Sugar substitutes tied to higher insulin, HbA1c

Clinical takeaway: When counseling patients to cut added sugar, non-nutritive sweeteners remain a reasonable swap for soda. But routine or high-volume use is not a clearly safe default.

Non-nutritive sweeteners have spread through the US food supply on the assumption that swapping out calories is enough to make them metabolically harmless. A new review and meta-analysis complicates that assumption. Pooling 21 randomized trials that compared sweeteners against non-caloric controls such as water, researchers isolated the sweeteners' own effect on the body, separate from the calories they replace. The result: higher fasting insulin and HbA1c, with a trend toward worse insulin sensitivity.

Non-nutritive sweeteners are pitched as a lower-risk stand-in for added sugar, but the evidence isn't uniform across compounds. Sucralose has the strongest negatives: separate trials tie it to higher insulin and reduced insulin sensitivity, and one linked it to gut dysbiosis. Erythritol carries its own cardiovascular signal from cohort data.

Saccharin and stevia, by contrast, showed no change to gut microbiota or glucose tolerance in the trials reviewed. The new meta-analysis pools sweeteners together, which may blur these differences. It also inherits a reverse-causation problem from the observational cohorts it reviewed: people already at higher cardiometabolic risk may be more likely to use these products in the first place.

The pooled trials found the clearest signal in glycemic markers. Compared with non-caloric controls such as water or placebo, non-nutritive sweeteners raised fasting insulin and HbA1c. They also trended toward worse insulin sensitivity, though that result didn't reach statistical significance. Prior meta-analyses that used sugar as the control couldn't separate a sweetener's own effect from the effect of removing sugar's calories, but this design did.

In detailed microbiome profiling, paired with transfer experiments moving human gut microbes into mice, certain sweeteners altered both the makeup and function of gut bacteria. That's a single trial, not a replicated finding, but it gives the insulin and HbA1c signals a plausible biological path.

This narrative review and meta-analysis pooled randomized controlled trials in adults comparing non-nutritive sweeteners against non-caloric comparators (water or placebo), alongside a separate review of prospective cohort studies on non-nutritive sweeteners and cardiometabolic disease.

The authors call for more randomized trials, both on cardiometabolic risk factors directly and on the microbiome pathway itself, since the mechanism so far rests on a single trial. They also flag a policy gap: US labeling requires sweeteners to appear on ingredient lists but not in what amount, which makes it hard for future cohort studies to quantify exposure with any precision.

“The rapidly increasing use of these sweeteners has outpaced our understanding of their long-term health effects,” said senior study author Dariush Mozaffarian, MD, DrPH, cardiologist and director of the Food is Medicine Institute at the Friedman School of Nutrition Science and Policy at Tufts University. “Until we know more, caution is needed. If you're replacing large amounts of added sugar in your diet, such as in multiple servings of soda, these low-calorie sweeteners may be a better alternative. But we can't simply assume they are safe and innocuous, and avoiding them whenever possible appears a prudent choice,"

Source: Wang M, et al. (2026 Jun 25) Curr Atheroscler Rep. Artificial and other non-nutritive sweeteners, the microbiome, and cardiometabolic health