Impact of Erythritol and Xylitol on Glucose Metabolism and Gut Microbiota in Obese Individuals: A Randomized Controlled Trial

Objective The aim of this study was to examine whether glucose tolerance, metabolomic profile, and gut microbiota composition is influenced by regular intake of erythritol or xylitol in non-diabetic, obese patients. Background Sugar alcohols such as erythritol and xylitol are on the rise as substitutes for sugar in the food industry and are being suggested for individuals managing obesity and diabetes. Despite their widespread usage and the general perception of safety, there is a notable absence of human trials that delve into their possible impacts on metabolism and the composition of gut microbiota. Methods Obese volunteers (n = 64; BMI: 37.9 ± 0.7 kg/m ² ; age: 34.2 ± 1.1 years, 70.3% female) were recruited in Switzerland, Russia and Norway. Participants were randomized to three groups (A = xylitol, B = erythritol, C = controls), and received: i) 36 g erythritol, or ii) 24 g xylitol per day, or iii) no treatment (control group) over a period of 7 weeks. At baseline and 7 weeks post-treatment, body weight, abdominal circumference, and blood pressure were measured; an oral glucose tolerance test (oGTT) was carried out; blood samples were collected to assess glycemic control (fasting glucose and insulin, HOMA-IR, fructosamine, AUC glucose/insulin during an oGTT), blood lipid profile, hepatic enzymes, inflammatory markers, and markers of intestinal permeability. Food diaries were collected, gastrointestinal symptoms were assessed by means of a validated questionnaire (GSRS), and stool consistency and form described by use of the Bristol Stool Scale (BSS). Changes in metabolomics patterns (plasma and urine) and gut microbiota composition were assessed by Ultra-Performance Liquid Chromatography coupled to Mass Spectrometry (UPLC-MS) and by 16S rDNA amplicon sequencing. Results Although there was no effect on instantaneous and dynamic metabolic homeostasis readouts (fasting blood glucose, insulin HOMA-IR and oGTT), we report a significant effect of sweeteners intake on medium-term glycemic control (fructosamine) after 7 weeks. We show that gut microbiota and metabolomic profiles track the individual fructosamine response to sweetener intake through a range of taxa and pathways. Lipids and carnitines of microbial origin (propionyl-, butyryl- and phenylacetyl- carnitines), indoles and bile acids modulate the fructosamine response to sweetener intake. Whilst some of these metabolites show concordant associations in normoglycemic and prediabetic volunteers, we found that microbiome metabolite associations differed based on prediabetes status, typically for phenylacetylcarnitine, trimethylamine N -oxide and benzoate metabolism. Conclusions Our study adds to the ongoing debate on sweeteners, and how the microbiome impacts glycemic responses to sweetener intake, suggesting a switch from glucose to lipid utilization and a shift in immune metabolic mediators mediated by the microbiome.