Effect of Psyllium on Gastric Emptying, Hunger Feeling and Food Intake in Normal Volunteers: a Double Blind Study

Effect of psyllium on gastric emptying, hunger feeling and food intake in normal volunteers: a double blind study

D Rigaud1, F Paycha2, A Meulemans3, M Merrouche1 and M Mignon1

1Service de GastroenteÂrologie et Nutrition, HoÃpital Bichat, 75877 Paris Cedex 18, France, 2MeÂdecine NucleÂaire, HoÃpital Louis Mourier, 92300 Colombes, and 3Laboratoire de MeÂdecine NucleÂaire, FaculteÂ X. Bichat, 75018 Paris, France

Objective: To assess whether psyllium, a soluble dietary ®bre, could, at an acceptable dose (7.4 g), delay gastric emptying of a low-calorie meal, and reduce hunger feeling and energy intake, without requiring intimate mixing with the meal. Design: A double blind randomized cross over study with 14 normal volunteers, to evaluate the effect of this dose of psyllium on postprandial serum glucose, triglycerides and insulin levels, and on gastric fullness, hunger feeling and food intake. Methods: Gastric emptying was measured using a standard double-radiolabeled 450 kcal meal and feelings by visual analogic scales. The postprandial serum glucose, triglycerides and insulin levels were also determined. Results: No delay in the gastric emptying of the solid and liquid phases of the meal was observed with psyllium. After the meal, hunger feelings and energy intake were signi®cantly lower during the psyllium session than during the placebo session (13% and 17% lower respectively; P < 0.05). Postprandial increase in serum glucose, triglycerides and insulin levels was less with psyllium than with placebo (P < 0.05). Conclusions: Psyllium reduces hunger feelings and energy intake in normal volunteers at reasonable dose and without requiring mixing with the meal. It does not act by slowing down the gastric emptying of hydrosoluble nutrients, but by increase in the time allowed for intestinal absorption, as suggested by the ¯attening of the postprandial serum glucose, insulin and triglycerides curves. Sponsorship: University, FaculteÂ de MeÂdecine X. Bichat, 75018 Paris, France. Descriptors dietary ®bre; soluble ®bre; psyllium; gastric emptying; hunger feeling; appetite control; satiety; food intake

Introduction most important factors of the in¯ux into the body. The postprandial reduction in the blood glucose increment In Western countries, a low dietary ®bre intake is asso- following the addition of soluble ®bre has been shown to ciated with some common diseases: chronic constipation, be dependent upon the ability of the soluble ®bre to obesity, dyslipoproteinemia, diabetes mellitus and coronary develop viscosity in water and upon intimate mixing to heart disease (Anderson et al, 1994; Glore et al, 1994; the meal (Jenkins et al, 1978). Few studies have dealt with Rigaud & Royer, 1988). A bene®cial effect of dietary ®bre the gastric emptying time of a glucose solution (Blackburn supplementation is likely for some of them (Enzi et al, et al, 1984; Ralphs & Lawaetz, 1978; Rydning et al, 1982) 1980; Frati-Munari et al, 1983). This may be due to the or a mixed meal (Grimes & Goddard 1977; Schwartz et al, high carbohydrate intake observed in people being on high 1982): they suggest that hydrophilic dietary ®bres are more dietary ®bre intake, as compared with the high energy and potent than insoluble dietary ®bres (Grimes & Goddard fat intake linked to the low dietary ®bre intake (Heaton, 1977). None of these studies was double blind. One of them 1993). It seems that high-®bre diet or dietary ®bre supple- (Blackburn et al, 1984) did not ®nd any relationship mentation are associated with lower energy intake and between gastric emptying rate and reduction in the post- lower energy bioavailability, as compared with occidental prandial glycemic response. diet (Burley et al, 1989; Forman et al, 1968; Gustafsson et Psyllium is a soluble dietary ®bre extracted from Plan- al, 1994; Jenkins et al, 1994). This effect could be related tago ovata. The dried ripe seeds of Plantago ovata contain to a decrease in hunger feelings due to a delay in nutrient mucilage in their outer epidermitis and swell in contact of absorption. Previous studies suggest that addition of dietary moisture. The mucilaginous portion has been used as bulk ®bre to a test meal could lower glucose and insulin laxative in the treatment of constipation. The hydrophilic responses (Enzi et al, 1980; Gustafsson et al, 1994; Jenkins properties of psyllium (each gram of psyllium retains about et al, 1994; Holt et al, 1979). Soluble dietary ®bres appear 10 g water) leads to the formation of a viscous gel. Thus more ef®cient than insoluble ones (Rigaud & Royer, 1988; psyllium could be able to pick up hydrophilic molecules Jarjis et al, 1984; Jenkins et al, 1978). such as simple carbohydrates (namely disaccharides) and The delay in nutrient absorption may be due to delay in proteins (Rigaud & Royer, 1988; Enzi et al, 1980; Frati- gastric emptying. Indeed gastric emptying is one of the Munari et al, 1983; Heaton, 1993). Psyllium could thus retain in the stomach such energetic molecules, as suggested for pectin or ispaghula (Holt et al, 1979; Jarjis et al, 1984; Blackburn et al, 1984; Ralphs & Lawaetz, 1978; Correspondence: D Rigaud. Received 7 February 1997; revised 25 April 1997; accepted 18 August Schwartz et al, 1982). This may also be true for the 1997 emulsion phase of the lipid part of the meal. The nutrient Inhibition of food intake by psyllium D Rigaud et al 240 ¯ux in the small bowel could thus be delayed and laid out. given to each subject to earn how to use it. Then the Moreover high viscosity soluble ®bres also increase the subjects reported to the Nuclear Department. The gastric unstirred water layer in the small intestine. emptying study began at 11.30 am. The subjects ingested This psyllium-related delay in the intestinal absorption two doses of either psyllium (Techni-Pharma, Monaco, of energy could modify food intake in different ways: (a) France) or placebo, namely 7.4 g psyllium or 6.9 g pla- by decreasing hunger feeling; (b) by inducing abdominal cebo, 15 min before the test meal, in a cross over double discomfort; (c) by increasing satiety. blind fashion. This was made to be sure that mental factors These effects may be useful in the treatment of obesity did not alter the rate of gastric emptying or hunger fellings. during programs including low-calorie meals (Enzi et al, Placebo and psyllium were contained in identical bags. The 1980; Rigaud et al, 1990; Vido et al, 1993). But one placebo consisted of gelatin and sweetener in order to be important condition is that the dietary ®bre could be similar in taste and look as psyllium after dilution in administered before meal, without requiring intimate 100 mL water. Psyllium or placebo were ingested just mixing with the meal. We thus studied the effects of after dilution and not mixed with any dishes of the meal. psyllium, administered before meal, on gastric emptying The water-holding capacity of our psyllium preparation of a low-calorie meal, gastric fullness, hunger feelings and (Techni-Pharma, Monaco, France) was 25 mL water=g food intake. This study was conducted in a double blind psyllium; its viscosity at 37C was 56 500.00 centipoises and cross over fashion, vs placebo in normal volunteers. at 5 min and 69 500.00 centipoises at 10 min. The placebo Psyllium and placebo were not mixed with the dishes of the solution had a water holding capacity of 4 mL water=g; its meal, but administered 15 min before the meal. viscosity at 37C was 2300.00 centipoises at 10 min.

Subjects and methods Gastric emptying study: After 15 min, the test-meal was offered. It had to be ingested in about 15 min. It consisted Subjects of one omelet which contain two eggs, 70 g white bread, Fourteen healthy subjects of normal weight were enrolled 10 g butter, 220 mL water. It contained 450 kcal (1.88 MJ) in the study (Table 1). Criteria for inclusion were the as follows: proteins: 21 g; carbohydrates: 40 g; lipids: following: a body mass index (BMI) between 19 and 23 g; total volume: 400 mL (1.1 kcal=mL). 25 kg=(m)2; and aged 18±50 y. Criteria for exclusion were the following: unstable body weight (namely weight variations higher than 3 kg in the Gastric emptying measurement procedures: The test meal last two months); any restrictive diet in the last two months; was a doubly labeled solid±liquid meal (Rigaud et al, any medication likely to modify food intake, gastric emp- 1988). The solid phase consisted of a two-egg omelet tying or eating behavior; consumption of more than 10 labeled by 37 MBq (1 mCi) of 99mtechnetium sulfur col- cigarettes or more than 20 g of alcohol per day, dyspepsia. loid. Approximately 5.5 MBq (150 mCi) of 111indium- All subjects gave informed consent to the study, the DTPA (diethylene triamine pentaacetic acid) was mixed design of which having been approved by the ethical into 220 mL tap water by way of liquid phase. The time committee of Bichat Hospital. required to consume the meal did not exceed 15 min. A large ®eld-of-view single headed gamma camera equipped Methods with medium energy collimator and interfaced to a com- This was a prospective randomized double blind cross over puter was used for data acquisitions. Solid and liquid study against placebo. Two weeks before the experimental phases data were recorded by setting the pulse height sessions, the subjects ingested the 440 kcal radio-labeled analyzer on the 140 keV photopeak of 99mtechnetium and meal in order to be sure that they had no alteration of the the 245 keV photopeak of 111indium, with 15 and 20% gastric emptying (this session is called the `no preload energy windows. Successive 1 min anterior and 1 min session'). Then there were two sessions, one each week, in posterior images were acquired at 10 min intervals with a randomized Latin squared design: placebo and psyllium the subject in an upright position. External 57cobalt point sessions, in which the placebo or psyllium administration sources were taped on the anterior and posterior thorax and was followed by the test meal 15 min later. abdomen in order to allow accurate repositioning of the subject between image acquisition. The image matrix was The sessions 64664 by 16 bits deep. A region of interest was drawn by The day before each session, the subjects reported their hand on the computer display, in an attempt to include the evening food intake. Only two subjects were studied per whole stomach while excluding the small intestine. The day, one receiving the psyllium and the other the placebo. data were corrected for radioactive decay, radionuclide At 07.00 am, the subjects had a small breakfast consisting crosstalk and septal penetration. We calculated the geo- of coffee or tea, 5 or 10 g sugar, 30 g white bread and 20 g metric mean of the anterior and posterior images count jam; no butter was allowed. At 11.00 am, the subjects met rates. The study lasted 200 min. the Nutrition department staff, where the details of the The fractional meal retention time was obtained by study were given. A notebook for recording hunger feel- dividing the geometric mean counts at each time interval ings, epigastric fullness and reporting all food items were by the maximum (zero time) value. Zero time was de®ned as the time of meal completion. Time activity curves were thus derived for both phases. From the curves, several Table 1 Characteristics of the healthy subjects parameters were obtained for statistical analysis: the lag phase (namely the time at which a 2% decrease from the Sex (M=F) 7=7 1 maximal point was measured), the emptying half-time (T2; Age (y) 21 Æ 2 the time elapsed until 50% of the label has left the area of Body weight (kg) 56 Æ 9 Body mass index (kg=m2) 20.9 Æ 2.1 interest, namely the stomach) and the area under the curve (AUC; namely the area between the curve of decay in Inhibition of food intake by psyllium D Rigaud et al 241 radioactivity and the basal (zero) line). For AUC, which corresponds to the mean transit time, the maximal value was 200 (no emptying). The procedure has been described in detail elsewhere (Rigaud et al, 1988).

Hunger feeling, epigastric fullness and food intake after the test-meal: Before 5, 15, 30, 45 min after the test-meal, then every 30 min during 6 h, the subjects recorded their hunger feeling and epigastric fullness on visual analogic scales (VAS) of 100 mm, without partition, with at the left end `not at all' and at the other end `very much'. During this part of the study, they were allowed to sit down, walk, but not to go out, of the laboratory. At the end of the gastric emptying measurements, the subjects were free to go and to eat what they wanted, without restriction. There were invited to do their usual activities. No food items were provided to them. All the Figure 1 Gastric emptying curves of the liquid phase of the meal for psyllium and placebo session Mean Æ s.d. White squares: psyllium; black food items consumed from the test meal to bedtime points: placebo; black triangles: no pre-load. The curves obtained with (snacks, nibbling and dinner) were weighted and recorded psyllium and placebo were signi®cantly different from that of the no pre- in diaries. One of us checked the diaries with the subjects. load session (P < 0.05). The psyllium curve did not differ from the placebo Energy and nutrient inputs were determined using a com- one. puterized food table for PC (CIQUAL 1991 for Windows). The morning following each session, the subjects were asked to say whether they thought they had taken psyllium or placebo before the test-meal.

Second experiment: serum glucose, insulin and triglycerides levels after the test meal: In order not to modify subjective responses to the meal, determination of serum glucose, insulin and triglycerides levels was done in a second set of experiments. In a double blind and cross- over fashion, as described before, the same 450 kcal meal was given to the same subjects 15 min after administration of psyllium or placebo. Before and 15, 30, 45, 60, 120, 180, 240 min after the test-meal, blood samples were drawn through an indwelling catheter for determination of serum glucose, insulin and triglycerides levels. Serum glucose was measured by the glucose-oxydase method, insulin by radioimmunoassay and serum triglycerides by the enzy- Figure 2 Gastric emptying curves of the solid phase of the meal for matic method using standard kit (Biotrol Inc., Paris, psyllium and placebo session. Mean Æ s.d. White squares: psyllium; black France). points: placebo; black triangles: no pre-load. The three curves were not signi®cantly different. Statistical analyses The results are expressed as mean Æ s.d. The integrated 1 obtained after placebo administration (Table 2). The T2 decrease in hunger feeling and the integrated epigastric of the liquid phase during both the psyllium and the fullness were calculated as the sum of all records minus the placebo sessions was higher than that obtained during the basal value (namely the pre-psyllium or the pre-placebo no-preload `inclusion' session (Table 2). value). Comparisons between `psyllium' and `placebo' sessions were made using variance analysis. The correla- Epigastric fullness, hunger feelings and energy intake tions between gastric emptying data and hunger feeling For the psyllium session, the epigastric fullness experi- or epigastric fullness at a given time were made using enced before the test meal (10 min after psyllium ingestion) variance analysis. A P-value < 0.05 was considered as as well as the one experienced during or after the test meal signi®cant. were not signi®cantly different from those observed during the placebo session. Considering all measures until dinner, Results the hunger feelings during the psyllium session was lower than that observed after placebo administration (P < 0.05 Gastric emptying by ANOVA with repeated measures; Figure 3). The cumu- The lag phase time (for solids) was similar for the two lated hunger feeling of the psyllium session was also sessions (psyllium or placebo), namely 7.3 Æ 3.6 min (psy- signi®cantly lower (713%) than that observed after llium) and 5.1 Æ 3.4 min (placebo). For both the liquid and placebo administration (P < 0.05 by ANOVA; Figure 4). the solid phases, the pro®les of the gastric emptying curves The total energy intake following the test meal was of the psyllium and the placebo sessions were very similar signi®cantly lower (17% after psyllium administration 1 (Figures 1 and 2). The T2 and the area under the curve of than after placebo (Figure 5). This lower energy intake the solid and liquid phases of the test-meal after psyllium was related to a 40% lower energy intake during snacks, administration did not signi®cantly differ from those while energy intake during dinner did not differ between Inhibition of food intake by psyllium D Rigaud et al 242 Table 2 Gastric emptying of the solid and liquid phases of the meal in 14 healthy subjects after psyllium or placebo administration, as compared to a session without pre-charge.

Placebo Psyllium No pre-load

Lag phase (min) 5.1 Æ 3.2 7.3 Æ 3.6 4.8 Æ 2.7 (0.85) (0.96) (0.72) 1 T2 solids (min) 88.7 Æ 24 88.4 Æ 22 93.1 Æ 21 (6.4) (5.88) (5.61) 1 T2 liquids (min) 67.0 Æ 22 67.5 Æ 17 47.3 Æ 11* (5.87) (4.54) (2.94) AUC solids (min) 91.6 Æ 19 92.2 Æ 15 93.3 Æ 16 (5.08) (4.02) (4.28) AUC liquids (min) 94.3 Æ 13 93.9 Æ 15 62.0 Æ 9* (3.47) (4.01) (2.40) Figure 5 Snacks, dinner and total energy intake following the test meal Mean Æ s.d.; s.e. in brackets. for the psyllium and the placebo session. The subjects were allowed to AUC area under the curve. have snacks from 16.00 pm to dinner. Mean Æ s.d. White columns: * p < 0.05 vs the two other sessions. placebo; black columns: psyllium. ** P < 0.02 for snacks and P < 0.05 None of the values obtained under psyllium was signi®cantly different for total energy intake, when psyllium and placebo session were compared. from those obtained with placebo.

the sessions (8% lower, but not signi®cantly, after psyllium than after placebo). The total macronutrient intake, namely fats, was lower but not signi®cantly after psyllium administration than after placebo (data not shown). The energy supply by fats and carbohydrates during snacks was lower for psyllium than for placebo (P < 0.02): 9 Æ 5g vs 15 Æ 6 g for fats and 27 Æ 11 vs 42 Æ 15 g for carbohydrates. Since the food items consumed during the tests were not de®ned, we are unable to settle whether the difference between the two sessions was related to changes in the type or in the quantity of foods selected. It seems that this difference was related both to the quantity and to the quality of foods. There was no correlation between energy intake or hunger feeling following the test meal and any parameter of the gastric emptying (solid and liquid phases). The day after the sessions, six subjects gave the right sequence of treatment and 8 failed to recognize it (NS).

Figure 3 Hunger feeling curves after psyllium or placebo administration. Serum glucose, insulin and triglycerides levels after the test Mean Æ s.d. White dotted line and white circles: placebo; solid line and meal black circles: psyllium. *P < 0.05 by ANOVA with repeated measures. After psyllium administration, there was a delayed and After 200 min, the hunger feeling could be affected by food intake, since small postprandial rise in serum glucose, as compared to the subjects were allowed to eat what they wanted. those observed after placebo administration (Figure 6; P < 0.05). As expressed as an integrated positive area under the curve, the glycemic response was 1.68 Æ 0.64 and 3.42 Æ 0.48 mmol h L71 for psyllium and placebo respectively (Figure 6; P < 0.02). Similarly, delayed and smaller postprandial rise in serum insulin level was observed during the psyllium session. The 300 min integrated insulin response to the meal was also signi®cantly lower than that observed after placebo administration: 80 Æ 21 mU h L71 for psyllium and 128 Æ 29 mU h L71 for placebo (P < 0.05). During the psyllium session, the triglycerides level did not signi®cantly increase after the test meal, as compared to the pre-meal level, while a little but signi®cant rise in triglycerides was observed during the placebo session (Figure 6; P < 0.05).

Discussion Figure 4 Cumulated hunger feeling after psyllium or placebo adminis- After administration of 7.4 g of psyllium 15 min before a tration. Mean Æ s.d. Grey columns: placebo; black columns: psyllium. * P < 0.05 by ANOVA, when psyllium and placebo session were com- low-calories meal (450 kcal), there was a 13% signi®cant pared. decrease in hunger feelings and a 17% lower energy intake Inhibition of food intake by psyllium D Rigaud et al 243 Some authors studied the relationship between hunger feelings and dietary ®bres (Burley, 1992; Burley et al, 1993; de Graaf et al, 1992; Di Lorenzo et al, 1988; French & Read 1994; Gustafsson et al, 1993; Gustafsson et al, 1994; Gustafsson et al, 1995; Raben et al, 1994; Rigaud et al, 1990; Vido et al, 1993). But there is very little information on the effects of a psyllium- or of another soluble dietary ®bre preparations on hunger feelings before and after a test meal in the literature (Di Lorenzo et al, 1988; French & Read, 1994). During a six month randomized double blind placebo-controlled study in overweight volunteers, we found that an insoluble dietary ®bre was able to reduce hunger feeling for long periods (Rigaud et al, 1990). Di Lorenzo et al, (1988) found that 15 g pectin Figure 6 Serum glucose and triglycerides levels after the test meal. Mean Æ s.d. Black circles, dotted lines: placebo; white circles, solid lines: delayed gastric emptying and hunger feeling in nine adult psyllium. * P < 0.05 by ANOVA with repeated measures, when psyllium obese subjects in a double blind cross over study. Burley et and placebo session were compared. al (1993) showed a signi®cant postingestive effect of dietary ®bre (a sugar beet-®bre supplemented breakfast) on satiety. De Graaf et al (1992) observed that the satieting effect of high-®bre or high-protein spreads were stronger during the hours following the test meal. This was shown that those of high-sweet carbohydrate or high-fat spreads in while psyllium was administered before and not mixed to 101 healthy subjects. In 10 healthy subjects, Raben et al the meal. This was associated with smaller insulin response (1994) demonstrated decreased postprandial thermogenesis and lower glucose and triglycerides levels. These effects and lower fat oxidation, together with higher fullness and seemed not to be related to differences in gastric emptying lower desire to eat after a high-®bre meal compared to a of solids or liquids, indeed we did not observe any low-®bre meal. Gustafsson et al evaluated the effect of difference between the placebo and the psyllium session different types (spinach, carrots, peas, Brussels sprouts) and for any parameter of the gastric emptying. amounts (96±250 g) of high-®bre foods on plasma glucose The psyllium is a soluble dietary ®bre. Its usefulness in and insulin response and on satiety after a mixed meal in metabolic diseases such as diabetes or hypercholesterole- healthy male volunteers (Gustafsson et al, 1993, 1994, mia has been suggested by many studies. In fact, it could 1995b). They found that high amount of spinach and raw improve the control of hyperglycemia in insulin and non- carrots induced signi®cantly lower postprandial glucose, insulin dependent diabetes (Frati-Munari et al, 1983) and insulin response and higher postprandial satiety than the could reduce serum cholesterol level by about 10±15% in control mixed meal. moderate hypercholesterolemia (Frati-Munari et al, 1983; The effect of soluble dietary ®bres on energy intake has Glore et al, 1994). One of the mechanisms involved in been rarely investigated. Some studies were performed with these metabolic effects could be the lowering in gastric other types of soluble or insoluble ®bres. Di Lorenzo et al emptying and thus in intestinal transport of the energetic (1988) found that the addition of 15 g pectin to the meal substrates. Several studies examined the role of dietary signi®cantly prolonged the time to the next meal from 6.0 ®bres in the emptying of different meals (Blackburn et al, to 10.6 h. Tomlin (1995) studied the effects of a soluble 1984; Ralphs & Lawaetz, 1978; Rydning et al, 1982; dietary ®bre in 17 healthy overweight subjects. He Grimes & Goddard 1977; Schwartz et al, 1982). Some observed that this novel liquid ®bre reduced the hunger authors observed that soluble ®bres as pectin, guar gum or feeling and the amount of food desired, increased fullness ispaghula were able to delay the emptying of a liquid or a and signi®cantly delayed the next food intake. However, mixed meal (Blackburn et al, 1984; Holt et al, 1979; although decreased, the amount of energy consumed later Ralphs & Lawaetz, 1978; Schwartz et al, 1982); other in the day and during the following day was not signi®- authors were unable to show any differences between cantly different. dietary ®bre and control diet, except at very high doses The decrease in hunger feelings and in energy intake (Rydning et al, 1982). This may be related to the viscosity that we observed could not be explained with certainty: of the dietary ®bre at the dose used, since not all soluble ®bres are viscous at low concentration. (1) We did not ®nd, contrarily to other authors for other Taken altogether these studies clearly suggest that solu- ®bres, that psyllium was retained in the gastric lumen ble dietary ®bres could delay the gastric emptying of liquid and slow down gastric output of the hydrosoluble hydrosoluble phase (water, energy-containing liquids such nutrients, as suggested for pectin, guar gum or ispa- as glucose or amino acids solutions) but not the emptying ghula (Blackburn et al, 1984; Holt et al, 1979; Ralphs of most of the solid phase (digestible and non digestible & Lawaetz, 1978; Rydning et al, 1982; Schwartz et al, solids, solid lipids fats). No information was available 1982). concerning the lipid phase (emulsi®ed and non-emulsi®ed (2) With the dose of psyllium we used, we did not ®nd any fats). In the present study, the reason for the lack of effect difference with placebo concerning abdominal feelings. of psyllium on the gastric emptying rate of the hydrosoluble Thus we can exclude that epigastric fullness or dis- phase could be related to the dose used, indeed pectin has comfort of the subjects after psyllium ingestion been shown to delay selectively the emptying of liquids, explained the decrease in hunger feelings and food when a dosage higher than 10 g, namely of high viscosity intake, as for high-®bre meal (Raben, 1994). ®bres, was used (Blackburn et al, 1984; Holt et al, 1979; (3) Another possibility is that psyllium retains alimentary Ralphs & Lawaetz, 1978; Rydning et al, 1982; Schwartz et fats into the gastric lumen by modifying motor activity al, 1982). No information was available for psyllium. of the gastric muscle (Jenkins et al, 1978). Fat content Inhibition of food intake by psyllium D Rigaud et al 244 is able to modify the effect of dietary ®bre on hunger constipation, two factors of poor patient compliance and (Low et al, 1988; Cummings et al, 1979). Unfortu- poor effectiveness. nately, we did not study the gastric emptying time of fats, because of the inaccuracy of the methods using labeled fat in mixed meal. References (4) A explanation might be that viscous gels and soluble Anderson JW, Smith BM & Gustafson NJ (1994): Health bene®ts and dietary ®bres are able to increase the thickness of the practical aspects of high-®ber diets. Am. J. Clin. Nutr. 59, Suppl 5, unstirred water layer covering the apex of the ente- 1242±1247. rocytes (Cherbut et al, 1994). A slowing of the lipids Berggren AM, BjoÈrck IME, Nyman EM & Eggum BO (1993): Short chain absorption in the small bowel could result, since the fatty acid content and pH in the cecum of rats given various sources of carbohydrates. J. Sci. Food. Agr. 63, 397±406. crossing through the unstirred water layer is the major Blackburn NA, Redfern JS, Jarjis HA, Holgate AM, Hanning I & Scarpello limiting factor for the transfer of fat to the body. JH (1984): The mechanism of action of guar gum in improving glucose (5) This satiating effect of psyllium may be explained by tolerance in man. Clin. Sci. 66, 329±36. modi®cations of plasma release of some regulatory Burley VJ (1992): The action of dietary ®ber on satiety. Nutr. Food. Sci. peptides, namely insulin or gastroentero-pancreatic 23, 11±13. Burley VJ, Paul AW & Blundell JE (1989): In¯uence of a high-®bre food regulatory peptides (Blackburn et al, 1984; Gustafsson on appetite: effects on satiation and satiety. Eur. J. Clin. Nutr. 47, 409± et al, 1994). In some studies, insulin response was less 418. important with dietary ®bre than without, or with high- Burley VJ, Paul AW & Blundell JE (1993): Sustained postingestive action ®bre meal than with low-®bre meal (Jenkins et al, of dietary ®ber effect of a sugar beet-®ber supplemented breakfast on satiety. J. Hum. Nutr. Diet. 6, 253±260. 1978; Cherbut et al, 1994). Cherbut C, des Varannes SB, Schnee M, Rival M, Galmiche JP & Delort- Laval J (1994): Involvement of small intestinal motility in blood Anyway, this reduced plasma glucose and insulin glucose response to dietary ®bre in man. Br. J. Nutr. 71, 675±685. response by psyllium seems to be related to a delaying Cummings JH, Southgate DTA, Branch WJ, Wiggins HS, Houston H & effect of psyllium on the absorption of some energetic Jenkins DJA (1979): The ingestion of pectin in the human gut and its substrate at the small bowel level. It has been shown that effect on calcium absorption and large bowel function. Br. J. Nutr. 41, the increased viscosity of the food obtained with soluble 477±485. de Graaf C, Sta¯eu A, Staal P & Wijne M (1992): Beliefs about the ®bres may further reduce the rate of small intestinal satieting effect of bread with spread varying in macronutrient content. absorption. Indeed an increase in the viscosity of food Appetite 18, 121±128. may have several effects: (a) to slow down the small Di Lorenzo C, Williams CM, Hajnal F & Valenzuela JE (1988): Pectin intestinal transit (Low, 1988; Cherbut et al, 1994); but delays gastric emptying and increases satiety in obese subjects. Gastro- Cummings et al (1979) did not ®nd any modi®cation of enterology 95, 1211±1215. Enzi G, Inelmen EM & Crepaldi g (1980): Effect of a hydrophilic oro-cecal transit time with pectin; (b) to reduce the inter- mucilage in the treatment of obese patients. Pharmacotherap. 2, 421± action between food nutrients and digestive enzymes 428. (FlourieÂ et al, 1984; Cummings et al, 1979; (c) to increase FlourieÂ B, Vidon N, Florent C & Bernier JJ (1984). Effect of pectin on the time for nutrient diffusion, by increasing the thickness jejunal glucose absorption and unstirred layer thickness in normal men. Gut 25, 936±941. of the unstirred water layer (FlourieÂ et al, 1984; Rydning & Forman DT, Garvin GE, Forestner JE & Taylor CB (1968): Increased Berstad, 1986). This is consistent with the studies which excretion of fecal bile acids by an oral hydrophilic colloid. Proc. Soc. have demonstrated that soluble ®bres could decrease the Exp. Biol. Med. 127, 1060±1068. area under the curve of serum glucose after a duodenal Frati-Munari AC, Fernandez-Harp JA, Becerril M, Chavez-Negrete A & Basales-Ham M (1983): Decrease in serum lipids, glycemia and body glucose load (Berggren et al, 1993). This is also consistent weight by plantago psyllium in obese and diabetic patients. Arch. Invest. with our own study: serum insulin response to the test meal, Med. (Mex.) 14, 259±268. and postprandial serum glucose and triglycerides levels French SJ & Read NW (1994): Effect of guar gum on hunger and satiety were of less magnitude during the psyllium session than after meals of differing fat content: relationship with gastric emptying. during the placebo one. But it could also be related to motor Am. J. Clin. Nutr. 59, 87±91. Glore S, Van Treeck D, Knehans A & Guild M (1994): Soluble ®ber and or metabolic actions of psyllium, such as increase in the serum lipids: a literature review. J. Am. Diet Assoc. 94, 425±436. production of short chain fatty acids by the large bowel in Grimes DS & Goddard J (1977): Gastric emptying of wholemeal and white response to the dietary ®bre ingestion (Berggren et al, bread. Gut 18, 725±729. 1993). This needs further investigation. Gustafsson K, Asp NG, Hagander B & Nyman M (1993): Effects of different vegetables in mixed meals on glucose homeostasis and satiety. Eur J. Clin. Nutr. 47, 92±100. Gustafsson K, Asp NG, Hagander B & Nyman M (1994): The satiety, Conclusions glucose and hormonal responses after mixed meals with vegetables. Am. J. Clin. Nutr. 59, Suppl 3, 793. Psyllium is able to reduce both hunger feeling and food Gustafsson K, Asp NG, Hagander B & Nyman M (1995b): In¯uence of intake in the hours following a low-calorie meal. This processing and cooking of carrots in mixed meals on satiety, glucose inhibitory effect of psyllium seems to be related neither and hormonal response. Int. J. Food. Sci. Nutr. 46, 3±12. Gustafsson K, Asp NG, Hagander B & Nyman M (1995b): Satiety effects to modi®cation of gastric emptying nor to abdominal of spinach in mixed meals: comparison with other vegetables. Int. J. discomfort. This is not a `ballast' effect of psyllium. One Food. Sci. Nutr. 46, 327±334. could suggest that psyllium is able to increase the time for Heaton KW. Food ®bre as an obstacle to energy intake (1993): Lancet 2, complete intestinal absorption and thus the duration of the 14±17. phase during which energetic substrates are available for Holt S, Heading RC, Cater DC, Prescott LF & Tothill P (1979): Effect of gel ®bre on gastric emptying and absorption of glucose and paraceta- body cells. Hence a sustained satiety and consequently a mol. Lancet 1, 636±639. decrease in food intake. Jarjis HA, Blackburn NA, Redfern JS & Read NW (1984): The effect of This decrease in hunger feelings and food intake occurs ispaghula and guar gum on glucose tolerance in man. Br. J. Nutr. 51, at an acceptable dosage concerning viscosity and smell, 371±378. Jenkins DJA, Jenkins AL, Wolever TMS, Vuksan V, Rao AV, Thompson allowing acceptable compliance. Its may be bene®cial in LU & Josse RG (1994). Low glycemic index: lente carbohydrates and overweight patients, in whom slimming diets often induce physiological effects of altered food frequency. Am. J. Clin. Nutr. 59, increase in hunger feelings and occurrence of chronic suppl 3, 706±709. Inhibition of food intake by psyllium D Rigaud et al 245 Jenkins DJA, Wolever TMS, Leeds AR, Gassul MA, Haisman P & Rigaud D, Ryttig KR, Angel LA & Apfelbaum M (1990). Overweight Dilawari J (1978). Dietary ®bres, ®bre analogues and glucose tolerance: treated with energy restriction and dietary ®bre supplement: a 6-month importance of viscosity. Br. Med. J. 1, 1392±1394. randomized double blind placebo controlled trial. Int. J. Obes. 14, 763± Low AG (1988): Gut transit and carbohydrate uptake. Proc. Am. Nutr. Soc. 769. 47, 153±159. Rydning A, Berstad A, Aadland E & Odegaard B (1982). Prophylactic Raben A, Christensen NJ, Madsen J, Holst JJ & Astrup A (1994): effect of dietary ®ber in duodenal ulcer disease. Lancet 2, 736±738. Decreased postprandial thermogenesis and fat oxidation but increased Rydning A & Berstad A (1986): Dietary ®ber and peptic ulcer. Scand. J. fullness after a high-®ber meal compared to a low-®ber meal. Am. J. Gastroenteroe. 21, 1±5. Clin. Nutr. 59, 1386±1394. Schwartz SE, Levine RA, Singh A, Scheidecker JR & Track NS (1982): Ralphs DNL & Lawaetz NJG (1978): Effect of a dietary ®bre on gastric Sustained pectin ingestion delays gastric emptying. Gastroentero. 83, emptying in dumpers. Gut 19, A 986±987. 812±817. Rigaud D, Bedig G, Merrouche M, Vulpillat M, Bon®ls S & Apfelbaum M Tomlin J (1995). The effect of a gel-forming liquid ®bre on eating (1988): Delayed gastric emptying in anorexia nervosa is improved by behaviour in man. Br. J. Nutr. 74, 427±436. the completion of a renutrition program. Dig. Dis. Sci. 33, 919±925. Vido L, Facchin P, Antonello I, Gobber D & Rigon F (1993): Childhood Rigaud D & Royer I (1988): Dietary ®ber: facts and fancy. Gastroenterol. obesity treatment: double blind trial on dietary ®bres versus placebo. Clin. Biol. (French) 12, 133±1748. Paediatr. Padol. 28, 133±136.