Acute glycemic and insulinemic effects of low-energy sweeteners: a systematic review and meta-analysis of randomized controlled trials

Autor(en) : Greyling A, Appleton KM, Raben A, Mela DJ.
Name der Veröffentlichung : American Journal of Clinical Nutrition 2020; 112(4);1002-1014.
Erscheinungsjahr : 2020


Background: It has been suggested that low-energy sweeteners (LES) may be associated with an increased risk of metabolic diseases, possibly due to stimulation of glucose-responsive mechanisms.
Objective: We conducted a systematic review and meta-analysis of human intervention studies examining the acute effect of LES intake on postprandial glucose (PPG) and postprandial insulin (PPI) responses, in order to comprehensively and objectively quantify these relations.
Methods: We systematically searched the Medline, OVID FSTA, and SCOPUS databases until January 2020. Randomized controlled trials comparing acute postprandial effects on PPG and/or PPI after exposure to LES, either alone, with a meal, or with other nutrient containing preloads to the same intervention without LES were eligible for inclusion. PPG and PPI responses were calculated as mean incremental area under the curve divided by time. Metaanalyses were performed using random effects models with inverse variance weighing.
Results: Twenty-six papers (34 PPG trials and 29 PPI trials) were included. There were no reports of statistically significant differences in the effects of LES on PPG and PPI responses compared with control interventions. Pooled effects of LES intake on the mean change difference in PPG and PPI were -0.02 mmol/L (95% CI: -0.09, 0.05) and -2.39 pmol/L (95% CI: -11.83, 7.05), respectively. The results did not appreciably differ by the type or dose of LES consumed, cointervention type, or fasting glucose and insulin levels. Among patients with type 2 diabetes, the mean change difference indicated a smaller PPG response after exposure to LES compared with the control (-0.3 mmol/L; 95% CI: -0.53, -0.07).
Conclusions: Ingestion of LES, administered alone or in combination with a nutrient-containing preload, has no acute effects on the mean change in postprandial glycemic or insulinemic responses compared with a control intervention. Apart from a small beneficial effect on PPG (-0.3 mmol/L) in studies enrolling patients with type 2 diabetes, the effects did not differ by type or dose of LES, or fasting glucose or insulin levels. This review and meta-analysis was registered at PROSPERO as CRD42018099608.


This is the latest systematic review and meta-analysis of controlled human intervention trials studying the acute effects of low/no calorie sweeteners intake on postprandial glucose and insulin responses. In line with previous reviews, it found that the consumption of low/no calorie sweeteners had no impact on glucose and insulin responses post-prandially (120 min after consumption) compared with a control intervention, whether administered alone or in combination with a nutrient/ carbohydrate-containing load.

Furthermore, the results showed that effects do not differ appreciably by the type or the dose of sweetener consumed, confirming that all low/no calorie sweeteners tested do not increase glucose and insulin levels after their consumption. The study results are based on a total of 452 individual participants who took part in the 34 trials (55 comparisons) for postprandial glucose, and 394 participants in 29 trials (50 comparisons) for postprandial insulin.

The findings are in agreement with previous systematic reviews that also showed no impact of low/no calorie sweeteners on glucose and insulin levels, making them a significant alternative to sugar for people with diabetes. The current analysis found that among patients with type 2 diabetes, there was a smaller postprandial glucose response after exposure to low/no calorie sweeteners compared with the control, which might suggest a potential direct glucose-lowering benefit of sweetener intake for these individuals. However, effect sizes were small and based on limited comparisons with few participants, and therefore more studies are needed to investigate this finding in the future.

View full paper