European Journal of Clinical Nutrition (2006) 60, 567–572

& 2006 Nature Publishing Group All rights reserved 0954-3007/06 $30.00
www.nature.com/ejcn

ORIGINAL ARTICLE
Oligofructose promotes satiety in healthy human:
a pilot study
PD Cani1, E Joly1, Y Horsmans2 and NM Delzenne1

1
Unit of Pharmacokinetics, Metabolism, Nutrition and Toxicology, Department of Pharmaceutical Sciences, Université catholique de
Louvain, Brussels, Belgium and 2Unit of Gastroenterology, Univeristé catholique de Louvain, Avenue Hippocrate, Bruxelles, Belgium

Objective: The administration of a fermentable dietary fibre (oligofructose) in rats increases satietogenic gut peptides and
lowered spontaneous energy intake. The aim of the study was to assess the relevance of those effects of oligofructose on satiety
and energy intake in humans.
Design: Single-blinded, crossover, placebo-controlled design, pilot study.
Subjects: Volunteers included five men and five women aged 21–39 years, BMI ranging from 18.5 to 27.4 kg/m2, were
randomly assigned as described below.
Interventions: Subjects were included in two 2-week experimental phases during which they received either fibre (oligofructose
(OFS)) or placebo (dextrine maltose (DM)); a 2-week washout period was included between crossover phases. In total, 8 g OFS
or 8 g DM were ingested twice daily (16 g/day in total). Energy intake, hunger, satiety, fullness and prospective food
consumption were assessed with analogue scales at the end of each experimental phase.
Results: During breakfast, OFS significantly increases the satiety (P ¼ 0.04) without any difference on other sensations as
compared to DM treatment periods. After lunch, no significant differences are observed between treatment period. At dinner,
OFS significantly increases satiety (P ¼ 0.04), reduces hunger (P ¼ 0.04) and prospective food consumption (P ¼ 0.05). The
energy intake at breakfast and lunch are significantly lower (P ¼ 0.01, 0.03, respectively) after OFS treatment than after DM
treatment. Total energy intake per day is 5% lower during OFS than in DM period.
Conclusion: Oligofructose treatment increases satiety following breakfast and dinner, reduces hunger and prospective food
consumption following dinner. This pilot study presents a rationale to propose oligofructose supplements in the management of
food intake in overweight and obese patients.
Sponsorship: This project is supported by an FSR grant from the Université catholique de Louvain.
European Journal of Clinical Nutrition (2006) 60, 567–572. doi:10.1038/sj.ejcn.1602350; published online 7 December 2005

Keywords: oligofructose; fermentation; dietary fibre; GLP-1 (7–36) amide; inulin-type fructans; satiety

Introduction such as guar gum, pectin or mucilages, reduce postprandial

glycemia by delaying gastric emptying, namely through
Dietary fibres would be of particular interest regarding their
their gel-forming effect (Nuttall, 1993). But other dietary
putative role in the management of metabolic syndrome;
fibres, which do not exhibit gel-forming properties, seem
they are prone to modulate food intake, body weight,
promising in the control of food intake and metabolic
glucose homeostasis, plasma lipid profile and associated
disorders associated with glucose intolerance and obesity. It
cardiovascular diseases. (Jenkins et al., 1999; Davy and
is the case of dietary fructans, namely inulin-type fructans,
Melby, 2003; Venn and Mann, 2004). Soluble dietary fibres,
which were recently recognized as dietary fibres (Delzenne
et al., 2003; Delzenne, 2003). They are commonly found in
Correspondence: Professor N Delzenne, UCL-PMNT 7369, 73 Avenue several vegetables and cereals (onion, garlic, wheat, etc) and
Mounier, B-1200 Brussels, Belgium.
in food products in which they are added for their
E-mail: [email protected]
Guarantors: NM Delzenne and PD Cani. nutritional or organoleptic properties (fat or sugar replacer)
Contributors: PDC performed the experiments, did the statistical analysis and (Van Loo et al., 1995; Roberfroid and Delzenne, 1998).
drafted the manuscript. EJ carried out the experiments. YH took part in the Animal studies suggest that dietary consumption of oligo-
coordination of the study. NMD coordinated the study. All authors
fructose (OFS) – a short-chain fructan obtained from inulin –
contributors helped with the revision of the paper.
Received 30 December 2004; revised 14 July 2005; accepted 15 August 2005; might enhance satiety, thereby resulting in greater reduc-
published online 7 December 2005 tions in energy intake in normal, Zucker fa/fa rats and
 Oligofructose promotes satiety
PD Cani et al
568
streptozotocin-treated diabetic rats (Daubioul et al., 2002; consume the entire amount of the two bags daily. Energy
Cani et al., 2005a). Although the mechanisms that might be intake, hunger, satiety, fullness and prospective food con-
responsible for such effects are not fully understood, OFS sumption were measured at intervals during the study as
stimulates the release of a satiety-inducing gut hormone, described below. Potential adverse effects were monitored
GLP-1 (7–36) amide, and its precursor proglucagon mRNA in daily during each period of treatment (thirst, nausea,
the proximal colon of rats (Cani et al., 2004, 2005a, b). diarrhea, pain, flatulence, abdominal rumbling, gastric
Consistent with these studies, intravenous infusion of GLP-1 reflux). OFS consisted of Raftilose P95 (6.27 kJ/g) kindly
in humans enhances satiety and decreases energy intake provided by Orafti (Tienen, Belgium), which is the fully
during the period of infusion (Flint et al., 1998, 2001). One soluble and highly fermentable. Caloreens (16.7 kJ/g) (Nes-
human study demonstrates that GLP-1 plasma level signifi- tlé Clinical Nutrition, Brussels, Belgium) was used as fully
cantly increases after OFS treatment, but no study specifi- soluble, totally absorbed and nonfermentable DM.
cally analysed the putative effect of OFS on satiety and
hunger sensations in humans (Piche et al., 2003).
We therefore performed a short-term pilot study to Free-choice buffet and diets
examine whether consumption of supplements of oligofruc- Before the study, the subjects were invited to participate in a
tose would promote satiety and decrease energy intake in day of acclimatization with three free choice buffet meals
healthy subjects. (breakfast, lunch, dinner); the objective was to define clearly
the meaning of the appetite sensation scores, and to test the
feasibility and adequacy of the appetite scale. The instruc-
Subjects and methods tions for OFS or placebo ingestion (inclusion in food and/or
adequate beverage) were given to the subjects. At the end of
Subjects each 2 two-week period of OFS and DM treatment, subjects
In total, 10 healthy subjects (five men and five women) aged were invited to an excess day free-choice buffet meal
21–39 year (mean7s.e.m. 27.271.6) with mean7s.e.m. (namely, breakfast, lunch and dinner). Food and drinking
weight 67.573.8 kg, mean7s.e.m. height 17373 cm, BMI were weighed before and after meals, and caloric intake was
ranging from 18.5 to 27.4 kg/m2 (mean7s.e.m. 22.370.7) calculated. Appetite ratings were made on 100 mm visual
were recruited by local advertisement, and were free from analogue scales (VAS) with text expressing the most positive
acute and chronic diseases or use of medications that might and the negative rating anchored at each end. (A) Satiety: ‘I
influence study outcomes. A diet evaluation consisting of cannot eat another bite’; (B) Hunger: ‘I have never been
both food-frequency questionnaires and a 2-day diet record more hungry’; (C) Fullness: ‘I am totally full’; (D) Prospective
was obtained to identify and exclude individuals with a food consumption: ‘I can eat a lot’. (Raben et al., 1995). VAS
usual fibre intake 430 g/day (22 subjects participated to the were used to assess satiety, hunger, fullness and prospective
validation of food frequency questionnaire; 10 subjects were food consumption of the test meals. Sensations were
selected for the study). Throughout the study, subjects lived recorded at the beginning of each meal (time 0) and
at home and prepared their own meals, while consuming throughout the period after breakfast, lunch and dinner.
oligofructose or placebo supplement – described in the The subjects were instructed to abstain from alcohol and
protocol section – during the two 2-week experimental strenuous physical activity for the 2 days before the test days
phases. Subjects were instructed to eat until they were in order to ensure similar macronutrient balance on the test
comfortably full and to try to not gain or lose weight days. The subjects were invited to report food consumed
consciously. The study protocol was approved by the Ethical between meals. Food consumption was recorded daily
committee of Université catholique de Louvain and written during the period test (OFS and DM) by the subjects and
informed consent was obtained from each subject. analysed by our dietitian; a food frequency questionnaire
and 24 h recall were recorded on the days before test meals to
validate the self-daily food record. Reported energy, macro-
Protocol nutrients and fibre intakes were calculated using the
Subjects were randomly assigned in a single-blind, crossover, program Diet-Expert 2000, and fructans intake was calcu-
placebo-controlled design. Outpatient investigation con- lated taking into account the content reported in food stuff
sisted of two 2-week experimental phases when a fibre by Van Loo et al. (1995).
(Oligofructose-OFS) or placebo (Maltodextrin-DM) supple-
ment was consumed, separated by a 2-week washout period.
Five subjects first received the OFS supplement and five Statistical analysis
subjects received the DM supplement. Daily supplements Results are expressed as mean7s.e.m. The effects of OFS and
were divided into two portions of 8 g each to be eaten during placebo between test meals were compared by ANOVA using
breakfast and dinner. To assess the compliance, subjects kept repeated measures model with fixed factors of treatment,
a preweighed bag (8 g) of daily supplement consumption and time, treatment  time, and a random factor of patient.
returned empty bags for monitoring. They were instructed to Energy intake and macronutrients between test meals were

European Journal of Clinical Nutrition

 Oligofructose promotes satiety
PD Cani et al
569
compared using a paired t-test, SPPS 9.0.0 for Windows between men and women (10.372.4 and 770.8 g, for men
system (SPSS, Chicago IL, USA). The level of significance was and women, respectively; mean value 9.1 g71.3 g/day). The
set at Po0.05. total fibre and dietary energy and macronutrient intake
(protein, carbohydrate, fat) during 2-week periods consum-
ing OFS or DM supplements were equivalent and are
Results presented in Table 1. Compliance was excellent; minor
gastrointestinal disorders (abdominal rumbling, flatulence)
The amount of fructans (including inulin and oligofructose) were reported only on the first 3 days of oligofructose
consumed before the study was not significantly different treatment.

Table 1 Dietary energy and nutrients intakes in healthy subjects during 2-week periods corresponding to oligofructose (OFS) supplement or placebo
(DM) supplementa

OFS DM P-value

Energy, kJ/day (kcal/day) 89377920 (21357220) 94407703 (22557168) P ¼ 0.05

Protein, % energy 14.771 13.971 P40.05
Carbohydrate, % energy 58.172 58.171.5 P40.05
Fat, % energy 27.471.5 27.270.9 P40.05
Dietary fiber (g/day) 22.472.6 23.973.2 P40.05
Total fiber (g/day) (including supplements) 38.472.6 23.973.2 Po0.05
a
Results are mean7s.e.m., n ¼ 10, statistical analysis are performed through a paired Student’s t-test.

Figure 1 Breakfast visual analogue score (relative scale) (satiety, hunger, fullness and prospective food consumption) after 2 weeks of OFS
(triangle) or DM (squares) supplements in healthy subjects. The results are presented as change from baseline scores and are mean7s.e.m. for all
subjects. Visual analogue scale corresponds to (a) Satiety: ‘I cannot eat another bite’; (b) Hunger: ‘I have never been more hungry’; (c) Fullness: ‘I
am totally full’; (d) Prospective food consumption: ‘I can eat a lot’. By ANOVA: Satiety: treatment effect, P ¼ 0.045; time effect, P ¼ 0.002;
time  treatment interaction effect, P ¼ 0.18. Hunger: treatment effect, P ¼ 0.88; time effect, P ¼ 0.001; time  treatment interaction effect,
P ¼ 0.68, Fullness: treatment effect, P ¼ 0.58; time effect, P ¼ 0.05; time  treatment interaction effect, P ¼ 0.3. Prospective food consumption:
treatment effect, P ¼ 0.23; time effect, P ¼ 0.17; time  treatment interaction effect, P ¼ 0.49.

European Journal of Clinical Nutrition

 Oligofructose promotes satiety
PD Cani et al
570
Appetite scores Discussion
After breakfast, OFS significantly increases satiety (P ¼ 0.04),
without any significant difference in hunger, fullness and Previous research has suggested that increasing total fibre
prospective food consumption as compared to that in the intake could help to reduce energy intake by decreasing
DM treatment period (Figure 1). After lunch, no significant hunger and/or increasing satiety and be helpful in the
differences were observed between OFS and DM treatment management of type 2 diabetes (Howarth et al., 2001; Venn
period (Figure 2). After dinner, OFS significantly increases and Mann, 2004). Dietary fibres have thus been proposed
and maintains a higher satiety (P ¼ 0.04), and OFS treatment as key food components taken into account in the control
reduces hunger (P ¼ 0.04) and prospective food consumption of the current high prevalence of obesity and overweight.
(P ¼ 0.05) (Figure 3). Nowadays, there is little information on the putative effects
of fermentable fibres on food intake and energy regulation.
In the present study, oligofructose or maltodextrin (as
Energy Intake during ad libitum free choice buffet placebo) were given twice daily, 8 g at breakfast and 8 g at
The energy intake at breakfast and lunch, expressed as % dinner. The observation of food intake related sensations and
from energy intake observed at DM period buffet, were behaviour was measured for several hours following break-
significantly lower (by about 10%) (P ¼ 0.01 and 0.03, fast, lunch and dinner proposed as controlled buffet.
respectively) after OFS treatment as compared to that after Tolerance towards oligofructose was good and explains the
DM treatment (Table 2). At dinner, there was no difference in excellent compliance observed. Interestingly, we found that
the energy intake between OFS and DM. The total energy oligofructose promotes satiety following breakfast and
intake during the 24-h period of the buffet was significantly dinner, and reduces hunger and prospective food consump-
lower corresponding to 95% (P ¼ 0.05) of the total energy tion after the dinner. During oligofructose feeding, breakfast,
intake recorded during DM treatment period (Table 2). lunch and total energy intake were moderately (by about

Figure 2 Lunch visual analogue score (relative scale) (satiety, hunger, fullness and prospective food consumption) after 2 weeks of OFS
(triangle) or DM (squares) supplements in healthy subjects. The results are presented as change from baseline scores and are mean7s.e.m. for all
subjects. Visual analogue scale corresponds to (a) Satiety: ‘I cannot eat another bite’; (b) Hunger: ‘I have never been more hungry’; (c) Fullness: ‘I
am totally full’; (d) Prospective food consumption: ‘I can eat a lot’. By ANOVA: Satiety: treatment effect, P ¼ 0.92; time effect, P ¼ 0.25;
time  treatment interaction effect, P ¼ 0.36. Hunger: treatment effect, P ¼ 0.84; time effect, P ¼ 0.83; time  treatment interaction effect,
P ¼ 0.44. Fullness: treatment effect, P ¼ 0.98; time effect, P ¼ 0.77; time  treatment interaction effect, P ¼ 0.78. Prospective food consumption:
treatment effect, P ¼ 0.35; time effect, P ¼ 0.64; time  treatment interaction effect, P ¼ 0.88.

European Journal of Clinical Nutrition

 Oligofructose promotes satiety
PD Cani et al
571

Figure 3 Dinner visual analogue score (relative scale) (satiety, hunger, fullness and prospective food consumption) after 2 weeks of OFS
(triangle) or DM (squares) supplements in healthy subjects. The results are presented as change from baseline scores and are mean7s.e.m. for all
subjects. Visual analogue scale corresponds to (a) Satiety: ‘I cannot eat another bite’; (b) Hunger: ‘I have never been more hungry’; (c) Fullness:
‘I am totally full’; (d) Prospective food consumption: ‘I can eat a lot’. By ANOVA: Satiety: treatment effect, P ¼ 0.04; time effect, Po0.001;
time  treatment interaction effect, P ¼ 0.03. Hunger: treatment effect, P ¼ 0.04; time effect, P ¼ 0.001; time  treatment interaction effect,
P ¼ 0.06. Fullness: treatment effect, P ¼ 0.62; time effect, P ¼ 0.004; time  treatment interaction effect, P ¼ 0.21. Prospective food consumption:
treatment effect, P ¼ 0.05; time effect, P ¼ 0.001; time  treatment interaction effect, P ¼ 0.02.

Table 2 Dietary energy intake in healthy subjects during 2-week interesting effects on fat mass development, steatosis and
periods corresponding to oligofructose (OFS) supplement or placebo
hyperglycemia – namely through the promotion of intest-
(DM) supplement a
inal synthesis and portal release of GLP-1 (7–36) amide (Cani
Meals DM OFS P-value et al., 2004, 2005a, b). GLP-1 (7–36) amide is a satiety
hormone causing weight loss in humans when administered
Breakfast (% from DM) 100 9173.3 Po0.01
exogenously at levels ranging from physio to supraphysio-
Lunch (% from DM) 100 89.573 Po0.05
Dinner (% from DM) 100 9576.8 P40.05 logic doses (Flint et al., 1998, 2001; Verdich et al., 2001). An
Total energy intake (% from DM) 100 94.671.8 Po0.05 increase in serum GLP-1 (7–36) amide by oligofructose has
a
been reported in one interventional study performed in
Values are mean7s.e.m., n ¼ 10. Statistical analysis is performed through a
paired Student’s t-test.
patients presenting gastric reflux, but this results has not
been related to food intake and satiety (Piche et al., 2003).
The authors suggest that the ‘kinetics’ of fermentation –
5–10%), but significantly lower than those observed during assessed by hydrogen breath test – is important to take into
DM period. The significant increase of satiety observed account when assessing the influence of fermented nutrients
following breakfast could explain the decrease of energy on circulating peptides. The increase in expired hydrogen
intake during the following meal, that is lunch. In contrast, (marker of fermentation) correlates with the modulation of
this observation, the similar satiety observed following lunch serum GLP-1 (7–36) amide level, which could explain the
could predict the same energy intake during the following link between intestinal fermentation and this peptide
meal, that is dinner. Two questions may arise from this secretion. Thus, on the basis of these results, it is reasonable
study: (1) by which mechanism could oligofructose mod- to suggest a role of oligofructose in enhancing satiety and
ulate food intake behaviour? (2) Could this effect also be reducing energy intake in humans consuming a diet ad
relevant for other fermentable dietary fibres? In rats, libitum. Recently, Archer et al. (2004) have demonstrated that
oligofructose supplementation decreases food intake – with another fermentable fructans dietary fibre – inulin – added in

European Journal of Clinical Nutrition

 Oligofructose promotes satiety
PD Cani et al
572
food as fat-replacer was able to induce a lower energy intake Cani PD, Daubioul CA, Reusens B, Remacle C, Catillon G, Delzenne
during a test day, despite no effect on satiety at breakfast, NM (2005a). Involvement of endogenous glucagon-like peptide-
1(7–36) amide on glycaemia-lowering effect of oligofructose in
suggesting, as mentioned by the authors, a late postabsorp-
streptozotocin-treated rats. J Endocrinol 185, 457–465.
tive satiety trigger related to the complete fermentation of Cani PD, Dewever C, Delzenne NM (2004). Inulin-type fructans
this fibre (Archer et al., 2004). Nevertheless, all fermentable modulate gastro-intestinal peptides involved in appetite regula-
dietary fibres do not have the same potency to increase tion – Glucagon-like peptide-1 and Ghrelin – in rats. Br J Nutr 92,
521–526.
satietogenic peptides, at least in rats: long-chain inulin,
Cani PD, Neyrinck AM, Maton N, Delzenne NM (2005b). Oligo-
which is largely fermented in the distal colon, whereas fructose promotes satiety in rats fed a high-fat diet: involvement
oligofructose is fermented in the proximal colon, was not of glucagon-like peptide-1. Obes Res 13, 1000–1007.
able to produce the same effects of oligofructose in terms of Daubioul C, Rousseau N, Demeure R, Gallez B, Taper H, Declerck B
et al. (2002). Dietary fructans, but not cellulose, decrease
GLP-1 (7–36) amide and proglucagon mRNA modulation
triglyceride accumulation in the liver of obese Zucker fa/fa rats.
(Cani et al., 2004). Besides, other authors have compared the J Nutr 132, 967–973.
putative effects of fermentable fibres (pectin and b-glucan, Davy BM, Melby CL (2003). The effect of fiber-rich carbohydrates on
ratio 2:1) and nonfermentable fibres (hydroxypropyl methyl- features of Syndrome X. J Am Diet Assoc 103, 86–96.
Delzenne N, Cherbut C, Neyrinck A (2003). Prebiotics: actual and
cellulose) on satiety, hunger and body weight and have
potential effects in inflammatory and malignant colonic diseases.
found no effect of these two fibres (Howarth et al., 2003). The Curr Opin Clin Nutr Metab Care 6, 581–586.
last observation leads us to think that the place (proximal or Delzenne NM (2003). Oligosaccharides: state of the art. Proc Nutr Soc
distal colon) and the pattern of fermentation (in terms of 62, 177–182.
Flint A, Raben A, Astrup A, Holst JJ (1998). Glucagon-like peptide 1
short-chain fatty acids (SCFA) production) (Van Loo et al.,
promotes satiety and suppresses energy intake in humans. J Clin
1999) of fermentable fibre would be important, but this Invest 101, 515–520.
remains speculative. Flint A, Raben A, Ersboll AK, Holst JJ, Astrup A (2001). The effect of
In conclusion, 2 weeks of oligofructose treatment increases physiological levels of glucagon-like peptide-1 on appetite, gastric
emptying, energy and substrate metabolism in obesity. Int J Obes
satiety following breakfast and dinner, reduces hunger and
Relat Metab Disord 25, 781–792.
prospective food consumption following dinner. Breakfast, Howarth NC, Saltzman E, Roberts SB (2001). Dietary fiber and weight
lunch and total energy intake were significantly reduced as regulation. Nutr Rev 59, 129–139.
compared to that in the DM treatment. These results suggest Howarth NC, Saltzman E, McCrory MA, Greenberg AS, Dwyer J,
Ausman L et al. (2003). Fermentable and nonfermentable fiber
a role of oligofructose in promoting a moderate negative
supplements did not alter hunger, satiety or body weight in a pilot
energy balance in humans consuming a diet ad libitum. study of men and women consuming self-selected diets. J Nutr
These results remain to be confirmed in a longer period of 133, 3141–3144.
time and assessed in obese and type 2 diabetes patients. Jenkins DJ, Kendall CW, Vuksan V (1999). Inulin, oligofructose and
intestinal function. J Nutr 129, S1431–S1433.
Nuttall FQ (1993). Dietary fiber in the management of diabetes.
Diabetes Rev 42, 503–508.
Acknowledgements Piche T, des Varannes SB, Sacher-Huvelin S, Holst JJ, Cuber JC,
Galmiche JP (2003). Colonic fermentation influences lower
esophageal sphincter function in gastroesophageal reflux disease.
We thank all the subjects who participated in this study, as Gastroenterology 124, 894–902.
well as the Paul Lambin Institute, Haute Ecole Leonard de Raben A, Tagliabue A, Astrup A (1995). The reproducibility of
Vinci (dietetic section) for their help in elaborating the test subjective appetite scores. Br J Nutr 73, 517–530.
Roberfroid MB, Delzenne NM (1998). Dietary fructans. Annu Rev Nutr
meals. This work was supported by an FSR grant from the
18, 117–143.
Université catholique de Louvain. PDC is teaching assistant, Van Loo J, Coussement P, de Leenheer L, Hoebregs H, Smits G (1995).
Université catholique de Louvain, Brussels, Belgium. On the presence of inulin and oligofructose as natural ingredients
in the western diet. Crit Rev Food Sci Nutr 35, 525–552.
Van Loo J, Franck A, Roberfroid M (1999). Functional food properties
of non-digestible oligosaccharides. Br J Nutr 82, 329.
References Venn BJ, Mann JI (2004). Cereal grains, legumes and diabetes. Eur J
Clin Nutr 58, 1443–1461.
Archer BJ, Johnson SK, Devereux HM, Baxter AL (2004). Effect of fat Verdich C, Flint A, Gutzwiller JP, Naslund E, Beglinger C, Hellstrom
replacement by inulin or lupin-kernel fibre on sausage patty PM et al. (2001). A meta-analysis of the effect of glucagon-like
acceptability, post-meal perceptions of satiety and food intake in peptide-1 (7–36) amide on ad libitum energy intake in humans.
men. Br J Nutr 91, 591–599. J Clin Endocrinol Metab 86, 4382–4389.

European Journal of Clinical Nutrition