eMJA: Obesity, weight loss and bariatric = surgery

Clinical Update

Obesity, weight loss and bariatric surgery

Paul=20 E O=92Brien, Wendy A = Brown and John B Dixon

MJA 2005; = 183 (6): = 310-314

Introduction

=97

Weight loss is a powerful health care = measure

=97

Options for tackling obesity

=97

Prevention

=97

Interventions

=97

Weight loss after bariatric surgery

=97

Benefits of surgically induced weight loss on health and = quality=20 of life

=97

Who should be considered for bariatric = surgery?

=97

What challenges need to be overcome?

=97

Competing interests

=97

References

=97

Author details

Abstract

Obesity is shaping up to be the major health care problem = and one=20 of the most frequent causes of preventable death in Western countries = in the=20 21st century.
Bariatric surgery is the only current treatment that has = been shown=20 to achieve major and durable weight loss.
Major weight loss in the severely obese leads to total or = partial=20 control of a wide range of common and serious diseases, such as = diabetes,=20 heart disease and hypertension.
Laparoscopic adjustable gastric banding is the most = common type of=20 obesity surgery performed in Australia. It is effective, relatively = safe and=20 minimally invasive.
The blocks to broader application of bariatric surgery = should be=20 identified and resolved.

Obesity is shaping up to be = one of the=20 greatest health challenges of the 21st century in Western countries. It = causes a=20 great deal of unnecessary morbidity and premature mortality, = substantially=20 reducing quality of life and increasing demand for health care = services.

About 20% of Australian adults (an estimated = 2.6 million) are=20 obese (body mass index [BMI] > 30 kg/m2)1=20 and the incidence of obesity has doubled since 1980. The problem is = especially=20 prevalent in more disadvantaged socioeconomic groups, Aboriginal and = Torres=20 Strait Islander people and rural women.2=20 Alarmingly, obesity often starts in childhood =97 7% of our teenagers = are already=20 obese.3

Obesity is the consummate =93pathogen=94. It increases the = risk of type=20 2 diabetes, coronary heart disease, stroke, hypertension, = obstructive sleep=20 apnoea, depression, a range of cancers (including breast, gynaecological = and=20 gastrointestinal malignancies), dyslipidaemia, polycystic ovary = syndrome,=20 non-alcoholic steatohepatitis and osteoarthritis of the lower spine and=20 weight-bearing joints. The metabolic syndrome is the paradigm of obesity = disease. It is a cluster of problems centring on abdominal obesity and = insulin=20 resistance, with the key clinical manifestations being type = 2 diabetes,=20 hypertension and dyslipidaemia. It is associated with non-alcoholic=20 steatohepatitis, obstructive sleep apnoea and polycystic ovary syndrome = and=20 affects up to one in five Australian adults.4

The US Centers for Disease Control estimated that there = were=20 112 000 excess deaths in the year 2000 from = obesity-related=20 disease.5=20 Using Centers for Disease Control data, Fontaine et al6=20 have predicted that teenagers entering adulthood with a BMI=20 > 40 kg/m2 have their life = expectancy reduced=20 by up to 13 years for males and 8 years for females.

Weight loss is a=20 powerful health care measure

Substantial weight loss has major health benefits. After = major weight=20 loss, more than two-thirds of people with type 2 diabetes no longer = need=20 treatment and return to normal fasting blood glucose, glycosylated = haemoglobin=20 (HbA1c) and serum insulin levels.7=20 Sixty per cent of people with obesity-associated hypertension revert to = having=20 normal blood pressure without need for treatment.8=20 In many people, obstructive sleep apnoea resolves,9=20 hypertriglyceridaemia and high-density-lipoprotein (HDL) cholesterol = levels=20 return to normal,10=20 and the fibrosis associated with non-alcoholic steatohepatitis improves = or=20 disappears.11=20 Fertility increases in women with polycystic ovary syndrome,12=20 and depression at least partially resolves in most people.13=20 In general, quality of life returns to normal14=20 and life expectancy improves.15=20 In view of these outcomes, weight loss in obese people should be a major = priority of doctors, public health practitioners and politicians = committed to=20 good health care.

Options for=20 tackling obesity

Prevention

Prevention is the first choice. It has to be our first = commitment,=20 particularly to our children and adolescents. Unfortunately, no = successful=20 program for primary prevention has yet been established. We have a = pervasive=20 culture, reinforced by powerful commercial forces, that promotes eating = and=20 physical inactivity. To challenge this established culture will take = strong=20 political will and a multidisciplinary approach.16

Interventions

Lifestyle

Solutions that involve lifestyle change are simple to = prescribe, yet=20 rarely achieve sustainable outcomes. The bottom line is that we need to = consume=20 less energy and expend more. Although a healthy lifestyle through good = eating=20 and exercise has been promoted for many years, the problem of obesity is = not=20 only continuing but is actually increasing. A recent systematic review = of the=20 options for weight control found that there is no strong evidence of a = durable=20 effect from any current lifestyle methods.17=20 We should continue to seek effective lifestyle interventions, but must = require=20 evidence of benefit before embracing new options. It is encouraging that = lifestyle interventions have had significant benefits (eg, improvements = in=20 diabetes control, reduced blood pressure and cholesterol levels) for = some people=20 at risk.18,19

Drug = therapy

Orlistat and sibutramine are the only two drugs that can be = recommended for long-term use to achieve and maintain weight loss. = Neither has=20 been shown to induce sufficient weight loss for treating obesity, and = their=20 long-term efficacy and safety are unknown. A recent meta-analysis of=20 pharmacological treatments for obesity showed mean weight losses, after=20 1 year, of 2.9 kg (95% CI, 2.3=963.5 kg) for orlistat and = 4.5 kg (95% CI, 3.6=965.3 kg) for sibutramine.20=20 For obese individuals, who need to lose 10 kg or more, these = results are=20 suboptimal. On the horizon, rimonabant is a potential alternative. In a = recent=20 multicentre trial,21=20 this cannabinoid-1 receptor antagonist produced a mean weight loss = of=20 6.6 kg (SD, 7.2 kg) after 1 year.

Endoscopic=20 procedures

Endoscopic procedures do not currently play an important = role in=20 obesity treatment. Intragastric balloons were enthusiastically adopted = for a=20 period in the 1980s, until randomised controlled trials showed them to = be=20 ineffective.22=20 A new possibility, the BioEnterics intragastric balloon (INAMED Health, = Santa=20 Barbara, Calif), which is approved for clinical use in Australia, is yet = to be=20 proven effective in equivalent trials.

Bariatric=20 surgery

The surgical treatment of obesity (bariatric surgery) is = the most=20 rapidly growing area of surgical practice in Australia today.23=20 This reflects both the ability of bariatric surgical procedures to = provide a=20 solution to an otherwise insoluble problem and the evolution of safer, = less=20 invasive procedures. Today there are essentially three procedures in = use:=20 laparoscopic adjustable gastric banding (LAGB), Roux-en-Y gastric bypass = (RYGB)=20 and biliopancreatic diversion (Box=20 1). There is variation in preferred procedure between countries, = often=20 reflecting local regulatory and insurance factors. In Australia, LAGB is = the=20 procedure of choice in more than 90% of cases, with RYGB making up most = of the=20 rest. Health Insurance Commission data (1994=962004) for Medicare = Benefits=20 Schedule item numbers 30511 (LAGB and other gastroplasty = procedures) and=20 30512 (RYGB) are shown in Box=20 2. The attributes of an ideal bariatric procedure are listed in Box=20 3, along with an estimate of the degree to which each procedure = possesses=20 those attributes. The key features of LAGB that have led to its = acceptance are=20 its safety, minimal invasiveness, adjustability, reversibility and = overall=20 effectiveness.

The LAGB procedure has led to a greater acceptance of = bariatric=20 surgery in Australia for a number of reasons. First, it appears to be=20 considerably safer than the other procedures. A systematic review by the = Australian Safety and Efficacy Register of New Interventional Procedures = =97=20 Surgical (ASERNIP-S) found that the mean short-term mortality rate = associated=20 with LAGB was 0.05%.24=20 Our group has treated more than 2700 severely obese patients with = the LAGB=20 procedure since 1994 without a single perioperative death. In = contrast,=20 mortality from RYGB is reported at between 0 and 5%, with the = ASERNIP-S=20 systematic review showing a mean short-term mortality rate of 0.5% =97 = ten times=20 the risk of LAGB. Second, gastric band placement is minimally invasive. = It is=20 placed laparoscopically, with very little tissue dissection, and can be=20 performed safely as a day-patient procedure.25=20 Third, its primary mode of effect is to induce a feeling of satiety, = even in the=20 absence of meals.26=20 It also has a secondary restrictive effect so that when the person does = eat,=20 there is an early feeling of fullness and very small meals will suffice. = The=20 total food intake is markedly reduced. Fourth, the LAGB is adjustable = =97 perhaps=20 its most important attribute. The degree of induced satiety and = restriction can=20 be controlled by an injection of saline into a subcutaneous access port = during a=20 visit to the clinic (Box=20 4). No other bariatric surgical option provides this important = flexibility.=20 Finally, the LAGB is easily and completely reversible, as no organs or = parts=20 thereof have been stapled or removed and no redirection of gut pathways = has been=20 created. Although there is no intention of removing the device until a = better=20 option is available, such a development may be expected in the future. = It is=20 comforting, particularly to younger patients, to know that a return to = baseline=20 is possible.

It is important to be aware of potential problems = associated with=20 LAGB. Although there are very few significant perioperative problems, = late=20 events such as prolapse of the stomach through the band, erosion of the = band=20 into the stomach and tubing breaks have led to the need for revisional = surgery=20 or removal of the band. Prolapse or slippage was more frequent in our = early=20 experience,27=20 but, over the past 4 years, the incidence has fallen to 4.4%. The=20 revisional procedure is performed laparoscopically and requires an = overnight=20 stay in hospital. Erosion has become a rare event, with only four = erosions=20 (0.4%) in the last 1000 patients treated. Erosion follows a benign = course=20 and is treated laparoscopically. Tubing breaks have now almost = disappeared with=20 redesign of the access port. Overall, we have removed the band in 3.7% = of our=20 patients over a period of 10 years.

Weight loss after=20 bariatric surgery

All bariatric procedures have been able to achieve loss of = more than=20 50% of excess weight.24,27-29=20 The ASERNIP-S systematic review showed greater weight loss after RYGB = than LAGB=20 during the first 2 years after the procedure, but the difference in = weight=20 loss was not significant at 3 and 4 years.24=20 In a recent review,23=20 we extended the data of the ASERNIP-S review by including all studies = that=20 included at least 50 patients, reported up to March 2004 (Box=20 5). This showed a substantial weight loss after both procedures, = with an=20 initial greater weight loss after RYGB but similar effectiveness for = both=20 procedures at 4, 5 and 6 years.

Benefits of=20 surgically induced weight loss on health and quality of life

Weight loss after bariatric surgery has a major impact on = health and=20 quality of life. No other currently available therapies are as effective = in=20 resolving a number of common and serious diseases in the severely = obese.

Type 2 diabetes. = Multiple studies=20 have now demonstrated that, in association with substantial weight loss = induced=20 by bariatric surgery, more than two-thirds of patients with type = 2 diabetes=20 return to having no clinical evidence of the disease, and have normal = fasting=20 blood glucose, serum insulin and HbA1c levels = without=20 therapy. Insulin resistance is reduced and beta-cell function is = improved.7,30

Hypertension.=20 In a study of 88 patients with hypertension followed for = 12 months=20 after LAGB placement, 59% had normal blood pressure (without any = treatment) and=20 a further 33% had improved blood pressure control with less therapy.8

Dyslipidaemia. Several = hundred patients=20 with dyslipidaemia were followed for variable periods up to 4 years = after=20 LAGB placement. Serum triglyceride levels, initially elevated, returned = to=20 normal over this period. HDL cholesterol levels, which were abnormally = low=20 before weight loss, returned to normal values.10

Reflux oesophagitis. We had = reported=20 earlier on a smaller group31=20 but have now followed 82 patients with moderate or severe reflux=20 oesophagitis for at least 12 months after LAGB and have achieved = full=20 control of the disease (no symptoms and requiring no therapy) in 89%, = with=20 improvement in a further 5%. Twenty-four-hour oesophageal pH studies = have=20 confirmed the reduction of reflux, and manometry has not revealed any = change in=20 oesophageal motility (unpublished data).

Asthma. All = 33 patients with=20 asthma who were followed for 1 year after LAGB showed improvement, = and a=20 third of these had no episodes of asthma and were off all therapy during = the=20 12 months after band placement. The mean asthma severity score was = reduced=20 from 44 to 13.32

Depression. Two hundred and = sixty-two=20 patients with depression were studied before and up to 4 years = after LAGB.=20 Before the procedure, about 25% of patients fell into each of four = categories =97=20 major depressive illness, moderate depression, mild depression and = normal status=20 =97 as measured by the Beck Depression Inventory. At 1 year, there = were major=20 reductions in the proportion of patients in each of the depression = groups, with=20 75% of patients assessed to be normal. Four years after the procedure, = the=20 beneficial effect persisted.13

Non-alcoholic steatohepatitis, = obstructive=20 sleep apnoea and polycystic ovary syndrome. These three diseases, = which=20 are commonly found in association with insulin resistance, hypertension, = type=20 2 diabetes and obesity, are all improved with weight loss. Out of=20 21 patients with non-alcoholic steatohepatitis who had repeat liver = biopsies after a weight loss of over 33 kg, the disease had = resolved in=20 19.11=20 Only one of 12 women who had polycystic ovary syndrome before LAGB = showed=20 continued evidence of the disease at 1-year follow-up.33=20 In another study, a group of 25 patients with obstructive sleep = apnoea were=20 using continuous positive airway pressure treatment before LAGB. After = obesity=20 surgery, which resulted in a mean weight loss of 45 kg, there was a = mean=20 reduction in the apnoea/hypopnoea index on polysomnography from = 62 to 13,=20 with 74% of patients no longer needing continuous positive airway = pressure=20 treatment.34

Quality of life. The = Medical Outcomes=20 Trust SF-36 health questionnaire has been used to evaluate quality = of life=20 among 459 patients before LAGB placement and annually thereafter. = Scores on=20 all eight subscales of the SF-36 were abnormally low before = operation. At=20 1 year, scores on all subscales had returned to normal and the = beneficial=20 effect was seen to persist for at least 4 years.14

Who = should be=20 considered for bariatric surgery?

Currently, we consider patients for bariatric surgery if = they have a=20 BMI > 35 kg/m2; have identifiable = medical,=20 physical or psychosocial problems associated with their obesity; and = have made=20 prolonged efforts to relieve those problems by non-surgical means. When = serious=20 comorbidities, such as type 2 diabetes, are present, we may = consider=20 patients with a BMI of less than 35 kg/m2. = Potential=20 candidates undergo a comprehensive clinical and laboratory assessment = and are=20 provided with extensive information about the problem and the procedure=20 (including its potential benefits and risks), and are advised about = their role=20 in the partnership to achieve success. Only when they have fulfilled all = the=20 requirements and appear to have a clear understanding do we accept them = into the=20 treatment program.

What = challenges=20 need to be overcome?

No bariatric procedure has yet been proven to fulfil all of = the=20 ideals listed in Box=20 3, but LAGB comes the closest and is therefore currently the initial = procedure of choice in Australia. There remain several challenges. = First, all=20 bariatric surgical patients require a multidisciplinary long-term = follow-up care=20 program to obtain optimal benefit. Physicians who treat the = comorbidities and=20 general practitioners who care for the patient need to become part of = the team.=20 Second, the durability and cost-effectiveness of bariatric surgery need = to be=20 improved. Published data indicate that LAGB is effective for up to = 8 years,=20 but longer follow-up is desirable. The adjustability of LAGB gives = confidence=20 that this will happen. Third, we should seek further reduction of the = need for=20 revisional procedures that may occur some years after the initial = bariatric=20 procedure. For LAGB, the need for revisional procedures has been reduced = from=20 about 30% during our initial experience to 5% or less, but could be = reduced=20 further. Finally, we should recognise that obesity remains one of the = last areas=20 in which discrimination is regarded as acceptable. Our public hospitals = and=20 colleges continue to disregard their responsibilities for patient care = and=20 training in this area.35=20 As a medical community we should push these bodies to better recognise = the high=20 morbidity associated with obesity, the options for effective management = and the=20 impressive health benefits that can be achieved by weight loss.

1 The three current forms of bariatric surgery that = may be=20 considered for use in Australia=

2 Annual number of bariatric procedures performed in = Australia,=20 1994=962004*

LAGB =3D laparoscopic adjustable gastric = banding.=20 RYGB =3D Roux-en-Y gastric bypass. * Medicare Benefits = Schedule item=20 numbers 30511 (LAGB and other gastroplasty procedures) and=20 30512 (RYGB) (from unpublished Health Insurance Commission = data).

3 Key attributes of an ideal bariatric procedure, and = assessment of each procedure against these attributes*

Attribute	LAGB	RYGB	BPD

Safe	+++	++	+
Effective (substantial weight loss, improved = health and=20 quality of life)	++	++	+++
Easily and fully reversible	Yes	No	No
Side effects	+	++	++
Durable (effective over time)	++	++	+++
Minimal invasiveness=86	+++	++	+
Controllable/adjustable	Yes	No	No
Low reoperation/revision rate	+	+	+

BPD =3D biliopancreatic diversion.=20 LAGB =3D laparoscopic adjustable gastric banding.=20 RYGB =3D Roux-en-Y gastric bypass. * Assessment levels = are based=20 on our synthesis of the published data. =86+++ means=20 the least invasive.

4 Two views of the 10 cm LAP-BAND LAGB system, = with and=20 without added fluid, illustrating how the feeling of satiety in the = contained=20 proximal stomach can be adjusted by adding fluid*

LAGB =3D laparoscopic adjustable gastric = banding. * The=20 adjustable band consists of a ring of silicone with a balloon on its = inner=20 surface connected to an access port. The ring is placed around the top = of the=20 stomach and the port is placed on the abdominal wall muscles under the = skin.=20 There are two forms of the device for which there are significant = published data=20 on effectiveness: the LAP-BAND system (Inamed Health, Santa Barbara, = Calif) and=20 the Obtech band (Ethicon Endo-Surgery, Cincinnati, Ohio).

5 A comparison of the percentage weight loss after = LAGB and=20 RYGB*

LAGB =3D laparoscopic adjustable gastric = banding.=20 RYGB =3D Roux-en-Y gastric bypass. *Based on a systematic = review of all=20 studies published before Dec 2003 that included more than = 50 patients.=20 (Reproduced with permission from O=92Brien PE, Dixon JB, Brown W. = Obesity is a=20 surgical disease: overview of obesity and bariatric surgery. ANZ J Surg [Blackwell Publishing].23)

Competing=20 interests

Paul O=92Brien has received research support (grant-in-aid) = from INAMED=20 Health and Novartis and contributions to training programs from INAMED = Health.=20 John Dixon has received research grants for clinical studies from INAMED = Health=20 and Novartis. Paul O=92Brien and Wendy Brown are bariatric = surgeons.

References

Australian Institute of = Health and=20 Welfare. Australia's health 2004. Canberra: AIHW, 2004. (AIHW Cat. No. = AUS-44.)=20
Australian Institute of = Health and=20 Welfare. Heart, stroke and vascular disease =97 Australian facts 2001. = Canberra:=20 AIHW, 2001. (AIHW Cat. No. CVD-13.)=20
Booth ML, Wake M, Armstrong = T, et al. The=20 epidemiology of overweight and obesity among Australian children and=20 adolescents, 1995=9697. Aust N Z J Public = Health 2001;=20 25: 162-169. <PubMed>=20
Eckel RH, Grundy SM, Zimmet = PZ. The=20 metabolic syndrome. Lancet 2005; 365: = 1415-1428. <PubMed>=20
Flegal KM, Graubard BI, = Williamson DF,=20 Gail MH. Excess deaths associated with underweight, overweight and = obesity.=20 JAMA 2005; 293: 1861-1867. <PubMed>=20
Fontaine KR, Redden DT, Wang = C, et al.=20 Years of life lost due to obesity. JAMA = 2003; 289:=20 187-193. <PubMed>=20
Dixon JB, O=92Brien P. Health = outcomes of=20 severely obese type 2 diabetic subjects 1 year after = laparoscopic=20 adjustable gastric banding. Diabetes Care = 2002; 25:=20 358-363. <PubMed>=20
Dixon JB, O=92Brien PE. = Changes in=20 comorbidities and improvements in quality of life after LAP-BAND = placement.=20 Am J Surg 2002; 184(6B): 51S-54S. <PubMed>=20
Dixon JB, Schachter LM, = O=92Brien PE. Sleep=20 disturbance and obesity: changes following surgically induced weight = loss.=20 Arch Intern Med 2001; 161: 102-106. <PubMed>=20
Dixon JB, O=92Brien PE. Lipid = profile in=20 the severely obese: changes with weight loss after lap-band surgery. = Obes Res 2002; 10: 903-910. <PubMed>=20
Dixon JB, Bhathal PS, Hughes = NR, O=92Brien=20 PE. Nonalcoholic fatty liver disease: improvement in liver = histological=20 analysis with weight loss. Hepatology = 2004; 39:=20 1647-1654. <PubMed>=20
Dixon J, O=92Brien P. Ovarian = dysfunction,=20 androgen excess and neck circumference in obese women: changes with = weight=20 loss [abstract]. Obes Surg 2002; 12: 193.=20
Dixon JB, Dixon ME, O=92Brien = PE.=20 Depression in association with severe obesity: changes with weight = loss. Arch Intern Med 2003; 163: 2058-2065. <PubMed>=20
Dixon JB, Dixon ME, O=92Brien = PE. Quality=20 of life after lap-band placement: influence of time, weight loss, and=20 comorbidities. Obes Res 2001; 9: 713-721. = <PubMed>=20
Christou NV, Sampalis JS, = Liberman M, et=20 al. Surgery decreases long-term mortality, morbidity, and health care = use in=20 morbidly obese patients. Ann Surg 2004; = 240:=20 416-423. <PubMed>=20
Mullis RM, Blair SN, Aronne = LJ, et al.=20 Prevention Conference VII. Obesity, a worldwide epidemic related to = heart=20 disease and stroke. Group IV: prevention/treatment. Circulation 2004; 110: e484-e488. <PubMed>=20
Glenny AM, O=92Meara S, = Melville A, et al.=20 The treatment and prevention of obesity: a systematic review of the=20 literature. Int J Obes Relat Metab Disord = 1997; 21:=20 715-737. <PubMed>=20
Tuomilehto J, Lindstrom J, = Eriksson JG,=20 et al. Prevention of type 2 diabetes mellitus by changes in = lifestyle=20 among subjects with impaired glucose tolerance. N = Engl J=20 Med 2001; 344: 1343-1350. <PubMed>=20
Diabetes Prevention Program = Research=20 Group. Within-trial cost-effectiveness of lifestyle intervention or = metformin=20 for the primary prevention of type 2 diabetes. Diabetes Care 2003; 26: 2518-2523. <PubMed>=20
Li Z, Maglione M, Tu W, et = al.=20 Meta-analysis: pharmacologic treatment of obesity. Ann=20 Intern Med 2005; 142: 532-546. <PubMed>=20
Van Gaal LF, Rissanen AM, = Scheen AJ, et=20 al. Effects of cannabinoid-1 receptor blocker rimonabant on = weight=20 reduction and cardiovascular risk factors in overweight patients: = 1-year=20 experience from the RIO-Europe study. Lancet 2005;=20 365: 1389-1397. <PubMed>=20
Mathus-Vliegen EM, Tytgat GN, = Veldhuyzen-Offermans EA. Intragastric balloon in the treatment of = super-morbid=20 obesity. Double-blind, sham-controlled, crossover evaluation of = 500-milliliter=20 balloon. Gastroenterology 1990; 99: = 362-369. <PubMed>=20
O=92Brien PE, Dixon JB, Brown = W. Obesity is=20 a surgical disease: overview of obesity and bariatric surgery. ANZ J Surg 2004; 74: 200-204. <PubMed>=20
Chapman A, Kiroff G, Game P, = et al.=20 Laparoscopic adjustable gastric banding in the treatment of obesity: a = systematic literature review. Surgery = 2004; 135:=20 326-351. <PubMed>=20
Watkins BM, Ahroni JH, Erlitz = MD, et al.=20 Adjustable gastric banding in an ambulatory surgery center. Obes Surg 2005; 15: 1045-1049. <PubMed>=20
Dixon AF, Dixon JB, O=92Brien = PE.=20 Laparoscopic adjustable gastric banding induces prolonged satiety: a=20 randomized blind crossover study. J Clin = Endocrinol=20 Metab 2005; 90: 813-819. <PubMed>=20
O=92Brien PE, Dixon JB, Brown = W, et al. The=20 laparoscopic adjustable gastric band (Lap-Band): a prospective study = of=20 medium-term effects on weight, health and quality of life. Obes Surg 2002; 12: 652-660. <PubMed>=20
Buchwald H, Avidor Y, Braunwald E, et al. Bariatric = surgery: a=20 systematic review and meta-analysis. JAMA = 2004; 292:=20 1724-1737. <PubMed>=20
Maggard MA, Shugarman LR, = Suttorp M, et=20 al. Meta-analysis: surgical treatment of obesity. Ann=20 Intern Med 2005; 142: 547-559. <PubMed>=20
Dixon JB, Dixon AF, O=92Brien = PE.=20 Improvements in insulin sensitivity and beta-cell function (HOMA) with = weight=20 loss in the severely obese. Homeostatic model assessment. Diabet Med 2003; 20: 127-134. <PubMed>=20
Dixon JB, O=92Brien PE. = Gastroesophageal=20 reflux in obesity: the effect of lap-band placement. Obes=20 Surg 1999; 9: 527-531. <PubMed>=20
Dixon JB, Chapman L, = O=92Brien P. Marked=20 improvement in asthma after Lap-Band surgery for morbid obesity. Obes Surg 1999; 9: 385-389. <PubMed>=20
Dixon JB, O=92Brien PE. Neck = circumference=20 a good predictor of raised insulin and free androgen index in obese=20 premenopausal women: changes with weight loss. Clin=20 Endocrinol (Oxf) 2002; 57: 769-778. <PubMed>=20
Dixon JB, Schachter LM, = O=92Brien PE.=20 Polysomnography before and after weight loss in obese patients with = severe=20 sleep apnea. Int J Obes Relat Metab Disord = 2005; Apr=20 19 [Electronic publication ahead of print].=20
Talbot ML, Jorgensen JO, Loi = KW.=20 Difficulties in provision of bariatric surgical services to the = morbidly=20 obese. Med J Aust 2005; 182: 344-347. = <eMJA full text> <PubMed>

(Received 12 May 2005, accepted = 1 Aug=20 2005)