Sustained Weight Loss in Obese Subjects Has Benefits That Are Independent of Attained Weight
The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Abstract
Objective: To explore the hypothesis that sustained weight loss in severely obese patients may have benefits that are independent of their attained BMI.
Research Methods and Procedures: We conducted a comparison of two weight-stable groups with BMI in the 30 to 35 kg/m2 range. Subjects (n = 79) were selected obese patients 3 years after laparoscopic adjustable gastric band surgery, and controls (n = 79) were obese patients seeking weight loss therapy. Subjects were selected in a de-identified manner from our database to best match the control group. A range of clinical, biochemical, and questionnaire measures were obtained to assess obesity-related health status
Results: Subjects maintained a mean weight loss of 32.8 ± 18 kg after surgery. The weight loss subjects had significantly lower fasting plasma glucose, insulin, and triglyceride concentrations, along with higher high-density lipoprotein-cholesterol levels and better indirect measures of insulin sensitivity when compared with controls (p < 0.05 for all). In addition, aminotransferase levels, neutrophil counts, and globulin levels were also significantly lower in weight loss subjects. All differences in laboratory variables remained significant after controlling for BMI. The subjects also reported better health-related quality of life, fewer symptoms of depression, and greater satisfaction with their appearance than controls.
Discussion: These findings suggest that the post-weight loss state conveys benefits that are greater than predicted by the attained BMI. These findings may have important implications regarding the expectations of weight loss therapy, and mechanisms for this effect should be carefully sought.
Introduction
Weight loss is the most effective treatment for obesity-related physical and psychological comorbidity. The extent of weight loss and its effect on obesity-related comorbidity have been of great interest. Many studies have shown that there are major improvements in comorbidity even with modest weight loss, and total resolution of obesity-related comorbidity can occur without attaining normal weight (1,2,3,4). The implication that modest weight loss has disproportionate benefit is interesting and may imply that weight loss state per se has advantages in obese subjects.
Postobese patients or patients who have lost considerable weight have several important characteristics that distinguish them from naturally lean patients of similar weight. These characteristics include lower energy expenditure, lower rates of fat oxidation, enhanced dietary fat clearance, and lower leptin and postprandial insulin levels. Many of these characteristics may help explain the predisposition of these subjects to weight gain (5,6,7,8). But do these or other changes explain the apparent health advantages of modest weight loss?
Obesity surgery provides significant sustained weight loss with improvement or remission in a large range of obesity comorbidity. Following laparoscopic adjustable gastric banding (LAGB)1 surgery, patients lose between 50% and 60% of excess weight loss by 2 years after surgery, and this weight loss has been shown to be maintained for up to 8 years (9,10,11). In our series, as in most bariatric surgical series, the mean BMI of patients at presentation was ∼45 kg/m2, and when weight stabilized after 2 or 3 years, the mean BMI was 32 kg/m2 (9). Therefore, despite considerable weight loss, many patients remained obese (i.e., BMI > 30). At that level of BMI, it may be presumed that these subjects would have the health problems of this level of obesity. These patients, however, were consistently shown to achieve major improvements in the comorbidities of obesity and quality of life (QOL) with outcomes similar to normal community values (12,13,14). These patients who have lost weight yet remain obese present a paradox. Their continued state of obesity should be driving a compromised state of health and QOL, and, yet, the data indicate that the benefits of weight loss seem to override the harm of obesity.
We hypothesized that the weight loss state has advantages to health and QOL that are independent of the BMI. Using a broad range of anthropometric, clinical, biochemical, and questionnaire measures, we explored this hypothesis by comparing a group of weight-stable controls with a weight-stable group of subjects who had similar post-weight-loss BMI levels after LAGB surgery.
Research Methods and Procedures
Two groups of patients were used in this analysis. Both groups were sufficiently concerned regarding the problem of obesity that they would consider surgery an acceptable treatment for their condition. The study was carried out in accordance with the Declaration of Helsinki, with hospital ethics committee approval for the collection of prospective observational data (subjects) and for the randomized controlled trial (RCT; controls). Subjects and controls were recruited from the Melbourne area and were predominantly of middle class with a high level of employment. All were of European background with the exception of one of the controls who had a South Asian background.
Controls
Eighty obese subjects were recruited using a general advertising and information program for a randomized trial comparing LAGB surgery and intensive nonsurgical therapy for the treatment of obesity. Entry criteria included: BMI between 30 and 35 kg/m2, age between 20 and 50 years, a history of multiple attempts to lose weight over a period of at least 5 years, acceptance of the randomized process, no contraindication to either study arm, and self-recognition that their level of obesity is a significant problem to them. In addition, prospective candidates needed to be able to understand the treatment options and to comply with the requirements of each program. No specific or serious obesity-related comorbidity was required. Because only one of the 80 controls had type 2 diabetes, it was decided to exclude patients with type 2 diabetes from this analysis. The control group, therefore, consisted of 79 subjects.
Subjects
Criteria for patients eligible for the usual entry into the clinical LAGB program included: a BMI >35 kg/m2, a history of multiple attempts to lose weight over a period of at least 5 years, and suffering significant medical, physical, or psychosocial problems associated with obesity. In addition, prospective candidates must have had no contraindication to placement of the Lap-Band and were able to understand and comply with the requirements of the indefinite follow-up program.
Criteria were set to identify suitable post-weight-loss subjects to match the prerandomization control group. Patients were generally weight stable at later than 2 years after LAGB surgery. We elected to select all subjects based on their characteristics at 3 years after surgery. Criteria for inclusion required: completion of all components of the routine 3-year annual review (see below), a BMI between 30 and 35 kg/m2 at review, and age between 20 and 50 years at review. Potential subjects were excluded if they had a current or past history of type 2 diabetes, a past history of obesity surgery before the LAGB, or had had the band explanted before 3 years. If more prospective candidates were identified, then first sex and then BMI at 3 years would be used to select the best matched subject for each control. Selection was performed from our database by applying the inclusion and exclusion criteria without reference to the potential subject's identity or pre-weight-loss data. A total of 79 subjects were selected as the best match from a total of 126 eligible candidates.
Common Preoperative and Follow-Up Assessment
Preoperative patients entering our Lap-Band program have a standardized set of assessments that are repeated at yearly intervals after surgery. These include blood pressure (BP) and basic anthropometric measurements including height, weight, and waist, hip, and neck circumferences. Biochemical assessment includes metabolic and nutritional components including: fasting plasma glucose, insulin, lipid profile, liver function tests, homocysteine, hemoglobin, red cell indices and white cell count, iron studies, vitamin B12, and folate. All biochemical tests are performed in an approved pathology laboratory and have been described in previous publications (12,15,16). Patients also complete a number of questionnaires, which include: the Medical Outcomes Trust Short Form 36 (SF-36) (17,18), Beck Depression Inventory (19), and the appearance orientation and evaluation sections of the multidimensional body self-relations questionnaire (20). Collection of these data forms part of a long-term prospective study of this case series. Those recruited for the RCT (control group) had undergone an identical assessment protocol before randomization.
A validated indirect measure of insulin sensitivity based on the log-transformed insulin-glucose product was used (21). This quantitative insulin sensitivity check index (QUICKI) has been shown to correlate well with the hyperinsulinemic euglycemic clamp.
Statistical Analysis
This retrospective analysis involves a comparison of two groups of 79 closely matched patients all in the BMI range of 30 to 35 kg/m2. Data were described using mean ± SD for normally distributed variables, median ± interquartile range for other variables, and percentages for some ordinal groups. Differences between groups were tested using the Student's t test, Mann-Whitney U test, and χ2 test as appropriate. To exclude the potential effect of the small difference in BMI between the groups, linear regression analysis was used to assess the significance of an affect after controlling for BMI. The subject's baseline or pre-Lap-Band data were included for interest only and were not subject to statistical analysis because we have previously published the effect of weight loss on these parameters (22). All statistical analysis was performed using the SPSS/PC statistical program (version 10.0.5 for Windows; SPSS, Inc., Chicago, IL) (23).
Results
After exclusion of the one patient with type 2 diabetes in the BMI 30 to 35 kg/m2 control group, there were 79 controls. In this group recruited for an RCT of Lap-Band surgery compared with best conventional or nonsurgical therapy, there was a tendency for controls to have a BMI in the higher proportion of the range of 30 to 35 kg/m2, with a mean BMI of 33.4 kg/m2. There were 79 subjects selected, using sex and BMI criteria, from 126 possible candidates. After selecting “best match” weight loss subjects, there was a small but significantly higher BMI in the control group (Table 1). There were no significant differences in sex, age, or weight. There was no significant correlation between BMI and any of the biochemical measures in this group of subjects and controls; all, of course, were 30 to 35 kg/m2, so that the very small difference in mean BMI between subjects and controls was unlikely to be relevant. Nevertheless, all significant findings were checked for significance after controlling for BMI using linear regression analysis. Subjects were, as a group, weight stable at the time of the 3-year assessment. These 79 subjects had an initial mean BMI of 44.6 ± 6.8 kg/m2 and mean BMIs of 33.8 ± 4.2, 32.8 ± 2.8, and 33.0 ± 1.4 kg/m2 at 1, 2, and 3 years, respectively.
| Weight loss subjects of BMI 30 to 35 kg/m2 (mean, %) | Baseline controls of BMI 30 to 35 kg/m2 (mean, %) | p value of subjects and controls of BMI 30 to 35 kg/m2 | Baseline for subjects | |
|---|---|---|---|---|
| Number | 79 | 79 | 79 | |
| Age (years) | 39.4 ± 7.8 | 39.5 ± 7.6 | 0.96 | |
| Men in group (%) | 24 | 24 | 24 | |
| BMI (kg/m2) | 33.0 ± 1.4 | 33.4±1.6 | 0.03 | 44.6 ± 6.8 |
| Weight (kg) | 92.7 ± 10.3 | 94.7 ± 11.1 | 0.25 | 125.7 ± 22.4 |
| Weight loss at 3 years | 32.8 ± 17.6 kg | |||
| Excess weight loss at 3 years (%) | 51.5 ± 11.3 | |||
| Ideal weight (kg) | 65.1 ± 6.5 | 65.1 ± 7.1 | 0.99 | 65.1 ± 6.5 |
| Waist (cm) | 101.1 ± 9.8 | 101.3 ± 9.7 | 0.83 | 124.9 ± 12.0 |
| Hip (cm) | 118.5 ± 6.5 | 119.1 ± 6.2 | 0.62 | 138.8 ± 8.8 |
| Waist-to-hip ratio | 0.85 ± 0.08 | 0.85 ± 0.08 | 0.93 | 0.90 ± 0.1 |
| Neck (cm) | 37.7 ± 3.2 | 38.9 ± 3.5 | 0.04 | 42.4 ± 4.7 |
| Systolic BP (mm Hg) | 127.2 ± 17.3 | 130.8 ± 13.7 | 0.16 | 136.4 ± 14.9 |
| Diastolic BP (mm Hg) | 79.8 ± 10.6 | 82.1 ± 10.6 | 0.17 | 86.0 ± 11.6 |
| Fasting glucose (mM) | 4.81 ± 0.48 | 5.01 ± 0.57 | 0.017 | 5.14 ± 0.67 |
| Fasting insulin (mU/L)* | 8.5 ± 4.6 | 11.1 ± 5.4 | 0.003 | 19.0 ± 9.5 |
| QUICKI† | 0.29 ± 0.04 | 0.26 ± 0.04 | 0.001 | 0.23 ± 0.03 |
| Total cholesterol (mM) | 5.3 ± 0.9 | 5.6 ± 1.2 | 0.124 | 5.6 ± 1.0 |
| Fasting triglycerides (mM) | 1.25 ± 0.9 | 1.58 ± 0.9 | 0.009 | 1.70 ± 0.74 |
| HDL-cholesterol (mM) | 1.58 ± 0.38 | 1.37 ± 0.38 | 0.002 | 1.24 ± 0.26 |
| Low-density lipoprotein-cholesterol (mM) | 3.2 ± 0.8 | 3.5 ± 1.1 | 0.08 | 3.6 ± 1.0 |
| Total cholesterol-to-HDL-cholesterol ratio | 3.5 ± 0.9 | 4.4 ± 1.5 | <0.001 | 4.7 ± 1.2 |
- Continuous variables mean ± SD, p values from Student's t test. Test comparing weight loss subjects with controls, both groups with BMI in the range 30 to 35 kg/m2.
- * Insulin log transformed prior to analysis.
- † QUICKI = 1/[(loge fasting plasma glucose) + (loge fasting plasma insulin)].
Metabolic Syndrome Markers
The primary analysis for this study was a comparison of the weight loss subjects and controls. Data regarding the weight loss subjects before Lap-Band surgery are included for interest only. The weight loss subjects had significantly lower fasting plasma glucose levels, insulin, and better calculated insulin sensitivity than controls (Table 1). The conventional lipid abnormalities associated with obesity and the metabolic syndrome were also more favorable in the weight loss subjects, with lower fasting triglyceride, higher high-density lipoprotein (HDL)-cholesterol, and lower total cholesterol-to-HDL-cholesterol ratio (Table 1). Thus, the biochemical measures that are markers of the metabolic syndrome were all favorably affected. On the other hand, BP and anthropometric measures were similar with the exception of neck circumference, which was slightly smaller in the weight loss subjects (Table 1). There were 30 of 79 (38%) patients with the metabolic syndrome in the control group and 11 of 79 (14%) in the weight loss subjects group (χ2, p < 0.001). There was no significant difference in the numbers regularly taking antihypertensive (subjects, n = 3; controls, n = 8) and lipid-lowering medication (subjects, n = 2; controls, n = 2) in each group.
Other Laboratory Studies
There were some significant differences in other laboratory tests between the weight loss subjects and controls (Table 2). Basic hematology showed that hemoglobin levels and mean corpuscular volume were unchanged, but there was a lower total white cell count in the weight loss group, largely driven by lower neutrophil levels. Iron status was largely similar with no difference in iron, transferrin, and percentage transferrin saturation levels, but although median ferritin levels were markedly lower in the weight loss group, abnormally low ferritin levels of <15 were no more common in the weight loss group. There was a significantly higher fasting plasma homocysteine concentration in the weight loss subjects, and many more had concentrations >10. The higher homocysteine levels remained after controlling for vitamin B12 and red cell folate concentrations. Subjects taking regular multivitamin supplements had significantly lower fasting homocysteine concentrations (n = 35, 9.0 ± 2.0) than those taking them intermittently or not at all (n = 44, 11.1 ± 3.7, p = 0.007). The weight loss subjects had lower total plasma protein concentrations largely as a result of lower globulin concentrations. Alkaline phosphatase and alltransaminase levels were significantly lower in the weight loss group.
| Weight loss subjects of BMI 30 to 35 kg/m2 (mean, %) | Baseline controls of BMI 30 to 35 kg/m2 (mean, %) | p value for subjects and controls of BMI 30 to 35 kg/m2 | Baseline for subjects | |
|---|---|---|---|---|
| Number | 79 | 79 | 79 | |
| Hemoglobin (g/L) | 133 ± 14 | 136 ± 16 | 0.22 | 138 ± 11.8 |
| Mean corpuscular volume | 87 ± 6.1 | 88 ± 4.2 | 0.28 | 86.3 ± 5.0 |
| White cell count (×109/L) | 5.9 ± 1.7 | 6.7 ± 1.7 | 0.013 | 6.9 ± 1.8 |
| Neutrophils (×109/L) | 3.5 ± 1.5 | 4.2 ± 1.4 | 0.015 | 4.3 ± 1.4 |
| Lymphocytes (×109/L) | 1.8 ± 0.5 | 1.9 ± 0.5 | 0.10 | 2.0 ± 0.5 |
| Iron (μM) | 16.7 ± 7.8 | 16.2 ± 6.5 | 0.67 | 15.5 ± 5.8 |
| Transferrin (g/L) | 2.7 ± 0.5 | 2.6 ± 0.5 | 0.16 | 2.6 ± 0.4 |
| Saturation (%) | 25.8 ± 10.7 | 25.8 ± 12.2 | 1.0 | 24.9 ± 10 |
| Ferritin* (μg/L) | 47 ± 98 | 85 ± 146 | 0.001 | 88 ± 108 |
| Ferritin less than 15 (n)† (μg/L) | 7 | 5 | 0.50 | 2 |
| Homocysteine* (μM) | 10.0 ± 3.6 | 8.0 ± 2.8 | <0.001 | 8.4 ± 4.0‡ |
| Homocysteine > 10 (n)† (μM) | 35 | 7 | <0.001 | ‡ |
| Vitamin B12 (pM) | 396 ± 177 | 451 ± 178 | 0.060 | 408 ± 149 |
| Red cell folate (nM) | 441 ± 325 | 401 ± 391 | 0.51 | 311 ± 178 |
| Total protein (g/L) | 71.4 ± 4.3 | 74.0 ± 4.8 | 0.001 | 72.6 ± 4.4 |
| Albumin (g/L) | 42.4 ± 3.1 | 43.2 ± 3.9 | 0.065 | 43.2 ± 2.8 |
| Globulin (g/L) | 28.5 ± 3.6 | 30.7 ± 3.6 | <0.001 | 29.4 ± 4.2 |
| Alanine aminotransfereas (IU)* | 16.0 ± 21 | 21.0 ± 25.5 | <0.001 | 27.5 ± 31 |
| Aspartate aminotransferase (IU)* | 17.0 ± 7.3 | 19.0 ± 12 | 0.010 | 20.5 ± 11 |
| Gamma glytamyltransferase (IU)* | 17.5 ± 11 | 21.0 ± 26 | 0.002 | 38.0 ± 32 |
| Alkaline phosphatase* (IU) | 71 ± 25.5 | 88 ± 33 | <0.001 | 88 ± 31 |
| Total bilirubin* (μM) | 11.0 ± 6 | 9.0 ± 3.5 | 0.06 | 10.0 ± 4 |
- All p values from tests comparing weight loss subjects with controls, both groups with BMI in the range 30 to 35 kg/m2. Continuous variables mean ± SD, p values from Student's t test.
- * Variable was not normally distributed median ± interquartile range, p value from Mann Whitney U test.
- † χ2 test.
- ‡ Only 28 of the later subjects had fasting plasma homocysteine at baseline.
QOL, Depression, and Appearance
The weight loss subjects scored significantly better for all eight of the SF-36 domain scores and the physical component summary score when compared with the controls (Table 3). Subjects also had QOL scores comparable with community reference levels (Figure 1). They also reported fewer symptoms of depression as measured by the Beck Depression Inventory. Appearance orientation, or the value one places on grooming and appearance, was the same, and in the community normal range, in both groups. The weight loss group more favorably scored their appearance evaluation when compared with the control group.
| Weight loss subjects BMI 30 to 35 kg/m2 | Controls BMI of 30 to 35 kg/m2 | p value | |
|---|---|---|---|
| Physical function | 85.3 ± 18.3 | 69.4 ± 22.2 | <0.001 |
| Role physical* | 100 ± 0 | 75 ± 75 | <0.001 |
| Pain | 85.1 ± 20.4 | 71.5 ± 23.0 | <0.001 |
| General health | 78.0 ± 18.7 | 54.7 ± 22.4 | <0.001 |
| Energy or vitality | 70.0 ± 18.0 | 41.4 ± 20.5 | <0.001 |
| Social function | 83.7 ± 23.5 | 68.1 ± 25.7 | <0.001 |
| Role emotional* | 100 ± 33.3 | 100 ± 66.6 | 0.024 |
| Mental health | 75.7 ± 17.5 | 64.0 ± 19.6 | <0.001 |
| Physical component summary | 52.7 ± 8.0 | 44.7 ± 10.4 | <0.001 |
| Mental component summary | 49.0 ± 7.2 | 47.0 ± 8.2 | 0.12 |
| Beck Depression Inventory | 5.5 ± 6.4 | 13.4 ± 8.3 | <0.001 |
| Appearance orientation | 3.66 ± 0.68 | 3.56 ± 0.69 | 0.40 |
| Appearance evaluation | 2.72 ± 0.70 | 1.92 ± 0.57 | <0.001 |
- Domain score of the SF-36.
- Normally distributed response scores mean ± SD, p values from Student's t test.
- * Response scores were not normally distributed median ± interquartile range, p value from Mann-Whitney U test.
SF-36 domain scores for subjects and controls plotted in a standard manner. Expected community normal values or reference values have been added (18). Popn, population.
Discussion
This analysis of the two groups of well-matched obese individuals (BMI 30 to 35 kg/m2) found very substantial differences between those who were weight stable after weight loss and those presenting naturally at this weight. Almost all these differences indicated a more favorable status in the weight loss group when compared with age- and BMI-matched controls presenting for obesity management.
There are weaknesses with this type of retrospective analysis. Patients were not recruited at the same time or in the same manner, and, of course, one group had been regularly followed after LAGB surgery. The controls in this study were not control subjects of the classic definition. They were unusual controls in that they were selected to match the weight loss state of the subjects. Therefore, one must be cautious and recognize that there are limitations in attributing the differences found solely to weight loss. However, the unusual nature of the study design and analysis may give us a unique insight into some of the effects of weight loss in obese subjects independent of their initial weight.
Our findings extend those of previous research looking at the health status of postobese subjects (24). We demonstrated improved indirect measures of insulin sensitivity and dyslipidemia associated with the metabolic syndrome in the weight loss subjects, suggesting advantages of the intentional weight loss state per se, and provided support for the observation that even the modest weight loss state has a substantial effect in reducing the risk of developing type 2 diabetes and improves the cardiovascular risk profile even in subjects who remain overweight or obese (2,25).
We have not routinely looked at specific inflammatory markers associated with obesity and the metabolic syndrome. However, several laboratory measures have suggested a reduction in inflammation in the weight loss subjects. The total white cell and neutrophil counts have been lower (26), and total globulin and ferritin concentrations have also been lower. The lower ferritin is likely to reflect its acute phase effect rather than lower iron stores, as other markers of iron status are the same in each group. LAGB surgery has no malabsorptive component to its action, and iron deficiency is not expected. Enzyme concentrations indicative of hepatocellular inflammation were all significantly lower in the weight loss group. These differences in indirect inflammatory markers suggest that in conjunction with better conventional lipid and glucose markers of the metabolic syndrome, inflammation markers may also be better in the weight loss group. Improvements in both the metabolic proinflammatory cytokines and inflammatory markers of the metabolic syndrome have improved with weight loss (27,28).
Mechanisms for the apparent advantage of the weight loss state need to be sought. Neuroendocrine settings in the weight loss state are quite different from those in the normal state and are very much geared to favor weight gain. Conversely, it is also possible that metabolic factors associated with obesity actually put a brake on further weight gain and are very active in the region of the subject's natural weight. Therefore, any weight reduction, however small, may result in an exaggerated beneficial metabolic effect. In support of this hypothesis is the observation that hyperinsulinemia at baseline in nondiabetic Pima Indians has been associated with less weight gain over the following years (29). The concept of a metabolic brake on weight gain may also be supported by the observation that the insulin-resistant subjects have great difficulty achieving and sustaining weight loss. We have demonstrated hyperinsulinemia and diseases related to the metabolic syndrome to be indicators of poorer weight loss after LAGB surgery (30). In this situation with the brake off, through weight loss, the drive for weight gain may be very powerful. Clearly, more needs to be known about this post-weight-loss state, and careful assessment of weight distribution, dietary intake, and energy expenditure would be interesting.
Raised fasting plasma homocysteine levels have been linked to increased vascular risk (31). Higher homocysteine levels have been reported after weight loss despite normal micronutrient concentrations, and it appears that with weight loss, higher concentrations of folic acid and vitamin B12 are required to maintain optimally low levels (16,32). The present analysis again demonstrated higher homocysteine levels in the weight loss group and some reduction of the problem in those taking regular multivitamin supplements. The observation suggests that raised homocysteine levels continue in weight-stable post-weight-loss subjects as previously reported by the Swedish Obese Subjects study (33). It is disappointing that we have not been able to encourage >44% of our patients to take regular multivitamin supplements.
In addition to metabolic and, perhaps, inflammatory differences between groups, there also seem to be functional and psychosocial advantages in the weight loss. We have previously demonstrated that the extent of weight loss is a poor predictor of improved QOL after LAGB surgery and that mean SF-36 scores return to normal community values (14). Here, in two groups of subjects with class I obesity, we found major differences in all aspects of self-reported QOL, especially the physical aspects, with the weight loss group scores comparable with expected community values (18). Although differences may be partly explained by the relationship to therapy, with one group having had generally successful surgery and the other seeking weight loss therapy, it remains highly likely that the weight loss per se provides significant benefit. The mechanism for any benefit is unclear, and hypotheses could include: relief of more serious QOL, psychological and body image impairment, improved metabolic or inflammatory status, and improved physical status of joints, muscles, and other structures.
We have previously shown that symptoms of depression are reduced with substantial weight loss (13). This analysis suggests that the weight loss state, in addition to the actual weight attained, may reduce depressive symptoms.
This analysis provides some insight into the value of the most commonly used bariatric surgical procedures because both LAGB and RYGB provide 50% to 60% of excess weight loss at >3 years after surgery. Patients can be advised that realistic expectations are for 50% to 60% excess weight loss, but although they may still be heavier than the community mean, they can anticipate substantial improvements in obesity-related comorbidity, QOL, body image, and psychological status.
Acknowledgement
Inamed Health Corporation (Santa Barbara, CA) the manufacturers of the Bioenterics Lap Band, provided a research grant for The Monash University, Centre of Obesity Research and Education.
