LCS as a means for body weight control: the evidence

Professor Raben then focused on LCS and weight control. There have been several studies on the effectiveness of the sweeteners, mainly aspartame, on weight control, and the very early studies were summarised by de la Hunty et al.^{(Reference De La Hunty, Gibson and Ashwell17)}.

A study by Raben's group using a 10-week ad libitum design demonstrated that body weight and fat mass decreased significantly in adults with overweight after the intake of foods and drinks containing LCS compared with similar products containing added sugar (sucrose), resulting in an average difference in body weight of 2⋅6 kg^{(Reference Raben, Vasilaras and Møller18)}. Ten years later, a 1⋅5-year study in children (aged 4–11 years) also showed a very clear picture^{(Reference de Ruyter, Olthof and Seidell19)}. In the present study, participants received 250 ml/d of an LCSB or a similar SSB while at school and the LCSBs reduced body weight gain and fat accumulation compared with the sugar-containing beverages. In both these studies, the volunteers were blinded to the intervention arms.

Comprehensive reviews and meta-analyses were published a few years later by Miller and Perez^{(Reference Miller and Perez20)} and by Rogers et al.^{(Reference Rogers, Hogenkamp and de Graaf21)}. The review by Rogers et al.^{(Reference Rogers, Hogenkamp and de Graaf21)} was the first to consider both acute and longer-term animal, human, cross-over, RCT and cohort studies on LCS, appetite, energy intake and body weight regulation. A meta-analysis of short-term RCTs (129 comparisons) showed a significantly reduced ad libitum consumption (94 kcal) after intake of LCS v. sugar-sweetened foods or beverages, with no difference when compared with water. A meta-analysis of intervention RCTs ranging from 1 to 40 months showed that LCS v. sugar led to a reduction in body weight of 1⋅35 kg (nine comparisons), and a similar relative reduction in body weight v. water (three comparisons). The systematic review from Azad et al.^{(Reference Azad, Abou-Setta and Chauhan22)} was neutral for RCTs showing no benefit nor weight gain in the LCS groups compared with the controls. Toews et al.^{(Reference Toews, Lohner and de Gaudry23)} concluded from their systematic review, which included thirty-five observational studies, that ‘In adults, evidence of very low and low certainty from a limited number of small studies indicated a small beneficial effect of NSSs [i.e. LCS] on body mass index’ and ‘For all other outcomes, no differences were detected between the use and non-use of NSSs, or between different doses of NSSs.’

In some reviews, the different LCS have been considered separately, which is very relevant, given that LCS have different metabolic fates. Recent reviews on aspartame alone or steviol glycosides alone showed no significant differences in body weight between the LCS and a control or sucrose^{(Reference Santos, Caria and Gotardo24,Reference Anker, Rafiq and Jeppesen25)} . Such analyses are, however, hampered by the limited number of studies, and the majority of RCTs have used a combination of sweeteners. More recently, a 12-week RCT included four different LCSBs (aspartame, saccharin, sucralose and rebaudioside A) and sucrose in a parallel-arm design including 154 participants^{(Reference Higgins and Mattes26)}. The beverages contained 400–560 kcal/d (sucrose treatments) or <5 kcal/d (LCS treatments). The results showed that sucrose and saccharin led to significantly increased body weight (by 1⋅85 and 1⋅18 kg, respectively) when compared with aspartame, rebaudioside A and sucralose. The change in body weight observed was directionally negative and significantly lower with sucralose when compared with the three other LCS (weight difference ≥ 1⋅37 kg). Energy intake also decreased with sucralose, supporting the observed weight loss in this group^{(Reference Higgins and Mattes26)}.

Water is generally believed to be the preferred choice over LCS beverages. To clarify this, the role of LCS in comparison with water was investigated in a 1-year RCT with 303 people with overweight and obesity. All participants took part in a 12-week behavioural weight loss programme and were then assigned to either 710 ml of water or LCSBs per day for 1 year. Convincingly, the study showed that participants drinking LCSBs maintained more than twice the weight loss (6⋅2 kg) compared with participants drinking water (2⋅6 kg), i.e. an improvement of weight control with LCS compared with water^{(Reference Peters, Beck and Cardel27)}.

To summarise, the balance of evidence indicates that the use of LCS in place of sugar (sucrose), in children and adults, can lead to reduced energy intake and body weight. The current evidence, although still limited, points to no difference or similar effects when LCS are compared with water.

LCS as a means of glycaemic control: the evidence

Professor Raben then summarised the studies on LCS and glycemic control. In a 10-week ad libitum study, postprandial glucose (PPG) and insulin concentrations after 10 weeks were significantly lower after LCS vs sucrose^{(Reference Raben, Møller and Flint28)}. This was also the case after adjusting for differences in body weight changes and fasting values at week 10. After further adjusting for differences in energy and sucrose intake, postprandial insulin (PPI) was still significantly lower on the LCS diet. The possible effect of LCS on gastrointestinal hormones (e.g. GIP and GLP-1) was also investigated. The results showed that postprandial GIP and GLP-1 concentrations at 10 weeks were significantly lower after LCS compared with sucrose, even after adjusting for differences in body weight changes, fasting GIP and GLP-1 values, energy and sucrose intake^{(Reference Raben, Møller and Flint28)}.

An RCT investigating the effect of 0, 350 or 1050 mg aspartame/d in a beverage for 12 weeks reported no differences in glucose, insulin, GLP-1 or GIP during an oral glucose tolerance test (OGTT) in healthy, lean adults^{(Reference Higgins, Considine and Mattes29)}. A similar result for glucose and insulin was seen during an OGTT after a 6-month intake of either 1 Diet Coke, Regular Coke, water or milk^{(Reference Engel, Tholstrup and Bruun30)}. In a systematic review and meta-analyses, LCS were also not found to elevate blood glucose level – rather a gradual decline in glucose was seen after LCS consumption^{(Reference Nichol, Holle and An31)}. It was also seen that the glycaemic impact of LCS did not differ by the type of LCS (aspartame, saccharin, steviosides and sucralose), but to some extent by age, body weight and diabetes state. These latter factors are, therefore, important to consider when comparing results from different studies on LCS.

Two other systematic reviews on LCS and glycaemia were published in the past 5 years^{(Reference Romo-Romo, Aguilar-Salinas and Brito-Córdova32,Reference Tucker and Tan33)} . Based on fourteen observational studies, LCS seemed to be related to the development of metabolic diseases, but adiposity was often an important confounder. In twenty-eight clinical trials, contradictory results were seen and, furthermore, studies were not comparable^{(Reference Romo-Romo, Aguilar-Salinas and Brito-Córdova32)}. Another analysis of forty-one studies showed that some LCS triggered physiological responses, although this was inconsistent. Without co-ingestion of carbohydrates, LCS acted similarly to water, and with co-ingestion of carbohydrates, LCS reduced plasma glucose compared with sucrose^{(Reference Tucker and Tan33)}. Very recently, a review and meta-analyses on steviol glycosides reported no effect of LCS on glucose^{(Reference Anker, Rafiq and Jeppesen25)}.

Professor Raben took the opportunity to present some preliminary results from a systematic review and meta-analysis of human intervention studies. It examined the acute effect of LCS intake on PPG and PPI responses and found that the ingestion of LCS has no acute effects on the mean change in postprandial glycemic or insulinemic responses compared with a control intervention^{(Reference Greyling, Appleton and Raben34)}.

Professor Raben concluded her presentation by looking to the future and describing the Horizon-2020 project ‘SWEET’ (www.sweetproject.eu, 2018–2023, Grant Agreement No. 774293), which aims to dig further into the potential risks and benefits of sweeteners and sweetness enhancers (S&SEs). The focus is on health, obesity, safety and sustainability in a multidisciplinary approach. A core part of SWEET is a 2-year RCT across Europe, which will investigate the effect of the prolonged use of S&SEs in a whole healthy diet approach (foods and drinks) on diet compliance, weight control and obesity-related risk factors (e.g. glycaemia and lipidemia) and safety (e.g. gut microbiota and allergenicity) in both adults and children.

More specifically, SWEET consists of a consortium of twenty-nine pan-European research, consumer and industry partners. In different work packages, diverse S&SE containing products will be developed and new databases generated (health, technological and sweetness). The potential toxicity and the regulatory frameworks affecting S&SE use will be assessed. In short to medium-term studies, the impact of specific S&SEs alone or in combination pertaining to gut hormone release, microbiota, central nervous system response, eating behaviour, satiety, reward, cravings and food choice in differing populations (gender, BMI and weight status) will be investigated. Besides the large-scale 2-year RCT, epidemiological evidence is being re-examined using multiple data sets with up to 170,000 individuals across different European populations. A comprehensive analysis of dietary composition and urinary biomarkers will be done to validate self-reported S&SE intake. The environmental, social and economic sustainability of increasing the production of S&SEs through life cycle analysis will also be investigated in the 5-year project.

Attitudes and beliefs about LCSBs

To address current gaps in knowledge, a questionnaire on the Attitudes and Beliefs towards LCS Beverages has recently been developed and this quantifies key factors associated with LCSBs by measuring attitudes and beliefs associated with their consumption^{(Reference Maloney, Christiansen and Harrold45)}. In this research, frequent consumers were defined as individuals who consumed over 825 ml of LCSBs per day, as determined using a self-reported online Food Frequency Questionnaire (see Appleton and Conner^{(Reference Appleton and Conner42)}). Results indicated that frequent and non-consumers of LCSBs had polarised attitudes and beliefs towards LCSBs^{(Reference Maloney, Christiansen and Harrold45)}. Specifically, frequent consumers had more positive beliefs that LCSBs were palatable and effective in controlling their appetite and body weight in comparison with non-consumers. As such, beliefs about hedonic enjoyment and health appear to influence consumer decisions about consumption or avoidance of LCSBs. These contrasting beliefs, and the more negative views among non-consumers, are not surprising, given that several studies have raised public awareness of potential adverse health effects of LCSBs,^{(Reference Sylvetsky and Rother36,Reference Fowler, Williams and Resendez46–Reference Swithers, Laboy and Clark48)} which likely discourages their consumption among some consumers. The importance of overcoming such misinformation and scepticism about LCS will be discussed in further detail later in the paper.

In line with these findings, Catenacci et al.^{(Reference Catenacci, Pan and Thomas49)} examined the motivations behind the consumption of LCSBs in individuals who had successfully maintained weight loss. They found that 78 % of consumers believed that LCSBs helped them control or reduce their total calorie consumption whilst also avoiding weight gain. In addition, palatability was another important factor driving the consumption of LCSBs in this sample. This is further evidence that LCSBs may help people to control their appetite and satisfy their food cravings when dealing with the continuous challenge of maintaining weight loss over time. Taken together, goals concerning body image and weight, coupled with positive hedonic reward and palatability beliefs, appear to be significant factors in motivating the consumption of LCSBs.

Psychological mechanisms underpinning the effect of LCSB on energy intake

As noted previously, in a systematic review and meta-analysis of short-term experimental studies, energy intake was significantly reduced when foods or beverages containing LCS were consumed relative to their sugar-containing counterparts^{(Reference Rogers, Hogenkamp and de Graaf21)}. However, the psychological mechanisms that underpin the effect of LCSB on energy intake are unclear.

One possibility is that consumers are using these beverages as a strategy to satisfy their desire for hedonic pleasure whilst simultaneously controlling their energy intake. As established earlier, frequent consumers of LCSB typically have higher levels of restrained eating, and as such, their eating behaviour is likely to be characterised by cycles of food restriction and disinhibited eating, ultimately making them more susceptible to weight gain^{(Reference Gorman, Allison and Allison50,Reference Lowe, Fairburn and Brownell51)} . According to the goal conflict model of eating behaviour, individuals with high dietary restraint find it difficult to regulate their food intake because they are juggling two conflicting goals: the hedonic goal of eating enjoyment whilst also satisfying the longer-term goal of weight maintenance^{(Reference Stroebe, Mensink and Aarts52)}. These goals frequently conflict with each other, because low-energy ‘diet’ foods are often less hedonically pleasing than foods of higher energy content^{(Reference Drewnowski53)}. Consumption of LCSBs may play an important role in this context. Sweet-tasting foods and beverages are hedonically pleasing to many people and LCSBs may, therefore, satisfy food cravings and hedonic eating goals. In addition, due to their very low energy content, LCSBs may simultaneously preserve weight control goals, thereby realigning previously conflicting goals in the goal conflict model. In this respect, LCSBs may offer great potential as a means of managing hedonic food motivations and cravings in individuals who struggle with their weight.

To explore this idea, a recent study used a ‘chocolate craving’ manipulation to examine the effect of priming hedonic eating goals on ad libitum energy intake in frequent consumers and non-consumers of LCSBs^{(Reference Maloney, Christiansen and Harrold54)}. It was hypothesised that energy intake would be greater after the hedonic eating prime relative to a control prime in non-consumers, but that frequent LCSB consumers would be protected from this effect (due to their consumption of LCSBs satisfying their hedonic eating motivations). Findings from two experiments did not consistently support this hypothesis. However, in the second experiment, the frequent consumers ingested fewer calories (less energy intake) overall when LCSBs were available relative to a condition when they were unavailable. Furthermore, when LCSBs were unavailable, frequent consumers reported lower perceived behavioural control (i.e. lower self-efficacy), lower meal enjoyment and higher eating-related guilt relative to the condition when LCSBs were available. These findings suggest that LCSBs assist frequent consumers in exercising self-control over food choices and weight control. This is important, given that previous research has found that emotions, such as eating-related guilt, can lead to negative outcomes including selection of indulgent foods, increased food consumption and long-term weight gain^{(Reference Kuijer and Boyce55,Reference Kuijer, Boyce and Marshall56)} .

Interestingly, Maloney et al.^{(Reference Maloney, Christiansen and Harrold54)} also found that frequent LCSB consumers had a visual attentional bias towards images of LCSBs, relative to water and SSB. This finding suggests that frequent consumers view LCSBs as hedonically desirable. This attentional bias was not evident for non-consumers, which is consistent with research showing that individuals selectively attend to personally relevant environmental stimuli^{(Reference Field, Werthmann and Franken57,Reference Kemps and Tiggemann58)} . Importantly, frequent consumers showed a visual preference for LCSBs relative to sugar-containing beverages, which indicates a specific bias towards LCSBs rather than a general bias towards sweet-tasting products. This finding is contrary to the hypothesis that LCSBs encourage a generalised preference for sweet-tasting foods^{(Reference Casperson, Johnson and Roemmich59–Reference Swithers, Martin and Davidson61)}. Indeed, evidence to date suggests that exposure to sweet taste does not promote a subsequent preference for sweet products but, in fact, leads to a reduced preference for sweetness in the short term^{(Reference Appleton, Tuorila and Bertenshaw62)}. However, the authors of this systematic review highlighted that the existing evidence base is weak and that there is a need for longer-term, adequately powered studies.

Collectively, findings to date indicate that LCSBs appear to be fulfilling a psychological role for consumers by satisfying their hedonic food motivations without violating dieting goals. In doing so, LCSBs could play a meaningful role in reducing energy intake by facilitating self-regulation in the face of high-calorie food temptation, without the accompanying caloric intake and guilt. Nevertheless, further research is still needed to understand how these beverages affect cognitions and subsequent appetitive behaviours.

Professor Halford ended his presentation by emphasising that, whilst the beneficial effects of LCS beverages in weight management have been reported^{(Reference Peters, Beck and Cardel27,Reference Peters, Wyatt and Foster63)} , the effects of LCS over longer time periods need to be further elucidated. The SWITCH trial (Trial registration: Clinical Trials: NCT02591134; registered: 23 October 2015) is addressing these research gaps by exploring the longer-term effects of LCSBs in weight management and underpinning physiological and psychological mechanisms^{(Reference Masic, Harrold and Christiansen64)}. Specifically, this ongoing trial is assessing the effect of LCSBs, relative to water, on both short and long-term weight management, as well as examining several candidate behavioural and biological mechanisms (e.g. changes in glycaemic control, fasting lipid profiles, appetite, energy intake, food choice, mood and attitudes) through which these effects may arise relative to water. This research will provide a new and detailed understanding of the role that LCSBs play in weight loss and maintenance, particularly their psychological impact, and of the behavioural mechanisms that mediate these effects.