The sugar that leaves unchanged.
Erythritol is a sugar with one piece changed. Your tongue still reads it as sweet, about 70 percent as sweet as table sugar. None of your enzymes can burn it and your gut bacteria do not ferment it, so it goes into your blood intact and your kidneys filter it straight back out: about 90 percent of a dose is in your urine within a day, at under 0.4 calories a gram. This page has that path drawn out, the fermenter and the crystalliser that make it with a titre slider and the oxygen balance that comes with it, and the 2023 heart study set next to what the regulator said after reading it.
The tank and the crystalliser are the parts an engineer models. Reflux Student runs that model in your own Aspen Plus, from a sentence. Try it free on Aspen Plus →, or jump straight to the molecule.
Out now
Try Reflux Student free on Aspen Plus
Reflux Student drives your own Aspen Plus V14 from plain English. You type what you want, it opens the case, makes the change in Aspen, runs it, reads the result back and tells you what it verified. Aspen does the math. You keep the judgment. Windows, your own licence, and three free runs to start: it is a trial with a limit on it, not a free product.
Enter your email on the next page and the download link lands in your inbox, so you can open it on the Windows machine Aspen lives on. Windows will say it does not recognise the app the first time: click More info, then Run anyway. The certificate is new and Windows trusts it by reputation, which takes downloads to build. Nothing is wrong with the file.
Not on Aspen Plus? Tell me which simulator you use → and you hear the day your build exists.
One piece changed, and your enzymes lose their grip.
Start from erythrose, a four carbon sugar. Its first carbon is an aldehyde: a carbon double bonded to an oxygen, which is the reactive end and the handle almost every enzyme that works on sugars grabs. Reduce it, meaning add hydrogen across that bond, and the double bond becomes a single one, an H lands on the carbon, and the aldehyde becomes an ordinary alcohol. That is the whole change. C4H8O4 becomes C4H10O4.
Two things follow from losing that one group, and they are the reason the label says zero.
In, through, and out, with one number at the end.
It crosses the wall of the small intestine passively and enters the portal blood intact. Then it reaches the kidney, and this is where the interesting part is. The glomerulus filters small molecules out of the blood indiscriminately, so erythritol goes into the filtrate along with everything else its size. Glucose would then be pulled straight back through the tubule wall by a dedicated transporter. Erythritol has no transporter, so it stays in the filtrate and leaves.
measured caloric value: under 0.4 kcal/g, against 4.0 kcal/g for sucrose
Labelled as zero in the United States and 0 kcal/g in Japan. The residual is the small fraction that is oxidised rather than excreted.
So the trick is not that your body cannot absorb it. Your body absorbs it very well. The trick is that once it is in, nothing can do anything with it, and the kidney has no reason to keep it.
The tank and the crystalliser.
It is made by fermentation, not by chemistry on a bench. An osmophilic yeast, Moniliella pollinis, is fed sugarcane or corn derived sugar in a stirred, aerated, fed batch tank and secretes erythritol into the broth, reported at about 95 g/L at a yield of 0.38 g of erythritol per g of sugar. The broth then goes to a cooling crystalliser, where dropping the temperature pushes it past saturation and the crystals come out at about 96 percent purity. What does not crystallise is mother liquor and goes round again.
Left: the fermenter in section, with the sparger at the base, two impellers that shear the bubbles smaller on their way up, and the cooling jacket. Right: the crystalliser, with its internal coil taking the heat out and the crystals settling into the cone.
Why the titre is an oxygen problem
Getting a higher titre is not free. Every gram of erythritol costs sugar that was not turned into product, and burning that sugar consumes oxygen, which has to be dragged out of bubbles and into a broth that is getting thicker as the titre climbs. The oxygen demand falls straight out of the yield:
carbon in = 2.63 × 0.400 = 1.053 g C
carbon out (product) = 1 × 48/122 = 0.393 g C
carbon respired = 0.660 g C → 2.42 g CO2
oxygen, at RQ = 1 = 2.42 × 32/44 = 1.76 g O2 per g erythritol Glucose is 40 percent carbon by mass; erythritol is C4H10O4, so 48 of its 122 g/mol are carbon. RQ near 1 is the right assumption for a carbohydrate substrate. Biomass and byproducts take a slice of the respired carbon, so this is an upper bound on the oxygen.
A 100 m³ fed batch, over 120 hours
A stirred, sparged industrial fermenter runs somewhere between about 100 and 500 per hour, so this is a normal tank working hard.
Drag the titre up and the required transfer coefficient rises with it, in a straight line, because the oxygen demand is proportional to how much product you make in the same time. What the line does not show is the second effect: the broth gets more viscous as the titre climbs, which lowers both the oxygen solubility and the transfer coefficient the same impeller can deliver. That is why a real titre curve flattens out, and it is exactly the sort of thing you find out by running the case rather than by arithmetic.
The 2023 study, and what the regulator said after reading it.
In March 2023 a paper in Nature Medicine reported that among patients already being evaluated for cardiac risk, the ones with the most erythritol in their blood had more heart attacks and strokes over the following three years. It is a real finding across three separate cohorts, and it is an association, not a demonstration of cause.
What the study found
A consistent, statistically significant association in three independent cohorts, plus preclinical evidence that erythritol can enhance platelet activation.
- Discovery: HR 2.95 (1.70 to 5.12)
- United States: HR 1.80 (1.18 to 2.77)
- Europe: HR 2.21 (1.20 to 4.07)
What it could not rule out
The FDA reviewed it in June 2023 and did not question the existing safety conclusions. Its memo names the gaps.
- No diet data was collected on any of the subjects
- Your body makes erythritol from glucose through the pentose phosphate pathway, so a high blood level can follow a high sugar diet rather than erythritol intake
- The subjects all had known cardiovascular risk factors, which limits extrapolation to everyone else
The number from that paper worth keeping is the pharmacokinetic one, because it has nothing to do with the argument about cause. Give someone a 30 g drink, which is roughly what is in a large serving of a sweetened dessert, and plasma erythritol goes up about 1,500 fold within half an hour and is still far above baseline more than two days later.
That is a question worth answering, and it is being worked on. It is not an answer yet, and nobody serious is telling you to stop.
Sources: sweetness, caloric value, urinary recovery and the laxative threshold from Mazi and Stanhope, Nutrients 2023. Cohorts, hazard ratios and the 30 g pharmacokinetics from Witkowski et al., Nature Medicine 2023. Endogenous production, the missing diet data and the GRAS position from the FDA memorandum of 15 June 2023. The European re evaluation and its 0.5 g per kg per day intake limit, set on the laxative effect, is EFSA, December 2023. Fermentation titre and yield on sugarcane juice from Deshpande et al. 2022.
The tank is the part I model.
How much oxygen you can push into a broth that thick, and how much crystal you get per pass, are both questions a simulator answers and arithmetic does not. Reflux Student runs them in your own Aspen Plus, from a sentence. Free to start, Windows, your own licence.
Nathan
← reflux.sh