartile of plasma erythritol had roughly twice the three-year risk of a major adverse cardiovascular event compared to those in the lowest quartile (Witkowski et al., Nature Medicine, 2023; NIH Research Matters summary).
The adjusted hazard ratios were 1.80 in the U.S. cohort and 2.21 in the European replication group. "Roughly twice" is an accurate description of those numbers.
A second study, published in JACC: Advances in March 2025, followed 4,006 older adults from the Atherosclerosis Risk in Communities (ARIC) cohort over a median of 8.41 years. Higher plasma erythritol was significantly associated with increased risk of heart failure hospitalization, heart failure with preserved ejection fraction, cardiovascular death, and all-cause mortality, after adjusting for covariates (JACC: Advances, 2025; PMC full text).
In June 2025, a University of Colorado Boulder lab study added a cellular-level finding. Berry and colleagues exposed human brain microvascular endothelial cells to 6 millimolar erythritol, a concentration equivalent to a single 30-gram sweetened-drink serving. Within three hours, reactive oxygen species were significantly increased, nitric oxide production dropped from 7.3 to 5.8 micromoles per litre, and markers of endothelial stress rose (Berry et al., Journal of Applied Physiology, 2025; 138(6):1571-1577; DOI 10.1152/japplphysiol.00276.2025).
Three caveats that belong alongside those findings:
Both the ARIC and Witkowski studies are observational. They show association, not causation. The FDA reviewed the 2023 Witkowski study and concluded it "did not establish a causal link between consuming erythritol and the observed effects" (FDA evaluation). People with higher plasma erythritol might differ in diet, metabolic profile, or other risk factors, and erythritol may be a marker rather than a driver.
The Berry et al. lab study used cultured cells, not human volunteers. A published editorial response to that paper (DOI 10.1152/japplphysiol.00904.2025) explicitly cautions against extrapolating in-vitro results to clinical outcomes.
Should You Stop Eating Erythritol-Sweetened Foods?
The science does not say erythritol causes cardiovascular harm. It says that high circulating erythritol appears alongside higher cardiovascular risk in large observational cohorts, and that drink-level concentrations stress blood-vessel cells in a lab dish. That is a pattern worth knowing about. It is not a proven mechanism.
A few things worth understanding before you decide what to do:
Your body also makes erythritol on its own. Endogenous production (from glucose via the pentose phosphate pathway) may account for some of the erythritol measured in plasma studies, particularly in people with metabolic syndrome or consistently high blood glucose. Plasma erythritol is not a clean proxy for erythritol consumed from food alone.
The dose context matters. The Berry lab study used a concentration equivalent to a single 30-gram serving. Many keto protein bars and sugar-free ice creams contain 15 to 25 grams per serving, and some people eat more than one serving in a day. Whether cumulative daily exposure over months or years looks different from a single-dose cell study is not yet answered in humans.
If you have existing cardiovascular risk factors, treating erythritol-sweetened products as a sometimes-food rather than a daily unlimited sugar substitute is a reasonable position consistent with the current evidence. Your doctor can help you apply that to your specific situation.
How Do You Know If a Product Contains Erythritol?
Erythritol is a sugar alcohol. Sugar alcohols appear in the "Total Carbohydrates" section of Canadian and U.S. nutrition labels. The name is not always front-and-centre on the packaging.
Look for these in the ingredients list:
- "erythritol" (the most direct)
- "monk fruit sweetener" (most commercial monk-fruit blends use erythritol as the bulk ingredient)
- "stevia blend" (same setup)
- "natural sweeteners" on keto-positioned products (often erythritol-based)
Products most likely to carry significant erythritol: keto bars and cookies, sugar-free chocolate, low-carb ice cream, sugar-free gummy candy, and some protein powders.
If you want to check before you eat, NoJunk scans the barcode and surfaces ingredients worth reviewing, including sugar alcohols. The ingredient glossary inside the app gives more context on specific additives.
Frequently Asked Questions
Is erythritol FDA-approved as safe?
Yes. Erythritol has GRAS status from the FDA, meaning it is recognized as safe based on the evidence available at the time of that determination. GRAS status does not mean a substance carries zero risk under any condition, or that new research cannot raise new questions. The FDA's own evaluation of the 2023 Witkowski study found the data did not establish a causal link between consuming erythritol and cardiovascular events (FDA evaluation). The current GRAS determination remains in place as of this writing.
Does erythritol raise blood sugar?
No. Erythritol has a glycemic index of essentially zero. Unlike sorbitol or xylitol, it is absorbed in the small intestine and excreted largely unchanged in urine, which also means it typically does not cause the digestive side effects other sugar alcohols produce at moderate doses. That profile is why it became common in diabetic-friendly and keto products.
What is "plasma erythritol" and is it different from dietary erythritol?
Plasma erythritol is the concentration of erythritol measured in a blood sample. It reflects both erythritol you consume in food and erythritol your body produces naturally through metabolism. The observational studies (Witkowski 2023, ARIC 2025) measure plasma levels, not dietary intake logs. This distinction matters: high plasma erythritol in some people may reflect metabolic processes that are themselves markers of cardiovascular risk, independent of how much erythritol those people eat.
What exactly did the FDA say about erythritol?
The FDA reviewed the Witkowski 2023 Nature Medicine paper and published a written evaluation (FDA document). The agency concluded that the observational studies "did not establish a causal link between consuming erythritol and the observed effects" and noted that the cohorts consisted of people already at higher cardiovascular risk, making extrapolation to the general healthy adult population uncertain. The FDA did not call for restrictions or label changes based on that review.
How much erythritol is typically in a "sugar-free" snack?
It varies significantly by product. As a rough range: a sugar-free chocolate bar typically contains 10 to 20 grams, a serving of keto ice cream 15 to 22 grams, and a single monk-fruit or stevia packet for coffee around 1 gram. The CU Boulder Berry study used a dose equivalent to 30 grams in a single beverage. Checking the "sugar alcohols" line in the nutrition facts panel of any specific product gives you the most reliable number.
Last reviewed: June 2026