A sugar alcohol at industrial dose

Xylitol is easy to miss because it was designed to be missed. It is the sugar alcohol in gum, toothpaste, jam, yogurt, and ice cream, the ingredient that lets a label say sugar-free without tasting like a punishment. It occurs in plants in tiny amounts. That is not how people now encounter it. It is being added at doses about 1,000 times what appears in nature.

The gap between a birch tree and a stick of gum is the whole story. A compound that a metabolism can handle as a trace is not automatically a compound a metabolism can handle as a daily flood. Food chemists did not wait for cardiology to settle that question. They found a sweetener that does not rot teeth the way sucrose does, that does not spike blood sugar the way glucose does, and that slots neatly into the industrial logic of “better for you” snacks. The packets multiplied. The blood levels followed.

At the European Society of Cardiology congress in Munich, Berlin cardiologist Marco Witkowski reported what he framed as the largest look yet at whether those blood levels track with the outcomes cardiology actually cares about: death, heart attack, and stroke. The sample was 17,710 people. The comparison was simple on purpose: top versus bottom blood-level quartile.

What a quartile comparison is, and is not

A blood-level quartile is not a grocery receipt. Witkowski’s group was not counting sticks of gum. They were ranking people by how much xylitol was circulating in their blood, then asking whether the people in the highest quarter of that distribution fared worse than the people in the lowest quarter. That design has a strength and a weakness, and both are obvious.

The strength is biological proximity. If xylitol is doing something to vessels or to clotting, the relevant exposure is not the marketing copy on a yogurt lid. It is the molecule in plasma. People who chew a lot of gum, swallow a lot of sugar-free mints, or eat a lot of “lite” ice cream will tend to sit higher in that distribution. So will people whose own metabolism, kidney function, or diet pattern leaves more of the compound unprocessed. The blood level captures the sum.

The weakness is the usual observational weakness. The people in the top quartile are not randomly assigned there. They may be trying to lose weight. They may have diabetes and have been told to avoid sugar. They may drink diet products because they already have heart disease in the family. They may be heavier, older, more hypertensive. If the analysis does not account for those differences, xylitol becomes a mascot for a lifestyle rather than a suspect in a mechanism.

Witkowski’s report tried to meet that objection on its own terms. The signal held after age, sex, BMI, hypertension, diabetes, and lipids. That list is not decorative. Those are the variables a critic reaches for first. Body mass index captures the dieting pathway. Diabetes captures the “I was told to swap sugar” pathway. Hypertension and lipids capture the already-at-risk pathway. Age and sex capture the basic actuarial facts of cardiovascular disease. After those adjustments, the highest xylitol group was still different from the lowest.

Fifty-seven percent at six years, eighteen percent at thirty

The headline numbers are large enough to travel and specific enough to be misquoted. The highest xylitol group was 57 percent more likely to die or have a heart attack or stroke within six years, and still 18 percent more likely after 30 years.

Those are relative risks, not a claim that 57 percent of heavy xylitol users will have an event. They describe a comparison between top and bottom quartiles of blood xylitol. The six-year window is the kind of interval contemporary cohorts can measure with some confidence. The thirty-year window is a rarer, longer look, and the residual 18 percent elevation is the part that keeps the finding from being dismissed as a short-term artifact of people who were already unwell.

Risk rose with dose. That matters more than a single top-versus-bottom split. A step-up from quartile to quartile, or from lower to higher circulating levels, is what pharmacologists look for when they are trying to decide whether they are staring at a threshold, a smooth curve, or noise. A dose relationship does not prove causation. The absence of one would have made causation harder to argue. Witkowski had the relationship.

The composite endpoint — die or have a heart attack or stroke — is the composite cardiology uses when it does not want to slice a rare outcome so thin that the statistics starve. Heart attack and stroke share a clotting biology. Death, in a middle-aged and older cohort, is often cardiovascular even when the death certificate is messier than that. Lumping them together is a choice. It is also why the platelet work, from earlier studies, sits so neatly beside this one.

Platelets that are easier to provoke

Earlier work showed xylitol makes platelets more eager to clot. That is the mechanistic sentence hanging under the epidemiology. Platelets are the small cell fragments that plug holes in blood vessels. Eagerness is useful if you are bleeding. It is less useful if a plaque in a coronary artery has cracked and the question is whether a clot will finish the job.

A sweetener that heightens platelet reactivity would not have to raise LDL, wreck the kidneys, or inflame every artery in the body in order to show up as more heart attacks and more strokes. It would only have to tilt the last step of a process that already happens millions of times a day without becoming a myocardial infarction. Cardiology has seen this movie with other exposures. Smoking does it. Some inflammatory states do it. Certain drugs do it on purpose, which is why antiplatelet pills exist.

The earlier platelet findings are why a blood-level study in 17,710 people is not arriving as a curiosity from nowhere. They are also why an observational association is harder to wave away as pure confounding, even though confounding remains possible. A critic can still say that people who consume a lot of sugar-free products are different. The critic now has to say they are different in a way that survives adjustment for age, sex, BMI, hypertension, diabetes, and lipids, tracks with dose, lasts into a thirty-year look, and happens to point in the same direction as a lab finding about clotting.

None of that is a courtroom verdict. It is a pattern.

Sold as the healthy swap

Artificial sweeteners are sold as the healthy swap and rated safe by regulators. That sentence is the cultural setup for Witkowski’s warning, and it is accurate as a description of the last twenty years of food marketing. Sugar became the villain. Products that could taste sweet without sugar became the moral alternative. Gum with xylitol was not only permitted. It was praised, especially in dentistry, because sugar alcohols do not feed the bacteria that cause cavities the way sucrose does. Toothpaste followed. Then jam. Then yogurt. Then ice cream. The same molecule moved from a niche oral-care claim into the general pantry.

Regulator safety ratings for sweeteners have historically been built around cancer bioassays, developmental toxicity, and acceptable daily intakes derived from those tests. They have not, as a class, been built around thirty-year coronary and stroke follow-up in tens of thousands of people ranked by blood levels. That mismatch is what Witkowski was pointing at in Munich. Witkowski said the long-term heart data say we barely know.

“Barely know” is not the same as “know it is deadly.” It is a statement about the evidence base that was used to declare these products the adult thing to eat if you are trying to be good. A compound added at about 1,000 times natural levels, in foods that are marketed to people already worried about their weight and their glucose, is a compound whose cardiovascular file should be thick. Witkowski’s argument is that the file is thin, and that the largest look yet is not reassuring.

The marketing has a second problem. Ultra-processed sugar-free foods are not gum plus water. They are formulations: bulkers, flavors, gums, oils, colors, and a sweetener that lets the front of the package do ethical work the back of the package cannot. If xylitol is the molecule in the blood, the product around it is still an ultra-processed object. Separating the sweetener from the food pattern is exactly what a blood-level analysis tries to do, and exactly what a shopper cannot do in an aisle.

Observational, not proof, and not a license for sugar

The pushback was as important as the finding, and it came from inside the same professional world. An ESC reviewer and the British Heart Foundation called it observational, not proof of cause, and not a green light for more sugar.

That is three claims in one, and all three are correct as far as they go. Observational means people were not randomized to xylitol or to placebo gum. Unmeasured confounding can still hide in the data. People in the top quartile may differ in exercise, diet quality, medication adherence, or the sheer volume of packaged food they eat, even after BMI and diabetes are on the table. Proof of cause, in the sense a drug trial would demand, is not what a congress presentation of this type can deliver.

The third clause is the one the internet will skip. A red flag on xylitol is not a medical argument for returning to sucrose-sweetened soda as a cardioprotective act. Sugar has its own, much larger, evidence base linking it to obesity, fatty liver, cavities, and metabolic disease. Not a green light for more sugar is the British Heart Foundation and the ESC reviewer trying to keep a complicated finding from being flattened into a cartoon in which the only two characters are “chemicals” and “natural cane sugar.”

Witkowski’s own framing sits next to that caution rather than against it. He did not, in the report as given, tell the public to dump toothpaste or to treat a stick of gum as a cigarette. He said the long-term heart data say we barely know. The barely-knowing is the news. The products are already on the shelf. The doses are already about a thousand times nature. The labels already say healthy.

A red flag on the sugar-free aisle

It is a red flag on ultra-processed ‘sugar-free’ foods as xylitol use climbs. Red flag is a precise metaphor. It is not a ban. It is not a recall. It is not a revised acceptable daily intake. It is a signal that the story consumers were sold — that sweetness without sugar is metabolically quiet — is not a story cardiology can currently underwrite with long-term data.

Use is climbing because the incentive is climbing. Every public campaign against sugar creates a market for replacements. Every dentist who recommends xylitol gum for cavities, for reasons that can be locally valid, adds a little more of the compound to a mouth that may also be eating sugar-free yogurt and sugar-free ice cream. The exposures stack. Blood levels in the top quartile are how that stack becomes a number.

What a cautious reader does with a red flag is boring, which is why it will be less popular than the headline. Treat “sugar-free” as a marketing phrase, not a health claim. Notice xylitol on a gum wrapper, a toothpaste tube, a jam jar, a yogurt cup, an ice cream tub. Understand that about 1,000 times natural dose is a different exposure than a trace in fruit. Understand that 17,710 people, ranked by blood level, showed a 57 percent higher chance of death, heart attack, or stroke over six years in the top quartile, 18 percent over thirty years, a dose relationship, and persistence after the usual risk-factor adjustments. Understand that platelets were already known, from earlier work, to become more eager to clot. Understand that Marco Witkowski, a Berlin cardiologist speaking at the ESC in Munich, says the long-term heart data say we barely know. Understand that an ESC reviewer and the British Heart Foundation refuse to call that proof, and refuse to bless a swing back to sugar.

The honest position after Munich is uncomfortable. A sweetener that dentistry likes, that regulators have rated safe, and that food companies have poured into the healthy-swap aisle now has the largest blood-level look yet tying high circulating amounts to the outcomes that fill coronary care units. The look is observational. The mechanism is biologically plausible. The doses are not natural. The use is climbing. The flag is red. It does not tell anyone to panic over a piece of gum. It tells them the word sugar-free was never a cardiac clearance, and that the molecule behind much of that word is no longer entitled to the benefit of the doubt it has enjoyed by default.