A diet soft drink in a shopping basket now carries an argument with it. One headline says aspartame is a possible carcinogen. Another says erythritol makes blood clot more readily. A third says sweeteners do not help with weight anyway. All three appeared within months of each other, and all three were narrowly accurate.
The trouble is the word "sweeteners". Aspartame, sucralose, acesulfame K, saccharin, erythritol, steviol glycosides and monk fruit share exactly one property, sweetness without much sugar, and almost nothing else. Different chemistry, different absorption, different regulatory files, different quality of evidence. Filing them under a single verdict is what makes this topic feel unresolvable.
The honest answer before the detail. At the doses people actually consume, the approved sweeteners have not been shown to cause harm, and the safety margins are wider than almost anyone expects. Erythritol carries the one genuinely open question. And swapping a sugared drink for a sweetened one removes a large sugar exposure, which matters because the realistic alternative in most lives is not water.
What does an acceptable daily intake actually mean in cans
Aspartame has an acceptable daily intake of 40 mg per kilogram of body weight per day. EFSA arrived at that figure in its 2013 re-evaluation and JECFA uses the same number (EFSA Journal 2013). An ADI is not the point where harm begins. It is the intake judged safe daily for a lifetime, usually the highest no-effect dose in animal studies divided by 100.
Translated into cans, the number startles people. WHO published its own worked example: with 200 to 300 mg of aspartame per can, a 70 kg adult would have to drink more than 9 to 14 cans a day to exceed the ADI, assuming no other source (WHO, 2023). Many diet colas sit well under that range, which pushes the figure higher still, as the Coke Zero question works through.
Acesulfame K moved against the mood of the coverage. EFSA re-evaluated it in April 2025 and raised the ADI from 9 to 15 mg/kg per day (EFSA Journal 2025). The one absolute exception: aspartame is a source of phenylalanine, so the ADI does not apply in phenylketonuria.
Why did WHO call aspartame a possible carcinogen and safe on the same day
On 14 July 2023 two WHO bodies published simultaneously, and the coverage read as self-contradiction. IARC placed aspartame in Group 2B, possibly carcinogenic to humans, citing limited evidence for hepatocellular carcinoma in humans, limited evidence in animals and limited mechanistic evidence. JECFA, looking at the same literature, found no sufficient reason to change the acceptable daily intake of 0 to 40 mg/kg and reaffirmed it.
They were not answering the same question. IARC identifies hazards, meaning whether something can cause cancer under some exposure, and its four categories describe how strong the evidence is, not how large the risk is. JECFA assesses risk at realistic exposure. WHO's own nutrition director put the reconciliation on the record: safety is not a major concern at commonly used doses, and the effects described need better studies. The US FDA went further and said it disagreed with IARC's conclusion outright.
Why does WHO advise against sweeteners for weight control
In May 2023 WHO issued a guideline recommending that non-sugar sweeteners not be used to control body weight or reduce the risk of noncommunicable diseases. The small print comes from WHO itself: the recommendation is conditional, rests on low certainty evidence, and does not apply to people with pre-existing diabetes (WHO, 2023).
Certainty is low because the evidence points two ways. Randomised trials tend to show small reductions in body fat, prospective cohorts the reverse, and WHO names the likely reason: the link may be confounded by baseline characteristics and complicated patterns of use. That is reverse causation, stated politely. People who drink diet soft drinks are disproportionately people already trying to manage weight or blood sugar.
Substitution trials give the cleaner read. Across six randomised trials in 1,729 participants, replacing an existing intake of sugar-sweetened drinks with non-caloric alternatives produced a BMI reduction of 0.31 kg/m2, roughly 0.5 to 1 kg, and weight drifted back once one intervention stopped (Obesity Reviews 2024). Modest, real, and conditional. For scale on the other side, NutriNet-Santé reported a hazard ratio of 1.09 for cardiovascular disease in higher consumers, 346 versus 314 events per 100,000 person years (Debras et al., BMJ 2022).
Is erythritol safe after the platelet studies
Erythritol is the sweetener where "not yet known" is the accurate answer rather than a dodge. In 2023 Witkowski and colleagues reported in Nature Medicine that circulating erythritol predicted three-year major adverse cardiovascular events across three cohorts of cardiac patients, with adjusted hazard ratios of 1.80 and 2.21 for the highest versus lowest quartile in the two validation cohorts. Erythritol also enhanced platelet aggregation in vitro and thrombosis in mice (Nat Med 2023). In 2024 the same group ran the controlled comparison the first paper lacked: 30 g of erythritol, but not 30 g of glucose, acutely increased platelet aggregation and granule release in healthy volunteers (ATVB 2024).
The mechanism looks real. Whether an ordinary sweetener habit produces it is a separate claim, and three caveats sit in between. Those cohort samples were drawn from 2007 to 2009, two supporting datasets go back to 1987, before erythritol was in wide use as a sweetener at all, and no dietary data was collected. Humans also synthesise erythritol from glucose via the pentose phosphate pathway, and that production rises with hyperglycaemia and oxidative stress, so a high plasma level may be a readout of metabolic trouble rather than its cause. And the doses are far apart: 30 g in 300 ml drunk in two minutes on an empty stomach, against EU rules capping erythritol as a flavour enhancer in energy-reduced flavoured drinks at 1.6%, or about 4.8 g per 300 ml (Cramer et al., Front Nutr 2023).
The largest follow-up reinforces the caution rather than the alarm. Among 4,006 older adults in the ARIC study, erythritol and its metabolite erythronate were associated with heart failure, cardiovascular death and total mortality, and the authors concluded that both are markers of cardiometabolic health (JACC Adv 2025). Marker carries most of the weight there. Erythritol in large daily doses is neither established as harmful nor as safe, and someone baking with it by the tablespoon has more reason to follow this literature than someone whose exposure is a square of sugar-free chocolate.
What about sucralose acesulfame K stevia and monk fruit
The strongest human trial here is Suez and colleagues in Cell in 2022: 120 healthy adults, two weeks, four sweeteners at doses below the ADI. Each altered stool and oral microbiome composition and the plasma metabolome. Saccharin and sucralose significantly impaired glycaemic responses, while aspartame and stevia did not. Germ-free mice colonised with stool from the strongest and weakest responders reproduced their donor's glycaemic pattern (Suez et al., Cell 2022). Two weeks in 120 people does not show that a diet drink damages metabolism, only that treating sweeteners as inert is wrong for some people.
Plant-derived is a marketing fact, not a physiological one. Steviol glycosides carry an ADI of 4 mg/kg per day as steviol equivalents, set by JECFA and matched by EFSA. Monk fruit is a European oddity: EFSA's 2019 opinion found the toxicological database insufficient to conclude on safety as a food additive (EFSA Journal 2019), and since Regulation (EU) 2024/2345 only one aqueous extract is authorised, with purified mogrosides still outside. An incomplete file is a different statement from unsafe.
Why the sweetener answer is an individual one
Averages across 120 people hide the two at either end, and that is the finding here that has held up best.
Take two readers with the same litre of diet cola a day. The first replaced a litre of sugared cola, roughly 100 g of sugar, so the swap removed an exposure whose evidence base is far stronger, and far less flattering, than anything published about sweeteners. The second added the diet version to an already sweet diet, so nothing was removed. Identical intake, opposite meaning, and the same logic runs through the energy drinks question.
Then the responder problem. Whether sucralose moves your glucose curve appears to depend on your microbiome. Whether your plasma erythritol reflects your diet or your pentose phosphate pathway depends on your glycaemic control. Neither is answerable from a label.
The honest limit belongs before the panel rather than after it: no routine blood test measures a sweetener's effect. There is no aspartame marker. A panel shows whether the metabolic picture around the habit is holding steady or drifting, and five markers do most of that work.
Five markers that answer the sweetener question indirectly
| Marker |
What it tells you here |
How fast it responds |
| Fasting glucose |
Where baseline glucose control sits before any meal |
Days to weeks, and a single reading is noisy |
| HbA1c |
Average glucose exposure over roughly three months |
Eight to twelve weeks after a real change |
| Fasting insulin |
How hard the pancreas works to hold glucose steady |
Weeks, often before glucose moves at all |
| hs-CRP |
Low-grade inflammation across the whole diet |
Days, so an infection distorts one reading |
| Triglycerides |
The lipid most responsive to sugar and alcohol intake |
Two to four weeks, and fasting state matters |
Together they answer a narrower question than "are sweeteners bad", and a more useful one: whether glucose control around this habit is holding.
How to see your own numbers
An Aniva membership costs EUR 199 per year and covers more than 100 biomarkers across 10 physiological systems, fasting glucose, HbA1c, fasting insulin, hs-CRP and the full lipid panel among them. The draw takes place at a partner location in Germany or Finland and is performed by licensed physicians. Results arrive in the app in about a week with a personalised action plan, a biological age estimate and an ongoing personal health concierge chat. Membership is up to 100% reimbursable via German private health insurance (PKV), depending on your plan.
Retesting sits inside the annual membership cycle. Hold the habit constant for a season and see what the numbers did, or change it deliberately and look again. Getting started takes a few minutes, and the daily oats question follows the same method.
Frequently asked questions
Are artificial sweeteners bad for you
At the doses people actually consume, the approved sweeteners have not been shown to cause harm, and safety margins are wide. Sweeteners are not one category, so the useful question is which one and how much. The strongest evidence in this area is not about sweeteners at all, it is about the sugar they replace.
How much aspartame is too much
The acceptable daily intake is 40 mg per kilogram of body weight per day, set by both EFSA and JECFA. WHO's worked example is that a 70 kg adult would need more than 9 to 14 cans of a diet soft drink a day to exceed it. The exception is phenylketonuria, where the ADI does not apply.
Is erythritol safe
The honest answer is unresolved. A 2023 Nature Medicine study linked higher plasma erythritol to cardiovascular events and to increased platelet reactivity, and a 2024 trial confirmed the platelet effect after a 30 g dose. The caveats are substantial: those cohorts collected no dietary data, humans produce erythritol themselves via the pentose phosphate pathway, and the test dose was several times what a sweetened drink contains.
Do sweeteners raise blood sugar or insulin
On average, very little. In a randomised trial of 120 adults published in Cell in 2022, saccharin and sucralose significantly impaired glycaemic responses over two weeks while aspartame and stevia did not, and responses varied strongly between individuals. That is a reason to measure rather than to assume.
Can a blood test show whether sweeteners are affecting me
Not directly. No routine panel measures a sweetener or its effect. Fasting glucose, HbA1c, fasting insulin, hs-CRP and triglycerides show whether the metabolic picture around a habit is stable or drifting, which is a narrower answer but an actionable one.
This article is general information about sweeteners and the blood markers that surround them. It is not medical advice, and nothing here diagnoses anything. Reference ranges are set by the analysing laboratory and differ between labs and methods, so a glucose or insulin result flagged in one report may sit inside range in another. Discuss your own results, medical history and any medication with a clinician.
