Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
Surgical treatment of morbid obesity An update
AGENCE D’ÉVALUATION DES TECHNOLOGIES ET DES MODES D’INTERVENTION EN SANTÉ
41
Surgical Treatment of Morbid Obesity
An Update
August 2006(Original French version published in October 2005)
Report prepared for AETMIS by by Raouf Hassen-Khodja and Jean-Marie R. Lance
This report was translated from an offi cial French publication of the Agence d’évaluation des technologies et des modes d’intervention en santé (AETMIS). Both the original report, titled Le traitement chirurgical de l’obésité morbide : mise à jour and the English report are available in PDF format on the Agency’s Web site.
Scientifi c reviewDr. Véronique Déry, Chief Executive Offi cer and Scientifi c Director
TranslationJocelyne Lauzière, M.A., Certifi ed Translator
Editorial supervisionSuzie Toutant
Page layoutJocelyne Guillot
ProofreadingFrédérique Stephan
Bibliography researchDenis Santerre
Co-ordinationLise-Ann Davignon
Communications and disseminationDiane Guilbault
For further information about this publication or any other AETMIS activity, please contact:
Agence d’évaluation des technologies et des modes d’intervention en santé2021, Union Avenue, Suite 1040Montréal (Québec) H3A 2S9
Telephone: (514) 873-2563Fax: (514) 873-1369E-mail: [email protected]
How to cite this document:
Agence d’évaluation des technologies et des modes d’intervention en santé (AETMIS). Surgical Treatment of Morbid Obesity: An Update. Report prepared by Raouf Hassen-Khodja and Jean-Marie R. Lance (AETMIS 05-04). Montréal: AETMIS, 2006, xvii-113 p.
Legal depositBibliothèque et Archives nationales du Québec, 2006National Library of Canada, 2006ISBN 2-550-45724-2 (Print) (French edition ISBN 2-550-45464-2)ISBN 2-550-45725-0 (PDF) (French edition ISBN 2-550-45465-0)
© Gouvernement du Québec, 2005.This report may be reproduced in whole or in part provided that the source is cited.
MISSION
The mission of the Agence d’évaluation des technologies et des modes d’intervention en santé (AETMIS) is to contribute to improving the Québec health-care system and to participate in the implementation of the Québec government’s scientifi c policy. To accomplish this, the Agency advises and supports the Minister of Health and Social Services as well as the decision-makers in the health-care system, in matters concerning the assessment of health services and technologies. The Agency makes recommen-dations based on scientifi c reports assessing the introduction, diffusion and use of health technologies, including assistive devices for disabled persons, as well as the modes of providing and organizing ser-vices. The assessments take into account many factors, such as effi cacy, safety and effi ciency, as well as ethical, social, organizational and economic implications.
EXECUTIVE
Dr. Luc DeschênesCancer Surgeon, President and Chief Executive Offi cer of AETMIS, Montréal, and Chairman, Conseil médical du Québec, Québec
Dr. Véronique DéryPublic Health Physician, Chief Executive Offi cer and Scientifi c Director
BOARD OF DIRECTORS
Dr. Jeffrey BarkunAssociate Professor, Department of Surgery, Faculty of Medicine, McGill University, and Surgeon, Royal Victoria Hospital (MUHC), Montréal
Dr. Marie-Dominique BeaulieuFamily Physician, Holder of the Dr. Sadok Besrour Chair in Family Medicine, CHUM, and Researcher, Unité de recherche évaluative, Hôpital Notre-Dame (CHUM), Montréal
Dr. Suzanne ClaveauSpecialist in microbiology and infectious diseases, Hôtel-Dieu de Québec (CHUQ), Québec
Roger JacobBiomedical Engineer, Coordinator, Capital Assets and Medical Equipment, Agence de la santé et des services sociaux de Montréal, Montréal
Louise MontreuilAssistant Executive Director, Direction générale de la coordination ministérielle des relations avec le réseau, ministère de la Santé et des Services sociaux, Québec
Dr. Jean-Marie MoutquinObstetrician/Gynecologist, Research Director, and Executive Director, Département d’obstétrique-gynécologie, CHUS, Sherbrooke
Dr. Réginald NadeauCardiologist, Hôpital du Sacré-Cœur, Montréal, Board Member of the Conseil du médicament du Québec
Guy RocherSociologist, Professor, Département de sociologie, and Researcher, Centre de recherche en droit public, Université de Montréal, Montréal
Lee SoderströmEconomist, Professor, Department of Economics, McGill University, Montréal
i
Dr. Reiner BankenPhysician, Deputy Chief Executive Offi cer, Development and Partnerships
Dr. Alicia FramarinPhysician, Deputy Scientifi c Director
Jean-Marie R. LanceEconomist, Senior Scientifi c Advisor
Lucy BoothroydEpidemiologist, Scientifi c Advisor
iii
FOREWORD
SURGICAL TREATMENT OF MORBID OBESITY: AN UPDATE
Obesity is now considered a major public-health problem and has even been declared a “global epidemic” by the World Health Organization (WHO). This chronic disease, which results from numerous biological, environmental and behavioural factors, leads to several health problems, including hypertension, dyslipidemia, diabetes and some cardiovascular disorders. Québec has not been spared from this epidemic, with a rate of obesity that reached 21.8% in 2004.
In practice, obesity is defi ned as a body-mass index (BMI) greater than 30 kg/m2. A more critical threshold is attained when the BMI reaches 40, or even only 35 if it is associated with co-morbidities. This is referred to as morbid obesity. In such cases, behavioural and medical therapy fail to achieve long-term weight reduction and, according to the WHO, surgery is the only effective treatment. However, owing to the expanding range of surgical techniques and the recent introduction of laparoscopic approaches, legitimate questions have been raised about the effi cacy and risks of these major procedures.
As early as 1998, the Agence d’évaluation des technologies et des modes d’intervention en santé (AETMIS) had produced a report on the surgical treatment of morbid obesity, or bariatric surgery. At that time some techniques had been classifi ed as accepted technologies, but another, performed in a Québec hospital, was still considered experimental. The rapid evolution of bariatric surgery and of the scientifi c evidence on this topic has prompted the need for an update. Moreover, faced with the growing prevalence of morbid obesity and concerned about the effi cacy of the different techniques and the need to provide effective management for those affected by this health problem, the Ministère de la Santé et des Services Sociaux (MSSS) asked AETMIS to assess this surgical treatment.
This report examines the effi cacy and risks of complications pertaining to the four major types of procedures, including those performed in Québec. It also compares abdominal-incision approaches with laparoscopic approaches, and deals with the economic aspects of this treatment. The primary sources for this analysis were scientifi c articles and health-technology assessment reports published since 1998.
Results confi rm the long-term effi cacy of surgical treatment in terms of maintaining weight loss and reducing co-morbidities. The different techniques available in Québec are considered effi cacious and safe. Some have proven effi cacy, while others continue to require close follow-up so that patient indications and eligibility may be better identifi ed. In conclusion, AETMIS recommends that an action plan be developed to clearly defi ne the needs for bariatric surgery and establish the means to meet those needs; that key conditions be determined to ensure that hospital centres offer high-quality bariatric treatment; and that a registry on morbid obesity and its management be established.
In submitting this report, AETMIS hopes to contribute to improving the health and quality of life of people with morbid obesity.
Dr. Luc DeschênesPresident and Chief Executive Offi cer
iv
ACKNOWLEDGEMENTS
This report was prepared at the request of AETMIS by Raouf Hassen-Khodja, MD, MSc (health administration), physician (hemobiology) and consultant researcher, and by Jean-Marie R. Lance, MSc (economics), senior scientifi c advisor.
AETMIS would like to thank the following external reviewers for their valuable comments on this report:
Mitiku Belachew, MDProfessor, University Surgery Department, Centre hospitalier régional de Huy, Huy , Belgium
Nicholas V. Christou, MDSurgeon and professor of surgery, McGill University Health Centre, Montréal, Québec
Slim Haddad, MDAssociate professor, Department of Social and Preventive Medicine, Faculty of Medicine, and economist, Université de Montréal, Montréal, Québec
Picard Marceau, MDProfessor, Department of Surgery, Université Laval, and surgeon, Hôpital Laval, Quebec City, Québec
Jean Mouïel, MDProfessor, Nice Faculty of Medicine, and specialist in digestive surgery, Centre de chirurgie et laparoscopie, Nice, France
Nicola Scopinaro, MDProfessor, Department of Surgery, Faculty of Medicine, Università di Genova, Ospedale San Martino, Genoa, Italy
Rudolf Steffen, MDFMH specialist in visceral surgery, Bern, Switzerland
DISCLOSURE OF CONFLICTS OF INTEREST
None declared.
v
SUMMARY
MORBID OBESITYSince 1998, in the wake of a report by the World Health Organization (WHO), obesity has been considered a major public-health problem and has even been declared a “global epidemic.” The WHO even calls it a chronic disease requiring long-term strategies for effective prevention and management. Obesity is the result of complex interactions of metabolic, endocrine, genetic, socio-economic, environmental, cultural, psychological and behavioural factors. It causes many diseases, including hypertension, hyperlipidemia, diabetes, some cardiovascular disorders, sleep apnea, osteoarthritis and some cancers, and even death.
Defi nition
Obesity is characterized by excess body fat and is generally defi ned by the body-mass index (BMI), which takes into account weight and height. This index is expressed in kilograms per square metre (kg/m2). The term obesity applies when the BMI is greater than or equal to 30 kg/m2. A BMI between 25 and 29.9 kg/m2 is called overweight. Morbid obesity refers to a BMI that is greater than or equal to 40, or 35 kg/m2 if associated with co-morbidities.
Prevalence
The prevalence of obesity (BMI ≥ 30) in the household population aged 18 and older (excluding pregnant women) is growing steadily. Whereas obese people accounted for only 13.8% of Canada’s population in 1978–1979, this proportion rose to 23.1% in 2004; these fi gures are based on directly measured height and weight. The gap between men and women is small: 22.9% vs 23.2%. The situation is comparable in Québec, which has an overall rate of 21.8% (20.9% for men and 22.7% for women). Morbid obesity (BMI ≥ 40) has also risen dramatically in Canada, from 0.9% in
1978–1979 up to 2.7% in 2004, with women being more affected by this problem (3.8% vs 1.6% for men). This fi gure is not available for Québec.
Consequences
Obesity gives rise to a considerable epidemiological and economic burden. According to studies in the United States, where the rate of obesity during the 1999–2002 period reached 31.1% among people aged 20 to 74, this problem caused at least 112,000 deaths per year, although other estimates combining overweight and obesity yielded more than 300,000 deaths. Controversy persists over the magnitude of this burden. From an economic viewpoint, Canada’s 1997 direct medical costs attributable to adult obesity were estimated to be $1.8 billion, or 2.4% of total direct medical costs. One study estimated that obesity in Québec led to expenditures totalling $700 million, or 5.8% of the province’s health-care budget (1999–2000 fi scal year), and to productivity losses in excess of $800 million.
ROLE OF SURGICAL TREATMENT IN THE THERAPEUTIC APPROACH TO OBESITYThe therapeutic approach to obesity is multi-faceted and complex. It requires a specially adapted treatment structure and the availability of a multidisciplinary team.
Management of obesity
Obesity management is based on a minimum of three key measures:
1) intensive patient education on improving food patterns;
2) counselling on the need for regular physical activity; and
vi
3) behavioural approaches designed to help people better regulate the lifestyle habits needing to be modifi ed.
Weight-loss objectives must be clearly defi ned with the patient. Physicians may suggest drug therapy for patients unable to meet their target objectives through diet and physical activity.
Management of morbid obesity
The multi-dimensional approach described above is not effective for treating morbid obesity. According to the WHO, the only effective treatment is bariatric surgery (from the Greek word baros, which means weight).
Bariatric surgery currently encompasses a range of techniques that can be classifi ed into two main types of procedures:
gastric-restriction techniques, which decrease food intake by reducing gastric capacity: gastroplasty: a pouch or partition is
created by horizontal or vertical stapling or banding (vertical banded gastroplasty), gastric banding: a fi xed or adjustable band
is inserted to form a small-volume gastric reservoir;
hybrid techniques, which combine gastric restriction with the principle of intestinal malabsorption by creating either a bypass or a diversion system: gastric bypass techniques, including Roux-
en-Y, the most common variant performed worldwide, biliopancreatic diversion with distal
gastrectomy or duodenal switch.
Although all these surgical techniques were developed for the abdominal-incision, or open-surgery, approach (laparotomy), surgeons continued to explore new ways of performing this procedure, chiefl y in terms of the surgical approach. As a result, laparoscopic techniques appeared in the mid-1990s and soon became widely used in several countries. In fact, according to an international survey, 62.85% of
the procedures performed worldwide in 2003 were done laparoscopically, especially gastric bypass and adjustable gastric banding.
Although the effi cacy and safety of each of these techniques, whether open or laparoscopic, are established to varying degrees, they still raise legitimate questions.
ASSESSMENT OBJECTIVESIn 1998, the Conseil d’évaluation des technologies de la santé (the predecessor of AETMIS) published a report on the surgical treatment of morbid obesity. The need to clarify the status of biliopancreatic diversion with duodenal switch (the procedure used in Québec), the rapid expansion of laparoscopic techniques and the growing prevalence of morbid obesity are the reasons for this update. This report also responds to a request from the MSSS asking AETMIS to examine the evolving effi cacy of bariatric-surgery techniques and the best conditions for managing people with severe obesity. Lastly, this assessment explores the economic aspects of this treatment.
METHODOLOGYThis report is based on a review of the scientifi c literature and health-technology assessments published between 1998 (publication date of the previous report) and April 2005. Standard databases—Medline, Cochrane Library and HTA Database—were searched on the following keywords: obesity, morbid obesity, surgical treatment, bariatric surgery, gastroplasty, gastric bypass, gastric banding.
The search retrieved a large number of studies published since 1998. However, given the scarcity of comparative trials, whether randomized or not, and the predominance of case series, studies were selected by means of a simplifi ed grid containing the following elements: study design, publication date, number of patients treated, length of follow-up, and relevance of clinical outcome measures.
vii
The main evaluation criteria for this analysis were:
Clinical effi cacy: excess weight loss (EWL), defi ned in relation to ideal weight, or, alternatively, weight loss and a decrease in body-mass index; Safety: complications characterized by their
onset (short, medium or long term), type and severity; Co-morbidity: reduction or not of associated
conditions; Consumption of health goods or services
or other resources: days of hospitalization, mean length of stay, operating time; Effi ciency: resource costs or savings; cost-
effectiveness and cost-utility ratios.
RESULTSSurgical treatment in general
Despite the large number of primary studies on the surgical treatment of morbid obesity, most cover either treatments with established effi cacy or new approaches, especially laparoscopic procedures. Few provide long-term outcomes, however. A single major study (Swedish Obese Subjects Intervention Study, or SOS) compares surgical treatment with the medical approach. It uses a prospective design with 18 subject-matching variables. The other comparative studies, some of which were randomized, examined either the effects of the surgical approach (open or laparoscopy) or variants of the same technique. The studies are therefore mostly retrospective case series, while a few are prospective.
Surgical treatment is currently recognized as being a more effective therapeutic option than non-surgical treatment for patients who are morbidly obese. Although most of the evidence refers to short-term outcomes, several studies are beginning to demonstrate long-term sustained weight loss. Moreover, the SOS (the best controlled study available) found that bariatric surgery achieved a sustained weight loss of 16.1% in people with BMIs of at least 40,
or at least 35 if associated with co-morbidities, including diabetes, hyperlipidemia and hypertension. Surgery itself has some potentially serious complications. Although these adverse effects are generally managed appropriately, they require continual assessment.
Bariatric surgery remains an expensive procedure because it requires a multidisciplinary team, a specialized technical platform and long-term follow-up. In return, the resulting weight loss decreases the prevalence of co-morbidities and their consequences (prescription drug spending), serving to reduce productivity losses caused by sick leave and disability, and improves quality of life. Nevertheless, the favourable cost-effectiveness (or cost-utility) ratio and the effi ciency suggested by the current state of the evidence need to be confi rmed by longer-term well-designed economic studies.
Surgical techniques
Although bariatric surgery relies on a wide range of techniques, current evidence does not yet favour any one over the others, owing to the variety of contexts in which they are applied, the diversity of patient characteristics and the lack of well-designed controlled studies. Moreover, a single procedure may involve several techniques.
The choice of surgical technique depends on a number of factors:
Patient profi le: age, personality, BMI, food patterns, personal understanding and commitment, co-morbidities, contra-indications; Reversibility or non-reversibility of the
technique; Risks linked to each technique (e.g., wound
dehiscence, hernias, device slippage, staple-line failure); Potential effects of nutritional defi cits; Availability of human and material
resources; Support provided by the expertise of a
multidisciplinary team; and
viii
Surgical team’s experience in bariatric surgery and, where applicable, in laparoscopy, which requires a lengthy learning curve.
In terms of overall effi cacy, current evidence generally indicates that hybrid techniques combining gastric restriction and intestinal malabsorption are superior to those designed only to restrict gastric capacity. The following provides details regarding the main techniques under review, of which three are used in Québec.
Roux-en-Y gastric bypass (RYGB): This technique has proven effi cacy in terms of stable weight loss, low complication rates and reduction of co-morbidities. Considered the gold standard of weight-loss surgery, the RYGB is the most commonly used gastric-bypass technique.
Vertical banded gastroplasty (VBG): Although this technique has established effi cacy, it has achieved lower than expected weight loss and has lost favour with North American surgeons (including those in Québec). Combined with the RYGB, VBG yields good long-term results.
Adjustable gastric banding (AGB): This technique is generally recognized as being effective in terms of both weight loss and low complication rates. It has the advantage of being reversible and is increasingly replacing VBG.
Biliopancreatic diversion with duodenal switch (BPD-DS): Despite the fact that this technique is used only in a few centres because of its stringent requirements for post-operative patient management and follow-up, its long years in use (over 20 years), the cumulative number of procedures performed to date and its positive weight-reduction results mean that this procedure is no longer considered experimental. In addition, some studies suggest that BPD-DS would be appropriate for super-obese patients with BMIs over 50.
Laparoscopic approach
Laparoscopic procedures offer many advantages: they reduce hospital stays and decrease, if not eliminate, complications
associated with open surgery; however, they do have other types of complications. Surgeons must train in the best conditions to master this approach.
The two most advanced laparoscopic techniques are Roux-en-Y gastric bypass (LRYGB) and adjustable gastric banding (LAGB), and they are no longer considered experimental. They must nonetheless be introduced in an environment that permits the ongoing study of their effects. After one year of follow-up, the LRYGB achieves the same outcomes as the open version, and their early complications differ only slightly. However, it is still necessary to obtain longer-term comparative.
The LAGB techniques appear to be safe and effective (in terms of excess weight loss) and have the extra advantage of being reversible. Furthermore, major complications are rare, and complication and re-operation rates are acceptable. Yet these effects have been measured only in the short term and need to be confi rmed by longer studies.
The other laparoscopic techniques are still classifi ed as experimental, owing to the uncertainty surrounding their effects.
Lastly, in addition to offering no comparisons, the evidence on the surgical treatment of adolescents and children with morbid obesity is insuffi cient to draw valid conclusions. Although this assessment has not examined the consequences of substantial weight loss (e.g., the need for reconstructive plastic surgery), this aspect must not be overlooked in the therapeutic treatment plan because it has a potentially signifi cant psychological impact.
COST EFFECTIVENESS OF BARIATRIC SURGERYAccording to the current state of evidence, even if the published economic studies and models have their limitations, the surgical treatment of morbid obesity would appear to be a cost-effective procedure. Although this
ix
type of surgery is relatively expensive, mostly because of the costs incurred by the procedure itself and by the management of early or late complications than can result, and because of the requirement for annual follow-ups and the possible need for plastic surgery, the positive effects linked to weight reduction would appear to compensate for these costs. Indeed, bariatric surgery lowers the prevalence of co-morbidities (e.g., cardiovascular diseases and diabetes) and their impact on resource utilization (hospitalizations, drug expenditures). It also reduces productivity losses caused by sick leave and disability, and improves quality of life.
These initial results must nevertheless be confi rmed by more well-designed economic evaluations based on factual data on long-term effectiveness and resource utilization and on valid comparisons of the different surgical techniques and approaches (laparoscopy or open surgery).
CHALLENGES FOR QUÉBEC’S MEDICAL PRACTICEDifferent bariatric-surgery techniques are currently being used in Québec by highly experienced surgeons in the fi eld. Yet there is a lack of data on the quality and effectiveness of these procedures and on the population of treated patients. The supply of services also appears insuffi cient, given the steady growth of waiting lists and wait times. In such a context:
it is crucial to know and share all the different information about the treated population and the outcomes achieved in bariatric-surgery centres; it is necessary to effectively measure the
evolution and extent of bariatric-surgery needs resulting from the growing prevalence of morbid obesity; it is advisable to promote the development
of practice guidelines on the management of patients with morbid obesity in order to ensure that service offerings are of high quality.
Québec’s Association of General Surgeons (QAGS) has developed a policy on the surgical treatment of morbid obesity. The QAGS emphasizes the following points: need for an interdisciplinary team; designation of referral centres; information and training for surgical residents; and increased bariatric-surgery training opportunities. Furthermore, it would be advisable, in the management of any bariatric-surgery plan, to anticipate the potential need for reconstructive plastic surgery.
RECOMMENDATIONS1) It is recommended that the Ministère de
la Santé et des Services Sociaux and other decision makers concerned with the problem of morbid obesity identify current and future needs in bariatric surgery, establish an action plan to increase the capacity to provide this treatment, and ensure that patients in the different settings and regions have fair access to these services.
2) It is recommended that, at the organizational level, all hospital bariatric-surgery programs comply with the conditions listed below, which will be subject to a quality-assurance process. Such programs must:
Establish a strict patient-selection process (e.g., patients who have BMIs of 40 kg/m2 or more, or 35 with co-morbidities, who have acceptable operative risks, who are motivated and well informed of the inherent risks of the procedure and of the need for lifelong follow-up) and a system for prioritizing patients on scheduled waiting lists.
Have available facilities and equipment adapted to the specifi c profi le of the patients concerned (e.g., recovery rooms, intensive-care units, beds and furniture, diagnostic investigation tables, operating tables, and adapted surgical instruments).
Have an experienced multidisciplinary team capable of supplying the full range of care and services tied to this type of
x
treatment: surgical team, psychologist, nutritionist, medical specialists (e.g., diabetologists, cardiologists, pneumologists).
Provide closely monitored lifelong follow-up, and cover the physical and psychological dimensions of this treatment, which consequently includes consultations linked to the need for plastic surgery.
3) It is recommended that a Québec registry on morbid obesity and its management be established. This registry will offer key support in implementing a regional follow-up program for operated patients by linking the different health-care structures (hospitals, health centres) and by including specifi c patient education on nutritional approaches appropriate for this type of patient. This data source will make it possible to determine the prevalence and categorization of the different patients, to evaluate the surgical procedures that are currently being performed and to rule on the new bariatric-surgery approaches.
xi
AGB Adjustable gastric banding
AHAL Ad hoc alimentary limb
AHFMR Alberta Heritage Foundation for Medical Research
AHRQ Agency for Healthcare Research and Quality
AHS Ad hoc stomach
ALOS Average hospital length of stay
ANAES Agence Nationale d’Accréditation et d’Évaluation en Santé
AOT Average operating time
ASBS American Society of Bariatric Surgery
ASERNIP-S Australian Safety and Effi cacy Register of New International Procedures – Surgical
BMI Body mass index (kg/m2). It is calculated by dividing weight (in kilograms) by height squared (in metres).
BPD Biliopancreatic diversion (open procedure)
BPD-DS Biliopancreatic diversion with duodenal switch
DG Distal gastrectomy
DS Duodenal switch
EWL Excess weight loss
FDA Food and Drug Administration
GB Gastric banding
GBP Gastric bypass (open procedure)
GPI Genuine Progress Index
INAHTA International Network of Agencies for Health Technology Assessment
LAGB Laparoscopic adjustable gastric banding
LBPD Laparoscopic biliopancreatic diversion
LBPD-DS Laparoscopic biliopancreatic diversion with duodenal switch
LGBP Laparoscopic gastric bypass
LRYGB Laparoscopic Roux-en-Y gastric bypass
LSAGB Laparoscopic Swedish adjustable gastric banding
LVBG Laparoscopic vertical banded gastroplasty
MAS Medical Advisory Secretariat
LIST OF ABBREVIATIONS
xii
MSAC Medical Services Advisory Committee
MUHC McGill University Health Centre
NAGB Non-adjustable gastric banding
NHMRC National Health and Medical Research Council
NICE National Institute for Clinical Excellence
OHTAC Ontario Health Technology Advisory Committee
QAGS Québec Association of General Surgeons
QALY Quality-adjusted life year
RYGB Roux-en-Y gastric bypass
SAGB Swedish adjustable gastric banding
SF-36 36-Item Short Form Health Survey
SOS Swedish Obese Subjects Intervention Study
TEC Technology Evaluation Center
VBG Vertical banded gastroplasty or silastic ring vertical gastroplasty
VBG-RYGB Vertical banded gastroplasty combined with Roux-en-Y gastric bypass
WHO World Health Organization
xiii
GLOSSARY
AnastomosisConnection between two vessels and, by extension, between two conduits of the same type or between two nerves. It may be natural or surgically created.
ConversionIn this report, conversion refers to a surgical intervention that begins as a laparoscopic procedure and is completed as an open procedure.
DentitionSet of natural teeth.
Dumping syndromeSyndrome involving the rapid early gastric emptying of the operated stomach (partial gastrectomy and/or gastrojejunal anastomosis). It occurs when food or liquid passes too quickly into the intestine, causing digestive problems, discomforts, etc.
Excess weightExcess weight in relation to ideal weight calculated according to height and sex.
Excess weight loss (EWL) Excess weight loss achieved through diet or through medical or surgical treatment. EWL is measured by weight units (pounds or kilograms) or by a percentage (initial weight – current weight) / (initial weight – ideal weight).
Ideal weightThis weight is based on the tables produced by the Metropolitan Life Insurance Company (1979). It is evaluated according to average values that take into account height and sex.
Laparoscopy Visual examination directly in the abdominal cavity previously distended by means of an endoscope introduced through the abdominal wall for diagnostic or therapeutic purposes.
LaparotomySurgical incision through the abdominal wall and peritoneum. (Also called open surgery.)
PlicationSurgical technique which consists in folding an anatomical structure or organ to modify its position, shape or function, or to modify the position, shape or function of an adjacent organ.
Quality-adjusted life year (QALY)Calculation method allowing situations to be compared in relation to two criteria taken into account simultaneously, that is, effi cacy (number of life years gained) and the quality of life of those years.
SecretinHormone produced in the duodenum that activates the secretion of pancreatic juices (especially alkaline salts) and, to a lesser extent, bile, intestinal juices and saliva.
xiv
TABLE OF CONTENTS
MISSION .......................................................................................................................................................... i
FOREWORD .................................................................................................................................................. iii
ACKNOWLEDGEMENTS ............................................................................................................................ iv
SUMMARY ..................................................................................................................................................... v
LIST OF ABBREVIATIONS ......................................................................................................................... xi
GLOSSARY .................................................................................................................................................. xiii
1 INTRODUCTION ...................................................................................................................................... 1
1.1 Defi nition of obesity .......................................................................................................................... 1
1.2 Burden of obesity ............................................................................................................................... 1
1.2.1 Prevalence .............................................................................................................................. 11.2.2 Mortality and economic impact .............................................................................................. 2
1.3 Role of surgical treatment in the therapeutic approach to obesity ..................................................... 2
1.4 Objective ............................................................................................................................................ 3
2 METHODOLOGY ..................................................................................................................................... 4
3 SURGICAL TECHNIQUES ...................................................................................................................... 6
3.1 Objectives of the surgical treatment of morbid obesity ..................................................................... 6
3.2 Description of the techniques ............................................................................................................ 6
3.2.1 Gastric-restriction techniques ................................................................................................. 63.2.2 Gastric bypass ......................................................................................................................... 83.2.3 Laparoscopic techniques ...................................................................................................... 103.2.4 Bariatric surgery with hand-assisted laparoscopy .................................................................11
4 STUDY OUTCOMES .............................................................................................................................. 12
4.1 Surgical vs non-surgical treatment of morbid obesity ..................................................................... 12
4.2 Effi cacy of the different surgical procedures ................................................................................... 13
4.2.1 By type of procedure ............................................................................................................ 134.2.2 Comparison of the techniques .............................................................................................. 24
4.3 Complications .................................................................................................................................. 30
4.3.1 Complications from gastric bypass ...................................................................................... 314.3.2 Complications from gastroplasty .......................................................................................... 314.3.3 Complications from biliopancreatic diversion ..................................................................... 314.3.4 Complications from laparoscopic procedures ...................................................................... 31
xv
4.4 Impact of bariatric surgery on obesity co-morbidities ..................................................................... 32
5 ECONOMIC OUTCOMES...................................................................................................................... 35
5.1 Results of the analysis of primary-data articles ............................................................................... 35
5.2 Modelling results ............................................................................................................................. 38
5.2.1 Model developed by Clegg ................................................................................................... 385.2.2 Model developed by Craig and Tseng .................................................................................. 39
5.3 Cost of bariatric-surgery procedures ................................................................................................ 39
5.4 Recapitulation of the economic evaluation ...................................................................................... 40
6 REVIEW OF THE VARIOUS HEALTH-TECHNOLOGY ASSESSMENT REPORTS ........................ 41
6.1 Bariatric surgery .............................................................................................................................. 41
6.2 Laparoscopic surgical procedures .................................................................................................... 43
7 DISCUSSION .......................................................................................................................................... 45
8 CONCLUSION ........................................................................................................................................ 49
8.1 General role of bariatric surgery ...................................................................................................... 49
8.2 The different surgical techniques ..................................................................................................... 49
8.3 Challenges for Québec’s medical practice ....................................................................................... 50
9 RECOMMENDATIONS .......................................................................................................................... 52
ABBREVIATIONS USED IN THE APPENDICES ..................................................................................... 53
APPENDIX A STATUS OF HEALTH TECHNOLOGIES : AETMIS CLASSIFICATION .................... 55
APPENDIX B OUTCOMES OF STUDIES ON VERTICAL BANDED GASTROPLASTY ................. 56
APPENDIX C OUTCOMES OF STUDIES ON GASTRIC BYPASS ..................................................... 58
APPENDIX D OUTCOMES OF STUDIES ON BILIOPANCREATIC DIVERSION ............................ 66
APPENDIX E OUTCOMES OF STUDIES ON GASTRIC BANDING ................................................. 70
APPENDIX F OUTCOMES OF STUDIES COMPARING DIFFERENT TYPES OF BARIATRIC SURGERY ................................................................................................... 85
APPENDIX G META-ANALYSIS OF THE IMPACT OF BARIATRIC SURGERY ON OBESITY CO-MORBIDITIES .................................................................................. 90
APPENDIX H DETAILED OUTCOMES OF ECONOMIC STUDIES ................................................... 92
APPENDIX I METROPOLITAN LIFE INSURANCE COMPANY TABLES ....................................... 98
APPENDIX J BAROS SCORING KEY .................................................................................................. 99
REFERENCES ............................................................................................................................................ 100
xvi
TABLES AND FIGURES
Figure 1 Vertical banded gastroplasty ...................................................................................................... 7Figure 2 Silicone gastric banding with injection reservoir and calibration tube ..................................... 7Figure 3 Basic gastric bypass .................................................................................................................. 8Figure 4 Roux-en-Y gastric bypass ........................................................................................................ 8Figure 5 Biliopancreatic diversion with distal gastrectomy (Scopinaro) ................................................ 9Figure 6 Biliopancreatic diversion with duodenal switch ..................................................................... 10Table 1 Difference between distal gastrectomy (DG) (Scopinaro) and
duodenal switch (DS) .............................................................................................................. 10Table 2 Outcomes of studies on open (VBG) or laparoscopic (LVBG)
vertical banded gastroplasty ................................................................................................... 14Table 3 Outcomes of studies on open or laparoscopic gastric bypass ................................................. 16Table 4 Outcomes of studies on biliopancreatic diversion ................................................................... 19Table 5 Outcomes of studies on adjustable gastric banding (AGB) .................................................... 21Table 6 Studies comparing biliopancreatic diversion with other open-surgery techniques ................. 25Table 7 Comparative study of open vertical banded gastroplasty ........................................................ 27Table 8 Outcomes of studies comparing VBG with other bariatric-surgery techniques ...................... 27Table 9 Outcomes of the study by Biertho et al. [2003] ...................................................................... 28Table 10 Outcomes of the comparative review of AGB and NAGB ..................................................... 29Table 11 Outcomes of the meta-analysis by Buchwald et al. [2004]
comparing the main types of bariatric surgery ........................................................................ 30Table 12 Outcomes of the meta-analysis by Maggard et al. [2005]
comparing the main types of bariatric surgery ........................................................................ 31Table 13 Type of procedures performed by the members of the International Bariatric
Surgery Registry ...................................................................................................................... 46Table B-1 Laparoscopic vertical banded gastroplasty ............................................................................. 56Table B-2 Comparisons between open and laparoscopic vertical banded gastroplasty ........................... 57Table C-1 Open gastric bypass ................................................................................................................ 58Table C-2 Laparoscopic gastric bypass ................................................................................................... 59Table C-3 Comparisons between open and laparoscopic gastric bypass ................................................ 62Table D-1 Open biliopancreatic diversion ................................................................................................ 66Table D-2 Laparoscopic biliopancreatic diversion ................................................................................... 68Table D-3 Comparisons between open and laparoscopic biliopancreatic diversion ................................ 69Table E-1 Swedish adjustable gastric banding......................................................................................... 70Table E-2 Comparisons between two laparoscopic gastric-banding techniques ..................................... 72Table E-3 Comparisons between the Lap-Band and the Heliogast bands ............................................... 74Table E-4 Laparoscopic adjustable gastric banding ................................................................................. 75Table E-5 Comparisons between open and laparoscopic adjustable gastric banding .............................. 82Table F-1 Comparisons between biliopancreatic diversion and other types of
open procedures....................................................................................................................... 85
xvii
Table F-2 Comparisons between vertical banded gastroplasty and other bariatric-surgery techniques .................................................................................................... 86
Table F-3 Comparisons between Roux-en-Y gastric bypass and laparoscopic adjustable gastric banding ....................................................................................................... 88
Table G-1 Impact of bariatric surgery on obesity co-morbidities ............................................................ 90Table H-1 Description of studies on bariatric surgery with an economic analysis .................................. 92Table I-1 Metropolitan Life Insurance Company table (women with medium frames) ......................... 98Table I-2 Metropolitan Life Insurance Company table (men with medium frames) .............................. 98
1
1 INTRODUCTION
Since 1998, in the wake of a report prepared by the World Health Organization (WHO) from the work of the International Obesity Task Force, obesity has been considered a major public-health problem and has even been declared a “global epidemic.” The WHO even calls it a chronic disease requiring long-term strategies for effective prevention and management: obesity is the result of complex interactions of metabolic, endocrine, genetic, socio-economic, environmental, cultural, psychological and behavioural factors. Obesity causes many diseases, including hypertension, hyperlipidemia, diabetes, some cardiovascular disorders, sleep apnea, osteoarthritis, some cancers, and even death [WHO, 2003].
1.1 DEFINITION OF OBESITYObesity is characterized by excess body fat and is generally defi ned by the body mass index (BMI), which takes into account weight and height. This index is calculated by dividing weight in kilograms by height in metres squared: it is therefore expressed in kilograms per square metre (kg/m2). The term obesity applies when the BMI is greater than or equal to 30 kg/m2. If the BMI is between 25 and 29.9 kg/m2, it is called overweight. According to the International Obesity Task Force, obesity can be divided into three categories: Obese Class I (BMI from 30.0 to 34.9 kg/m2), Obese Class II (from 35.0 to 39.9) and Obese Class III (greater than or equal to 40 kg/m2). Morbid obesity refers to Obese Class III, or to Obese Class II if it is associated with other co-morbidity factors.
1.2 BURDEN OF OBESITY
1.2.1 Prevalence
In the United States, the incidence of obesity (BMI ≥ 30) in the population aged 20 to 74 has
been growing steadily: the rate of obesity rose from 15.1% (age-standardized rate) in the years 1976–1980 to a mean of 23.3% between 1988 and 1994, and to 31.1% in the 1999–2002 period [NCHS, 2004].
In Canada, the percentage of obese people (BMI ≥ 30) in the household population aged 18 and older (excluding pregnant women) in 2004 was estimated to be 23.1%, while it was 13.8% in 1978–1979. These rates are respectively estimated to be 5.1% and 2.3% for people with BMIs between 35 and 39.9 kg/m2, and 2.7% and 0.9% when the BMI is greater than or equal to 40. Although the difference in the obesity rate between men (22.9%) and women (23.2%) is small, it is larger for people with BMIs greater than or equal to 40 kg/m2: 1.6% for men and 3.8% for women. The base data used to calculate these rates were obtained by directly measured height and weight, although an adjustment was necessary to minimize non-response bias (42.5%), (owing to those who did not respond to that part of the survey) [Tjepkema, 2005].
In Québec, according to the data from the 1998 Québec social and health survey, 29.0% (34.1% of men and 23.5% of women) of the population aged 20 to 64 was overweight (BMI ≥ 27 kg/m2). This rate rose to 12.7% (13.5% for men and 11.7% for women) for a BMI ≥ 30 kg/m2, and to 3.0% for a BMI ≥ 35 kg/m2 [Institut de la statistique du Québec, 2001]. Note that these data were collected through a self-administered questionnaire, a method which tends to yield lower obesity rates. In 2004, according to Canadian source data, the obesity rate (BMI ≥ 30 kg/m2) in the Québec household population aged 18 and older was 21.8%, the rate in women (22.7%) being higher than that in men (20.9%) [Tjepkema, 2005]. That publication does not provide more detailed statistics for each class of obesity in Québec.
2
1.2.2 Mortality and economic impact
In the United States, obesity causes a large number of deaths each year, owing both to its complications and to its co-morbidities, although its estimation sparked a major controversy. While a fi rst report by researchers from the Centers for Disease Control and Prevention (CDC) estimated that the number of deaths attributable to overweight and obesity was 365,000 per year [Mokdad et al., 2004], a second report by other CDC researchers yielded a fi gure of 111,909 for obesity alone (BMI ≥ 30) [Flegal et al., 2005]. Even if several factors explain a large part of this difference, it is motivating researchers to develop even more rigorous approaches to estimating mortality attributable to obesity.
Obesity carries an economic burden representing from 5.5% to 7.0% of total health-care spending [Thompson and Wolf, 2001]; it accounted for 27% of the rise in actual spending per person between 1987 and 2001 [Thorpe et al., 2004]. According to a study that examined the relationship between the BMI of people aged 18 to 65 and Medicare spending for these same people at age 65 and older, obese people (30 ≤ BMI < 35) and severely obese people (BMI ≥ 35) generated costs totalling US$9,612 and US$12,342 per person per year respectively, compared with US$6,224 for non-overweight people (18.5 ≤ BMI < 25) [Daviglus et al., 2004].
In Canada, for 1997 alone, direct medical costs attributable to obesity (BMI greater than or equal to 27) in adults were estimated to be $1.8 billion, or 2.4% of total direct medical costs [Birmingham et al., 1999]. A study by the research group Atlantic GPI (Genuine Progress Index), which adopted and refi ned Birmingham’s method, estimated Québec’s direct health-care costs to be $700 million, or 5.8% of the province’s health-care budget (1999–2000 fi scal year). Costs attributable to productivity losses were evaluated as being in excess of $800 million, and the sum of these two estimates could represent nearly 1% of Québec’s gross domestic product [Colman and Dodds,
2000]. Colman also provided the same estimates for seven other provinces.1
1.3 ROLE OF SURGICAL TREATMENT IN THE THERAPEUTIC APPROACH TO OBESITY The therapeutic approach to obesity is multi-faceted and complex. It requires an adapted treatment structure and the availability of a multidisciplinary team.
Management of obesity
Obesity management is based on a minimum of three key measures:
1) intensive patient education aimed at improving food patterns;
2) counselling on the need for regular physical activity; and
3) behavioural approaches designed to help people better regulate the lifestyle habits needing to be modifi ed [Kushner, 2003; NHLBI/NIH, 1998].
Weight-loss objectives must be clearly defi ned with the patient [Snow et al., 2005]. Physicians may suggest drug therapy for obese patients unable to meet their target objectives through diet and physical activity. The use of a pharmacological agent requires a doctor–patient discussion before such treatment is initiated. The side effects of the prescribed medication, the lack of long-term safety data and the modest weight loss associated with it are all points that must be covered [Snow et al., 2005]. According to a recent meta-analysis (April 2005) of the pharmacological treatment of obesity, the mean weight loss achieved, after adjustment for the placebo effect, is less than 5 kg after one year [Li et al., 2005].
Management of morbid obesity
In the latest WHO technical report on the prevention and management of obesity, surgery
1. Colman’s articles are available online: http://www.gpiatlantic.org/publications/health.shtml#obesity.
3
is considered to be the only effective treatment for morbid obesity. This type of surgery is called bariatric surgery (from the Greek word baros, which means weight). Besides its positive effects on weight loss and its acceptable rates of weight-loss maintenance, bariatric surgery is the treatment offering the best cost-effectiveness ratio in the medium term [WHO, 2003; Näslund et al., 2001]. Bariatric surgery encompasses a wide range of techniques, and the effectiveness of each is relatively well established. The choice of one technique over the other is subject to a number of criteria, such as the patient’s clinical and psychological characteristics, the availability of the appropriate infrastructure, the surgeon’s preference and the medical team’s expertise.
The growing “epidemic” of obesity and morbid obesity has prompted the medical body to take greater interest in bariatric surgery and to explore new treatment methods, not only in terms of the techniques themselves but also in terms of the surgical approach, especially laparoscopy. This development has led to a sharp rise in the number of surgical procedures being performed in this fi eld. In the United States, for example, the American Society of Bariatric Surgery (ASBS) reports that this fi gure reached 140,640 in 2004, or more than double the number recorded in 2002
(63,100 procedures) [Colwell, 2005]. This rise can be explained in part by the greater availability of services (e.g., membership in the American Society of Bariatric Surgery doubled between 2000 and 2002 [ASBS, 2001]), but also by a greater reliance on laparoscopy. Moreover, even if the criteria defi ning obesity and the established and generally recognized surgical-candidate profi le have not changed, waiting lists (which may vary according to the surgeon’s expertise) have been growing.
1.4 OBJECTIVEThis is the particular context in which AETMIS proposed to update its information on the effi cacy of the surgical techniques used in the treatment of morbid obesity, which it had already examined in a previous report when it was known as the Conseil d’évaluation des technologies de la santé [CETS, 1998]. In the present update, special attention will be given to laparoscopic techniques, which had barely begun to enter into practice at that time and had therefore not been examined. This report also responds to an assessment request from the Ministère de la Santé et des Services sociaux asking AETMIS to examine the evolution of bariatric surgery and the best patient-care conditions for people with morbid obesity.
4
2 METHODOLOGY
This study is interested in both the reports produced by the different assessment agencies that have dealt with the topic of bariatric surgery and the more recent studies published since the release of the fi rst AETMIS report. A literature search was performed using the major databases: Medline, Cochrane Library, HealthStar (a database that ceased to exist in October 2003) and HTA Database (a health-technology assessment database created jointly by the Centre for Reviews and Dissemination based at the University of York in England, and the INAHTA (International Network for Agencies in Health Technology Assessment). Keywords used were obesity, morbid obesity, surgical treatment, bariatric surgery, gastroplasty, gastric bypass, gastric banding.
The search identifi ed a large number of studies published since 1998. However, given the scarcity of controlled trials, whether randomized or not, and the predo minance of case series, studies were selected by means of a simplifi ed grid containing the following elements:
study design; publication date; number of patients treated; length of follow-up; relevance of clinical and economic outcome
measures.
The main outcome measures for bariatric surgery selected for this analysis were the following (not all were applicable to each of the studies):
Clinical effi cacy: excess weight loss (EWL), defi ned in relation to ideal weight,2 or a decrease in body mass index (BMI);
2. In studies, ideal weight is often based on the tables produced by the Metropolitan Life Insurance Company. It is evaluated according to mean values that take into account height and sex (see Appendix I).
Safety: complications characterized by their time of onset (short, medium or long term) and by their type and severity; Co-morbidity: reduction or not of diseases
associated with obesity; Consumption of health goods or services
or other resources: hospitalization days, mean length of hospital stay, operating time (average length of the surgical procedure) (these criteria can be evaluated as clinical health-status indicators or resource-utilization indicators from an economic perspective), days of absenteeism from work, etc.; Cost effectiveness: resource costs or savings;
cost-effectiveness and cost-utility ratios.
To defi ne the study designs, we used a basic classifi cation system adapted from the one proposed by the Agence Nationale d’Accréditation et d’Évaluation en Santé3 [ANAES, 2001]. This system identifi es a large number of non-comparative studies, regardless of quality:
Randomized comparative study (RC); Non-randomized comparative study (C),
which may be either controlled (CC) or non-controlled (NCC), depending on how much effort was made to ensure that the study groups were as comparable as possible. In addition, the temporal aspect, either prospective (P) or retrospective (R), is indicated by the addition of its corresponding letter: PCC or RCC PNCC or RNCC
Prospective non-comparative study (P); Retrospective non-comparative study (R).
3. On January 1, 2005, this agency was granted additional mandates and renamed the Haute Autorité de Santé.
5
Here are a few salient facts about the articles selected for this analysis.
Most of the published studies deal with surgical techniques that have established effi cacy, or with new approaches, especially laparoscopic procedures. This analysis selected 83 studies published
between January 1998 and April 2005; however, after 2000, most studies deal with adjustable gastric banding. Most of the studies rely on a methodological
design that from the outset is not conducive to achieving the most valid results (non-randomized studies, mostly non-comparative).
Less commonly, some of the studies compare various open or laparoscopic techniques (different types of adjustable bands) or yet again two approaches to the same surgical procedure. The eight randomized studies compare
the laparoscopic and open approaches for the same surgical technique, or different techniques for implanting gastric bands. Two meta-analyses published in 2004 and
2005, respectively, compare the effi cacy of the main procedures. One of them also deals with their impact on the progression of obesity co-morbidities. A single study compares surgical treatment
with non-surgical treatment, and has given rise to several publications dealing with different aspects of the study.
6
3 SURGICAL TECHNIQUES
gastric banding: a fi xed or adjustable band is implanted to form a small-volume gastric reservoir;
hybrid or mixed techniques. These combine gastric restriction with the principle of intestinal malabsorption by creating a bypass system or a diversion system. This group includes: gastric bypass, biliopancreatic diversion.
3.2.1 Gastric-restriction techniques
3.2.1.1 GASTROPLASTY According to the level of the procedure, there are two types of gastroplasty:
Horizontal gastroplasty: Many variants have been proposed, from Mason’s model, which consists in partitioning the stomach horizontally (lesser curvature) by leaving a narrow outlet (stoma) for food passage, to Gomez’s model, which consists in placing a staple line transversely across the proximal part of the stomach and creating a reinforced opening at the level of the greater curvature. Although the sutures were reinforced, the staples often failed and the conduits re-expanded. This type of procedure is no longer performed.
Vertical banded gastroplasty: As described by Mason, vertical banded gastroplasty has the advantage of being easy to perform. This technique consists in creating a small gastric pouch (15–20 ml) that empties into the residual portion of the stomach through a small channel built along the lesser curvature of the stomach and calibrated by means of a polypropylene collar (Figure 1). Patients must not only have good dentition, they must also restrict themselves to eating food in small quantities. Those patients affl icted with frequent vomiting suffer from vitamin and mineral defi ciencies,
3.1 OBJECTIVES OF THE SURGICAL TREATMENT OF MORBID OBESITY Bariatric surgery is based on the principle of restricting food intake (by decreasing the gastric reservoir) or of reducing nutrient absorption (by decreasing the contact time between the food bolus and the digestive juices and bile by shortening the section of the intestine that promotes such contact). The effi cacy of surgical treatment is often measured as a percentage of excess weight loss (EWL) in relation to ideal weight: a technique is considered effective if the EWL is greater than or equal to 50% (based on the criteria defi ned by Reinhold [1982]) [CETS, 1998; Hall et al., 1990]. The outcome is excellent if the EWL is greater than 75%, good if between 50% and 75%, and fair if between 25% and 50%.
Most medical associations recommend that bariatric surgery be reserved for obese patients who meet a certain number of criteria, such as:
body mass index (BMI) greater than or equal to 40, or 35 if associated with other co-morbidity factors; presence of severe co-morbidities; failure of a diet followed for several years; acceptance of long-term (even lifelong)
follow-up and the inherent risks involved in this type of procedure.
3.2 DESCRIPTION OF THE TECHNIQUESThe two most commonly performed bariatric-surgery procedures are:
techniques based on gastric restriction. These decrease food intake by reducing gastric capacity. This group includes: gastroplasty: a pouch or partition is
created by stapling or banding,
7
including iron. Lost weight is frequently regained. Gastric leaks are considered surgical emergencies (risk of septicemia and death).
3.2.1.2 GASTRIC BANDING
Silicone gastric banding has been performed for nearly 20 years. The objective of this procedure is to achieve, by means of a restrictive mechanism, a reduction in dietary intake by creating a small gastric pouch. The gastric fi bres of this pouch stretch more quickly, which stimulates the vagus nerve fi bres and triggers satiety refl exes. In 1987, a silicon part was added to this system, which helps adjust and calibrate the collar that controls the speed of food passage from the upper gastric pouch to the lower portion of the stomach (hour-glass shape). The band can be adjusted by injecting liquid into a subcutaneous reservoir (Figure 2).
This procedure is relatively easy to perform and also reversible, which explains why it is used so frequently. There are a few types of adjustable gastric bands, which basically have differing elasticity and closure systems. The main trademarks are Lap-Band, SAGB and Heliogast. Only the Lap-Band is commercially available in Canada.
Improvements in surgical techniques (type, fi xation method and band placement) and the laparoscopic approach have both contributed to signifi cantly reducing complications. Of these, dilatation of the newly formed gastric pouch, whether or not associated with band slippage, remains the most frequent (5–20% of cases). Adjustable gastric banding is the bariatric-surgery technique that is most often performed laparoscopically.
FIGURE 1 FIGURE 2
Vertical banded gastroplasty Silicone gastric band with injection reservoir and calibration tube
8
3.2.2 Gastric bypass
The key objective of gastric bypass is to reduce the digestion of absorbed nutrients. This procedure induces marked weight loss but is associated with complications that are more or less severe, depending on the technique used. Techniques combining gastric restriction and malabsorption are used more frequently with morbidly obese patients with BMIs greater than 50.
3.2.2.1 GASTRIC-BYPASS PROCEDURES
Since the 1960s, several modifi cations have been made to Mason’s gastric bypass (Figure 3), which consisted in creating a small proximal gastric pouch by surgically dividing the stomach and its duodenal opening (gastric bypass with retrocolic loop). Other gastric-bypass models consist in creating a reservoir by stapling. This type of procedure leads to leaks and the dumping syndrome (an adverse event that
occurs with all hybrid techniques that include gastric bypass).
3.2.2.2 ROUX-EN-Y GASTRIC BYPASS
Now considered the gold standard in bariatric surgery, this technique consists in creating a small proximal gastric pouch anastomosed to a segment of the jejunum. This system is shaped like a Y; hence the name Roux-en-Y gastric bypass (RYGB). This technique has several variants, one of which involves creating a small pouch with a line of staples (Figure 4). Different RYGB procedures are performed, including short-limb (50–100 cm) Roux-en-Y gastric bypass and long-limb Roux-en-Y gastric bypass, considered by some authors to be biliopancreatic diversions. The main complications associated with RYGB are metabolic, and they require patients to take supplemental vitamins (especially vitamin B12) and minerals (calcium and iron).
FIGURE 3 FIGURE 4
Basic gastric bypass Roux-en-Y gastric bypass
9
3.2.2.3 BILIOPANCREATIC DIVERSION
3.2.2.3.1 Scopinaro’s biliopancreatic diversion
Scopinaro et al. [2000] showed that it was possible to decrease nutrient absorption by shortening the intestine and to reduce fat (lipids) absorption by diverting bile juices and decreasing the contact between food and enzymes. Scopinaro’s biliopancreatic diversion shares many similarities with RYGB, but differs in that it involves a complete resection of the lower, or distal, portion of the stomach (Figure 5). The Scopinaro procedure is in fact a biliopancreatic diversion combined with a distal gastrectomy (BPD-DG).
3.2.2.3.2 Biliopancreatic diversion with duodenal switch
Unlike Scopinaro’s biliopancreatic diversion with distal gastrectomy, this technique applies a duodenal switch with an end-to-end duodeno-ileal anastomosis (Figure 6). In theory,
biliopancreatic diversion with duodenal switch (BPD-DS) has the advantage of permitting near-normal functioning of the stomach and avoiding a gastro-enteric anastomosis [Hess and Hess, 1998]. This type of procedure conserves normal vagal innervation (control of the satiety centre) and preserves the antropyloric junction, which plays a role in triggering the secretion of secretin (Table 1). A duodenal switch involves resecting the gastric fundus, helping to reduce the secretion of hydrochloric acid. Maintaining both the integrity of this junction and a small portion of the duodenum (a few centimetres of the fi rst duodenal loop) seems to prevent gastroduodenal ulcers, perforations and the dumping syndrome [DeMeester et al., 1987]. In 1992 Marceau et al., from the Hôpital Laval in Quebec City, modifi ed this technique by eliminating plication in creating the anastomosis and by replacing it with an end-to-end anastomosis [Baltasar et al., 1995; Marceau and Biron, 1993]. Post-operative complications are most often metabolic, and supplemental vitamins and calcium are required.
FIGURE 5
Biliopancreatic diversion with distal gastrectomy (Scopinaro)
10
FIGURE 6
Biliopancreatic diversion with duodenal switch
TABLE 1
Difference between distal gastrectomy (DG) (Scopinaro) and duodenal switch (DS)
DG (SCOPINARO) DS*
Insertion Stomach/ileum Duodenum/ileum
Common limb 50 cm 100 cm
Vagal innervation Not preserved Preserved
* In 1992 Marceau abandoned plication for an end-to-end anastomosis.
3.2.3 Laparoscopic techniques
In theory, laparoscopic surgery techniques have the advantages of shorter recovery times and lower peri-operative and post-operative complications. Other than the surgical approach itself, laparoscopic techniques are generally identical to open-surgery techniques. While the most commonly used laparoscopic techniques since 1990 are based on gastric restriction, procedures targeting malabsorption are recent and still infrequent.
Vertical banded gastroplasty (VBG) and adjustable gastric banding (AGB) are the
most frequently performed laparoscopic procedures. The LVBG4 technique requires considerable expertise in both bariatric surgery and laparoscopy, and it is increasingly being replaced by LAGB, a technique that has been greatly improved in recent years.
Gastric bypass (LGB) is also one of the bariatric surgeries most frequently performed laparoscopically. It is therefore expected that LRYGB performed by experienced surgeons will become a therapeutic option to consider.
4. LVBG: laparoscopic VBG. The same type of abbreviation is used for the other laparoscopic techniques: LAGB, LDG, LRYGB and LBPD.
11
Laparoscopic biliopancreatic diversion (LBPD) is used only in exceptional cases because of its complexity and recent entry into the arsenal of bariatric surgery.
3.2.4 Bariatric surgery with hand-assisted laparoscopy
This technique combines two surgical approaches. First, a small incision (6–8 cm) is made to the abdomen. This incision, which
is large enough to fi t a hand, allows the surgeon to palpate the organs and makes it easier to mobilize them (e.g., the colon). The essential part of the procedure is then done laparoscopically (resection, ligature, etc.). This novel approach remains limited to some centres or is used as a training tool before surgeons proceed to exclusively laparoscopic procedures [DeMaria et al., 2002b; Bleier et al., 2000; Naitoh et al., 1999; Memon and Fitzgibbons, 1998; Watson and Game, 1997].
12
4 STUDY OUTCOMES
4.1 SURGICAL VS NON-SURGICAL TREATMENT OF MORBID OBESITY
As part of the major prospective SOS study, Ågren et al. [2002b] compared patients who had undergone open surgery (vertical banded gastroplasty, gastric banding, gastric bypass) with patients who had followed medical (non-surgical) therapy. The SOS study was initially designed to compare, over a 20-year period, 2010 surgically treated obese patients and 2037 matched patients who were offered conventional treatment in primary-care centres. Control subjects were matched according to 18 basic anthropometric variables or to variables linked to risk factors for morbidity and mortality. The researchers fi nally studied the fi rst 962 consecutive obese patients (BMI > 34 kg/m2 for men and > 38 kg/m2 for women) aged 37 to 60. These patients were recruited between 1987 and 1991, and were followed for at least six years. The study examined therapeutic effectiveness in relation to weight loss and hospital costs associated with each of the treatment options: bariatric surgery vs services commonly offered to obese patients in primary-care centres (conventional treatment). No specifi c information was given on the treatments actually received by the “conventional treatment” group. Comparisons of hospital costs will be covered in Chapter 5 on economic outcomes.
Results show that, at one year, the surgical patients had lost more weight (mean weight loss of 25.1 ± 10.1% for 450 patients) than those in the conventionally treated group (mean weight loss of 0.7 ± 6.5% for 425 patients). This major benefi t of bariatric surgery was maintained after six years (mean weight loss of 16.7 ± 11.8% for 401 patients, compared with a mean weight gain of 0.9 ± 10.1% for 344 non-surgical patients).
Virtually all the studies on bariatric surgery are of adult subjects after the failure of conventional treatment (diet and medical therapy) and after review of the patient’s psychological profi le. Until recently, bariatric surgery for adolescents (ages 11 to 17) with morbid obesity was used only in exceptional circumstances [Abu-Abeid et al., 2003]. A recent meta-analysis of the surgical treatment of morbid obesity identifi ed 12 case series with a combined total of 172 adolescents. However, insuffi cient data and a lack of comparators do not permit valid conclusions to be drawn [Maggard et al., 2005].
Among all the articles identifi ed and selected for the present assessment, a single study compared the effi cacy of surgical vs non-surgical treatment for patients with morbid obesity: the Swedish Obese Subjects (SOS) Intervention Study [Sjöström et al., 2004; Ågren et al., 2002b]. The outcomes of this comparative study, which is of good quality although non-randomized, confi rm that surgery has a role to play in the management of morbid obesity. This study includes a clinical component and an economic component. The economic component will be covered in Chapter 5.
The studies dealing with the impact of bariatric surgery on obesity co-morbidities will be analyzed in section 4.4.
This section begins by examining study outcomes by type of procedure and then deals with comparisons of the effi cacy of different techniques. Given that the techniques used are the same, whether open or laparoscopic, the effi cacy outcomes will not be presented separately by approach. However, special attention will be given to post-operative length of hospital stay and to some complications that do depend on the type of surgical approach taken.
13
A more recent article by Sjöström et al. [2004] on the same study indicates that weight loss measured at 10 years was roughly the same at 6 years (16.1% vs 16.7%). The control group had had a mean weight gain of 1.6% at 10 years, and the difference between the two groups is statistically signifi cant (p < 0.001). The data were based on 641 surgical patients and 627 controls. It should be mentioned that the lost to follow-up rates were high, 24.7% and 26.4% respectively. Finally, the post-operative mortality rate among the 2010 surgical patients remained very low (0.25%).
In terms of weight reduction alone, the SOS study shows that bariatric surgery is more effective than non-surgical treatment, and the difference is statistically signifi cant. Even if the preliminary outcomes (measured at one or two years) are higher, the benefi ts achieved in terms of weight loss are nevertheless sustained. In addition, as will be seen in section 4.4 on the impact of bariatric surgery on co-morbidities and in Chapter 5 on the economic component, the benefi ts of bariatric surgery go beyond weight loss.
In their summary report of a retrospective comparative study recently conducted in Québec, Christou et al. [2004] present their comparative outcomes resulting from a 20-year (1983–2002) follow-up of 6781 patients with morbid obesity (1035 surgical patients and 5746 patients matched by age and sex but not surgically treated). The mean initial BMI was 50 kg/m2 (range, 36–98). The two cohorts were followed up for a maximum of fi ve years, and the data were extracted from provincial health-insurance databases on hospitalizations, medical services and medications. Of the surgical patients, 194 underwent vertical banded gastroplasty (VBG), 68 then underwent Roux-en-Y gastric bypass (RYGB), and 841 underwent isolated RYGB (21 of which were performed laparoscopically). Seven different surgeons affi liated with the same university health centre took part in the procedures during the 16.4 study years.
There were no signifi cant differences between the two groups in terms of age, sex and length of follow-up. Overall, after a mean follow-up of 5.3 years, excess weight loss (EWL) after bariatric surgery was 67.1% (standard deviation: 23.7%; p < 0.001). This reduction tended to persist until 16 years after surgery. Nevertheless, the follow-up rate, which had stayed relatively stable for 11 years, dropped dramatically thereafter. According to Reinhold’s criteria (EWL ≥ 50%), the authors consider the procedure successful for 83% of the morbidly obese patients and for 73% of the super-obese patients (BMI > 50).
4.2 EFFICACY OF THE DIFFERENT SURGICAL PROCEDURES
4.2.1 By type of procedure
4.2.1.1 VERTICAL BANDED GASTROPLASTY
The outcomes (excess weight loss) of the studies of vertical banded gastroplasty (VBG) vary according to the settings and the follow-up periods under consideration; these studies are only case series (except for the randomized trial by Dávila-Cervantes comparing open vs laparoscopic VBG (Table 2).
A prospective study of 60 patients indicates that the BMI, initially 44.4 kg/m2, dropped to 37 kg/m2 at 36 months, and that the outcomes were practically identical, whether the gastroplasty was performed as an open procedure or laparoscopically [Näslund et al., 1999] (Table B-1, Appendix B). In two studies, weight loss was maintained in patients who were followed up for fi ve years [Hell and Miller, 2000; Capella and Capella, 1996]. With the laparoscopic approach, the mean hospital stay is generally from three to four days, and this approach does not usually lead to the complications that arise from open surgery (Tables B-1 and B-2, Appendix B).
14
TABLE 2
Outcomes of studies on open (VBG) or laparoscopic (LVBG) vertical banded gastroplasty*
AUTHORS AND YEAR (STUDY DESIGN)
TYPE OF PROCEDURE
BMI(kg/m2)
EWL
Capella and Capella, 1996 (RNCC)
VBG (n = 328) 52 ± 9 30–42 months: 48 ± 23%54–66 months: 47 ± 23%
Alle et al., 1998 (R)
LVBG (n = 261) 43.3 18 months: 75%
Salval et al., 1999 (R)
LVBG (n = 87) 43.8 18 months: 75%
Toppino et al., 1999 (R)
LVBG (n = 170) 43.9 36 months: 61%
Bajardi et al., 2000 (RNCC)
VBG (n = 93) 48.7 (37–65.6) 24 months: 48%
Hell and Miller, 2000 (PNCC)
VBG (n = 101) 46.9 ± 9.0 2 years: 61% (n = 98)5 years: 69% (n = 15)
Hell et al., 2000 (PCC)
VBG (n = 30) 46.9 ± 9.9 40.1 ± 8.3 months: 0–24% (n = 1) 50–74% (n = 15) 25–49% (n = 12) 75–100% (n = 2)
Dávila-Cervantes et al., 2002 (CR) VBG (n = 14)
LVBG (n = 16)43 (37–50)45 (38–50)
12 months (NS): 55% (30–88) 47% (22–97)
* See appendix for detailed data in Tables B-1, B-2, F-1 and F-2.
PCC: prospective, controlled comparative study; PNCC: prospective, non-controlled comparative study; RNCC: retrospective, non-controlled comparative study; RC: randomized comparative study; R: retrospective non-comparative study; n: number of patients; NS: non-signifi cant difference.
patients enrolled in this study greatly reduces the clinical bearing of these outcomes (Table B-2, Appendix B).
4.2.1.2 ROUX-EN-Y GASTRIC BYPASS AND VARIANTS
Analysis of the outcomes in the different studies on this topic confi rms the effi cacy of Roux-en-Y gastric bypass (RYGB), evaluated according to the criteria of stable weight loss and a low complication rate (Table 3). In clinical practice, RYGB is now considered the gold standard of open bariatric-surgery techniques. The EWL achieved generally exceeds 50% and even nears 80% two years after surgery, and according to studies with longer follow-up periods, this weight loss seems to be maintained [Hell et al.,
The article on the most recent randomized study comparing open vertical banded gastroplasty (VBG) with the laparoscopic approach (LVBG) provides outcomes for a limited number of patients (14 for the open-surgery group and 16 for the laparoscopy group) for a relatively short follow-up period. At one year, there was no signifi cant difference in terms of EWL: 55% for the VBG group and 47% for the LVBG group (Table 2). The mean hospital stay was similar for both groups (four days). Patients in the LVBG group required fewer analgesics on the fi rst post-operative day, stopped taking them sooner (two days vs three days for the VBG group) and returned to their normal activities sooner than those in the VBG group [Dávila-Cervantes et al., 2002]. Nevertheless, the limited number of
15
2000; Wittgrove and Clark, 2000; Rabkin, 1998; Capella and Capella, 1996]. The mean hospital stay associated with this procedure is from four to six days, according to the two studies providing that fi gure [Westling et al., 2002; Fobi et al., 1998]. The differences in weight-loss outcomes largely depend on the patients’ characteristics and on the surgical team’s learning curve. In 57% of the super-obese patients studied by MacLean et al. [2000], the EWL was greater than 50% (with improved quality of life in most cases), while another team, which had combined VBG with RYGB, reported the same outcome in 97.6% of the patients in that particular group [Capella and Capella, 2002].
A randomized study on gastric bypass compared the laparoscopic approach with open surgery. A total of 104 patients were randomly assigned to open surgery (51 subjects) or to laparoscopy (53). There were no signifi cant differences in terms of male-female ratio, age, initial weight and BMI (see Table C-3 in Appendix C). The authors found no signifi cant difference between the two groups after a mean follow-up of 23 months, but the results pertaining to changes in BMI are presented only graphically. The patients assigned to laparoscopy had shorter operating times (186.4 vs 201.7 minutes: p < 0.05) and hospital stays (5.2 vs 7.9 days: p < 0.05). Early complications (less than 30 days post-operatively) rose to 22.6% in the laparoscopy group and 29.4% in the open-surgery group. Although the difference is not signifi cant, it is worth mentioning that the two groups experienced different types of
complications and that one death occurred in the open-surgery group during a re-operation. The conventionally treated patients had more late complications (24%, mainly post-operative abdominal-wall hernias) than those who had been operated laparoscopically (11%: p < 0.05). Two deaths were nevertheless recorded in the laparoscopy group, one intra-operatively and the other presumably due to pulmonary embolism [Luján et al., 2004].
Despite the major advantages that the laparoscopic approach may offer, the authors conclude that it has a more complex learning curve and that the risk of post-operative complications is greater in the early stages of the curve.
For laparoscopic gastric-bypass procedures, the outcomes expressed as EWL are rather similar, varying from 68.8% to 82% at 12 months, depending on the study (Table 3). This similarity was also observed in a study with a longer follow-up period [Wittgrove and Clark, 2000]. Hospital stays are generally shorter than with conventional GBPs and RYGBs (1.6–4 days) (Tables C-2, C-3 and F-3, Appendices C and F). One study, which did not use EWL as an outcome measure, reported a mean initial weight loss of 35% and a decrease in BMI from 51.5 to 32 kg/m2, at 18 months of follow-up. This study included 195 patients: 159 underwent laparoscopic procedures and 36, open surgery [Brody, 2004] (Table C-2, Appendix C). Results of the same order had previously been observed in a series of 52 patients, the mean BMI having dropped from 55 to 34 kg/m2 [Gagner et al., 1999] (Table C-2, Appendix C).
16
TABLE 3
Outcomes of studies on open or laparoscopic gastric bypass*
AUTHORS AND YEAR (STUDY DESIGN)
TYPE OF PROCEDURE
BMI(kg/m2)
EWL
Capella and Capella,1996 (RNCC)
VBG-RYGB (n = 560) 52 ± 9 30–42 months: 70 ± 19%54–66 months: 62 ± 17%
Fobi et al., 1998 (R)
RYGB (n = 944) 46 24 months (mean): 80%
Rabkin, 1998 (RNCC)
RYGB (n = 138) 49 24 months: 74%48 months: 63%
Wittgrove and Clark, 2000 (R)
LRYGB (n = 500) Unspecifi ed 54 months: 73%
Hell et al., 2000 (PCC)
RYGB (n = 30) 45.2 ± 8.2 60 ± 8.1 months: 0–24% (n = 0) 25–49% (n = 2) 50–74% (n = 6) 75–100% (n = 22)
Higa et al., 2000 (R)
LGBP (n = 400) 46 12 months: 69%
MacLean et al., 2000 (R)
RYGB (n = 243) 49 66 ± 18 months: ≥ 50%(93% of obese or morbidly obese patients and 57% of super-obese patients)
Schauer et al., 2000 (P)
LRYGB (n = 275) 48 30 months: 77%
Nguyen et al., 2001 (RC) RYGB (n = 76)
LRYGB (n = 79)48.4 ± 5.447.6 ± 4.7
6 months (p = 0.01): 1 year (p = 0.07):45 ± 12% 62 ± 14%54 ± 14% 68 ± 15%
Capella and Capella,2002 (R)
VBG-RYGB (n = 652) 50 (38–86)(42% of super-obese patients)
5 years:77 ± 17% (all patients)74 ± 15% (super-obese patients)
DeMaria et al., 2002b (R)
LGBP (n = 281)Hand-assisted LGB (n = 25)
48.1 ± 6.5
(40.3–71) 12 months: 70 ± 15%
Frezza et al., 2002 (R)
LRYGB (n = 238) 48 (39–67.9) 12 months: 68.8%
Biertho et al., 2003 (RNCC)
LRYGB (n = 456) 49.4 ± 8.3 6 months: 51.6% (88% of patients)12 months: 67.0% (57%)18 months: 74.6% (37%)
Courcoulas et al., 2003a (PCC) GBP (n = 80)
LGBP (n = 80)4644
6 months (p < 0.05): 1 year (NS):45% 64.9%52.6% 69.2%
Stoopen-Margain et al., 2004 (P)
LRYGB (n = 100) 50 ± 9(33% of patients with BMIs > 50)
6 months: 47 ± 2% (n = 82)12 months: 62 ± 4% (n = 70) 18 months: 66 ± 5% (n = 63) 24 months: 67 ± 8% (n = 35)
* See appendix for detailed data in Tables C-1, C-2, C-3, F-1, F-2 and F-3.P: prospective non-comparative study; PCC: Prospective, controlled, non-randomized comparative study; R: restrospective non-comparative study; RC: randomized comparative study; RNCC: retrospective, non-controlled, non-randomized comparative study; n: number of patients; NS: non-signifi cant difference.
17
Two randomized studies compared open vs laparoscopic gastric-bypass procedures. The fi rst presented outcomes related to excess weight loss for 76 patients who had undergone open RYGB and for 79 patients who had undergone LRYGB over a mean follow-up of 9.6 months (1–23 months) [Nguyen et al., 2001]. In terms of excess weight loss measured at one year, laparoscopy tended to be more effective, although the difference is not signifi cant (68 ± 15% vs 62 ± 14% for the open-surgery group; p = 0.07) (Table 3). Even though operating times were longer in the LRYGB group, patients lost less blood and required less intensive care than those in the open RYBG group (signifi cant differences). The mean length of hospital stay and the return to normal activities were shorter with laparoscopy: 3 hospitalization days and 8.4 recovery days (vs 4 hospitalization days and 17.7 recovery days with open surgery). From a quality of life perspective, even if the SF-365 scores at one month were better for the LRYGB patients, at three months they were similar to those in the group assigned to open surgery (Table C-3, Appendix C).
These outcomes are supported by the other study, which was conducted by Westling and Gustavsson [2001] (since these authors do not provide their outcomes in relation to EWL, the details of this study are presented in Table C-3 in Appendix C). The outcome differences between the two groups are not statistically signifi cant: at one year, the BMI fell from 41 to 27 ± 4 kg/m2 in the group of 30 LRYGB patients, and from 43.9 to 30.6 ± 4 kg/m2 in the open-surgery group (21 patients). The mean length of hospital stay and recovery times were shorter in the laparoscopy group (n = 30) (signifi cant differences: p < 0.025). Weight loss remained
5. The SF-36 (36-Item Short-Form Health Survey) is a self-administered questionnaire that serves to measure quality of life (QoL) as related to health. It contains 36 questions that evaluate 8 health domains: physical functioning, social functioning, bodily pain, general health perceptions, vitality, limitations as a consequence of mental or physical health, mental health, and perceived changes in general health.
relatively identical at one year in both groups. However, in the laparoscopy group, the conversion rate to an open procedure was high (23%, while it was 2.5% in the study by Nguyen et al.). The authors suggest that this rate could be appreciably reduced by stricter patient selection.
Another study, which was non-randomized, compared the open and laparoscopic approaches to gastric bypass in two groups of 80 patients matched by age, sex, pre-operative BMI and number of co-morbidities, with a 12-month follow-up. Even if the laparoscopic approach led to greater weight loss after the fi rst six months, with an EWL of 52.6% in the LRYGB group and 45% in the RYGB group (p < 0.05), this difference had decreased within one year and was no longer statistically signifi cant (69.2% vs 64.9% respectively) (Table 3). The differences in the major and minor complication rates are not signifi cant. The preliminary quality-of-life analysis based on an SF-36 survey indicates similar positive results close to healthy-population normative data: all the surgical subjects appreciated the benefi ts of improved quality of life after their operations, and there was no signifi cant difference between the two groups. The authors conclude that the two surgical approaches to gastric bypass seem equally effective in reducing weight and complications [Courcoulas et al., 2003a] (Table C-3, Appendix C).
Other more recent case series [Brody, 2004; Dresel et al., 2004; Stoopen-Margain et al., 2004] corroborate the effi cacy and benefi ts of LRYGB in terms of decreasing hospital stays and post-operative complications (Table 3 and Table C-2, Appendix C). Brody reports a decrease in the mean BMI from 51.5 to 32.0 kg/m2 in 195 patients after 18 months of follow-up. Stoopen-Margain et al., who analyzed the outcomes for 100 consecutive patients treated with LRYGB, reported an excess weight loss identical to that obtained with open surgery, 66 ± 5% at 18 months (n = 63) and 67 ± 8% at 24 months (n = 35), along with a signifi cant improvement in the patients’ health status
18
(diabetes and hypertension). They nevertheless emphasize that long-term follow-up remains necessary. Dresel et al., after having compared operating times, mean length of stay and complications in 60 obese patients (BMI < 50) and 60 super-obese patients (BMI > 50), observed no signifi cant differences. They conclude that the laparoscopic approach can safely be used with super-obese patients. They do not provide weight-loss outcomes, however.
As for hand-assisted RYGB procedures, the outcomes of the prospective, randomized trial conducted by Sundbom and Gustavsson [2004] are comparable to those documented in patients who underwent open RYGB: the mean BMI fell from 44 to 29 kg/m2 and from 45 to 30 kg/m2 respectively, with similar post-operative data. The authors conclude that there is no advantage to hand-assisted procedures. They confi rm the previous outcomes published in 2002 by DeMaria et al. [2002a].
4.2.1.3 BILIOPANCREATIC DIVERSION (BPD)
Even if biliopancreatic diversions with duodenal switch (BPD-DS) have been performed for more than 20 years, there is still only a limited number of studies providing detailed results. The CETS report [1998] had previously examined BPD-DS, which had been raising questions in Québec at that time. In selected patients who had undergone a BPD-DS, the EWL seemed to remain above 70% four or more years after surgery [Marceau et al., 1998; Rabkin, 1998], and even up to eight years in some cases [Hess and Hess, 1998] (Table 4 and Tables D-1 and F-1, Appendices D and F). Marceau et al. [1998] also compared BPD-DS with BPD-DG and attempted to minimize the differences in the follow-up periods by analyzing the subgroup of patients on whom they had operated during the year before the adoption of BPD-DS and those who had had this surgery during the year
after its adoption. The difference in the EWL is signifi cant, in favour of BPD-DS (70 ± 21% for a mean follow-up of 74 ± 4 months, vs 63 ± 21% for a follow-up of 85 ± 3 months) (Table F-1, Appendix F).
In a study published in 2000, Scopinaro et al. presented the outcomes of 23 years of experience with 2316 patients treated with BPD-DG (with different surgical variants). Data show a permanent decrease of about 75% in the initial excess weight of patients followed up for long periods (EWL = 75 ± 15% at 6 years in 1054 patients, and 76 ± 15% at 10 years in 381 patients) (Table 4), with an operative mortality rate of less than 0.5%. The late-complication rates remained less than 5% (Table D-1, Appendix D). The authors found improvements in the patients’ clinical and biological characteristics. They conclude that BPD-DG, when correctly performed by a surgeon with a sound knowledge of its mechanisms of action, is an effective and safe procedure.
More recently, Biron et al. [2004] presented the outcomes that they obtained through 20 years of performing BPDs (this retrospective study is not included in Table 4 because the outcomes are not expressed as EWL; details are presented in Table D-1 in Appendix D). From February 1984 to December 2002, 1271 consecutive patients underwent BPD, and 997 of them were enrolled in the study. After a follow-up of 7.9 ± 4.2 years, the BMI had dropped from 48.4 ± 9.4 to 31.3 ± 6.5 kg/m2, but in 36% of patients, the procedure was considered a failure: at the end of follow-up, 10% of patients had BMIs greater than or equal to 40 kg/m2, and the BMIs of the remaining 26% were between 35 and 40. The authors explain these results by the fact that the most obese patients lose relatively less weight and tend to regain lost weight more quickly.
19
TABLE 4
Outcomes of studies on biliopancreatic diversion*
AUTHORS AND YEAR
(STUDY DESIGN)
TYPE OF PROCEDURE
BMI(kg/m2)
EWL
Hess and Hess, 1998 (R)
BPD-DS (n = 440) 50 From 9 to 108 months: 80% of patients had a minimum EWL of 80%.
Marceau et al., 1998 (RNCC)
BPD-DG (n = 252)BPD-DS (n = 465)
46 ± 9 (n = 233)47 ± 9 (n = 457)
100 ± 20 months: 61 ± 22%
51 ± 25 months: 73 ± 21% Rabkin, 1998 (RNCC) BPD-DG (n = 32)
BPD-DS (n = 105)4549
24 months: 48 months (NS):69% 73% 78% 73%
Bajardi et al., 2000 (RNCC)
BPD-DS (n = 142) 50 (35–81) 2 years: 60%
Ren et al., 2000 (P)
LBPD-DS (n = 40) 60 (42–85) 6 months: 46 ± 2%9 months: 58 ± 3%
Scopinaro et al., 2000 (R)
BPD-DG (different variants) (n = 2316)
47 (29–87) 6 years: 75 ± 15% (n = 1054)10 years: 76 ± 15% (n = 381)
Kim et al.,2003 (RNCC) BPD-DS (n = 28)
LBPD-DS (n = 26)68.8 ± 10.166 ± 7.5
6 months (NS): 9 months (NS): 1 year (NS):44.3 ± 5.7% 48.7 ± 4.1% 56.8 ± 26.3% 56.9 ± 20.4% 68.1 ± 26.5% 76.7 ± 19.7%
Marinari et al.,2004 (R)
BPD-AHS (n = 858) 47 ± 7 2 years: 67 ± 18% (n = 800) 4 years: 67 ± 18% (n = 738) 6 years: 68 ± 18% (n = 659) 8 years: 69 ± 18% (n = 532) 10 years: 68 ± 18% (n = 334) 12 years: 66 ± 18% (n = 131) 14 years: 69 ± 15% (n = 60)
* See appendix for detailed data in Tables D-1, D-2, D-3 and F-1.
P: prospective non-comparative study; R: retrospective non-comparative study; RNCC: retrospective, non-controlled, non-randomized comparative study; n: number of patients; NS: non-signifi cant difference.
weight loss remained between 66% and 69% for 14 years on average, but the difference in the mean excess weight loss between the two groups was not signifi cant (BPD-AHS: 70.5 ± 23%; BPD-AHS-AHAL: 64.7 ± 17%). The re-operation rate was nevertheless higher (8.6%) in the BPD-AHS group (1.1% for the BPD-AHS-AHAL group; p < 0.001). The reduction of surgical revisions and improvement of quality of life (QoL score), and even resolution of certain co-morbidities documented among the patients in the BPD-AHS-AHAL group increased the success rate (sum of excellent, very good
For their part, Marinari et al. [2004] analyzed the results of 858 biliopancreatic diversions performed between June 1984 and December 1998 with creation of an ad hoc stomach (AHS) (the volume is adapted to patient characteristics) and a 50-cm common limb. Patients operated on between June 1984 and August 1992 received a 200-cm alimentary limb (BPD-AHS), while those operated on from September 1992 onwards received an ad hoc alimentary limb (AHAL) that varied in length depending on individual characteristics (BPD-AHS-AHAL). As shown in Table 4, excess
20
and good BAROS6 results) to 92% (83% for the BPD-AHS group). The complication and mortality rates with BPD did not seem higher than those recorded for the other procedures (Table D-1, Appendix D). It is nevertheless worth pointing out that these outcomes were obtained by teams who were highly experienced, if not specialized, in this type of procedure.
Studies on laparoscopic biliopancreatic diversion are rare (fi ve studies: Kim et al., 2003; Baltasar et al., 2002; Scopinaro et al., 2002; Paiva et al., 2001; Ren et al., 2000), their results are most often incomplete, their enrollments are limited and their follow-up is too short or unspecifi ed (see Table 4 and Tables D-2 and D-3 in Appendix D for outcomes not expressed as EWL). In a case series presenting more complete data on 40 BPD-DS patients at 9 months, the mean excess weight loss was 58%, and the complication rate was 17.5% [Ren et al., 2000]. Hospital stays ranged from four to eight days and were shorter than those following open surgery.
A single study retrospectively compared open biliopancreatic diversion (BPD-DS) and laparoscopic biliopancreatic diversion (LBPD-DS) in super-obese patients (BMI > 60 kg/m2). Outcomes at one year of follow-up tend to show that laparoscopy is more effi cacious and has a greater incidence of post-operative complications, but these differences are not signifi cant (Table D-3, Appendix D). While recognizing the limitations of their study design, the authors conclude that, even if LBPD is associated with a steep learning curve, this technique seems to be effective and safe for super-obese patients. Nevertheless, both approaches lead to appreciable mortality and
6. BAROS (Bariatric Analysis and Reporting Outcome System): This system is based on assigning scores to quality of life (QoL), EWL, complications and improvements in co-morbidities (the scoring key is presented in Appendix J).
morbidity rates, and further studies are needed to determine the best treatment for this particular population [Kim et al., 2003].
4.2.1.4 GASTRIC BANDING
Adjustable gastric banding (AGB) is a technique most often performed laparoscopically. In a recent prospective study, Steffen et al. [2003] published fi ve-year follow-up outcomes for 824 patients who had undergone laparoscopic Swedish adjustable gastric banding (SAGB). This study is characterized by a rather long follow-up period (fi ve years for 97% of patients), complete data on complications (time of onset and type) and a clear defi nition of the effi cacy outcome measure used (insuffi cient weight loss is defi ned as an EWL of less than 50% with no weight loss for three months or a 10% regain of lost weight, with the EWL being calculated from the Metropolitan Life Insurance tables [Appendix I]). Outcomes show that 82.9% of the patients lost more than 50% of their excess weight and that their quality of life (based on the BAROS score) varied according to the presence or absence of co-morbidities. At three years, the BAROS score was good, very good or excellent in 29 of the 40 patients who did not have any co-morbidities (72.5%) and in 157 of the 177 patients who did (88.7%). The authors report a mortality rate of 0.4% (three deaths) and a conversion rate of 5.2%. The long-term complication rate was 23.2% (191 patients), and 135 complications were related to the gastric band (Table E-1, Appendix E).
In conclusion, these study outcomes show that LAGB is effective in reducing excess weight, as long as fi ve years post-operatively (Table 5), with complication rates that the authors qualify as acceptable.
21
TABLE 5
Outcomes of studies on adjustable gastric banding (AGB)*
AUTHORS AND YEAR
(STUDY DESIGN)
TYPE OF PROCEDURE
BMI(kg/m2)
EWL
Belachew et al., 1998 (R)
LAGB (n = 550) 43 5 years: 50%
Dargent, 1999 (R)
LAGB (n = 500) 43 (36–60) 6 months: 45% (n = 443)1 year: 56% (n = 270)2 years: 65% (n = 96)3 years: 64% (n = 19)
Fielding et al., 1999 (R)
LAGB (n = 335) 46.7 (34–86) 12 months: 52% (n = 125)18 months: 62% (n = 58)
O’Brien et al., 1999 (P)
LAGB (Lap-Band)
(n = 277 out of 302 patients selected prospectively)
44.5 ± 6 1 year: 51.0 ± 17% (n = 120)2 years: 58.3 ± 20% (n = 43) 3 years: 61.6 ± 2% (n = 25)4 years: 68.2 ± 21% (n = 12)
Zimmermann et al.,1999 (R)
LAGB (n = 864)
LSAGB (n = 33)
42 (35–72) 6 months: 32% (n = 676)12 months: 40% (n = 233)18 months: 46.5% (n = 89)24 months: 39% (n = 47)
Hell et al., 2000 (PCC)
LSAGB (n = 30) 46.9 ± 7.8 Follow-up of 39.7 ± 7.6 months:0–24% (n = 1)25–49% (n = 13)50–74% (n = 15) 75–100% (n = 1)
Hell and Miller,2000 (PNCC)
LSAGB (n = 99) 46.9 ± 7.8 2 years: 59% (n = 97)5 years: 71% (n = 16)
DeMaria et al.,2001 (R)
LAGB (Lap-Band)(n = 37)
44.5 ± 4 12 months: 34.5 ± 20% (n = 28)24 months: 36 ± 23% (n = 24)36 months: 38 ± 27% (n = 15)48 months: 44% (n = 4)
Nehoda et al., 2001 (R)
LSAGB (n = 320) 44.29 12 months: 68%
Belachew et al., 2002 (R)
LAGB (n = 763) 42 (35–65) 6 months: 30%1 year: 40%2 years: 50%4 years: 50–60%
* See appendix for detailed data in Tables E-1, E-3, E-4, E-5, F-2 and F-3.
CR: comparative randomized study; P: prospective non-comparative study; PCC: prospective, controlled, non-randomized comparative study; PNCC: prospective, non-controlled, non-randomized comparative study; R: retrospective non-comparative study; RNCC: retrospective, non-controlled, non-randomized study; n: number of patients.
22
TABLE 5
Outcomes of studies on adjustable gastric banding (AGB) (cont’d)
AUTHORS AND YEAR
(STUDY DESIGN)
TYPE OF PROCEDURE
BMI(kg/m2)
EWL
Dixon and O’Brien, 2002 (R)
LSAGB (n = 50 patients with type 2 diabetes)
48.2 ± 8 1 year: 38 ± 14%
Doherty et al., 2002 (PNCC)
Group 1: Kusmak AGB(n = 40) (March 1992 to May 1995)
Group 2: Lap-BandLAGB (n = 17)AGB (n = 5)(1995 to January 7, 1997)
50
47
Group 1 Group 21 year 44% (n = 40) 27% (n = 19)2 years 47% (n = 40) 28% (n = 18)3 years 33% (n = 31) 25% (n = 18)4 years 40% (n = 24) 17% (n = 17)5 years 30% (n = 20) 21% (n = 15)6 years 32% (n = 18) 15% (n = 13)7 years 33% (n = 15)8 years 32% (n = 13)
Rubenstein, 2002 (R)
LAGB (n = 63) 48.8 ± 8 (36.8–67)
6 months: 27.2 ± 14.2% (n = 62)1 year: 38.3 ± 15.6% (n = 59)2 years: 46.6 ± 19.5% (n = 19)3 years: 53.6 ± 23.8% (n = 13)
Blanco-Engert et al.,2003 (CR)
LAGB (Lap-Band) (n = 30)LAGB (Heliogast) (n = 30)
43.4
41.2
3 months: 6 months: 12 months: (p < 0.0001) (p < 0.0001) (p < 0.0001)12.2 ± 1.3% 26.4 ± 2.88% 41.7 ± 2.71%
9.4 ± 1.39% 17.1 ± 1.65% 28.3 ± 2.40%
Biertho et al.,2003 (RNCC)
LAGB (n = 805) 42.2 ± 4.9(29–64)
6 months: 21.9%12 months: 33.3%18 months: 40.4% (97% of the patients)
Steffen et al., 2003 (P)
LSAGB (n = 824) 42.4 ± 1(31–69)
1 year: 29.5 ± 0.5% (n = 821)2 years: 41.1 ± 0.7% (n = 744)3 years : 48.7 ± 0.9% (n = 593)4 years: 54.5 ± 1.2% (n = 380)5 years: 57.1 ± 1.9% (n = 184)
Angrisani et al., 2004a (R)
LAGB (Lap-Band) (n = 381 out of 573) A: 30–39.9 (n = 166)
B: 40–49.9 (n = 302)C: 50–59.9 (n = 96)D: ≥ 60 (n = 9)
5 years:A: 54.6 ± 32.3% (n = 96)B: 54.1 ± 17.2% (n = 214)C: 51.6 ± 35% (n = 64)D: 59.1 ± 17.1% (n = 7)
Angrisani et al.,2004b (R)
LAGB (Lap-Band) (n = 210 patients with BMIs ≤ 35)
33.9 ± 1.1(25.1–35)
6 months: 28.1 ± 20.7% (n = 210/210)1 year: 52.5 ± 13.2% (n = 182/197)2 years: 61.3 ± 14.7% (n = 119/148)3 years: 64.7 ± 12.2% (n = 75/99) 4 years: 68.8 ± 15.3% (n = 49/73)5 years: 71.9 ± 10.7% (n = 21/29)
23
A multicentre retrospective study by Angrisani et al. [2004a] presented the outcomes obtained for 381 (66.5%) of the 573 subjects studied fi ve years after placement of a Lap-Band adjustable gastric band. Of these patients, 155 were lost to follow-up, 24 had to have the band removed because of complications, 8 had had another type of bariatric surgery, and 5 (0.9%) died from causes unrelated to the procedure. Detailed outcomes by patient group classifi ed by initial BMI (Table 5) show that the EWL exceeded 50% in all the groups. The mortality rate was signifi cantly higher in the group with BMIs from 50 to 59.9 kg/m2: 3.1% vs 0.87% for all the groups combined. Gastric pouch dilatation occurred in 4.1% of the patients, but in 7.2% of those with BMIs from 30 to 39.9 inclusively. Intragastric band migration (erosion) was diagnosed in 2.1% of cases (Table E-4, Appendix E).
Angrisani et al. [2004b] also specifi cally studied 210 patients with BMIs less than or equal to 35 kg/m2. For the study patients overall, the BMI dropped from 28.7 ± 3.8 kg/m2, with an EWL estimated to be 61.3 ± 14.7% at two years (for more than half of the surgical patients). These outcomes seemed stable at fi ve years (BMI = 28.2 ± 0.09 kg/m2 and EWL = 71.9 ± 10.7% in 21 of the 29 patients who were followed up) (Table 5). A single death was reported at 20 months; it was due to sepsis caused by perforation of the dilated gastric pouch (Table E-4, Appendix E).
The other LAGB studies involved patients with mean BMIs varying between 41 and 48.8 kg/m2, but their post-operative follow-up periods were very heterogeneous, from 12 to 96 months. Outcomes also varied widely in terms of clinical indicators such as mean operating time (35–193 minutes) and post-operative hospital stay (1–7 days). These variations can largely be explained by the surgical team’s learning curve: as patient numbers increase, operating times decrease [Rubenstein, 2002] and fewer complications arise (most often band-related: slippage, erosion, leakage, etc.). The effi cacy of LAGB expressed as post-operative EWL (actual
value or percent decrease) varied with the length of the follow-up period. At one year, the EWL was between 30% and 68%, and longer-term (fi ve-year) values were generally between 50% and 60% (Table 5 and Tables E-1, E-3, E-4 and E-5, Appendix E). Post-operative outcomes after LAGB procedures were similar to those obtained after open procedures, and they had a positive impact on patients’ quality of life [Hell et al., 2000].
Two randomized comparative studies of different methods of implanting adjustable gastric bands (conventional or retrogastric vs esophagogastric) arrive at different conclusions [Weiss et al., 2002; Weiner et al., 2001]. Even if the outcomes are similar from the standpoint of effi cacy (weight loss), the practitioners have diverging opinions on esophagogastric placement (Tables 5 and E-2, Appendix E). According to Weiner et al. [2001], this method is easier to perform and safer, and the complication rate is lower than with retrogastric placement (signifi cant difference). The study by Weiss et al. [2002], for its part, reports higher complication and re-operation rates with esophagogastric placement. The limited number of subjects (101 for Weiner and 54 for Weiss) and the relatively short follow-up (18 and 24 months) do not permit defi nitive conclusions.
In a randomized study, Blanco-Engert et al. [2003] compared the outcomes achieved after having implanted two types of adjustable bands (Lap-Band® and Heliogast®)7 in two groups of 30 patients (Table 5). At 12 months, they found that the Heliogast patients had much lower EWLs (28.3 ± 2.40 vs 41.7 ± 2.71%; p < 0.0001) and a higher number of complications than the Lap-Band patients (signifi cant difference). According to the authors, the Heliogast should no longer be used or should at least be reserved to specifi c patients such as those over the age of 60 (Table E-3, Appendix E).
The randomized comparative study by de Wit et al. [1999] involved 50 patients divided
7. The Lap-Band is manufactured by the U.S. fi rm Inamed, and the Heliogast, by the French fi rm Hélioscopie.
24
equally into two groups (open AGB vs LAGB) with no signifi cant differences as to their characteristics (male-female ratio, weight, BMI and co-morbidities). (This study does not appear in Table 5 because the outcomes are not expressed as EWL: see Table E-5 in Appendix E.) At one year, there was no signifi cant difference in the number of post-operative or early complications, while weight loss was similar in the two groups. However, laparoscopy was associated with a shorter hospital stay and a lower number of re-admissions than with the open procedure (signifi cant differences).
The outcomes of a retrospective, comparative study by Fried [2000] (presented only in Table E-5 in Appendix E) show a signifi cant difference in post-operative length of stay in favour of patients who underwent laparoscopic procedures (2.8 days vs 10.5 days for the open procedure) and a substantial decrease in the number of early complications related to open surgery (1%), with appreciably equivalent effi cacy (mean weight loss of 37.6 kg in the laparoscopic group vs 38.4 kg in the open-surgery group).
The non-randomized comparative study by Doherty et al. [2002] was carried out with two groups of patients who underwent adjustable gastric banding (the Kusmak type of SAGB for the open procedure, and the Lap-Band for laparoscopic placement [17 cases] and for open placement [5 cases]). The outcomes indicate a large number of re-operations and band removals caused by the onset of complications or band intolerance. These outcomes seem exceptional because, according to a review article, most of the recent studies report conversion rates of less than 5% and a much smaller number of complications [Fried et al., 2002] (Table E-5, Appendix E). With respect to long-term effi cacy, in the study by Doherty et al. [2002], excess weight loss rose to 47% at two years but gradually dropped to 32% at six years of follow-up. Excess weight loss was even lower in the Lap-Band group (28% at two years and 15% at six years) (Table 5). These outcomes do not agree with those of the other studies under
review, and their validity is weakened by the small sample size.
4.2.2 Comparison of the techniques
4.2.2.1 BILIOPANCREATIC DIVERSION VS OTHER TYPES OF BARIATRIC-SURGERY PROCEDURES
Only two retrospective studies have compared biliopancreatic diversion with other surgical techniques [Bajardi et al., 2000; Rabkin, 1998] (Table 6).
4.2.2.1.1 Biliopancreatic diversion (BPD ) and biliopancreatic diversion with duodenal switch (BPD-DS ) vs Roux-en-Y gastric bypass (RYGB)
When Rabkin [1998] retrospectively compared three bariatric-surgery techniques (RYGB, Scopinaro’s BPD and BPD-DS), he noted that the medium-term effi cacy of BPD-DS (at two and four years of follow-up) was similar to that of the other two techniques. However, in the fi rst 37 patients in the group of 105 subjects who underwent BPD-DS, four cases of severe operative complications occurred (two cases of peritonitis, one pancreatitis, and one thrombophlebitis associated with pulmonary embolism). Given that the results on the complications and the mortality rates are neither complete nor available for the two other surgical techniques, it is diffi cult to pass judgment on the overall superiority of any one of them (Table F-1, Appendix F).
4.2.2.1.2 Biliopancreatic diversion (BPD) vs vertical banded gastroplasty (VBG)
A comparative study of two groups of obese adult patients treated with VBG (n = 93) and with BPD (n = 142) indicates that, after two years of post-operative follow-up, the mean percentage of EWL was higher in the BPD group: 60% vs 48% in the VBG group. It is not indicated, however, if this difference is statistically signifi cant. Moreover, the operating time, the mean hospital stay and the mortality rate were higher in the BPD group. Virtually
25
all the early complications occurred among the BPD patients (7.3%), which could explain the longer operating time for this procedure. The incidence of late complications was the same in both groups, but there was a greater prevalence of malnutrition syndromes in the BPD group and of gastric stenoses in the VBG group. In addition to its weight-reduction effi cacy, the authors emphasize that it reduced and even resolved some of the co-morbidities present before the procedure. All the patients reported improvements in their physical activity and the resolution of sleep apnea [Bajardi et al., 2000].
The authors conclude that the choice of either technique depends on the patient’s clinical and psychological characteristics. They are of the opinion that BPD should be reserved to super-obese patients (BMI > 50 kg/m2) who present with severe dyslipidemia and agree to long-term follow-up. However, the weakness of their study design (retrospective) limits the validity and scope of their study outcomes (Table F-1 Appendix F).
4.2.2.2 VERTICAL BANDED GASTROPLASTY (VBG) VS VERTICAL BANDED GASTROPLASTY COMBINED WITH ROUX-EN-Y GASTRIC BYPASS (VBG-RYGB)
In their retrospective study, Capella and Capella [1996] compared vertical banded gastroplasty combined with Roux-en-Y gastric bypass (VBG-RYGB) with VBG alone. The VBG-
RYGB procedure is analogous to the basic gastric-bypass procedure performed at the Royal Victoria Hospital, McGill University Health Centre (MUHC). This retrospective study analyzed 888 surgically treated obese subjects (328 with VBG and 560 with VBG-RYGB, all by the same surgeon) followed up from 30 to 66 months. There was no signifi cant difference between the two groups in terms of age, male-female ratio, initial BMI and relative number of super-obese patients.
The outcomes show a signifi cantly higher EWL rate in those who underwent VBG-RYGB (62 ± 17% at fi ve years’ mean follow-up) compared with those in the VBG group (47 ± 23%) (Table 7). While the early complication rates were 0.3% and 1% in the VBG and VBG-RYGB groups respectively, late complications rose to 9% and 12% respectively. The authors point out that, among the fi rst 272 people who underwent VBG-RYGB, the late-complication rate was 22% but dropped to 1% in the following 351 patients. In most cases, these complications were due to staple-line disruption (Table F-2, Appendix F). In a more recent study of a series of 652 consecutive patients, of whom 72 were followed up for fi ve years, the same authors confi rm the effi cacy of VBG-RYGB for super-obese patients (mean EWL of 77% at fi ve years) and recommend it without reservations [Capella and Capella, 2002] (Table C-1, Appendix C).
TABLE 6
Studies comparing biliopancreatic diversion with other open-surgery techniques*
AUTHORS AND YEAR
(STUDY DESIGN)
TYPE OF PROCEDURE
BMI(kg/m2)
EWL
Rabkin, 1998 (RNCC) BPD (n = 32)
RYGB (n = 138)BPD-DS (n = 105)
454949
24 months 48 months 69% 73% 74% 63% 78% 73%
Bajardi et al., 2000 (RNCC)
VBG (n = 93) (1990–1995)BPD (n = 142) (1993–1998)
48.7 (37–65.6)50 (35–81)
48% after 2 years60% after 2 years
* See detailed data in Table F-1, Appendix F.
RNCC: retrospective, non-controlled, non-randomized comparative study; n: number of patients.
26
4.2.2.3 VERTICAL BANDED GASTROPLASTY (VBG) VS ROUX-EN-Y GASTRIC BYPASS (RYGB) VS LAPAROSCOPIC SWEDISH ADJUSTABLE GASTRIC BANDING (LSAGB)
In a prospective study, Hell et al. [2000] compared VBG, RYGB and LSAGB in terms of effi cacy and quality of life. Each of the three groups included 30 patients matched by sex, BMI and age. While excess weight loss (EWL) was similar in the VBG and LSAGB groups (17 and 16 patients respectively had lost more than 50% of their initial excess weight), this fi gure rose to 28 patients in the RYGB group, in which 22 patients had lost excess weight greater than or equal to 75% (compared with two patients in the VBG group and one in the LSAGB group) (Table 8). However, unlike the weight-loss outcomes, the BAROS scores did not indicate such great superiority in terms of improved quality of life (QoL) for the patients in the RYGB group (BAROS score: 7.15 for RYGB, 6.13 for VBG and 5.99 for LSAGB). The outcomes obtained by Hell and Miller [2000] with a larger group of patients (101 for VBG and 99 for LSAGB) showed similar performances in terms of EWL, both at two years and at fi ve years. This study had also stratifi ed patients by age, sex and initial BMI (Table 8).
4.2.2.4 LAPAROSCOPIC ROUX-EN-Y GASTRIC BYPASS (LRYGB) VS LAPAROSCOPIC ADJUSTABLE GASTRIC BANDING (LAGB)
The most recent comparative study presents the outcomes obtained for more than 1200 patients treated with LRYGB at the Mount Sinai Medical Center in New York (n = 406) or with LAGB at the Obex Institute in Switzerland (n = 805). For patients with BMIs > 50 kg/m2, the length of the Roux-en-Y limb was 150 cm [Biertho et al., 2003]. Data were collected retrospectively for the LRYGB group and prospectively for the
LAGB group. At the time of inception, the two comparison groups had signifi cantly different weights and BMIs (p = 0.0001), which were greater in the patients treated with LRYGB in New York, as shown in Table 9.
Eighteen months after the procedure, the LRYGB patients had lost a signifi cantly higher percentage of excess weight than the LAGB patients (74.6% vs 40.4%; p < 0.0001). However, LRYGB led to more complications than LAGB, especially major early complications (4.2% vs 1.7%; p = 0.02), intra-operative complications (2% vs 1.3%) and deaths (0.44% vs 0%): in the last two cases, the difference was not signifi cant (Table F-3, Appendix F.) According to the authors, it is not yet possible to pass judgment on the best indication for each technique, but it probably depends on a certain number of factors such as initial BMI, dietary patterns and co-morbidities. Nevertheless, LAGB could be indicated for patients with BMIs between 30 and 40 kg/m2, while LRYGB would be better suited to more obese patients (40–50 kg/m2). The authors add that, for subjects with BMIs greater than 50 kg/m2, another type of procedure—such as biliopancreatic diversion—would be preferable.
Despite its large size, this comparative study suffers from some major methodological biases. First of all, the groups were heterogeneous, given that the weight of the patients in the LRYGB group was greater than that of the patients in the LAGB group, and the percentages of patients followed up at 18 months differed (37% vs 97%). In addition, the study compared surgeons practising in very different settings (Switzerland and the United States), even though several authors have emphasized the relationship between bariatric-surgery outcomes and the surgical learning curve [DeMaria, 2003; Gagner and Rogula, 2003].
27
TABLE 7
Comparative study of open vertical banded gastroplasty*
AUTHORS AND YEAR
(STUDY DESIGN)
TYPE OF PROCEDURE
BMI(kg/m2)
OUTCOMES
Capella and Capella,1996(RNCC)
VBG (329 operations on 328 patients)
VBG-RYGB (623 operations on 560 patients: in 351 cases, gastric segments were completely separated)
52 ± 9
52 ± 9
30 to 42 months 54 to 66 monthsEWL (%) 48 ± 23 47 ± 23BMI (kg/m2) 39 ± 9 40 ± 9
EWL (%) 70 ± 19 62 ± 17BMI (kg/m2) 32 ± 6 34 ± 6
* See detailed data in Table F-2, Appendix F.
RNCC: retrospective, non-controlled, non-randomized comparative study.
TABLE 8
Outcomes of studies comparing VBG with other bariatric-surgery techniques*
AUTHORSAND YEAR
(STUDY DESIGN)
TYPE OF PROCEDURE
BMI OR INITIAL
WEIGHTEWL
Hell et al., 2000 (PCC) VBG (n = 30)
LSAGB (n = 30)RYGB (n = 30)
46.9 ± 9.9 kg/m2
46.9 ± 7.8 kg/m2
45.2 ± 8.2 kg/m2
0–24% 25–49% 50–74% 75–100% Follow-up (mos)n = 1 12 15 2 40.1 ± 8.3 n = 1 13 15 1 39.7 ± 7.6 n = 0 2 6 22 60.0 ± 8.2
Hell and Miller, 2000 (PNCC) VBG (n = 101)
46.9 ± 9.0 kg/m2
133.7 ± 33.3 kg2 years: 61% (40 kg) (n = 98)5 years: 69% (48 kg) (n = 15)
LSAGB (n = 99)46.9 ± 7.8 kg/m2
133 ± 22.7 kg2 years: 59% (46 kg) (n = 97)5 years: 71% (56 kg) (n = 16)
* See detailed data in Table F-2, Appendix F.
PCC: prospective, controlled, non-randomized comparative study; PNCC: prospective, non-controlled, non-randomized comparative study; n: number of patients.
28
TABLE 9
Outcomes of the study (RNCC) by Biertho et al. [2003]*
NUMBER OF
PATIENTS
INITIAL WEIGHT AND BMI
HOSPITAL STAY (MEDIAN
IN DAYS)
FOLLOW-UP† EWL (%) IN RELATION TO PRE-OPERATIVE BMI
Patients (%) Length 30–40 40–50 50–60 All‡
456
(LRYGB)
Weight (kg): 135.4 ± 26.3 (76–221)
BMI (kg/m2): 49.4 ± 8.3 (27–77)
3 ± 0.3(2–94)
88 6 months57 12 months37† 18 months
55 56 47 51.6 75 72 57 67.0 – 81 69 74.6
805
(LAGB)
Weight (kg):117 ± 17.1(75–224)
BMI (kg/m2):42.2 ± 4.9(29–64)
5 ± 2.4(2–22)
97 6 months97 12 months97† 18 months
24 21 18 21.9 37 32 26 33.3 41 40 33 40.4
* Detailed outcomes for this study can be found in Table F-3, Appendix F.† The authors attribute the difference in survival rates at 18 months (37% and 97%) to the gastric-bypass procedure itself (which requires more assessments, whereas the patients come from diverse regions that are sometimes quite far), making long-term follow-up diffi cult.‡ All the differences between the two groups, regardless of the length of follow-up, are signifi cant: p < 0.0001.
RNCC: retrospective, non-controlled, non-randomized comparative study.
4.2.2.5 ADJUSTABLE GASTRIC BANDING (AGB) VS NON-ADJUSTABLE GASTRIC BANDING (NAGB)
After reviewing studies published between September 2000 and September 2001, Fried et al. [2002] compared the effi cacy of adjustable gastric banding (AGB) with that of non-adjustable gastric banding (NAGB). The effi cacy outcome measures selected for this comparison were early and late complications, number of re-operations and weight loss after a minimum post-operative follow-up of three years.
Four centres performing NAGB procedures were chosen (Spain, Israel, Czech Republic and United States). The technique and materials used during the four study years were considered to be similar. In total, 1812 patients treated with NAGB were compared with 1968 patients treated with AGB (Table 10). The outcomes of this study show that at 48 months, the percentages of EWL and the early complication rates were virtually the same in both groups. However, the authors found signifi cant
differences in the rates of late complications (6.7% vs 1.9%) and re-operations (7.2% vs 3.4%) in favour of NAGB. They add that the materials used in NAGB offer greater fl exibility and are better adapted to gastric peristalsis, which could lessen irritation and give the band a greater physiological effect.
4.2.2.6 COMPARISON OF THE FOUR MAIN TYPES OF BARIATRIC SURGERY: TWO META-ANALYSES
In a meta-analysis of 136 studies, Buchwald et al. [2004] evaluated the effi cacy of different types of bariatric surgery (gastroplasty, gastric bypass, adjustable and non-adjustable gastric banding, and biliopancreatic diversion) in relation to weight outcome measures (absolute weight loss in kilograms, decrease in BMI, percentage of initial weight loss or percentage of excess weight loss). They also examined the impact of these procedures on the progression of obesity co-morbidities (mainly type 2 diabetes, hypertension, hyperlipidemia and sleep apnea).
29
TABLE 10
Outcomes of the comparative review of AGB and NAGB*
AUTHORS OF PRIMARY
STUDIES
AGB1968 PATIENTS
MEAN BMI: 44.0 kg/m2
NAGB1812 PATIENTS
MEAN BMI: 42.4 kg/m2
Forsell(2000)
Hell and Miller, (2000)
Belva et al.
(2001)
Favretti et al.
(2001)
Mean values†
Ballesta et al.
(1998)
Dudai(1999)
Friedet al.(1999 and
2000)
Molina(2000 and
2001)
Mean values†
Number of patients 376 99 + 69 763 830 306 (336)
1919(931)487
(6906)512
EWL (%)at 48 months 68 65 – 55 53 68 70 54 – 54.2
Early complications (%)
2.1 2.0 0.8 0.2 1.6 2.5 1.5 1.1 0.6 1.4
Late complications (%)
4.0 2.2 9.0 3.3 6.7 1.8 1.8 2.2 1.7 1.9
Re-operations (%) 7.0 9.0 11.1 3.9 7.2 1.95 1.78 6.3 3.1 3.4
Source: Fried et al., 2002.
* The results are presented here as they appear in the article. For further details on the case series studied, please consult the references cited by Fried et al. † The values in bold are the variables in which a signifi cant difference is observed between the two groups.
This meta-analysis involved a total of 22,094 patients identifi ed in studies published until July 2003: 5 controlled randomized trials, 28 non-randomized controled studies, 101 case series and 2 health-care economic studies. Of these studies, 58 were conducted in Europe, 56 in North America and the remainder in various countries around the world (Australia, New Zealand, South America, etc.). Some of the studies included in this meta-analysis will be evaluated separately in this report.
The outcomes confi rm the general effi cacy of bariatric surgery, with EWL rates on the order of 61.23% (95% CI: 64.40–58.06) for all the study patients combined (10,172), although this rate varies by type of procedure: 47.45% EWL for gastric banding, 61.56% for gastric bypass, 68.17% for gastroplasty and 70.12%
for biliopancreatic diversion (Table 11). Note, however, that each treatment group included several variants and that the number of treated patients differed (e.g., 4204 patients underwent gastric bypass and 506, gastroplasty). In addition, the outcomes are not presented according to the number of months or years of follow-up, except it is mentioned that weight-loss outcomes did not differ signifi cantly at assessments at two years or less compared with those at more than two years. The post-operative mortality rate (within 30 days of the procedure) was 0.1% for restrictive procedures (gastroplasty and gastric banding), 0.5% for gastric bypass, and 1.1% for biliopancreatic diversion. Outcomes related to a reduction in the signs and symptoms of the main co-morbidities will be dealt with in section 4.4.
30
TABLE 11
Outcomes of the meta-analysis by Buchwald et al. [2004] comparing the main types of bariatric surgery*
GASTROPLASTY(MAINLY VBG)
RYGB(INCLUDING VARIANTS)
GASTRIC BANDING (ADJUSTABLE AND NON-ADJUSTABLE)
BPD(INCLUDING VARIANTS)
ALL PROCEDURES COMBINED†
Weight loss† (kg)
936 patients(28 groups)
–39.82(–44.74 to –34.90)‡
2742 patients(20 groups)
–43.48(–48.14 to
–38.82)
482 patients(13 groups)
–28.64(–32.77 to
–4.51)
1282 patients(10 groups)
–46.39(–51.58 to
–41.20)
7588 patients(83 groups)
- 39.71(- 42.23 to
- 37.19)
Decrease in BMI†
(kg/m2)
942 patients(27 groups)
–14.20(–16.14 to
–12.27)
2705 patients(22 groups)
–16.70(–18.43 to
–14.98)
1959 patients(25 groups)
–10.43(–11.52 to
–9.33)
984 patients(12 groups)
–17.99(–19.40 to
–16.59)
8232 patients(96 groups)
–14.20(–15.13 to
–13.27)
EWL(%)
506 patients(15 groups)
–68.17(–74.81 to
–61.53)
4204 patients(22 groups)
–61.56(–66.45 to
–56.68)
1848 patients(12 groups)
–47.45(–54.23 to
–40.68)
2480 patients(7 groups)
–70.12(–73.91 to
–66.34)
10,172 patients(67 groups)
–61.23(–64.40 to
–58.06)
NB.: Unlike in the other tables, negative results are included to respect the author’s presentation, especially for the confi dence intervals.
* Includes Roux-en-Y gastric bypass, gastric banding, gastroplasty, biliopancreatic diversion and mixed techniques, along with other less common procedures (biliary-intestinal bypass, ileogastrostomy, jejuno-ileal bypass, and unspecifi ed bariatric surgery.† Whenever possible, outcome time points representing at least 50% of the patient population undergoing surgery were used.‡ 95% confi dence interval. The p-value is signifi cant for heterogeneity (cross-study variation), except for the initial weight loss induced by gastric bypass or by biliopancreatic diversion with or without duodenal switch.
Another very recent meta-analysis (April 2005) also compares the same four techniques in terms of weight loss at 12 months and at 36 months or more, mortality and complications [Maggard et al., 2005]. Of the 167 studies identifi ed, 89 were eligible to be selected for weight-loss analysis. The outcomes (Table 12) generally tally with those of the meta-analysis by Buchwald et al., except for VBG, which achieved lower outcomes. Overall, the outcomes at 36 months differ little from those at 12 months, despite slight downward trends for RYGB and upward trends for AGB and BPD. It can also be seen that the surgical approach (laparoscopy or open surgery) has little impact on the outcomes achieved with RYGB, although a single study presents three-year outcomes for the laparoscopic procedure. Finally, overall mortality remains less than 1%, whether the deaths occur 30 or fewer days post-operatively.
4.3 COMPLICATIONS
Examining surgical complications is a diffi cult task. In the selected studies, the incidence and nature of the complications vary by type of procedure and surgical approach; early and late complications are not systematically differentiated; and the studies generally use different analytical frameworks.
There are different types of bariatric-surgery complications: they are mainly infectious, lesional (wound dehiscence, hernias, etc.) and metabolic (anemia, hypocalcemia, etc.). Other complications are linked directly to the gastric band. Gallstone formation, not a rare event, is attributable to excessively radical weight loss, and may occur with non-surgical approaches. The pathogenesis has not yet been clarifi ed [Stocker, 2003].
31
TABLE 12
Outcomes of the meta-analysis by Maggard et al. [2005] comparing the main types of bariatric surgery*
PROCEDURE
FOLLOW-UP AT 12 MONTHS FOLLOW-UP AT 36 MONTHS AND MORE
WEIGHT LOSSIN KG
(95% CI)
STUDIES (PATIENTS)
WEIGHT LOSSIN KG
(95% CI)
STUDIES (PATIENTS)
RYGB (all) 43.46 (41.24–43.46) 32 (n = 2937) 41.46 (37.36–45.56) 21 (n = 1281)
RYGB (open) 43.89 (41.09–46.69) 25 (n = 2074) 41.58 (37.38–45.78) 20 (n = 1266)
LRYGB 42.17 (38.95–45.38) 10 (n = 863) 38.32 (28.04–48.60) 1 (n = 15)
VBG (all) 32.16 (29.92–34.41) 21 (n = 2080) 32.03 (27.67–36.38) 18 (n = 1877)
AGB (all) 30.19 (27.95–32.42) 27 (n = 5562) 34.77 (29.47–40.07) 17 (n = 3076)
BPD (all) 51.93 (45.10–58.75) 3 (n = 735) 53.10 (47.36–58.84) 1 (n = 50)
CI: confi dence interval; n: number of patients.
(Tables B-1, B-2, F-1 and F-2, Appendices B and F).
4.3.3 Complications from biliopancreatic diversion
According to the limited data available, complication rates from biliopancreatic diversion vary and may reach 24% for incisional hernias; the mortality rate is between 1% and 4%. It should be mentioned that these outcomes were obtained in patients with particular characteristics (high BMIs or co-morbidities). In addition to common bariatric-surgery complications (hernias, infections, pulmonary embolisms), studies indicate a high percentage of malnutrition syndromes for specifi c nutrients (vitamins, iron, and hyperproteinemia) (19% in the study by Bajardi et al. [2000]) (Tables D-1, D-2, D-3 and F-1, Appendices D and F).
4.3.4 Complications from laparoscopic procedures
With respect to complications from laparoscopic procedures, adjustable gastric banding achieves the best results. Complication rates vary by time of onset (early or late), type (hernias, erosion, infections, pouch dilatation) and degree of severity (from simple gastro-esophageal refl ux
4.3.1 Complications from gastric bypass
The rates of early and late gastric-bypass complications, especially from Roux-en-Y, vary between 0.3% and 2.7% and between 2.2% and 13% respectively. These relatively low values refl ect improvements in the performance of this procedure, given that a study published in 1995 reported early complication rates that were 10 times higher (25.5%) [Pories et al., 1995]. Late complications are most often due to staple-line disruption or hernia development. The risk of pulmonary embolism is nevertheless lower with gastric bypass. The systematic administration of nutritional supplements (iron, vitamin B12, folic acid and calcium) after these operations contributes to the absence of metabolic complications. The mortality rate is less than or equal to 1% (Tables C-1, C-2, C-3, F-1 and F-3, Appendices C et F).
4.3.2 Complications from gastroplasty
In their comparative study, Capella and Capella [1996] documented a 9% complication rate in 328 patients treated with VBG. Complications were most often related to leakage caused by band migration or staple-line disruption. In the other studies, these rates varied between 3% and 21%. The mortality rate remained less than 0.5%
32
to complications that are severe, and even fatal, or that require re-operation). In most cases, complications are minor and re-operations are required to remove, replace or adjust the gastric band. The decline in complication rates is mainly due to the surgical learning curve and to enhanced gastric bands [Steffen et al., 2003; Doherty et al., 2002; Fried et al., 2002] (Tables E-1 to E-5, Appendix E).
In a review of 3464 cases identifi ed in 10 different studies (published between January 1, 1994, and December 31, 2002), Podnos et al. [2003] evaluated the number and type of complications resulting from laparoscopic RYGB. These data were compared with 2771 cases of conventional RYGB (open procedure) from 8 studies published over the same period. According to 9 of the 10 studies, the conversion rate from laparoscopic to open surgery is estimated to be 2.2%, the main reason being hepatomegaly (48.7% of cases). The laparoscopic approach leads to decreased wound-related complications (infections and hernias), iatrogenic splenectomies and mortality. In return, it is associated with an increase in complications that rarely occur with open surgery, including early or late intestinal obstruction, gastro-intestinal tract bleeding, and stomal stenosis. The authors attribute part of this increase to the learning curve for this new technique. As a result, they recommend that professional associations provide better training for surgeons interested in performing this type of procedure. These conclusions confi rm the outcomes previously reported by Luján et al. [2004] and those by Westling et al. [2002] (Table C-3, Appendix C), who also support stricter patient selection as a means of reducing the number of conversions.
4.4 IMPACT OF BARIATRIC SURGERY ON OBESITY CO-MORBIDITIES
Besides the weight-reduction effi cacy of bariatric surgery, several authors affi rm that it
reduces and even resolves some pre-operative co-morbidities. Essentially on the strength of the SOS study, the Blue Cross and Blue Shield Association published a report on the impact that post-operative weight loss could have on the progression of diseases associated with morbid obesity [BCBS, 2003a]. According to the outcomes achieved by Sjöström et al. [1999], the incidence of diabetes, two years after surgery, was 30 times lower among the 767 surgical patients than it was among the 712 non-surgical patients (two-year incidence: 0.2% vs 6.3%; p < 0.001). This reduction was on the order of 10 for hypertriglyceridemia (0.8 vs 7.7%; p < 0.001) and 2.5 for hypertension (5.4 vs 13.6%; p < 0.001). In the fi rst two groups of patients (517 surgical and 539 non-surgical patients) followed up for 10 years, the difference was maintained for diabetes, the incidence of which was four times lower in the surgically treated group (odds ratio [OR] = 0.25; 95% confi dence interval [CI]: 0.17–0.38; p < 0.001). However, if only the fi rst groups of patients are analyzed, the difference decreases for hypertriglyceridemia (OR = 0.61; CI: 0.39–0.95; p < 0.03) and is no longer signifi cant for hypertension (OR = 0.75; CI: 0.52–1.08; p < 0.06) [Sjöström et al., 2004]. All these outcomes were associated with decreased body mass in the obese patients.
The Swedish study also shows reductions of 56% and 48% in the number of patients meeting the diagnostic criteria for depression or anxiety after the surgical procedure [Karlsson et al., 1998]. The authors of the Blue Cross and Blue Shield Association report also analyzed the outcomes of 11 additional studies, including one randomized trial. Their data confi rm the results of the Swedish study [BCBS, 2003a].
Although some authors have tried to establish the precise relationship between the amount of weight loss and the improvement in some outcome measures (diabetes, arthritis, glycemia, glycosylated hemoglobin, SF-36, etc.) and to derive clinical signifi cance from it, the Blue Cross and Blue Shield Association [2003a] concludes that it is not possible to generalize
33
from the conclusions they reached. According to them, neither is it possible to set an objective weight-loss cut-off point to determine the success of a surgical procedure.
According to the results of other studies and depending on the surgical technique (VBG, RYGB, GB, LAGB), the quality of life enjoyed by 90% to 98% of the patients improved, and co-morbidity symptoms were reduced in 50% to 80% of cases [Schauer et al., 2003a; Stocker, 2003; DeMaria et al., 2002b; Frezza et al., 2002]. However, most of the studies do not examine the progression of health problems associated with obesity. A few rare studies, such as those by Bajardi et al. [2000] and by Kim et al. [2003], indicate the effects of bariatric surgery on co-morbidities. After having studied its impact on 142 BPD-DS patients and 93 VBG patients, Bajardi et al. recorded a reduction in the clinical symptoms associated with lower-limb chronic venous insuffi ciency in 95% of the cases, discontinuation of anti-hypertensive treatment in 90% of the cases, and a return to normal blood lipid levels (cholesterol and triglycerides) in 100% of the cases. All the diabetic patients required to take insulin stopped all injections after one year, and only 21% of diabetics continued taking oral hypoglycemic medications. Finally, most of the patients reported increased physical activity because weight loss reduced their osteoarthropathy or improved their cardiopulmonary performance [Bajardi et al., 2000].
The outcomes of the study by Kim et al. [2003], who compared open BPD-DS (28 patients) and laparoscopic BPD-DS (26 patients), show improvement in some of the patients’ clinical conditions with respect to co-morbidities, mainly diabetes, sleep apnea, hypertension, asthma, and arthritis. These improvements translated into lower doses of medication and were signifi cantly greater in the group that had undergone laparoscopic BPD-DS (eight patients in the laparoscopy group and two in the open-surgery group). Hypertension and asthma were reduced by 20% in the laparoscopy group, while in the open-surgery group, hypertension was
reduced by 8.3% and asthma by 17%. As for the other co-morbidities, only the data from the LBPD group were published. Similar outcomes were achieved in 51 obese patients with type 2 diabetes who had undergone LAGB: remission of diabetes occurred in 32 of these patients [Dixon and O’Brien, 2002].
These outcomes are confi rmed by four studies published in 2004 (three in the United State and one in Québec), which analyzed the effects of bariatric surgery on the progression of co-morbidities. In the U.S. studies, the assessments included estimated cost savings derived from bariatric surgery as a result of a reduction in the signs and symptoms, or resolution, of co-morbidities associated with clinical obesity.
Potteiger et al. [2004] retrospectively reviewed the use of hypertension and diabetes medications pre-operatively and at 9 months post-operatively in 51 consecutive patients with morbid obesity (BMIs greater than 40 kg/m2 with an associated co-morbidity, or BMIs greater than 45 alone) who underwent RYGB (30 with the open approach and 21 laparoscopically). Total control or a favourable progression of the disease was achieved by 92% (n = 47) of the patients with diabetes, and by 78% (n = 40) of the patients with hypertension. This progression was associated with a signifi cant decrease in medication usage. It fell from 1.12 ± 1.15 to 0.12 ± 0.48 (p < 0.001) for diabetes medications, and from 1.32 ± 1.25 to 0.44 ± 0.64 for anti-hypertensive agents (p < 0.001) (Appendix H).
In their retrospective study, Monk et al. [2004] focused on the progression of the fi ve major obesity co-morbidities (type 2 diabetes, hypertension, sleep apnea, gastro-esophageal refl ux disease, and asthma). Of the 100 patients selected, 87 were taking medications to treat these problems before surgery (RYGB); this number was reduced to 64 patients, owing to relocation or to incomplete pharmacology records; fi nally, follow-up varied from 6 to 60 months (mean of 16 months). The authors acknowledge that the number of patients lost to follow-up and the inadequate follow-up
34
period weaken the validity of their study. The mean BMI of the 64 patients was 57 kg/m2 (range, 36.6–85.4). The outcomes show that 21 of the 23 diabetic patients had discontinued all their medications and that the remaining 2 had reduced their therapeutic doses; the same was true for the hypertensive patients (11 discontinued treatment and the remaining 15 reduced their doses). Patients suffering from sleep apnea (treated with continuous positive-airway pressure), gastro-esophageal refl ux disease and asthma had similar outcomes (Appendix H).
Snow et al. [2004] reported similar results for 78 patients with morbid obesity (BMIs from 36 to 70 kg/m2) who underwent LRYGB between March 2001 and March 2003 (78 patients were followed for one year, and 26 for two years) (Appendix H). In this retrospective study, the authors selected from a database of 1060 patients all those over the age of 54 who had been followed for a minimum of six months: 82 cases were extracted, but 4 were excluded (three deaths before 180 days and one lost to follow-up). Pre-operatively, 324 prescriptions were recorded for 70 patients, and one year later, the number of prescriptions had dropped to 112 for 53 patients. All co-morbidities decreased: hypertension or hypertension with cardiovascular disease, type 2 diabetes, pulmonary insuffi ciency, osteoarthritis, anxiety or depression, hyperlipidemia, gastro-esophageal refl ux disease, and urinary incontinence.
The outcomes of the retrospective comparative study conducted in Québec by Christou et al. [2004] also show improvements in the surgical patients’ health status, chiefl y with respect to cardiovascular disease (4.7% vs 26.7% for non-surgical patients: relative risk [RR] = 0.18), cancer (2.0% vs 8.5%: RR = 0.24), infectious diseases (8.7% vs 37.3%: RR = 0.23), endocrine disorders (9.5% vs 27.3%: RR = 0.35) and psychiatric or mental disorders (4.4% vs 8.2%: RR = 0.61). All these reductions in relative risk are signifi cant (p < 0.001). In return, the rates
of digestive problems were higher in the group of surgical patients (36.4% vs 24.7%: RR = 1.48; p < 0.01). Finally, the mortality rate for the surgical patients was roughly 10 times lower than that for the non-surgical patients (0.68% vs 6.17%, RR = 0.11; 95% CI: 0.04–0.27).
The meta-analysis by Buchwald et al. [2004], cited earlier in section 4.2.2.6 dealing with the weight-reduction effi cacy of bariatric surgery, also shows several positive effects on obesity co-morbidities (Appendix G). According to the data on 63 treatment groups (1846 patients), diabetes was completely resolved in 1417 patients (76.8%; 95% CI: 70.7–82.9%). In 30 other treatment groups (485 patients), it was resolved or improved in 414 patients (86.0%; 95% CI: 78.4–93.7%). The most effective techniques were gastric bypass and gastroplasty. Hyperlipidemia improved in 846 of 1019 patients in 23 treatment groups (79.3%; 95% CI: 68.2–90.5%). Hypertriglyceridemia decreased in 912 of 983 patients in 11 treatment groups (82.4%; 95% CI: 71.1–93.7%). Finally, hypercholesterolemia also improved in 1777 of 2051 patients in 14 treatment groups (71.3%; 95% CI: 55.5–87%). For these three hyperlipidemia outcome measures, the best results were documented with gastric-bypass procedures and biliopancreatic diversions.
Hypertension was controlled in 61.7% of the 4806 patients in 67 treatment groups (95% CI: 55.6–67.8%). In 43 other groups totalling 2141 patients, these problems were resolved or improved in 78.5% of the subjects (95% CI: 70.8–86.1%). Gastric bypass and gastroplasty were more effective in that regard. Lastly, sleep apnea was resolved in 85.7% of the 1195 patients in 38 studies (95% CI: 79.2–92.2%). According to the results of 24 other studies combining 726 patients, sleep apnea was resolved or improved in 83.6% of the patients (95% CI: 71.8–95.4%). It is diffi cult to determine which surgical technique is the most effective, given that effi cacy varies according to the outcome of interest.
35
5 ECONOMIC OUTCOMES
Although these studies were not randomized, they established a comparison between subjects eligible for surgery and willing to undergo it, and subjects matched by sex and 18 variables related to morbidity and mortality. The study by Sjöström et al. [1995], for its part, is more of a modelling exercise than a rigorous analysis. Below are the main fi ndings drawn from analysis of these studies.
A single study attempts to estimate the cost-utility ratio, which proves favourable to the surgical approach: it would result in savings from US$3,928 to US$4,004 per QALY (quality-adjusted life year)8 [van Gemert et al., 1999]. However, this study conducted in the Netherlands is not based on a true comparison between surgical and medical options because it uses a cost-of-illness approach to provide an economic counterpart for comparison. Furthermore, it includes only 21 patients who underwent vertical banded gastroplasty and who were followed for two years. The weight reduction, estimated to be 38.6%, was associated with a gain of 12 QALYs in a lifetime scenario. Productivity gains were estimated in part from the data provided by the subjects.
Several studies have compared resource-utilization outcomes and their associated costs, and have generally demonstrated the advantage of surgery:
First, Narbro et al. [1999] note that, at the fourth year of follow-up, the number of sick-leave and disability days is signifi cantly lower among the surgical patients than among the controls, and that these benefi ts are more pronounced for people older than 46.7 years (median age).
8. QALY (quality-adjusted life year): Calculation method allowing situations to be compared in relation to two criteria taken into account simultaneously, that is, effi cacy (number of life years gained) and the quality of life of those years.
The objective of this section is to report on the costs and the cost-effectiveness and cost-utility ratios of bariatric surgery, in light of the available scientifi c literature. The literature search was based on the results provided in the fi rst report published on this topic by AETMIS [CETS, 1998] and on an in-depth review conducted by British researchers [Clegg et al., 2002]. Since the British report took into account data identifi ed in Medline until October 19, 2001, this database was consulted to fi nd articles published until December 2004.
5.1 RESULTS OF THE ANALYSIS OF PRIMARY-DATA ARTICLES
The British report cited above identifi ed only four relevant articles [van Gemert et al., 1999; Chua and Mendiola, 1995; Martin et al., 1995; Sjöström et al., 1995]. The fi rst two had been cited in the previous AETMIS report [CETS, 1998]. Ten other more recent studies were also identifi ed [Monk et al., 2004; Potteiger et al., 2004; Sampalis et al., 2004; Snow et al., 2004; Angus et al., 2003; Gallagher et al., 2003; Ågren et al., 2002a; 2001; Nguyen et al., 2001; Narbro et al., 1999], three of which were related to the SOS study (Swedish Obese Subjects). Another article published in 2002 [Narbro et al., 2002] used the same data found in the recent study by Ågren et al., while another [Cooney et al., 2003], of more limited scope, analyzed the factors contributing to higher hospital costs incurred for the same type of surgical procedure. A detailed description of these studies can be found in Appendix H, except for the last study.
Unfortunately, none of these studies is complete enough to contribute to a valid evaluation of the cost-effectiveness or cost-utility ratio of bariatric surgery. Only one is randomized [Nguyen et al., 2001], but the analyses derived from the SOS study do present an adequate level of validity.
36
According to Ågren et al. [2002a], the surgical approach helps signifi cantly lower medication usage for obesity-related cardiovascular disease and diabetes by reducing not only the proportion of patients already under drug therapy but also that of patients required to begin such therapy (six-year follow-up). Nevertheless, weight loss has to reach at least 15% before the mean cost of total drug expenditures can decrease signifi cantly (7.8% reduction).
At six years of follow-up, hospitalization rates and costs remained signifi cantly higher for the surgical patients than for the controls. If the costs of the surgical procedure itself and the costs associated with its common problems are excluded, hospitalization costs for other problems do not differ between the two groups. In other words, at six years of follow-up, the decrease in the risk of heart disease associated with the reduction of obesity still did not translate into a reduction in hospitalization days.
Martin et al. [1995] estimate that a weight-reduction surgery program costs US$24,000 for a one-year period, while medical treatment costs US$3,000: these estimates relate only to direct health-care costs and do not include expenses resulting from complications. Furthermore, at three years, 95% of the subjects in the surgical group and 52% of those in the medical-therapy group had successfully reduced their weight by at least one third and had maintained this weight loss. However, the non-similarity of the comparison groups (especially in terms of initial weight and BMIs, age and medical history) and the large number of patients lost to follow-up weaken the validity of these outcomes.
Gallagher et al. [2003], who studied the impact of bariatric surgery on the U.S. Veterans Administration health-care system (n = 25), conclude that the costs of Roux-en-Y gastric bypass are offset by a reduction in health-care costs in the fi rst post-operative year. Mean pre-operative costs of out-patient visits (up to one
year before surgery) and hospital costs (up to three years before surgery) totalled US$10,558 per patient, while corresponding post-operative costs were reduced to US$2,840 (p = 0.005). Mean costs of peri-operative care, including eight days of hospitalization, totalled US$8,976. According to the authors, this signifi cant cost reduction, as similarly shown in other studies, would appear to be linked to a reduction in obesity-related diseases and symptoms.
Two studies comparing laparoscopic vs conventional (open) surgery conclude that the laparoscopic option offers greater cost savings [Nguyen et al., 2001; Chua and Mendiola, 1995]. According to the randomized study by Nguyen et al., hospital costs for laparoscopic gastric bypass are similar to those incurred with the conventional approach (approximately US$14,100). Operative costs were higher (p < 0.01) in the laparoscopic group, however, although they were offset by the lower costs (p < 0.02) for other hospital services, chiefl y because of shorter hospital stays and a lesser need for intensive care. In addition, subjects in the laparoscopy group were faster to return to their normal activities and work, and to attain a quality of life comparable to the one enjoyed by healthy people, according to U.S. standards (signifi cant deviations at three months). Another study, with weaker validity, was designed to analyze the specifi c issue of public and private health-care reimbursements. It indicates that laparoscopic Roux-en-Y gastric bypass costs less than open surgery (p < 0.001), despite higher direct operative costs. However, there were more complications in the laparoscopy group [Angus et al., 2003].
One particular study (not presented in the appendix) attempted to understand why certain gastric-bypass cases treated in compliance with a standardized clinical pathway were more expensive (greater than one standard deviation above the mean). The authors compared a group of 15 patients defi ned as having a cost-outlier profi le with a group of 73 patients having a normal cost profi le for total hospital care. The deviation can presumably be explained by the
37
difference in the prevalence of co-morbidities: there was a greater incidence of diabetes in the cost-outlier group (60% vs 33% ; p < 0.05), along with other severe medical conditions (60% vs 9.2%; p < 0.05), mainly sleep apnea with obesity-hypoventilation syndrome. Another factor that might explain it is that the incidence of major complications, especially gastro-intestinal and pulmonary, was also greater in the cost-outlier group (53.3% vs 7.9%; p < 0.05). Finally, the cost-outlier group had more re-admissions (46.7% vs 13%). The authors also mention that the laparoscopic procedures do not appear to cost signifi cantly more than the open-surgery procedures [Cooney et al., 2003].
Three recent studies reveal the pharmaceutical savings that bariatric surgery can achieve as a result of the resolution or improvement of the diseases associated with morbid obesity [Monk et al., 2004; Potteiger et al., 2004; Snow et al., 2004]. These three studies (described in section 4.4) were of patients who had undergone either laparoscopic or open RYGB.
According to Monk et al. [2004], the data available on 64 patients show the savings achieved in monthly medication expenditures following LRYGB: per-patient savings are estimated to be US$182, and this result is signifi cant (p < 0.01) (Appendix H). Extrapolating from these results to the nearly 75 million potentially concerned U.S. citizens, the authors estimate that total cost savings would amount to over US$100 billion per year. In a study of 51 consecutive patients,
Potteiger et al. [2004] estimated that monthly savings per patient would be US$145 (or US$1,736 per year) (p < 0.01). Considering the mean age of the study population (45 years) and projecting the medications prescribed to these patients over 30 years, the authors arrive at a total cost of US$52,080 and compare this estimate with the cost of an RYGB procedure (mean cost of US$14,700 at the medical centre in the study). This comparison does not take into account QALYs, the progression of morbid obesity
or other subsequent obesity co-morbidities (Appendix H). The authors conclude that a more detailed analysis is required. In their study of 78 patients who underwent
LRYGB, Snow et al. [2004] report a decrease in prescription-medication costs on the order of 68% as early as the fi rst year post-operatively (monthly savings of US$250 per patient) and of 72% for the second year (US$264 per patient per month). Considering that the mean cost of LRYGB is US$8,090 (US$631,000 for the 78 patients), the authors conclude that this treatment is signifi cantly more effective and economical than pharmaceutical treatments for morbid obesity and its co-morbidities, since it yielded annual savings of US$240,000 for the 78 patients. In other words, the cost of surgery is offset by the savings achieved by the end of 32 months (Appendix H).
The study conducted by Christou et al. [2004] at the McGill University Health Centre (MUHC) (described in section 4.4) also had an economic component [Sampalis et al., 2004]. Evaluated direct costs include human and material resource expenditures associated with hospital care9 and payment for medical services10. The necessary data were extracted from provincial administrative databases on hospitalizations and medical procedures and included internal MUHC data; costs are expressed in 1996 Canadian dollars. The surgical-treatment cohort incurred higher direct health-care costs in the fi rst year than did the control cohort, per 1000 patients: $12,461,938 vs $3,609,680 respectively. At fi ve years, mean cumulative costs were $19,516,667 vs $25,264,608, for a mean difference of almost $6 million per 1000 patients. Surgery costs were amortized after three and a half years.
9. Hospital costs include the costs for hospital bed use, nursing care, intensive care, food, medications, operating-room costs, diagnostic procedures, medical and surgical supplies, dietetics services and other paramedic services, including physiotherapy.
10. Medical services include out-patient visits, consultations, and medical and surgical procedures (anesthesia, surgery, psychiatric and other concerned specialties).
38
5.2 MODELLING RESULTS
Given the dearth of primary data and the lack of studies establishing valid cost-effectiveness or cost-utility ratios, two studies relied on modelling in an attempt to estimate these ratios [Clegg et al., 2002; Craig and Tseng, 2002]
5.2.1 Model developed by Clegg
The model developed by Clegg et al. [2002] compares the impact and costs of three types of surgical procedures—gastric bypass (mainly Roux-en-Y), vertical banded gastroplasty and adjustable gastric banding—with those of the non-surgical management of patients with morbid obesity. The outcomes compared include excess-weight reduction, quality-of-life gains and life-expectancy gains. Costs include only direct health-care costs: pre-operative care, surgical procedures, treatment for complications, surgery revisions and additional procedures, and follow-up. Calculations are based on the follow-up of a hypothetical cohort of 100 patients with the following characteristics: mean age of 40; 90% women; mean body weight and BMI of 135 kg and 45 kg/m2 respectively. The authors do not justify the values assigned to the variables they use. The time horizon is 20 years after surgery, and the discount rate is 6%.
Weight-reduction effi cacy scenarios are based on the data available in the literature for each of the fi rst fi ve years, after which the weight achieved is considered to be stable. The effi cacy of non-surgical treatment is assumed to be nil. After having considered different sources, the authors selected, as a quality-of-life gain measure, the utility values categorized by the patients’ ages and BMIs supplied by Roche Pharmaceuticals in its submission for approval for orlistat, a weight-loss medication. In terms of co-morbidity reduction, only diabetes was selected by the authors because studies had shown a substantial and lasting impact on this disease, while the effects on hypertension were transient, and there were insuffi cient data on the
other health problems. The impact on morbidity was measured by the reduction in treatment costs. Finally, the authors cautiously assumed no effect on life expectancy, owing to the lack of specifi c clinical data.
Cost parameters are based on published data, expert opinion, common practices in England, and unit costs established by the Scottish Health Service and by a British study. To offset their choice of fi xed values, which depend on assumptions that are often non-reproducible, the authors add a sensitivity analysis. This analysis varies hospital length of stay, pre-admission and post-admission costs, weight loss, in both the surgical and the medical scenarios, the effect of the surgical learning curve, costs associated with diabetes and utility gains.
Model results indicate that, compared with the medical option, gastric bypass, adjustable gastric banding and vertical banded gastroplasty cost £6,289, £8,527 and £10,237 respectively per QALY gained, which is lower than the £20,000 cut-off point suggested by England’s National Institute for Clinical Excellence [NICE, 2004]. The authors of this model also indicate that these ratios are lower than those of the most relevant comparators, that is, pharmaceutical treatment with orlistat or sibutramine. The different scenarios used in the sensitivity analysis help support the robustness of this conclusion. Furthermore, comparisons between the different surgical procedures still have too many uncertainties to classify any one of them as the treatment of choice. As the authors point out, the selected assumptions, which were more conservative, had an unfavourable effect on the cost-utility ratios. For example, a positive effect on life expectancy or on the reduction of related or secondary diseases other than diabetes would be associated with an improvement in these ratios, as would larger proportions of men or young adults. Furthermore, their analysis did not include non-medical benefi ts, such as productivity gains, in accordance with the recommendations in the NICE guidance.
39
5.2.2 Model developed by Craig and Tseng
In developing their model for the U.S. context, Craig and Tseng [2002] selected the perspective of the payer to estimate the cost-utility ratio of gastric bypass in relation to no treatment. They obtained this ratio by dividing the difference in total lifetime medical costs by the difference in QALYs between the two options, with a 3% discount rate. The morbidly obese population was defi ned as follows: BMI between 40 and 50 kg/m2, aged 35 to 55, non-smoker, no cardiovascular disease, drug addiction or major psychological problems, and unsuccessful medical therapy. A determinist model (decision tree) specifi ed the different rates and probabilities associated with each option, that is, life expectancy, discounted QALYs, surgical success rates, intra-operative deaths, surgery revisions and reconstructive plastic surgery (associated with complication rates), and the costs related to each possible clinical pathway. The model was based on different assumptions: for example, weight loss after a successful operation was considered to be stable at fi ve years.
Data on rates, probabilities and costs came from different single sources: for example, weight-loss estimates and complications rates were derived from a single clinical study published by Pories et al. [1995], while estimates of life expectancy and the burden of obesity (lifetime medical costs) came from a study by Thompson et al. [1998]. To estimate quality of life, the authors used national-survey data adjusted by means of a regression method and different assumptions.
In the base-case model, cost-utility ratios varied between US$5,000 and US$16,000 per QALY for women, and between US$10,000 and US$35,600 per QALY for men, depending on their ages and initial BMIs. In a few subgroups of older and less obese men, varying some key parameters infl uenced the cost-utility ratios by increasing them beyond the cut-off of US$50,000 (a reference cut-off point used by the authors, but without justifi cation). This
sensitivity analysis did not lead to appreciable changes for the other subgroups.
In summary, the authors conclude that, although gastric bypass does not lead to health-care cost savings, it has an acceptable cost-utility ratio. They indicate that their analysis has several limitations, some of which were taken into account in the sensitivity analysis, especially the fact that the weight-loss percentage was derived from a single study with a large number of patients lost to follow-up and that their QoL evaluation was not based on specifi c data. In addition, their analysis does not include patients with co-morbidities such as diabetes, cardiovascular disease or hypertension: intra-operative risks would have been greater, but so would weight-loss benefi ts. Finally, given that the study is based on the payer perspective, it excludes non-medical costs, such as those related to patients’ decreased productivity and lost wages.
5.3 COST OF BARIATRIC-SURGERY PROCEDURES
According to the estimates derived from countries where public health-care systems predominate (Australia, Netherlands, United Kingdom and Sweden), direct costs may vary between CAN$4,968 and CAN$10,870 [MSAC, 2003; Ågren et al., 2002a; Clegg et al., 2002; van Gemert et al., 1999]. This variation depends on insured costs (hospital care and physician fees, general hospital-service charges), the type of surgery, the inclusion or not of potential complications, and consideration or not of the follow-up period. It is impossible to analyze the differences in these estimates because the information is either not detailed enough or incomplete, and the organizational settings differ. According to the analyses by Clegg et al. (which include surgery revisions, re-operations, treatment for complications and follow-up) and by Australia’s Medical Services Advisory Committee (which include only the main procedure itself), the cost of LAGB is higher than that of Roux-en-Y gastric bypass (RYGB)
40
and that of vertical banded gastroplasty (VBG), whereas VBG costs less than RYGB [MSAC, 2003; Clegg et al., 2002].
According to the previous AETMIS report, the procedure performed at the Royal Victoria Hospital at the MUHC—a basic Roux-en-Y gastric bypass—would cost roughly CAN$10,500 [CETS, 1998]. According to the data provided by the bariatric-surgery follow-up committee at Hôpital Laval in 2001, the mean cost of a biliopancreatic diversion would be CAN$10,719, excluding surgeons’ fees [Comité de suivi de la chirurgie bariatrique, 2001]. Finally, the MUHC’s Technology Assessment Unit estimated direct costs for LAGB and LRYGB to be $9,418 and $7,064 respectively. These costs include the expenses associated with the surgical procedure, conversion, follow-up, complications and physician fees. The higher cost of LAGB is chiefl y due to the cost of adjustable bands [Chen and McGregor, 2004]. If general hospital-service costs were added (based on a rate of 30%), total costs would rise to $11,634 for LAGB and $8,666 for LRYGB.
In its assessment of bariatric surgery, the Medical Advisory Secretariat (MAS) of Ontario’s Ministry of Health and Long-Term Care estimates the total cost of gastric bypass or gastroplasty to be $6,185, based on 2003 fi nancial data. That amount covers only the procedure itself, not pre-operative consultations, follow-up, surgery revisions and hospitalizations due to complications. In that total, hospital costs rise to an estimated $4,890, based on a resource-use index by diagnostic category, whereas professional services (surgeon, anesthetist and surgical assistant) account for the remaining $1,295. Professional fees are added if patients have BMIs greater than 45 kg/m2, totalling $23.54 for the anesthetist and $61.30 for the surgeon. If the gastric banding were to be covered as an insured service, roughly $800 would need to be added. It is interesting to note that the Ontario Ministry of Health and Long-Term Care planned to reimburse $10,842 on average per bariatric-surgery procedure performed outside the province [MAS, 2005].
5.4 RECAPITULATION OF THE ECONOMIC EVALUATION
According to the current state of evidence, despite the limitations of the published studies and models, bariatric surgery would improve health and quality of life at rather high additional costs, but comparable to those of other medical treatments or health-care services. Its effi cacy in producing sustained weight loss is well established, and studies seem to indicate that this weight loss lowers the prevalence of co-morbidities (e.g., cardiovascular disease, diabetes) and their consequences (medication costs), reduces productivity losses caused by sick leave and disability, and improves quality of life. Bariatric surgery is a relatively expensive operation. It also leads to a certain number of complications and the need for an annual follow-up, which requires additional health-care resources. Although this point has been overlooked in the published studies, a large number of patients might also need plastic surgery after this treatment. If plastic surgery were to be deemed medically necessary, it would increase the burden of treatment costs. According to the model by Clegg et al. [2002], the net cost of surgery to treat morbid obesity (i.e., taking into account the savings attributable to diabetes control) would total between CAN$22,000 and CAN$24,700 over a 20-year period. By comparison, a less effi cacious treatment program would cost a net total of CAN$15,925. Although bariatric surgery seems cost-effective, it is still necessary to conduct well-designed economic studies and to be able to rely on longer-term actual data not only on effi cacy and quality of life but also on resource-utilization costs (or savings). Similarly, studies are needed to solidly establish the impact and relative costs of the different surgical approaches, especially laparoscopy, given that this technique is rapidly becoming widespread and the follow-up period associated with this technique remains less than fi ve years.
41
6 REVIEW OF THE VARIOUS HEALTH-TECHNOLOGY ASSESSMENT REPORTS
6.1 BARIATRIC SURGERY no clinical or psychological contra-indications to surgery under anesthesia. In addition, patients must agree to long-term follow-up. The choice of surgical technique depends on patient preferences and characteristics and on the surgeon’s learning curve.
In its latest clinical practice guidelines on the management of patients with morbid obesity, Australia’s National Health and Medical Research Council [2003] considers that bariatric surgery is the treatment of choice for these patients when all other therapeutic approaches have failed. The NHMRC adds that this treatment must take place under strict medical and nutritional follow-up.
From its systematic review of the literature published between 1992 and 2002, the Basque agency Osteba concludes that purely restrictive surgical techniques are safer, although their long-term effi cacy is weak. Conversely, procedures inducing malabsorption have greater risks of complications and a steeper learning curve. They nevertheless yield better weight-loss outcomes. Osteba considers that, “in light of available data, gastric bypass may be the procedure that best balances safety and effi cacy, that restrictive procedures should be reserved to a select subgroup of patients, and that super-obese patients may benefi t from a technique that includes induced malabsorption” (Free translation) [Rico Iturrioz et al., 2003]. These conclusions support the fi ndings of its prior report, which examined the effi cacy of the main techniques in use (vertical banded gastroplasty, gastric banding and Roux-en-Y gastric bypass), while pointing to the contextual elements to be taken into consideration, that is, the learning curve and multidisciplinarity of the team in charge of the procedure as well as patient characteristics [Egino Sasiain et al., 2000]. In a similar vein, Osteba suggests that all surgical cases of morbid obesity be regularly recorded in
The literature review performed by the Agence Nationale d’Accréditation et d’Évaluation en Santé [2001] in France considered different reports published worldwide on the surgical treatment of morbid obesity: Scottish Intercollegiate Guideline Network, 1996; Centre for Review and Dissemination, University of York, 1997; National Health Services, 1997; National Heart, Lung and Blood Institute, 1998; and the Conseil d’évaluation des technologies de la santé, 1998. The ANAES selected three types of procedures for its evaluation: vertical banded gastroplasty, adjustable gastric banding and gastric bypass.
While acknowledging the place occupied by bariatric surgery in the management of morbid obesity and the sharp rise in the number of surgical procedures, the ANAES concludes that, “when the indication is correctly diagnosed, vertical banded gastroplasty and gastric bypass present an acceptable risk-benefi t ratio. The same holds true for gastric banding, based on its short-term evaluation” [Free translation]. Nevertheless, this agency emphasizes that each technique leads to specifi c complications that may sometimes impair the patient’s vital prognosis. The ANAES points to the lack of long-term follow-up that does not permit evaluation of the actual incidence of late complications from gastric banding. Consequently, it deplores the fact that this technique is being put into widespread use without ever having been subject to evaluation.
The National Institute for Clinical Excellence [2002] in Great Britain recommends bariatric surgery for people over the age of 18 who have morbid obesity, who are followed up by a bariatric specialist in a hospital centre and who have been unsuccessfully treated with other non-surgical therapeutic options. Patients must have
42
order to provide better patient care by offering an appropriate choice of procedure based on patient characteristics. This registry would also help evaluate the advisability of introducing new surgical techniques.
In a synthesis study on the effi cacy of pharmacotherapy, counselling or behaviourist approaches and surgery for the treatment of obesity, which was prepared for the U.S. Preventive Services Task Force, McTigue et al. [2003] found certain limitations in previous systematic reviews:
They have widely differing eligibility criteria, treatment classifi cations and data-synthesis methods; Aggregate values of their fi ndings do not
refl ect the variations in sample size, follow-up periods, and treatment differences that are found in randomized comparative rials.
Despite the limitations of these studies, McTigue et al. point out that they did achieve consistent fi ndings. They conclude that only surgery permits signifi cant long-term weight loss, even if it is associated with a low risk for severe complications. Body size, health status, and prior weight-loss history may all infl uence the choice of appropriate treatment.
In its review of the evidence on the effi cacy of pharmacological and surgical treatments for obesity, the U.S. Agency for Healthcare Research and Quality (AHRQ) concludes that bariatric surgery is more effective than currently available non-surgical treatments for patients with BMIs greater than 40 kg/m2. Surgery also translates into improved health outcomes (reduced diabetes and sleep apnea, and better quality of life). Although the data seem to indicate that it leads to sustained weight loss in patients with BMIs between 35 and 40, they are not conclusive enough and further studies are required. Despite the scarcity of comparative studies on the different techniques, the AHRQ affi rms that Roux-en-Y gastric bypass (RYGB), vertical banded gastroplasty (VBG) and adjustable gastric banding (AGB) bring about substantial weight loss, but that
RYGB seems more effi cacious than VBG for people with BMIs greater than 40. This assessment agency nevertheless suggests that the various techniques may result in considerably different complications and adverse effects. In addition, the proportion of patients experiencing complications may be quite large, greater than 20% (although most are minor in severity). A post-operative mortality rate of less than 1% has been achieved by a fair number of surgeons and bariatric-surgery centres, although it may be higher in other settings. The authors recommend that randomized controlled trials be performed to compare the effi cacy and safety of the different types of surgical procedures and that well-designed comparative studies (with pharmacological and behaviourist approaches) be conducted to study the effects of bariatric surgery on obesity co-morbidities. These conclusions are based on 245 studies published until 2003 (78 on medical therapy and 167 on bariatric surgery) [Shekelle et al., 2004].
The Ontario assessment agency recently published its fi ndings from its analysis of 15 systematic reviews or assessment reports and English-language articles on bariatric surgery published between July 2002 and September 2004. In its conclusions, the Medical Avisory Secretariat (MAS) confi rms the effi cacy of surgical treatment in general for patients with BMIs greater than or equal to 40 kg/m2 or with BMIs greater than or equal to 35 in the presence of co-morbidities (major weight being given to the evidence in the SOS study). It also confi rms that malabsorptive techniques are superior to restrictive ones. However, the lack of solid evidence cross-comparing the different techniques does not allow it to pass judgment on the superiority of any one of them. In that regard, no conclusion is drawn on laparoscopic or open approaches [MAS, 2005]. The Ontario Health Technology Advisory Committee (OHTAC), which guides and approves the work of the MAS, is responsible for the recommendations derived from MAS systematic reviews. In terms of bariatric surgery, the OHTAC endorses the conclusion regarding
43
its effi cacy and recommends the development of detailed selection criteria and the establishment of centres of excellence. These centres should be properly equipped to accommodate patients who are severely obese and should have an interdisciplinary team of specialists to conduct pre-operative assessments and post-operative follow-up for those patients. Moreover, these centres should offer gastric banding as an insured service. Finally, the OHTAC recommends that the particular type of bariatric procedure should be chosen after a surgeon–patient discussion that includes examination of the risks and benefi ts of each procedure.
6.2 LAPAROSCOPIC SURGICAL PROCEDURESIn its 2001 assessment report, the Alberta Heritage Foundation for Medical Research (AHFMR) concludes that laparoscopic bariatric surgery is still in its “learning stage” and that this approach requires two types of expertise: bariatric-surgery skills and laparoscopic skills. Reported complications and the lack of long-term follow-up make it diffi cult to foresee the impact that this technique might have on the management of obese patients [Hailey and Harstall, 2001].
In its April 2000 report, the Australian Safety and Effi cacy Register of New Interventional Procedures – Surgical (ASERNIP-S), while drawing attention to the lack of valid comparative data, points out some of the potential benefi ts of laparoscopic procedures. Its assessment specifi cally covered laparoscopic adjustable gastric banding and indicated that the mean length of surgery was generally less than two hours, and the mortality rate less than 1 in 1000. The morbidity rate was similar to that of the other types of procedures, the hospital stay was shorter, and quality of life better. The long-term effi cacy of this surgical technique remains to be demonstrated [Chapman and Kiroff, 2000]. In a second report published in June 2002, the ASERNIP-S reiterated its conclusions, while expressing reservations about the superiority
of Roux-en-Y gastric bypass (RYGB) over laparoscopic adjustable gastric banding (LAGB) for weight-loss maintenance in the medium term (from two to four years). This organization stresses the need for randomized controlled trials and longer-term follow-up [Chapman et al., 2004].
Following the report by the ASERNIP-S and in the context of the steadily rising number of bariatric surgeries, the Australian Minister for Health and Ageing (through the Medicare Benefi ts Branch) asked the Medical Services Advisory Committee (MSAC) to assess the effi cacy and safety of LAGB (Lap-Band device). The comparators used for this assessment were RYGB, the current gold standard, and vertical banded gastroplasty (VBG), the most commonly performed procedure in Australia. According to their report, LAGB may eventually replace VBG. For its analysis, the MSAC selected 174 studies and three health-technology assessment reports published up to July 2002. In its conclusions, the MSAC fi rst indicates that LAGB is as safe as the other procedures. In terms of weight loss, LAGB is less effi cacious than RYGB, but as effi cacious as VBG, although preliminary evidence seems to indicate that LAGB maintains weight loss for a longer period. Even if LAGB costs more than the comparators, the incremental cost-utility ratio is considered acceptable. On the strength of these fi ndings, the MSAC recommends that this procedure continue to be publicly funded [MSAC, 2003].
In a report published in 2003, the Technology Evaluation Center (TEC) at the Blue Cross and Blue Shield Association examined the most recent bariatric-surgery techniques. The assessment was based on the following specifi c criteria: fi nal approval from a government regulatory body for the technology used, conclusive scientifi c evidence on the impact of the technology on health outcomes, improvement in the net health outcome, superior or equivalent benefi ts by comparison with other established techniques, and the reproduction of the last two effects in current practice. The techniques considered established were RYGB,
44
VBG, horizontal gastroplasty, and AGB. In light of its analysis, the TEC concludes that laparoscopic gastric bypass and vertical banded gastroplasty, as well as biliopancreatic diversion, and distal or long-limb gastric bypass, do not meet its criteria and are therefore not eligible for coverage [BCBS, 2003b].
In its most recent clinical policy bulletin, the U.S. insurer AETNA [2004], after analysis of evidence-based clinical information, concludes that RYGB is medically necessary (and therefore eligible for coverage). The same applies to VBG and LAGB, but only for patients who are at increased risk of complications from RYGB, owing to the presence of any of the following co-morbid medical conditions: hepatic cirrhosis, chronic infl ammatory bowel disease, radiation enteritis, demonstrated complications from extensive adhesions involving the intestines from prior major abdominal surgery, poorly controlled systemic disease. Some of the procedures expressly excluded from coverage, owing to a lack of suffi cient evidence, are loop gastric bypass, horizontal gastroplasty, BPD-DS, BPD (Scopinaro), along with LAGB and VBG, except for the previously mentioned specifi c indication (at-risk patients).
In a report released in April 2004, the Technology Assessment Unit of the McGill University Health Centre (MUHC) published its assessment of laparoscopic adjustable gastric banding (LAGB) (Swedish adjustable gastric band). This study relied on the report by the Australian agency ASERNIP-S [Chapman et al., 2004] and selected 19 additional studies published between May 2001 and February 2004. The selected comparator was LRYGB. The authors conclude that, according to the data from a follow-up period covering up to fi ve years, LAGB is effective (in terms of excess weight loss) and safe, with complication and mortality rates comparable to, if not lower than, those of RYGB. This procedure, however, costs 39% more than its comparator. However, so long as this procedure is not approved and partially covered by the Québec government, it recommends that the MUHC not use it except in special cases, and only after consultation with the surgeon. Furthermore, even if the government did approve LAGB, it would be necessary to demonstrate its long-term clinical superiority before the MUHC would accept it as the procedure of choice [Chen and McGregor, 2004].
45
7 DISCUSSION
international survey, RYGB accounted for 85% of the operations performed in the United States and Canada in 2003, while VBG represented only 1.4% of the procedures [Buchwald and Williams, 2004; Fisher and Schauer, 2002]. This technique is performed more often in other countries, such as Sweden, Poland, Greece, Italy and Egypt. The international survey estimates that VBG accounted for 5.43% of the procedures. The major reason is that VBG is a purely restrictive technique, whereas RYGB combines restriction and induced malabsorption. American surgeons have pointed out that VBG produces unsatisfactory outcomes in terms of sustained weight loss, re-operation rates and adverse events such as gastro-esophageal refl ux and inability to tolerate solid food [Balsiger et al., 2000; Mason et al., 1992; Sugerman et al., 1987]. Belachew12 adds that a very high rate of vertical-staple disruptions has been observed. Moreover, in Australia, laparoscopic adjustable gastric banding (LAGB) is quickly supplanting VBG: according to the international-survey data, the number of LAGB procedures reached 80% in 2003.
In the case of biliopancreatic diversion (BPD), there are fewer assessments and long-term follow-up. As a general rule, this lack is due to the technical requirements for this type of procedure or to the fact that this practice is limited to specifi c institutions, or to both factors. It is important to point out the similarity between long-limb Roux-en-Y gastric bypass and Scopinaro’s biliopancreatic diversion. The characteristics shared by these two procedures have led some authors to underestimate or overestimate the number of BPD procedures, depending on whether or not they consider this type of RYGB to be a biliopancreatic diversion. Over the past few years, use of the laparoscopic variations of these procedures (gastric resection
12. Written personal communication, April 5, 2004.
Among the different techniques, gastric bypass is the most frequently performed procedure (Table 13). The American Society of Bariatric Surgery estimates, from data obtained from its members, that 103,200 bariatric surgeries11 were performed in 2003, and that this fi gure reached 140,640 in 2004 [Colwell, 2005]. Practitioners have diverging opinions on biliopancreatic diversion (BPD) because of its technical complexity and problems linked to malabsorption, especially if patient management is inadequate (e.g., vitamin defi ciencies or malnutrition).
In Québec, fi fteen or so surgeons perform surgical procedures to treat morbid obesity. Current bariatric centres are located in Montréal, Quebec City, Sherbrooke, Longueuil, Val-d’Or and Drummondville. Surgical techniques used are:
gastric bypass; vertical banded gastroplasty; biliopancreatic diversion; adjustable gastric banding (Lap-Band).
Currently, the effi cacy and safety of vertical banded gastroplasty (VBG) and Roux-en-Y gastric bypass (RYGB) are well established (follow-up > 15 years). However, most bariatric-surgery specialists consider RYGB to be the gold standard and the comparator of choice in most comparative studies in the fi eld. The latest data from an international survey (26 out of 31 countries responded to the survey) indicate that gastric bypass (mostly different versions of RYGB) accounts for 65.11% of the total number of procedures [Buchwald and Williams, 2004].
For its part, VBG is quickly losing impetus, especially in the United States, and is being increasingly replaced by RYGB. According to an
11. This estimate is close to the one by Pandolfi no et al. [2004], who report a fi gure closer to 100,000.
46
and other combined methods) has become more frequent, but these techniques remain limited to some surgical teams and are currently under assessment.
It is worth mentioning that the World Health Organization [2003] states that it has yet to see an in-depth assessment of the safety and long-term effi cacy of biliopancreatic diversion and of the laparoscopic versions of restrictive techniques (e.g., adjustable gastric banding) and mixed techniques. Intestinal bypass is not recommended as a fi rst-line surgical treatment for obesity.
In the CETS report [1998] on the surgical treatment of morbid obesity, biliopancreatic diversion with duodenal switch (BPD-DS) had been classifi ed as an experimental technology (see Appendix A for the classifi cation of health technologies). Technological advances, the number of surgical patients, the length of follow-up for surgical patients and published fi ndings have led to recognizing this technique as effective: it contributes to sustained weight loss comparable to that achieved by other technical procedures, and even better in some cases, according to some authors. Although this technique is no longer considered experimental,
too few comparative studies have been done so far, leaving a number of points that still need clarifi cation with respect to its most appropriate indications and its place among the range of bariatric techniques for treating morbid obesity.
In addition, biliopancreatic diversion, along with its variant—biliopancreatic diversion with duodenal switch (BPD-DS)—is still not widely used since it represents only 4.85% of the procedures performed worldwide, according to the international survey by Buchwald and Williams [2004]. In actual fact, BPD-DS is performed by only 128 surgeons around the world,13 including one team in Québec. Its relatively low use is related more to surgeons’ belonging to a particular “school” and/or to their experience than to concerns about its effi cacy. As a few authors have pointed out, insurers’ different decisions and even lack of clear policies on covering this procedure (except for the Blue Cross and Blue Shield Association) show that the problem is not related to the outcomes achieved with this technique but to the fact that it is used infrequently. For example, in its assessment of the surgical treatment of morbid obesity, the ANAES [2001] excluded BPD-DS because it is rarely performed in France and elsewhere in Europe.
13. Personal communication from Dr. N. Scopinaro, April 7, 2004.
TABLE 13
Types of procedures performed by the members of the International Bariatric Surgery Registry
TYPE OF PROCEDURE* RELATIVE SHARE
Gastric bypass Roux-en-Y gastric bypass (and variants) Other gastric bypass procedures (with transverse dissection of stomach)
“Simple” procedures Vertical banded gastroplasty Silicone banded gastroplasty Others
“Complex” procedures: biliopancreatic diversion, ileogastrectomies, etc.
71%(48%)(23%)21%
(12%) (8%) (1%) 8%
Source: IBSR, 2001 (data confi rmed on June 7, 2004).* No distinction is made between open and laparoscopic procedures.
47
Laparoscopic bariatric surgery is becoming increasingly used to treat morbid obesity. Numerous studies on this topic have been published, but to date too few randomized clinical trials have been conducted to assess the effi cacy or safety of this technique compared with other types of open bariatric surgery, especially Roux-en-Y gastric bypass, which has become the gold standard. Most of the studies on this approach do not yet have a long enough perspective (generally less than fi ve years) to be able to assess its level of weight-loss maintenance or the incidence of late complications. Nevertheless, the clinical profi le for this procedure, including elements such as a low operative risk, an acceptable morbidity rate and reversibility, makes it a promising procedure for the management of morbid obesity, even though a certain number of complications directly linked to the laparoscopic approach have yet to be resolved.
According to the international survey by Buchwald and William [2004], 62.85% of the procedures performed worldwide in 2003 were done laparoscopically: gastric bypass LRYGB accounted for 34.59% of them, and adjustable gastric banding (LAGB) for 24.16%. These data indicate that 45% of all RYGB procedures are done laparoscopically in North America (United States and Canada). Furthermore, the approval of LAGB (Lap-Band) in June 2001 by the U.S. Food and Drug Administration (FDA) opened the door to the rapid proliferation of this technology. In Canada, three devices have been approved and are commercially available. No province has yet agreed to specifi cally cover this laparoscopic technique [Chen and McGregor, 2004], even though the OHTAC recommended in 2005 that it become an insured service [MAS, 2005]. In Europe, LAGB is the most frequent type of procedure. As a point of illustration, a national survey in France found that this procedure accounted for 88.3% of all bariatric surgeries [CNAMTS, 2004].
The rapid spread of this technology has led several authors to raise questions about the consequences related to the learning curve, which permits surgeons to acquire the necessary expertise to perform this technique in all
safety. In a recent study of 600 consecutive patients treated surgically for morbid obesity (LRYGB with hand-sewn sutures), Ballesta-Lopez et al. [2005] noted that most outcome measures (length of surgery, complication rate, conversion rate, and mortality rate) improved with the number of procedures performed. The worst outcomes were documented for the fi rst 100 surgical patients. These fi ndings confi rm the conclusions of other authors, who have stated that the improvement curve for these outcome measures reaches a plateau between 75 and 100 procedures [Oliak et al., 2003; Schauer et al., 2003b; Papasavas et al., 2002].
The relationship between the volume of procedures and the incidence of complications has also been demonstrated in an analysis of hospital-discharge data in the state of Pennsylvania [Courcoulas et al., 2003b]. Among the 4685 patients who underwent gastric bypass, 28 deaths (0.6%) and 813 adverse outcomes (17.4%) were documented. For surgeons who performed fewer than 10 operations per year, the risk of adverse outcomes was 28% and the risk of mortality was 5%, compared with 14% (p < 0.05) and 0.3% (p < 0.06) respectively for higher-volume surgeons. Although the relationship between adverse outcomes and hospital volume did not reach signifi cance, the authors nevertheless noted that surgeons who treated from 10 to 50 cases per year had a 55% risk of adverse outcomes. These fi ndings argue in favour of concentrating bariatric surgery in centres that are well equipped and able to rely on experienced surgeons.
Regardless of the satisfactory outcomes published for each surgical technique studied individually, the diversity of the characteristics of patients with morbid obesity and the inadequate number of well-designed comparative studies do not yet allow one technique to be systematically favoured over the other. The outcomes associated with a given surgical technique depend to a great extent on the following factors (other than its intrinsic value, the surgical learning curve and the patient’s personal preference):
exclusion of patients at risk of post-operative complications;
48
psychological preparation;
clinical preparation: for example, at the Mercy and Unity Hospital in Minneapolis, which set up a bariatric centre for patients with morbid obesity (36 bariatric surgeries per month, mean length of stay from two to three days, and mortality rate less than 1%), potential patients are “requested” to lose 10% of their excess weight before surgery, which could be a way of evaluating how determined they are to reach the objectives set by their treating physicians; acceptance of long-term, if not lifelong,
follow-up.
The treatment of super-obese patients presents several challenges that can be overcome by some of the surgical techniques evaluated, including RYGB, VBG-RYGB, and biliopancreatic diversion. More recent data reveal the fi rst promising experiences with LRYGB for patients with BMIs greater than 50 kg/m2 and even 70 kg/m2 [Dresel et al., 2004; Kreitz and Rovito, 2003]. In addition, as a result of a trial conducted with patients with BMIs greater than 60, a group from the Mount Sinai School of Medicine (New York) suggest that this procedure be performed in two stages: a sleeve gastrectomy followed by RYGB from six to nine months later [Regan et al., 2003].
There appears to be a growing reliance on consensus for developing standards for assessing the quality of surgical procedures in this particular fi eld: in the United States, a post-operative leak rate from 1% to 2% and a rate of infectious wounds less than 5% in operated obese patients are considered “usual” values for this type of procedure. The same is true for hospital stays from 2.5 to 4 days (laparoscopy) and from 3.5 to 4 days (open surgery).
As a general rule, as emphasized by the ANAES [2001], no surgical technique may be considered harmless, and recommendations must be strictly followed (surgeon’s learning curve with the technique, appropriate clinical setting, management by a multidisciplinary team, and long-term medical follow-up). Many authors stress that bariatric surgery performed by inexperienced surgeons or the lack of strict,
long-term follow-up may lower the quality of patient management.
A further aspect involved in patient management concerns the impact of signifi cant weight loss, given that it may require reconstructive plastic surgery. Although the present assessment has not examined this issue, recommendations for effective bariatric surgery specify that therapeutic treatment plans must include this aspect [Mouïel et al., 2004]. A recent report from the Scottish Executive Health Department [2004] confi rms this recommendation by emphasizing that plastic surgery is an integral part of an overall bariatric-surgery management program and, consequently, patient selection criteria must be clearly defi ned. Plastic surgery restores body image and contributes to psychological healing; however, it must be considered only after patients have achieved a stable weight, at least 12 months after bariatric surgery.
Finally, the techniques for treating morbid obesity have been constantly evolving, as shown by the application of robotics to bariatric surgery and the use of gastric stimulators to promote weight loss. Preliminary results from the use of a Da Vinci robot to perform a laparoscopic Roux-en-Y gastric bypass procedure seem encouraging, although its long-term benefi ts and drawbacks require further investigation [Mohr et al., 2005]. Gastric stimulators, one type of which has already been approved in Canada, are surgically implantable devices that send electrical impulses to the abdominal-wall muscles causing different reactions that lead to reduced appetite and a rapid sensation of satiety. Despite positive weight-loss outcomes, more studies are required to better understand their mechanism of action and to identify the type of patient who would benefi t from this device [Cigaina, 2004; De Luca et al., 2004]. It would therefore be advisable to implement a horizon-scanning system to track the evolving status of these new techniques, especially because one of them, gastric simulation, is the subject of trials at the McGill University Health Centre in Montréal.14
14. Information taken from http://www.weightlosssurgery.ca/en/surgery.php (accessed on September 9, 2005).
49
8 CONCLUSION
The choice of surgical technique depends on a number of factors:
Patient characteristics: age, personality, BMI, food patterns, personal understanding and commitment, co-morbidities, contra-indications; Reversibility or non-reversibility of the
technique; Risks linked to each technique (e.g., wound
dehiscence, hernias, device slippage, staple-line disruption); Potential effects of nutritional defi cits; Availability of human and material resources; Support provided by the expertise of a
multidisciplinary team; and Surgical team’s experience in both general
and specifi c terms: for example, laparoscopy requires great dexterity and therefore involves a rather steep learning curve.
Although the laparoscopic approach raises specifi c issues, which are described further on, examination of the evidence on the main techniques used in Québec lead to the following observations:
Roux-en-Y gastric bypass (RYGB): This technique has established effi cacy in terms of stable weight loss, a low complication rate and a positive impact on co-morbidities. For most bariatric specialists, it has become the gold standard. It is the most commonly used technique in the general group of gastric-bypass procedures.
Vertical banded gastroplasty (VBG): Although this technique has established effi cacy, it has achieved lower than expected weight loss and has lost favour with North American surgeons. Combined with RYGB, VBG yields good long-term results.
This analysis of the various studies and assessment reports currently available on the surgical treatment of morbid obesity leads AETMIS to make the following conclusions.
8.1 GENERAL ROLE OF BARIATRIC SURGERYSurgical treatment is recognized today as a more effective therapeutic option than non-surgical treatment for patients who are morbidly obese. Although most of the evidence refers to short-term outcomes, several studies are beginning to demonstrate long-term sustained weight loss. Surgery itself has some potentially serious complications. Although these adverse effects are generally managed appropriately, they require continual assessment.
Even if bariatric surgery continues to be an expensive procedure, the weight loss that results decreases the prevalence of co-morbidities (e.g., cardiovascular disease and diabetes) and their consequences (prescription drug spending), reduces productivity losses caused by sick leave and disability, and improves quality of life. Nevertheless, the favourable cost-effectiveness (or cost-utility) ratio and the effi ciency suggested by the current state of the evidence need to be confi rmed by longer-term well-designed economic studies.
8.2 THE DIFFERENT SURGICAL TECHNIQUESAlthough bariatric surgery relies on a wide range of techniques, current evidence does not yet favour one over the other, owing to the variety of contexts in which they are performed, the diversity of patient characteristics and the lack of well-designed comparative tudies. Moreover, a single procedure may involve several techniques.
50
Adjustable gastric banding (AGB): This technique is generally recognized as being effective in terms of both weight loss and low complication rates. It has the advantage of being reversible and is increasingly replacing VBG.
Biliopancreatic diversion with duodenal switch (BPD-DS): Despite the fact that this technique is used only in a few centres because of its stringent requirements for post-operative patient management and follow-up, its long years in use (over 20 years), the cumulative number of procedures performed to date and its positive weight-reduction outcomes mean that this pro cedure is no longer considered experimental. In addition, some studies suggest that BPD-DS would be appropriate for super-obese patients.
Laparoscopic bariatric-surgery techniques have been developing at a rapid pace and have spread to a growing number of countries.
Compared with conventional surgical techniques, laparoscopic procedures offer many advantages. They reduce hospital stays and decrease, if not eliminate, complications associated with open surgery; however, they do have other types of complications. Recall that surgeons must train in the best conditions to master this approach.
Given that the Roux-en-Y gastric bypass procedure (RYGB) is well advanced and recognized as the gold standard in its open version, laparoscopic RYGB may be considered an innovative procedure. The two approaches yield the same effects at one year of follow-up, and their range of short-term complications differs only slightly. However, longer-term comparative data are still required.
Adjustable gastric banding (LAGB) may also be considered an innovative technology. Not only is this technique reversible, it also appears safe and effi cacious (in terms of
excess weight loss): major complications are rare and its complication and re-operation rates are acceptable. Although an Australian assessment agency recommends that this technology be covered by the public-health system, other agencies, while recognizing its effi cacy, specify that well-designed comparative studies based on longer follow-up periods (more than fi ve years) are required to confi rm its outcomes.
As for the other laparoscopic techniques, it remains necessary to closely monitor patients, to identify their indications more clearly and to pursue more research in this fi eld with solid long-term comparative studies. They must still be classifi ed as experimental, owing to the uncertainty that continues to surround their effi cacy and safety.
8.3 CHALLENGES FOR QUÉBEC’S MEDICAL PRACTICE Different bariatric-surgery techniques are currently being used in Québec by highly experienced surgeons in the fi eld. Yet there is a lack of data on the quality and effec tiveness of these procedures and on the population of treated patients. The supply of services also appears insuffi cient, given the steady growth of waiting lists and waiting times. In such a context:
it is crucial to know and share all the different information about the treated population and the outcomes achieved in bariatric-surgery centres;
it is necessary to effectively measure the evolution and extent of bariatric-surgery needs resulting from the growing prevalence of morbid obesity;
it is advisable to promote the development of practice guidelines on the management of patients with morbid obesity in order to ensure that service offerings are of high quality.
51
Following a meeting of an ad hoc committee on bariatric surgery, the Québec Association of General Surgeons (QAGS) developed a policy on the surgical treatment of morbid obesity.15 This policy contains fi ve key points:
1. Surgeons wishing to perform this surgery must work in conjunction with an interdisciplinary team. The members (QAGS ad hoc committee) are of the opinion that it is crucial that this team include anesthetists and internists (endocrinologists – pneumologists). It is strongly advised that nutritionists and psychologists be part of the team. A bariatric nurse could ultimately join the team.
2. In the current state of affairs, we do not believe that all surgeons will set up a bariatric-surgery practice. It is possible, however, that a surgeon, in a peripheral or even remote hospital centre, would want to perform this type of procedure.
15. Québec Association of General Surgeons (QAGS), written communication, March 21, 2005.
That is why it would be benefi cial for the surgeon to become affi liated to a reference hospital, upon which the surgeon could rely on for advice or in the event of a problem. These centres could be defi ned in relation to bariatric-surgery volume.
3. It is important that residents in training be made aware of and even trained in this surgical practice in the same way as any other surgical procedure.
4. Independent training sessions or even continuing medical-education activities will be encouraged and promoted in a bid to raise awareness among all surgeons, thereby giving surgeons interested in this fi eld the opportunity to perform this type of surgery.
5. It would be to surgeons’ advantage to inform their departmental colleagues of the establishment of a bariatric-surgery practice.
52
9 RECOMMENDATIONS
Have an experienced multidisciplinary team capable of supplying the full range of care and services tied to this type of treatment: surgical team, psychologist, nutritionist, medical specialists (e.g., diabetologists, cardiologists, pneumologists).
Provide closely monitored lifelong follow-up, and cover the physical and psychological dimensions of this treatment, which consequently includes consultations linked to the need for plastic surgery.
3) It is recommended that a Québec registry on morbid obesity and its management be established. This registry will offer key support in implementing a regional follow-up program for operated patients by linking the different health-care structures (hospitals, health centres) and by including specifi c patient education on nutritional approaches appropriate for this type of patient. This data source will make it possible to determine the prevalence and categorization of the different patients, to evaluate the surgical procedures that are currently being performed and to rule on the new bariatric-surgery approaches.
1) It is recommended that the Ministère de la Santé et des Services Sociaux and other decision makers concerned with the problem of morbid obesity identify current and future needs in bariatric surgery, establish an action plan to increase the capacity to provide this treatment, and ensure that patients in the different settings and regions have fair access to these services.
2) It is recommended that, at the organizational level, all hospital bariatric-surgery programs comply with the conditions listed below, which will be subject to a quality-assurance process. Such programs must:
Establish a strict patient-selection process (e.g., patients who have BMIs of 40 kg/m2 or more, or 35 with co-morbidities, who have acceptable operative risks, who are motivated and well informed of the inherent risks of the procedure and of the need for lifelong follow-up) and a system for prioritizing patients on scheduled waiting lists.
Have available facilities and equipment adapted to the specifi c profi le of the patients concerned (e.g., recovery rooms, intensive- care units, beds and furniture, diagnostic investigation tables, operating tables, and adapted surgical instruments).
53
ABBREVIATIONS USED IN THE APPENDICES
AGB Adjustable gastric banding (open procedure)
AHAL Ad hoc alimentary limb
AHS Ad hoc stomach
ALOS Average length of hospital stay
AOT Average operating time
BAROS Bariatric Analysis and Reporting Outcome System
BMI Body mass index
BPD Biliopancreatic diversion (open procedure)
BPD-DS Biliopancreatic diversion with duodenal switch
CI Confi dence interval
CVD Cardiovascular disease
EWL Excess weight loss
F Females
GB Gastric banding
GBP Open gastric bypass
GERD Gastro-esophageal refl ux disease
LAGB Laparoscopic adjustable gastric banding
LBPD Laparoscopic biliopancreatic diversion
LGBP Laparoscopic gastric bypass
LRYGB Laparoscopic Roux-en-Y gastric bypass
LSAGB Laparoscopic Swedish adjustable gastric banding
LVBG Laparoscopic vertical banded gastroplasty
M Males
MO Morbid obesity
N Number of patients
NCC Non-controlled non-randomized comparative study
NS Non-signifi cant difference
O Obese patients
OR Odds ratio
P Prospective non-comparative study
54
PCC Prospective, controlled, non-randomized comparative study
PNCC Prospective, non-controlled, non-randomized comparative study
R Retrospective non-comparative study
RC Randomized comparative study
RNCC Retrospective, non-controlled, non-randomized comparative study
RR Relative risk
Rx Prescribed medication
RYGB Open Roux-en-Y gastric bypass
SAGB Swedish adjustable gastric banding
SO Super-obese patients
SOS Swedish Obese Subjects Intervention Study
T Total
VBG Open vertical banded gastroplasty
VBG-RYGB Vertical banded gastroplasty combined with Roux-en-Y gastric bypass
55
APPENDIX A
STATUS OF HEALTH TECHNOLOGIES: AETMIS CLASSIFICATION
AETMIS has developed the following classifi cation to identify the status of health technologies under review [CETS, 1998]:
Experimental status
The term experimental will be used here to describe a procedure whose effectiveness has yet to be established. Such a procedure should therefore not be used in health-care facilities, except in the context of research projects.
Innovative status
The term innovative will be used to describe a technology which has passed the experimental stage and whose effectiveness has been established. However, because of a lack of experience, certain indications for its use and various aspects of its application are not yet clearly defi ned. To gain further knowledge of such technology, it would be important to gather systematically all the information acquired from its utilization and to communicate it to the medical community in the form of a clinical research report or systematic review or to an appropriate registry. To further these objectives and to prevent its premature widespread use, such technology should be restricted to certain authorized university hospitals which have the necessary resources and knowledge.
Accepted status
The term accepted will describe a well-established technology for which there is a long history of use and a knowledge of, or failing that, universal acceptance of its effectiveness in all its applications.
56
APPE
ND
IX B
OU
TCO
MES
OF
STU
DIE
S O
N V
ERTI
CAL
BAN
DED
GAS
TRO
PLAS
TYTA
BLE
B-1
Lap
aros
copi
c ve
rtic
al b
ande
d ga
stro
plas
ty
AUTH
OR
SN
UM
BER
O
F PA
TIEN
TS
BM
I(K
G/M
2)
FOLL
OW
-UP
(MO
NTH
S)AO
T(M
IN)
ALO
S(D
AYS)
EAR
LY
COM
PLIC
ATIO
NS
MO
RTA
LITY
R
ATE
(%)
LATE
CO
MPL
ICAT
ION
S*
OU
TCO
MES
Alle
et a
l.,19
98(R
)26
143
.328
102
41.
9%0.
4 (1
pul
mon
ary
embo
lism
)5.
7%EW
L: 7
5% a
t 18
mon
ths
Näs
lund
et a
l., 1
999
(P)
6044
.4 ±
1.0
23 ±
1.5
85–2
253
6.7%
1.7
2.2%
R
e-op
erat
ions
: 25
%
(15
case
s)
BM
I at 3
6 m
onth
s (N
S):
LVB
G: 3
7.0
kg/m
2 (2
5.8–
53.3
)V
BG
: 36.
9 kg
/m2
(24.
6–50
.7)
Salv
alet
al.,
199
9 (R
)87
43.8
6–18
Unspecifi ed
Unspecifi ed
12.6
%1.
157.
4%EW
L: 7
5% a
t 18
mon
ths
Topp
ino
et a
l., 1
999
(R)
170
43.9
1–36
95
Unspecifi ed
4.7%
04.
0%R
e-op
erat
ions
: 0.
6%EW
L: 6
1% a
t 36
mon
ths
* La
te c
ompl
icat
ions
incl
ude
absc
esse
s, le
aks, fi s
tula
s, pu
lmon
ary
embo
lism
s, st
enos
es, r
efl u
x, b
leed
ing,
her
nias
, gas
tric-
pouc
h di
lata
tion
and
food
into
lera
nce.
57
TAB
LE B
-2
Com
pari
sons
bet
wee
n op
en a
nd la
paro
scop
ic v
erti
cal b
ande
d ga
stro
plas
ty
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I(K
G/M
2)
AOT
(MIN
)AL
OS
(DAY
S)
MEA
N
FOLL
OW
-U
P(M
ON
THS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Dáv
ila-
Cer
vant
es
et a
l., 2
002
(RC
)
VB
G: 1
4 (c
o-m
orbi
ditie
s pr
esen
t in
8 ca
ses)
43(3
7–50
)10
5(6
5–15
0)4
(3–4
2*)
12
8 pa
tient
s ha
d co
mpl
icat
ions
:
6
wou
nd-r
elat
ed p
robl
ems
2 fi s
tula
s at t
he g
astri
c pa
rtitio
n re
quiri
ng
re-o
pera
tion
N
o de
aths
V
BG
LV
BG
Susp
ensi
on o
f an
alge
sics
(day
s)
3 2
EWL:
at 1
yea
r 55
%
47%
(3
0–88
) (2
2–97
)
N
o si
gnifi
cant
diff
eren
ce in
pai
n in
tens
ity
betw
een
patie
nts i
n th
e 2
grou
ps. H
owev
er,
patie
nts i
n th
e V
BG
gro
up re
quire
d hi
gher
ex
tra d
oses
of n
arco
tics d
urin
g th
e fi r
st p
ost-
oper
ativ
e da
y.
Spiro
met
ric p
aram
eter
s wer
e si
mila
r for
the
2 gr
oups
dur
ing
the fi r
st 3
pos
t-ope
rativ
e da
ys. I
nspi
rato
ry a
nd e
xpira
tory
forc
e va
lues
wer
e ne
verth
eles
s hig
her i
n th
e LV
BG
pat
ient
s 3 d
ays p
ost-o
pera
tivel
y (r
espe
ctiv
ely
60 a
nd 6
5 cm
H2O
for t
he L
VB
G
patie
nts v
s 50
and
54.5
cm
H2O
for t
he G
VC
pa
tient
s).
Pa
tient
s in
the
LVB
G g
roup
s wer
e fa
ster
to
retu
rn to
thei
r nor
mal
act
iviti
es.
G
reat
er sa
tisfa
ctio
n ex
pres
sed
by th
e LV
BG
pa
tient
s.
LVB
G: 1
6(c
o-m
orbi
ditie
s pr
esen
t in
9 ca
ses)
45(3
8–50
)13
0(9
0–24
0)4
(3–9
7*)
4 pa
tient
s ha
d co
mpl
icat
ions
:
2
min
or c
ompl
icat
ions
: pu
lmon
ary
atel
ecta
sis (
1)
and
wou
nd in
fect
ion
(1)
2 fi s
tula
s at t
he g
astri
c pa
rtitio
n re
quiri
ng
re-o
pera
tion
N
o de
aths
* Ex
trem
e va
lues
are
due
to th
e le
ngth
of h
ospi
tal s
tay
of p
atie
nts w
ith c
ompl
icat
ions
such
as fi
stu
las (
one
in e
ach
grou
p).
58
APPE
ND
IX C
O
UTC
OM
ES O
F ST
UD
IES
ON
GAS
TRIC
BYP
ASS
TAB
LE C
-1
Ope
n ga
stri
c by
pass
AUTH
OR
SN
UM
BER
O
F PA
TIEN
TS
BM
I OR
IN
ITIA
L W
EIG
HT
AOT
(MIN
)AL
OS
(DAY
S)
MEA
N
FOLL
OW
-UP
(MO
NTH
S)EA
RLY
CO
MPL
ICAT
ION
SM
OR
TALI
TY
RAT
ELA
TE C
OM
PLIC
ATIO
NS
OU
TCO
MES
Mac
Lean
et
al.,
20
00
(R)
RYG
B: 2
43
- O
bese
(O):
13- P
atie
nts w
ith
mor
bid
obes
ity
(MO
): 13
4- S
uper
-obe
se
(SO
): 96
49 k
g/m
2
Unspecifi ed
Unspecifi ed
66 ±
18
Uns
pecifi e
d0.
41%
Her
nias
: 16%
EWL:
≥ 5
0%-
93%
of p
atie
nts w
ith o
besi
ty
or m
orbi
d ob
esity
- 57
% o
f sup
er-o
bese
Fina
l BM
I:O
and
MO
: 29
± 4
kg/m
2
SO: 3
5 ±
7 kg
/m2
Fobi
et
al.,
1998
(R)
RYG
B: 9
4446
kg/
m2
Unspecifi ed
4 24
2.7%
0.4%
Pulm
onar
y em
bolis
m: 0
.6%
Leak
: 3.1
%H
erni
a: 4
.7%
EWL:
80%
Wes
tling
et a
l.,
2002
(R)
RYG
B: 4
4 pa
tient
s pr
evio
usly
trea
ted
with
SAG
B: 2
6V
BG
: 13
GB
P: 5
35 k
g/m
2
(21–
49)
155
(88–
240)
6 (4
–15)
24In
tra-o
pera
tive
hem
orrh
age:
1In
fect
ed h
emat
oma:
1M
inor
surg
ery:
5
1 de
ath
(lung
can
cer)
11%
- Se
ptic
emia
: 1-
Stric
ture
in th
e ga
stro
-en
tero
anas
to m
osis
: 2-
Bow
el o
bstru
ctio
n: 4
- D
eep
veno
us
thro
mbo
sis:
2-
Re-
oper
atio
ns: 8
(1 e
arly
and
7 la
te)
At 2
yea
rs:
BM
I: 28
kg/
m2 (1
8–42
)Im
prov
ed q
ualit
y of
life
in
98%
of p
atie
nts (
wei
ght
redu
ctio
n an
d ab
senc
e of
the
com
plic
atio
ns a
ssoc
iate
d w
ith th
eir p
revi
ous g
astri
c re
stric
tions
in 7
0% o
f pa
tient
s)C
apel
la
and
Cap
ella
,20
02(R
)
RYG
B-V
BG
: 652
11
2 pa
tient
s (7
2 co
ntac
ted)
fr
om a
n or
igin
al
serie
s of 6
52
140
kg
(94–
288)
50 k
g/m
2 (3
8–86
)
42%
of
patie
nts
wer
e cl
assifi e
d as
su
per-o
bese
: B
MI o
f 60
kg/
m2
(48–
86)
Unspecifi ed
Unspecifi ed
60Ea
rly c
ompl
icat
ions
re
quiri
ng re
-ope
ratio
n:-
Bow
el o
bstru
ctio
n: 3
- O
bstru
ctio
n of
exc
lude
d lim
b: 1
- G
astro
-inte
stin
al b
leed
ing:
1
Early
com
plic
atio
ns le
adin
g to
an
ALO
S >
3 da
ys:
- St
rictu
re g
astro
jeju
nost
omy:
3-
Acu
te g
out a
ttack
: 2-
Wou
nd in
fect
ion:
4-
Pulm
onar
y em
bolis
m: 1
- R
espi
rato
ry: 2
- R
enal
: 1-
Dia
rrhe
a an
d C
lost
ridia
di
ffi ci
le in
fect
ions
: 20
2 de
aths
Late
com
plic
atio
ns
requ
iring
re
-ope
ratio
n:-
Esop
hage
al st
rictu
re: 3
- In
cisi
onal
her
nias
: 26
Late
com
plic
atio
ns n
ot
requ
iring
surg
ery:
- C
hron
ic o
r rec
urre
nt
anas
tom
otic
ulc
er: 3
- G
astro
-inte
stin
al
blee
ding
: 6-
Stom
al st
rictu
re: 6
At 5
yea
rs:
Wei
ght l
oss:
58
kg (1
4–14
3)B
MI:
29 k
g/m
2 (2
0–43
)EW
L: 7
7% (3
2–10
8)
(EW
L >
50%
in 9
3% o
f pa
tient
s)
Supe
r-obe
se g
roup
(3
0 pa
tient
s):
BM
I: 32
± 6
kg/
m2
EWL:
74
± 15
%
(97%
of p
atie
nts l
ost m
ore
than
50%
of t
heir
exce
ss
wei
ght)
59
TAB
LE C
-2
Lap
aros
copi
c ga
stri
c by
pass
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
INIT
IAL
BM
I (K
G/M
2)
AOT
(MIN
)AL
OS
(DAY
S)CO
MPL
ICAT
ION
S
Dre
sel
et a
l.,20
04(P
)
120
patie
nts u
nder
wen
t LRY
GB
be
twee
n Ja
nuar
y 20
02 a
nd A
ugus
t 200
2
60 o
bese
pat
ient
s (56
wom
en +
4 m
en)
with
BM
Is le
ss th
an 5
0 (O
gro
up)
rece
ived
a 1
00-c
m R
oux
limb
60 su
per-o
bese
pat
ient
s (50
wom
en
+ 10
men
) with
BM
Is g
reat
er th
an 5
0 (S
O g
roup
) rec
eive
d a
150-
cm R
oux
limb
Trea
tmen
t was
the
sam
e in
bot
h gr
oups
, ex
cept
for t
he le
ngth
of t
he R
oux
limbs
.
Age
(NS)
: -
O g
roup
: 41
(19–
64)
- SO
gro
up: 4
0 (1
9–60
)
Co-
mor
bidi
ties (
NS)
:-
O g
roup
: 5-
SO g
roup
: 6
O: 4
4.6
(39–
49)
SO: 5
8.4
(50–
100)
Non
-sig
nifi c
ant
diffe
renc
e
O: 1
28(7
5–22
5)SO
: 144
(75–
240)
O: 2
SO: 2
Late
ana
stom
otic
stric
ture
s, tre
ated
with
end
osco
pic
dila
tion
Re-
oper
atio
ns
Ana
stom
otic
ble
edin
g
Con
vers
ions
Dea
ths
Oth
er c
ompl
icat
ions
Ear
ly c
ompl
icat
ions
(< 7
day
s)- P
neum
onia
- Pul
mon
ary
edem
a - W
ound
infe
ctio
n- B
owel
obs
truct
ion
Late
com
plic
atio
ns- P
ulm
onar
y em
bolis
m- G
astro
gast
ric fi
stul
a- T
roca
r her
nia
O g
roup
3 0 3 0 0 1 – – – – – –
SO g
roup
4 2(1
her
nia
and
1 ga
stric
fi st
ula)
2 0 0 – 1 1 1 1 1 1
60
TAB
LE C
-2
Lap
aros
copi
c ga
stri
c by
pass
(con
’d)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I O
R
INIT
IAL
WEI
GH
T
FOLL
OW
-UP
(MO
NTH
S)AO
T(M
IN)
ALO
S(D
AYS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Gag
ner
et a
l.,
1999
(R)
5255
kg/
m2
3627
74
Ear
ly c
ompl
icat
ions
: 15.
0%La
te c
ompl
icat
ions
: 3.8
%D
eath
s: 0
BM
I: 34
kg/
m2 at
18
mon
ths
Hig
a et
al.,
2000
(R)
400
46 k
g/m
222
Uns
pecifi e
d1.
6
Ear
ly c
ompl
icat
ions
: 15.
0%La
te c
ompl
icat
ions
: 15.
0%R
e-op
erat
ions
: 3%
Dea
ths:
0
EWL:
69%
at 1
2 m
onth
s
Scha
uer
et a
l.,
2000
(P)
275
48 k
g/m
230
247
2.6
Ear
ly c
ompl
icat
ions
: 3.3
%La
te c
ompl
icat
ions
: 27.
0%R
e-op
erat
ions
: 1%
D
eath
s: 1
EWL:
77%
at 3
0 m
onth
s
Witt
grov
e an
d C
lark
,20
00(R
)
500
Uns
pecifi e
d60
120
2.6
Ear
ly c
ompl
icat
ions
: 10.
4%La
te c
ompl
icat
ions
: 2.2
%D
eath
s: 0
EWL:
73%
at 5
4 m
onth
s
Frez
zaet
al.,
20
02(R
)
238
(LRY
GB
)
(152
vol
unte
ers w
ith
gast
ro-e
soph
agea
l refl
ux
dise
ase,
of w
hom
one
third
ha
d al
read
y un
derg
one
surg
ery)
289.
1 lb
s(2
21–4
57)
48 k
g/m
2
(39–
67.9
)
12(6
–36)
Uns
pecifi e
dU
nspe
cifi e
dC
onve
rsio
ns: 0
.7%
EWL:
68.
8% a
t 12
mon
ths
Res
olut
ion
of sy
mpt
oms i
n 80
% o
f ca
ses
Impr
oved
qua
lity
of li
fe in
90%
of
patie
nts
DeM
aria
et a
l.,
2002
b(R
)
281
(LG
BP)
pat
ient
s su
rgic
ally
trea
ted
betw
een
Mar
ch 1
998
and
Oct
ober
200
1
(25
with
LG
BP
+ ha
nd-
assi
sted
pro
cedu
re)
291
± 46
.6 lb
s(1
71–4
46)
48.1
± 6
.5 k
g/m
2
(40.
3–71
)
12(6
9 pa
tient
s)U
nspe
cifi e
d4
± 9
(273
pat
ient
s)
15.8
± 2
5(8
pat
ient
s)
Com
plic
atio
ns:
Stom
al st
enos
is: 6
.6%
Gas
trodu
oden
al u
lcer
s: 5
.1%
Ana
stom
otic
leak
: 5.1
%O
ther
: 7.1
% (h
erni
as;
infe
ctio
ns; p
ulm
onar
y em
bolis
ms)
Con
vers
ions
/re-
oper
atio
ns:
2.8%
Dea
ths:
0
At 1
yea
r:EW
L: 7
0 ±
15%
; BM
I: 30
.5 ±
5.1
kg/
m2
Wei
ght:
180
± 30
lbs
Dia
bete
s: 9
3% o
f pat
ient
s no
long
er
need
ed m
edic
atio
n.G
astro
-eso
phag
eal r
efl u
x di
seas
e w
as
reso
lved
in 9
5% o
f cas
es.
Hyp
erte
nsio
n: re
turn
ed to
nor
mal
in 5
2%
of c
ases
.B
enefi
cia
l im
pact
on
obes
ity-r
elat
ed
orth
oped
ic p
robl
ems.
61
TAB
LE C
-2
Lap
aros
copi
c ga
stri
c by
pass
(con
t’d)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I(K
G/M
2)
AOT
(MIN
)AL
OS
(DAY
S)CO
MPL
ICAT
ION
SO
UTC
OM
ES
Stoo
pen-
Mar
gain
et a
l.,
2004
(P)
100
cons
ecut
ive
patie
nts
who
und
erw
ent L
RYG
B
betw
een
Febr
uary
200
0 an
d Se
ptem
ber 2
002
63 fe
mal
es a
nd 3
7 m
ales
Mea
n ag
e: 3
1 ±
5 ye
ars
Pre-
oper
ativ
e co
-mor
bidi
ties:
Hyp
erte
nsio
n 46
Dia
bete
s 24
Slee
p ap
nea
24H
yper
lipid
emia
22
Ost
eoar
thro
path
y 13
Veno
us in
suffi
cien
cy
12H
yper
uric
emia
9
Pneu
mop
athy
9
Dep
ress
ion
5H
ypot
hyro
idis
m
4G
astro
-eso
phag
eal
refl u
x di
seas
e 2
Pulm
onar
y hy
perte
nsio
n 2
Isch
emic
car
diop
athy
1
50 ±
9
33 su
per-o
bese
pa
tient
s with
B
MIs
> 5
0
228
± 42
6Le
ak: 1
as a
resu
lt of
st
aple
-line
dis
rupt
ion
Ana
stom
otic
sten
osis
: 4B
owel
obs
truct
ion:
4H
emat
oma:
1W
ound
infe
ctio
n: 4
at t
he
troca
r site
R
habd
omyo
lysi
s: 2
2 Th
rom
bo-e
mbo
lism
: 1
Con
vers
ions
: 2
(1 c
ause
d by
the
pres
ence
of n
umer
ous
adhe
sion
s and
1 b
ecau
se
of a
n ex
tens
ivel
y en
larg
ed le
ft liv
er lo
be)
Dea
ths:
2 p
ost-o
pera
tive
deat
hs1
thro
mbo
-em
bolis
m1
acut
e re
spira
tory
di
stre
ss
BM
I (kg
/m2 )
6 m
onth
s: 4
7 ±
2 (n
= 8
2)12
mon
ths:
62
± 4
(n =
70)
18 m
onth
s: 6
6 ±
5 (n
= 6
3)24
mon
ths:
67
± 8
(n =
35)
Impa
ct o
n co
-mor
bidi
ties:
B
efor
e A
fter
su
rger
y (2
0 ±
11 m
os)
BM
I (kg
/m2 )
50 ±
9
36.4
± 6
.4
Dia
bete
s- T
reat
ed w
ith
hypo
glyc
emic
dru
gs
20
11- T
reat
ed w
ith d
iet
4
3
Hyp
erte
nsio
n- 1
dru
g 42
23
- 2 d
rugs
4
1
Hyp
erlip
idem
ia- H
yper
trigl
ycer
idem
ia
8
4- H
yper
chol
este
role
mia
14
12
Bro
dy,
2004
(R)
195
patie
nts t
reat
ed w
ith L
RYG
B(u
ntil
Dec
embe
r 31,
200
3)
Follo
w-u
p of
18
mon
ths
A to
tal o
f 36
of th
ese
patie
nts
dire
ctly
und
erw
ent o
pen
surg
ery
beca
use
of v
entil
ator
y pr
oble
ms
upon
ane
sthe
sia
or a
his
tory
of
prio
r sur
gerie
s.
51.5
(mea
n)LR
YG
B:
2.96
RYG
B:
5.11
Con
vers
ions
:4.
08
Leak
age:
2(1
ana
stom
otic
leak
tre
ated
by
drai
nage
, and
1
gast
ric-p
ouch
leak
)W
ound
infe
ctio
n: 9
(sur
gica
lly tr
eate
d)Ev
isce
ratio
n: 1
Sple
nect
omy:
3
(1 d
ue to
a ru
ptur
e an
d 2
to re
tract
ion
inju
ries)
Bow
el o
bstru
ctio
n: 5
re
quiri
ng su
rger
yC
onve
rsio
ns: 3
6 (1
8.5%
)R
e-op
erat
ions
: 17
At 1
8 m
onth
s, th
e pa
tient
s had
lost
35%
of t
heir
initi
al
wei
ght.
The
mea
n B
MI f
ell f
rom
51.
5 to
32.
0 kg
/m2 .
62
TAB
LE C
-3
Com
pari
sons
bet
wee
n op
en a
nd la
paro
scop
ic g
astr
ic b
ypas
s
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
IN
ITIA
L W
EIG
HT
AOT
(MIN
)AL
OS
(DAY
S)FO
LLO
W-U
P(M
ON
THS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Ngu
yen
et a
l.,
2001
(RC
)
RYG
B: 7
648
.4 ±
5.4
kg/
m2
195
± 41
49.
6–6.
3(1
–23)
Ear
ly c
ompl
icat
ions
: 9.2
% (7
cas
es)
- Ana
stom
otic
leak
: 1- B
owel
obs
truct
ion:
1- R
espi
rato
ry fa
ilure
: 1- P
ulm
onar
y em
bolis
m: 1
- Inf
ectio
n: 2
- Ret
aine
d la
paro
tom
y sp
onge
: 1
Late
com
plic
atio
ns: 1
5.8%
(12
case
s)- A
nast
omot
ic st
rictu
re: 2
- Ven
tral h
erni
a: 6
- Ane
mia
: 2- O
ther
gas
tro-in
test
inal
co
mpl
icat
ion:
2
Re-
oper
atio
ns: 6
.6%
of c
ases
No
deat
hs
EW
L
RY
GB
LR
YG
B
p th
resh
old
3 m
onth
s 32
± 1
0%
37 ±
10%
0.
016
mon
ths
45 ±
12%
54
± 1
4%
0.01
1 ye
ar
62 ±
14%
68
± 1
5%
0.07
LRY
GB
: 79
47.6
± 4
.7 k
g/m
222
5 ±
403
9.6–
6.5
(1–2
3)E
arly
com
plic
atio
ns: 7
.6%
(6 c
ases
)- A
nast
omot
ic le
ak: 1
- Bow
el o
bstru
ctio
n: 3
- Per
fora
tion:
1- G
astro
-inte
stin
al b
leed
ing:
1
Late
com
plic
atio
ns: 1
8.9%
(15
case
s)- A
nast
omot
ic st
rictu
re: 9
- Cho
lelit
hias
is: 3
- Pro
tein
-cal
orie
mal
nutri
tion:
1- O
ther
gas
tro-in
test
inal
co
mpl
icat
ion:
2
Con
vers
ions
: 2.5
% (2
pat
ient
s)
Re-
oper
atio
ns: 7
.6%
of t
he c
ases
No
deat
hs
63
TAB
LE C
-3
Com
pari
sons
bet
wee
n op
en a
nd la
paro
scop
ic g
astr
ic b
ypas
s (co
nt’d
)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
IN
ITIA
L W
EIG
HT
AOT
(MIN
)AL
OS
(DAY
S)
MEA
N
FOLL
OW
-UP
(MO
NTH
S)
COM
PLIC
ATIO
NS
OU
TCO
MES
Wes
tling
an
d G
usta
vsso
n,20
01(R
C)
LRY
GB
: 30
Cho
lecy
stec
tom
y: 3
Join
t and
bac
k pa
in: 1
0H
yper
tens
ion:
7A
sthm
a: 5
Rhe
umat
oid
arth
ritis
: 0D
iabe
tes:
0
41 k
g/m
2Ex
clud
ing
conv
ersi
ons*
:23
5 (1
65–3
90)
(n =
23)
Incl
udin
g co
nver
sion
s:24
5 (1
35–3
90)
(n =
30)
With
out
conv
ersi
on*:
4 ±
0.8
(n =
222
)
With
co
nver
sion
:4.
5 ±
1.2
(n =
292
)
12C
onve
rsio
n to
ope
n su
rger
y: 2
3%
(7 c
ases
)
Intra
-ope
rativ
e bl
eedi
ng:
- 20
0 m
l (50
–400
): n
= 23
(exc
ludi
ng
conv
ersi
ons)
- 25
0 m
l (50
–150
0): n
= 3
0 (in
clud
ing
conv
ersi
ons)
Early
and
late
com
plic
atio
ns (i
nclu
ding
co
nver
sion
s):
- R
e-op
erat
ions
for b
owel
ob
stru
ctio
n: 6
- Pu
lmon
ary
embo
lism
: 1-
Gas
tric
ulce
r: 1
- G
astro
jeju
nal s
trict
ure:
1-
Oth
er re
-adm
issi
ons:
2 fo
r epi
gast
ric
pain
and
1 fo
r pne
umon
ia
BM
I at
1 y
ear:
LRY
GB
: 27
± 4
kg/m
2 (dec
reas
e in
BM
I: 14
± 3
)
RYG
B: 3
0.6
± 4
kg/m
2 (dec
reas
e in
BM
I: 13
± 3
)N
on-s
ignifi c
ant d
iffer
ence
AO
T: 2
tim
es lo
nger
than
in th
e LR
YG
B g
roup
Pain
: lap
aros
copy
pat
ient
s (ex
clud
ing
conv
ersi
ons)
seem
ed to
hav
e le
ss p
ain
than
the
open
-sur
gery
pat
ient
s (ev
alua
ted
acco
rdin
g to
m
orph
ine
dose
s adm
inis
tere
d).
Mor
phin
e do
se:
- LR
YG
B g
roup
: 69
± 46
.4 m
g fo
r the
22
2 pa
tient
s who
did
not
und
ergo
con
vers
ion
(98
± 71
.5 fo
r the
gro
up c
ombi
ned
[n =
292
], in
clud
ing
conv
ersi
ons)
- RY
GB
gro
up (n
= 2
1): 1
40 ±
90
mg
Sign
ifi ca
nt d
iffer
ence
(p <
0.0
05)
ALO
S: sh
orte
r with
LRY
GB
The
diffe
renc
e is
sign
ifi ca
nt, a
nd e
spec
ially
pr
onou
nced
in th
e gr
oup
incl
udin
g co
nver
sion
s (p
< 0
.025
).
Rec
over
y pe
riod:
shor
ter w
ith L
RYG
B
(p <
0.0
25)
Ther
e is
no
corr
elat
ion
betw
een
the
pre-
oper
ativ
e B
MI,
the
mor
phin
e do
se a
dmin
iste
red
and
the A
LOS.
Deg
ree
of p
atie
nt sa
tisfa
ctio
n: It
was
hig
h an
d id
entic
al fo
r pat
ient
s in
both
gro
ups (
92%
of t
he
patie
nts r
epor
ted
that
they
wer
e ve
ry sa
tisfi e
d an
d 8%
satisfi e
d; 5
% o
f the
pat
ient
s que
stio
ned
repo
rted
expe
rienc
ing
sym
ptom
s ass
ocia
ted
with
the
dum
ping
synd
rom
e bu
t wer
e no
t di
stur
bed
by th
em.
RYG
B: 2
1
Cho
lecy
stec
tom
y: 6
Join
t and
bac
k pa
in: 8
Hyp
erte
nsio
n: 0
Ast
hma:
5R
heum
atoi
d ar
thrit
is: 1
Dia
bete
s: 1
43.9
kg/
m2
100
(70–
150)
6 ±
3.8
Intra
-ope
rativ
e bl
eedi
ng:
- 30
0 m
l (20
0–50
0)
Early
and
late
com
plic
atio
ns:
- R
e-op
erat
ion
for l
eaks
: 1-
Gas
tric
ulce
r: 2
- M
inor
wou
nd in
fect
ion:
3-
Her
nia:
1-
Oth
er re
-adm
issi
ons:
1 fo
r un
expl
aine
d fe
ver
* B
leed
ing
was
the
maj
or c
ause
of c
onve
rsio
n (la
paro
scop
y to
ope
n su
rger
y) fo
r 4 p
atie
nts.
Ble
edin
g ca
me
from
a tr
ocar
site
, the
shor
t gas
tric
vess
els,
a st
aple
line
in th
e R
oux
limb
and
an
intra
mur
al g
astri
c ar
tery
. Oth
er re
ason
s for
con
vers
ions
wer
e la
ck o
f exp
osur
e, d
iffi c
ultie
s in
defi n
ing
the
mes
ocol
ic tu
nnel
, unf
avou
rabl
e tro
car p
ositi
onin
g an
d th
e ac
cide
ntal
dis
cove
ry o
f a
hiat
al h
erni
a. O
ne la
paro
scop
ic p
roce
dure
was
con
verte
d to
ope
n su
rger
y w
hen
it w
as d
isco
vere
d th
at th
e je
junu
m h
ad b
een
trans
ecte
d m
ore
than
one
met
re fr
om th
e lig
amen
t of T
reitz
.
64
TAB
LE C
-3
Com
pari
sons
bet
wee
n op
en a
nd la
paro
scop
ic g
astr
ic b
ypas
s (c
ont’d
)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
IN
ITIA
L W
EIG
HT
AOT
(MIN
)AL
OS
(DAY
S)
MEA
N
FOLL
OW
-U
P (M
ON
THS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Cou
rcou
las
et a
l.,
2003
a(P
CC
)
80 p
airs
mat
ched
by
age
, sex
, pr
e-op
erat
ive
BM
I and
num
ber
of c
o-m
orbi
ditie
sG
BP:
80
LGB
P: 8
0
GB
P: 4
6 kg
/m2
LGB
P: 4
4 kg
/m2
(mea
n)
Unspecifi ed
3.6
12
GB
LG
BB
leed
ing
1 1
Pulm
onar
y em
bolis
m
1 (d
eath
) 1
Infe
ctio
n 9
3Su
ture
dis
rupt
ion
6 –
Min
or b
owel
obs
truct
ion
– 6
Pneu
mon
ia
– 1
Ana
stom
otic
stric
ture
–
5Ile
us
– 1
Leak
–
2
St
atis
tical
G
B
LGB
te
stM
ean
EWL
6 m
onth
s:
45.0
%
52.6
%
p <
0.05
9 m
onth
s:
51.4
%
68.4
%
p <
0.05
1 ye
ar:
64.9
%
69.2
%
NS
BM
I(m
ean
in
kg/m
2 )1
year
: 31
.3
29.6
–
Ret
urn
to
norm
al
activ
ities
(wee
ks):
9.
8 ±
16
7.3
± 16
–
Qua
lity
of li
fe (S
F-36
): no
sign
ifi ca
nt d
iffer
ence
65
TAB
LE C
-3
Com
pari
sons
bet
wee
n op
en a
nd la
paro
scop
ic g
astr
ic b
ypas
s (c
ont’d
)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
IN
ITIA
L W
EIG
HT
AOT
(MIN
)AL
OS
(DAY
S)
MEA
N
FOLL
OW
-U
P (M
ON
THS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Lujá
n et
al.,
20
04(R
C)
LGB
P: 5
3(1
0 m
ales
and
43
fem
ales
)A
ge: 3
7 ye
ars
(18–
64)
GB
P: 5
1(1
3 m
ales
and
38
fem
ales
)A
ge: 3
8 ye
ars
(20–
63)
130.
7 kg
(92–
208)
48.5
3 kg
/m2
(36–
78)
137.
57 k
g(9
6–21
4)52
.20
kg/m
2
(37–
80)
186.
4(1
25–2
90)
201.
712
9–31
0)(p
< 0
.05)
5.2
(1–1
3)
7.9
(2–2
8)(p
< 0
.05)
23
LGB
G
BE
arly
(<
30
days
) 22
.6%
29
.4%
- Int
estin
al su
bocc
lusi
on
3 –
- Asy
mpt
omat
ic le
ak
2 –
- Int
ra-a
bdom
inal
ble
edin
g 2
–
- Upp
er g
astro
-inte
stin
al h
emor
rhag
e 2
3- I
ntra
-abd
omin
al h
emor
rhag
e 1
–- T
hrom
boph
lebi
tis
1 –
- Ste
nosi
s of t
he g
astro
-ent
ero-
anas
tom
osis
1
–- S
ubph
reni
c ab
sces
s –
4- W
ound
infe
ctio
n –
4- P
ulm
onar
y in
fect
ion
– 3
- Evi
scer
atio
n (d
eath
) –
1
Late
(>
10
days
) 11
%
24%
- Par
tial b
owel
obs
truct
ion
3*
1†
- Pan
crea
titis
(cho
lecy
stec
tom
y)
2 –
- Sud
den
deat
h 1
–- A
bdom
inal
-wal
l her
nia
– 10
- Sub
phre
nic
absc
ess
1 –
Afte
r a (m
ean)
follo
w-u
p of
23
mon
ths,
the
auth
ors d
id n
ot
note
a si
gnifi
cant
diff
eren
ce in
the
chan
ges i
n B
MI b
etw
een
the
two
grou
ps o
f pat
ient
s.
* O
ne p
atie
nt re
ceiv
ed m
edic
al tr
eatm
ent a
nd th
e tw
o ot
hers
surg
ical
trea
tmen
t (1
deat
h).
† A
seco
nd p
roce
dure
was
nec
essa
ry.
66
APPE
ND
IX D
O
UTC
OM
ES O
F ST
UD
IES
ON
BIL
IOPA
NCR
EATI
C D
IVER
SIO
N
TAB
LE D
-1
Ope
n bi
liopa
ncre
atic
div
ersi
on
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I O
R IN
ITIA
L W
EIG
HT
AOT
(MIN
)AL
OS
(DAY
S)FO
LLO
W-U
PCO
MPL
ICAT
ION
SO
UTC
OM
ES
Hes
s and
H
ess,
1998
(R)
BPD
-DS:
440
50 k
g/m
2
199–158
Unspecifi ed
9–108 months
Peri-
oper
ativ
e m
edic
al c
ompl
icat
ions
: 8- D
eep
vein
thro
mbo
phle
bitis
: 0.7
5% (3
) - P
ulm
onar
y em
bolis
m: 0
.5%
(2)
- Pne
umon
ia: 0
.5%
(2)
- Acu
te re
spira
tory
dis
tress
synd
rom
e: 0
.25%
(1)
Peri-
oper
ativ
e su
rgic
al c
ompl
icat
ions
: 23
- Gas
tric
leak
and
fi st
ula
(per
fora
tion
or
stap
le-li
ne d
isru
ptio
n): 2
% (9
)- G
astro
duod
enal
leak
: 1.5
% (6
)- S
plen
ecto
my
(inci
dent
al):
0.9%
(4)
- Dis
tal R
oux-
en-Y
leak
: 0.2
5% (1
)- P
ost-o
pera
tive
blee
ding
requ
iring
su
rger
y: 0
.5%
(2)
- Abs
cess
(not
rela
ted
to le
aks)
: 0.2
5% (1
)La
te c
ompl
icat
ions
: 11
- Duo
dena
l sto
mal
obs
truct
ion:
0.7
5% (3
)- S
mal
l bow
el o
bstru
ctio
n: 2
% (8
)
Post
-ope
rativ
e at
elec
tasi
s in
20%
of p
atie
nts
17 su
rgic
al re
visi
ons
Dea
ths:
1.2
% (2
ear
ly a
nd 3
late
)
EWL:
361
pat
ient
s (ba
sed
on la
st
wei
ght m
easu
rem
ent)
- Exc
elle
nt (a
t lea
st 8
0%):
79.8
%- G
ood
(at l
east
60%
): 13
.0%
- Fai
r (at
leas
t 40%
): 6.
9%- P
oor (
at le
ast 2
0%):
0.3%
- Fai
lure
(< 2
%):
0.0%
Year
NSa
tisfa
ctor
y(a
t lea
st 4
0%)
Goo
d–ex
celle
nt(a
t lea
st 6
0%)
1
345
344
320
2
264
263
240
3
187
186
167
4
132
132
117
5
92
92
80 6
51
51
45
7
29
29
25 8
11
11
9
Scop
inar
o et
al.,
20
00(R
)
BPD
-DG
: 231
6 (d
iffer
ent v
aria
nts)
47 k
g/m
2
(29–
87)
128
kg
(73–
236)
Unspecifi ed
Unspecifi ed
Maximum of 15 years
Ope
rativ
e m
orta
lity
rate
: < 0
.5%
Late
com
plic
atio
ns: >
5%
EWL:
6 ye
ars:
75
± 15
% (n
= 1
054)
10 y
ears
: 76
± 15
% (n
= 3
81)
Biro
n et
al.,
2004
(R)
BPD
-DG
: 997
ou
t of 1
271
cons
ecut
ive
patie
nts
(81.
4% b
y sl
eeve
ga
stre
ctom
y st
artin
g in
199
0)(F
ebru
ary
1984
to
Dec
embe
r 200
2)
48.4
±
9.4
kg/m
2 13
0 ±
30 k
g Unspecifi ed
Unspecifi ed
7.9 ± 4.2 years
Ope
rativ
e m
orta
lity
rate
: 0%
Late
com
plic
atio
ns: >
5%
BM
I (m
ean)
:Pr
e-op
erat
ive:
48.
4 ±
9.4
kg/m
2
Post
-ope
rativ
e: 3
1.3
± 6.
5 kg
/m2
The
oper
atio
n w
as c
onsi
dere
d a
failu
re
for p
atie
nts w
ho h
ad B
MIs
≥ 3
5 af
ter
the
oper
atio
n (1
0%: B
MI ≥
40;
and
26
%: 3
5 ≤
BM
I < 4
0).
67
TAB
LE D
-1
Ope
n bi
liopa
ncre
atic
div
ersi
on (c
ont’d
)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
IN
ITIA
L W
EIG
HT
FOLL
OW
-U
P(M
ON
THS)
OU
TCO
MES
EWL
(%)
IMPA
CT O
N C
O-M
OR
BID
ITIE
S AN
D C
OM
PLIC
ATIO
NS
Mar
inar
i et
al.,
2004
(R)
- 858
of 1
800
patie
nts
who
und
erw
ent B
PD
betw
een
June
198
4 an
d D
ecem
ber 1
998
com
plet
ed th
e en
tire
ques
tionn
aire
- 709
of 8
58 p
atie
nts
had
co-m
orbi
ditie
s
- 615
fem
ales
- Age
: 38
± 11
yea
rs
- 596
pat
ient
s ope
rate
d be
twee
n Ju
ne 1
984
and
Sept
embe
r 199
2: B
PD
with
ad
hoc
stom
ach
(ada
pted
to p
atie
nt
char
acte
ristic
s), w
ith a
50
-cm
com
mon
lim
b an
d a
200-
cm st
anda
rd
alim
enta
ry li
mb
(BPD
-AH
S)
- 262
pat
ient
s ope
rate
d af
ter S
epte
mbe
r 199
2:ad
hoc
alim
enta
ry li
mb
(BPD
-AH
S-A
HA
L)
BM
I: 47
± 7
kg/
m2
Wei
ght:
128
± 26
kg
Exce
ss w
eigh
t: 11
7 ±
38%
24–1
802
year
s: 6
7 ±
18 (n
= 8
00)
4 ye
ars:
67
± 18
(n =
738
)
6 ye
ars:
68
± 18
(n =
659
)
8 ye
ars:
69
± 18
(n =
532
)
10 y
ears
: 68
± 18
(n =
334
)
12 y
ears
: 66
± 18
(n =
131
)
14 y
ears
: 69
± 15
(n =
60)
BPD
-AH
S: 7
0.5
± 23
BPD
-AH
S-A
HA
L:
64.7
± 1
7 (N
S)
Co-
mor
bidi
ties
Dis
appe
ared
Im
prov
ed
Unc
hang
edH
yper
tens
ion
(52%
) 87
9
18D
yslip
idem
ia (4
6%)
100
– –
Type
2 d
iabe
tes (
14%
) 10
0 –
–O
besi
ty-h
ypov
entil
atio
nsy
ndro
me
(8%
) 10
0 –
–Sl
eep
apne
a (4
%)
100
– –
BA
RO
S B
PD-A
HS
B
PD-A
HS-
AH
AL
Failu
re
28 (3
.3%
)Fa
ir 97
(11.
3%)
Goo
d 19
5 (2
2.7%
)Ve
ry g
ood
341
(39.
7%)
83%
92
%Ex
celle
nt
197
(23%
)
Com
plic
atio
ns:
- Duo
dena
l lea
k: 1
- Com
plic
ated
pep
tic u
lcer
dis
ease
: 12
- Pul
mon
ary
embo
lism
: 1- P
rote
in d
efi c
ienc
y: 3
2 (r
ehos
pita
lized
)- H
erni
a re
quiri
ng h
ospi
tal s
tay
> 7
days
: 15
- Dep
ress
ion:
1- S
urge
ry re
visi
ons:
54
(6.3
%)
Re-
oper
atio
ns:
- BPD
-AH
S: 8
.6%
- BPD
-AH
S-A
HA
L: 1
.1%
68
TAB
LE D
-2
Lap
aros
copi
c bi
liopa
ncre
atic
div
ersi
on
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
INIT
IAL
WEI
GH
T
FOLL
OW
-UP
(MO
NTH
S)AO
T(M
IN)
ALO
S(D
AYS)
MAJ
OR
CO
MPL
ICAT
ION
SO
UTC
OM
ES
Ren
et a
l.,20
00(P
)
40(B
PD-D
S)60
kg/
m2
(42–
85)
6(1
–12)
210
± 9
(110
–360
)4
17.5
%
Ble
edin
g: 4
Leak
: 1
Veno
us th
rom
bosi
s: 1
Subp
hren
ic a
bsce
ss: 1
Dea
ths:
1
EWL:
6 m
onth
s: 4
6 ±
2%
9 m
onth
s: 5
8 ±
3%
Paiv
a et
al.,
2001
(R)
10 (l
apar
osco
pic
Scop
inar
o B
PD)
40–5
5 kg
/m2
119.
3 kg
(100
–150
)
Uns
pecifi e
d4.
3 ho
urs
(3.5
–7)
5.2
(4–7
)C
onve
rsio
ns o
r re-
oper
atio
ns: 0
Dea
ths:
0W
eigh
t los
s: 1
.6–3
3 kg
/day
(m
ean
2.3
kg)
Bal
tasa
r et a
l.,20
02(R
)
16(B
PD-D
S)>
40 k
g/m
2U
nspe
cifi e
d19
5–27
05–
8(1
3 pa
tient
s)In
tra-o
pera
tive
com
plic
atio
ns: 3
Inte
rnal
ble
edin
g: 1
Parti
al st
enos
is o
f the
gas
tric
tube
: 1C
onve
rsio
ns o
r re-
oper
atio
ns: 0
Dea
ths:
0
No
BM
I or E
WL
outc
omes
Lapa
rosc
opic
BPD
with
du
oden
al sw
itch
is c
ompl
ex
but f
easi
ble.
Scop
inar
o et
al.,
2002
(R)
26(B
PD-D
G)
43 k
g/m
2 (m
ean)
6–12
U
nspe
cifi e
dU
nspe
cifi e
dU
nspe
cifi e
dTh
e au
thor
s con
clud
e th
at
outc
omes
are
sim
ilar t
o th
ose
achi
eved
with
lapa
rosc
opic
B
PD.
69
TAB
LE D
-3
Com
pari
sons
bet
wee
n op
en a
nd la
paro
scop
ic b
iliop
ancr
eati
c di
vers
ion
AUTH
OR
SN
UM
BER
O
F PA
TIEN
TS
BM
I OR
IN
ITIA
L W
EIG
HT
(MED
IAN
)
AOT
(MIN
)AL
OS
(DAY
S)
MEA
N
FOLL
OW
-UP
(MO
NTH
S)
COM
PLIC
ATIO
NS
OU
TCO
MES
Kim
et a
l.,20
03(R
NC
C)
BPD
-DS:
28
68.8
± 1
0.1
kg/m
2
196
± 29
.1 k
g
259
± 60
(180
–400
)5
± 47
10.1
(4–1
9)
BPD
LB
PDM
ajor
com
plic
atio
ns (N
S):
- Sub
phre
nic
absc
ess
0 1
- Ana
stom
otic
leak
0
1- R
espi
rato
ry fa
ilure
0
1- W
ound
dis
rupt
ion
1 –
- Wou
nd in
fect
ion
2 –
Min
or c
ompl
icat
ions
(NS)
:- L
ow-e
xtre
mity
ede
ma
0 1
- Wou
nd in
fect
ion
0 1
- Inc
isio
nal h
erni
a 0
1- U
rinar
y-tra
ct in
fect
ion
1 0
Cau
se o
f dea
th (N
S):
- Nec
rotiz
ing
panc
reat
itis
1 –
- Res
pira
tory
failu
re
– 1
- Ana
stom
otic
leak
–
1
EW
L (m
edia
n)
BPD
LB
PD3
mon
ths
32.2
± 1
5.6%
35
.6 ±
15.
6% (N
S)
(9.9
–60.
8)
(15.
4–77
.2)
6 m
onth
s 44
.3 ±
5.7
%
56.
9 ±
20.4
% (N
S)
(42.
6–56
.2)
(32.
6–91
.7)
9 m
onth
s 48
.7 ±
4.1
%
68.
1 ±
26.5
% (N
S)
(58.
5–68
.7)
(42.
0–10
2.6)
1 ye
ar
56.8
± 2
6.3%
76
.7 ±
19.
7% (N
S)
(32.
2–94
.8)
(68.
8–11
2.9)
BM
I (m
edia
n in
kg/
m2 )
1 ye
ar
48.2
± 6
.3
37.3
± 5
.6
(40.
7–54
) (3
1.2–
43.4
)
LBPD
-DS:
26
66 ±
7.5
kg/
m2
189
± 31
.7 k
g21
0 ±
68(1
45–4
03)
4 ±
4118
.5 (6
–13)
70
APPE
ND
IX E
O
UTC
OM
ES O
F ST
UD
IES
ON
GAS
TRIC
BAN
DIN
GTA
BLE
E-1
Swed
ish
adju
stab
le g
astr
ic b
andi
ng
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
INIT
IAL
WEI
GH
T AN
D B
MI
AOT
(MIN
)AL
OS
(DAY
S)CO
MPL
ICAT
ION
SO
UTC
OM
ES
Stef
fen
et a
l.,
2003
(P)
824
(ope
rate
d be
twee
n A
pril
1996
and
M
ay 1
997)
79: fi
rst-
gene
ratio
n LS
AG
B
754:
seco
nd-
gene
ratio
n LS
AG
B
(ope
rate
d be
twee
n M
ay 1
997
and
Febr
uary
200
1)
118
± 1
kg
(75–
224)
42.4
± 1
kg/
m2
(31–
69)
Exce
ss w
eigh
t: 90
± 1
%
(35–
368%
)
49 ±
17
3.7
Intra
-ope
rativ
e co
mpl
icat
ions
: 1.4
%
(12
case
s)
- 12
cas
es o
f tra
umat
ic in
tuba
tion.
All
intra
-ope
rativ
e co
mpl
icat
ions
occ
urre
d in
the fi r
st 1
00 p
atie
nts.
- Li
ver h
emat
oma
(5);
sple
nic
hem
orrh
age
(3):
hem
orrh
age
from
gas
tro-e
pipl
oic
vein
s (2)
; CO
2-em
bolis
m (1
); es
opha
geal
per
fora
tion
(1)
Post
-ope
rativ
e co
mpl
icat
ions
: M
inor
ear
ly c
ompl
icat
ions
: 3%
(25
case
s)-
13 p
atie
nts r
equi
red
antib
iotic
ther
apy
for p
ulm
onar
y at
elec
tasi
s or p
neum
onia
- 2
patie
nts r
equi
red
enem
as fo
r a p
rolo
nged
su
b-ile
us-
10 p
atie
nts h
ad m
inor
wou
nd p
robl
ems
Com
plic
atio
ns r
elat
ed to
the
band
: 6.3
%
(51
of 8
24 p
atie
nts i
n 5
year
s)-
Ban
d le
akag
e (1
4)*;
infe
ctio
n (2
)† ; sl
ippa
ge (2
2)‡ ;
eros
ion
(13)
§
Com
plic
atio
ns d
ue to
the
acce
ss p
ort o
r tu
be|| :
6.8%
(56
out o
f 824
pat
ient
s in
5 ye
ars)
- In
fect
ion
(8);
hem
atom
a (2
); di
scom
fort/
prom
inen
ce
(19)
; dis
loca
tion
(8);
tube
leak
(10)
; tub
e di
scon
nect
ing/
kink
ing
(9)
EW
L (%
) B
MI (
kg/m
2 )
1 ye
ar (n
= 8
21)
29.5
± 0
.5
35.8
± 0
.2
2 ye
ars (
n =
744)
41
.1 ±
0.7
33
.2 ±
0.1
3 ye
ars (
n =
593)
48
.7 ±
0.9
31
.5 ±
0.2
4 ye
ars (
n =
380)
54
.5 ±
1.2
30
.0 ±
0.3
5 ye
ars (
n =
184)
57
.1 ±
1.9
29
.2 ±
0.4
- W
eigh
t los
s was
con
side
red
insu
ffi ci
ent i
n 14
1 (1
7.1%
) pa
tient
s. A
n ad
juva
nt tr
eatm
ent w
ith li
pid-
abso
rptio
n in
hibi
tors
(orli
stat
) or s
ibut
ram
ine
was
adm
inis
tere
d to
86
of t
hese
pat
ient
s, w
ith a
succ
ess r
ate
of 6
6.3%
; 26
of
141
patie
nts (
2.1%
) req
uire
d a
re-o
pera
tion.
- Q
ualit
y of
life
was
eva
luat
ed b
ased
on
the
BA
RO
S sc
ore¶
in 3
48 p
atie
nts 2
, 3 a
nd 5
yea
rs a
fter s
urge
ry.
- In
pat
ient
s with
out c
o-m
orbi
ditie
s, ou
tcom
es w
ere
good
to
exc
elle
nt fo
r 42
of 6
8 pa
tient
s (61
.7%
) 2 y
ears
afte
r su
rger
y, a
nd fo
r 29
of 4
0 pa
tient
s (72
.5%
) 3 y
ears
afte
r su
rger
y.-
In p
atie
nts w
ith c
o-m
orbi
ditie
s, ou
tcom
es w
ere
good
to
exce
llent
for 2
50 o
ut o
f 280
pat
ient
s (89
%) 2
yea
rs a
fter
surg
ery,
and
for 1
57 o
ut o
f 177
pat
ient
s (88
.7%
) 3 y
ears
af
ter s
urge
ry.
71
TAB
LE E
-1
Swed
ish
adju
stab
le g
astr
ic b
andi
ng (c
ont’d
)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
INIT
IAL
WEI
GH
T AN
D B
MI
AOT
(MIN
)AL
OS
(DAY
S)CO
MPL
ICAT
ION
SO
UTC
OM
ES
Func
tiona
l com
plic
atio
ns++
: 7%
(58
of 8
24 p
atie
nts i
n 5
year
s)P
rim
ary
band
into
lera
nce:
1%
(8)
Seco
ndar
y ba
nd in
tole
ranc
e: 6
.1%
(50)
Con
vers
ions
: 5.2
% (8
24 p
atie
nts)
; 2.1
%
betw
een
2000
and
200
1
Dea
ths:
0%
at 1
mon
th; 0
.4%
** a
t 1 m
onth
Aut
hors
’ com
men
ts:
The
auth
ors c
oncl
ude
that
LSA
GB
is e
ffi ca
ciou
s, w
ith a
ccep
tabl
e co
mpl
icat
ion
and
mor
talit
y ra
tes.
* C
onsi
dera
tion
is g
iven
onl
y to
leak
s pro
duce
d w
ith th
e 2n
d-ge
nera
tion
SAG
Bs;
32%
of t
he 1
st-g
ener
atio
n SA
GB
s im
plan
ted
betw
een
Apr
il 19
96 a
nd M
arch
199
7 w
ere
repl
aced
bec
ause
of a
leak
at
the
seam
.†
One
ear
ly a
nd o
ne la
te b
and
infe
ctio
n.‡
Ban
d sl
ippa
ges o
ccur
red
betw
een
the
10th
and
the
48th
pos
t-ope
rativ
e m
onth
s, w
ith a
pea
k in
cide
nce
durin
g th
e 2n
d an
d 3r
d po
st-o
pera
tive
year
s.§
Ban
d er
osio
ns o
ccur
red
betw
een
the
11th
and
the
42nd
mon
ths.
|| Mos
t of t
he c
ompl
icat
ions
occ
urre
d du
ring
the fi r
st tw
o po
st-o
pera
tive
year
s (3.
2% a
nd 2
.2%
). Th
ese
com
plic
atio
ns d
ropp
ed to
1.7
% (3
rd y
ear)
, to
1.0%
(4th
yea
r) a
nd to
0%
(5th
yea
r).
¶ B
AR
OS
(Bar
iatri
c Ana
lysi
s and
Rep
ortin
g O
utco
me
Syst
em).
This
syst
em is
bas
ed o
n th
e sc
orin
g of
pat
ient
s’ qu
ality
of l
ife, E
WL,
com
plic
atio
ns a
nd im
prov
emen
t in
co-m
orbi
ditie
s. **
Co-
mor
bidi
ties c
ause
d 2
deat
hs, a
nd th
e 3r
d pa
tient
die
d fo
llow
ing
sept
ic c
ompl
icat
ion
afte
r con
vers
ion
to g
astri
c by
pass
(ban
d in
tole
ranc
e).
++ T
he a
utho
rs d
efi n
e pr
imar
y ba
nd in
tole
ranc
e as
tota
l sol
id-f
ood
into
lera
nce
with
an
empt
y ba
nd. S
econ
dary
ban
d in
tole
ranc
e is
cha
ract
eriz
ed b
y a
tota
l or s
ubto
tal s
olid
-foo
d in
tole
ranc
e at
min
imal
ba
nd fi
lling
, ind
ucin
g or
mai
ntai
ning
wei
ght l
oss.
72
TAB
LE E
-2
Com
pari
sons
bet
wee
n tw
o la
paro
scop
ic g
astr
ic-b
andi
ng t
echn
ique
s
AUTH
OR
STY
PE O
F PR
OCE
DU
RE
NU
MB
ER
OF
PATI
ENTS
WEI
GH
T(K
G)
EXCE
SS
WEI
GH
T(K
G)
BM
I(K
G/M
2)
AOT
(MIN
)AL
OS
(DAY
S)FO
LLO
W-U
P(M
ON
THS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Wei
ner
et a
l.,20
01(R
C)
Ret
roga
stric
pl
acem
ent o
f LA
GB
(Bel
ache
w a
nd
Cal
dier
e)
5114
5.8
± 17
.466
.2 ±
10.
248
.5 ±
3.4
58.2
± 4
.84.
9 ±
0.8
18Po
st-o
pera
tive
com
plic
atio
ns: 3
(p <
0.0
01)
Rel
ated
to p
ort s
ites:
4 (N
S)R
elat
ed to
the
band
: 3 (p
< 0
.001
)-
Slip
page
: 2 (p
< 0
.001
)-
Pouc
h di
lata
tion:
1 (p
< 0
.05)
- Es
opha
geal
dila
tatio
n: 1
(dia
met
er >
30
mm
)O
ther
sym
ptom
s:
- H
unge
r sen
satio
n: 3
- D
ysph
agia
: 1-
Rec
urre
nt v
omiti
ng: 1
Dea
ths:
0
Wei
ght
loss
(gra
phic
ally
dep
icte
d da
ta)
Ret
roga
stric
pl
acem
ent:
(n =
50)
- 1
year
: 37
kg-
18 m
onth
s: >
40
kg
Esop
hago
gast
ric
plac
emen
t: (n
= 4
9)
- 1
year
: 34
kg-
18 m
onth
s: >
40
kg
Esop
hago
gast
ric
plac
emen
t of
LAG
B
(Wei
ner)
5014
2.9
± 14
.369
.7 ±
12.
349
.5 ±
4.2
56.5
± 5
.24.
5 ±
0.4
Post
-ope
rativ
e co
mpl
icat
ions
:
Rel
ated
to p
ort s
ites:
0R
elat
ed to
the
band
: 0
Esop
hage
al d
ilata
tion:
1(d
iam
eter
> 3
0 m
m)
Oth
er sy
mpt
oms:
-
Hun
ger s
ensa
tion:
1-
Dys
phag
ia: 1
- R
ecur
rent
vom
iting
: 1
Dea
ths:
0
73
TAB
LE E
-2
Com
pari
sons
bet
wee
n tw
o la
paro
scop
ic g
astr
ic-b
andi
ng t
echn
ique
s (co
nt’d
)
AUTH
OR
STY
PE O
F PR
OCE
DU
RE
NU
MB
ER
OF
PATI
ENTS
WEI
GH
T(K
G)
EXCE
SS
WEI
GH
T(K
G)
BM
I(K
G/M
2)
AOT
(MIN
)AL
OS
(DAY
S)FO
LLO
W-U
P(M
ON
THS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Wei
ss
et a
l.,20
02(R
C)
Con
vent
iona
l pl
acem
ent o
f LA
GB
(The
ban
d is
pl
aced
from
1 to
3
cm b
elow
the
low
er e
soph
agea
l sp
hinc
ter,
base
d on
For
sell’
s te
chni
que.
)
28
(24
fem
ales
an
d 4
mal
es)
Uns
pecifi e
dU
nspe
cifi e
d42
.5(3
9.3–
47.3
)77
(55–
152)
3.2
(3–1
0)23
(20–
26)
Con
vers
ions
: 3.6
%
Re-
oper
atio
ns:
10.7
%(3
cas
es, i
nclu
ding
1 o
pen
surg
ery)
- B
and
mig
ratio
n in
to
the
stom
ach:
1-
Pouc
h di
lata
tion:
1-
Port
disc
onne
ctio
n: 1
Oth
er sy
mpt
oms:
-
Dys
phag
ia: 0
%-
Hea
rtbur
n: 1
1.1%
BM
I (k
g/m
2 )
25.1
(22.
4–36
.3)
Esop
hago
gast
ric
plac
emen
t of
LAG
B
(The
ban
d is
pl
aced
aro
und
the
esop
hagu
s an
d th
eref
ore
over
laps
the
low
er e
soph
agea
l sp
hinc
ter,
base
d on
Niv
ille’
s te
chni
que.
)
26
(23
fem
ales
an
d 3
mal
es)
Uns
pecifi e
dU
nspe
cifi e
d41
.8(3
9–44
.1)
82(6
5–16
0)3.
5(3
–10)
24(2
0–26
)C
onve
rsio
ns:
3.8%
Re-
oper
atio
ns:
19.2
%(5
cas
es, i
nclu
ding
1 o
pen
surg
ery)
- Ea
rly b
and
mig
ratio
n in
to
the
esop
hagu
s: 1
- La
te b
and
mig
ratio
n in
to th
e es
opha
gus:
1-
Port
disl
ocat
ion:
1-
Psyc
holo
gica
l pro
blem
s: 1
Oth
er sy
mpt
oms:
-
Dys
phag
ia: 5
7.1%
-
Hea
rtbur
n: 1
4.3%
22.9
(21.
9–35
)
74
TAB
LE E
-3
Com
pari
sons
bet
wee
n th
e L
ap-B
and
and
Hel
ioga
st b
ands
AUTH
OR
STY
PE O
F PR
OCE
DU
RE
NU
MB
ER O
F PA
TIEN
TSB
MI
(KG
/M2
)AO
T(M
IN)
COM
PLIC
ATIO
NS
OU
TCO
MES
Bla
nco-
Enge
rt et
al.,
200
3(R
C)
LAG
B(L
ap-B
and®
)30
43.4
50
Lap-
Ban
d H
elio
gast
Slip
page
0
0M
igra
tion
0 1
(3.3
%) (
NS)
Def
ects
0
3 (1
0%) (
NS)
Inad
equa
te
stom
a si
ze
0 26
(87%
) (p
< 0.
0001
)O
ther
1*
8† (
p =
0.02
37‡ )
- 14
of th
e 26
pat
ient
s, in
who
m it
was
not
pos
sibl
e to
re
ach
a fu
nctio
nal s
tom
a si
ze, p
rese
nted
with
sym
ptom
s su
ch a
s pai
n, n
ause
a, b
rady
card
ia a
nd sw
eatin
g, a
fter
an in
crea
se in
the fi l
l vol
ume
from
5 to
9 c
c. T
hese
sy
mpt
oms w
ere
pres
umab
ly d
ue to
vag
us re
actio
ns
caus
ed b
y in
crea
sed
pres
sure
on
the
sutu
red
stom
ach.
- In
the
Hel
ioga
st g
roup
, 3 re
-ope
ratio
ns (L
RYG
B) w
ere
perf
orm
ed (1
bec
ause
of a
n ul
cer a
nd in
fect
ion
and
2 af
ter b
and
dila
tatio
n ag
grav
ated
by
atte
mpt
s to
incr
ease
th
e ba
nd v
olum
e).
%
EW
L
Lap-
Ban
d H
elio
gast
p
valu
e1
mon
th
4.2
± 0
.86
4.8
± 0
.55
0.00
473
mon
ths
12.2
± 1
.30
9.4
± 1
.39
0.00
016
mon
ths
26.4
± 2
.88
17.1
± 1
.65
0.00
0112
mon
ths
41.7
± 2
.71
28.3
± 2
.40
0.00
01
- At 6
mon
ths,
the
perc
enta
ge o
f exc
ess w
eigh
t los
s in
the
Hel
ioga
st g
roup
was
low
er th
an th
at in
the
Lap-
Ban
d gr
oup.
The
diff
eren
ce is
sign
ifi ca
nt.
- The
com
plic
atio
n ra
te w
as si
gnifi
cant
ly h
ighe
r in
the
Hel
ioga
st g
roup
.
LAG
B(H
elio
gast
®)
30
3 (1
st
gene
ratio
n:
larg
e)
27 (2
nd
gene
ratio
n:
med
ium
)
41.2
54
* 1
band
rupt
ure
(3.3
%).
† 6
case
s of b
and
rota
tion
(20%
) and
2 c
ases
of i
nfec
tion
(one
afte
r a re
visi
on a
nd th
e ot
her a
ssoc
iate
d w
ith b
and
mig
ratio
n).
‡ Th
e di
ffere
nce
is si
gnifi
cant
onl
y fo
r ban
d ro
tatio
n.
75
TAB
LE E
-4
Lap
aros
copi
c ad
just
able
gas
tric
ban
ding
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I (K
G/M
2)
FOLL
OW
-UP
(MO
NTH
S)CO
MPL
ICAT
ION
S O
UTC
OM
ES
Ang
risan
iet
al.,
20
04a
(R)
LAG
B(L
ap-B
and)
- In
itial
ly, 3
562
patie
nts o
pera
ted
betw
een
Janu
ary
1996
and
200
3 in
26
cen
tres i
n Ita
ly
- 57
3 pa
tient
s (4
51 fe
mal
es a
nd
122
mal
es) w
ere
follo
wed
for 5
yea
rs
(ope
rate
d be
twee
n Ja
nuar
y 19
96 a
nd
Dec
embe
r 199
7).
Thes
e pa
tient
s wer
e gr
oupe
d ac
cord
ing
to th
eir p
re-o
pera
tive
BM
Is.
Gro
up A
(n =
166
)
30–3
9.9
(1
38 M
/ 25
8 F)
Mea
n ag
e:36
.9 ±
12.
1 ye
ars
(18–
61)
Gro
up B
(n =
302
)
40–4
9.9
(6
1 M
/ 24
1 F)
Mea
n ag
e:37
.8 ±
10.
9 ye
ars
(21–
63)
Gro
up C
(n =
96)
50–5
9.9
(19
M /
77 F
)M
ean
age:
39 ±
12.
5 ye
ars
(18–
74)
Gro
up D
(n =
9)
≥ 60
(1
M /
8 F)
Mea
n ag
e:37
.1 ±
14.
7 ye
ars
(23–
65)
5-ye
ar d
ata
avai
labl
e on
38
1 pa
tient
sG
roup
sA
BC
DN
umbe
r of
patie
nts
T =
573
166
302
969
Gas
tric-
pouc
h di
latio
n
T =
24
(4. %
)
12(7
.2%
)p
< 0.
05
10(3
.3%
)1
(1%
)1
(11.
1%)
p <
0.01
Intra
gast
ric
band
m
igra
tion/
eros
ion
T =
12
(2.1
%)
3(1
.8%
)6
(1.9
%)
3 (3
.1%
)p
< 0.
01
0
Post
-op
erat
ive
mor
talit
y ra
teT
= 5
(0.8
7%)
1(0
.6%
)1
(0.3
%)
3 (3
.1%
)0
Con
vers
ions
to o
ther
type
s of b
aria
tric
surg
ery:
1.
4% (8
/573
)
Rem
oval
of g
astri
c ba
nd: 4
.2%
(24/
573)
Gro
ups
AB
CD
Num
ber
of p
atie
nts
T =
381/
573
(66.
5 %
)
96/1
66(5
7.8%
)21
4/30
2(7
0.8%
)64
/96
(66.
6%)
7/9
(77.
7%)
Initi
al B
MI
(kg/
m2 )
T =
30.2
± 8
.3
27.5
± 5
231
.6 ±
4.7
37.6
± 1
7.3
41.4
± 6
.9
Lost
BM
I(k
g/m
2 )
T =
12 ±
6
9.8
± 5.
412
.9 ±
5.2
15.8
± 8
.123
.2 ±
4.9
EWL
(%)
T =
54.8
± 2
2.3
54.6
± 3
2. 3
54.1
± 1
7.2
51.6
± 3
559
.1 ±
17.
1
< 25
% E
WL
T =
22/3
65(6
%)
9(9
.7%
)4
(1.9
%)
p <
0.05
9(9
.3 %
)0
< 50
% E
WL
T =
91/3
65
(24.
9 %
)
17(1
8.4%
)51
(24.
8%)
23(2
3.9%
)0
76
TAB
LE E
-4
Lap
aros
copi
c ad
just
able
gas
tric
ban
ding
(con
t’d)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
INIT
IAL
WEI
GH
T FO
LLO
W-U
P (M
ON
THS)
AOT
(MIN
)AL
OS
(DAY
S)CO
MPL
ICAT
ION
S O
UTC
OM
ES
Ang
risan
iet
al.,
20
04b
(R)
Ret
rosp
ectiv
e,
mul
ticen
tre st
udy
of 2
10 o
f 33,
129
Lap-
Ban
d pa
tient
s bet
wee
n Ja
nuar
y 19
96 a
nd
Dec
embe
r 200
2
176
wom
en a
nd
34 m
en
Age
:38
.9 ±
11.
8 ye
ars
(17–
66)
BM
I ≤ 3
5 kg
/m2
Co-
mor
bidi
ties:
10
9 in
55
patie
nts
33.9
± 1
.1 k
g/m
2
(25.
1–35
)
Exce
ss w
eigh
t:29
.7 ±
7.1
kg
(8–4
1)
6–60
––
Tube
-por
t lea
k: 4
- tub
e-po
rt re
conn
ectio
n: 2
- p
ort r
epla
cem
ent:
2
Pouc
h di
lata
tion:
11
- ban
d defl a
tion:
5- b
and
repo
sitio
ning
: 2- b
and
rem
oval
: 2
Intra
gast
ric m
igra
tion:
2
- ban
d re
mov
al: 1
Dea
ths:
1(a
t 20
mon
ths f
ollo
win
g se
ptic
emia
cau
sed
by
perf
orat
ion
of th
e ga
stric
po
uch)
Follo
w-u
pEW
L (%
)B
MI
(kg/
m2 )
6 m
onth
s (n
= 21
0/21
0)28
.1 ±
20.
731
.1 ±
2.1
51
year
(n =
182
/197
)52
.5 ±
13.
229
.7 ±
2.1
92
year
s (n
= 11
9/14
8)61
.3 ±
14.
728
.7 ±
3.8
3 ye
ars (
n =
75/9
9)64
.7 ±
12.
226
.7 ±
4.3
4 ye
ars (
n =
49/7
3)68
.8 ±
15.
327
.9 ±
3.2
5 ye
ars (
n =
21/2
9)71
.9 ±
10.
728
.2 ±
0.9
Bef
ore
Afte
r(1
yea
r)Su
cess
(%)
Anx
iety
and
dep
ress
ion
472
95.8
Ost
eoar
thro
path
y43
490
.8H
yper
tens
ion
91
88.9
Gas
tro-e
soph
aege
al
refl u
x di
seas
e5
010
0
Non
-insu
lin-d
epen
dent
di
abet
es4
010
0
Res
pira
tory
dis
orde
rs1
10
At o
ne y
ear,
all c
o-m
orbi
ditie
s had
bee
n re
solv
ed in
49
of th
e 55
pat
ient
s (89
.1%
)
77
TAB
LE E
-4
Lap
aros
copi
c ad
just
able
gas
tric
ban
ding
(con
t’d)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
INIT
IAL
WEI
GH
T FO
LLO
W-U
P (M
ON
THS)
AOT
(MIN
)AL
OS
(DAY
S)CO
MPL
ICAT
ION
SO
UTC
OM
ES
Bel
ache
w
et a
l.,19
98(R
)
550
43.0
kg/
m2
6062
Not
ava
ilabl
eE
arly
com
plic
atio
ns: 0
.4%
*- G
astri
c pe
rfor
atio
n: 1
Late
com
plic
atio
ns:
- Pou
ch d
ilata
tion:
28
- Ban
d er
osio
n: 1
8
Re-
oper
atio
ns: 5
6%
Dea
ths:
0
EWL:
50%
Bel
ache
wet
al.,
20
02(R
)
763:
Lap
-Ban
d(3
cen
tres)
42 k
g/m
2
(35–
65)
Ove
r 48
patie
nts
(90%
of p
atie
nts)
U
nspe
cifi e
dU
nspe
cifi e
dE
arly
com
plic
atio
ns:
- Gas
tric
perf
orat
ion:
0.5
%- L
arge
-bow
el p
erfo
ratio
n: 0
.1%
- Ble
edin
g: 0
.1%
Late
com
plic
atio
ns:
- Ban
d er
osio
n: 0
.9%
- Foo
d in
tole
ranc
e: 8
%- A
cces
s-po
rt pr
oble
ms:
2.5
%
(11.
1% re
-ope
rate
d)- C
onve
rsio
ns: 1
.3%
Dea
ths:
0.1
%
BM
I: 30
kg/
m2 at
4 y
ears
(2
8.6%
dec
reas
e)
EWL:
6 m
onth
s: 3
0%
1 ye
ar: 4
0%2
year
s: 5
0%4
year
s: 5
0–60
%
Dar
gent
,19
99(R
)
500
43.0
kg/
m2
(36–
60)
28N
ot a
vaila
ble
Not
ava
ilabl
eE
arly
com
plic
atio
ns:
0.8%
- Gas
tric
perf
orat
ion:
1- B
and
mig
ratio
n: 2
Late
com
plic
atio
ns:
- Pou
ch d
ilata
tion:
25
- Ban
d er
osio
n: 3
- Acc
ess-
port
com
plic
atio
ns: 5
Re-
oper
atio
ns:
3.6%
Dea
ths:
0
EWL:
6 m
onth
s: 4
5% (n
= 4
43)
1 ye
ar: 5
6% (n
= 2
70)
2 ye
ars:
65%
(n =
96)
3 ye
ars:
64%
(n =
19)
* Ex
clud
ing
food
into
lera
nce.
78
TAB
LE E
-4
Lap
aros
copi
c ad
just
able
gas
tric
ban
ding
(con
t’d)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
INIT
IAL
WEI
GH
T FO
LLO
W-U
P (M
ON
THS)
AOT
(MIN
)AL
OS
(DAY
S)CO
MPL
ICAT
ION
SO
UTC
OM
ES
DeM
aria
et
al.,
2001
(R)
Lap-
Ban
d: 3
744
.5 ±
4 k
g/m
2
276
± 38
lbs
12–4
8U
nspe
cifi e
dU
nspe
cifi e
d- E
soph
agea
l dila
tatio
n: 7
1%- B
and
mig
ratio
n: 8
.5%
- Lea
kage
: 5.4
%- I
nfec
tion:
5.4
%
- Con
vers
ions
: 2.7
%- D
eath
s: 0
Follo
w-u
pB
MI
(kg/
m2 )
EWL
12 m
onth
s (n
= 28
)37
.0 ±
634
.5 ±
20%
24 m
onth
s (n
= 24
)37
.2 ±
736
± 2
3%
36 m
onth
s (n
= 15
)35
.8 ±
638
± 2
7%
48 m
onth
s (n
= 4)
3444
%
Dix
on
and
O’B
rien,
20
02(R
)
LSA
GB
: 50
patie
nts
with
type
2 d
iabe
tes
from
a c
ohor
t of 5
00
cons
ecut
ive
patie
nts
48.2
± 8
kg/
m2
137
± 30
kg
12U
nspe
cifi e
dU
nspe
cifi e
dE
arly
com
plic
atio
ns (6
% o
f pat
ient
s)- I
nfec
tion:
4%
- Res
pira
tory
: 2%
Late
com
plic
atio
ns:
30%
- Sto
mac
h pr
olap
se: 2
0%- B
and
eros
ion:
6%
- Lea
kage
: 4%
- Con
vers
ions
: 0
At 1
yea
r:- B
MI:
38.7
± 6
kg/
m2
- Wei
ght:
110
± 24
kg
- EW
L: 3
8 ±
14%
(47
± 17
%) (
coho
rt of
50
0 pa
tient
s)
Rem
issi
on o
f dia
bete
s in
32 p
atie
nts
(unc
hang
ed o
r hig
h gl
ycem
ic c
ontro
l in
18 p
atie
nts)
Impr
oved
qua
lity
of li
fe fo
r 34
of th
e la
st
35 o
pera
ted
patie
nts
Favr
etti
et a
l.,19
98(R
)
180
45.5
kg/
m2
1880
2–3
Ear
ly c
ompl
icat
ions
:
- Per
fora
tion:
1- B
and
mig
ratio
n: 1
Late
com
plic
atio
ns:
- Ban
d m
igra
tion:
6- B
and
eros
ion:
1
Re-
oper
atio
n: 2
.8%
Dea
ths:
0
BM
I:6
mon
ths:
37
kg/m
2 18
mon
ths:
35
kg/m
2
79
TAB
LE E
-4
Lap
aros
copi
c ad
just
able
gas
tric
ban
ding
(con
t’d)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
IN
ITIA
L W
EIG
HT
FOLL
OW
-UP
(MO
NTH
S)AO
T(M
IN)
ALO
S(D
AYS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Fiel
ding
et a
l.,
1999
(R)
335
46.7
kg/
m2
(34–
86)
1871
1.4
Ear
ly c
ompl
icat
ions
: 2.
1%
- Inf
ectio
n: 4
- Ban
d m
igra
tion:
2
Late
com
plic
atio
ns:
- Ban
d m
igra
tion:
12
- Acc
ess-
port
com
plic
atio
ns: 5
Dea
ths:
0
EWL:
12
mon
ths:
52%
(n =
125
)18
mon
ths:
62%
(n =
58)
Gre
enst
ein
et a
l.,19
99(P
)
250
48 k
g/m
260
Uns
pecifi e
dU
nspe
cifi e
dC
ompl
icat
ions
: 5.6
% (1
3 ca
ses)
Dea
ths:
1
EWL:
40
kg (m
ean)
Gus
tavs
son
and
Wes
tling
,20
02(R
)
90
(199
4–19
96)
43 k
g/m
284
Uns
pecifi e
dU
nspe
cifi e
d58
% o
f pat
ient
s req
uire
d re
-ope
ratio
nB
MI a
t 5 y
ears
: 33
.7 k
g/m
2 (24
–44)
Mill
er a
nd
Hel
l,19
99(N
CC
)
A: 1
02 (s
ilico
ne)
B: 5
4 (S
wed
ish)
45 k
g/m
2
43 k
g/m
228
(1.5
–46)
Uns
pecifi e
dA
: 4.3
B: 3
.3E
arly
com
plic
atio
ns:
- Sili
cone
: hem
atom
a (1
) - S
wed
ish:
wou
nd in
fect
ion
(1)
Late
com
plic
atio
ns:
- Sili
cone
: pou
ch d
ilata
tion
(2);
band
rupt
ure
(2)
- Sw
edis
h: p
ouch
dila
tatio
n (1
); ba
nd e
rosi
on (1
); ac
cess
-por
t com
plic
atio
ns (1
); re
-ope
ratio
ns (7
%)
Dea
ths:
0
BM
I:1
year
: 34
kg/m
2 3
year
s: 2
8 kg
/m2
80
TAB
LE E
-4
Lap
aros
copi
c ad
just
able
gas
tric
ban
ding
(con
t’d)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
IN
ITIA
L W
EIG
HT
FOLL
OW
-UP
(MO
NTH
S)AO
T(M
IN)
ALO
S(D
AYS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Neh
oda
et a
l.,20
01(R
)
320
(LSA
GB
)44
.29
kg/m
2
127.
8 kg
12(6
–28)
Uns
pecifi e
dU
nspe
cifi e
dR
e-op
erat
ions
: 10.
3%EW
L: 6
8% (3
1.0
kg: 4
.3 k
g/m
onth
the fi r
st 3
mon
ths)
Impr
oved
qua
lity
of li
fe
in 9
7% o
f cas
es
O’B
rien
et a
l.,19
99(P
)
277
of 3
02(L
ap-B
and)
44.5
kg/
m2
4857
(45–
110)
3.9
(1–7
)E
arly
com
plic
atio
ns: 4
.3%
(ban
d m
igra
tion)
Late
com
plic
atio
ns:
- Her
nia:
27
- Eso
phag
itis:
1
Dea
ths:
0
EWL:
1
year
: 51.
0 ±
17%
(n =
120
)2
year
s: 5
8.3
± 20
% (n
= 4
3)3
year
s: 6
1.6
± 2%
(n =
25)
4 ye
ars:
68.
2 ±
21%
(n =
12)
Rub
enst
ein,
2002
(R
)
63
(46
with
co
-mor
bidi
ties)
(Nov
embe
r 199
6 to
May
199
9)
48.8
kg/
m2
(36.
8–67
)6
(62
patie
nts)
12(5
9 pa
tient
s)
24(1
9 pa
tient
s)
36(1
3 pa
tient
s)
193
± 72
.1(s
tart)
120
± 25
.7(e
nd)
1.4
(1–2
)E
arly
com
plic
atio
ns:
- Gas
tric
perf
orat
ion:
1.5
%
- Acc
ess-
port
prob
lem
s: 7
.9%
Late
com
plic
atio
ns:
Ban
d m
igra
tion:
14.
2% (3
con
verte
d to
gas
tric
bypa
ss a
nd 6
requ
ired
band
rem
oval
); er
osio
n: 1
.5%
(1);
infe
ctio
n: 1
.5%
(1 g
astri
c sy
ndro
me)
Re-
oper
atio
ns:
0
Dea
ths:
0
EWL
6 m
onth
s: 2
7.2
± 14
.2%
1
year
: 38.
3 ±
15.6
%
2 ye
ars:
46.
6 ±
19.5
%
3 ye
ars:
53.
6 ±
23.8
%
81
TAB
LE E
-4
Lap
aros
copi
c ad
just
able
gas
tric
ban
ding
(con
t’d)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
IN
ITIA
L W
EIG
HT
FOLL
OW
-UP
(MO
NTH
S)AO
T(M
IN)
ALO
S(D
AYS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Szol
d an
d A
bu-A
beid
,20
02
(R)
715
(Nov
embe
r 199
6 to
May
199
9)
43.1
kg/
m2
(35–
66)
4878
(36–
165)
1.2
Ear
ly c
ompl
icat
ions
: 1.
1%
(0.8
% in
tra-o
pera
tive)
Late
com
plic
atio
ns: 7
.4%
- Inf
ectio
n: 1
- Ban
d er
osio
n: 3
- Spl
enic
abs
cess
: 2- A
cces
s-po
rt pr
oble
ms:
18
Re-
oper
atio
ns:
57
Dea
ths:
0
26.5
% d
ecre
ase
in in
itial
BM
I in
2 ye
ars
Zim
mer
man
n et
al.,
19
99
(R)
894
(199
5–19
98)
864
(Lap
-Ban
d)
33(S
wed
ish
band
)
42.0
kg/
m2
(35–
72)
(115
with
B
MI >
50)
1235
3.0
Ear
ly c
ompl
icat
ions
: 0.3
3%
- Hep
atic
ble
edin
g: 2
- Pne
umot
hora
x: 2
Late
com
plic
atio
ns:
- Pou
ch d
ilata
tion:
49
- Ban
d ru
ptur
e: 4
8- L
eaka
ge: 5
Re-
oper
atio
ns: 1
8
Dea
ths:
1
B
MI
EWL
(k
g/m
2 ) (%
)6
mon
ths (
n =
676)
35
32 12
mon
ths (
n =
233)
34
40 18
mon
ths (
n =
89)
33
46.5
24 m
onth
s (n
= 47
) 33
39
82
TAB
LE E
-5
Com
pari
sons
bet
wee
n op
en a
nd la
paro
scop
ic a
djus
tabl
e ga
stri
c ba
ndin
g
AUTH
OR
SN
UM
BER
O
F PA
TIEN
TS
BM
I OR
IN
ITIA
L W
EIG
HT
AOT
(MIN
)AL
OS
(DAY
S)FO
LLO
W-U
P(M
ON
THS)
COM
PLIC
ATIO
NS
OU
TCO
MES
De
Wit
et a
l.,19
99(R
C)
AG
B: 2
5W
eigh
t:14
6.4
± 19
.9 k
g
BM
I: 49
.7 ±
5.6
kg/
m2
76 ±
20
7.2
(5–1
3)12
Surg
ical
com
plic
atio
ns:
- Inc
isio
nal h
erni
a: 7
(3 p
atie
nts)
- Ban
d m
igra
tion:
1
Acc
ess-
port
com
plic
atio
ns:
6 (5
pat
ient
s)- R
e-op
enin
g: 1
- Dis
loca
tion:
4- I
nfec
tion:
1
Re-
adm
issi
on:
15 (7
pat
ient
s)
No
deat
hs
At 1
yea
r:
LAG
BA
BG
pth
resh
old
Wei
ght
(kg)
11
7.2
± 25
.211
2.0
± 19
.1ns
Wei
ght
loss
(k
g)35
34.4
ns
BM
I(k
g/m
2 )39
.7 ±
8.7
39.1
± 8
.2ns
- At o
ne y
ear,
no si
gnifi
cant
diff
eren
ces w
ere
foun
d in
th
e nu
mbe
r of p
ost-o
pera
tive
or e
arly
com
plic
atio
ns.
- Wei
ght l
oss w
as si
mila
r in
both
gro
ups.
- LA
GB
was
ass
ocia
ted
with
a sh
orte
r hos
pita
l st
ay a
nd a
low
er re
-adm
issi
on ra
te (s
ignifi c
ant
diffe
renc
es).
LAG
B: 2
5W
eigh
t:15
2.2
± 31
.4 k
g
BM
I: 51
.3 ±
10.
4 kg
/m2
150
± 48
5.9
(4–1
0)Su
rgic
al c
ompl
icat
ions
:- U
mbi
lical
her
nia:
1
Acc
ess-
port
com
plic
atio
ns:
7 (5
pat
ient
s)- R
e-op
enin
g: 2
- Dis
loca
tion:
5
Re-
adm
issi
on:
6 (5
pat
ient
s)
Con
vers
ions
: 2
- Due
to th
e in
abili
ty to
obt
ain
pneu
mop
erito
neum
(adh
esio
ns: 1
; he
pato
meg
aly:
1)
No
deat
hs
83
TAB
LE E
-5
Com
pari
sons
bet
wee
n op
en a
nd la
paro
scop
ic a
djus
tabl
e ga
stri
c ba
ndin
g (c
ont’d
)
AUTH
OR
SN
UM
BER
O
F PA
TIEN
TS
BM
I OR
INIT
IAL
WEI
GH
T AO
T (M
IN)
ALO
S(D
AYS)
FOLL
OW
-UP
(MO
NTH
S)CO
MPL
ICAT
ION
SO
UTC
OM
ES
Doh
erty
et a
l.,
2002
(PN
CC
)
Gro
up 1
: K
usm
ak
AG
B*:
40
(Mar
ch 1
992
to M
ay 1
995)
Mea
n B
MI:
50 k
g/m
2U
nspe
cifi e
dU
nspe
cifi e
dU
p to
96
- In
fect
ed b
and:
16%
-
Obs
truct
ive
aneu
rysm
al d
efor
mity
of
the
infl a
tabl
e bl
adde
r com
pone
nt o
f th
e ba
nd: 3
.2%
(AG
B: 2
) -
Enla
rged
pou
ch w
ith o
bstru
ctiv
e an
gula
tion
of th
e ou
tlet c
hann
el:
17.7
% (A
GB
: 7; L
ap-B
and:
4)
- H
erni
atio
n of
the
dist
al st
omac
h:
22.6
% (A
GB
: 11;
Lap
-Ban
d: 3
)-
30 re
-ope
ratio
ns re
quire
d to
co
rrec
t com
plic
atio
ns re
late
d to
the
impl
ante
d ba
nd
- B
and
rem
oval
in 2
7 ca
ses:
18
in
grou
p 1
and
9 in
gro
up 2
- 9
of 2
6 pa
tient
s com
plai
ned
of fr
eque
nt v
omiti
ng, g
astro
-es
opha
geal
refl u
x an
d ve
ry li
mite
d so
lid-f
ood
choi
ces
Year
12
34
56
78
Gro
up 1
:
Patie
nts
4040
3124
2018
1513
Wei
ght (
kg)
112
112
117
116
122
126
121
123
BM
I39
3833
3641
4440
44
% E
WL
4447
3340
3032
3332
Gro
up 2
:
Patie
nts
1918
1817
1513
Wei
ght (
kg)
118
121
120
128
134
127
BM
I40
4141
4447
43
% E
WL
2728
2517
2115
Gro
up 2
:(L
ap-B
and)
LAG
B: 1
7A
GB
: 5(1
995
to
Janu
ary
7,
1997
)
Mea
n B
MI:
47 k
g/m
2U
p to
72
Frie
d,20
00(R
NC
C)
310
AG
B(1
983–
1993
)B
MI
50.9
kg/
m2
(35.
4–86
.5)
Wei
ght
138.
6 kg
(93.
5–24
8.0)
Uns
pecifi e
d10
.536
(92%
of
patie
nts)
Ear
ly c
ompl
icat
ions
: 6.9
%
- M
ain
wou
nd-r
elat
ed
Late
com
plic
atio
ns:
12.6
%
of w
hich
9.4
% w
ere
hern
ias
- C
onve
rsio
ns o
r re-
oper
atio
ns:
0.6%
pos
t-ope
rativ
e7.
1% la
te
No
deat
hs
3 ye
ars:
LA
GB
A
GB
Mea
n w
eigh
t los
s (kg
): 37
.6
38.4
Mea
n B
MI (
kg/m
2 ):
37.1
36
.8
621
LAG
B(1
993–
1998
)B
MI
47.2
kg/
m2
(37.
6–58
.3)
Wei
ght
141.
3 kg
(90.
8–19
6.5)
2.8
Ear
ly c
ompl
icat
ions
: 1.
0%
- A
cces
s-po
rt co
mpl
icat
ions
Late
com
plic
atio
ns:
0.2%
- C
onve
rsio
ns o
r re-
oper
atio
ns:
1.3%
pos
t-ope
rativ
e6.
3% la
te
No
deat
hs
* In
this
stud
y, A
GB
refe
rs to
the
Kus
mak
type
of a
djus
tabl
e ga
stric
ban
d, a
nd L
ap-B
and
to th
e m
odel
mod
ifi ed
in 1
993.
84
TAB
LE E
-5
Com
pari
sons
bet
wee
n op
en a
nd la
paro
scop
ic a
djus
tabl
e ga
stri
c ba
ndin
g (c
ont’d
)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
BM
I OR
IN
ITIA
L W
EIG
HT
AOT
(MIN
)AL
OS
(DAY
S)FO
LLO
W-U
P(M
ON
THS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Wes
tling
et a
l.,
1998
(RN
CC
)
AG
B: 2
743
kg/
m2
(34–
57)
913
36
A
GB
LA
GB
C
onve
rsio
nsE
arly
com
plic
atio
ns:
- Inc
isio
nal h
erni
a 0
1
–- I
nfec
tion
3 4
–
- Ban
d er
osio
n 1
6
–- E
soph
agiti
s 0
33
–
Late
com
plic
atio
ns:
- Inf
ectio
n 3
4
–- B
and
eros
ion
1 6
3
Dea
ths
0 0
0
Out
com
es a
re n
ot
pres
ente
d by
surg
ical
ap
proa
ch.
BM
I:1
year
: 32
kg/m
2
2 ye
ars:
31
kg/m
2
LAG
B: 6
316
52
Con
vers
ions
: 16
170
4
85
APPE
ND
IX F
O
UTC
OM
ES O
F ST
UD
IES
COM
PAR
ING
DIF
FER
ENT
TYPE
S O
F B
ARIA
TRIC
SU
RG
ERY
TAB
LE F
-1
Com
pari
sons
bet
wee
n bi
liopa
ncre
atic
div
ersi
on a
nd o
ther
typ
es o
f op
en p
roce
dure
s
AUTH
OR
STY
PE O
F PR
OCE
DU
RE
(NU
MB
ER
OF
PATI
ENTS
)
BM
I(K
G/M
2)
AOT
(MIN
)AL
OS
(DAY
S)CO
MPL
ICAT
ION
SM
OR
TALI
TY R
ATE
OU
TCO
MES
:EW
L
Mar
ceau
et
al.,
1998
(RN
CC
)
BPD
-DG
: 252
BPD
-DS:
465
(239
end
-to
-end
an
asto
mos
es,
a te
chni
que
intro
duce
d in
199
2)
46 ±
9(n
= 2
33)
(BPD
)
47 ±
9(n
= 4
57)
(BPD
-DS)
Uns
pecifi e
dA
nnua
l re-
oper
atio
n ra
te:
- BPD
: 1.7
%- B
PD-D
S: 0
.1%
- BPD
: 4%
(4 e
arly
de
aths
, 6 la
te)
- BPD
-DS:
3.8
%
(9 e
arly
dea
ths,
9 la
te)
100
± 20
mon
ths (
BPD
): 61
± 2
2%
(n =
233
)51
± 2
5 m
onth
s (B
PD-D
S): 7
3 ±
21%
(n
= 4
57)
85 ±
3 m
onth
s (B
PD):
63 ±
21%
(1
08 o
pera
ted
in 1
989)
74 ±
4 m
onth
s (B
PD-D
S): 7
0 ±
21%
(5
2 op
erat
ed in
199
0)
Sign
ifi ca
nt d
iffer
ence
: p <
0.0
01
Rab
kin,
1998
(RN
CC
)
BPD
-DG
: 32
RYG
B: 1
38B
PD-D
S: 1
05
45 49 49
Uns
pecifi e
dB
PD: u
nspe
cifi e
dRY
GB
: uns
pecifi e
dB
PD-D
S: 2
per
itoni
tis; 1
thro
mbo
phle
bitis
as
soci
ated
with
pul
mon
ary
embo
lism
; 1
panc
reat
itis;
2 w
ound
infe
ctio
ns; 9
her
nias
; 1
re-o
pera
tion
(dat
a on
the fi r
st 3
7 pa
tient
s)
- BPD
: uns
pecifi e
d- R
YG
B: u
nspe
cifi e
d- B
PD-D
S: 1
dea
th
(alc
ohol
ism
)
24
mon
ths
48 m
onth
sB
PD:
69%
73
%RY
GB
: 74
%
63%
BPD
-DS:
78
%
73%
Baj
ardi
et
al.,
2000
(RN
CC
)
VB
G: 9
3(1
990–
1995
)48
.7(3
7–65
.6)
60(4
2–12
0)9
- Gas
tric
outle
t ste
nosi
s: 2
1.5%
(req
uirin
g ba
nd re
mov
al in
17.
2% o
f cas
es)
- Pos
t-ope
rativ
e he
rnia
: 15%
0%2
year
s: 4
8%
BPD
-DG
: 142
(199
3–19
98)
50(3
5–81
)17
0(9
0–31
5)16
- Pos
t-ope
rativ
e he
rnia
: 24%
- Dee
p ve
nous
thro
mbo
sis:
2.1
%- H
emor
rhag
e: 0
.8%
- Bow
el o
bstru
ctio
n: 0
.8%
- P
ulm
onar
y em
bolis
m: 1
.2%
- Ana
stom
otic
leak
: 0.4
%- M
alnu
tritio
n sy
ndro
me:
19%
(vita
min
s, iro
n, h
ypop
rote
inem
ia)
3.5%
(1 d
eath
due
to
pulm
onar
y em
bolis
m
shor
tly a
fter s
urge
ry,
2 du
e to
chr
onic
he
patit
is a
nd 1
to
perit
oniti
s)
2 ye
ars:
60%
86
TAB
LE F
-2
Com
pari
sons
bet
wee
n ve
rtic
al b
ande
d ga
stro
plas
ty a
nd o
ther
bar
iatr
ic-s
urge
ry t
echn
ique
s
AUTH
OR
STY
PE O
F PR
OCE
DU
RE
BM
I(K
G/M
2)
AOT
(MIN
)AL
OS
(DAY
S)CO
MPL
ICAT
ION
SO
UTC
OM
ES
Cap
ella
and
C
apel
la,
1996
(RN
CC
)
VB
G
(329
ope
ratio
ns
perf
orm
ed o
n 32
8 pa
tient
s)
52 ±
9U
nspe
cifi e
dE
arly
com
plic
atio
ns:
0.3%
(1 c
ase)
Late
com
plic
atio
ns: 9
%
(29
case
s of l
eaka
ge, b
and
disl
ocat
ion
or
pulm
onar
y em
bolis
m)
Dea
ths:
1 d
ue to
pul
mon
ary
embo
lism
EW
L (%
)
BM
I (kg
/m2 )
30–4
2 m
onth
s 48
± 2
3 39
± 9
54–6
6 m
onth
s 47
± 2
3 40
± 9
VB
G-R
YG
B(6
23 o
pera
tions
on 5
60 p
atie
nts
(in 3
51 c
ases
, ga
stric
segm
ents
w
ere
com
plet
ely
sepa
rate
d)
52 ±
9E
arly
com
plic
atio
ns:
1% (6
cas
es)
Late
com
plic
atio
ns:
12%
(7
3 ca
ses o
f fi s
tula
s, ul
cers
and
stap
le-li
ne
disr
uptio
n)
Dea
ths:
0
30–4
2 m
onth
s 70
± 1
9 32
± 6
54–6
6 m
onth
s 62
± 1
7 34
± 6
87
TAB
LE F
-2
Com
pari
sons
bet
wee
n ve
rtic
al b
ande
d ga
stro
plas
ty a
nd o
ther
bar
iatr
ic-s
urge
ry t
echn
ique
s (c
ont’d
)
AUTH
OR
STY
PE O
F PR
OCE
DU
RE
BM
I OR
IN
ITIA
L W
EIG
HT
AOT
(MIN
)FO
LLO
W-U
P(M
ON
THS)
COM
PLIC
ATIO
NS
OU
TCO
MES
Hel
l et a
l.,20
00(P
CC
)
VB
G: 3
0LS
AG
B: 3
0RY
GB
: 30
46.9
± 9
.9 k
g/m
2
46.9
± 7
.8 k
g/m
2
45.2
± 8
.2 k
g/m
2
48 88 135
40.1
± 8
.339
.7 ±
7.6
60 ±
8.1
Agg
rava
tion
of c
linic
al st
atus
in
3%
of R
YG
B p
atie
nts
V
BG
LS
AG
B
RYG
B
(n)
(n)
(n)
EWL
0–24
%
1 1
025
–49%
12
13
2
50–7
4%
15
15
675
–100
%
2 1
22
Hel
l and
M
iller
,20
00
(PN
CC
)
VB
G: 1
01
46.9
± 9
.0 k
g/m
2
133.
7 ±
33.3
kg
Uns
pecifi e
d40
± 8
.3E
arly
com
plic
atio
ns:
- Wou
nd in
fect
ion:
3%
- Hem
atom
a: 1
%
Late
com
plic
atio
ns:
- Eso
phag
ogas
tric
obst
ruct
ion*
: 12%
- Eso
phag
itis:
2%
Con
vers
ions
or
re-o
pera
tions
: 2%
per
yea
r
EWL:
2 ye
ars (
n =
98):
61%
(40
kg)
5 ye
ars (
n =
15):
69%
(48
kg)
Impr
ovem
ent i
n co
-mor
bidi
ties:
B
efor
e A
fter
Type
2 d
iabe
tes
15%
0%
Hyp
erte
nsio
n 50
%
15%
LSA
GB
: 99
46.9
± 7
.8 k
g/m
2
133
± 22
.7 k
gU
nspe
cifi e
d39
.7 ±
7.6
Ear
ly c
ompl
icat
ions
:
- Wou
nd in
fect
ion:
1%
- Hem
atom
a: 1
%
Late
com
plic
atio
ns:
- Ste
nosi
s: 3
%- E
soph
agiti
s: 4
%
Con
vers
ions
or
re-o
pera
tions
:
- 1.8
–3%
per
yea
r
EWL:
2 ye
ars
59%
(46
kg) (
n =
97)
5 ye
ars
71%
(56
kg) (
n =
16)
Impr
ovem
ent i
n co
-mor
bidi
ties:
B
efor
e A
fter
Type
2 d
iabe
tes
15%
0%
Hyp
erte
nsio
n 50
%
15%
* C
ause
d by
the
pass
age
of th
e fo
od b
olus
in p
atie
nts w
ho a
re n
ot u
sed
to th
eir n
ew e
atin
g pa
ttern
s.
88
TAB
LE F
-3
Com
pari
sons
bet
wee
n R
oux-
en-Y
gas
tric
byp
ass
and
lapa
rosc
opic
adj
usta
ble
gast
ric
band
ing
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
WEI
GH
T AN
D
INIT
IAL
BM
I AO
T (M
IN)
ALO
S(M
EDIA
NIN
DAY
S)FO
LLO
W-U
PCO
MPL
ICAT
ION
SO
UTC
OM
ES
Bie
rtho
et a
l.,20
03(R
NC
C)
LRY
GB
: 456
(4
0.2
± 10
.5 y
ears
)W
eigh
t:13
5.4
± 26
.3 k
g(7
6–22
1)
BM
I:49
.4 ±
8.3
kg/
m2
(27–
77)
Uns
pecifi e
d3
± 0.
3 (2
–94)
3 m
onth
s (89
%)
6 m
onth
s (88
%)
12 m
onth
s (5
7%)
18 m
onth
s (3
7%)*
Maj
or in
tra-
oper
ativ
e co
mpl
icat
ions
: 2%
- Maj
or le
akag
e of
gas
troje
juna
l an
asto
mos
is: 1
- Rou
x lim
b to
o sh
ort:
3- E
nd-to
-end
ana
stom
osis
dev
ice
pulle
d th
roug
h st
omac
h: 1
- Nas
ogas
tric
tube
stap
led:
1
Ear
ly p
ost-
oper
ativ
e co
mpl
icat
ions
: 4.
2%- M
ajor
leak
age
of g
astro
jeju
nal
anas
tom
osis
: 6- D
eep
veno
us th
rom
bosi
s: 4
- Gas
tric
dila
tatio
n: 2
- Int
ra-a
bdom
inal
or i
ntes
tinal
bl
eedi
ng: 2
- Fis
tula
on
Rou
x lim
b: 1
- Pne
umon
ia: 1
- Par
esia
left
arm
: 1
Late
pos
t-op
erat
ive
com
plic
atio
ns:
8.1%
- Ste
nosi
s of g
astro
jeju
nal
anas
tom
osis
: 15
- Lap
aros
copi
c ch
olec
yste
ctom
y: 1
1- O
bstru
ctio
n in
retro
colic
tu
nnel
: 4- P
erfo
ratio
n : 3
- Pan
crea
titis
: 1- O
ther
: 3
Con
vers
ions
: 2%
Dea
ths:
0.4
4% (2
)
EWL
acco
rdin
g to
pre
-ope
rativ
e B
MI:
LR
YG
B
30–4
0 40
–50
50–6
06
mon
ths
55%
56
%
47%
12 m
onth
s 75
%
72%
57
%18
mon
ths
– 81
%
69%
LA
GB
6 m
onth
s 24
%
21%
18
%12
mon
ths
37%
32
%
26%
18 m
onth
s 41
%
40%
33
%
89
TAB
LE F
-3
Com
pari
sons
bet
wee
n R
oux-
en-Y
gas
tric
byp
ass
and
lapa
rosc
opic
adj
usta
ble
gast
ric
band
ing
(con
t’d)
AUTH
OR
SN
UM
BER
OF
PATI
ENTS
WEI
GH
T AN
D
INIT
IAL
BM
I AO
T (M
IN)
ALO
S(M
EDIA
NIN
DAY
S)FO
LLO
W-U
PCO
MPL
ICAT
ION
SO
UTC
OM
ES
Bie
rtho
et a
l.,(c
ont’d
)
LAG
B: 8
05(4
1.7
± 10
.9
year
s)
Wei
ght:
117
± 17
kg
(75–
224)
BM
I: 42
.2 ±
4.9
kg/
m2
(29–
64)
Uns
pecifi e
d5
± 2,
4(2
–22)
3 m
onth
s (97
%)
6 m
onth
s (97
%)
12 m
onth
s (97
%)
18 m
onth
s (97
%)*
Maj
or in
tra-
oper
ativ
e co
mpl
icat
ions
: 1.
3%- B
leed
ing:
4- L
iver
hem
atom
a: 4
- Eso
phag
eal p
erfo
ratio
n: 1
- Ble
edin
g fr
om g
astro
-epi
ploi
c ar
tery
: 1- G
as e
mbo
lism
: 1
Ear
ly p
ost-
oper
ativ
e co
mpl
icat
ions
: 1.
7%- P
neum
onia
: 7- P
ulm
onar
y em
bolis
m: 2
- Por
t hem
atom
a: 2
- Acu
te a
bdom
en: 1
- Por
t inf
ectio
n: 1
Late
pos
t-op
erat
ive
com
plic
atio
ns:
9.1%
- Rel
ated
to th
e ba
nd: 4
6- R
elat
ed to
the
port
or tu
be: 2
3- O
ther
: 4
Con
vers
ions
: 3.
0%
Dea
ths:
0%
* Th
e au
thor
s attr
ibut
e th
e di
ffere
nce
in su
rviv
al ra
tes a
t 18
mon
ths (
37%
and
97%
) to
the
gast
ric-b
ypas
s pro
cedu
re it
self
(whi
ch re
quire
s mor
e as
sess
men
ts, w
here
as th
e pa
tient
s com
e fr
om v
ario
us
regi
ons t
hat a
re so
met
imes
qui
te fa
r), w
hich
mak
es lo
ng-te
rm fo
llow
-up
diffi
cult.
90
APPE
ND
IX G
M
ETA-
ANAL
YSIS
OF
THE
IMPA
CT O
F B
ARIA
TRIC
SU
RG
ERY
ON
O
BES
ITY
CO-M
OR
BID
ITIE
STA
BLE
G-1
Impa
ct o
f ba
riat
ric
surg
ery
on o
besi
ty c
o-m
orbi
diti
es [
Buc
hwal
d et
al.,
200
4]
CO-M
OR
BID
ITIE
SG
RO
UPS
OF
SUR
GIC
ALLY
TR
EATE
D P
ATIE
NTS
*
GAS
TRO
PLAS
TY(M
AIN
LY V
BG
)
RYG
B(IN
CLU
DIN
G
VAR
IAN
TS)
GAS
TRIC
BAN
DIN
G
(AD
JUST
ABLE
AN
D
NO
N-A
DJU
STAB
LE)
BPD
(INCL
UD
ING
VA
RIA
NTS
)
Type
2 d
iabe
tes
Res
olut
ion:
1417
(76.
8) o
f 184
6 st
udy
patie
nts
(63
grou
ps)
Mea
n an
d 95
% C
I: 76
.8%
(70.
7–82
.9%
)p
< 0.
01 (t
est f
or h
eter
ogen
eity
)
Impr
ovem
ent o
r re
solu
tion:
414
(85.
4%) o
f 485
(3
0 gr
oups
) M
ean:
86.
0% (7
8.4–
93.7
%)
p <
0.01
New
or
wor
se:
12 (0
.7%
) of 1
835
(10
grou
ps)
45 (6
8.2%
) of 6
6 (1
1 gr
oups
)
71.6
% (5
5.1–
88.2
%)
p <
0.10
34 (8
9.5%
) of 3
8 (8
gro
ups)
90.8
% (7
6.2–
10.%
)p
< 0.
10
1 (6
.7%
) of 1
5 (1
gro
up)
829
(83.
8%) o
f 989
(2
6 gr
oups
)
83.7
% (7
7.3–
90.1
%)
p <
0.01
115
(90.
6%) o
f 127
(6
gro
ups)
93.2
% (7
9.3–
100%
)p
< 0.
01
6 (0
.5%
) of 1
142
(3 g
roup
s)
98 (4
7.8%
) of 2
05
(9 g
roup
s)
47.9
% (2
9.1–
66.7
%)
p <
0.01
174
(80.
2%) o
f 217
(9
gro
ups)
80.8
% (7
2.2–
89.4
%)
p <
0.10
1 (0
.2%
) of 5
21
(2 g
roup
s)
282
(97.
9%) o
f 288
(6
3 gr
oups
)
98.9
% (9
6.8–
100%
)N
S
89 (8
8.1%
) of 1
01
(6 g
roup
s)76
.7%
(42.
2–10
0%)
p <
0.01
No
data
pro
vide
d
Hyp
erlip
idem
ia†
Hyp
erch
oles
tero
lem
ia†
Hyp
ertri
glyc
erid
emia
†
Impr
ovem
ent:
846
(83.
0%) o
f 101
9(2
3 gr
oups
)M
ean:
79.
3% (6
8.2–
90.5
%)
p <
0.01
1777
(86.
6%) o
f 205
1(1
4 gr
oups
)M
ean:
71.
3% (5
5.5–
87.0
%)
p <
0.01
912
(92.
8%) o
f 983
(1
1 gr
oups
)M
ean:
82.
4% (7
1.1–
93.7
%)
p <
0.01
174
(80.
9%) o
f 215
(7
gro
ups)
73.6
% (6
0.8–
86.3
%)
p <
0.01
40 (3
9.2%
) of 1
02
(4 g
roup
s)38
.4%
(25.
4–51
.4%
)N
S
15 (7
1.4%
) of 2
1 (2
gro
ups)
72.4
% (5
3.4–
91.4
%)
NS
117
(93.
6%) o
f 125
(6
gro
ups)
96.9
% (9
3.6–
100%
)N
S
417
(95.
0%) o
f 439
(5
gro
ups)
94.9
% (9
0.7–
99.1
%)
p <
0.10
255
(94.
1%) o
f 271
(4
gro
ups)
91.2
% (8
3.6–
98.8
%)
p <
0.01
303
(71.
1%) o
f 426
(6
gro
ups)
58.9
% (2
8.2–
89.6
%)
p <
0.01
18 (7
8.3%
) of 2
3 (1
gro
up)
78.0
% (6
1.1–
94.9
%)
NS
10 (7
6.9%
) of 1
3 (1
gro
up)
77.0
% (5
4.1–
99.9
%)
NS
199
(99.
5%) o
f 200
(3
gro
ups)
99.1
% (9
7.6–
100%
)N
S
1234
(99.
7%) o
f 123
8(3
gro
ups)
87.2
% (5
9.2–
100%
)p
< 0.
10
588
(100
%) o
f 588
(2
gro
ups)
100%
(98.
1–10
0%)
NS
91
TAB
LE G
-1
Impa
ct o
f ba
riat
ric
surg
ery
on o
besi
ty c
o-m
orbi
diti
es [
Buc
hwal
d et
al.,
200
4] (c
ont’d
)
CO-M
OR
BID
ITIE
SG
RO
UPS
OF
SUR
GIC
ALLY
TR
EATE
D P
ATIE
NTS
*G
ASTR
OPL
ASTY
(MAI
NLY
VB
G)
RYG
B(IN
CLU
DIN
G
VAR
IAN
TS)
GAS
TRIC
BAN
DIN
G
(AD
JUST
ABLE
AN
D
NO
N-A
DJU
STAB
LE)
BPD
(INCL
UD
ING
VAR
IAN
TS)
Hyp
erte
nsio
nR
esol
utio
n:
3151
(65.
6%) o
f 480
5(6
7 gr
oups
)M
ean:
61.
7% (5
5.6–
67.8
%)
p <
0.01
Impr
ovem
ent o
r re
solu
tion:
1752
(81.
8%) o
f 214
1(4
3 gr
oups
)M
ean:
78.
5% (7
0.8–
98.1
%)
p <
0.01
277
(72.
5%) o
f 382
(2
0 gr
oups
)69
.0%
(59.
1–79
.0%
)p
< 0.
01
83 (8
0.6%
) of 1
03
(12
grou
ps)
85.4
% (7
4.1–
96.7
%)
p <
0.01
1594
(75.
4%) o
f 211
5(2
0 gr
oups
)67
.5%
(58.
4–76
.5%
)p
< 0.
01
379
(87.
1%) o
f 435
(1
1 gr
oups
)87
.2%
(78.
4–95
.9%
)p
< 0.
01
232
(38.
4%) o
f 604
(1
2 gr
oups
)43
.2%
(30.
4–55
.9%
)p
< 0.
01
490
(71.
5%) o
f 685
(1
0 gr
oups
)70
.8 %
(61.
9–79
.6%
)p
< 0.
01
629
(81.
3%) o
f 774
(7
gro
ups)
83.4
% (7
3.2–
93.6
%)
p <
0.10
718
(91.
8%) o
f 782
(7
gro
ups)
75.1
% (4
4.7–
100%
)p
< 0.
01
Slee
p ap
nea
Res
olut
ion:
1051
(87.
9%) o
f 119
5(3
8 gr
oups
)M
ean:
85.
7% (7
9.2–
92.2
%)
p <
0.01
Impr
ovem
ent o
r re
solu
tion:
585
(80.
6%) o
f 726
(2
4 gr
oups
)M
ean:
83.
6% (7
1.8–
95.4
%)
p <
0.01
33 (7
6.7%
) of 4
3 (1
0 gr
oups
)78
.2%
(53.
6–10
0%)
p <
0.01
25 (8
9.3%
) of 2
8 (6
gro
ups)
90.7
% (7
8.5–
100%
)N
S
776
(86.
6%) o
f 896
(1
3 gr
oups
)80
.4%
(68.
4–92
.3%
)p
< 0.
01
167
(94.
9%) o
f 176
(6
gro
ups)
94.8
% (9
1.5–
98.1
%)
NS
53 (9
4.6%
) of 5
6 (5
gro
ups)
95.0
% (8
8.8–
100%
)N
S
10 (5
5.6%
) of 1
8 (3
gro
ups)
68.0
% (2
6.2–
100%
)p
< 0.
10
157
(95.
2%) o
f 165
(6
gro
ups)
91.9
% (8
1.9–
100%
)p
< 0.
01
144
(86.
7%) o
f 166
(6
gro
ups)
71.2
% (3
4.5–
100%
)p
< 0.
01
* In
clud
es R
oux-
en-Y
gas
tric
bypa
ss, g
astri
c ba
ndin
g, g
astro
plas
ty, b
iliop
ancr
eatic
div
ersi
on a
nd m
ixed
gro
ups,
alon
g w
ith o
ther
less
com
mon
pro
cedu
res (
bilia
ry-in
test
inal
byp
ass,
ileog
astro
stom
y,
jeju
no-il
eal b
ypas
s and
uns
pecifi e
d ba
riatri
c su
rger
y).
† Fi
gure
s in
clud
e pa
tient
s w
ho h
ave
elim
inat
ed o
r dec
reas
ed th
eir m
edic
atio
n do
ses,
patie
nts
repo
rted
to h
ave
impr
oved
lipi
d pa
ram
eter
s, an
d al
l pat
ient
s w
ho h
ave
been
clin
ical
ly o
r bio
logi
cally
ev
alua
ted
for i
mpr
ovem
ent.
92
APPE
ND
IX H
D
ETAI
LED
OU
TCO
MES
OF
ECO
NO
MIC
STU
DIE
STA
BLE
H-1
Des
crip
tion
of
stud
ies
on b
aria
tric
sur
gery
wit
h an
eco
nom
ic a
naly
sis
AUTH
OR
S,
COU
NTR
YST
UD
Y D
ESCR
IPTI
ON
MAI
N C
LIN
ICAL
O
UTC
OM
ESM
AIN
ECO
NO
MIC
OU
TCO
MES
MET
HO
DO
LOG
Y D
ETAI
LS
Van
Gem
ert
et a
l., 1
999
Net
herla
nds
(mod
el)
Type
of m
odel
: ret
rosp
ectiv
e be
fore
–afte
r stu
dy c
ombi
ned
with
da
ta o
n th
e co
st o
f the
bur
den
of m
orbi
d ob
esity
in o
rder
to
calc
ulat
e a
cost
-effe
ctiv
enes
s rat
ioN
umbe
r of s
ubje
cts:
21
(con
secu
tive)
Leng
th o
f fol
low
-up:
24
mon
ths
Lost
to fo
llow
-up:
non
eSu
rgic
al te
chni
que:
VB
GB
MI:
47.2
2 ±
7.15
kg/
m2
BM
I: 30
.1 ±
7.1
kg/
m2
(at 1
yea
r)29
.17
± 6.
75 k
g/m
2 (a
t 2 y
ears
)
Δ 12
QA
LYs
Paid
em
ploy
men
t (be
fore
vs a
fter s
urge
ry):
19%
vs 4
8% (p
< 0
.05)
Sick
leav
e: 3
8% v
s 10%
(p <
0.0
5)C
ost o
f sur
gery
: US$
5,86
5C
ost o
f mor
bid
obes
ity: f
rom
US$
9,36
7 (p
reva
lenc
e: 0
.25%
) to
US$
8,30
4 (1
%)
Indi
rect
cos
ts o
f obe
sity
: US$
45,8
79Sa
ving
s: U
S$3,
928–
$4,0
04/Q
ALY
QA
LY: c
ombi
natio
n of
the
resu
lts o
f 3
QoL
que
stio
nnai
res a
nd e
stim
ated
nu
mbe
r of l
ife y
ears
gai
ned
(3.6
ba
sed
on p
ublis
hed
data
)B
urde
n of
obe
sity
: cos
t fi g
ures
from
a
natio
nal s
ourc
eD
isco
unt r
ate:
5%
Non
-incl
usio
n of
follo
w-u
p co
sts a
nd
patie
nts’
pers
onal
exp
ense
s
Sjös
tröm
et a
l.,19
95Sw
eden
(mod
el)
Type
of m
odel
: dat
a fr
om a
non
-ra
ndom
ized
stud
y co
mbi
ned
with
an
eco
nom
ic m
odel
ling
exer
cise
Leng
th o
f fol
low
-up:
24
mon
ths
Dire
ct c
osts
(SEK
) at 1
0 ye
ars p
er
100
patie
nts:
16.5
mill
ion
(sur
gica
l gro
up) v
s 14.
5 m
illio
n (c
ontro
l gro
up)
Prel
imin
ary
data
from
the
SOS
stud
y af
ter t
wo
year
s of f
ollo
w-u
pC
ompl
emen
tary
dat
a fr
om v
ario
us
sour
ces
Nar
bro
et a
l.,19
99
Swed
en(P
CC
)
Num
ber o
f sub
ject
s:
36
9 (S
= su
rger
y)
371
(C =
con
trols
)
Leng
th o
f fol
low
-up:
60
mon
ths (
48
post
-ope
rativ
ely)
Lost
to fo
llow
-up:
0 (d
ata
on si
ck
leav
e or
dis
abili
ty p
ensi
on)
Surg
ical
tech
niqu
e: G
B, V
BG
, AG
BB
MI:
41.6
kg/
m2 (
S); 4
1.0
kg/m
2 (C
)W
eigh
t: 11
9.6
kg (S
); 11
7.1
kg (C
)
Wei
ght l
oss:
S:
30.7
± 1
4.0
kg
(a
t 1 y
ear)
23
.5 ±
15.
3 kg
(at 4
yea
rs)
C:
1.1
± 7.
20 k
g
(at 1
yea
r)
+0.8
± 1
0.1
kg
(a
t 4 y
ears
)
Mea
n nu
mbe
r of s
ick-
leav
e da
ys w
ith
disa
bilit
y pe
nsio
n (S
vs C
):Ye
ar 1
: +50
% (p
< 0
.001
)Ye
ar 3
: –14
% (p
< 0
.01)
Year
4: N
S
Mea
n nu
mbe
r of d
ays o
f dis
abili
ty p
ensi
on
(S v
s C):
Year
1 :
+68%
(C) v
s 81%
(S) (
NS)
Year
s 3 a
nd 4
: S <
C (p
< 0
.05)
Thes
e be
nefi t
s are
sign
ifi ca
ntly
hig
her f
or
subj
ects
age
d 46
.7 a
nd o
lder
.
Med
ian
age:
46.
7 ye
ars
93
TAB
LE H
-1
Des
crip
tion
of
bari
atri
c-su
rger
y st
udie
s w
ith
an e
cono
mic
ana
lysi
s (c
ont’d
)
AUTH
OR
S,
COU
NTR
YST
UD
Y D
ESCR
IPTI
ON
MAI
N C
LIN
ICAL
OU
TCO
MES
MAI
N E
CON
OM
IC O
UTC
OM
ESM
ETH
OD
OLO
GY
DET
AILS
Ägr
en e
t al.,
2002
bSw
eden
(PC
C)
Num
ber o
f sub
ject
s:
510
(S =
surg
ery)
455
(C =
con
trols
)Le
ngth
of f
ollo
w-u
p: 7
2 m
onth
sLo
st to
follo
w-u
p: <
2%
Surg
ical
tech
niqu
e: G
B, V
BG
, AG
BB
MI:
41.8
± 4
.1 k
g/m
2 (S)
; 39
.9 ±
4.6
kg/
m2
(C)
Wei
ght l
oss a
t 6 y
ears
:S:
16.
2 %
± 1
1.6
C: –
0.8%
± 9
.6Pr
opor
tion
S vs
C:
Initi
ally
und
er d
rug
ther
apy:
CV
D: R
R =
0.7
7; p
< 0
.05)
Dia
bete
s: R
R =
0.7
1; p
< 0
.05)
Not
initi
ally
und
er d
rug
ther
apy:
CV
D: R
R =
0.8
0 (N
S)D
iabe
tes:
RR
= 0
.20
(p <
0.0
5)
Varia
tion
in m
edic
atio
n co
sts a
t 6 y
ears
for
subj
ects
initi
ally
und
er d
rug
ther
apy:
- CV
D: –
455
SEK
for w
eigh
t los
s > 1
5%
(p <
0.0
5; c
ompa
rison
in re
latio
n to
w
eigh
t los
s < 5
%)
- Dia
bete
s: –
673
SEK
for w
eigh
t los
s > 1
5%
(p <
0.0
5)
for a
ll su
bjec
ts a
nd b
oth
dise
ases
:- <
15%
: 439
–481
SEK
(77–
97%
)- ≥
15:
–20
SEK
(–7.
8)
Rel
ativ
e w
eigh
t los
s cla
ssifi
ed
acco
rdin
g to
4 c
ateg
orie
s:
< 5;
5–1
0%; 1
0–15
%; ≥
15%
Prop
ortio
n on
med
icat
ion:
dat
aga
ther
ed b
y qu
estio
nnai
re
Indi
vidu
al c
ost f
or e
ach
med
icat
ion,
taki
ng in
to a
ccou
nt
dosa
ge a
nd o
ffi ci
al p
rice
(bas
ed
on d
ecla
red
cons
umpt
ion
durin
g th
e 3
mon
ths p
rior t
o th
e as
sess
men
t)
Ägr
en e
t al.,
2002
aSw
eden
(PC
C)
Num
ber o
f sub
ject
s:
481
(S =
surg
ery)
481
(C =
con
trols
)Le
ngth
of f
ollo
w-u
p: 8
4 m
onth
sLo
st to
follo
w-u
p: 3
% (h
ospi
taliz
atio
n da
ta)
Surg
ical
tech
niqu
es: G
B, V
BG
, AG
BB
MI:
41.9
± 4
.2 k
g/m
2 (S)
; 40
.1 ±
5.0
kg/
m2
(C)
Wei
ght l
oss a
t 6 y
ears
:S:
16.
7% ±
1.8
(401
/481
subj
ects
)C
: –0.
9% ±
10.
1 (3
44/4
81 su
bjec
ts)H
ospi
taliz
atio
n ra
tes:
Year
0: 2
7.7%
(S);1
4.1%
(C)
Year
s 2 to
6: O
R ra
ngin
g fr
om 2
.70
(p <
0.0
001)
to 1
.32
(p =
0.1
1)Ye
ar 0
–6: O
R =
2.7
2 (9
5% C
I: 2.
06–3
.59)
Hos
pita
lizat
ion
days
:Ye
ar 0
: 0.9
(S);
0.7
(C) (
p =
0.42
)Ye
ar 1
(sur
gica
l tre
atm
ent):
9.4
Year
s 1 to
6: 6
.2 (S
); 0.
9 (C
) for
com
mon
po
st-o
p co
nditi
ons:
7.8
(S);
6 (C
) for
oth
er
cond
ition
s
Dis
coun
ted
cost
s in
US$
:Su
rger
y: 4
,207
(S)
Com
mon
pos
t-op
cond
ition
s: 2
,579
(S);
363
(C) (
p <
0.00
1)O
ther
con
ditio
ns: 2
,747
(S);
2,17
7 (C
) (p
= 0
.17)
Tota
l: 9,
553
(S);
2,54
0 (C
) (p
< 0.
001)
Diff
eren
ce b
etw
een
com
mon
he
alth
con
ditio
ns re
late
d to
ba
riatri
c su
rger
y an
d ot
her n
on-
rela
ted
heal
th c
ondi
tions
Dis
coun
t rat
e: 3
%
Mar
tin e
t al.,
1995
Uni
ted
Stat
es(R
NC
C)
Num
ber o
f sub
ject
s:
201
(S =
surg
ery)
161
(C =
con
trols
, med
ical
trea
tmen
t)Le
ngth
of f
ollo
w-u
p: 2
4–84
mon
ths
Lost
to fo
llow
: 75%
(S);
79%
(C) a
t the
5-
year
follo
w-u
p ex
amin
atio
nSu
rgic
al te
chni
que:
RY
GB
BM
I: 49
.3 k
g/m
2 (S)
; 41.
2 kg
/m2 (C
)
Perc
enta
ge o
f pat
ient
s who
redu
ced
thei
r exc
ess w
eigh
t by
at le
ast
one
third
and
who
mai
ntai
ned
this
ou
tcom
e:89
% (S
); 21
% (C
) (at
5 y
ears
)
Uni
t cos
t: U
S$3,
000
vs U
S$24
,000
for
med
ical
trea
tmen
t (C
) and
surg
ical
trea
tmen
t (S
) res
pect
ivel
y
Tota
l cos
ts p
er p
ound
lost
: hig
her v
alue
in th
e co
ntro
l gro
up (fi
gur
es n
ot p
rovi
ded
and
no
stat
istic
al te
st)
Fixe
d un
it co
st c
alcu
late
d ac
cord
ing
to e
stim
ated
fees
ch
arge
d fo
r exa
min
atio
ns a
nd
test
s, m
edic
al tr
eatm
ent a
nd
hosp
italiz
atio
n w
ith su
rgic
al
treat
men
t (ex
clud
ing
treat
men
t fo
r com
plic
atio
ns)
94
TAB
LE H
-1
Des
crip
tion
of
bari
atri
c-su
rger
y st
udie
s w
ith
an e
cono
mic
ana
lysi
s (c
ont’d
)
AUTH
OR
S,
COU
NTR
YST
UD
Y D
ESCR
IPTI
ON
MAI
N C
LIN
ICAL
OU
TCO
MES
MAI
N E
CON
OM
IC O
UTC
OM
ESM
ETH
OD
OLO
GY
DET
AILS
Ngu
yen
et a
l.,20
01U
nite
d St
ates
(RC
)
Num
ber o
f sub
ject
s:
LV
BG
: 79
V
BG
: 76
Leng
th o
f fol
low
-up:
24
mon
ths
Lost
to fo
llow
-up:
num
ber
varie
s acc
ordi
ng to
out
com
e m
easu
re a
nd ti
me
of
mea
sure
men
tSu
rgic
al te
chni
ques
: LV
BG
, V
BG
BM
I:
LVB
G: 4
7.6
± 4.
7 kg
/m2
V
BG
: 48.
4 ±
5.4
kg/m
2
Exce
ss w
eigh
t los
s:LV
BG
: 68
± 15
%V
BG
: 62
± 14
(p =
0.0
7)In
tens
ive
care
:LV
BG
: 6 (7
.6%
)V
BG
: 16
(21.
1%) (
p =
0.03
)R
e-op
erat
ions
: LV
BG
: 6 (7
.6%
)V
BG
: 5 (6
.6%
) (N
S)
Com
pari
sons
bet
wee
n LV
BG
and
VB
G:
Med
ian
hosp
ital l
engt
h of
stay
: 3 v
s 4 d
ays
(p <
0.0
01)
Ope
ratin
g tim
e: 2
25 v
s 195
min
utes
(p <
0.0
01)
Rec
over
y tim
e:-
norm
al a
ctiv
ities
: 8.4
vs 1
7.7
days
(p <
0.0
01)
- w
ork:
32.
2 vs
46.
1 da
ys (p
= 0
.02)
SF-3
6 sc
ores
: som
e di
ffere
nces
at 1
mon
th,
but s
imila
r at 3
mon
ths
BA
RO
S sc
ores
: 97%
vs 8
2% (f
rom
goo
d to
ex
celle
nt)
Tota
l dire
ct c
osts
: US$
7,47
8 vs
US$
7,44
0 (N
S)
- su
rger
y: U
S$4,
922
vs U
S$3,
591
(p <
0.0
1)-
othe
r ser
vice
s: U
S$2,
519
vs U
S$3,
742
(p =
0.0
2)To
tal c
osts
: US$
14,0
87 v
s US$
14,0
98 (N
S)
Inte
ntio
n-to
-trea
t prin
cipl
e
Qua
lity
of li
fe (S
F-36
and
BA
RO
S qu
estio
nnai
res)
Dire
ct su
rger
y co
sts d
istin
guis
hed
from
oth
er h
ospi
tal-s
ervi
ce c
osts
Dire
ct se
rvic
e-ut
iliza
tion
cost
s de
rived
from
a d
ecis
ion
supp
ort
syst
em d
atab
ase
Tota
l cos
ts: i
nclu
de g
ener
al se
rvic
es
Chu
a an
d M
endi
ola,
1995
U
nite
d St
ates
(RN
CC
)
Num
ber o
f sub
ject
s:
a) L
VB
G (1
993–
1994
): 11
b) G
B (1
987–
1992
): 11
c) G
B (1
986)
: 11
Leng
th o
f fol
low
-up:
no
follo
w-
upSu
rgic
al te
chni
ques
: VB
G, G
BB
MI:
a)
48.
2 ±
5.3
kg/m
2
b) 4
6.5
± 6.
79 k
g/m
2
c) u
nspe
cifi e
d
Wei
ght l
oss:
uns
pecifi e
dC
ompl
icat
ions
:
a) n
one
b)
3 su
bjec
ts
c) u
nspe
cifi e
d
Aver
age
leng
th o
f sta
y:
a)
3.9
days
b) 7
.4 d
ays (
afte
r exc
lusi
on o
f one
at
ypic
al c
ase)
c)
7.2
days
Ope
ratin
g tim
e:
a)
202
min
utes
b)
105
min
utes
c)
unsp
ecifi
edH
ospi
tal c
harg
es:
a)
US$
12,8
00b)
US$
16,6
69 (a
fter e
xclu
sion
of o
ne
atyp
ical
cas
e)c)
U
S$14
,131
No
clin
ical
info
rmat
ion
on 1
986
surg
ical
gro
up (c
)
Hos
pita
l cha
rges
: no
deta
ils
prov
ided
on
com
pone
nts o
r dat
a-co
llect
ion
met
hod
Cha
rges
for g
roup
s b) a
nd c
) wer
e co
nver
ted
into
cur
rent
cas
h va
lues
.
95
TAB
LE H
-1
Des
crip
tion
of
bari
atri
c-su
rger
y st
udie
s w
ith
an e
cono
mic
ana
lysi
s (c
ont’d
)
AUTH
OR
S,
COU
NTR
YST
UD
Y D
ESCR
IPTI
ON
MAI
N C
LIN
ICAL
OU
TCO
MES
MAI
N E
CON
OM
IC O
UTC
OM
ESM
ETH
OD
OLO
GY
DET
AILS
Ang
us e
t al.,
2003
Uni
ted
Stat
es(R
NC
C)
Num
ber o
f sub
ject
s:
a) R
YG
B (2
001)
: 122
b) L
RYG
B (2
001)
: 11
Leng
th o
f fol
low
-up:
30
days
Surg
ical
tech
niqu
es: R
YG
B, L
RYG
BB
MI:
a)
55.
32 ±
5.6
4 kg
/m2
b)
49.
54 ±
6.5
1 kg
/m2 (
p <
0.05
)
Blo
od lo
ss (m
l):a)
305
± 8
3b)
125
± 6
8 (p
< 0
.001
)In
tens
ive
care
(no.
of p
atie
nts)
:a)
3
b) 0
Com
plic
atio
ns a
t 30
days
:a)
13%
(16)
b) 4
5% (5
)
Aver
age
leng
th o
f sta
y (d
ays)
: a)
4.8
± 1
.2b)
3.5
± 0
.69
(p <
0.0
01)
Ope
ratin
g tim
e (m
inut
es):
a) 1
55 ±
48
b) 2
85 ±
50
(p <
0.0
01)
Dire
ct c
osts
(US$
):a)
3,1
79 ±
101
b) 4
,180
± 3
82 (p
< 0
.001
)In
dire
ct c
osts
(US$
):a)
2,1
37 ±
285
b) 1
,792
± 2
63 (p
< 0
.001
)To
tal c
osts
(US$
): a)
7,8
94 ±
264
b) 6
,350
± 7
5 (p
< 0
.001
)
Inte
ntio
n-to
-trea
t prin
cipl
e
Cos
ts b
ased
on
hosp
ital c
harg
es
Dire
ct c
osts
: sur
gery
cos
ts
(ope
ratin
g tim
e, o
pera
ting-
room
su
pplie
s, po
st-a
nest
hesi
a ca
re,
excl
udin
g re
usab
le la
paro
scop
ic
equi
pmen
t); h
ospi
tal-s
ervi
ce
cost
s (nu
rsin
g, p
harm
aceu
tical
, ra
diol
ogy)
Indi
rect
cos
ts: h
ouse
keep
ing,
ov
erhe
ad a
nd e
mpl
oyee
ben
efi ts
)
Gal
lagh
er
et a
l.,
2003
Uni
ted
Stat
es(R
)
Num
ber o
f sub
ject
s: 2
5
Tim
e pe
riod
unde
r stu
dy:
3 ye
ars f
or p
re-o
pera
tive
adm
issi
ons,
and
up to
3 y
ears
for
post
-ope
rativ
e ad
mis
sion
s (m
ean
follo
w-u
p of
18
mon
ths)
Surg
ical
tech
niqu
e: R
YG
B
BM
I: 52
± 2
kg/
m2
Age
: 52
± 2
year
s
Sex:
72%
(M);
28%
(F)
Inte
nsiv
e ca
re (d
ays)
: 1.4
± 0
.5
Re-
adm
issi
ons:
6 p
atie
nts f
or
12 a
dmis
sion
s
Cos
t of p
re-o
pera
tive
out
-pat
ient
car
e: 8
± 0
.5 d
ays
Cos
t of p
re-o
pera
tive
out-p
atie
nt c
are
5,47
6 ±
682
(US$
)A
dmis
sion
s 13
,211
± 6
,906
Hom
e-he
alth
dev
ices
1,
383
± 34
9
TOTA
L 10
,558
± 2
,470
Peri-
oper
ativ
e co
sts :
- S
urge
ry
1,90
0 (fi
xe)
- Hos
pita
lizat
ion
7,07
6 ±
497
TOTA
L 8,
976
± 49
7Po
st-o
pera
tive
outp
atie
nt c
are
≈ 1
,800
± 3
00A
dmis
sion
s ≈
5,00
0 ±
1600
TOTA
L 2,
840
± 62
2
Excl
usio
n of
adm
issi
ons a
nd
cons
ulta
tions
not
rela
ted
to
obes
ity o
r to
its c
o-m
orbi
ditie
sSe
rvic
es in
clud
e ho
me-
heal
th
devi
ces.
Cos
t and
oth
er d
ata
are
base
d on
in
form
atio
n pr
ovid
ed b
y th
e ad
min
istra
tive
supp
ort s
ervi
ces o
f th
e su
rger
y de
partm
ent.
Fixe
d un
it co
sts w
ere
assi
gned
to
the
diffe
rent
type
s of s
ervi
ces.
96
TAB
LE H
-1
Des
crip
tion
of
bari
atri
c-su
rger
y st
udie
s w
ith
an e
cono
mic
ana
lysi
s (c
ont’d
)
AUTH
OR
S,
COU
NTR
YST
UD
Y D
ESCR
IPTI
ON
MAI
N C
LIN
ICAL
OU
TCO
MES
MAI
N E
CON
OM
IC O
UTC
OM
ESM
ETH
OD
OLO
GY
DET
AILS
Potte
iger
et
al.,
2004
Uni
ted
Stat
es(R
)
Num
ber o
f sub
ject
s: 5
3 (b
ut 2
exc
lude
d)Fo
llow
-up:
pre
-ope
rativ
e as
sess
men
t, th
en a
t 3-m
onth
and
9-m
onth
in
terv
als p
ost-o
pera
tivel
ySu
rgic
al te
chni
ques
: RY
GB
(30)
and
LR
YG
B (2
1)B
MI g
reat
er th
an 4
0 kg
/m2
Age
: 45
year
s (27
–63)
Sex:
16
mal
es a
nd 3
7 fe
mal
es
Co-
mor
bidi
ties
- D
iabe
tes +
hyp
erte
nsio
nin
34%
of p
atie
nts (
18/5
3)-
Dia
bete
s: 5
5.7%
(29/
53)
- H
yper
tens
ion:
44.
3% (2
4/53
)
Mor
talit
y: 0
Com
plic
atio
ns: 2
5.5%
(13
patie
nts)
Impr
ovem
ent o
r res
olut
ion:
D
iabe
tes:
92%
(47)
Hyp
erte
nsio
n: 7
8% (4
0)
Aver
age
post
-ope
rativ
e st
ay (d
ays)
: 3.
29 ±
20.
7 (5
1/53
)R
educ
tion
in n
umbe
r (n)
and
cos
t (U
S$) o
f m
edic
atio
ns:
Dia
bete
sB
efor
eA
fter
n1.
12 ±
1.1
50.
12 ±
0.4
8U
S$13
6.9
± 20
6.6
26.6
± 1
07.1
Hyp
erte
nsio
n n1.
32 ±
1.2
50.
44 ±
0.6
4U
S$50
.4 ±
59.
9115
.97
± 24
.6To
tal
n2.
44 ±
1.8
60.
56 ±
0.8
1U
S$18
7.24
± 2
37.4
142
.53
± 11
6.6
(for
all
devi
atio
ns: p
< 0
.001
)
Cos
t per
pat
ient
(US$
):H
ospi
tal c
osts
: 10
,508
± 3
,704
Clin
icia
n co
sts:
4,16
8 ±
1,59
5(in
clud
ing
surg
eon)
: 2,
340
± 1,
031
Tota
l cos
ts:
14,6
76 ±
5,2
99
Cos
ts b
ased
on
actu
al c
osts
to
the
med
ical
cen
tre
Hos
pita
l cos
ts: h
ospi
tal s
uppl
ies
and
serv
ices
Clin
icia
n co
sts:
fees
for s
urge
ry,
med
ical
trea
tmen
t and
co
nsul
tatio
ns
Excl
usio
ns: o
ne o
f the
subj
ects
ex
clud
ed (S
ubje
ct 1
) had
slee
p ap
nea
and
chro
nic
obst
ruct
ive
pulm
onar
y di
seas
e an
d w
as
hosp
italiz
ed fo
r 50
days
for
recu
rren
t pne
umon
ia a
nd
pulm
onar
y re
habi
litat
ion;
th
e ot
her e
xclu
ded
subj
ect
(Sub
ject
2) w
as h
ospi
taliz
ed
for 1
36 d
ays f
or a
n an
asto
mot
ic le
ak c
ompl
icat
ed
by a
cute
resp
irato
ry d
istre
ss
synd
rom
e.To
tal c
osts
(US$
):Su
bjec
t 1: 1
35,3
91Su
bjec
t 2: 3
33,8
56
Mon
k et
al.,
20
04U
nite
d St
ates
(R)
Num
ber o
f sub
ject
s: 8
7 of
100
pat
ient
s us
ing
pres
crip
tion
med
icat
ion
pre-
oper
ativ
ely
(but
23
lost
to fo
llow
-up)
Leng
th o
f fol
low
-up:
16
mon
ths
(6–6
0)
Surg
ical
tech
niqu
e: R
YG
BM
ean
BM
I: 57
kg/
m2 (
36.6
–85.
4)A
ge: 4
4 ye
ars (
27–6
4)Se
x: 1
3 m
ales
and
51
fem
ales
Co-
mor
bidi
ties (
64 p
atie
nts)
:- S
leep
apn
ea (3
8)- T
ype
2 di
abet
es (2
3)- H
yper
tens
ion
(31)
- Gas
tro-e
soph
agea
l refl
ux
dise
ase
(21)
- Ast
hma:
23
Slee
p ap
nea:
of t
he 2
5 pa
tient
s who
pr
e-op
erat
ivel
y re
quire
d ai
rway
m
anag
emen
t at n
ight
, onl
y 2
cont
inue
d to
do
so.
Type
2 d
iabe
tes:
21
patie
nts d
isco
ntin
ued
thei
r med
icat
ion
and
2 ot
hers
dec
reas
ed
thei
r dos
ages
by
50%
.H
yper
tens
ion:
11
disc
ontin
ued
thei
r m
edic
atio
n an
d 15
dec
reas
ed th
eir
dosa
ges.
Gas
tro-e
soph
agea
l refl
ux
dise
ase:
5
of th
e 8
patie
nts d
isco
ntin
ued
thei
r med
icat
ions
.A
sthm
a: 1
8 pa
tient
s red
uced
thei
r dos
ages
or
dis
cont
inue
d at
leas
t one
of t
heir
med
icat
ions
.
Med
ical
exp
ense
s ($U
S) re
late
d to
co-
mor
bidi
ties:
Dis
ease
B
efor
e A
fter
Savi
ngs
Slee
p ap
nea
225.
00
18.0
0 20
7.00
Dia
bete
s 75
.60
4.60
71
.00
Hyp
erte
nsio
n 26
.59
10.3
3 16
.26
GER
D
81.7
3 47
.86
33.8
7A
sthm
a 52
.45
32.7
9 19
.66
All
317.
30
135.
20
182.
10
Wilc
oxon
test
: p <
0.0
1
Cos
t of m
edic
atio
n al
one
Excl
usio
n of
pat
ient
s who
di
d no
t use
med
icat
ion
pre-
oper
ativ
ely
Inco
mpl
ete
data
on
23 p
atie
nts
97
TAB
LE H
-1
Des
crip
tion
of
bari
atri
c-su
rger
y st
udie
s w
ith
an e
cono
mic
ana
lysi
s (c
ont’d
)
AUTH
OR
S,
COU
NTR
YST
UD
Y D
ESCR
IPTI
ON
MAI
N C
LIN
ICAL
OU
TCO
MES
MAI
N E
CON
OM
IC O
UTC
OM
ESM
ETH
OD
OLO
GY
DET
AILS
Snow
et a
l.,20
04U
nite
d St
ates
(R)
Num
ber o
f sub
ject
s: 7
8 ov
er th
e ag
e of
54
who
had
bee
n fo
llow
ed a
m
inim
um o
f 6 m
onth
s (ex
tract
ed
from
a d
atab
ase
of 1
060
patie
nts)
Leng
th o
f fol
low
-up:
up
to 2
yea
rs
Surg
ical
tech
niqu
e: L
RYG
B
BM
I: 48
kg/
m2 (
36–7
0)
Wei
ght:
136
kg (8
5–22
6)
Age
: 60
year
s (55
–75)
Sex:
17
mal
es a
nd 6
1 fe
mal
es
Num
ber o
f pat
ient
s with
co
-mor
bidi
ties r
equi
ring
med
icat
ion:
- H
yper
tens
ion
and/
or C
VD
: 51
- Ty
pe 2
dia
bete
s: 2
8-
Pulm
onar
y in
suffi
cien
cy: 2
4-
Ost
eoar
thrit
is: 3
2-
Anx
iety
or d
epre
ssio
n: 3
4-
Hyp
erlip
idem
ia: 2
0-
GER
D: 2
1-
Urin
ary
inco
ntin
ence
: 3-
All
co-m
orbi
ditie
s: 7
0
Intra
-ope
rativ
e m
orta
lity:
0
Dea
ths (
betw
een
30 a
nd
180
days
pos
t-ope
rativ
ely)
: 3
Con
vers
ions
: 10
(12.
4%)
Num
ber o
f co-
mor
bidi
ties
(per
pat
ient
ave
rage
): 3
vs 1
fo
r pat
ient
s < 5
5 ye
ars
Mea
n bl
ood
loss
: 132
cc
Aver
age
oper
atin
g tim
e:
124
min
utes
Impr
ovem
ent o
r res
olut
ion:
Dia
bete
s: 9
2% (4
7)
Hyp
erte
nsio
n: 7
8% (4
0)
Cha
nges
in n
umbe
r of p
resc
riptio
ns (R
x), c
osts
and
sa
ving
s (U
S$) p
er p
atie
nt p
er m
onth
:
Bef
ore
Afte
r6
mos
1 yr
2 yr
s*N
o. o
f Rx
324
112
113
38N
o. o
f pa
tient
s4.
21.
41.
41.
5
Cos
t36
911
911
910
5Sa
ving
s25
025
026
4R
educ
tion
68%
68%
72%
Ann
ualiz
ed c
osts
and
savi
ngs (
US$
) per
pr
escr
iptio
n:In
terv
alC
osts
Savi
ngs
Pre-
op34
5,05
6.40
6 m
os p
ost-o
p11
1,47
7.60
233,
578.
801
year
pos
t-op
111,
057.
1223
3,98
1.28
2 ye
ars p
ost-o
p97
,961
.76
247,
150.
80
Tota
l cos
ts o
f car
e: $
US6
31,0
00 (b
ased
on
aver
age
cost
of $
US8
,090
per
inte
rval
)
Num
ber o
f pre
scrip
tions
: dat
a w
ere
obta
ined
from
pat
ient
s an
d ch
ecke
d ag
ains
t med
ical
re
cord
s.
Pres
crip
tion
cost
s: c
ost o
f a
30-d
ay su
pply
of e
ach
drug
w
as o
btai
ned
from
thre
e re
tail
sour
ces a
nd a
vera
ged.
Aver
age
savi
ngs:
diff
eren
ce
betw
een
the
aver
age
cost
at e
ach
stud
y in
terv
al a
nd th
e av
erag
e pr
e-op
erat
ive
cost
.
Cos
t per
stud
y in
terv
al: a
vera
ge
cost
of s
urge
ry a
nd m
edic
al
fees
, con
sulta
tions
and
hos
pita
l se
rvic
es o
btai
ned
from
the
data
bank
of a
sing
le in
sure
r (in
surin
g 75
% o
f the
pat
ient
s in
the
stud
y da
taba
se).
* 25
of t
he in
itial
78
patie
nts
98
APPENDIX I METROPOLITAN LIFE INSURANCE COMPANY TABLES
(Prepared from the 1979 Build Study)TABLE I-1
Metropolitan Life Insurance Company table (women with medium frames)
Height (cm) Weight (kg) Height (cm) Weight (kg)
148 49.6–55.1 166 58.1–64.5149 50.0–55.5 167 58.7–65.0150 50.3–55.9 168 59.2–65.5151 50.7–56.4 169 59.7–66.1152 51.1–57.0 170 60.2–66.6153 51.5–57.5 171 60.7–67.1154 51.9–58.0 172 61.3–67.6155 52.2–58.6 173 61.8–68.2156 52.7–59.1 174 62.3–68.7157 53.2–59.6 175 62.8–69.2158 53.8–60.2 176 63.4–69.8159 54.3–60.7 177 64.0–70.4160 54.9–61.2 178 64.5–70.9161 55.4–61.7 179 65.1–71.4162 55.9–62.3 180 65.6–71.9163 56.4–62.8 181 66.1–72.5164 57.0–63.4 182 66.6–73.0165 57.5–63.9 183 67.1–73.5
TABLE I-2
Metropolitan Life Insurance Company table (men with medium frames)
Height (cm)
158159 160 161 162 163164 165 166 167 168 169 170 171 172 173 174 175
Weight (kg)
5 9 . 6 – 6 4 . 2 5 9 . 9 – 6 4 . 5 6 0 . 3 – 6 4 . 9 6 0 . 6 – 6 5 . 2 6 1 . 0 – 6 5 . 6 6 1 . 3 – 6 6 . 0 6 1 . 7 – 6 6 . 5 6 2 . 1 – 6 7 . 0 6 2 . 4 – 6 7 . 6 6 2 . 8 – 6 8 . 2 6 3 . 2 – 6 8 . 7 6 3 . 8 – 6 9 . 3 6 4 . 3 – 6 9 . 8 6 4 . 8 – 7 0 . 3 6 5 . 4 – 7 0 . 8 6 5 . 9 – 7 1 . 4 6 6 . 4 – 7 2 . 4 66.9–72.4
Height (cm)
176 177 178 179 180 181 182 183 184 185 186187188189190191192193
Weight (kg)
67.5–73.0 68.1–73.5 68.6–74.0 69.2–74.6 69.7–75.1 70.2–75.8 70.7–76.5 71.3–77.2 71.8–77.9 72.4–78.6 73.0–79.3 73.7–80.0 74.4–80.7 74.9–81.5 75.4–82.2 76.1–83.0 76.8–83.9 77.6–84.8
* For people aged 25 to 69, these fi gures assume that women are wearing shoes with 2.5-cm heels and outdoor clothing weighing 1.4 kg and that men are wearing outdoor clothing weighing 2.3 kg.
99
APPENDIX J BAROS SCORING KEY(Bariatric Analysis and Reporting Outcome System)
Source: Oria and Moorehead, 1998.
WEIGHT LOSS
% OF EXCESS MEDICAL CONDITIONS QUALITY OF LIFE QUESTIONNAIRE
Weight gain Aggravated 1. SELF-ESTEEM
(–1) (–1)
–1.0 –0.50 0 +0.50 +1 0–24 Unchanged 2. PHYSICAL
(0) (0)
–0.50 –0.25 0 +0.25 +0.50 25–49 Improved 3. SOCIAL
(1) (1)
–0.50 –0.25 0 +0.25 +0.50 50–74 One major resolved 4. LABOUR
(2) Others improved (2)
–0.50 –0.25 0 +0.25 +0.50 5. SEXUAL
75–100 All major resolved (3) Others improved (3)
–0.50 –0.25 0 +0.25 +0.50
Subtotal: Subtotal: Subtotal:
Complications: minor: deduct 0.2 points major: deduct 1 pointRe-operation: deduct 1 point
TOTAL SCORE
Outcome groupsScoring keyFailure 1 point or lessFair > 1 to 3 pointsGood > 3 to 5 pointsVery good > 5 to 7 pointsExcellent > 7 to 9 points
100
REFERENCES
Abu-Abeid S and Szold A. Results and complications of laparoscopic adjustable gastric banding: An early and intermediate experience. Obes Surg 1999;9(2):188–90.
Abu-Abeid S, Gavert N, Klausner JM, Szold A. Bariatric surgery in adolescence. J Pediatr Surg 2003;38(9):1379–82.
Aetna. Obesity surgery. Clinical Policy Bulletin No. 0157. Avalailable at: http://www.aetna.com/cpb/data/CPBA0157.html (accessed on May 18, 2004).
Agence Nationale d’Accréditation et d’Évaluation en Santé (ANAES). Chirurgie de l’obésité morbide de l’adulte. Paris: ANAES; 2001.
Ågren G, Narbro K, Jonsson E, Naslund I, Sjöström L, Peltonen M. Cost of in-patient care over 7 years among surgically and conventionally treated obese patients. Obes Res 2002a;10(12):1276–83.
Ågren G, Narbro K, Naslund I, Sjöström L, Peltonen M. Long-term effects of weight loss on pharmaceutical costs in obese subjects―A report from the SOS intervention study. Int J Obes Relat Metab Disord 2002b;26(2):184–92.
Alle JL, Poortman M, Chelala E. Five years’ experience with laparoscopic vertical banded gastroplasty. Obes Surg 1998;8:373–4.
Allen J. Surgical Internet at a glance: Bariatric surgery. Am J Surg 2000;179(1):33.
Allgood P. Surgical interventions for morbid obesity. In: Foxcroft DR, Muthu V, ed. STEER: Succinct and Timely Evaluated Evidence Reviews 2001;1(8). Wessex Institute for Health Research and Development, University of Southampton. Available at: http://www.wihrd.soton. ac.uk/projx/signpost/steers/STEER_2001(18).pdf.
American College of Surgeons (ACS). Recommendations for facilities performing bariatric surgery. Bull Am Coll Surg 2000;85(9):20–3. Available at: http://www.facs.org/fellows_info/statements/st-34.html (accessed on October 29, 2004).
American Society for Bariatric Surgery (ASBS). Rationale for the surgical treatment of morbid obesity. Gainsville, FL: ASBS; 2001. Available at: http://www.asbs.org/html/rationale/rationale.html (accessed on October 23, 2004).
Angrisani L, Di Lorenzo N, Favretti F, Furbetta F, Iuppa A, Doldi SB, et al. The Italian Group for Lap-Band: Predictive value of initial body mass index for weight loss after 5 years of follow-up. Surg Endo 2004a;18:1524–7.
Angrisani L, Favretti F, Furbetta F, Iuppa A, Doldi SB, Paganelli M, et al. Italian Group for Lap-Band System: Results of multicenter study on patients with BMI < or = 35 kg/m2. Obes Surg 2004b;14(3):415–8.
101
Angrisani L, Furbetta F, Doldi SB, Basso N, Lucchese M, Giacomelli M, et al. Lap Band adjustable gastric banding system: The Italian experience with 1863 patients operated on 6 years. Surg Endosc 2003;17(3):409–12.
Angrisani L, Alkilani M, Basso N, Belvederesi N, Campanile F, Capizzi FD, et al. Laparoscopic Italian experience with the Lap-Band. Obes Surg 2001;11(3):307–10.
Angus LD, Cottam DR, Gorecki PJ, Mourello R, Ortega RE, Adamski J. DRG, costs and reimbursement following Roux-en-Y gastric bypass: An economic appraisal. Obes Surg 2003;13(4):591–5.
Asp N-G, Bjorntorp P, Britton M, Carlsson P, Kjellstrom T, Marcus C, et al. Obesity―Problems and interventions: A systematic review. Stockholm, Sweden: Swedish Council on Tech nology Assessment in Health Care (SBU); 2002.
Bajardi G, Ricevuto G, Mastrandrea G, Branca M, Rinaudo G, Cali F, et al. Surgical treatment of morbid obesity with biliopancreatic diversion and gastric banding: Report on an 8-year experience involving 235 cases. Ann Chir 2000;125(2):155–62.
Ballesta-Lopez C, Poves I, Cabrera M, Almeida JA, Macias G. Learning curve for laparoscopic Roux-en-Y gastric bypass with totally hand-sewn anastomosis: Analysis of 600 conse cutive patients. Surg Endosc 2005;19(4):519–24.
Balsiger BM, Poggio JL, Mai J, Kelly KA, Sarr MG. Ten and more years after vertical banded gastroplasty as primary operation for morbid obesity. J Gastrointest Surg 2000;4(6):598–605.
Baltasar A, Bou R, Miro J, Bengochea M, Serra C, Perez N. Laparoscopic biliopancreatic diversion with duodenal switch: Technique and initial experience. Obes Surg 2002;12(2):245–8.
Baltasar M, Bou R, Cipagauta LA, Marcote E, Herrera GR, Chisbert JJ. Hybrid bariatric surgery: Bilio-pancreatic diversion and duodenal switch―Preliminary experience. Obes Surg 1995;5(4):419–23.
Belachew M and Zimmermann JM. Evolution of a paradigm for laparoscopic adjustable gastric banding. Am J Surg 2002;184(6B):21S–25S.
Belachew M, Belva PH, Desaive C. Long-term results of laparoscopic adjustable gastric banding for the treatment of morbid obesity. Obes Surg 2002;12(4):564–8.
Belachew M, Legrand M, Vincent V, Lismonde M, Le Docte N, Deschamps V. Laparoscopic adjustable gastric banding. World J Surg 1998;22(9):955–63.
Biertho L, Steffen R, Ricklin T, Horber FF, Pomp A, Inabnet WB, et al. Laparoscopic gastric bypass versus laparoscopic adjustable gastric banding: A comparative study of 1,200 cases. J Am Coll Surg 2003;197(4):536–44.
Birmingham CL, Muller JL, Palepu A, Spinelli JJ, Anis AH. The cost of obesity in Canada. CMAJ 1999;160(4):483–8.
Biron S, Hould FS, Lebel S, Marceau S, Lescelleur O, Simard P, Marceau P. Twenty years of biliopancreatic diversion: What is the goal of the surgery? Obes Surg 2004;14(2):160–4.
102
Blanco-Engert R, Weiner S, Pomhoff I, Matkowitz R, Weiner RA. Outcome after laparoscopic adjustable gastric banding, using the Lap-Band and the Heliogast band: A prospective randomized study. Obes Surg 2003;13(5):776–9.
Bleier JI, Krupnick AS, Kreisel D, Song HK, Rosato EF, Williams NN. Hand-assisted laparo scopic vertical banded gastroplasty: Early results. Surg Endosc 2000;14(10):902–7.
Blue Cross and Blue Shield Association Technology Evaluation Center (BCBS). Special report: The relationship between weight loss and changes in morbidity following bariatric surgery for morbid obesity. BCBS 2003a;18(9):23 p.
Blue Cross and Blue Shield Association Technology Evaluation Center (BCBS). Newer techniques in bariatric surgery for morbid obesity. BCBS 2003b;18(10):51 p.
Brody F. Minimally invasive surgery for morbid obesity. Cleve Clin J Med 2004;71(4):289, 293, 296–8.
Buchwald H. Overview of bariatric surgery. J Am Coll Surg 2002;194(3):367–75.
Buchwald H and Williams SE. Bariatric surgery worldwide 2003. Obes Surg 2004;14(9):1157–64.
Buchwald H, Avidor Y, Braunwald E, Jensen MD, Pories W, Fahrbach K, Schoelles K. Bariatric surgery: A systematic review and meta-analysis. JAMA 2004;292(14):1724–37.
Caisse Nationale de l’Assurance Maladie des Travailleurs Salariés (CNAMTS). Chirurgie digestive de l’obésité – Résultats enquête nationale. CNAMTS, Direction du Service Médical, 2004. Available at: http://www.ameli.fr/135/DOC/1321/article_pdf.html# (accessed on May 26, 2004).
Capella JF and Capella RF. An assessment of vertical banded gastroplasty-Roux-en-Y gastric bypass for the treatment of morbid obesity. Am J Surg 2002;183(2):117–23.
Capella JF and Capella RF. The weight reduction operation of choice: Vertical banded gastro plasty or gastric bypass? Am J Surg 1996;171(1):74–9.
Champion JK, Hunt T, DeLisle N. Laparoscopic vertical banded gastroplasty and Roux-en-Y gastric bypass in morbid obesity. Obes Surg 1999;9(2):123.
Chapman A and Kiroff G. Laparoscopic adjustable gastric banding for treatment of obesity: ASERNIP-S Report No. 9. Adelaide, Australia: ASERNIP-S; 2000.
Chapman A, Kiroff G, Game P, Foster B, O’Brien P, Ham J, Maddern, G. Laparoscopic adjustable gastric banding in the treatment of obesity: A systematic literature review. Surgery 2004;135(3):326–51.
Chapman A, Game P, O’Brien P, Maddern G, Kiroff G, Foster B, Ham J. Systematic review of laparoscopic adjustable gastric banding for the treatment of obesity: Update and re-appraisal. ASERNIP-S Report No. 31. 2nd ed. Adelaide, Australia: ASERNIP-S; 2002.
103
Chen J and McGregor M. The gastric banding procedure. An evaluation. Technology Assessment Unit of the McGill University Health Centre, Montréal, Canada; 2004. Available at: http://upload.mcgill.ca/tau/Gastric_Banding_FINAL_Apr27.pdf (accessed on May 6, 2004).
Christou NV, Sampalis JS, Liberman M, Look D, Auger S, McLean AP, MacLean LD. Surgery decreases long-term mortality, morbidity, and health care use in morbidly obese patients. Ann Surg 2004;240(3):416–24.
Chua TY and Mendiola RM. Laparoscopic vertical banded gastroplasty: The Milwaukee experience. Obes Surg 1995;5(1):77–80.
Cigaina V. Long-term follow-up of gastric stimulation for obesity: The Mestre 8-year experience. Obes Surg 2004;14 Suppl 1:S14–22
Clegg AJ, Colquitt J, Sidhu MK, Royle P, Loveman E, Walker A. The clinical effectiveness and cost-effectiveness of surgery for people with morbid obesity: A systematic review and economic evaluation. Health Technol Assess 2002;6(12):1–153.
Colman R. The cost of obesity in Nova Scotia. Prepared for Cancer Care Nova Scotia and presented to the Healthy Weights Conference, Dartmouth, Nova Scotia; 2000.
Colman R and Dodds C. Cost of obesity in Quebec. Genuine Progress Index: Measuring Sustainable Development. GPI Atlantic; November 2000. Available at: http://www.gpiatlantic.org/pdf/health/obesity/que-obesity.pdf (accessed on August 15, 2005).
Colwell J. New ACP guidelines target obesity management. ACP Observer; April 2005. Available at: http://www.acponline.org/journals/news/apr05/obesity.htm.
Comité de suivi de la chirurgie bariatrique. Pour répondre aux besoins pressants d’une population souffrant d’obésité morbide : un programme suprarégional de chirurgie bariatrique. Québec : Hôpital Laval, Institut universitaire de cardiologie et de pneumologie (cf. résolution CA-01-03-[09]-01); 2001.
Conseil d’évaluation des technologies de la santé (CETS). Le traitement chirurgical de l’obésité morbide. Montréal: CETS; 1998.
Cooney RN, Haluck RS, Ku J, Bass T, MacLeod J, Brunner H, Miller CA. Analysis of cost outliers after gastric bypass surgery: What can we learn? Obes Surg 2003;13(1):29–36.
Cottam DR, Mattar SG, Schauer PR. Laparoscopic era of operations for morbid obesity. Arch Surg 2003;138(4):367–75.
Courcoulas A, Perry Y, Buenaventura P, Luketich J. Comparing the outcomes after laparoscopic versus open gastric bypass: A matched paired analysis. Obes Surg 2003a;13(3):341–6.
Courcoulas A, Schuchert M, Gatti G, Luketich J. The relationship of surgeon and hospital volume to outcome after gastric bypass surgery in Pennsylvania: A 3-year summary. Surgery 2003b;134(4):613–23.
104
Craig BM and Tseng DS. Cost-effectiveness of gastric bypass for severe obesity. Am J Med 2002;113(6):491–8
Dargent J. Laparoscopic adjustable gastric banding: Lessons from the fi rst 500 patients in a single institution. Obes Surg 1999;9(5):446–52.
Daviglus ML, Liu K, Yan LL, Pirzada A, Manheim L, Manning W, et al. Relation of body mass index in young adulthood and middle age to Medicare expenditures in older age. JAMA 2004;292(22):2743–9.
Dávila-Cervantes A, Borunda D, Dominguez-Cherit G, Gamino R, Vargas-Vorackova F, Gonzalez-Barrenco J, Herrera MF. Open versus laparoscopic vertical banded gastroplasty: A randomized controlled double blind trial. Obes Surg 2002;12(6):812–8.
De Luca M, Segato G, Busetto L, Favretti F, Aigner F, Weiss H, et al. Progress in implantable gastric stimulation: Summary of results of the European multi-center study. Obes Surg 2004;14 Suppl 1:S33–9.
De Wit LT, Mathus-Vliegen L, Hey C, Rademaker B, Gouma DJ, Obertop H. Open versus laparoscopic adjustable silicone gastric banding: A prospective randomized trial for treatment of morbid obesity. Ann Surg 1999;230(6):800–7.
DeMaria EJ. Is gastric bypass superior for the surgical treatment of obesity compared with malabsorptive procedures? J Gastrointest Surg 2004;8(4):401–3.
DeMaria EJ. Laparoscopic adjustable silicone gastric banding: Complications. J Laparoendosc Adv Surg Tech A 2003;13(4):271–7.
DeMaria EJ, Schweitzer MA, Kellum JM, Meador J, Wolfe L, Sugerman HJ. Hand-assisted laparoscopic gastric bypass does not improve outcome and increases costs when compared to open gastric bypass for the surgical treatment of obesity. Surg Endosc 2002a;16(10):1452–5.
DeMaria EJ, Sugerman HJ, Kellum JM, Meador JG, Wolfe LG. Results of 281 consecutive total laparoscopic Roux-en-Y gastric bypasses to treat morbid obesity. Ann Surg 2002b;235(5):640–7.
DeMaria EJ, Sugerman HJ, Meador JG, Doty JM, Kellum JM, Wolfe L, et al. High failure rate after laparoscopic adjustable silicone gastric banding for treatment of morbid obesity. Ann Surg 2001;233(6):809–18.
DeMeester TR, Fuchs KH, Ball CS, Albertucci M, Smyrk TC, Marcus JN. Experimental and clinical results with proximal end-to-end duodenojejunostomy for pathologic duodeno gastric refl ux. Ann Surg 1987;206(4):414–26.
Dixon JB and O’Brien PE. Health outcomes of severely obese type 2 diabetic subjects 1 year after laparoscopic adjustable gastric banding. Diabetes Care 2002;25(2):358–63.
Dixon JB, Chapman L, O’Brien P. Marked improvement in asthma after Lap-Band surgery for morbid obesity. Obes Surg 1999;9(4):385–9.
105
Doherty C, Maher JW, Heitshusen DS. Long-term data indicate a progressive loss in effi cacy of adjustable silicone gastric banding for the surgical treatment of morbid obesity. Surgery 2002;132(4):724–8.
Doherty C, Maher JW, Heitshusen DS. An interval report on prospective investigation of adjustable silicone gastric banding devices for the treatment of severe obesity. Eur J Gastroenterol Hepatol 1999;11(2):115–9.
Dresel A, Kuhn JA, McCarthy TM. Laparoscopic Roux-en-Y gastric bypass in morbidly obese and super morbidly obese patients. Am J Surg 2004;187(2):230–2.
Egino Sasiain E, Rico Iturrioz R, Guttierez Ibarluzea I. El tratamiento quirurgico de la obesidad morbida. Vitoria-Gasteiz, Spain: OSTEBA IR-00-06; 2000.
Favretti F, Cadiere GB, Segato G, Busetto L, Loffredo A, Vertruyen M, et al. Bariatric Analysis and Reporting Outcome System (BAROS) applied to laparoscopic gastric banding patients. Obes Surg 1998;8(5):500–4.
Fielding GA, Rhodes M, Nathanson LK. Laparoscopic gastric banding for morbid obesity―Surgical outcome in 335 cases. Surg Endosc 1999;13(6):550–4.
Fisher BL and Schauer P. Medical and surgical options in the treatment of severe obesity. Am J Surg 2002;184(6B):9S–16S.
Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA 2005;293(15):1861–7.
Fobi MA, Lee H, Holness R, Cabinda D. Gastric bypass operation for obesity. World J Surg 1998;22(9):925–35.
Frezza EE, Ikramuddin S, Gourash W, Rakitt T, Kingston A, Luketich J, Schauer P. Symptomatic improvement in gastroesophageal refl ux disease (GERD) following laparoscopic Roux-en-Y gastric bypass. Surg Endosc 2002;16(7):1027–31.
Fried M. Open and laparoscopic non-adjustable gastric banding. In: Deitel M and Cowan Jr GSM, ed. Update: Surgery for the morbidly obese patients. Toronto: FD-Communications; 2000:333–50.
Fried M, Kasalicky M, Melechovsky D, Kormanova K. Current status of non-adjustable gastric banding. Obes Surg 2002;12(3):395–8.
Furbetta F and Gambinotti G. Functional gastric bypass with an adjustable gastric band. Obes Surg 2002;12(6):876–80.
Gagner M and Rogula T. Laparoscopic reoperative sleeve gastrectomy for poor weight loss after biliopancreatic diversion with duodenal switch. Obes Surg 2003;13(4):649–54.
Gagner M, Garcia-Ruiz A, Arca MJ, Heniford BT. Laparoscopic isolated gastric bypass for morbid obesity. Surg Endosc 1999;13(Suppl 1):S19.
106
Gallagher SF, Banasiak M, Gonzalvo JP, Paoli DP, Alwood J, Morris D, et al. The impact of bariatric surgery on the Veterans Administration healthcare system: A cost analysis. Obes Surg 2003;13(2):245–8.
Greenstein RJ, Martin L, MacDonald KJ, Chapman W, McIntyre R, Wittgrove AC, et al. The Lap-Band®
system as surgical therapy for morbid obesity: Intermediate results of the USA, multicenter, prospective study. Surg Endosc 1999;13(Suppl 1):S1.
Gustavsson S and Westling A. Laparoscopic adjustable gastric banding: Complications and side effects responsible for the poor long-term outcome. Semin Laparosc Surg 2002;9(2):115–24.
Hailey D and Harstall C. Decisions on the status of health technologies. Edmonton, AB: Alberta Heritage Foundation for Medical Research (AHFMR); 2001.
Hall JC, Watts JM, O’Brien PE, Dunstan RE, Walsh JF, Slavotinek AH, Elmslie RG. Gastric surgery for morbid obesity: The Adelaide Study. Ann Surg 1990;211(4):419–27.
Hell E and Miller K. [Criteria for selection of patients for bariatric surgery]. Zentralbl Chir 2002;127(12):1035–7 (Article in German whose original title is: Kriterien zur Selektion von Patienten fur bariatrische Eingriffe).
Hell E and Miller K. Comparison of vertical banded gastroplasty and adjustable silicone gastric banding. In: Deitel M and Cowan Jr GSM, ed. Update: Surgery for the morbidly obese patients. Toronto: FD-Communications; 2000: 379–86.
Hell E, Miller KA, Moorehead MK, Norman S. Evaluation of health status and quality of life after bariatric surgery: Comparison of standard Roux-en-Y gastric bypass, vertical banded gastroplasty and laparoscopic adjustable silicone gastric banding. Obes Surg 2000;10(3):214–9.
Hess DS and Hess DW. Biliopancreatic diversion with a duodenal switch. Obes Surg 1998;8(3):267–82.
Higa KD, Boone KB, Ho T, Davies OG. Laparoscopic Roux-en-Y gastric bypass for morbid obesity: Technique and preliminary results of our fi rst 400 patients. Arch Surg 2000;135(9):1029–34.
Institut de la statistique du Québec. Enquête sociale et de santé 1998. 2nd ed. Québec: ISQ; 2001.
International Bariatric Surgery Registry (IBSR). IBSR 2000-2001 winter pooled report. Iowa City: IBSR; 2001: 19.
Karlsson J, Sjöström L, Sullivan M. Swedish obese subjects (SOS)―An intervention study of obesity. Two-year follow-up of health-related quality of life (HRQL) and eating behavior after gastric surgery for severe obesity. Int J Obes Relat Metab Disord 1998;22(2):113–26.
Kim WW, Gagner M, Kini S, Inabnet WB, Quinn T, Herron D, Pomp A. Laparoscopic vs. open biliopancreatic diversion with duodenal switch: A comparative study. J Gastrointest Surg 2003;7(4):552–7.
107
Kreitz K and Rovito PF. Laparoscopic Roux-en-Y gastric bypass in the “megaobese”. Arch Surg 2003;138(7):707–10.
Kushner RF. Pharmacological management. In: Kushner RF. Roadmaps for clinical practice: Case studies in disease prevention and health promotion—Assessment and management of adult obesity; a primer for physicians. Chapter 6. Chicago, Ill: American Medical Association; 2003.
Li Z, Maglione M, Tu W, Mojica W, Arterburn D, Shugarman LR, et al. Meta-analysis: Pharmacologic treatment of obesity. Ann Intern Med 2005:142(7):532–46.
Luján JA, Frutos MD, Hernandez Q, Liron R, Cuenca JR, Valero G, Parrilla P. Laparoscopic versus open gastric bypass in the treatment of morbid obesity: A randomized prospective study. Ann Surg 2004;239(4):433–7.
MacLean LD, Rhode BM, Nohr CW. Late outcome of isolated gastric bypass. Ann Surg 2000; 231(4):524–8.
Maggard MA, Shugarman LR, Suttorp M, Maglione M, Sugarman HJ, Livingston EH, et al. Meta-analysis: Surgical treatment of obesity. Ann Intern Med 2005;142(7):547–59.
Marceau P and Biron S, ed. Le traitement chirurgical de l’obésité. Sillery, Québec: Septentrion; 1993.
Marceau P, Hould FS, Simard S, Lebel S, Bourque RA, Potvin M, Biron S. Biliopancreatic diversion with duodenal switch. World J Surg 1998;22(9):947–54.
Marinari GM, Murelli F, Camerini G, Papadia F, Carlini F, Stabilini C, et al. A 15-year evalu ation of biliopancreatic diversion according to the Bariatric Analysis Reporting Outcome System (BAROS). Obes Surg 2004;14(3):325–8.
Martin LF, White S, Lindstrom W Jr. Cost-benefi t analysis for the treatment of severe obesity. World J Surg 1998;22(9):1008–17.
Martin LF, Tan TL, Horn JR, Bixler EO, Kauffman GL, Becker DA, Hunter SM. Comparison of the costs associated with medical and surgical treatment of obesity. Surgery 1995;118 (4):599–607.
Mason EE. Vertical gastroplasty. ISBR Newsletter 2002;17(1):1–2. Available at: http://aboutplastic.surgery.uiowa.edu/ibsr/wspring02.htm.
Mason EE. Surgery in the obese. Lancet 2003;361(9374):2001–2.
Mason EE, Tang S, Renquist KE, Barnes DT, Cullen JJ, Doherty C, Maher JW. A decade of change in obesity surgery. National Bariatric Surgery Registry (NBSR) Contributors. Obes Surg 1997;7(3):189–97.
Mason EE, Maher JW, Scott DH, Rodriguez EM, Doherty C. Ten years of vertical banded gastroplasty for severe obesity. Probl Gen Surg 1992;9(2):280–9.
108
McTigue KM, Harris R, Hemphill B, Lux L, Sutton S, Bunton AJ, Lohr KN. Screening and interventions for obesity in adults: Summary of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 2003;139(11):933–49.
Medical Advisory Secretariat (MAS). Bariatric surgery: Health technology literature review. Toronto: MAS, Ontario Ministry of Health and Long Term Care; 2005. Available at: http://www.health.gov.on.ca/english/providers/program/mas/reviews/docs/baria_0105.pdf
Medical Services Advisory Committee (MSAC). Laparoscopic adjustable gastric banding for morbid obesity. Canberra, ACT; Australian Department of Health and Aging; 2003. Available at: http://www.msac.gov.au/pdfs/reports/msacref14.pdf.
Memon MA and Fitzgibbons RJ Jr. Hand-assisted laparoscopic surgery (HALS): A useful tech nique for complex laparoscopic abdominal procedures. J Laparoendosc Adv Surg Tech A 1998;8(3):143–50.
Miller K and Hell E. Laparoscopic adjustable gastric banding: A prospective 4-year follow-up study. Obes Surg 1999;9(2):183–7.
Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States, 2000 [correction published in JAMA 2005;293(3):293–4, 298]. JAMA 2004;291(10):1238–45.
Monk JS Jr, Dia Nagib N, Stehr W. Pharmaceutical savings after gastric bypass surgery. Obes Surg 2004;14(1):13–5.
Mohr CJ, Nadzam GS, Curet MJ. Totally robotic Roux-en-Y gastric bypass. Arch Surg 2005;140(8):779–86.
Mouïel J, Benchetrit S, Bertrand JC, Cady J, Champault G, Descottes B, et al. Recommandations pour la bonne pratique de la chirurgie de l’obésité. Available at: http://www.obesite-info.com/recommandations.htm (also : J Coelio-Chir June 2004;50:96).
Msika S. La chirurgie de l’obésité morbide chez l’adulte. 1. Effi cacité clinique des différents procédés chirurgicaux. J Chir (Paris) 2002;139(4):194–204.
Naitoh T, Gagner M, Garcia-Ruiz A, Heniford BT, Ise H, Matsuno S. Hand-assisted laparoscopic digestive surgery provides safety and tactile sensation for malignancy or obesity. Surg Endosc 1999;13(2):157–60.
Narbro K, Ågren G, Jonsson E, Naslund I, Sjöström L, Peltonen M. Pharmaceutical costs in obese individuals—Comparison with a randomly selected population sample and long-term changes after conventional and surgical treatment: The SOS Intervention Study. Arch Intern Med 2002;(18)162:2061–9.
Narbro K, Ågren G, Jonsson E, Larsson B, Naslund I, Wedel H, Sjostrom L. Sick leave and disability pension before and after treatment for obesity: A report from the Swedish Obese Subjects (SOS) study. Int J Obes Relat Metab Disord 1999;23(6):619–24.
Näslund E, Hellström PM, Kral JG. The gut and food intake: An update for surgeons. J Gastrointest Surg 2001;5(5):556–67.
109
Näslund E, Freedman J, Lagergren J, Stockeld D, Granstrom L. Three-year results of laparo scopic vertical banded gastroplasty. Obes Surg 1999;9(4):369–73.
National Center for Health Statistics (NCHS). Health. United States, 2004 with chartbook on trends in the health of Americans. Hyattsville, Maryland: NCHS; 2004. Available at: http://www.cdc.gov/nchs/hus.htm.
National Health and Medical Research Council (NHMRC). Clinical practice guidelines for the management of overweight and obesity in adults. Canberra, Australia; 2003.
National Heart, Lung, and Blood Institute (NHLBI). Clinical guidelines on the identifi cation, evaluation, and treatment of overweight and obesity in adults. The evidence report. NIH Publication No. 98–4083. Bethesda: National Institute of Health, NHLBI; 1998.
National Institute for Clinical Excellence (NICE). Guide to the methods of technology appraisal. London, UK: NICE; 2004. Available at: http://www.nice.org.uk/pdf/TAP_Methods.pdf (accessed on May 20, 2004).
National Institute for Clinical Excellence (NICE). Guidance on the use of surgery to aid weight reduction for people with morbid obesity (Technology Appraisal Guidance no. 46). London, UK: NICE; 2002. Available at: http://www.nice.org.uk/pdf/Fullguidance-PDF-morbid.pdf.
Nehoda H, Hourmont K, Sauper T, Mittermair R, Lanthaler M, Aigner F, Weiss H. Laparoscopic gastric banding in older patients. Arch Surg 2001;136(10):1171–6.
Nguyen NT, Goldman C, Rosenquist CJ, Arango A, Cole CJ, Lee SJ, Wolfe BM. Laparoscopic versus open gastric bypass: A randomized study of outcomes, quality of life, and costs. Ann Surg 2001;234(3):279–91.
O‘Brien PE, Brown WA, Smith A, McMurrick PJ, Stephens M. Prospective study of a laparoscopically placed, adjustable gastric band in the treatment of morbid obesity. Br J Surg 1999; 86(1):113–8.
Oliak D, Ballantyne GH, Weber P, Wasielewski A, Davies RJ, Schmidt HJ. Laparoscopic Roux-en-Y gastric bypass: Defi ning the learning curve. Surg Endosc 2003;17(3):405–8.
Oria HE and Moorehead MK. Bariatric Analysis and Reporting Outcome System (BAROS). Obes Surg 1998;8(5):487–99.
Paiva D, Bernardes L, Suretti L. Laparoscopic biliopancreatic diversion for the treatment of morbid obesity: Initial experience. Obes Surg 2001;11(5):619–22.
Pandolfi no JE, Krishnamoorthy B, Lee TJ. Gastrointestinal complications of obesity surgery. MedGenMed 2004;6(2):15.
Papasavas PK, Hayetian FD, Caushaj PF, Landreneau RJ, Maurer J, Keenan RJ, et al. Outcome analysis of laparoscopic Roux-en-Y gastric bypass for morbid obesity: The fi rst 116 cases. Surg Endosc 2002;16(12):1653–7.
110
Podnos YD, Jimenez JC, Wilson SE, Stevens CM, Nguyen NT. Complications after laparoscopic gastric bypass: A review of 3464 cases. Arch Surg 2003;138(9):957–61.
Pories WJ, Swanson MS, MacDonald KG, Long SB, Morris PG, Brown BM, et al. Who would have thought it? An operation proves to be the most effective therapy for adult-onset diabetes mellitus. Ann Surg 1995;222(3):339–52.
Potteiger CE, Paragi PR, Inverso NA, Still C, Reed MJ, Strodel W 3rd, et al. Bariatric surgery: Shedding the monetary weight of prescription costs in the managed care arena. Obes Surg 2004;14(6):725–30.
Rabkin RA. Distal gastric bypass/duodenal switch procedure, Roux-en-Y gastric bypass and biliopancreatic diversion in a community practice. Obes Surg 1998;8(1):53–9.
Regan JP, Inabnet WB, Gagner M, Pomp A. Early experience with two-stage laparoscopic Roux-en-Y gastric bypass as an alternative in the super-super obese patient. Obes Surg 2003;13(6):861–4.
Ren CJ, Patterson E, Gagner M. Early results of laparoscopic biliopancreatic diversion with duodenal switch: A case series of 40 consecutive patients. Obes Surg 2000;10(6):514–24.
Reinhold RB. Critical analysis of long term weight loss following gastric bypass. Surg Gynecol Obstet 1982;155(3):385–94.
Rico Iturrioz R, Diez del Val I, Salcedo Fernandez F, Martinez Blazquez C, Mazagatos Uriarte B, Arrizabalaga Abasolo JJ, et al. Tratamiento quirurgico de la obesidad morbida. Vitoria-Gasteiz, Spain: OSTEBA; 2003.
Rubenstein RB. Laparoscopic adjustable gastric banding at a US center with up to 3-year follow-up. Obes Surg 2002;12(3):380–4.
Salval M, Allietta R, Bocchia P, et al. Laparoscopic Mason-MacLean vertical banded gastroplasty (LVBG) in the treatment of morbid obesity. Results in 87 patients with 6-18 months follow-up. J Ceolio-Chir 1999;29:77.
Sampalis John S, Lieberman M, Auger S, Christou NV. The impact of weight reduction surgery on health-care costs in morbidly obese patients. Obes Surg 2004;14(7):939–47.
Schauer PR, Burguera B, Ikramuddin S, Cottam D, Gourash W, Hamad G, et al. Effect of laparoscopic Roux-en Y gastric bypass on type 2 diabetes mellitus. Ann Surg 2003a;238(4):467–85.
Schauer P, Ikramuddin S, Hamad G, Gourash W. The learning curve for laparoscopic Roux-en-Y gastric bypass is 100 cases. Surg Endosc 2003b;17(2):212–5.
Schauer PR, Ikramuddin S, Gourash W, Ramanathan R, Luketich J. Outcomes after laparoscopic Roux-en-Y gastric bypass for morbid obesity. Ann Surg 2000;232(4):515–29.
111
Scopinaro N, Marinari GM, Camerini G. Laparoscopic standard biliopancreatic diversion: Tech nique and preliminary results. Obes Surg 2002;12(3):362–5.
Scopinaro N, Adami GF, Marinari GM, Traverso E, Papadia F, Camerini G. Biliopancreatic diversion: Two decades of experience. In: Deitel M and Cowan Jr GSM, ed. Update: Surgery for the morbidly obese patients. Toronto: FD-Communications; 2000: 227–58.
Scottish Executive Health Department. Review of bariatric surgical services in Scotland: (Weight loss surgery for adults who are severely obese). A report by a working group of the Scot tish Medical and Scientifi c Advisory Committee. Edinburgh: Scottish Executive; 2004. Available at: http://www.scotland.gov.uk/Resource/Doc/35596/0012569.pdf (accessed on August 22, 2005).
Shekelle PG, Morton SC, Maglione MA, Suttorp M, Tu W, Li Z, et al. Pharmacological and surgical treatment of obesity. Evidence Report/Technology Assessment No. 103. Rockville, MD: Agency for Healthcare Research and Quality; 2004.
Sjöström L, Lindroos AK, Peltonen M, Torgerson J, Bouchard C, Carlsson B, et al. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. NEJM 2004;351(26):2683–93.
Sjöström CD, Peltonen M, Sjöström L. Blood pressure and pulse pressure during long-term weight loss in the obese: The Swedish Obese Subjects (SOS) Intervention Study. Obes Res 2001;9(3):188–95.
Sjöström CD, Lissner L, Wedel H, Sjöström L. Reduction in incidence of diabetes, hypertension and lipid disturbances after intentional weight loss induced by bariatric surgery: The SOS Intervention Study. Obes Res 1999;7(5):477–84.
Sjöström L, Narbro K, Sjöström D. Costs and benefi ts when treating obesity. Int J Obes Relat Metab Disord 1995;19(Suppl 6):S9–12.
Snow LL, Weinstein LS, Hannon JK, Lane DR, Ringold FG, Hansen PA, Pointer MD. The effect of Roux-en-Y gastric bypass on prescription drug costs. Obes Surg 2004;14(8):1031–5.
Snow V, Barry P, Fitterman N, Qaseem A, Weiss KL, Clinical Effi cacy Subcommittee of the American College of Physicians. Pharmacologic and surgical management of obesity in primary care: A clinical practice guideline from the American College of Physicians. Ann Intern Med 2005;142(7);525–31.
Statistics Canada and Canadian Institute for Health Information. Body Mass Index (BMI). Health Indicators, vol. 2004, no. 1. Available at: http://www.statcan.ca/english/freepub/82-221-XIE/00604/whatsnew.htm.
Steffen R, Biertho L, Ricklin T, Piec G, Horber FF. Laparoscopic Swedish adjustable gastric banding: A fi ve-year prospective study. Obes Surg 2003;13(3):404–11.
Stocker DJ. Management of the bariatric surgery patient. Endocrinol Metab Clin North Am 2003; 32(2):437–57.
112
Stoopen-Margain E, Fajardo R, Espana N, Gamino R, Gonzalez-Barranco J, Herrera MF. Laparoscopic Roux-en-Y gastric bypass for morbid obesity: Results of our learning curve in 100 consecutive patients. Obes Surg 2004;14(2):201–5.
Sugerman HJ, Starkey JV, Birkenhauer R. A randomized prospective trial of gastric bypass versus vertical banded gastroplasty for morbid obesity and their effects on sweets versus non-sweets eaters. Ann Surg 1987;205(6):613–24.
Sundbom M and Gustavsson S. Randomized clinical trial of hand-assisted laparoscopic versus open Roux-en-Y gastric bypass for the treatment of morbid obesity. Br J Surg 2004;91(4):418–23.
Szold A and Abu-Abeid S. Laparoscopic adjustable silicone gastric banding for morbid obesity―Results and complications in 715 patients. Surg Endosc 2002;16(2):230–3.
Thompson D and Wolf AM. The medical-care cost burden of obesity. Obes Rev 2001;2(3):189–97.
Thompson D, Edelsberg J, Kinsey KL, Oster G. Estimated economic costs of obesity to U.S. business. Am J Health Promot 1998;13(2):120–7.
Thorpe KE, Curtis SF, Howard DH, Joski P. The impact of obesity on rising medical spending. Health Aff 2004;W4:480–6.
Tjepkema M. Measured obesity – Adult obesity in Canada: Measured height and weight. Nutrition: Findings from the Canadian Community Health Survey. Issue no. 1. Component no. 82-620-MWE2005001. Ottawa: Statistics Canada; 2005. Avail able at: http://www.statcan.ca/english/research/82-620-MIE/2005001/pdf/aobesity.pdf (accessed on August 15, 2005).
Toppino M, Morino M, Garrone C, et al. Coelioscopic vertical banded gastroplasty: 3 years experience on 170 cases. J Coelio-Chir 1999;29:81–2.
Van Gemert WG, Adang EM, Kop M, Vos G, Greve JW, Soeters PB. A prospective cost-effectiveness analysis of vertical banded gastroplasty for the treatment of morbid obesity. Obes Surg 1999;9(5):484–91.
Watson DI and Game PA. Hand-assisted laparoscopic vertical banded gastroplasty―Initial report. Surg Endosc 1997;11(12):1218–20.
Weiner R, Bockhorn H, Rosenthal R, Wagner D. A prospective randomized trial of different laparoscopic gastric banding techniques for morbid obesity. Surg Endosc 2001;15(1):63–8.
Weiss HG, Nehoda H, Labeck B, Peer-Kuehberger R, Oberwalder M, Aigner F, Wetscher GJ. Adjustable gastric and esophagogastric banding: A randomized clinical trial. Obes Surg 2002;12(4):573–8.
Westling A and Gustavsson S. Laparoscopic vs open Roux-en-Y gastric bypass: A prospective, randomized trial. Obes Surg 2001;11(3):284–92.
113
Westling A, Ohrvall M, Gustavsson S. Roux-en-Y gastric bypass after previous unsuccessful gastric restrictive surgery. J Gastrointest Surg 2002;6(2):206–11.
Westling A, Bjurling K, Ohrvall M, Gustavsson S. Silicone-adjustable gastric banding: Disap pointing results. Obes Surg 1998;8(4):467–74.
Wittgrove AC and Clark GW. Laparoscopic gastric bypass, Roux-en-Y- 500 patients: Technique and results, with 3-60 month follow-up. Obes Surg 2000;10(3):233–9.
World Health Organization (WHO). Obesity: preventing and managing the global epidemic. Report of a WHO consultation (WHO Technical Report Series no. 894). Geneva, Switzerland: WHO; 2003.
Zimmermann JM, Mashoyan PH, Michel G, Zimmermann E, Grimaldi JM. Laparoscopic adjustable silicon gastric banding : une étude préliminaire personnelle concernant 900 cas opérés entre juillet 1995 et décembre 1998. J Coelio-Chir 1999;29:77–80.
Zinzindohoue F, Chevallier JM, Douard R, Elian N, Ferraz JM, Blanche JP, et al. Laparoscopic gastric banding: A minimally invasive surgical treatment for morbid obesity: Prospective study of 500 consecutive patients. Ann Surg 2003;237(1):1–9.