August 19, 2022

expert reaction to study looking at the effect of non-nutritive sweeteners on human microbiomes and glycaemic levels

A study published Cell Press looks at microbiome-driven effects of non-nutritive sweeteners on human glucose tolerance.

Dr Sarah Berry, Senior Lecturer, King’s College London, said:

“There is little causal evidence in humans regarding the impact of sweeteners on the composition and function of the gut microbiome. Unlike previous studies this study has the strength of testing the most commonly consumed sweeteners, at realistic consumption doses in humans who don’t commonly consume sweeteners.

“This study contributes to our current knowledge by providing evidence in humans at realistic intake doses that sweetener consumption impacts the microbiome composition and function, which has previously only been well studies in animal models. This study confirms what has been previously shown in mice in relation to the impact of sweeteners on the microbiome and adds to the body of evidence that some sweeteners have an unfavourable effect on blood glucose control.

“This study shows that the effects of sweeteners on blood glucose is personalised, and that the microbiome may mediate some of the variability on how people respond to sweeteners. This may account for some of the conflicting evidence in previously published research in this area and lack of clarity on the effects of sweeteners on health. The finding that there are ‘responders’ and ‘non-responders’ to sweeteners also highlights the need to consider the health effects of foods at an individual level alongside population based advice.

“Although it is challenging to accurately measure sweetener intake from peoples typical diet, as sweeteners are found in such a wide range of foods, the authors tried to recruit individuals who do not typically consume sweeteners to more acutely study the effect of each sweetener on outcomes. The researchers recruited people who had not consumed sweetners in the past 6 months based on a food frequency questionnaire which allowed them to study the effect of the different sweeteners without confounders that may be introduced from previous exposure. However, by only recruiting individuals who did not consume sweeteners in their typical diet, we cannot be certain if we would see the same unfavourable effect for people that typically consume sweeteners, due to possible adaptation effects.

“The researchers studied the effect of sweeteners over a two-week period which is sufficient to study the effects on microbiome composition and blood glucose control as these can change quite rapidly, however it would be interesting to see if these unfavourable effects are sustained after long-term intake and if there are also subsequent effects on other health outcomes in the long term, such as blood pressure, insulin resistance and body weight.

“The intake of sweeteners has grown hugely in the last decade, largely alongside the growth in ultra-processed foods and the demand for low-sugar alternatives. Many people aren’t even aware that they are consuming sweeteners as they are in such a wide range of foods. Therefore, these results are relevant to a large proportion of the population, especially for those who select low-sugar drinks and foods as a healthy alternative.

“Although all 4 sweeteners impacted the microbiome composition, the findings that two of them (saccharin and sucralose) had an impact on blood glucose responses is especially interesting from a practical perspective. The higher blood glucose responses following two of the sweeteners (saccharin and sucralose) but not all (aspartame and stevia), could inform the food industry as well as consumers of healthier low-sugar options. However, more long-term studies would need to be conducted before re-formulation and consumer advice could be confidently delivered.

“The transient effect of sweeteners on blood glucose is interesting for consumers as this may suggest that the unfavourable effects of sweeteners can be reversed if they are removed from the diet. However, this study does not show the long-term health effects of sweeteners so whilst the research shows that their effect on blood glucose control is short term, they may have long-lasting unfavourable effects.

“The long-term effect of sweeteners on health cannot be extrapolated from this study and so whilst the effects on the microbiome and blood glucose control suggest an unfavourable impact of sweeteners on health, it is uncertain whether this would translate to long term unfavourable effects especially in people who typically consume sweeteners.

“The evidence from this new research and other studies show that sweeteners are not inert and whilst they are a better option than full sugar alternatives, my recommendation would be to try and avoid consuming them in excess.

“In the UK we are consuming too much sugar and there is very clear evidence that too much sugar unfavourably impacts our health from effects on inflammation and body weight, as well as how we feel such as our hunger and energy levels. Therefore, strategies to reduce sugar intake are needed and evidence from this research, to suggest that some sweeteners might be better for us than others, is an important step forward in informing the food industry and consumers of the potential for re-formulation and healthier sugar alternatives.”

Professor Kim Barrett, Vice Dean for Research and Distinguished Professor of Physiology and Membrane Biology, University of California Davis School of Medicine, said:

“Non-nutritive sweeteners (NNS) are highly popular and have been used extensively in attempts to combat the epidemics of obesity and metabolic diseases such as diabetes.  However, while NNS have traditionally been considered to be inert, evidence has emerged that they can impact both physiological signaling related to nutrient handling as well as the composition of the microbiome, at least in animals (although findings were sometimes conflicting).  The present study is a large step forward to understanding the impact of NNS in humans, with a very rigorous design and follow-up work in mice to assess underlying mechanisms and causality.  Importantly, the work is a major advance because the authors were careful to exclude participants who knowingly or unknowingly ingest NNS in their normal diet.  They found that, while there was variability between subjects, two NNS, saccharin and sucralose, provoked intolerance to a glucose load (a marker of metabolic dysfunction) and also caused specific changes in the gut microbiome and the chemicals these microorganisms produce.  Furthermore, by implanting the fecal microbiome from those subjects who showed greater or minimal glucose intolerance responses into germ-free mice, they could reproduce the respective glucose intolerance in the mice, implicating the effect of the NNS on the microbiome as the cause of the metabolic dysfunction.  Caveats about the study include the fact that only young, healthy subjects were included, and the NNS were only given for two weeks.  Nevertheless, this well-designed study indicates the potential for NNS to have adverse effects in at least some individuals, and should prompt additional work as well as, perhaps, providing an explanation that diet drink consumption is often associated with greater rather than reduced levels of obesity. While the findings have not been replicated in all studies, there are many that show that regular consumption of diet beverages is associated with an increased risk of obesity (as well as metabolic syndrome and other negative health outcomes).  The studies are, however, confounded by the possibility of reverse causality — those who are predisposed to obesity may be inclined to increase their intake of diet beverages — so additional well-controlled interventional trials are needed. Ultimately, perhaps we should not be surprised by the findings since NNS are recognized by human taste receptors even though they are non-caloric, and bacteria are also well-known to have the capacity to sense chemicals in their environment and to change their behaviour accordingly.”

Sarah Coe, Nutrition Scientist at the British Nutrition Foundation, said:

“Non-nutritive sweeteners have been critically evaluated by international authorities and are safe and approved for use in a range of food and drink products. There is often contradictory information in the media around the safety and health effects of sweeteners, which can cause confusion among consumers. These suggested effects of sweeteners on health include an influence on the gut microbiome. Evidence from animal studies shows that the consumption of non-nutritive sweeteners, in particular sucralose and saccharin, could influence the diversity and composition of the gut microbiome and in turn lead to abnormal blood sugar levels and impaired glucose tolerance. Current evidence in humans is limited and does not confirm an effect of non-nutritive sweeteners on the gut microbiome, and the findings from animal studies cannot be translated to human effects. 

“The findings of this latest study by Suez et al. add to the evidence to date suggesting an effect of the non-nutritive sweeteners sucralose and saccharin on the gut microbiome and the development of glucose intolerance. Unlike previous animal studies that have given very high doses of sweeteners, the researchers used doses below the Acceptable Daily Intake (ADI) levels set by the European Food Safety Authority (EFSA), which may be more comparable to current average intakes. Although the researchers collected dietary intake data during the study to monitor any changes in nutrient intake, diet was not controlled and this is something that may need to be considered to account for other dietary factors that may affect the gut microbiome. Participants were also healthy, something the researchers acknowledge as a limitation, as the effects of non-nutritive sweeteners may differ between healthy people and people with cardiometabolic disease. 

“This study does not indicate a need to change average consumption habits of non-nutritive sweeteners. It’s important that the science on the long-term effects of non-nutritive sweeteners continues to be reviewed, and further research with more long-term, randomised controlled trials in humans are needed to investigate the potential effects of non-nutritive sweeteners on the gut microbiota and how this relates to health outcomes and disease risk. Switching to foods with sweeteners (rather than sugar) is still one way consumers can manage their daily calorie intake as part of achieving a healthier, more balanced diet.”

Francisco Guarner, Director of the Digestive System Research Unit at the Vall d’Hebron University Hospital in Barcelona and former president of the Spanish Society of Microbiota, Probiotics and Prebiotics (SEMiPyP), said:

“It is neither definitive nor of high quality. They observed discrete alterations in the tolerance test after oral glucose overload in 20 individuals after two weeks of taking 180 mg of saccharin daily, or in 20 individuals after taking 102 mg of sucralose daily, compared to 20 controls [who did not take any sweetener]. The number of individuals in each arm [those taking a particular sweetener or serving as controls] is very small (N=20), although the study includes 120 in total.

“It is important to underline that they do not detect this negative effect in individuals treated with stevia or aspartame. Therefore, the discrete negative effect should not be attributed to all sweeteners.

“There are many discrepancies with the existing evidence. The new publication is not a clinical study, but an experimental study: the N (the number of individuals in the study) is small, they did not enrol people with insulin resistance and they use exorbitant doses of sweetener. This is a very important limitation: 180 mg of saccharin a day for two weeks is equivalent to taking 50 tablets or 18 sachets of saccharin every day. I do not know anyone who takes such quantities. In the case of sucralose, the intervention would involve taking 20 sucralose tablets every day. Nobody does that.

“Personally, I am of the opinion that there is only one acceptable conclusion, which is also important: the fact that a substance is not absorbable and therefore does not enter the blood does not mean that it is inert. The substance influences the microbiota of the large intestine and can induce negative or positive changes”.

Ascensión Marcos, Research Professor and Director of the Immunonutrition Group at the Institute of Food Science and Technology and Nutrition of the CSIC, said:

“This is a good study. However, as usual, one can always see ‘drawbacks’. Of the 19 sweeteners approved in the EU, only four appear in this article, so no results can be extrapolated. As we also found in a recently published review, the authors go so far as to observe an effect on the gut microbiota with altered glycaemic response for saccharin and sucralose.

“In any case, although the N (the number of participants in the study) is fine (120), there are only 20 in each arm of the study. In addition, the age range is very wide (18-70), both sexes are included (something that should always be taken into account because there are sex differences in terms of microbiota results) and the intervention time is short, only two weeks. In terms of nutritional status, there are several groups and obesity is not taken into account, only overweight.

“In principle, it would be preferable to differentiate by case, by nutritional situation, by pathology, even by geographical area, since in Latin America the population has been consuming different types of sweeteners for years and could possibly have adapted their microbiota.

“I think that, in general, taking these points into account, it may be difficult to draw conclusions”.

Dr Duane Mellor, Registered Dietitian and Senior Teaching Fellow, Aston Medical School, Aston University, said:

“This is a very detailed study looking at how 4 different non-nutritive sweeteners when consumed daily mixed with glucose for 2 weeks by people who normally do not consume sweeteners affect how their body handles a glucose load (glucose tolerance test). It is important to note that the sweeteners were not taken at the same time as the glucose load, and that not all sweeteners were linked to a worsening of glucose tolerance, this was seen with sucralose and saccharin, but not with aspartame and stevia.

“This study then looked at why sweeteners, which previously have been thought to be inert may be linked to this effect. This was done by looking at the different types of bacteria found in participants’ colons. The researchers initially thought that sucralose would alter colon bacteria levels the most as it is not well absorbed, but it was found that all four sweeteners were associated with a change in colon bacteria species, a change that was not seen in the control groups who consumed either just glucose or nothing. The reason for this is unclear as sweeteners such as aspartame will be almost completely broken down into amino acids (naturally found in protein-containing foods) and methanol (small amounts of this are found in fruit) which are mostly absorbed. Especially, when the levels of these compounds reaching the colon is also relatively small (milligrams per day), more work is needed to see how different sweeteners may influence colon bacteria levels. It was also apparent that not all individuals responded to sweeteners, to test the individual response the research team took samples from those who responded least and those who responded most to the sweeteners and transplanted these colonic bacteria into sterile rats. The rats given the bacteria from individuals who responded to sweeteners also had a decrease in tolerance to glucose, meaning their glucose level was higher for longer after a glucose load.

“This study suggests that sweeteners are not inert, which is the case for almost every compound in food, naturally occurring or not. The authors suggest that sugar is not ideal for metabolic health, the best option is water. However, this needs to be balanced with enjoying the food and drinks we consumed, and many people find pleasure from consuming sweet-tasting food and drinks. This study does not show a link between all non-nutritive sweeteners and higher blood glucose levels long-term (only after a glucose tolerance test). It did suggest thought that some individuals who do not normally consume sweeteners may not tolerate glucose as well after consuming 6 sachets of either saccharin or sucralose mixed with glucose per day. It does not provide any information about how people who normally consume sweeteners or people with either type 1 or type 2 diabetes respond to non-nutritive sweeteners. Therefore, for some people, it is likely to be a better option and more sustainable approach, to use sweeteners as a ‘stepping stone’ allowing them to reduce the amount of added sugar in foods and drinks, to reduce their sugar intake and still enjoy what they eat and drink, on the way to reducing both added sugar and sweeteners in their diet.”

Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“This is a complicated study, in its statistical aspects and in other ways. In statistical terms, the researchers have done a good job of using appropriate statistical methods and reporting them properly. I think most of the interesting points in the research relate to what it says about the possible effects of the sweeteners in the human body, and in the details of how those effects might arise and operate, and I’m no expert on those aspects. No research study can ever be perfect, but this one looks good to me. Subject to some limitations, I think it does establish that consuming these sweeteners can, in some people, have an effect on the microbes in their gut, and, in some people and with some sweeteners, that may lead to changes in the way their bodies deal with sugars. But exactly what the health consequences of all this, if any, might be, is a subject for future research.

“It’s important to understand that the research is not saying that these sweeteners are worse for us, in heath terms, than sugar. The final quote from Eran Elinav in the press release does explicitly say that sugar “is clearly most harmful to our metabolic health.

“The researchers found that all four of the sweeteners that they considered did lead to measurable changes in the microbes in the guts of the participants, and that two of the sweeteners also affected glucose tolerance (the way that the levels of sugars in the participants’ bodies changed after they consumed a given amount of the sugar glucose). Because this was a randomised trial, where participants were allocated a treatment (one of the sweeteners, or one of the two control treatments that did not include any of the four sweeteners) at random, really the only possible causes for these differences between people are either that they took different sweeteners (or no sweetener). The random allocation essentially means that the only possible explanations for differences between participant groups are either the sweeteners or the effects of chance, and the statistical analyses show that it’s pretty unlikely that only chance was involved.

“What this work in the human subjects can’t show directly is whether the changes in glucose tolerance were caused by the changes in the gut microbes, rather than being a response to the sweeteners that works in some different way. The researchers dealt with that by using laboratory mice that were germ-free, transferring microbes from some of the humans’ guts to the mice, and observing what happened to glucose tolerance in the mice. They found similar changes in the responses of the mice as were found in the humans, and interestingly, changes in the mice were seen only in mice that were given microbes from the humans who showed high levels of changes in their gut microbes.

“The researchers conclude that these sweeteners can affect gut microbes in humans, that the extent to which this happens does vary between individuals, and that it seems likely that these changes can, on average, affect the body’s response to glucose. Rightly, the researchers are somewhat cautious in describing these findings, and they don’t say what the consequences for human health might be, only that their findings indicate that further research should be done to see what the health consequences (if any) might be.”

“There are some limitations to take into account. Because the point of the study was to establish whether these effects exist, the research was (rightly) carried out in a fairly tightly controlled way, with only one way of administering the sweeteners, a fairly short time period of taking the sweeteners (or control treatments), and one specific way of measuring glucose tolerance. Also, the human participants involved weren’t typical of the whole population – they were not overweight, didn’t have known diabetes or quite a wide range of other health problems, and (very unusually in most high-income countries) they had not consumed any non-nutritive sweeteners in the six months before the study. It’s possible that things would have turned out differently in other groups of people; the researchers acknowledge this, but further investigations are for future research.

“A key aspect of the research is the use of the germ-free mice to establish that the changes in glucose tolerance are caused by the changes in gut microbes. That’s a good way of filling what would otherwise be a big hole in the understanding of how all this might work, but it does perhaps still raise the question of the extent to which the workings of gut microbes in mice and in humans are sufficiently similar – and I’m not qualified to comment on that.”

Dr Kathy Redfern, Lecturer in Human Nutrition, University of Plymouth, said:

“This study by Suez and colleagues is a well-designed randomised controlled trial which begins to fill the gap in the current literature, which currently mostly relies on results from observational studies. This study investigated four commonly consumed non-nutritive sweeteners that food manufacturers routinely add to our foods and drinks to give a sweet taste, without adding calories. It’s important to note, however, that there are other sweeteners that we commonly consume in the UK that weren’t tested in this study, one of which is Acesulfame K, which is the sweetener most commonly added to ‘diet’ soft drinks in the UK alongside aspartame – which was tested in this study, but did not appear to influence glucose tolerance in humans.

“Although participants were screened so that only those consuming no products containing NNS in the last six months, and who did not have obesity or pre-existing metabolic disease were included in the study, these findings do not tell us anything about how the human body may adapt to NNS products over time, especially as those consuming NNS typically consume them over years, if not decades, rather than the short period observed in the current study. This is an important study, and adds to the body of evidence that suggests that NNS may not as inert in the human body as we first thought, which certainly warrants further study.

“We still have a lot to learn about the human microbiome, and although this study suggests two of the sweeteners tested in this study (sucralose and saccharin) significantly affected glucose tolerance, these deviations were small, but nevertheless warrant further investigation to assess how small changes in glucose tolerance in response to NNS consumption may influence longer term glucose tolerance and risk for metabolic complications such as Type 2 Diabetes.

“It’s also important to note that the non-nutritive sweeteners added to our foods have undergone rigorous safety tests by the European Food Safety Authority in Europe, and the Food and Drug Administration in the USA, and are currently considered safe. Anyone who currently chooses foods containing non-nutritive sweeteners does not need to panic or dramatically change their dietary habits, however, as a Registered Nutritionist, and as highlighted by the authors of the paper, the best drink for hydration is water. Soft drinks, whether sweetened with non-nutritive sweeteners or sugar, can be enjoyed in moderation.”

‘Personalized microbiome-driven effects of non-nutritive sweeteners on human glucose tolerance’ by Jotham Suez et al. was published in Cell Press at 16:00 UK time on Friday 19th August 2022.

DOI: 10.1016/j.cell.2022.07.016

Declared interests

Dr Sarah Berry: “I provide consultancy for ZOE Ltd (a healthcare tech company).”

Prof Kim Barrett: “I have no conflicts of interest to declare.”

Sarah Coe: “Funding to support the British Nutrition Foundation’s charitable aims and objectives comes from a range of sources including membership, donations and project grants from food producers and manufacturers, retailers and food service companies, contracts with government departments; conferences, publications and training; overseas projects; funding from grant providing bodies, trusts and other charities. Further information about the British Nutrition Foundation’s activities and funding can be found at http://www.nutrition.org.uk/aboutbnf/”

Ascensión Marcos: “I have no conflicts of interest to declare.”

Francisco Guarner: “I have no conflicts of interest to declare.”

Dr Kathy Redfern: “I have no conflicts of interest to declare.”

Dr Duane Mellor has provided consultancy to the International Sweetener Agency and has worked on Food Standards Agency funded projects investigating health effects of aspartame.

Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee.  My quote above is in my capacity as an independent professional statistician.”

For all other experts, no reply to our request for DOIs was received.