| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
The Monash University Centre for Obesity Research and Education (CORE), Monash University, Alfred Hospital, Commercial Road, Prahran 3181, Victoria, Australia
Address all correspondence and requests for reprints to: John Dixon, Centre for Obesity Research and Education, Monash University, Alfred Hospital, Commercial Road, Prahran 3181, VIC, Australia. E-mail: john.dixon{at}med.monash.edu.au.
| Abstract |
|---|
|
|
|---|
Seventeen weight-stable subjects who had achieved LAGB-induced weight loss attended blind crossover breakfast tests, one with optimal band restriction and one with reduced restriction. Standardized meals were consumed (0900 h) after 14-h fasting. Satiety visual analog scales were completed hourly (07001100 h) before and after feeding. Blood glucose, plasma insulin, ghrelin, and leptin levels were measured. Seventeen body mass index-matched controls were tested.
Optimal restriction was associated with significantly greater fasting and postprandial satiety levels than reduced restriction (P < 0.01). Glucose, insulin, ghrelin, and leptin levels did not alter between optimal and reduced restriction. LAGB subjects displayed higher ghrelin (+12%, P = 0.13) and lower glucose (17%, P = 0.018), insulin (33%, P = 0.016), and leptin (32%, P = 0.05) 4-h area under the curve levels than controls.
Optimal LAGB restriction increased fasting and postprandial satiety levels. This supports the hypothesis that LAGB provides prolonged satiety, present even during fasting, favorably influencing energy homeostasis. Plasma insulin, leptin, and ghrelin appeared unrelated to the satiety effect and displayed orexigenic compensatory changes. Identifying the mechanisms underlying LAGB-induced satiety may assist the understanding of human energy homeostasis and obesity.
| Introduction |
|---|
|
|
|---|
Several peripheral signals are known to regulate energy homeostasis. Insulin and leptin are hormones that circulate in proportion to fat mass and act within the hypothalamus to reduce appetite (7). When fat mass is reduced during weight loss, insulin and leptin levels decline, strongly stimulating appetite and weight gain. Ghrelin, a hormone produced primarily by the stomach (8), appears to stimulate appetite and feeding (9, 10, 11). Ghrelin levels are inversely related to body weight, with obese people displaying low ghrelin levels (12). Diet-induced weight loss has been shown to increase ghrelin levels (13, 14), implying that ghrelin may play a role in countering such weight loss by increasing appetite and energy intake. It has been proposed that lower ghrelin levels may contribute to the success of some surgical weight loss procedures including Roux-en-Y gastric bypass (13, 15) and possibly LAGB (16); however, data have been inconsistent (17, 18, 19, 20).
LAGB surgery involves the laparoscopic placement of an adjustable silicone band around the stomach just below the gastroesophageal junction (Fig. 1A
). Tubing connects the inner shell of the band to a sc port where injection of saline allows the band stoma to be adjusted. When placed in a patient, the band initially offers minimal gastric restriction; over time, restriction is altered to achieve desired weight and nutritional outcomes without causing undue obstruction. Follow-up of the first 709 patients treated at this center revealed progressive weight loss over 2 yr and weight stabilization thereafter (21). Excess weight lost was 53% at 2 yr and 57% at 6 yr.
|
In this study, we investigated the hypothesis that LAGB induces a prolonged satiety effect that operates even during fasting and that plasma ghrelin plays a role in this effect. LAGB subjects who had achieved sustained weight loss attended fasted crossover breakfast meals with their band set to optimal restriction or reduced restriction. The band setting for each test was blind to the subject and direct investigator, and each subject was crossed over to allow paired comparison between restriction levels. Feelings of satiety and blood glucose, plasma insulin, ghrelin, and leptin levels were measured throughout testing. Body mass index (BMI)-matched controls were also tested.
| Subjects and Methods |
|---|
|
|
|---|
LAGB subjects were recruited as volunteers from the LAGB cohort attending The Centre for Bariatric Surgery. The Lap-Band system (Inamed Health, Santa Barbara, CA) was the LAGB device used. Controls were recruited through The Centre for Bariatric Surgery and Monash University Medical School, Alfred Hospital (Victoria, Australia).
Primary exclusion criteria for LAGB subjects and controls included age less than 18 yr, recent weight change (greater than 5 kg in the preceding 2 months), previous weight loss surgery, type 2 diabetes, thyroid disease, and unwillingness or inability to give informed consent. LAGB subjects were between 18 and 36 months post-LAGB surgery, had lost at least 35% of their excess body weight, had not received any change to their band restriction in the preceding month, and had at least 1.5 ml saline present within their band. This latter criterion ensured that reduced band restriction could be applied. All patients from the LAGB cohort who met the study criteria were approached to participate. The amount of satiety experienced by each patient did not influence selection. The selected subjects represented a slightly more successful weight loss group than our general LAGB cohort because normally 15% of patients fail to achieve 35% of excess weight loss after 18 months.
Secondary exclusion criteria included failure to consume the test meal within 20 min, regurgitation of any part of the test meal, and failure to return for a second breakfast test within 2 wk. This latter criterion was applied to minimize any individual changes in body weight and composition between breakfast tests. Due to the experimental nature of this study, the results of excluded subjects were not analyzed on an intention to treat basis.
Controls were individually matched to LAGB participants based on BMI. All but two of the participants had controls that were also matched for sex. All study procedures were in accordance with the Helsinki convention and approved by a Human Research Ethics Committee. Written informed consent was obtained from all participants.
Protocol for breakfast testing
Breakfast test protocol was the same for LAGB subjects and controls. Testing occurred at the Monash University Department of Surgery, Alfred Hospital. Participants arrived via sedentary transport at 0630 h having fasted from 1900 h the previous day. An intravenous catheter was inserted for blood sampling. Weight and height were measured using same device and operator. Testing was carried out from 07001100 h, whereas participants were sedentary. Participants recorded feelings of satiety hourly and directly after feeding on a 200-mm visual analog scale (29), a standard method for assessing satiety. Levels were measured to the nearest millimeter from the origin (no particular feeling, 0 mm; painfully hungry, 100 mm; full to nausea, 100 mm). Blood samples (2 x 7 ml and 1 x 2 ml EDTA) were taken at 0700, 0900, 1000, and 1100 h. A standardized meal containing 22.5 g of a wheat-based cereal (Sanitarium Weet-bix), 170 ml of low-fat milk (Physical), and 12 g of banana was served at 0900 h. This meal represented a common breakfast eaten by LAGB patients as determined by survey. The meal contained approximately 835 kJ (200 kcal) and was 64% carbohydrate, 20.5% protein, and 13.5% fat.
Protocol for randomized blind crossover testing
Each LAGB participant attended two breakfast tests; one with optimal band restriction and one with reduced restriction. Optimal restriction was defined as the amount of band restriction (saline volume) maintaining current weight. Individual settings varied and were determined by independent physicians monitoring weight progress, dietary intake, and obstructive symptoms over time. Reduced restriction was defined as optimal restriction less 2 ml saline or all saline if 1.52 ml was present (Fig. 1B
). Band adjustments occurred 2 d before each breakfast test and were blind to the subject and direct investigators. The order of tests was randomly allocated using a computer-generated blocked sequence at the time of first adjustment. Adjusters removed all saline before reinjecting the required amount. Readjustments were made after the conclusion of each test at 1100 h. Participants were asked to guess their band status immediately after adjustment to assess blinding.
Biochemical assays
Blood samples for hormone assays were mixed with a protease inhibitor, aprotinin (Trasylol, Bayer Pharmaceuticals Corp., West Haven, CT), before being centrifuged at 4 C. Plasma was then stored at 80 C. Blood glucose levels (milligrams per deciliter) were measured by hexokinase spectrophotometric assay (Roche Diagnostics Australia, Castle Hill, NSW, Australia) with inter and intraassay coefficients of variation < 2%. To convert blood glucose levels to millimoles per liter, divide by 18. Plasma insulin levels (microinternational units per milliliter) were measured by microparticle enzyme immunoassay (Abbott Diagnostics Division, Doncaster, Victoria, Australia) with inter- and intraassay coefficients of variation < 4%. To convert insulin values to picomoles per liter, multiply by 7.175. Plasma ghrelin levels (picograms per milliliter) were measured by Linco Research Inc. (St. Charles, MO) via RIA for total ghrelin (both octanoylated and des-octanoylated ghrelin) with intraassay coefficient of variation < 10%. Ghrelin values cannot be converted to picomoles per liter because two different molecular weight species were measured. The multiplication factor lies between 0.297 and 0.308. Plasma leptin levels (nanograms per milliliter) were measured in duplicate via commercial RIAs (Linco Research) with inter- and intraassay coefficients of variation < 5%. To convert leptin values to nanomoles per milliliter, multiply by 0.08. All samples from each LAGB subject were run in the same hormone assay with their matched control to reduce error.
Power calculations
Power calculations for this study were based on published ghrelin levels recorded in obese subjects (13). These data indicated that, for a power of 0.8, 16 subjects were required to provide 95% confidence of detecting a one SD difference in plasma ghrelin levels.
Statistical analysis
Percentage of excess weight lost was calculated using ideal weight from the Metropolitan Life Tables (30). The trapezoid rule was used to obtain area under the curve (AUC) levels for 2 h preprandial satiety, 2 h postprandial satiety, 4 h satiety, and 4 h biochemical levels. Demographic, satiety, and biochemical data were displayed as mean and SE. Differences between paired LAGB tests were compared by paired two-tailed Students t test. Differences between controls and LAGB subjects were compared by unpaired two-tailed Students t test. The influence of time between tests and time with reduced restriction, and change in satiety was tested using univariant general linear modeling. P < 0.05 was considered statistically significant. All analysis was performed using SPSS for Windows version 10 (31).
| Results |
|---|
|
|
|---|
|
All groups (optimal restriction, reduced restriction, and controls) became increasingly hungry preprandially, experienced maximal satiation immediately after the meal, and experienced decreasing satiety thereafter. Optimal restriction was associated with significantly greater feelings of satiety at all six time points when compared with reduced restriction, both during fasting and postprandially (P < 0.01 for all, Fig. 2
). Preprandial 2-h AUC satiety levels were higher in those with optimal restriction compared with reduced restriction levels (24.2 vs. 63.3, P = 0.002). The BMI-matched controls recorded a trend toward lower preprandial 2-h AUC satiety levels than optimal restriction (63.1 vs. 24.2, P = 0.069) and recorded levels comparable with the reduced restriction levels (63.1 vs. 63.3, P = 0.87). Postprandial 2-h AUC satiety levels were higher in those with optimal restriction (46.8) compared with reduced restriction levels (6.2, P = 0.001) and BMI-matched controls (10.6, P = 0.003). There were no significant satiety differences found between reduced restriction and controls at any time point or in AUC analysis. The number of days between paired test meals and the number of hours between band adjustment and test meal were not found to significantly influence change in satiety level (R2 = 0.11, P = 0.2 and R2 = 0.12, P = 0.17, respectively). When the three subjects who correctly guessed their band status were excluded, optimal restriction remained associated with significantly higher satiety levels than reduced restriction and controls at all time points and in AUC analysis.
|
|
There were no statistically significant differences in blood glucose or plasma insulin levels between optimal restriction and reduced restriction at any time points or in AUC analysis (Fig. 4
, A and B). There was, however, a trend toward lower 0900 h (fasting) glucose during reduced restriction (2.52 mg/dl lower, P = 0.07) and a preprandial drop in glucose of 2.70 mg/dl from 07000900 h with reduced restriction compared with 0.18 mg/dl with optimal restriction (P = 0.12). LAGB subjects had significantly lower glucose 4-h AUC levels (257.4 vs. 309.6, P = 0.018) and insulin 4-h AUC levels (46.1 vs. 69.0, P = 0.016) than controls.
|
There were no significant differences in plasma ghrelin or leptin levels between optimal restriction and reduced restriction at any time points or in AUC analysis (Fig. 4
, C and D). LAGB subjects had nonstatistically significant higher ghrelin 4-h AUC levels than controls (2803 vs. 2497, P = 0.13). LAGB subjects had a trend toward lower leptin 4-h AUC levels than controls (56.0 vs. 82.1, P = 0.05).
| Discussion |
|---|
|
|
|---|
Levels of satiety recorded by the LAGB subjects during reduced band restriction were found to be equivalent to those recorded by BMI-matched controls, despite the fact that some degree of gastric restriction was still likely to be present. This observation may simply be a product of subjectivity in satiety scoring, but may also represent a homeostatic response where feelings of satiety quickly decrease to normal after only a partial reduction in gastric restriction in an attempt to counteract the significant weight loss that has occurred. It is also possible that gastric restriction becomes negligible after a large reduction in band fluid volume.
We showed no significant changes in pre- or postprandial blood glucose, plasma insulin, plasma ghrelin, or plasma leptin levels between optimal LAGB restriction and reduced restriction, despite significant satiety changes. Circulating levels of these biochemical factors are, therefore, unlikely to be principal mediators of LAGB satiety induction. A statistical trend was observed toward lower fasting glucose and a greater drop in preprandial glucose during reduced LAGB restriction compared with optimal restriction. It is difficult to know whether this trend occurred randomly or whether a larger number of subjects providing higher statistical power would have shown a significant difference. Mayers glucostatic hypothesis postulates that small declines in glucose concentrations trigger preprandial hunger and meal initiation (32). Thus, it remains possible that optimally adjusted LAGB may reduce hunger by negating Mayers effect.
Plasma ghrelin 4-h AUC levels were found to be 12% higher among LAGB subjects compared with BMI-matched controls; however, with only 17 participants in each group this difference did not reach statistical significance (P = 0.13). Others have previously noted ghrelin levels to be higher among LAGB subjects (19, 20), whereas some have shown lower levels (16). In people who have achieved weight loss after Roux-en-Y gastric-bypass, ghrelin levels have varied between studies; some showing decrease (13, 15), others describing increase (17, 18). Our study suggests that ghrelin levels may be slightly higher than would be expected for BMI after LAGB-induced weight loss, a response that should stimulate appetite. If ghrelin levels do rise with LAGB-induced weight loss, it would be important to compare this to the physiological rise seen with diet-induced weight loss.
Glucose, insulin, and leptin 4-h AUC levels were all found to be lower among LAGB subjects after significant weight loss than BMI-matched controls (17%, P = 0.018; 33%, P = 0.016; 32%, P = 0.05 respectively). We have previously confirmed in a much larger study that LAGB-induced weight loss appears to result in greater improvement in insulin sensitivity than would be expected for BMI change alone (33). The reasons behind this phenomenon remain unclear. This study suggests that leptin levels may similarly reduce more than would be expected for BMI change alone. The combination of low insulin and low leptin levels among LAGB subjects compared with BMI-matched controls represents a response that should normally stimulate appetite and weight gain.
Although more participants and tighter matching of controls for weight, height, sex, and age may have proved more definitive, this study indicates that the prolonged satiety effect induced by LAGB operates in the face of, and despite, increased orexigenic signals from insulin, leptin, and possibly ghrelin. This suggests that the mechanisms that underlie the LAGB satiety effect act powerfully within the energy homeostasis system, such that they are able to allow weight loss and prevent weight regain despite orexigenic compensatory changes.
We hypothesize that the restricted proximal stomach pouch present during optimal LAGB adjustment induces satiety by altering neural and hormonal messages arising from the area. These messages, be they many or few, then act as satiety signals to favorably impact appetite and energy regulation within the central nervous system, allowing weight loss and preventing weight regain. This study clearly demonstrates that LAGB-induced satiety is present even after 14 h of fasting; thus, some of the satiety signals involved would appear to act well after food is clear of the pouch. Furthermore, we have shown that by disrupting the integrity of the restricted pouch, the effect on appetite can be significantly altered within days, implying that some of the satiety signals involved are likely to be relatively short-acting.
Roux-en-Y gastric-bypass and vertical banded gastroplasty are two common surgical procedures that, like LAGB, produce significant and sustained weight loss (21, 34) by reducing energy intake (22, 23, 35, 36, 37). Over the years, both procedures have evolved surgically to include the formation of a small proximal pouch of stomach (Fig. 5
), whereas Roux-en-Y gastric bypass also prevents food from entering the remainder of the stomach and the proximal small bowel. Based on our satiety findings in LAGB, we propose that an element in the durable long-term weight loss of each of these surgical procedures may be related to the common nature of the restricted proximal stomach pouch.
|
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online December 7, 2004
Abbreviations: AUC, Area under the curve; BMI, body mass index; LAGB, laparoscopic adjustable gastric banding.
Received August 3, 2004.
Accepted November 29, 2004.
| References |
|---|
|
|
|---|
Find additional patient-related information at:
This article has been cited by other articles:
![]() |
H. Ashrafian, C. W. le Roux, A. Darzi, and T. Athanasiou Effects of Bariatric Surgery on Cardiovascular Function Circulation, November 11, 2008; 118(20): 2091 - 2102. [Full Text] [PDF] |
||||
![]() |
J. M H Bennett, S. Mehta, and M. Rhodes Surgery for morbid obesity Postgrad. Med. J., January 1, 2007; 83(975): 8 - 15. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. E. O'Brien, J. B. Dixon, C. Laurie, S. Skinner, J. Proietto, J. McNeil, B. Strauss, S. Marks, L. Schachter, L. Chapman, et al. Treatment of Mild to Moderate Obesity with Laparoscopic Adjustable Gastric Banding or an Intensive Medical Program: A Randomized Trial Ann Intern Med, May 2, 2006; 144(9): 625 - 633. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Morinigo, V. Moize, M. Musri, A. M. Lacy, S. Navarro, J. L. Marin, S. Delgado, R. Casamitjana, and J. Vidal Glucagon-Like Peptide-1, Peptide YY, Hunger, and Satiety after Gastric Bypass Surgery in Morbidly Obese Subjects J. Clin. Endocrinol. Metab., May 1, 2006; 91(5): 1735 - 1740. [Abstract] [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 |