AJCN EB Program 2010
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Monsivais, P.
Right arrow Articles by Drewnowski, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Monsivais, P.
Right arrow Articles by Drewnowski, A.
Agricola
Right arrow Articles by Monsivais, P.
Right arrow Articles by Drewnowski, A.
American Journal of Clinical Nutrition, Vol. 86, No. 1, 116-123, July 2007
© 2007 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Sugars and satiety: does the type of sweetener make a difference?1,2,3

Pablo Monsivais, Martine M Perrigue and Adam Drewnowski

1 From the Nutritional Sciences Program, School of Public Health and Community Medicine (PM, MMP, and AD), and the Department of Dental Public Health Sciences, School of Dentistry (PM), University of Washington, Seattle, WA

2 Supported by a grant from the American Beverage Association, by the Corn Refiners Association, and by fellowship T32 DE07132 from the National Institute of Dental and Craniofacial Research (to PM).

3 Reprints not available. Address correspondence to A Drewnowski, 305 Raitt Hall, Box 353410, University of Washington, Seattle, WA 98195. E-mail: adamdrew{at}u.washington.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Widespread use of high-fructose corn syrup (HFCS) in beverages has been linked to rising obesity rates. One hypothesis is that HFCS in beverages has little satiating power.

Objective: The objective of the study was to compare the relative effect of commercial beverages containing sucrose or HFCS on hunger, satiety, and energy intakes at the next meal with the use of a within-subject design.

Design: Thirty-seven volunteers (19 men, 18 women) aged 20–29 y consumed isocaloric cola beverages (215 kcal) sweetened with sucrose, HFCS 42, or HFCS 55. HFCS 42 contains 42% fructose, and HFCS 55 contains 55% fructose. Diet cola (2 kcal), 1%-fat milk (215 kcal), and no beverage were the control conditions. The 5 beverages were consumed at 1010 (2 h after a standard breakfast). Participants rated hunger, thirst, and satiety at baseline and at 20-min intervals after ingestion. A tray lunch (1708 kcal) was served at 1230, and energy intakes were measured. The free sugars content of sucrose-sweetened cola was assayed at the time of the study.

Results: We found no differences between sucrose- and HFCS-sweetened colas in perceived sweetness, hunger and satiety profiles, or energy intakes at lunch. The 4 caloric beverages tended to partially suppress energy intakes at lunch, whereas the no-beverage and diet beverage conditions did not; the effect was significant (P < 0.05) only for 1%-fat milk. Energy intakes in the diet cola and the no-beverage conditions did not differ significantly.

Conclusion: There was no evidence that commercial cola beverages sweetened with either sucrose or HFCS have significantly different effects on hunger, satiety, or short-term energy intakes.

Key Words: Beverages • sucrose • fructose • high-fructose corn syrup • HFCS • sweetness • hunger • fullness • satiety • energy intakes


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The introduction of corn sweeteners into the US food supply is said to have contributed to the current obesity epidemic (1-3). High-fructose corn syrup (HFCS) began to replace sucrose in soft drinks at approximately the same time that obesity rates in the United States began their sharp increase (2, 4). However, temporal parallels between HFCS consumption patterns and body-weight trends are not sufficient to show causality. Obesity has also increased sharply in countries where beverage consumption is lower than in the United States and HFCS is not a common sweetener (5).

One of the criteria for establishing causality in evidence-based medicine is a biologically plausible mechanism. Attempts to establish a causal link between soft drink consumption and rising obesity rates have therefore relied on the notion that caloric beverages in general (6), and HFCS-sweetened beverages in particular (2, 7), lack satiating power. Research reports have suggested that liquids were less satiating than were solids (8); that sugars were less satiating than was either protein or fat (9); and that HFCS blunted the satiety response more than did other sweeteners (2, 10, 11). The metabolic and endocrinologic processes associated with the ingestion of free fructose have featured prominently in arguments that HFCS-sweetened beverages are the principal culprit in the obesity epidemic (2, 6, 7).

However, satiety-related arguments based on the ingestion of pure fructose or fructose-rich stimuli (12, 13) may not apply to sweetened beverages, given that the 2 most common forms of HFCS—HFCS 55 and HFCS 42—contain 55% and 42% free fructose, respectively, and the remainder is free glucose. Furthermore, the sharp distinctions made between HFCS-sweetened and sucrose-sweetened beverages (2, 14) may be incorrect. The low pH of carbonated soft drinks favors the breakdown of sucrose into free glucose and free fructose before consumption (15), and the rate of hydrolysis is dependent on storage variables, temperature, and time (16). Perhaps most important, the short-term satiating power of foods and beverages may have little to do with the long-term regulation of body weight (13, 17).

The present study was a direct test of the hypothesis that HFCS-sweetened carbonated soft drinks differ significantly from sucrose-sweetened soft drinks and from low-fat milk in their effect on satiety. Aiming to approximate naturalistic conditions of soft drink use, we compared the effect of commercially available cola beverages, sweetened with sucrose or with 2 types of HFCS (HFCS 42 and HFCS 55), on hunger, satiety, and energy intakes (EIs) at the test meal. Because so much has been made of the metabolic differences between free fructose and fructose bound within disaccharide sucrose molecules (2, 14), we sent samples of the sucrose-sweetened beverage to be analyzed for free sugars content at the time of the experiment.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Participants
Thirty-seven participants (19 M, 18 F) aged 20–29 y were recruited at the University of Washington with the use of advertisements and flyers. A telephone-administered screening interview was used to verify eligibility criteria. Eligible participants were normal-weight to overweight [body mass index (BMI; in kg/m2): 18–30], regularly consumed breakfast, did not smoke, and were not following a diet to gain or lose weight. Persons with food allergies or food restrictions; those who disliked ≥2 of the foods or beverages in the study; those taking prescription medications that were likely to affect taste, smell, or appetite; athletes in training; pregnant or lactating women; and persons reporting recent weight loss or weight cycling were excluded. Potential candidates were invited to report to the laboratory for a brief session, during which their weight and height were measured. The Eating Disorder Inventory (18) and the cognitive restraint subscale of the Eating Inventory (19) were administered as screening instruments to exclude persons with indications of eating disorders or restrained eating patterns. Persons who met all eligibility criteria were invited to participate and were given a reminder card stating the dates and times for the study sessions. To minimize variability, each participant was asked to report to the laboratory on the same day of the week throughout the study, to keep evening meals and activity levels on the day before each test as similar as possible, to refrain from drinking alcohol the day before each test, and to have a standardized breakfast at {approx}0800 on the mornings when they were scheduled to have a test. The participants’ standardized breakfasts were consumed at home and were composed of specified servings of hot or cold cereals with milk along with a medium-sized apple, orange, or banana or a specified serving of low-fat yogurt along with a medium-sized fruit.

All participants provided written informed consent. The study protocol was approved by the Institutional Review Board at the University of Washington. All 37 subjects completed the study and were compensated for their time.

Study design
The study followed a repeated-measures within-subject design, in which each participant returned for 6 separate test sessions. The sessions lasted from 0930 to 1310 and were spaced at least a week apart. The order of presentation of the 5 preloads and the no-beverage condition was counterbalanced. The same lunch foods were offered on all 6 testing occasions. The magnitude of the energy manipulation (0 or 215 kcal) was based on a review of previous studies in this area (20). Power analysis indicated that, with a power of 80% and an alpha of 0.05, a sample of 35 subjects was sufficient to detect a minimum difference of 150 kcal in compensation (21).

Beverage stimuli
The 5 beverages were cola sweetened with HFCS 42 (Sam's Choice Cola; Cott Beverages, Toronto, Canada), cola sweetened with HFCS 55 (Coca-Cola Classic; Coca-Cola Co, Atlanta, GA), cola sweetened with sucrose (Coca-Cola Classic), cola sweetened with aspartame (Diet Coke; Coca-Cola Co), and 1%-fat milk (Darigold; Wesfarm Foods, Seattle, WA). All preload beverages with the exception of the diet cola (2 kcal) were isoenergetic (894 kJ or 215 kcal) and of comparable sweetness, but they differed in sugar composition. To keep both sweetness and energy constant, the preload volume was allowed to vary within narrow limits (from 475 to 525 mL.). The composition and energy density of the 5 beverage preloads are shown in Table 1Go. All beverages were coded and were served chilled, without ice, in opaque containers with a lid and a straw. Participants were asked to consume the entire amount within 15 min.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Energy and macronutrient composition of the preloads1

 
Samples of the sucrose-sweetened beverage were analyzed at regular intervals during the data collection phase. The analyses, conducted by the Analytic Chemistry Department of the Coca-Cola Company, used samples from the same production run (25 May 2005) of Coca-Cola Classic and were conducted at the same time as the data collection phase of the present study.

Motivational ratings and hedonic evaluations
Participants used computerized, semi-anchored visual analogue scales (VASs) to rate their hunger, fullness, thirst, nausea, and desire to eat. The VAS software was custom-written by using the LABVIEW graphic programming software (version 6.1; National Instruments, Austin, TX) that was running on 10 identical Macintosh G3 computers (Apple Computers, Cupertino, CA). Motivational scales were presented one at a time (ie, one scale per screen) on the computer monitor. Each participant used a mouse to position a cursor along the 100-mm bar displayed on a flat-panel LCD computer monitor. The VAS bars were anchored at each extreme with the labels "not at all ..." and "extremely ..." (22). A semi-anchored VAS was also used for quantifying several sensory and hedonic attributes of each beverage. Each sensory attribute scale also was anchored with the labels "not at all..." and "extremely ...". Hedonic ratings and ratings along 11 sensory attribute scales were obtained for each beverage.

Test meal
A lunch meal served on a tray was provided at 1230. Identical meals were provided on each occasion. The set meal was 7120 kJ (1708 kcal) and included a variety of foods, both savory and sweet. Each lunch consisted of a selection of 2 grains, 2 types of fruit, 2 vegetables, 2 cheeses, 2 meats, 2 candies, 1 yogurt, 1 ice cream cup, hummus, chips, and water. A large cup containing 591 mL (20 fl oz) still water was provided with the test lunch. Participants were told that they could have as much or as little as they would like of any food or water and that they could request unlimited additional portions. All foods and water were weighed at the time of serving. Plate waste was collected and weighed by the experimenters. Food energy and nutrient values were calculated with FOOD PROCESSOR software (version 8.1; ESHA Research, Salem, OR) and from the manufacturer's food labels. The nutrient composition of the meal is shown in Table 2Go.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Energy and nutrient composition of foods provided at lunch1

 
Procedures
Participants were asked to consume a standard breakfast at home starting at 0800. They arrived at the laboratory at 0930 and were seated in separate cubicles. They remained there for the duration of the session and were allowed to read, listen to music with earphones, or use their portable computers. They were allowed to leave the room briefly to stretch or use the restroom. Participants were also asked to record all foods and beverages they had consumed for breakfast earlier that morning and to note the time of consumption. Motivational ratings were first obtained 10 min after arrival (baseline or time 0) and every 20 min thereafter until lunchtime (times 1 through 8). The preload was provided in the laboratory at 1010, and lunch was provided at 1230. The last set of ratings was obtained after lunch (time 9), after which participants left the laboratory.

Data analyses and statistical tests
We used SPSS for WINDOWS software [version 11.1 (23)] for all analyses. Normality was determined by the Kolmogorov-Smirnov test (normal if P > 0.05). Analyses of motivational ratings used a nested repeated-measures analysis of variance (ANOVA) with beverage type and time after ingestion (times 2–8) as the within-subjects factors and sex as the between-subjects factor. Analyses were conducted for all 6 conditions together and separately for the 3 sweetener conditions (sucrose, 42% HFCS, and 55% HFCS) and the 3 comparison conditions (aspartame, 1%-fat milk, and no beverage). Univariate tests of within-subject effects were subject to Huynh-Feldt correction when the sphericity assumption was violated. Bonferroni-adjusted pairwise comparisons were made when ANOVAs were significant. Because there were no significant main effects or sex-related interactions (P > 0.05 for all tests), the data were combined by sex for each beverage condition. Analyses of EIs and the weight of foods and water consumed at lunch used a repeated-measures ANOVA with beverage type as the within-subject factor and sex as the between-subjects factor. The strength of the association between prelunch (time 8) appetite ratings and energy or water intakes at lunch was tested by using Pearson's correlation coefficients. Prelunch appetite ratings were computed by averaging 3 proxies for appetite according to a method similar to that of Anderson et al (11): hunger, desire to eat, and the inverse of fullness (ie, 100 – fullness). Sweetness intensity and hedonic ratings were analyzed by repeated-measures ANOVA.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Participants and beverage stimuli
Mean (±SD) age was 22.6 ± 4.0 y for men and 23.4 ± 2.8 y for women. Mean body weight was 77.5 ± 10.7 kg for men and 60.2 ± 9.1 kg for women. Body mass index (BMI; in kg/m2) values were 23.4 ± 1.8 for men and 21.9 ± 2.7 for women.

Analysis of sweetness ratings for the 4 cola beverages found a significant (F3, 99 = 11.1, P < 0.001) main effect of beverage type. However, that was entirely due to the aspartame-sweetened cola, which was perceived as significantly less sweet than the 3 sugar-sweetened colas (P < 0.01 for all 3 comparisons). Cola beverages sweetened with sucrose, HFCS 55, and HFCS 42 did not differ significantly in perceived sweetness ratings. Milk (1% fat) was not perceived as sweet. All 5 beverages (including milk) were rated as equally palatable by the participants. The analysis of hedonic preference ratings showed no significant main beverage effect (F41, 32 = 2.5).

Motivational ratings
All 5 beverages (caloric and not) led to lower hunger ratings during the initial 20 min after ingestion than were seen in the no-beverage condition. Repeated-measures ANOVA showed significant main effects of both beverage type and time and a significant beverage x time interaction (P < 0.001 for all). Separate analyses were then conducted for the 3 colas and the 3 comparison conditions.

The time course of energy and volume effects on short-term hunger is shown in Figure 1Go (left). In the first 60 min after their ingestion, equal volumes of 1%-fat milk and noncaloric diet cola had comparable effects on perceived hunger. However, hunger ratings rose more rapidly in the diet cola condition than in the milk condition, and a separation in hunger ratings was visible before lunch. ANOVA found a significant (P < 0.001) main effect of beverage type and a significant (P < 0.005) beverage x time interaction.


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 1.. Temporal profiles of mean hunger ratings as a function of preload condition for control beverages (left) and sugar-sweetened carbonated beverages (right). n = 37. VAS, visual analogue scale; HFCS 42 and 55, high-fructose corn syrup containing 42% and 55% fructose, respectively. The no-beverage control was associated with higher hunger levels than were the milk and aspartame-sweetened diet cola controls (P < 0.005), and the beverage x time interaction was significant (P < 0.005). Hunger ratings for the 3 sugar-sweetened beverages did not differ significantly.

 
In contrast, the 3 sugar-sweetened cola beverages, although different from the no-beverage condition, did not differ significantly from each other (Figure 1Go, right). There was no significant main beverage effect or beverage x time interaction.

ANOVA of fullness ratings found significant main effects of beverage type and time (P < 0.001 for both) and a significant beverage x time interaction (P < 0.005). As shown in Figure 2Go (left), milk and diet cola had different effects on fullness ratings, depending on the time after ingestion. The beverage effect and the beverage x time interaction were significant (P < 0.01 for both). In contrast, the sucrose- and the HFCS-sweetened colas differed significantly (P < 0.001) from the no-beverage condition but not from each other. The beverage x time interaction was not significant (Figure 2Go, right).


Figure 2
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 2.. Temporal profiles of mean fullness ratings as a function of preload condition for control beverages (left) and sugar-sweetened carbonated beverages (right). n = 37. VAS, visual analogue scale; HFCS 42 and 55, high-fructose corn syrup containing 42% and 55% fructose, respectively. The no-beverage control was associated with significantly lower fullness ratings than were the milk and aspartame-sweetened diet cola controls (P < 0.001 for both). Diet cola was associated with significantly lower fullness ratings than was milk (P < 0.05), and the beverage x time interaction was significant (P < 0.01). Fullness ratings for the 3 sugar-sweetened beverages did not differ significantly.

 
Analysis of the desire-to-eat ratings found significant main effects of beverage type and time (P < 0.001 for both) and a significant beverage x time interaction (P < 0.01). It can be seen in Figure 3Go (left) that 1%-fat milk and diet cola had different effects on the desire to eat, depending on the time after ingestion. The effect of beverage type was significant (P < 0.001), as was the beverage x time interaction (P < 0.05). Again, the sucrose- and the HFCS-sweetened colas did not differ significantly from each other (Figure 3Go, right).


Figure 3
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 3.. Temporal profiles of the mean desire to eat as a function of preload condition for control beverages (left) and sugar-sweetened carbonated beverages (right). n = 37. VAS, visual analogue scale; HFCS 42 and 55, high-fructose corn syrup containing 42% and 55% fructose, respectively. The no-beverage control was associated with a significantly greater desire to eat than were the milk and aspartame-sweetened diet cola controls (P < 0.005), and the beverage x time interaction was significant (P < 0.05). Desire-to-eat ratings for the 3 sugar-sweetened beverages did not differ significantly.

 
Analysis of thirst ratings (Figure 4Go) found significant main effects of beverage type and time (P < 0.001 for both). The beverage x time interaction also was significant (P < 0.01). Subjects in the no-beverage condition reported significantly higher thirst ratings than did those in the other 5 conditions (P < 0.05 for all comparisons). The 5 beverages did not differ significantly from each other in thirst ratings. Participant nausea ratings did not vary significantly by preload condition, and nausea did not vary significantly as a function of time.


Figure 4
View larger version (22K):
[in this window]
[in a new window]

  		FIGURE 4.. Temporal profiles of mean thirst ratings as a function of preload condition for all 6 preload conditions. n = 37. VAS, visual analogue scale; HFCS 42 and 55, high-fructose corn syrup containing 42% and 55% fructose, respectively. The no-beverage control was associated with significantly greater thirst than were the milk and aspartame-sweetened diet cola controls (P < 0.001), and the beverage x time interaction was significant (P < 0.05). Thirst ratings for the 3 sugar-sweetened beverages did not differ significantly.

 
Energy and nutrient intakes
For each condition, energy and water intakes at lunch, as well as the weight of all foods consumed, are shown in Table 3Go. Across all 6 conditions, men consumed an average of 1077 kcal at lunch, whereas women consumed an average of 862 kcal. Whereas this effect of sex on EIs was significant (P < 0.005), the beverage type x sex interaction was not.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Energy and water intakes and the weight of foods and beverages consumed at lunch for each preload condition1

 
The 4 caloric beverages partially suppressed EIs at lunch as compared with the EIs in the no-beverage condition. The main effect of beverage type on EIs was significant (P < 0.05). However, pairwise comparisons showed that the effect was significant only for the 1%-fat milk preload (P < 0.05). EIs in the no-beverage condition and in the diet cola condition did not differ significantly, which indicated that preload volume had no effect on EIs by the time the test lunch was served (120 min after preload ingestion).

The combined energy content of the preload and the lunch also showed significant main effects of beverage type (P < 0.001). Pairwise comparisons showed that the 3 sugar conditions did not differ significantly from each other or from milk. However, all 4 caloric beverages differed significantly from both the diet cola and the no-beverage conditions.

Beverage type significantly affected the weight of food and the amount of water consumed at lunch (P < 0.001 for both). Participants in the no-beverage condition consumed significantly more water at lunch than did subjects in all of the other preload conditions except for the diet cola condition. The nutrients consumed in association with each beverage condition are shown in Table 4Go. Beverage type did not significantly affect the nutrient composition of the lunch meal. Overall, the meals selected and consumed by the participants provided 51.2% of energy as carbohydrate, 17.2% of energy as protein, and 32.8% of energy as fat. The nutrient composition of lunch did not differ significantly by sex.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Macronutrient intakes at lunch1

 
Motivational ratings and energy intakes
A composite score of appetite was calculated by using the method of Anderson et al (11). The correlation between appetite ratings and EIs at lunch was significant for women (r = 0.33, P < 0.001) but not for men (r = 0.17, NS). In contrast, the correlation between prelunch thirst ratings and water consumption at lunch was significant for both women (r = 0.39, P < 0.001) and men (r = 0.2, P < 0.05).

Free sugar content of sucrose-sweetened beverage
The progressive hydrolysis of sucrose in cola beverages over the course of the study is shown in Figure 5Go. Each point is based on the analysis of 3 samples. As sucrose hydrolyzed, its concentration declined from 36% of total sugars on June 30 to just above 10% on August 24, or {approx}3 mo after the beverages were manufactured (May 25). Free fructose increased from 32% to {approx}44%. Free glucose (not shown) followed the same course as fructose. During the time of the satiety study (period indicated by the shaded bar), the principal sugars in the sucrose-sweetened cola were free fructose and glucose. Dashed reference lines at 50.6% and 6.4% indicate concentrations of sucrose present in samples from a separate lot of sucrose-sweetened cola measured 10 d and 1 y after manufacture.


Figure 5
View larger version (26K):
[in this window]
[in a new window]

  		FIGURE 5.. Timeline of sucrose inversion in the sucrose-containing cola used in this study. Each data point represents the mean of 3 samples taken from the same lot on the indicated date. As sucrose hydrolyzed over the weeks of the study, its concentration declined from 36% of total sugars to just above 10%. Free fructose increased from 32% to {approx}45% of total sugar.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The argument that HFCS-sweetened beverages play a causal role in the obesity epidemic (24) rests, in part, on the notion that free fructose blunts the satiety response more strongly than do other sweeteners (2, 11, 13). Bray et al (14) made a sharp distinction between "free" and bound fructose in soft drinks, arguing further that HFCS-containing beverages could "never" have the same sweetness as sucrose-sweetened ones.

As the present study shows, the 3 cola beverages, which were sweetened with sucrose, HFCS 55, or HFCS 42, were perceived as equally sweet and significantly sweeter than diet cola. All 3 beverages showed identical temporal profiles of motivational ratings, which were different from the no-beverage condition. There were no differences between the reported temporal profiles for hunger, satiety, and the desire to eat obtained after the ingestion of HFCS- or sucrose-sweetened colas. Compared with the EIs under the no-beverage condition, the 3 colas and 1%-fat milk weakly suppressed EIs at lunch, whereas the diet cola did not. However, the effect was significant only for 1%-fat milk.

The present study used the conventional preload paradigm (25-27) and commercially available colas and 1%-fat milk. The statistical power was similar to that used in past research (9, 25). The study design was thus able to separate the effects of preload energy and preload volume. As had been noted in previous reports, diet cola suppressed hunger immediately after ingestion, but hunger ratings recovered sooner than did those after ingestion of caloric 1%-fat milk. Preload volume alone had no effect on EIs 120 min later. The amount of food consumed at lunch in the diet cola condition did not differ significantly from that in the no-beverage condition.

The notion that HFCS-sweetened beverages differ substantially from sucrose-sweetened beverages (2, 14) seems to have been based on the incorrect assumption that sucrose in solution remains stable. As documented by multiple laboratory assays conducted over the course of the present study, the sucrose content of the cola beverage dropped from 35% to 10%, whereas the free fructose content rose from 32% to 44% of total sugars. Other hydrolysis data indicate that 50% of the sucrose in carbonated beverages is hydrolyzed within 10 d of manufacture and that {approx}90% is hydrolyzed within 3 mo. Given the time lag between production and consumption, it is likely that most sugars in sucrose-sweetened beverages are already in the form of free fructose and glucose by the time the beverages are consumed (16).

Previous laboratory studies of sugars and satiety in humans, most often conducted with aqueous solutions of pure sugars (11, 12), produced inconsistent results. One early study did not find differences between glucose and fructose in hunger ratings or EIs (12). A study in 16 women showed that 50 g fructose in 500 mL water significantly reduced lunch intakes as compared with a water control (28). A study of different glucose-fructose mixtures showed that high-glucose (80%) stimuli elevated blood glucose concentrations more than did high-fructose stimuli (80%), but that the 2 mixtures had similar effects on appetite (T Akhavan and GH Anderson, unpublished observations, 2005). The same study showed that high-fructose stimuli (65% and 80%) were associated with lower short-term food intake than was sucrose.

The putative effect of HFCS on satiety hormones awaits further research. In a recent study, 30 lean women consumed cola beverages at lunch served at 1300 h, and blood samples were drawn before the meal and 30 min and 60 min afterward (29). There were no significant differences between plasma glucose, insulin, leptin, or ghrelin after the ingestion of HFCS- or sucrose-sweetened cola. It is possible that the responses to sucrose- and HFCS-sweetened cola beverages were similar because the sugar content of the stimuli was, in fact, much the same. Studies of the human response to sweetened liquids should assess the stimulus sugar composition at the time of testing.

The present study does not resolve the underlying question of whether liquids differ from solids in their satiating power. Past studies found no difference in satiating power between liquid cola and solid fat-free cookies (30) or between a drinkable liquid yogurt and the same yogurt that was eaten with a spoon (9). A review (20) showed that, whereas some studies found that liquids were less satiating than solids, other studies showed the opposite. Most recently, a study showing that apple juice had higher satiating power when it was called a "soup" attributed the difference to cognitive rather than physiologic factors (31).

In the present study, conducted with liquids only, a slightly but significantly different degree of compensation as compared with the no-beverage condition was obtained for 1%-fat milk only. This finding is of interest because 2 previous studies observed no differences in EIs among regular cola, orange juice, or 1%-fat milk, no matter whether the beverages were consumed 135 min before lunch (32) or with lunch (27). On the other hand, yogurts were associated with higher fullness ratings than were either dairy- or fruit-based drinks (9), possibly because of their higher protein content.

Although laboratory studies conducted with pure sugar solutions provide valuable data on fructose metabolism (12, 13), not all observations can be extrapolated to the human food supply. The hydrolysis of sucrose in soft drinks before consumption suggests that the substitution of HFCS for sucrose did not have the dramatic effects that had been claimed (2). The emerging view voiced in the scientific literature (33) and in the news media (34) is that any potential contribution of sugars to obesity is unlikely to be mediated by metabolic effects that are unique to HFCS.


   ACKNOWLEDGMENTS
 
We thank Susan L Adams for assistance in the planning of the test lunch, Robyn Sakamoto for assistance in conducting the study, and Chris Prall for VAS software development. The Nutritional Sciences Program has received prior research support from industry sources in the United States and the European Union.

The authors’ responsibilities were as follows—PM, MMP, and AD: study design and writing of the manuscript; PM and MMP: study implementation and data collection; PM, MMP, and AD: statistical analysis and revision of the dietary data. None of the authors had a personal or financial conflict of interest.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Gross LS, Li L, Ford ES, Liu S. Increased consumption of refined carbohydrates and the epidemic of type 2 diabetes in the United States: an ecologic assessment. Am J Clin Nutr 2004;79:774–9.[Abstract/Free Full Text]
  2. Bray GA, Nielsen SJ, Popkin BM. Consumption of high-fructose corn syrup in beverages may play a role in the epidemic of obesity. Am J Clin Nutr 2004;79:537–43.[Abstract/Free Full Text]
  3. Critser G. Fat land: how Americans became the fattest people in the world. Boston, MA: Houghton Mifflin Co, 2003.
  4. Ebbeling CB, Pawlak DB, Ludwig DS. Childhood obesity: public-health crisis, common sense cure. Lancet 2002;360:473–82.[Medline]
  5. Schorin MD. High fructose corn syrups, Part 2 health effects. Nutr Today 2006;41:70–7.
  6. Malik VS, Schulze MB, Hu FB. Intake of sugar-sweetened beverages and weight gain: a systematic review. Am J Clin Nutr 2006;84:274–88.[Abstract/Free Full Text]
  7. Wylie-Rosett J, Segal-Isaacson CJ, Segal-Isaacson A. Carbohydrates and increases in obesity: does the type of carbohydrate make a difference? Obes Res 2004;12(suppl):124S–9S.[Medline]
  8. Hulshof T, De Graaf C, Weststrate JA. The effects of preloads varying in physical state and fat content on satiety and energy intake. Appetite 1993;21:273–8.[Medline]
  9. Tsuchiya A, Almiron-Roig E, Lluch A, Guyonnet D, Drewnowski A. Higher satiety ratings following yogurt consumption relative to fruit drink or dairy fruit drink. J Am Diet Assoc 2006;106:550–7.[Medline]
  10. Elliott SS, Keim NL, Stern JS, Teff K, Havel PJ. Fructose, weight gain, and the insulin resistance syndrome. Am J Clin Nutr 2002;76:911–22.[Abstract/Free Full Text]
  11. Anderson GH, Catherine NL, Woodend DM, Wolever TM. Inverse association between the effect of carbohydrates on blood glucose and subsequent short-term food intake in young men. Am J Clin Nutr 2002;76:1023–30.[Abstract/Free Full Text]
  12. Kong MF, Chapman I, Goble E, Wishart J, Wittert G, Morris H, Horowitz M. Effects of oral fructose and glucose on plasma GLP-1 and appetite in normal subjects. Peptides 1999;20:545–51.[Medline]
  13. Teff KL, Elliott SS, Tschop M, et al. Dietary fructose reduces circulating insulin and leptin, attenuates postprandial suppression of ghrelin, and increases triglycerides in women. J Clin Endocrinol Metab 2004;89:2963–72.[Abstract/Free Full Text]
  14. Bray GA, Nielsen SJ, Popkin BM. Reply to MF Jacobson. Am J Clin Nutr 2004;80:1081–2 (letter).[Free Full Text]
  15. Hanover LM, White JS. Manufacturing, composition, and applications of fructose. Am J Clin Nutr 1993;58(suppl):724S–32S.[Abstract/Free Full Text]
  16. Hein GL, Storey ML, White JS, Lineback DR. Highs and lows of high fructose corn syrup: a report from the Center for Food and Nutrition Policy and its Ceres Workshop (Food Science). Nutr Today 2005;40:253–6.
  17. Drewnowski A. Palatability and satiety: models and measures. Ann Nestlé 1998;56:32–42.
  18. Garner DM, Olmstead MP, Polivy J. The Eating Disorder Inventory: a measure of cognitive-behavioral dimensions of anorexia nervosa and bulimia. In: Darby PL, Garfinkel PE, Garner DM, Coscina DV, eds. Anorexia nervosa: recent developments in research. New York, NY: AR Liss, 1983:173–84.
  19. Stunkard AJ, Messick S. The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. J Psychosom Res 1985;29:71–83.[Medline]
  20. Almiron-Roig E, Chen Y, Drewnowski A. Liquid calories and the failure of satiety: how good is the evidence? Obes Rev 2003;4:201–12.[Medline]
  21. Tabachnick B, Fidell L. Using multivariate statistics. 3rd ed. Mahway, NY: Harper Collins, 1996.
  22. Drewnowski A, Krahn DD, Demitrack MA, Nairn K, Gosnell BA. Taste responses and preferences for sweet high-fat foods: evidence for opioid involvement. Physiol Behav 1992;51:371–9.[Medline]
  23. Norusis MJ. SPSS/PC+. Chicago, IL: SPSS Inc, 1986.
  24. Gaby AR. Adverse effects of dietary fructose. Altern Med Rev 2005;10:294–306.[Medline]
  25. Rolls JB, Fedoroff IC, Guthrie JF, Laster LJ. Foods with different satiating effects in humans. Appetite 1990;15:115–26.[Medline]
  26. Rolls BJ, Bell EA, Thorwart ML. Water incorporated into a food but nor served with a food decreases energy intake in lean women. Am J Clin Nutr 1999;70:448–55.[Abstract/Free Full Text]
  27. DellaValle DM, Roe LS, Rolls BJ. Does the consumption of caloric and non-caloric beverages with a meal affect energy intake? Appetite 2005;44:187–93.[Medline]
  28. Guss JL, Kissileff HR, Pi-Sunyer FX. Effects of glucose and fructose solutions on food intake and gastric emptying in nonobese women. Am J Physiol 1994;267(6 Pt 2):R1537–44.
  29. Melanson KJ, Zukley L, Lowndes J, Nguyen V, Angelopoulos TJ, Rippe JM. Effects of high-fructose corn syrup and sucrose consumption on circulating glucose, insulin, leptin, and ghrelin and on appetite in normal-weight women. Nutrition 2007;23:103–12.[Medline]
  30. Almiron-Roig E, Flores SY, Drewnowski A. No difference in satiety or in subsequent energy intakes between a beverage and a solid food. Physiol Behav 2004;82:671–7.[Medline]
  31. Mattes R. Soup and satiety. Physiol Behav 2005;83:739–47.[Medline]
  32. Almiron-Roig E, Drewnowski A. Hunger, thirst, and energy intakes following consumption of caloric beverages. Physiol Behav 2003;79:767–73.[Medline]
  33. Jacobson MF. High-fructose corn syrup and the obesity epidemic. Am J Clin Nutr 2004;80:1081–90.[Free Full Text]
  34. Warner M. A sweetener with a bad rap. The New York Times 2006;2 July. Internet: nytimes/com/2006/07/02/business/yourmoney/02syrup.html (accessed 1 December 2006).
Received for publication December 4, 2006. Accepted for publication February 21, 2007.




This article has been cited by other articles:


Home page
 Am. J. Clin. Nutr.Home page
E. R Dove, J. M Hodgson, I. B Puddey, L. J Beilin, Y. P Lee, and T. A Mori
Skim milk compared with a fruit drink acutely reduces appetite and energy intake in overweight men and women
Am. J. Clinical Nutrition, July 1, 2009; 90(1): 70 - 75.
[Abstract] [Full Text] [PDF]

Home page
 J. Nutr.Home page
J. S. White
Misconceptions about High-Fructose Corn Syrup: Is It Uniquely Responsible for Obesity, Reactive Dicarbonyl Compounds, and Advanced Glycation Endproducts?
J. Nutr., June 1, 2009; 139(6): 1219S - 1227S.
[Abstract] [Full Text] [PDF]

Home page
 J. Nutr.Home page
T. H. Moran
Fructose and Satiety
J. Nutr., June 1, 2009; 139(6): 1253S - 1256S.
[Abstract] [Full Text] [PDF]

Home page
 J. Nutr.Home page
S. T. Borra and A. Bouchoux
Effects of Science and the Media on Consumer Perceptions about Dietary Sugars
J. Nutr., June 1, 2009; 139(6): 1214S - 1218S.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
K. J Melanson, T. J Angelopoulos, V. Nguyen, L. Zukley, J. Lowndes, and J. M Rippe
High-fructose corn syrup, energy intake, and appetite regulation
Am. J. Clinical Nutrition, December 1, 2008; 88(6): 1738S - 1744S.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
J. S White
No unique role for fructose sweeteners in obesity or cardiorenal disease
Am. J. Clinical Nutrition, April 1, 2008; 87(4): 1062 - 1063.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Monsivais, P.
Right arrow Articles by Drewnowski, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Monsivais, P.
Right arrow Articles by Drewnowski, A.
Agricola
Right arrow Articles by Monsivais, P.
Right arrow Articles by Drewnowski, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS