A review of non-nutritive sweetenersand sugar-sweetened beverages and theirimpact on body mass and health

Institutionen för Hälsovetenskap

Examensarbete

IV054G

Idrottsvetenskap GR (C), 15 hp, VT2020

A review of non-nutritive sweeteners and sugar-sweetened beverages and their

impact on body mass and health

Malin Esbjörnson

2020-05-07

Abstract

Consumption of sugar-sweetened beverages have increased and claims to have a strong association to being overweight and obesity; hence, a reduction could result in better health outcomes. A substitute to sugar-sweetened beverages are non-nutritive sweeteners, that provides sweetness without adding extra calories.

This research area is much disputed and need clarification; thus, this review describes knowledge about sugar-sweetened beverages and non-nutritive sweeteners, and its effect on body weight and general health, the last decade. A

total of 16 randomized controlled trials and cohort studies, published 2010 or later, were reviewed. The studies included all ages and normal weight to being overweight and obese, to provide a broad view. The results indicated that more research is needed on individual non-nutritive sweeteners; although, evidence supports that non-nutritive sweeteners had none or little negative effect on body

weight and health effects, when been compared to sugar-sweetened beverages, water and placebo. Sugar-sweetened beverages were claimed to have a negative impact on both health and body weight. It was concluded that there was a benefit

of replacing sugar sweetened beverages to non-nutritive sweetened, when it comes to both general health effects, as well as weight reduction and

maintenance.

Keywords: Artificial, drinks, soda, sucrose, sweetening, weight Abstrakt

Konsumtionen av sockersötade drycker har ökat och har påståtts ha en stark association till övervikt och obesitas; därav, månde en reducering resultera i bättre

hälsoresultat. Ett substitut till sockersötade drycker är icke-näringshaltiga sötningsmedel, som förser sötma utan att addera extra kalorier. Detta forskningsområde är mycket omdiskuterat och är i behov av klargörande; således, denna översikt beskrev kunskap om sockersötade drycker och icke-näringshaltiga sötningsmedel, och dess påverkan på kroppsvikt och generella hälsa, det senaste

årtiondet. Totalt granskades 16 randomiserade kontrollerade studier och kohortstudier, publicerade år 2010 eller senare. Studierna inkluderade alla åldrar,

normalviktiga till övervikt och obesitas, för att tillhandahålla en bred vy av forskningen. Sockersötade drycker har påståtts ha en negativ inverkan på både kroppsvikt och hälsan. Icke-näringshaltiga sötningsmedel hade ingen eller en liten

negativ effekt på både den generella hälsan och kroppsvikten, jämfört med socker-sötade drycker, vatten och placebo. Det har konkluderats att en kan med fördel ersätta sockersötade drycker till icke-näringshalt sötade, när det kommer till

både generella hälsoeffekter, samt viktminskning och att bibehålla vikten.

Nyckelord: Artificiell, drycker, läsk, sackaros, sötningsmedel, vikt

Innehållsförteckning

Introduction ... 1

Method ... 4

Discussion ... 7

Limitations and Strengths ... 7

NNS and Water ... 7

Aspartame ... 9

Acesulfame-K ... 12

Saccharin ... 14

Sucralose ... 15

SSB and NNS ... 19

Summary ... 20

Conclusion ... 21

References ... 22

1 Introduction

Obesity and being overweight has increased worldwide and has doubled between year 1980 to 2015 and has showed a continuous increase (GBD 2015 Obesity Collaborators 2017). Obesity and being overweight, i.e. high BMI is associated with several chronic diseases such as diabetes mellitus (Singh, et al. 2013), cardiovascular diseases (Singh, et al. 2013) and different cancers (Lauby-Secretan, et al. 2016). Metabolic conditions such as obesity and being overweight are complex and can therefore not be attributable by one single food, such as sugar. It is caused by an excess of energy compared to what your body requires, hence, it is not only caused by consuming sugar. There has been a large increase of overweight since 1978, the consumption of soda has increased with 40% and in the US, there has been recognized that an increase in energy consumption were made through beverages (Hill 1998). In the late 1970s to 2001 there has been an increase of consumption of sugar sweetened beverages (SSBs; beverages with added sugar or other sweeteners, such as high fructose corn syrup, sucrose and fruit juice concentrate) from 3.9% in 1977 to 9.2% in 2001 (Nielsen & Popkin 2004). The national cross-sectional study by Nielsen and Popkin (2004) also found that the consumption of SSBs have increased with 135% for children aged ≥2–9 years. In Sweden there has been a high increase of soda consumption the past decades and it is presented in Figure 1. Basu, et al. (2013) found in their cross-national study of 75 countries that an increase of SSBs with 1% is associated with 4.8 % increase with cases of being overweight and 2.3% of obesity. Furthermore Basu, et al. (2013) claims that the consumption of SSBs have a significant link to being overweight, diabetes and obesity, worldwide. The World Health Organization (WHO) recommend that an individual’s total energy intake should contain <10% of free sugars, but ideally it should be <5% (WHO, 2018). The description of free sugars is “all sugars added to foods or drinks by the manufacturer, cook or consumer, as well as sugars naturally present in honey, syrups, fruit, juices and fruit juice concentrates” (WHO, 2018). Despite these recommendations, the excessive intake of dietary sugars is high and has led to a pandemic of obesity and in Americans’

diets, where SSBs are the primary source of added sugars (Johnson, et al. 2009).

There is a strong correlation between SSBs and obesity, thus a reduction of SSBs would result in a decrease in obesity (Hu 2013).

Figure 1. Soda consumption and mineral water consumption from 1960-2013.

The Y-axis present liter per person and year for soda (blue) and mineral water (red) (Swedish Board of Agriculture 2015).

A substitute to SSBs are non-sugar sweeteners, that can be divided in two categories: artificial sweeteners and natural non-caloric sweeteners (Figure 2.).

Non-sugar sweeteners contain fewer or no calories, compared to sucrose (Toews, et al. 2019). The artificial sweeteners approved for use in the United States are, among others: aspartame, acesulfame-K, saccharin and sucralose, and natural non- caloric sweetener are stevia. For example; aspartame gives 4 kcal/gram, but it´s 180-200 times sweeter than sucrose (Chattopadhyay, Raychaudhuri, &

Chakraborty 2014); thus, aspartame provides a small amount of calories per grams,

but since its´ consumption is used in extremely small doses the calories are a

negligible. Although, aspartame is categorized as a low-calorie artificial sweetener

(EFSA), since it provides 4 kcal/gram, compared to other approved artificial

sweeteners that provides 0 kcal/gram (Kroger, Meister & Kava 2006) this overview

will refer all artificial sweeteners, as non-nutritive sweeteners (NNSs).

3

Figure 2. Types of sweeteners with categories and subcategories (Toews, et al. 2019).

NNSs can be used in beverages, and provides sweetness to the beverage, without adding extra calories. Hence, it can be used among individuals whom are attempting to lose weight, since a reduction of SSBs consumption is one common strategy to weight loss (Phelan et al. 2009). Flood, et al. (2006) examined how different beverages, both in size and energy, affected the total energy intake. The study compared cola, diet cola and water, in different sizes, along with the same amount of food. The results showed that by increasing the size of the beverage it significantly increased the amount that were consumed regardless of the type of the beverage. The consequence of this led to a significant increase in the total energy intake for the caloric beverage, compared with the non-nutritive sweetened beverages (NNSBs), since the food intake did not differ between the groups. Thus, the study by Flood, et al. (2006) suggests that when consuming SSBs one does not compensate with food intake; hence, it will lead to an overall energy expenditure.

An overview by Mattes & Popkins (2008) it is suggested that NNSs are associated

with increased hunger. NNSBs are often being consumed with foods and might

therefore lead to an overconsumption of foods. Furthermore, if heightened hunger

leads to increased energy intake is unclear. The evidence suggests that when NNSs

is used in free-living individuals, the weight loss or weight maintenance are

inadequate (Mattes & Popkins 2008). Although the research suggests that there is

a strong correlation between SSBs and being overweight (Hu 2013) there are some

concerns by the public population. Smith, et al. (2019) found that parents preferred

SSBs for their children (76%) and were concerned about NNSs and its´ potential

health effects (78%). The parents in this study felt safer to offer their child SSBs

than NNSBs and were more skeptical to artificial NNSs than the natural ones.

4 Mattes & Popkins (2008) claims that it is uncertain whether NNSBs stimulates eating, hence; leads to an excess of energy. These subjects are much-disputed and need clarification. This review describes a decade of research on SSBs and artificial NNSs (beverages and foods) effect on body weight and general health. The research questions that needs to be answered is: How does NNS affect the general health and bodyweight compared to no consumption? How does SSBs affect the general health and body weight compared to NNSBs?

Method

The searching was made by using the database Pubmed, by using following search terms: sugar sweetened beverages, non-nutritive sweeteners, artificial sweeteners, aspartame, saccharin, acesulfame-potassium/K, sucralose, body weight and health effects. The searching was made in March-April in 2020.

The inclusion criteria were randomized controlled trials (RCT), cohort studies, all ages, healthy and non-healthy, normal weight and being overweight to obese, as well as short-term and long-term interventions (i.e., one day to over a year). The exclusion criteria were studies published before 2010, pilot studies, reviews, meta- analyses, epidemiological studies and animal studies.

The original search of key words gave search results of: sugar sweetened beverages

– 2 792; non-nutritive sweeteners – 464, artificial sweeteners – 228 160; aspartame

– 1 507; saccharin – 5 057; acesulfame-potassium – 256; acesulfame-K - 340,

sucralose – 762; acesulfame-K and aspartame – 164; sugar sweetened beverages

body weight – 1 125; sugar sweetened beverages health effects – 868; non-nutritive

sweeteners body weight – 165; non-nutritive sweeteners health effects – 138. The

exclusion of the studies were made by reading titles, abstract and full papers, where

the papers were excluded when they did not match the inclusion criteria, or when

they matched the exclusion criteria.

5 Results

A total of 16 studies matched the inclusion criteria and the result is presented in Table 1 with an overview of each included study.

Table 1.

An overview of the 16 included studies.

Population Intervention

Duration Subjects NNS Comparison Result Study

Campos, et al. (2015)

Age:

unspecified BMI: >25

kg/m²

12 wk 27 Unspecified Unspecified SSB

Significant differences:

IHCL

RCT

Crézé, et al.

(2018)

Age:

unspecified BMI: ~21-

22 kg/m²

3 d 18 Acesulfame

-K

Sucrose and Water

NS*:

appetite and biochemical

effects

RCT

DeBoer, Scharf &

Demmer (2013)

Age: 2–5 y

”Normal weight”

Longitudinal between age

2-5 y

9600

No NNS, Nondrinker

of SSB

SSB regular drinker

Significant differences:

BMI

Cohort

de Ruyter, et al. (2012)

Age: 5–12 y

”Normal weight”

18 m

477

Sucralose and Ascesulfam

e-K

Sucrose Significant differences:

Weight gain, BMI and fat

mass

RCT

Fantino, et al. (2018)

Age: 18-45 y BMI: 19-28

kg/m²

9 wk 166 Unspecified Water

NS*: EI, satiety and consumption

of different foods

RCT

Grotz, et al.

(2017)

Age:

unspecified

”Normal weight”

12 wk 47 Sucralose Placebo

NS*:

metobonomi

cs RCT

Higgins, Considine &

Mattes (2018)

Age: 18-60 y BMI: 27-35

kg/m²

12 wk 93 Aspartame Placebo

NS*:

appetite, body weight,

body composition

and metobonomi

cs

RCT

Higgins &

Mattes (2019)

Age: 18-60 y BMI:

18-25 kg/m²

12 wk 124

Mixed (here: focus

on saccharin)

Sucrose

Weight gain NS*

Significant differences:

Fat mass, portion size

and EI

RCT

6

*=Non-significant

Population Intervention Duration Subjects NNS Comparison Result Study

Lertrit, et al.

(2018) Age: > 18 y

”Healthy” 4 wk 15 Sucralose Placebo

NS*:

EI Si gnificant differences:

Insulin resistance and secretion

RCT

Madjd, et al.

(2016)

Age: 18-50 y BMI: 27-35

kg/m²

6 m 65 Unspecified Water

Significant differences:

weight loss RCT

Madjd, et al.

(2017)

Age: 18-50 y BMI: 27-35

kg/m²

12 m 65 Unspecified Water

Significant differences:

weight loss RCT

Maersk, et al. (2012a)

Age: 20-50 y BMI: 28-36

kg/m²

1 d 24 Aspartame

Sucrose and Water

NS*: NNS and Water Significant differences (SSB): EI, metabonomi cs and appetite

RCT

Maersk, et al. (2012b)

Age: 20-50 y BMI: 26-40

kg/m²

6 m 47 Aspartame Sucrose and

Water

NS*: NNS and Water Significant differences (SSB): liver fat, skeletal muscle fat,

blood triglycerides

and cholesterol

RCT

Peters, et al.

(2015)

Age: 21-65 y BMI: 27-40

kg/m²

12 m 222 Unspecified Water

Significant differences:

weight loss RCT

Romo- Romo, et al.

(2018)

Age: 18-55 y BMI: 18-25

kg/m²

2 wk 61 Sucralose No

intervention

Significant differences:

Insulin sensitivity and insulin secretion

RCT

Sathyapalan, et al. (2015)

Aspartame sensitive and non- sensitive

1 d 96 Aspartame

(cereal bar)

Cereal bar

NS*:

biochemical effects and metabonomi

cs

RCT

7 Discussion

Limitations and Strengths

Ideally, all different NNSs should have been compared to placebo or water, rather than being compared with SSB, that some were. More studies on individual NNSs should have been included, but few studies have been done on some of the NNSs in humans; hence, it was impossible. Larger sample sizes and only long-term interventions would have been ideal; however, to get a bigger overview of the evidence provided the last decade and to include more studies, this was not one of the inclusions criteria. A decade of research was included since research that was made decades ago might not be current, and if some major findings were made, it would occur in current research as well. Also, this interprets that the evidence provided a decade ago is similar to current evidence. To provide a broad overview, all ages and normal weighted to obese were included, with both short-term and long-term intervention. Epidemiological studies were excluded since it is difficult to control for confounders, also due to reversed causality, since most epidemiological studies are observational. Thus, it does not have the same evidencable value as an RCT; hence, RCT and cohort studies were included to provide strong an evidence-based view of the research.

NNS and Water

222 weight-stable (within 4.5 kg the past 6 months) subjects being overweight and obese participated, and completed, in a RCT, included of a weight loss program where the subjects consumed 710 ml of NNSBs (no additional adding of NNSs in beverages such as coffee or tea were allowed; however, NNS food consumption were acceptable, but not a must) or water daily for one year (Peters, et al. 2015).

The weight loss program was 12 weeks followed by 40 weeks of weight

maintenance. The participants were both male and female, aged 21 to 65 and with

a BMI of 27 to 40 kg/m

. The NNS group significantly differenced with the water

group, where NNS decreased more in weight than water. The subjects were

relatively homogeneous; however, a large age difference between the subjects,

which could have affected the results.

8 Madjd, et al. (2016) performed a similar RCT, where 65 women being overweight and obese, with type 2 diabetes, whom usually consumed NNS, performed a six months weight loss program when consuming either 250 ml NNSBs or 250 ml water with a meal, five times a week. Before the intervention, each and every participant performed a 2-week washout period from all NNSs (i.e., NNS foods and beverages). Since all subjects were women, every subject started the intervention at the same phase of their menstrual cycle, that is necessary when a long-term intervention is performed, to eliminate possible effects that the menstrual cycle can have. There was a significant greater decrease in weight for water compared with NNSB, although both groups lost weight. One year after, Madjd, et al. (2017) published an 18-month follow-up RCT on the previous intervention. A 12-month follow-up period was made (after the end of the six months intervention), where a significant greater weight loss was observed in the water group compared with the NNS group, although both groups made a weight reduction; thus, the results were simliar to Madjd, et al. (2016). However, the results from Madjd, et al. (2016) and Madjd, et al. (2017) is not controlled for confounders, i.e., decreased EI or increased physical activity, that could have explained the significant effects that were found in both studies.

Another RCT was published by Fantino, et al. (2018) that provides results in between Peters, et al. (2016) and Madjd, et al. (2016), when 166 (86 men and 80 women) subjects, aged 18-45 years performed a 9-week intervention to compare water with a NNSB. The intervention consisted of two-day sessions at four different times (week 1, 2, 3 and 9). Day one was in a laboratory and day two was under free- living conditions, and food intake and appetite was measured in the weekly sessions. Water (330 ml) was consumed with each main meal (breakfast, lunch and dinner) during the first session. At the second (week 2) and third session (week 3), the subjects performed a randomized crossover with water and NNSB (330 ml);

half of the group consumed water at session 2, and NNSB at session 3, and the other

half of the group consumed NNSB at session 2 and water at session 3. Then, the

subjects were randomized again into two groups, consumption of NNSB (660 ml)

or water (660 ml; no NNSBs or foods were allowed) during week 4-8. At week 9,

the fourth session was done, where the group whom consumed NNSB during week

4-8 was tested with NNSBs and the group whom consumed water during week 4-8

9 was tested with water. There was no significant difference between NNSBs and water for neither, total energy intake (EI), satiety, nor consumption of different foods; hence the evidence from Fantino, et al. (2018) suggests that NNSBs has similar effect as water.

Although Madjd, et al. (2016) found a significant decrease in weight for NNSBs compared with water, Madjd, et al. (2017) found a significant weight reduction with water compared with NNSBs, but a weight reduction was made from both groups.

Fantino, et al. (2018) controlled for more confounders, and in both free-living and controlled environments, through a long-term intervention and a large sample size, which is ideal. It might not be claimed that NNSBs is better than water; however, it suggests that NNSBs has similar effects on weight and health effects as water.

Aspartame

An RCT was published in 2018 to investigate aspartame during a 12-week period

(Higgins, Considine & Mattes 2018). A total of 100 healthy, lean, subjects, men

and women, were randomized in three different groups, where 93 completed the

study – 31 in each group: no aspartame (2 capsules containing 680 mg dextrose and

80 mg PABA in total and 2 empty capsules), 350 mg aspartame/day (500 ml

beverage with 350 mg aspartame and 80 mg PABA, 2 capsules containing 680 mg

dextrose in total and 2 empty capsules) and 1050 mg aspartame/day (500 ml

beverage with 350 mg aspartame and 80 mg PABA, 4 capsules containing 700 mg

aspartame in total and 2 empty capsules). The subjects were aged 18-60 years, and

to investigate a more homogenous group, a smaller age difference would be

preferable, or separation of ages, to minimize any error sources from the subjects

and to distinguish the possible effect that age might have. A baseline test was made

where the participants reported appetitive sensation over all waking hours in a 24-

hours period on an hourly basis, also at week 4, 8 and 12. Measurements from a

blood sample were made and measurement of body composition at baseline and at

the end of week 12. The intervention period lasted for 12 weeks with a daily

consumption of assigned aspartame-sweetened beverage and capsules. No

significant difference between any of the groups for neither hunger, desire to eat,

fullness nor thirst. Insulin, GlP-1 and GIP responses were not significant different

10 for both doses of aspartame compared with control, neither were body composition nor fasting leptin. Hence, the evidence from this study claims that aspartame does not affect appetite or body weight negatively.

An acute effect on psychological symptoms, biochemical and metabonomic effects of aspartame were measured on 48 subjects whom self-reported sensitivity to aspartame (subjects with allergy were excluded) and compared with 48 matched non-sensitive subjects in a double blinded crossover RCT by Sathyapalan, et al.

(2015). Subjects were randomized to aspartame cereal bar (100 mg aspartame, equivalent to a can of diet soda) or control bar, with a cross-over with at least one week apart. Aspartame sensitive subjects reported more symptoms in the first session; however, this was reported for both the aspartame cereal bar and the control bar. None of the symptoms between neither NNS and control, nor between sensitive and non-sensitive, and none of the biochemical and metabonomic effects differenced in neither NNS bars nor control bars, also no difference in aspartame sensitive and non-sensitive were observed between the two sessions. More symptoms reported by the aspartame sensitive group at the first session cannot be attributed to aspartame, since it was reported for both bars. Also, self-reported data is always subjective; thus, whether the sensitive subjects were nocebo or control is unclear, as discussed by Sathyapalan, et al. (2015). Psychological data analysed symptoms separately, which was necessary; however, this increase the risk of Type 1 error and another measurement of this would have been preferable, to reduce the risk. Also, Higgins, Considine and Mattes (2018), a more recently published study, found a non-significant effect on neither of the appetite sensations on aspartame.

However, the results from Sathyapalan, et al. (2015) still suggest, and is stated in their conclusion that aspartame has no significant effect on biochemical and metabonomic outcomes, and no significant psychological effect was observed between the two sessions.

When performing a randomized controlled crossover trial (Maersk, et al. 2012a) on

24 overweight and obese, but apart from that, healthy subjects to compare SSBs

with isocaloric semi-skimmed milk and an aspartame-sweetened beverage the

results were likewise to these other studies. The subjects consumed either a 500 ml

SSB (900 kJ), semi-skimmed milk (950 kJ), aspartame-sweetened beverage (7.5

11 kJ) or water (control) with an Ad Libitum EI. The measurements were total EI, insulin, glucose, appetite, ghrelin, GLP-1 and GIP concentrations. No measurements of leptin were made, which would have been an interesting aspect of this, when measuring ghrelin. SSB led to significant increased subjective fullness and decreased hunger, also significant higher GLP-1 and GPI concentration, and ghrelin was significant lower for SSB compared to water. Insulin was significant higher for SSB, compared to water and the aspartame-sweetened beverage, and insulin was also significantly increased for SSB compared to all the others. The aspartame-sweetened beverage showed no statistically significant difference compared to water in none of the measurements. The Ad Libitum EI was not compensated at all for SSB, whereas the aspartame-sweetened beverage compared to water was not significant.

The same beverages (i.e., SSB, isocaloric semi-skimmed milk, NNSB (aspartame)

and water) were investigated in a 6-month on 47 subjects that were healthy,

nondiabetic (BMI 26-40 kg/m

), with normal blood pressure (<160/100 mm Hg)

and aged 20-50 years (Maersk, et al. 2012b). 1 L of the beverage, that the subjects

had been randomized to, was consumed each day, and they were allowed to drink

their usual amount of alcohol, to resemble free living. All lifestyle changes of diet

and physical activity were monitored at baseline, half-way and after the 6-month

intervention, using dietary record and questionnaire for physical activity. Every 1.5-

month anthropometric measurements were made and the SSB group were examined

by a dentist (no problems were developed during the intervention). A standardized

meal was consumed by each and every subject the night before each test and

performed overnight fasting. The SSB group had a significant higher liver fat,

skeletal muscle fat, blood triglycerides and total cholesterol, whereas milk and

NNSB significantly reduced the systolic blood pressure (10-15%) compared with

SSB, and NNSB showed similar effects to water. There was not a statistically

significant difference in body weight, however a daily intake of SSB is likely to

increase the risk of both cardiovascular and metabolic diseases. The non-significant

difference in body weight was measured in this 6-month intervention, can be

explained by the small sample size that affect the effect size negatively and multiple

tests; hence, the risk of statistical error is high. However, this study interprets free-

living individuals and supports the evidence of the benefits of NNSBs.

12 Other studies have found association between SSB and weight gain and in meta- analyses (Miller & Perez 2014; Malik, et al. 2013), where Malik, et al. (2013) included both cohort studies and meta analyses with 28 517 children and 174 544 adults, in total. It was concluded that consumption of SSB promotes weight gain in both children and adult. Thus, the result of no significant difference in weight for SSB group from Maersk, et al. (2012b) could be due to confounders.

Acesulfame-K

Crézé, et al. (2018) published a double-blinded crossover RCT, where the aim was to study the impact of SSB and NNSB (in this study: aspartame and acesulfame-K) beverage on food intake and brain responses to food. The study included 18 healthy males who consumed water, SSB and NNSB for three days, followed by a 3-week wash-out period, and a crossover. A three-day intervention period is not enough to predict the long-term effects, only acute effects. A five-day period of nutritional and lifestyle recommendations was performed, followed by participants receiving a controlled diet for weight-maintenance two days pre test day, where they ate at specific times of the day and all participants were asked to maintain the same physical activity (factor set at 1.5). The participants ate Ad Libitum at test days.

However, to resemble a natural habitat of free-living individuals, these recommendations could have been excluded, when the intervention period lasted for only three days. Although this was a randomized crossover; thus, all participants went through the same procedure, it would have been more interesting to perform this research without the recommendations. This to resemble their usual living.

Although a controlled environment is necessary to eliminate confounders, an overly controlled environment is not transferable. The main outcomes of the study were brain responses to food (i.e., hunger, thirst, satiety and taste cravings; rated on a scale) and physiological factors: plasma concentrations of ghrelin, glucose and insulin. The results showed a significant effect (P<0.05) on hunger, where hunger was lower in sucrose compared to water and NNSB. However, no statistical significance was found for neither thirst, satiety nor taste cravings for none of the beverages, and NNSB did not alter consumption of more food compared to water.

The plasma glucose concentration showed no significant effect on none of the three

13 beverages and the plasma insulin concentration was higher in sucrose than water and lowest for NNSB. Furthermore, the plasma ghrelin concentration showed no significance between water and NNSB, but a lower concentration for sucrose.

Another study looking at the acute effects of different beverages is Maersk, et al.

(2012a), whom contradicts Crézé, et al. (2018) and is supported by long-term studies. Crézé, et al. (2018) found no significance between water and NNSBs, but a significant difference compared to sucrose, where the plasma ghrelin concentration was lower; however, they did not measure leptin, that could have been of interest. They observed changes in brain responses that affect food intake regulation and suggests that it might cause an adaptation that lead to a fade of food intake regulation, although NNSBs did not alter in a higher food intake. To conclude that NNSBs could lead to consequences in food intake, i.e., increased food consumption, it would require a long-term study, which has been done by Maersk, et al. (2012b), among others. It has been suggested that replacement of SSBs to NNSBs leads to a decrease in intrahepatocellular lipids (IHCL) (Campos, et al.

2015), although the sample size was small, and a larger effect was found on subjects with already high IHCL and visceral adipose tissue volume (VAT). More studies need to be done on larger sample sizes and on subjects with higher BMI. No significant effect on body weight, nor body composition was observed, which can be explained by the relatively short intervention period, although a recently published meta-analysis claims that NNS consumption has significant effects on weight loss compared to SSB (Laviada-Molina, et al. 2020). Evidence still points to a benefit of NNSs compared to SSBs.

Regardless the results of brain responses made by Crézé, et al. (2018), several studies have showed that long-term consumption of NNSs leads to neither weight gain nor increased food consumption (Maersk, et al. 2012b; Higgins, Considine &

Mattes 2018; Miller & Perez 2014; Peter, et al. 2015; Laviada-Molina, et al. 2020), that NNSs have similar effects to water (Maersk, et al. 2012b, ) and that SSBs promotes weight gain and does not compensate for EI, including meta-analyses, long-term and short-term studies, (Maersk, et al. 2012a; Raben, et al. 2002;

Ebbeling 2014; Anton, et al. 2002; Malik, Willet & Hu 2013), although ghrelin

levels were 20% lower for SSB compared with water and NNS (Maersk, et al.

14 2012a). The similar effects of NNSBs (aspartame and acesulfame-K) and no consumption is also supported by Solomi, Rees and Redfern (2012), where a comparison between glucose, glucose + NNSB and SSB was made, and no statistically significant difference were observed in glycaemic response.

Saccharin

Body weight, body composition, EI, energy expenditure (EE), appetite and glycemia was measured on 124 subjects, who completed the study, (low consumers of NNSs; healthy; however, BMI 25-40 kg/m

) in an RCT over a 12-week period, published by Higgins and Mattes (2019). No significant difference between the subjects were observed. Baseline testing over five days, where EI, EE, body composition, body weight, glycemia and 24-hours appetite logs were measured and made. Subjects were randomly assigned to five groups: SSB (1.25-1.75 L/day, based on body weight) and NNSBs: saccharin, aspartame, rebA and sucralose. Both SSB and saccharin led to an increased body weight and was not significantly differed from each other, whereas the other NNSBs had no significant effect on body weight. The SSB contained 450-560 kcal/day and the NNSBs <5 kcal/day;

thus, it explains that SSB does not compensate for EI. However, total fat mass, portion size and EI (Figure 3.) was significantly increased for only SSB and not for any of the NNSBs, and no significant difference on EE of neither of the groups.

Thus, the significant increased body weight from saccharin, but no significant effect

on other factors, suggests that there is deficiency in the measurement, as a

suggestion, EE, or the self-reported EI. Saccharin group rated a significantly greater

hunger; however, no other appetitive sensations were significantly different and no

significantly difference was observed for glycemic response for neither of the

groups after the 12-week intervention. Thus, more studies are needed in humans to

provide sufficient evidence of saccharin and to replicate these results to claim that

saccharin does have a negative effect on body weight.

15

Figure 3. Change in energy intake over 12-week intervention for all sweeteners (Higgins & Mattes

Sucralose

Romo-Romo, et al. (2018) evaluated the effects on glucose metabolism from sucralose consumption, due to the questioning regarding metabolic effect of NNS.

An RCT with 61 healthy subjects, men and women, performed a 14-day intervention using sucralose sachets, where they consume 45% of acceptable daily intake (ADI) for sucralose (5 mg/kg/day) and the control group did no intervention.

The sucralose sachets (1 g) contained 958 mg dextrose, 30 mg maltodextrin and 12 mg sucralose and was consumed three times a day by adding them to beverages.

This provides 36 mg sucralose per day; thus, this amount is approximately 9% of ADI and should not provide any significant effect. However, the mean weight for all subjects were 58 kg, which would result in 130 mg of sucralose to consume 45%

of ADI. However, in the result the mean consumption of sucralose was 158 mg/day for men and 123 mg/day for women, although it was clear in the method that no other consumption of NNSs should occur. The actual consumption of sucralose is therefore unclear. Both groups were assigned to maintain the same physical activity and food intake and the variables were controlled through food record and a physical activity questionnaire. The subjects were assigned to not consume any other NNS products during the intervention and three-day food record was made;

however, this is self-reported data and does not exclude the possibility that some

16 foods were not reported. An intravenous glucose test was performed for both groups before and after the intervention period, with glucose variables: insulin sensitivity, acute insulin response to glucose, disposition index and glucose effectiveness. The results showed a significant decrease (P=0.04) in insulin sensitivity. The acute response to glucose were not significant (P=0.37), neither were disposition index (P=0.44) and glucose effectiveness (P=0.70). However, when performing further analysis with a subgroup of the participants, the acute insulin response was significant increased (P=0.04) and the insulin sensitivity was still significant decreased (P=0.04). Change in BMI, weight, physical activity, EI and consumption of sugars were not significant; however, it is not presented in the result and only written in text without a significant value and should have been presented as baseline testing, to provide a more accurate result.

However, these results are supported by Lertrit, et al. (2018) who performed a crossover, double-blinded RCT to determine the glycemic responses, insulin secretion, insulin sensitivity and GLP-1 release, on sucralose. 15 healthy subjects, both men and women, performed a 4-week intervention with 200 mg of sucralose a day, or placebo. After, there was a wash-out period for 1 week before the crossover. The potential effect of the cephalic response to sweet taste were eliminated by capsules to mask the sweet taste and dissolved in the stomach after 5-10 min. All subjects were advice to maintain the same physical activity and diet as normal. The results in this study showed a significant lower (P<0.005) Matsuda- index and insulin resistance (P<0.001), also a significant higher (P<0.001) homeostatic model assessment of insulin resistance (HOMA-IR) (sucralose: 1.85 ± 1.16, placebo: 1.48 ± 0.62) for sucralose compared to placebo, after 4 weeks. Furthermore, the insulin secretion was significant higher (P<0.001) for sucralose compared to placebo, after 4 weeks and active GLP-1 was significant higher (P<0.001) for sucralose than placebo after exposure. Thus, the results by Lertrit, et al. (2018) is supported by Romo-Romo, et al. (2018). However, no significant increase in EI between the groups were observed, but it was measured through a 24-hours dietary recall; thus, it was self-estimated, and a small change in the diet could have occurred without their knowledge and explain the results.

Lertrit, et al. (2018) claims that “Long-term consumption of sucralose can develop

insulin resistance” but none of the subjects showed criteria for insulin resistance,

17 i.e., HOMA-IR >3.60 and BMI >27.5 kg/m

(Stern, et al., 2005). The study had an intervention of 4 weeks and can therefore not claim that sucralose has that sort of metabolic effect, since all subjects had lower HOMA-IR and BMI than the criteria for insulin resistance.

Both Romo-Romo, et al. (2018) and Lertrit, et al. (2018) results showed an increased acute insulin response, i.e., increased insulin secretion and a decrease in insulin sensitivity, this does not result in a negative health effect since the reduction of insulin sensitivity is compensated by the increased insulin secretion. Although a significant effect was found, it does not per se alter in negative health effects, since the effects were small and could be an insignificant effect in a prospective perspective. Both of these studies included women, but none of them controlled for the menstrual cycle, nor hormonal contraception and the potential effects that is related to hormonal changes on neither blood glucose, nor insulin measures. Since both of the studies were ≥ 2 weeks, the women in the studies had gone through different phases in the menstrual cycle, at different times, also the possibility that some used hormonal contraception and Barrata, et al. (2013) found a differ in glucose control to the follicular and luteal phases. Furthermore, a study published the year after (Higgins & Mattes 2019) found a non-significant effect on insulin for sucralose in a 12-week intervention and the largest weight reduction was observed for sucralose, compared to other NNSs and SSB. Thus, sucralose still has more beneficial effects on health and body weight than SSBs.

Grotz, et al. (2017) refutes both Romo-Romo, et al (2018) and Lertrit, et al. (2018),

with their RCT where they investigated the effect from sucralose on glucose

homeostasis. A 12-week RCT with 47 healthy subjects, males only, to exclude the

potential effects from the menstrual cycle. All subjects were asked to maintain the

same physical activity and diet as normal. This study was performed in three

phases: 4-week screening, 12-week test and 4-week follow-up, where the main

outcomes where glucose, insulin, C-peptide and HbA1c. The intervention consisted

a consumption of 333 mg sucralose to take with meals or placebo (cellulose)

3x/day. The capsules were identical in size, weight and color and no other sucralose

products were consumed, since it was not sold in the country where the study was

performed. The results showed no statistically significant difference between the

18 groups, sucralose and placebo, in baseline for neither fasting glucose, insulin, C- peptide nor HbA1c. Furthermore, no statistically significant difference in change from baseline between the groups for none of the main outcomes (Figure 4.), and there was no statistically significant of the between-group differences. Grotz and Munro (2011) claims in their overview of sucralose that it has no effect on weight, however it is still effective when used for weight loss, since it reduces the calorie intake when being replaced from sugars. Furthermore, Grotz and Munro (2011) claims that sucralose can be safely used and is a complement to help manage calorie reduction for weight loss. The results from Grotz, et al. (2017) and the overview from Grots and Munro (2011) interprets that there is no long-term metabolic effect from sucralose on neither health effect nor body weight, however it does work for weight reduction, when being replaced from sugars.

Figure 4. Change in HbA1c, Fasting Glucose, Fasting Insulin and C-peptide from baseline and at

19 SSB and NNS

A double-blinded RCT was published by de Ruyter, et al. (2012) with an intervention period of 18 months, to study the effect of SSB and NNSB on body weight in 477 children (aged 4 years 10 month to 11 years 11 months). The children received either one 250 ml can of SSB (25 g of sugar) or NNSB (0 g of sugar) a day, where the drink where equal visually and in sweetness. The children consumed the drinks in school, where the consumers of SSBs did not change their caloric intake from beverages, but the NNSB consumers replaced their SSB to the NNS;

hence, they consumed 100 less calories from beverages. The NNSB group increased significantly less than the SSB group and when it was adjusted for height change, the BMI increased significantly less and less fat mass where measured for the NNSB group than SSB. Thus, these results provide evidence that there is no compensation for SSBs and no significant increase EI for NNSBs. This study was also performed on a large quantity of subjects. These results are similar to the cohort study (DeBoer, Scharf & Demmer 2013) made on 9600 children aged 2-5 years, where interviews with a parent was performed at 2-, 4-, and 5-years of age of their child, with questions about amount and frequency of SSB consumption for their child. Although the answers are subjective, the quantity in this study is large and provides more accurate mean values and a smaller margin of error. The results from this cohort study where that consumption of SSBs were associated with higher BMI. SSB has also been suggested to have an association to less satiety and also an increased EI, compared with solid carbohydrates that compensated the EI (Pan &

Hu 2011; DiMeglio & Mattes 2000).

Campos, et al. (2015) performed an RCT on 27 male and female subjects to evaluate if a replacement of SSB to NNSBs would reduce markers of non-alcoholic fatty liver disease in subjects with a BMI of 25-30 kg/m

. Firstly, this is a small sample size; hence, it may limit the statistical power. Secondly, it may have been of more importance to individuals with a higher BMI, to better the confirmation of the beneficial effects. A 4-week period was performed, where the subjects consumed their usual amount of SSB. Food intake was recorded over two days at the end of the fourth week, through an actimeter and a nutritionist calculated their food intake.

IHCL and VAT was measured, as well as body composition and fasting metabolic

markers. The subjects were thereafter randomly assigned into two groups: usual

20 SSB consumption (N=14) and NNSB (N=13), intervention period lasted for 12 weeks. Food intake and consumption of NNS were Ad Libitum for both groups. By replacing SSB with NNSB, IHCL were significantly decreased after a 12-week period, where the most important effect was found in subjects with already high IHCL and VAT. Thus, subjects with higher BMI would have been preferable and of more interest. No significant differences were observed with fasting metabolic markers, body weight, body composition and VAT. The study suggests that by replacing SSB with NNSB may have positive effects on liver metabolic health;

however, more studies with larger sample size is necessary and including individuals with already high IHCL and VAT, to confirm the most beneficial effects.

Summary

Out of all NNS studied in this overview, saccharin is the one with a potential

negative effect on body weight (Higgins & Mattes 2019); however, no significant

negative effect on health was observed and only one study was found, that matched

the inclusion and exclusion criteria of this overview. Thus, more studies need to be

done on humans to determine its effect. When founding an association between

NNSBs and type 2 diabetes it has been explained by BMI, dieting, health status,

high blood pressure, high triglycerides at baseline and pre-enrollment weight

change (de Koning, et al. 2011). The negative outcomes found to be associated with

NNSBs can therefore be that individuals already being overweight or obese are

more likely to consume NNSBs to succeed with weight loss and already have

negative health effects that comes with being overweight or obese. Saccharin is not

commonly used in beverages and the other NNSs in this overview, commonly used

in both beverages and foods, are with no doubt an overall better substitute to

SSBs. NNSs can replace SSBs for individuals being overweight and obese to

reduce weight and lead to a better general health, and also for normal weighted to

maintain their weight and general health.

21 Conclusion

Solid carbohydrates are suggested to have a better effect on body weight, than SSB,

due to compensation for EI. Evidence supports that there is no difference in NNS

beverages nor foods; it is claimed to have no effect on body weight, nor health

effects, and is similar to water. More research needs to be done on individual NNSs

to claim the effect on body mass and general health, that comes from single

acceptable NNS. However, evidence support a benefit of replacing SSBs to NNSs,

with both health effects and body weight. Thus, a NNS substituted is a better choice

for weight reduction, weight maintenance and health effects, than SSB.

22 References

Anton, S. D., Martin, C. K., Han, H., Coulon, S., Cefalu, W. T., Geiselman, P., &

Williamson, D. A. (2010). Effects of stevia, aspartame, and sucrose on food intake, satiety, and postprandial glucose and insulin levels. Appetite, 55(1), 37–43.

Barata, D. S., Adan, L. F., Netto, E. M., & Ramalho, A. C. (2013). The effect of the menstrual cycle on glucose control in women with type 1 diabetes evaluated using a continuous glucose monitoring system. Diabetes care, 36(5), e70-e70.

Basu, S., McKee, M., Galea, G., & Stuckler, D. (2013). Relationship of soft drink consumption to global overweight, obesity, and diabetes: a cross-national analysis of 75 countries. American journal of public health, 103(11), 2071-2077.

Beverage Digest Company. Fact Book 1998. Bedford Hill, NY 1998 pp 54–55 Flood, J. E., Roe, L. S., & Rolls, B. J. (2006). The effect of increased beverage portion size on energy intake at a meal. Journal of the American Dietetic Association, 106(12), 1984-1990.

Campos, V., Despland, C., Brandejsky, V., Kreis, R., Schneiter, P., Chiolero, A., Boesch, C., & Tappy, L. (2015). Sugar‐and artificially sweetened beverages and intrahepatic fat: A randomized controlled trial. Obesity, 23(12), 2335–2339.

Chattopadhyay, S., Raychaudhuri, U., & Chakraborty, R. (2014). Artificial sweeteners–a review. Journal of food science and technology, 51(4), 611-621.

Crézé, C., Candal, L., Cros, J., Knebel, J. F., Seyssel, K., Stefanoni, N., Schneiter, P., Murray, M.M., Tappy, L., & Toepel, U. (2018). The impact of caloric and non-caloric sweeteners on food intake and brain responses to food: a randomized crossover controlled trial in healthy humans. Nutrients, 10(5), 615.

DiMeglio, D. P., & Mattes, R. D. (2000). Liquid versus solid carbohydrate:

effects on food intake and body weight. International journal of obesity, 24(6), 794–800.

DeBoer, M. D., Scharf, R. J., & Demmer, R. T. (2013). Sugar-sweetened beverages and weight gain in 2-to 5-year-old children. Pediatrics, 132(3), 413–

420.

De Koning, L., Malik, V. S., Rimm, E. B., Willett, W. C., & Hu, F. B. (2011).

Sugar-sweetened and artificially sweetened beverage consumption and risk of type 2 diabetes in men. The American journal of clinical nutrition, 93(6), 1321–

1327.

de Ruyter, J. C., Olthof, M. R., Seidell, J. C., & Katan, M. B. (2012). A trial of sugar-free or sugar-sweetened beverages and body weight in children. New England Journal of Medicine, 367(15), 1397–1406.

Ebbeling, C. B., Feldman, H. A., Chomitz, V. R., Antonelli, T. A., Gortmaker, S.

L., Osganian, S. K., & Ludwig, D. S. (2012). A randomized trial of sugar-

23 sweetened beverages and adolescent body weight. N Engl J Med, 367, 1407–

1416.

EFSA [Internet] http://www.efsa.europa.eu/en/topics/topic/aspartame Fantino, M., Fantino, A., Matray, M., & Mistretta, F. (2018). Beverages containing low energy sweeteners do not differ from water in their effects on appetite, energy intake and food choices in healthy, non-obese French adults. Appetite, 125, 557-565.

GBD 2015 Obesity Collaborators. (2017). Health effects of overweight and obesity in 195 countries over 25 years. New England Journal of Medicine, 377(1), 13-27.

Grotz, V. L., & Munro, I. C. (2009). An overview of the safety of sucralose. Regulatory toxicology and pharmacology, 55(1), 1–5.

Grotz, V. L., Pi-Sunyer, X., Porte Jr, D., Roberts, A., & Trout, J. R. (2017). A 12- week randomized clinical trial investigating the potential for sucralose to affect glucose homeostasis. Regulatory Toxicology and Pharmacology, 88, 22–33.

Higgins, K. A., Considine, R. V., & Mattes, R. D. (2018). Aspartame

consumption for 12 weeks does not affect glycemia, appetite, or body weight of healthy, lean adults in a randomized controlled trial. The Journal of

nutrition, 148(4), 650-657.

Higgins, K. A., & Mattes, R. D. (2019). A randomized controlled trial contrasting the effects of 4 low-calorie sweeteners and sucrose on body weight in adults with overweight or obesity. The American journal of clinical nutrition, 109(5), 1288- 1301.

Hu, F. B. (2013). Resolved: there is sufficient scientific evidence that decreasing sugar‐sweetened beverage consumption will reduce the prevalence of obesity and obesity‐related diseases. Obesity reviews, 14(8), 606-619.

Johnson, R. K., Appel, L. J., Brands, M., Howard, B. V., Lefevre, M., Lustig, R.

H., ... & Wylie-Rosett, J. (2009). Dietary sugars intake and cardiovascular health:

a scientific statement from the American Heart Association. Circulation, 120(11), 1011-1020.

Kroger, M., Meister, K., & Kava, R. (2006). Low‐calorie sweeteners and other sugar substitutes: a review of the safety issues. Comprehensive reviews in food science and food safety, 5(2), 35–47.

Lauby-Secretan, B., Scoccianti, C., Loomis, D., Grosse, Y., Bianchini, F., &

Straif, K. (2016). Body fatness and cancer—viewpoint of the IARC Working Group. New England Journal of Medicine, 375(8), 794-798.

Laviada‐Molina, H., Molina‐Segui, F., Pérez‐Gaxiola, G., Cuello‐García, C.,

Arjona‐Villicaña, R., Espinosa‐Marrón, A., & Martinez‐Portilla, R. J. (2020).

24 Effects of nonnutritive sweeteners on body weight and BMI in diverse clinical contexts: Systematic review and meta‐analysis. Obesity Reviews.

Lertrit, A., Srimachai, S., Saetung, S., Chanprasertyothin, S., Chailurkit, L. O., Areevut, C., Katekao, P., Ongphiphadhanakul, B., & Sriphrapradang, C. (2018).

Effects of sucralose on insulin and glucagon-like peptide-1 secretion in healthy subjects: a randomized, double-blind, placebo-controlled trial. Nutrition, 55, 125–

130.

Madjd, A., Taylor, M. A., Delavari, A., Malekzadeh, R., Macdonald, I. A., &

Farshchi, H. R. (2016). Beneficial effects of replacing diet beverages with water on type 2 diabetic obese women following a hypo‐energetic diet: A randomized, 24‐week clinical trial. Diabetes, Obesity and Metabolism, 19(1), 125-132.

Madjd, A., Taylor, M. A., Delavari, A., Malekzadeh, R., Macdonald, I. A., &

Farshchi, H. R. (2017). Effects of replacing diet beverages with water on weight loss and weight maintenance: 18-month follow-up, randomized clinical

trial. International Journal of Obesity, 42(4), 835-840.

Maersk, M., Belza, A., Holst, J. J., Fenger-Grøn, M., Pedersen, S. B., Astrup, A.,

& Richelsen, B. (2012). Satiety scores and satiety hormone response after sucrose-sweetened soft drink compared with isocaloric semi-skimmed milk and with non-caloric soft drink: a controlled trial. European journal of clinical nutrition, 66(4), 523-529. (A)

Maersk, M., Belza, A., Stødkilde-Jørgensen, H., Ringgaard, S., Chabanova, E., Thomsen, H., Pedersen, S.B., Astrup, A., & Richelsen, B. (2012). Sucrose- sweetened beverages increase fat storage in the liver, muscle, and visceral fat depot: a 6-mo randomized intervention study. The American journal of clinical nutrition, 95(2), 283-289. (B)

Malik, V. S., Pan, A., Willett, W. C., & Hu, F. B. (2013). Sugar-sweetened beverages and weight gain in children and adults: a systematic review and meta- analysis. The American journal of clinical nutrition, 98(4), 1084-1102.

Mattes, R.D. and Popkin, B.M. (2009). Nonnutritive sweetener consumption in humans: effects on appetite and food intake and their putative mechanisms. The American journal of clinical nutrition, 89(1), pp.1–14.

Miller, P. E., & Perez, V. (2014). Low-calorie sweeteners and body weight and composition: a meta-analysis of randomized controlled trials and prospective cohort studies. The American journal of clinical nutrition, 100(3), 765-777.

Nielsen, S. J., & Popkin, B. M. (2004). Changes in beverage intake between 1977 and 2001. American journal of preventive medicine, 27(3), 205-210.

Pan, A., & Hu, F. B. (2011). Effects of carbohydrates on satiety: differences

between liquid and solid food. Current Opinion in Clinical Nutrition & Metabolic

Care, 14(4), 385-390.

25 Peters, J. C., Beck, J., Cardel, M., Wyatt, H. R., Foster, G. D., Pan, Z.,

Wojtanowski, A.C., Vander Veur, S.S., Herring, S.J., James, C.B., & Hill, J. O.

(2016). The effects of water and non‐nutritive sweetened beverages on weight loss and weight maintenance: a randomized clinical trial. Obesity, 24(2), 297-304.

Phelan, S., Nallari, M., Darroch, F. E., & Wing, R. R. (2009). What do physicians recommend to their overweight and obese patients? The Journal of the American Board of Family Medicine, 22(2), 115-122.

Romo-Romo, A., Aguilar-Salinas, C. A., Brito-Córdova, G. X., Gómez-Díaz, R.

A., & Almeda-Valdes, P. (2018). Sucralose decreases insulin sensitivity in healthy subjects: a randomized controlled trial. The American journal of clinical

nutrition, 108(3), 485-491.

Sathyapalan, T., Thatcher, N. J., Hammersley, R., Rigby, A. S., Courts, F. L., Pechlivanis, A., Gooderman, N.J., Holmes, E., le Roux, C.W., Atkin, S. L., &

Courts, F. (2015). Aspartame sensitivity? A double blind randomised crossover study. PloS one, 10(3), e0116212-e0116212.

Singh, G. M., Danaei, G., Farzadfar, F., Stevens, G. A., Woodward, M., Wormser, D., ... & Di Angelantonio, E. (2013). The age-specific quantitative effects of metabolic risk factors on cardiovascular diseases and diabetes: a pooled analysis. PloS one, 8(7), e65174.

Solomi, L., Rees, G. A., & Redfern, K. M. (2019). The acute effects of the non- nutritive sweeteners aspartame and acesulfame-K in UK diet cola on glycaemic response. International journal of food sciences and nutrition, 70(7), 894-900.

Smith, M. A., Wells, M. H., Scarbecz, M., Vinall, C. V., & Woods, M. A. (2019).

Parents' Preferences and Perceptions of Their Children's Consumption of Sugar and Non-nutritive Sugar Substitutes. Pediatric dentistry, 41(2), 119–128.

Stern, S. E., Williams, K., Ferrannini, E., DeFronzo, R. A., Bogardus, C., & Stern, M. P. (2005). Identification of individuals with insulin resistance using routine clinical measurements. Diabetes, 54(2), 333–339.

Swedish Board of Agriculture, 2015. [Internet].

Available at:

http://www2.jordbruksverket.se/download/18.7d09976314ed25cc1f6bd707/14381 64110182/ra15_15.pdf

Toews, I., Lohner, S., de Gaudry, D. K., Sommer, H., & Meerpohl, J. J. (2019).

Association between intake of non-sugar sweeteners and health outcomes:

systematic review and meta-analyses of randomised and non-randomised controlled trials and observational studies. bmj, 364.

World Health Organization (WHO). Health diet, 2018 [Internet].

Available at: https://www.who.int/news-room/fact-sheets/detail/healthy-diet.