The 2023 erythritol cardiovascular paper: signal, noise, or both?
A Nature Medicine study linked circulating erythritol to thrombosis. The dietary implication is more contested than the headlines suggested.
In February 2023, Witkowski and colleagues at the Cleveland Clinic published a Nature Medicine paper titled 'The artificial sweetener erythritol and cardiovascular event risk'. The study reported that people with higher blood erythritol levels had higher 3-year risk of major adverse cardiac events (MACE), and that adding erythritol to platelet-rich plasma promoted clotting. It made every front page.
Within 48 hours the food industry had pushed back. Within a week, methodologists had pushed back too — and on substantively different grounds.
What the paper actually shows
Two findings. First, in a discovery cohort of ~1,100 patients undergoing cardiac assessment, plasma erythritol predicted 3-year MACE independent of standard cardiovascular risk factors. This was replicated in two validation cohorts. Second, in vitro and in mice, erythritol at concentrations comparable to a single 30 g dose enhanced platelet aggregation.
Why the dietary inference is contested
Erythritol isn't only consumed — it's also produced endogenously. The pentose phosphate pathway converts glucose to erythritol via erythrose-4-phosphate. People with insulin resistance, diabetes and oxidative stress (the same people at high cardiovascular risk) make more endogenous erythritol. So the question is: does dietary erythritol cause cardiovascular events, or are high blood erythritol levels a marker of an underlying metabolic state that already predicts events?
The study's design — observational, in patients already presenting for cardiac assessment — cannot answer this. The follow-up acute-dose experiments showed plasma erythritol stayed elevated for >2 days after a 30 g serving, which is real, but doesn't address whether long-term exposure causes the observed associations.
How to read this if you use erythritol
If you're a healthy, low-cardiovascular-risk consumer using erythritol occasionally — handful of squares of sugar-free chocolate, occasional baked good — the signal in this paper does not translate to a clear personal risk.
If you have established cardiovascular disease, diabetes or significant risk factors and use erythritol heavily (multiple servings daily, keto baking), the precautionary read is to substitute. Allulose and monk fruit do not have an equivalent signal in the literature.
How we score it
Erythritol holds at 72 / 100 — Good — on the SweetSpot scale. The score reflects an excellent overall safety record over decades, the new but unresolved cardiovascular signal, and the existence of cleaner-positioned alternatives. We will revise upward or downward as the next generation of trials reports out.