Sweetener resolver
Is erythritol safe? The cardiovascular study vs. its GRAS status
A 2023 study linking erythritol to cardiovascular events made headlines. The FDA's safety classification didn't move. Both of those things are true at the same time — here's exactly why, with the actual source, not the headline.
Database v0.0.1 (draft) — updated 2026-07-21
In the repair
What did the 2023 study actually find?
Witkowski et al., published in Nature Medicine (February 2023), followed a large observational cohort (4,000+ people) and found that people with higher blood erythritol levels had roughly 1.8–2.2x higher odds of a major adverse cardiac event (top vs. bottom quartile). The same paper reported an in-vitro/ex-vivo lab finding that erythritol increased platelet reactivity and clotting behavior in blood samples.
Source: Witkowski et al., Nature Medicine, Feb 2023 (doi: 10.1038/s41591-023-02223-9), via Cleveland Clinic (Hazen lab) and NIH Research Matters institutional coverage.Is that the same as proving erythritol causes heart attacks?
No. An observational association shows a correlation across a population — it cannot, on its own, prove that erythritol causes cardiac events in a given person. The study's authors themselves describe it as an association requiring follow-up, not a causal human randomized controlled trial (RCT). The in-vitro platelet finding is a plausible mechanism, not proof that the mechanism plays out the same way at typical dietary intake in humans.
Source: same Nature Medicine 2023 paper, author caveats reported via Cleveland Clinic newsroom and NIH coverage.Why is erythritol still FDA GRAS (Generally Recognized As Safe)?
The FDA's GRAS determination for erythritol has not been withdrawn or revised since the 2023 paper. Regulatory safety status changes generally require a body of causal evidence (typically RCTs or a formal agency safety review), not a single observational cohort study — so as of this writing, GRAS and "an association was found in one 2023 cohort study" are not contradictory facts.
Source: FDA GRAS status (unchanged, per general regulatory record as reviewed alongside the 2023 paper's institutional coverage).Does erythritol raise blood sugar?
No, and this is a separate axis from the cardiovascular question entirely. About 90% of ingested erythritol is absorbed in the small intestine and excreted unchanged in urine — it isn't metabolized as sugar, giving it a glycemic index near zero. RecipeMend buckets sweeteners by blood-glucose impact; erythritol's cardiovascular disclosure doesn't change that bucket, and both facts are shown side by side rather than one crowding out the other.
Source: Nature Medicine 2023 discussion of erythritol pharmacokinetics; Database v0.1 sweetener entry "erythritol".How much can I eat before it upsets my stomach?
Monash University's FODMAP program has lab-tested erythritol as low-FODMAP up to about 1 tablespoon (~13g) per sitting; some cited studies found doses as high as 50g didn't produce significant symptoms in their samples, though individual tolerance varies substantially. This guidance comes from secondary-source synthesis of Monash-derived material rather than a directly queried Monash database entry, so confidence is kept medium here.
Source: Monash University FODMAP program (via secondary summaries) — see Sources for the full caveat.Does erythritol bake like sugar?
It swaps roughly 1:1 by volume in cookies and cakes, but erythritol is only about 60–80% as sweet as sugar by weight, and it recrystallizes to a slightly grainy texture on cooling — most noticeable in cakes, less so in cookies. Cakes typically need 1–2 extra tablespoons of liquid per cup swapped to counter a drier bake.
Source: Swerve baking conversion guidance (manufacturer) and general erythritol brand baking guides — see Database v0.1 swap_rules.Want this same rigor applied to your own recipe?
€19 one-time
Paste any dessert recipe and get every ingredient's swap, ratio, and source in one printable report.
Repair your own recipe →Noted — taking you to the tool.