New Meta-Analysis: Should We Be Concerned About Sweeteners?

5 days ago

Neue Meta-Analyse: Sind Süßungsmittel ein Problem?

5 days ago

Sweeteners are frequently the subject of criticism. A new meta-analysis of randomized human trials provides indications of potential effects on glucose metabolism. But anyone concluding from this that sweeteners are inherently problematic is overlooking a crucial point: stevia is not sucralose. And comparing sweeteners with water answers a different question than comparing them with sugar.

Sweeteners are one of those nutrition topics where simple answers are particularly tempting. Some see them as a useful way to reduce sugar intake. Others consider them problematic in principle.

That is why we have already taken an in-depth look at stevia, including the available research, different steviol glycosides, raw material quality and dosage. You can find our detailed article here:
→ Stevia fact check: How safe is this sugar substitute?

We are taking the publication of a new meta-analysis as an opportunity for an update. The authors evaluated 21 randomized controlled trials (RCTs) investigating different sweeteners. The comparison groups received calorie-free controls such as water or placebo, rather than sugar.

This is important. It means the analysis addresses a very specific question: What effects do sweeteners have when the benefit of reducing calories from sugar is not taken into account?

At first, the results look concerning

On average, the sweeteners studied were associated with slightly higher fasting insulin and a marginally higher HbA1c. No clear effect was found for fasting glucose or blood lipids.

But how large were these differences in practice? For HbA1c, a marker of long-term blood glucose levels, the difference was 0.07 percentage points. Relative to a baseline value of 5.3%, this corresponds to an increase of around 1.3%. Put simply, 5.3% would mathematically become approximately 5.37%. The difference in fasting insulin was also small.

Whether changes of this magnitude have any long-term relevance for the health of healthy people cannot be determined from these findings. There is another important consideration: the term sweeteners groups together substances with very different chemical structures.

Stevia, sucralose, aspartame and saccharin may all taste sweet. That does not mean they have identical effects in the body.

Stevia is not sucralose

How relevant this distinction may be is illustrated by a randomized human trial involving 120 healthy adults. Participants consumed saccharin, sucralose, aspartame or stevia for two weeks at doses below the respective acceptable daily intake.

In this study, saccharin and sucralose significantly impaired glycemic responses. No such significant effect was observed for stevia or aspartame.

At the same time, all four sweeteners altered certain features of the gut and oral microbiome. What these changes mean for long-term health is not yet sufficiently understood.

Other studies do not always reach the same conclusions. In addition to the specific substance, dosage, duration of consumption and individual differences are likely to be relevant.

The more useful question is therefore not: Are sweeteners healthy or unhealthy? Instead, we should ask: Which sweetener, at what dose, for how long and compared with what?

Comparing sweeteners with sugar changes the question

In everyday life, sweeteners are often used to replace sugar. This changes the scientific question.

A good example comes from a six-month randomized trial involving 47 overweight adults. Participants drank one liter per day of sugar-sweetened cola, diet cola, low-fat milk or water.

In the sugar-sweetened cola group, the increase in liver fat was 132 to 143% greater than in the other three groups. Less favorable changes were also observed in skeletal muscle fat, visceral fat, triglycerides and total cholesterol. For the parameters studied, diet cola produced largely similar effects to water.

This illustrates why the comparison matters. Asking whether a sweetener is metabolically neutral compared with water is not the same as asking whether replacing sugar with a sweetener can be beneficial.

What does the new analysis tell us?

The meta-analysis provides no reason to assume that sweeteners are universally harmless. But neither does it provide a basis for treating all sweeteners as a single metabolic problem.

What matters is the specific substance, the dose, the duration of consumption and what it is being compared with. At the same time, a marginal change in individual biomarkers does not automatically mean that it will lead to meaningful long-term health consequences.

And what does this mean for stevia?

This distinction is relevant to our own formulations because we use stevia extract.

The human studies discussed here do not show the same specific signal for impaired glucose tolerance with stevia as observed with saccharin and sucralose. This means that findings for other sweeteners cannot simply be extrapolated to stevia.

You can learn more about the research, steviol glycosides and safety assessment in our stevia fact check. We explain the quality we use ourselves and why we choose an extract containing 97% Rebaudioside A on our stevia page.