Artificial sweeteners and glucose intolerance – experts respond

A new study published in Nature suggests that consuming commonly used artificial sweeteners may actually increase the body’s post-meal uptake of sugars and the risk of developing glucose intolerance.

Credit: Steve Snodgrass (Flickr)
Credit: Steve Snodgrass (Flickr)

The Israeli researchers tested a range of non-caloric artificial sweeteners including saccharin, aspartame and sucralose, often used in diet soft drinks and sugar-free desserts, on mice and and collected data from almost 400 humans in a nutritional study over two and a half years.

They found that the sweeteners affect gut microbes, which in turn contributes to glucose intolerance and raised blood sugar levels. The study’s results suggest that artificial sweeteners may exacerbate, rather than prevent, metabolic disorders such as diabetes.

National media coverage of the study includes:

Radio New Zealand: Doubt cast on sweetener diabetes link
New Zealand Herald: Warning on artificial sweeteners
NZ Doctor: Artificial sweeteners linked to abnormal glucose metabolism

Our colleagues at the UK SMC have collected the following expert commentary. Feel free to use these quotes in your reporting. If you would like to contact a New Zealand expert, please contact the SMC (04 499 5476;

Prof Christine Williams, Professor of Human Nutrition at the University of Reading, comments:

“This is an extremely intriguing set of findings – with the animal studies carried out with a high degree of rigour. The use of the faecal transplant model is an impressive part of the study design, providing convincing evidence that the metabolic changes observed are linked directly with changes in the gut microbiome.

“However the human studies are much weaker; the observation that consumption of NAS is higher in individuals of heavier body weight and those showing metabolic disturbances is not in the least surprising, since many will have been attempting to lose weight over a numbers of years, with not unexpected observation of greater use of NAS in these individuals. Cause and effect has not been proven.

“Furthermore, the small human intervention study showed only 4 out of 7 individuals showed an increase in blood glucose following NAS administration. This could have been a chance finding. Together with the notable inter individual variability in glucose responses in the animals fed NAS, this suggests there are factors other than, or as well as, NAS consumption which are playing a part in these observations.”

Catherine Collins, Principal Dietitian at St George’s Hospital NHS Trust, comments:

“Suggesting sweeteners like saccharin may cause problems with blood sugar – when they are used by many attempting to lose weight or control blood sugar – is bound to generate worried interest from users. But if we separate the results from reality there’s no real human story here at all.

“If you’re a human trying to lose weight, you need to lose calories from your diet. Low calorie sweeteners help reduce your calorie intake from previous sugar or sugary drink use, and it’s a good lifestyle move for those wanting to lose weight and control blood sugar levels.

“However, if you’re a lab mouse, with your regular mouse chow providing a typical 60% of calories from fat, it seems that high dose sweeteners added to your drinking water alters blood glucose metabolism – and not in a good way. The researchers have shown that bacteria present in mouse bowels are altered when they’re fed sweetener-laced water, and they suggest these changes seem to influence blood sugar. On this evidence, I’d agree that lab mice shouldn’t have lots of sweeteners in their drinking water.

“But remember we’re talking lab mice here, and as their diet is normally 60% of calories from fat this means their usual carbohydrate intake must be much, much lower – perhaps 25% carbohydrate calories, and this mouse dietary profile is bound to influence the results as found.

“Conversely, we humans would find a 60% fat diet difficult to follow, with serious digestive symptoms a result in many. Around half the calories we humans eat each day are from carbohydrate with only 30% of our calories from fat. Our naturally higher carbohydrate intake has generated bowel bacteria happily digesting whatever we swallow, and their symbiotic relationship with our bowel cells and beyond is testimony to this. This is significantly different to the lab mouse and can’t be discounted when trying to suggest cross-species effect.

“Of dietetic concern is the ‘human study’ aspect, which studied 7 non-overweight, non-diabetic subjects, and recreated a false situation where the absolute maximum amount of saccharin suggested as an ‘acceptable daily intake’ was given to all subjects, divided into 3 daily doses. Four of the seven subjects showed a difference in bowel bacteria during the week, the other three showed no difference. With such a tiny human sample size, and such a balanced effect with massive saccharin intake, there is really nothing for we registered dieticians to take from this study.”

Dr John Menzies, Centre for Integrative Physiology, University of Edinburgh, comments:

“Artificial sweeteners are in widespread use as an alternative to sugar. In sharp contrast to sugar, these sweeteners contain next-to-no calories, and as such they are a useful and proven means of reducing total caloric intake to aid weight loss or maintain a healthy body weight. Artificial sweeteners have been used in foods and drinks for decades and current data indicate they are safe to eat, but some studies have shown an association between artificial sweetener consumption and weight gain. To date, no convincing mechanism for this association has been put forward.

“There is a physiological rise in blood glucose after sugar consumption that is restored by insulin release. However, in diabetes, blood glucose levels can be elevated higher than normal and for a longer period – this is the basis for many of the deleterious effects of the disease. This study describes an impairment in blood glucose control resembling that seen in diabetes in mice fed artificial sweeteners, not sugar, in their drinking water. It should be noted that high levels of artificial sweetener intake was needed to see an effect – equivalent to the FDA’s maximum acceptable daily intake – and that animals had no alternative fluid source. Caution is needed when comparing this experimental paradigm with human drinking behaviours.

“The study shows a clear role for the bacteria in the mice’s gut. There is an emerging understanding of the importance of gut bacteria in obesity. In humans, the types and activity of gut bacteria can depend on the diet of the host and certain profiles of gut bacteria are associated with an increased likelihood of the host developing metabolic diseases like non-alcoholic fatty liver disease. in the study, animals treated with antibiotics to suppress gut bacteria showed an improved response to the artificial sweeteners, and transplanting gut bacteria from artificial sweetener-fed mice to bacteria-free animals resulted in an impairment in the latter’s control of blood glucose.

“Is this relevant to humans? It is not currently clear whether artificial sweeteners can alter the profile of gut bacteria in humans or have a direct effect on metabolism. However, the study found a positive relationship between historical artificial sweetener consumption and some parameters of metabolic disease. But human diets are complex – consisting of many foods, the consumption of which can vary in amounts, and over time. Therefore, this relationship does not show that artificial sweeteners are responsible for changes in metabolism or increases in body weight. The study tested this directly by giving a small group of human volunteers high doses of an artificial sweetener for a short period while monitoring blood glucose. Only a small majority of the group (4 out of 7 tested) showed poorer responses to a glucose challenge.

“This animal study provides a basis for a mechanistic explanation of how artificial sweeteners may be involved in metabolic disease, but we do not yet know whether long-term artificial sweetener consumption in humans has the same effects, or whether the amounts of artificial sweeteners consumed in the typical human diet is high enough to put us at risk of metabolic diseases. More work is needed to understand better the metabolic effects of artificial sweeteners, but in contrast, the scientific evidence for the contribution of sugar over-consumption to obesity and its associated diseases – diabetes, hypertension and cancer – is very strong. As such, it seems premature to discourage the use of artificial sweeteners in favour of sugar.”

Dr Katarina Kos, Senior Lecturer and Consultant in Diabetes and Endocrinology at the University of Exeter, comments:

“The study is based primarily on mouse experiments and only seven human subjects were studied. The findings require further confirmation prior to making firm conclusions. Larger scale human studies and funding are urgently required controlling for overall calorie intake.

“Meanwhile, these findings support the widespread understanding that water is the healthiest drink option and that we should avoid sweet and sweetened drinks. Water is the best drink to control our blood sugar.”

Prof Sir Stephen O’Rahilly FRS, Professor of Clinical Biochemistry and Medicine at the University of Cambridge, comments:

“The evidence that the intake of sugared beverages contributes to recent increases in obesity is strong and growing. Recent randomised controlled trials in children have shown that substituting artificially sweetened fizzy drinks for sugared drinks reduces the risks of becoming overweight. This is the background against which the current study needs to be viewed.

“Most of the work reported by Suez et al is in mice. While the study of mice is an invaluable part of medical research it is always very difficult to know if disease mechanisms that are suggested in mice will be relevant to humans.

“The part of this report which reports effects in humans is not compelling as it is seriously underpowered. The authors report an association between artificially sweetened beverages and markers of diabetes in 381 people; however, a recent study involving more than a third of a million people showed no association between consumption of artificially sweetened drinks and the development of diabetes. Suez et al also report adverse effects on glucose level s after 7 days of saccharin ingestion. However these experiments were undertaken in only seven people, so must be deemed preliminary.

“In conclusion: this new report must be viewed very cautiously as it mostly reports findings in mice, accompanied by human studies so small as to be difficult to interpret.”

Dr Nita Forouhi, Programme Leader at the MRC Epidemiology Unit, University of Cambridge, comments:

“Through a series of elegant experiments in mice this research advances our understanding that non-caloric sweeteners have an important effect on changing the types and functions of the bacteria that colonise the gut. This in turn leads to metabolic changes in mice, such as glucose intolerance that have a knock-on effect on promoting type 2 diabetes.

“Though this is an important advance in understanding underlying pathways that might link NAS with metabolic disturbances in mice, the findings in humans are currently inconclusive, being based on 7 individuals, testing only one artificial sweetener (saccharin) in very high doses at the upper limit of the acceptable daily intake equivalent to about 40 cans per day of a typical artificially sweetened cola-type drink.

“This research raises caution that NAS may not represent the ‘innocent magic bullet’ they were intended to be to help with the obesity and diabetes epidemics, but it does not yet provide sufficient evidence to alter public health and clinical practice.

“The findings point to the importance and relevance of investing in future better designed specific research in humans, but until then we are not ready to adopt a change in current advice.”

Prof Brian Ratcliffe, Professor of Nutrition at Robert Gordon University, Aberdeen, comments:

“This interesting paper will raise some questions about how best to help people to lower their intakes of free sugars. SACN’s draft recommendations for the UK are in line with WHO’s suggesting that free sugars should be limited to a population average of 5% of dietary energy intake. Sheiham and James have argued just this week that the target should be less than 3% to avoid dental caries. Most of us greatly exceed this and people under 18 years consume an average that is around five times this amount.

“One strategy to reduce the intake of free sugars is to consume foods that are sweetened with ‘non-caloric artificial sweeteners’ (NAS). Unfortunately, the authors lump all these sweeteners together in the title of their paper. They looked at saccharin, sucralose and aspartame. Apart from providing sweetness, these NAS have very little in common chemically. But most of the effects that they report relate to saccharin with little or no effect of aspartame. Their paper ought to be limited to ‘saccharin’ in the title rather than attributing the effects to all NAS.

“The study raises some concerns about the widespread use of saccharin and should provoke further investigations. Sugar sweetened beverages (SSB) make a significant contribution to the intake of free sugars. The market leaders’ alternatives in the form of ‘diet’ drinks do not usually contain saccharin so there seems no reason to suggest that swapping to a diet version of your favourite fizzy drink is unwise as a strategy to reduce the intake of free sugars.”

Prof Naveed Sattar, Professor of Metabolic Medicine at the University of Glasgow, comments:

“This is an interesting finding in mice but we have to remember two important things.

“Animal data for many experiments do not show the same effect in humans, which can sometimes be quite the opposite. Hence one must be cautious in extrapolating the findings to humans.

“Current epidemiological data in humans do not support a meaningful link between diet drinks and risk for diabetes, whereas sugar rich beverages do appear to be associated with higher diabetes risk. So these findings would not make me choose sugary drinks over diet drinks.

“The findings of this study do not prove that sweeteners pose any real risk to humans. If there are any risks, we need well controlled studies in humans to find them.”

Gaynor Bussell, Dietitian and Public Health Nutritionist, comments:

“Our use of NAS is likely to rise, especially if the draft recommendations by SACN about needing to reduce sugar intake becomes main policy. The disruption of the human microbiota and its association with obesity and diabetes has been shown before, as has the change in microbiota with the use of NAS, but its significance is still not clear.

“The human element of this research was only conducted on 7 individuals and this is insufficient to be categorical about firm associations. Indeed, studies exist that demonstrate how NASs can be used effectively to reduce and prevent obesity. However, due to the likely increase in NAS use and with what we are beginning to find out about the gut microbiota and metabolic disease I would back the recommendations of the authors to conduct further studies in this area.”