In an effort to reduce sugar consumption to prevent diabetes mellitus and cardiovascular diseases, “sugar-free” or “no added sugar” products that substitute sugar with non-nutritive sweeteners (NNSs) (eg, Splenda, Sweet’N Low, and Stevia) have become increasingly popular. The use of these products during pregnancy has also increased, with approximately 30% of pregnant women reporting intentional NNS consumption. In clinical studies with nonpregnant participants and animal models, NNSs were shown to alter gut hormonal secretion, glucose absorption, appetite, kidney function, in vitro insulin secretion, adipogenesis, and microbiome dysbiosis of gut bacteria. In pregnant animal models, NNS consumption has been associated with altered sweet taste preference later in life and metabolic dysregulations in the offspring (eg, elevated body mass index, increased risk of obesity, microbiome dysbiosis, and abnormal liver function tests). Despite the accumulating evidence, no specific guidelines for NNS consumption are available for pregnant women. Furthermore, there are limited clinical studies on the effects of NNS consumption during pregnancy and postpartum and long-term outcomes in the offspring.
Studies have shown that almost all pregnant women in the United States have measurable levels of chemicals in their bodies that pose threats to the development of the fetus. , These chemicals include phthalates, bisphenol A, flame retardants, tobacco, pesticides, lead, polychlorinated biphenyls, and mercury and have been linked to preterm birth, congenital anomalies, and neurodevelopmental disorders. , In a critical time frame such as embryonic and fetal development, even minimal exposure to toxic chemicals can be detrimental to fetal development and have lasting negative effects leading to increased susceptibility and development of disease from childhood to adulthood. ,
In addition to environmental chemicals, there is a growing list of comestible products that should be limited or avoided while pregnant (ie, caffeine or alcohol) because they lead to impaired neurologic and metabolic development of the fetus. Among emerging food additives that are raising concerns, non-nutritive sweeteners (NNSs) approved for consumption have been challenged for their safety in children and adults. , NNSs are zero- or low-calorie alternatives to nutritive sweeteners, such as table sugar. The initial uses of NNSs were limited to tabletop packets and diet soft drinks, and they are now widely used in thousands of foods and beverages. Currently, there are no guidelines for NNS consumption in pregnancy despite the accumulating evidence of deleterious effects after their consumption in animal models and in humans. This article intends to summarize types of NNSs used in the United States, their prevalence in the diets of pregnant women, and current evidence on the effects of consumption of NNSs during pregnancy and lactation on maternal and child health. In addition, this article identifies areas for future research in perinatal NNS exposure.
Non-nutritive sweeteners
In the past 20 years, the rate of obesity in the United States has increased rapidly and is contributing to the prevalence of diabetes and heart disease. Several national associations, including the American Heart Association and the American Diabetes Association, recommend limiting added sugar by replacing sugary food and drinks with sugar-free options containing NNSs. NNSs approved for consumption in the United States include sucralose, acesulfame K, aspartame, saccharin, and stevioside, a natural NNS found in the stevia leaf extract ( Figure 1 ). Consequently, reported NNS consumption has increased from 8.7% to 25.1% in children and from 26.9% to 41.4% in adults from 1999–2000 to 2009–2012 with most reporting use once daily (80% of children and 56% of adults) and with consumption higher in females than males among adults. Products containing NNSs include diet and reduced-calorie beverages (including Pedialyte), condiments, yogurts, fruit cups, baked goods, candies, and low-calorie dessert products. In addition, tabletop packets of NNSs are regularly found in restaurants and are becoming a common at-home baking ingredient.
NNS regulation and labeling
Premarket safety review and approval for NNS differ on the basis of origin (eg, artificial or naturally occurring). The Food and Drug Administration (FDA) has approved 6 artificial NNSs and provided acceptable daily intake (ADI) values for each based on the amount safe to consume per kilogram of body weight per day over a lifetime ( Table 1 ). Natural NNSs (ie, Stevia) are categorized as “generally recognized as safe” food ingredients and are not required to go through the FDA’s premarket approval process.
NNS | ADI (mg/kg of body weight) | ADI equivalence (for an adult weighing 70 kg [154 lb]) | ||
---|---|---|---|---|
mg | 12-oz diet soda, n | Tabletop sweetener packets, n | ||
Saccharin | 15 | 1050 | 16 | 52.5 |
Aspartame | 50 | 3500 | 18 | 87 |
Acesulfame K | 15 | 1050 | 35 | 26 |
Sucralose | 5 | 350 | 8 | 27 |
Steviosides | 4 | 280 | – | 10.5 |
The FDA’s regulatory processes for NNSs are controversial owing to concerns with unknown effects of long-term consumption, and manufacturers are not required by the FDA to disclose NNS amounts per serving on food or beverage products. This lack of disclosure creates challenges for consumers with identifying NNSs in products and, if so, how much. In addition, NNSs are increasingly being found in unsuspecting products, such as “no sugar added” or “reduced sugar” products, and nondiet products, such as regular sodas, juices, and condiments, making it difficult for consumers to consciously avoid NNSs.
NNS basic properties
Products sweetened with NNSs elicit fewer reward responses in the central nervous system when compared with sugar-sweetened foods or beverages. However, the ranges of NNS response are mostly attributed to sweet taste receptor binding. Interestingly, this receptor is found not only in the tongue but also in the intestines, lungs, bones, adipose tissue, and testis. Most artificial sweeteners bind to taste receptors with greater affinity than sucrose (table sugar).
Consequences of NNS consumption in animals and humans
First thought to be harmless, an increasing number of research studies now advise moderation of NNS consumption in young adults and patients with metabolic disorders. Once bound to the sweet receptor, NNSs can trigger insulin secretion in humans. In human and mouse models, NNSs alter secretion of the gut hormone incretin. Incretin drives glucose absorption in the gut. Therefore, it is thought that when coingested orally with glucose, NNSs increase the amount of glucose absorbed. NNSs may also upregulate proinflammatory and adipogenesis-promoting pathways. , More specifically, saccharin and acesulfame K stimulate adipogenesis of precursor cells through the AKT (protein kinase B) signaling independently of sweet taste receptor presence, whereas sucralose increases proinflammatory gene expression in the livers of mice, specifically matrix metalloprotease and inducible nitric oxide synthase. Finally, NNSs have been linked to changes in the microbiome, generally decreasing the level of healthy and beneficial bacteria. In mice, saccharin triggers glucose intolerance because of microbiome alteration at ADI level. A bacteriostatic effect has also been suggested on oral microbes by saccharin, aspartame, and sucralose.
In human studies, NNS substitution has shown to moderately lower body mass index (BMI) in a subset of children with overweight. However, in children with normal weight, increased NNS consumption led to higher sugar intake and paradoxically has been correlated with an increased BMI. In adults, NNS consumption has been correlated with elevated risk of metabolic syndrome.
NNS absorption, distribution, and excretion during pregnancy and lactation
Absorption of NNSs into the bloodstream occurs in the small intestine and can be transferred to a fetus through the placenta and to an infant through breast milk. Although some NNSs are fully degraded (aspartame), most of them (sucralose, acesulfame K, saccharin) circulate in the body unmetabolized and are found in the blood, urine, and feces ( Figure 2 ).
Blood, urine, and feces excretion
Sucralose (20%–30%) is absorbed in the blood circulation and can be found in the urine up to 5 days after ingestion. Although 2 small unidentified metabolites have been observed from sucralose degradation, about 80%–90% is found unmetabolized in the urine and feces. Acesulfame K is not metabolized or stored; instead 99% of it is excreted unchanged through the urine. , Saccharin is not metabolized by animals or humans; 85%–95% is absorbed and eliminated through the urine, and the rest is excreted through feces. Aspartame undergoes full digestion in the gastrointestinal tract by esterases and peptidases providing secondary metabolites (methanol, aspartic acid, and phenylalanine) subsequently absorbed in the bloodstream. Although harmless to most of the adult population, people with a rare hereditary disease known as phenylketonuria will have a difficult time metabolizing the aspartame breakdown product phenylalanine and should, therefore, avoid aspartame consumption. Steviosides are not metabolized by human enzymes , but rather undergo metabolism by gut bacteria. , These bacteria provide an aglycone metabolite that can then enter the blood circulation.
In utero exposure
Using an animal model, our group revealed that sucralose can be found in the urine of newborns, suggesting in utero transmission likely through cord blood. Acesulfame K was found in amniotic fluids and fetuses’ urine even when high concentrations were found in the placenta, suggesting NNS filtering. , Saccharin was found in amniotic fluid and fetal bladder as well as maternal blood in similar amounts. Aspartame was not found to pass through the placenta because it is fully digested in the gastrointestinal tract. To date, we do not know if steviosides cross the placental barrier.
Breast milk
Sucralose, acesulfame K, and saccharin can transfer through breast milk. , These 3 NNSs start accumulating in breast milk 2–3 hours after ingestion of 1 diet soda. , Aspartame is not found in breast milk. To date, there are no data on the excretion of steviosides through breast milk.
NNS consumption during pregnancy and mothers’ health
About 92% of women are aware that consuming too much sugar leads to an increased risk for diabetes mellitus and cardiovascular diseases. PubMed search conducted by the authors (A.P. and S.O.V.S.) found 9 relevant human studies that investigated NNS consumption during pregnancy ( Figure 3 ). Evidence indicates that pregnant women consume as much or even more NNSs than the general population. On the basis of these studies, the median of self-assessed NNS exposure in pregnancy is around 22.4% weekly and 13.1% daily ( Figure 3 ). One study reported an increase in NNS consumption during pregnancy compared with preconception period. The rise in NNS consumption during pregnancy over the past decade corresponds with trends in the general population. With NNS use becoming more prevalent in beverages, foods, supplements, and medications, women may not be aware of their actual NNS consumption. Other women may make a conscious decision to consume NNSs in an attempt to reduce gestational weight gain or while following medical nutrition therapy for pregestational or gestational diabetes mellitus.
Safety of perinatal NNS exposure
The Academy of Nutrition and Dietetics states that NNS consumption is safe during pregnancy and childhood, whereas the US Institute of Medicine and the American College of Obstetricians and Gynecologists have not made any statement on NNS consumption during pregnancy. Two systemic reviews on this topic present inconsistent evidence on long-term metabolic effects of NNS exposure during gestation and infancy. , The first review from 2016 included studies from early-life exposure, but it did not include in utero or breast milk NNS exposure because no human studies on this topic were available. The second review published in 2018 included 2 studies of women consuming artificially sweetened beverages (ASBs) during pregnancy and lactation, which found that ASB exposure was associated with a higher risk of overweight or obesity compared with no ASB exposure.
Well-designed human studies evaluating NNS metabolic effects during pregnancies are needed to inform guidelines for pregnant women regarding NNS consumption. Similar to the FDA approval process, potential consequences and risks of in utero NNS exposure are often investigated using animal models. Below, we summarize findings derived from animal studies on the safety of NNS ingestion throughout pregnancy and lactation for mothers and infants.
Maternal effects of NNS exposure
Data on NNS consumption during pregnancy and maternal health outcomes are largely absent. Animal studies exploring this topic have focused on the health of offspring, and no study to date has investigated maternal glycemic outcomes, including glucose and insulin levels or glucose tolerance during pregnancy. In a few studies, both saccharin and sucralose decreased body weight and weight gain in pregnancy and lactation, whereas the majority showed no effects from NNS exposure. Although observational studies in adult human populations suggest an association between NNS consumption and development of metabolic diseases, , no firm conclusion has been drawn regarding maternal exposure to NNS and long-term health risks to the mother, leaving this area of research unexplored.
Offspring effects associated with in utero NNS exposure
Birthweight and weight gain
On the basis of findings in rodent models, litter size was not affected by in utero NNS exposure. However, continuous NNS exposure during the lactation period demonstrated a significant decrease in body weight at weaning (but not at birth) in male and female offspring. , , , , Offspring’s body weight at adulthood was also decreased when exposed to NNS during pregnancy and or lactation. , Interestingly, this was true for perinatal exposure to saccharin, sucralose, acesulfame K, aspartame, and even steviosides. , The mechanistic explanation of this finding is likely related to the activation of the sweet receptor as this is the only common feature of all NNS. Finally, few studies examined weight gain from birth to adulthood following in utero NNS exposure; results were highly inconsistent. , It is important to note that these effects are likely independent of the sweet receptor, considering most rodent models do not taste aspartame as sweet.
Glucose metabolism
Blood glucose was increased in rodent offspring of aspartame-fed mothers. , , In offspring of mothers fed with combined mixture of acesulfame K and sucralose, blood glucose was decreased at weaning.
Sweet tooth
Prenatal acesulfame K exposures have been associated with alteration of offspring sweet taste preference , at high concentration but not at ADI level.
Liver health
Liver detoxification was found to be less efficient in offspring of NNS-fed mothers leading to liver whitening associated with detoxification defect.
Gut microbiome
Microbiome profiling confirmed a significant increase in 1 of 2 major gut microbial phyla, Firmicutes, and a significant decrease of Akkermansia muciniphila in rodent offspring of NNS-fed mothers at ADI level. Similar microbiome alterations in humans have been linked to an increased rate of metabolic disease and obesity. Firmicutes increases have been linked to obesity in both mice and humans. Similarly, A muciniphila level is inversely correlated to fat mass gain, and supplementing these bacteria helps prevent fat mass gain on a high-fat diet. , Therefore, comparable offspring microbiome changes associated with in utero or lactational NNS exposure undoubtedly triggers metabolic deregulation across generations.
Mechanisms for the metabolic changes seen in the offspring of mice after perinatal NNS exposure are presumed to stem from NNS-induced fetal programming. These alterations in fetal programming can potentially activate the sweet taste receptors in the intestine, leading to increased glucose absorption, changes in the nascent microbiome, induction of oxidative stress, and compensatory energy intake.
Evidence from human studies
Although observational studies regarding NNS exposure are becoming more prevalent, data are still rare and complex to interpret. This is due to multiple factors associated with human cohorts including heterogeneity of samples and accuracy of self-reported NNS consumption. Two large cohort studies of Danish pregnant women observed a small increase in the risk of preterm delivery and childhood asthma after consumption of NNS-containing beverages during pregnancy. , Another cohort study in Canada found that maternal consumption of ASBs during pregnancy was significantly associated with infant BMI at 1 year of age. After controlling for maternal obesity and diet quality, daily ASB consumption was associated with a 0.2-unit increase in infant BMI z-scores and a 2-fold higher risk of overweight. Only 1 similar study was performed in the United States and showed no significant association between prenatal NNS exposure and infant BMI. Although NNS is found in many food products, abovementioned studies only looked at ASB consumption (eg, “diet” and “zero” soda) and did not assess all dietary NNS consumption occasions.
Conclusion
Emerging evidence from animal studies warns against NNS consumption and present specific effects that harm the metabolism in offspring. Although each NNS is different and can trigger different metabolic outcomes, researchers agree on advising moderation of NNS consumption, particularly for sensitive populations such as patients with diabetes mellitus, children, and pregnant women. Moderation is key!
As evidence accumulates and with about 30% of pregnant women self-reporting NNS consumption and possibly more with unintentional exposure, it is critical to question the safety of NNS consumption during pregnancy. Prospective observational cohorts and clinical studies that accurately analyze exposure and long-term health effects of NNS consumption during pregnancy in mothers and babies are needed to inform health organizations, dietary guidelines, and practitioners.
The authors report no conflict of interest.
Each author has indicated that he/she has met the journal’s requirements for authorship.