| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Nutrition Service (M.R., S.G., P.H.) and Department of Biochemistry (B.S.), Lille University Medical Center, F-59045 Lille, France; and Institut National de la Santé et de la Recherche Médicale U-508 (J.D.), Institut Pasteur de Lille, 59021 Lille, France
Address all correspondence and requests for reprints to Monique Romon, Service de Nutrition, Faculté de Médecine, F-59045 Lille Cedex, France. E-mail: mromon{at}univ-lille2.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to assess the influence of weight loss in obese women on the plasma ghrelin response to a fat- or carbohydrate-rich meal.
Design: Seventeen obese women (mean body mass index 37.6 ± 5 kg/m2) were given an energy-restricted diet (800 kcal/d) for 7 wk, followed by a maintenance diet for 1 wk. Before and after the weight reduction diet, each woman was given (in random order) two isoenergetic test meals, corresponding to 40% of daily energy needs. The test meals contained either 80% fat and 20% protein or 80% carbohydrate and 20% protein. Blood samples were collected over a 10-h period. Two-way ANOVA with repeated measures was used to assess the effect of the test meal on variables.
Results: Weight loss (–11.2 ± 1.4 kg) was associated with a significant decrease in baseline plasma insulin (9.7 ± 4.1 to 7.9 ± 2.4 mU/ml; P < 0.0001) and leptin (25.9 ± 8.3 to 17.2 ± 7.8 ng/ml; P < 0.0001) and an increase in plasma ghrelin (1.86 ± 1.05 to 2.28 ± 1.48 ng/ml; P < 0.05). Before weight loss, there was no significant difference in postprandial ghrelin response between the test meals. After weight reduction, the ghrelin response was more pronounced after the carbohydrate test meal than after the fat test meal (P < 0.02).
Conclusion: Weight loss is associated with an improved postprandial plasma ghrelin response to a carbohydrate meal, whereas the response to a fat meal is not modified.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Plasma ghrelin levels rise before meals and fall rapidly afterward, suggesting that ghrelin may be a meal-initiation signal (2, 3). The postprandial reduction is proportional to energy load (4) and is influenced by the meal’s macronutrient content. In rodents (5, 6) and normal-weight humans (7), the postprandial drop in ghrelin levels is more pronounced after a carbohydrate meal than after a fat meal. In contrast, the postprandial suppression of plasma ghrelin has not been observed in obese subjects (8, 9).
Weight loss is associated with higher fasting levels of ghrelin (2, 10, 11). The effect of weight loss on the postprandial ghrelin response remains subject to debate. Previous work has given contradictory results, showing either a normal postprandial ghrelin suppression (12) or a lack of ghrelin response after a meal (2, 13, 14). However, most of these studies were performed after gastric surgery and/or with mixed meals, and thus the specific response to macronutrients after weight loss is not known. Hence, the aim of the present study was to investigate the effect of diet-induced weight loss on the ghrelin response to carbohydrate and fat meals.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Seventeen obese women were recruited from Lille University Medical Center’s outpatient obesity clinic. They completed a questionnaire about their health status, eating, drinking and exercise habits, weight history, and weight concerns. Inclusion criteria were obesity [body mass index (BMI) > 30 kg/m2] and stable weight (body weight change of < 2 kg over the previous 3 months). Smoking, diabetes, hyperlipidemia, or use of medications (except oral contraceptives) was exclusion criteria. The Lille University Medical Center’s Investigational Review Board approved the protocol according to France’s current regulatory framework.
Weight loss protocol
Before initiating the regimen, subjects met a registered dietitian and were given instructions. All women were submitted to a 7-wk, energy-restricted diet that provided a total of 800 kcal/d (50% as protein, 30% as fat, and 20% as carbohydrate). Low-carbohydrate bars and shakes (Insudiet, Champtoceaux, France) were provided to all subjects. During the diet period itself, subjects met the dietitian twice to discuss progress and any problems. All subjects underwent a 7-d period of weight stabilization before and after the weight loss protocol. The dietitian gave instructions on achieving a weight-stabilization diet (50% carbohydrate, 35% fat, 15% protein), which corresponded to 1.5-fold their basic metabolic rate. Subjects were weighed weekly in the investigation center.
The postprandial experiment
The experiment used a comparative, repeated-measure design, in which each subject served as her own control. Two subsets of experimental conditions (namely high-carbohydrate and high-fat test meals) were performed both before and after weight loss. The two conditions (high-fat and high-carbohydrate) were presented in random order. There was an interval of 48 h between sessions. On the day before each test meal, the subject was asked to abstain from physical exercise and alcohol consumption and eat her last meal before 200 h. On test days, the subjects arrived in the investigation center at 0730 h after an overnight fast (i.e. at least 12 h). Subjects were weighed after voiding their bladder, and body composition was estimated by two-frequency (5 and 100 kHz) bioimpedance analysis (Analycor 2, Spengler, France) bioelectrical impedance analysis. A catheter was inserted into an antecubital arm vein. At 0800 h, a blood sample was taken, and hunger and satiety were rated using a visual analog scale (VAS). The subject was then given the test meal and instructed to consume it fully within 20 min at most. Next, the subject completed VAS ratings and gave blood samples every hour for 4 h and then every 2 h up to 10 h. During this time, the subject was allowed to read or watch TV. At 1800 h, a buffet meal was served to measure spontaneous food intake. The subject was instructed to take between 20 and 30 min to eat this meal, during which she was not allowed to read or watch TV. She was told to eat as much as she wanted until she felt full and comfortable. Food portions were weighed before and after the buffet meal to calculate energy intake. Energy intake and food composition at the buffet meal was calculated using the General Repertory of Food Items (15).
Test meals
The two test meals were composed of either cream or fat-free cottage cheese (either vanilla or coffee flavored, according to the subject’s preference). The energetic value of the two test meals was identical (3.4 MJ). Furthermore, the test meals had the appearance of milk shakes and were prepared so that volume, aspect, and taste were identical. The macronutrient composition was as follows: 1) fat test meal, 79% fat, 3% carbohydrate, 18% protein; and 2) carbohydrate test meal, 1% fat, 81% carbohydrate, 18% protein.
Hunger and satiety ratings
Subjects rated their hunger and satiety on a 100-mm VAS on which the most positive and most negative ratings were indicated at each end (16). Subjects completed baseline ratings at the beginning of each session and then every hour for 10 h after the test meal.
Biochemical and clinical measurements
Blood was collected in EDTA tubes. Plasma was separated by centrifugation (2500 rpm) for 20 min at 4 C. Glucose was determined enzymatically (glucose hexokinase method; Randox Laboratories Ltd., Antrim, UK). Hormone levels were determined using commercially available immunological assays. Insulin was measured by RIA (Bi-Insulin RIA, ERIA, Diagnostics Pasteur, Marnes la Coquette, France). The homeostasis model assessment (HOMA) was used as a surrogate measure of insulin sensitivity: HOMA = fasting insulin (microunits per milliliter) x fasting glucose (millimoles per liter)/22.5 (17). Leptin levels were measured in duplicate by RIA (Linco Diagnostics, St. Louis, MO). The lower limit of the assay’s detection range is 0.5 ng/ml and the intra- and interassay variabilities were 3.7 and 12.6%, respectively. Total plasma ghrelin was determined by RIA (GHRT-89HK; Linco Research, St. Charles, MO). The lower limit of the assay’s detection range is 93 pg/ml, and the intra- and interassay variabilities were 5 and 16.7%, respectively.
Statistical analysis
Two-way ANOVA with repeated measures on both factors was performed to assess the effect of macronutrients on biological variables and VAS ratings before and after weight loss. The two factors were test meal composition (two levels: fat or carbohydrate) and postprandial interval (time: 0 h, 1–9 h). The interaction term was calculated so as to compare the postprandial response for the two test meals. Pearson’s correlation analysis was performed to assess the relationship between baseline ghrelin or ghrelin area under the postprandial curve and other variables. The postprandial area under the curve (AUC) for ghrelin was calculated using the triangulation method. The plasma ghrelin changes was calculated as TX-T0, where T represents the plasma ghrelin level at a given time point and X varies from 0 to 9 h.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Biological and clinical baseline characteristic of the subjects before and after weight loss are shown in Tables 1
and 2. Mean body weight, BMI, and body fat mass were significantly reduced by the weight reduction program, falling from 99.9 ± 14 kg, 36.9 ± 4.6 kg/m2, and 42.9 ± 8.2 kg, respectively to 89.6 ± 13.3 kg (P < 0.0001), 33.1 ± 4.5 kg/m2 (P < 0.0001), and 34.8 ± 8.2 kg (P < 0.0001). Baseline biological values were compared using a two-way ANOVA with repeated measures, with meal (two levels: fat or carbohydrate) and diet (two levels: before and after weight loss) as independent variables. There was no statistical evidence for an effect of test meals on fasting plasma glucose, insulin, HOMA, leptin, and ghrelin levels. In contrast, there was a significant effect of weight loss on baseline insulin (P < 0.0001), HOMA (P < 0.0001), leptin (P < 0.0001), and ghrelin (P = 0.05) levels but not on plasma glucose (ns).
|
|
The results for the postprandial levels are presented in Fig. 1. After the carbohydrate test meal, glucose levels rose to a peak at 1 h and then decreased progressively (main effect: postprandial time, P < 0.0001). There was no evidence for any statistically significant difference in postprandial glucose levels before and after weight loss (interaction: ns; main effect of weight loss, ns). Insulin levels rose progressively to reach a peak 1 h after the carbohydrate test meal and then decreased progressively (main effect: postprandial time, P < 0.0001). Mean postprandial insulin levels were significantly lower after weight reduction (main effect: weight loss, P < 0.017). After the carbohydrate test meal, plasma ghrelin levels decreased progressively to reach a nadir at 4 h postprandially and then rose progressively (main effect: postprandial time, P < 0.0001). On average, mean postprandial ghrelin levels were higher after weight reduction (main effect: weight loss, P < 0.0001).
|
Effect of weight loss on postprandial ghrelin changes after the carbohydrate and fat test meals
The results of postprandial ghrelin changes are presented Fig. 2. After the carbohydrate meal (left panel), the ghrelin changes were characterized by an initial decrease followed by a raise both before and after weight loss (main effect: postprandial time, P < 0.0001; interaction, ns). Mean response was lower after weight loss than before (main effect: weight loss, P < 0.0005). In agreement with these results, the AUC for carbohydrate postprandial ghrelin was significantly lower after weight loss than before (–2.11 ± 2.87 vs. – 4.07 ± 4.58 AU; P < 0.045).
|
To further characterize the ghrelin changes, we compared the carbohydrate and fat response (Fig. 3). Before weight loss (left panel), there was no difference in ghrelin response after the carbohydrate and fat meal (main effect: meal, ns; interaction: ns). There was no difference in the AUC for carbohydrate and fat postprandial ghrelin before weight loss (–2.11 ± 2.87 vs. –2.46 ± 2.72 AU; ns). After weight loss (right panel), carbohydrate postprandial response was significantly different from fat response (main effect: meal, P < 0.04; interaction: P < 0.02). The initial lowering and the late increase of plasma ghrelin was more marked after the carbohydrate than after the fat meal. The AUCs for carbohydrate and fat postprandial ghrelin were not statistically different (–4.07 ± 4.58 vs. – 3.50 ± 6.36 AU; ns) most probably due the higher late rebound of ghrelin after weight loss (significant interaction: P < 0.02 between time and weight loss factors).
|
Results of correlation analyses are presented in Table 3. There was a significant, inverse correlation between baseline plasma ghrelin and both insulin and HOMA values before (r = –0.38, P < 0.04 and r = –0.36, P < 0.36, respectively) and after weight loss (r = –0.43, P < 0.02 and r = –0.40, P < 0.03). There was no statistically significant correlation between the postprandial ghrelin response (AUC) and BMI or glucose, insulin, or HOMA values after both the carbohydrate and fat meals. Finally, there was no evidence for any statistically significant association between weight reduction and changes in baseline biological variables.
|
Postprandial satiety and hunger ratings are presented in Fig. 4. After both the carbohydrate (P < 0.0001) and fat test meals (P < 0.0001), satiety ratings were significantly higher after weight loss than before. Similarly, hunger ratings after both the carbohydrate (P < 0.0001) and fat test meals (P < 0.005) were significantly lower after weight loss than before.
|
Spontaneous food intake is presented in Table 4. Ten hours after the test meals, there was no evidence of a statistically significant difference in energy, carbohydrate, lipid, or protein intake when comparing the test meals (carbohydrate and fat) before and after weight loss.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Before weight loss, the postprandial ghrelin levels of obese women decreased progressively after both the fat and carbohydrate test meals. This finding contrasts with earlier studies showing a defect in postprandial ghrelin suppression in obese subjects (8, 14). Several hypotheses may explain this apparent discrepancy. First, the energy load in the present study was equivalent to 40% of the subject’s daily energy needs, a higher load than was used in previous studies. Therefore, because ghrelin response is proportional to meal energy content (4), one possibility is that the relative energy load in present experiment was sufficient to produce a response. Second, the women in the present study did not display glucose intolerance. Because insulin sensitivity is an important determinant of postprandial ghrelin suppression (18, 19), the positive response could be accounted for by normal sensitivity to glucose. Finally, differences in blood sample timing might also explain part of the discrepancy. However, in contrast to healthy women [who display a more pronounced drop in plasma ghrelin levels after a carbohydrate meal than after a fat meal (7)], the preweight loss ghrelin levels decreased similarly after carbohydrate and fat meals. This observation is in agreement with another study that had shown that the postprandial ghrelin response of obese subjects was not affected by meal composition (9). Taken as a whole, these data could suggest the mechanism regulating the specific postprandial response to carbohydrate and fat may be altered in obese women.
The principal finding of the present study was that weight loss selectively improves the response of ghrelin to carbohydrate. Studies of the effect of weight loss on postprandial ghrelin responses have generally yielded inconsistent results (2, 13, 14). However, recent work has reported that weight loss induced by Roux-en-Y gastric bypass is associated with normal postprandial ghrelin suppression (12) and that a 12-wk diet-induced weight loss improved the postprandial ghrelin response to a standardized test meal (20). The results of the present study further extend these observations and show that after diet-induced weight loss, postprandial ghrelin suppression is improved only after a carbohydrate test meal.
The precise mechanism by which the macronutrient composition of the test meal affects postprandial ghrelin levels is not well defined. In the present study, potential intrastudy differences in the volume and energy contents of the test meals are unlikely to play a role because test meals were calibrated for these parameters. An improvement in insulin sensitivity is a possible explanation. This is supported by experimental clamp and glucose infusion studies in normal-weight and obese subjects, which suggest that postprandial ghrelin response is correlated to insulin response (21, 22). These results have been confirmed by postprandial studies showing that the ghrelin response depends on insulin response and insulin sensitivity (18, 23). According to this hypothesis, the increase in ghrelin response to a carbohydrate test meal may be explained by greater insulin sensitivity after weight loss. Although the obese women in the present study were not diabetic, their HOMA index was improved after weight loss. This may explain (at least in part) the increase in postprandial ghrelin response after a carbohydrate test meal. In addition to improved insulin sensitivity, other factors may explain postprandial ghrelin changes after weight loss. It has recently been demonstrated that peptide YY (PYY) infusion decreases ghrelin levels (24). Because PYY levels are low in obese subjects, we cannot exclude the possibility that a putative rise in PYY levels after weight loss could explain the improvement in postprandial ghrelin response. Furthermore, in healthy subjects, sham feeding induces a decrease in ghrelin levels similar to that seen with actual feeding, suggesting vagal involvement in the postprandial ghrelin fall (25, 26). Greater sensitivity to vagal stimulation after weight loss might also result in a more pronounced drop in ghrelin levels. However, increased PYY levels and vagal stimulation should produce similar improvements in ghrelin response after both carbohydrate and fat meals. The fact that we observed differences as a function of meal composition therefore suggests that PYY and vagal factors are unlikely contributors to the selective change in ghrelin response to carbohydrate after weight loss.
We also aimed at investigating the possible role of ghrelin on satiety. We found that after weight loss, the (baseline and postprandial) satiety and hunger ratings were higher and lower, respectively, suggesting an improvement in fasting satiety and hunger feelings after both the carbohydrate and fat test meals. However, food intake via the buffet meal eaten 10 h after the test meal was affected by neither weight loss nor the test meal’s nutrient content. Equally, there was no correlation between appetite rating and ghrelin levels. This agrees with a previous study that found similar results in obese subjects (27). The most probable explanation is lower sensitivity of hunger and satiety ratings in obese subjects (28). Furthermore, subjects were not blinded to external time cues that could potentially have influenced their subjective feelings of hunger and satiety.
There is still much controversy over the effects of high-fat vs. high-carbohydrate diets in the management of obesity (29, 30). Recent work on weight loss and ghrelin could help gain a better understanding of these clinical results. It has recently been demonstrated that weight loss induced by a high-carbohydrate diet is not associated with an increase in fasting ghrelin levels (31, 32). The results of the present study suggest that a preliminary weight loss induced by an energy-restricted low-carbohydrate diet could boost the effect of high-carbohydrate diets by improving the postprandial response. Long-term studies are necessary to confirm this hypothesis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online December 29, 2005
Abbreviations: AUC, Area under the curve; BMI, body mass index; HOMA, homeostasis model assessment; PYY, peptide YY; VAS, visual analog scale.
Received May 11, 2005.
Accepted December 16, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. Paddon-Jones, E. Westman, R. D Mattes, R. R Wolfe, A. Astrup, and M. Westerterp-Plantenga Protein, weight management, and satiety Am. J. Clinical Nutrition, May 1, 2008; 87(5): 1558S - 1561S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. E. Foster-Schubert, J. Overduin, C. E. Prudom, J. Liu, H. S. Callahan, B. D. Gaylinn, M. O. Thorner, and D. E. Cummings Acyl and Total Ghrelin Are Suppressed Strongly by Ingested Proteins, Weakly by Lipids, and Biphasically by Carbohydrates J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1971 - 1979. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Hayes, C. K. Miller, J. S. Ulbrecht, J. L. Mauger, L. Parker-Klees, M. D. Gutschall, D. C. Mitchell, H. Smiciklas-Wright, and M. Covasa A Carbohydrate-Restricted Diet Alters Gut Peptides and Adiposity Signals in Men and Women with Metabolic Syndrome J. Nutr., August 1, 2007; 137(8): 1944 - 1950. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Hendriks, W. Pasman, W. Blom, and A. Stafleu Reply to C Otto et al Am. J. Clinical Nutrition, September 1, 2006; 84(3): 665 - 666. [Full Text] [PDF]
|
|
|
|
|
|
B. Otto, W. Heldwein, C. Otto, S. Huptas, and K. G Parhofer Effect of a high-fat meal on the postprandial ghrelin response. Am. J. Clinical Nutrition, September 1, 2006; 84(3): 664 - 665. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
