sign in sign up
<<
Home , Gastric bypass surgery, Virtual colonoscopy

Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1251

Roux-en-Y Gastric Bypass

Hand-assisted and Investigation of the Excluded

BY MAGNUS SUNDBOM

ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2003 Dissertation for the Degree of Doctor of Philosophy (Faculty of Medicine) in Surgery presented at Uppsala University in 2003

ABSTRACT Sundbom, M. 2003. Roux-en-Y Gastric Bypass – Hand-assisted Laparoscopy and Investigation of the Excluded Stomach. Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1251. 62 pp. Uppsala. ISBN 91-554-5604-9.

Roux-en-Y gastric bypass (RYGBP) sustains and ameliorates diseases common in the morbid obese (BMI>40 kg/m2), but leaves the stomach and inaccessible. Morbidly obese patients have increased operative risks and in other fields minimal surgery is known to facilitate the postoperative course. The aim of this thesis was to evaluate hand-assisted laparoscopy in RYGBP and develop techniques to study the excluded stomach. The hand-assisted technique was developed in 13 patients and subsequently compared to open surgery in a blinded, prospective, randomised trial of 50 patients. Hand-assistance was feasible with a low need for conversions or re-operations. The duration of surgery was longer (150 versus 85 minutes) and postoperative results were similar to those in open surgery. Thus, the patients did not appear to derive benefits from hand-assisted laparoscopy. Interventional accessed the excluded stomach and allowed , barium studies and acid measurements. Chronic gastritis and low acid production were found. After RYGBP, 8 of 22 patients (36%) had duodenogastric bile reflux (DGBR), when studied by HIDA-. No DGBR was seen among controls. The was evaluated by serology (pepsinogen I (PGI), H. pylori and H,K-ATPase) in 64 patients before and 1-4 years after operation. RYGBP, in contrast to gastric restriction, had reduced PGI levels postoperatively. According to serology, the mucosa is atrophic or in a resting state. This study focuses on safety in RYGBP. Hand-assisted laparoscopy was feasible, but not favourable compared to an optimised open procedure. The excluded stomach is no longer inaccessible and characterised by chronic gastritis, low acid production and frequent bile reflux.

Key words: Gastric bypass, morbid , laparoscopy.

Magnus Sundbom, Department of Surgical Sciences, Uppsala University Hospital, SE- 751 85 Uppsala, Sweden

© Magnus Sundbom 2003

ISSN 0282-7476 ISBN 91-554-5604-9

Printed in Sweden by Uppsala University, Tryck & Medier, Uppsala 2003

2 To my family

3 This thesis is based on the following papers, which will be referred to in the text by their Roman numerals (I-V)

I Sundbom M and Gustavsson S. Hand-assisted Laparoscopic Roux-en-Y Gastric Bypass: Aspects of Surgical Technique and Early Results. Obes Surg 2000; 10(5): 420-427

II Sundbom M and Gustavsson S. Hand-assisted Laparoscopic versus Open Roux-en-Y Gastric Bypass: A Prospective, Randomised Study. Submitted for publication.

III Sundbom M, Nyman R, Hedenström H, Gustavsson S. Investigation of the Excluded Stomach after Roux-en-Y Gastric Bypass. Obes Surg 2001; 11(6): 25-27

IV Sundbom M, Hedenström H, Gustavsson S. Duodenogastric Bile Reflux After Gastric Bypass. A Cholescintigraphic Study. Dig Dis Sci 2002; 47(8): 1891-1896

V Sundbom M, Mårdh E, Mårdh S, Öhrvall M, Gustavsson S. Reduction in Serum Pepsinogen I After Roux-en-Y Gastric Bypass. Accepted for publication in J Gastrointest Surg

Reprints have been made with the permission of the publishers.

4 Table of contents

Abbreviations 6

Introduction 7

Background 9 Definition of obesity and BMI 9 Risks with morbid obesity 10 Operative treatment 11 Evolution of bariatric procedures 13 Summary of procedures used today 21 Laparoscopy 22

Aims of the thesis 26

Methods 27

Results and discussion 32 Paper I 32 Paper II 37 Paper III 42 Paper IV 43 Paper V 44

Conclusions 51

Summary in Swedish (svensk sammanfattning) 52

Acknowledgements 53

References 54

5 Abbreviations

AGB Adjustable gastric banding BAO Basal acid output BMI BPD Bileo-pancreatic diversion CCK CLO Campylobacter like organism CRP C-reactive protein CT Computed DGBR Duodeno-gastric bile reflux HIDA Hepatic iminodiacetic acid H,K-ATPase Hydrogen potassium adenosine tri-phosphatase H. pylori Lap-RYGBP Laparoscopic Roux-en-Y gastric bypass LCD Low calorie diet JIB Jejuno-ileal bypass MAO Maximal acid output NIH National Institute of Health NSAID Non-steroid anti-inflammatory drugs PPI Proton-pump inhibitor PGI Pepsinogen I PCA Patient controlled analgesia RYGBP Roux-en-Y gastric bypass VBG Vertical banded gastroplasty WHR Waist hip ratio WHO World Health Organisation

6 Introduction People who suffer from obesity bear the double burden of a serious health In the last 20 years, the number of hazard and a social stigma. obese people has doubled [1] and obesity has become the most common Dietary modification with an increase form of [2]. It is in physical activity and reduction in paradoxical that in a world where sedentary habits forms the basis of millions of humans are dying of hunger obesity therapy. Low calorie diets and starvation, many others suffer from (LCD) can be used to replace meals and overeating. By damaging our health, there are also drugs on the market to reducing the quality of our lives, and help promote weight loss. leading us to a premature death, obesity Unfortunately, for most obese patients is becoming the greatest health problem these medical treatments are in the developed world [3, 4]. unsuccessful in the long run.

Food type and availability are the most At present, surgery is necessary to likely reasons for the upward drift in counter extreme obesity [10-15]. the prevalence of obesity during the last However, most surgical procedures decades. However, the recent explosion imply major life-long rearrangement of is considered the result of rapidly the to obtain the decreasing levels of physical activity massive weight reduction. The changes [5]. The most common obesity-related in gastrointestinal physiology can often complications are non-- be controlled by medication and dependent , high blood supplements, but not all long-term pressure, myocardial infarction, sleep effects are fully understood. In apnoea, joint problems, addition, obese patients have an disease and some cancers [4, 6, 7]. increased risk of complications Obese people report a lower quality of secondary to surgery, such as wound life than the population at large [8, 9]. dehiscence [16], [17, 18], [19] and pulmonary

7 embolism [20]. In spite of all has components of both principles and precautions, operative mortality is one is well tolerated by the patients. percent. The patient must be aware of RYGBP is known to sustain weight loss these risks and be committed to life- and ameliorate diseases common in the long follow-up. morbid obese [10-12, 21]. However, the stomach and duodenum are bypassed The introduction of minimal invasive and concerns have been raised about techniques in general surgery has given the environment in the excluded patients benefits in form of reduced stomach. postoperative pain, earlier mobilisation, shorter hospital stay and reduced sick This thesis focuses on the safety of the leave. Obesity was initially considered Roux-en-Y gastric bypass and has a contraindication and laparoscopic evolved along two different lines. First, surgery in the morbid obese has mainly we have studied the role of minimal been performed in specialised centres invasive surgery for RYGBP, in due to the technical difficulties. The particular the so-called hand-assisted results have been promising and the laparoscopic technique. Second, we interest from patients and surgeons in have been concerned about the laparoscopic surgery for morbid obesity environment in the inaccessible is now enormous. excluded stomach and studied the state of the gastric mucosa and postoperative The development of surgical treatment changes in gastrointestinal physiology. has been investigational. The different procedures can be divided into two categories: gastric restrictive and malabsorptive. Gastric restriction mechanically prevents the patient from overeating and the latter interferes with the absorption of ingested nutrients. Roux-en-Y gastric bypass (RYGBP)

8 Background BMI (kg/m2) and corresponding weight at different heights (m):

BMI >25 >30 >40 A number of methods are used to determine whether an individual is 1.70 m 72 kg 87 kg 116 kg overweight or obese. Body mass index, 1.80 m 81 kg 97 kg 130 kg BMI, has been found by body 1.90 m 90 kg 108 kg 144 kg densitometry studies to indicate most accurately the degree of “fatness” for all heights [22]. BMI is defined as BMI does not consider whether the weight in kilograms divided by height excess weight is due to fat or muscular in meters squared. tissue, neither does it reflect the distribution of fat in the body. Studies BMI = kg / m2 indicate that abdominal fat is more hazardous than fat concentrated around the hips [23, 24]. This can be measured Normal weight is defined as a BMI by waist circumference and waist-hip between 18 and 25 kg/m2. WHO circumference ratio, WHR. identifies overweight as a BMI above 25 kg/m2 and obesity above 30 kg/m2 Excess body weight, EBW, is used to [1]. A BMI > 40 kg/m2 implies morbid describe weight above an ideal weight. obesity, a term that will be used The ideal weight is often taken from the extensively in this summary. In the 1983 Metropolitan Life Statistics Table American literature, morbid obesity can [25] and used as a reference when sometimes be defined as being 100 describing weight loss after surgery. pounds (~ 45 kg) overweight. The excess weight lost is presented in percent of the original excess body weight, EWL%. The concomitant use of BMI and EWL% in European and

9 American studies, respectively, is The aetiology of morbid obesity is confusing and does not facilitate multifactorial. Current research comparisons. However, the use of BMI suggests that in genetically is gaining in popularity. predisposed individuals, factors related to lifestyle (diet and Prevalence exercise) and society, culture and The prevalence of obesity is epidemic. behaviour determine whether or A majority of adults in Western not obesity develops [5, 27-31]. societies are overweight and approximately 20% are obese (BMI Risks with morbid obesity >30) [1]. In the last 20 years, the Young overweight people of both sexes number of obese people has nearly die sooner than their lean doubled [4]. It is estimated that 5 to 6 contemporaries [32]. A twelve-fold million Americans are morbidly obese increase in mortality in the 25- to 34- (BMI > 40). In some subgroups, such year age group and a six-fold increase as Afro-American women aged 40 to in the 35- to 44-year age group have 60, individuals with morbid obesity been described among morbidly obese exceed 10% [26]. [3].

Throughout the world obesity is The most frequent co-morbid increasing and a similar pattern as that conditions to morbid obesity are: in the USA is occurring in most of Europe, several South American and • Cardiovascular; , Asian countries [1]. In Sweden, the atherosclerosis, heart failure and number of overweight 18-year-olds was angina pectoris 6% in 1971 and had risen to 16% in • Metabolic; , impaired 1995. Today, the number of obese glucose tolerance and non-insulin- Swedes is estimated to be 500,000. dependent diabetes mellitus • Respiratory; dyspnea, and

10 • Gastrointestinal; hepatic steatosis, single meals over a longer period, often gallbladder disease, and colon 3-4 months. At present, the choice of carcinoma pharmacotherapy is limited and not • Genitourinary; infertility, targeted. In Sweden, two drugs are ® amenorrhea, incontinence, breast approved (Xenical ) [36, 37] and prostate carcinoma and sibutramine (Reductil¡) [38]. In • Degenerative; arthritis of weight- morbid obese, the non-operative bearing joints, chronic venous methods are seldom effective in insufficiency, hernias, and fungal achieving significant long-term weight skin infections loss [39]. • Psychological; depression, isolation and loss of self-esteem Operative treatment Since 1954 surgery has been performed Several of these improve or even in an attempt to control severe disappear with an appropriate weight overweight. These surgical procedures reduction [20, 33, 34]. Weight loss may have been called bariatric after Baros, also have an additional benefit in the Greek for fat. , like all psychological aspects of morbid obesity other surgical interventions, has been improving quality of life [9, 35]. developmental and has had an incidence of failure. However, the Non-operative treatment outcome has improved steadily, mainly Diet and increase in physical activity during the last 10 to 20 years [40]. are conventional strategies in obesity management. Patients need to change Several bariatric procedures are food preparation, adopt to low-fat currently used, having a variety of products, reduce take-away foods as advantages and drawbacks. Most initial dietary changes, as well as bariatric procedures have a mortality reduce portion size. LCD can be used to rate of about 1% and an early replace all meals for several weeks or postoperative morbidity of 10% or more (wound infections, various

11 pulmonary problems, thrombosis or Indications for bariatric surgery stomal stenosis). However, the most According to the National Institute of serious complications, anastomotic Health (NIH) consensus [43] and the leaks and pulmonary embolism occur, established criteria for bariatric surgery in only 1%. The operative treatment as expressed by the International does not involve removal of any Federation of Surgery for Obesity and is in no way (IFSO) [44], the weight requirement for cosmetic. After the massive weight loss an operation is a BMI > 40 kg/m2, or > however, plastic surgery is often 35 in the presence of severe co- required. morbidity. All patients should have failed in long-term professional weight The demand for bariatric surgery is reduction programs including rising world-wide. The understanding medication. The option of surgical of the serious obesity-related morbidity, treatment should be offered to patients in the general public and among health who are well informed and able to care professionals, has increased participate in treatment. Patients with markedly during the past decade. The manifest psychopathology that continuous improvement in both safety jeopardises co-operation in long-term and long-term integrity of the surgical follow-up may need to be excluded. A procedures, and the introduction of multidisciplinary team should evaluate laparoscopic techniques have resulted all patients and the operation should be in a “boom” for bariatric surgery. In performed at well-equipped medical 2001, the number of bariatric centres capable of managing all types procedures in the USA was 40,000; in of surgical complications. This team 2002, it was projected to be 86,000 has also the responsibility to arrange [41]. This can be compared to 6,425 in the life-long follow-up program after 1992 [42]. surgery. At present, only adults (> 18 years) are accepted for bariatric surgery. Morbid obesity has even affected the adolescents and some

12 centres have operated on adolescents Intestinal bypass with good results [45, 46]. The fact that individuals who had the short-bowel syndrome lost weight led Evolution of bariatric to the performance of the jejuno-colic procedures bypass in the 1950s. The intestinal limb lengths used resulted in persisting fluid Various surgical procedures have been and electrolyte imbalance, and tried for the treatment of obesity, many failure. of which later proved to have serious complications precluding their use. The These major disadvantages led to the more powerful the procedure is, the development of the jejuno-ileal bypass greater the risk of gastrointestinal (JIB) in the 1960s. complications and nutritional deficiencies.

Normal digestive tract Jejuno-ileal bypass

13 The JIB was performed by stomach was bypassed, instead of anastomosing the proximal to resected, to produce early satiety and the distal . sustained weight loss with fewer surgical complications. He divided the The malabsorption resulted in weight stomach horizontally across the upper loss. Initial fluid and electrolyte fundus and anastomosed a jejunal loop imbalances were controllable, but renal to the greater curvature. The early stones, migratory polyarthralgia, results were good, but the procedure abdominal bloating and liver resulted in a high incidence of bile dysfunction followed. These required reflux to the gastric pouch and ulcers in close surveillance and gave JIB a bad the anastomosed jejunum. The high reputation. The weight loss was good frequency of bile gastritis was reduced (60% loss of excess body weight) and by changing the jejunal loop to a Roux many patients alive today would have limb [48]. The gastric fundus would died without their JIBs. often dilate and could reach large proportions, allowing the patient to In spite this powerful modification, the resume large food intake. Mason digestive tract was able to adapt by reoriented the gastric pouch to the proliferation of the intestinal lesser curvature and developed another epithelium, leading to weight regain in bariatric procedure, the vertical banded the long run. The JIB is nowadays only gastroplasty (VBG) [49]. This vertical performed by a few surgeons after lesser-curvature pouch was included in additional modifications. Roux-en-Y gastric bypass in 1983 [50]. In 1993, the operation was Gastric bypass adapted to laparoscopy by Wittgrove Ed Mason observed in the 1960s, that and Clark [51]. This has increased its most patients lost weight after gastric popularity, making RYGBP the most resection (Billroth II) and he developed performed bariatric procedure world- the gastric bypass for weight loss wide today [42]. surgery [47]. The major part of the

14 We have chosen RYGBP as our The RYGBP produces massive weight standard procedure and have performed loss and is well tolerated by the almost 400 operations since 1996. We patients. However, bypassing the create a small completely transsected stomach and duodenum has two gastric pouch on the lesser curvature by disadvantages, nutritional deficiencies linear staplers. The Roux limb is made and the inaccessibility. 50-75 cm long and placed behind the and excluded stomach. Nutritional deficiencies

It is anastomosed to the gastric pouch Iron, B12, folic acid and fat-soluble by staplers or hand suturing. , D, and E deficiencies occur. Iron is absorbed in the duodenum and jejunum in the ferrous form and after RYGBP, the absorption in the first 50- cm of the jejunum (the Roux limb) is severely limited due to

[52]. Vitamin B12 absorption is severely affected by difficulties of releasing the protein-bound vitamin [53, 54] and insufficient secretion of

[55]. Therefore vitamin B12 must be available at doses compatible with passive diffusion rather than dependent on the intrinsic factor mechanism. We encourage all patients to take daily dietary supplements of vitamins and

minerals, and we prescribe vitamin B12 to all patients and iron to menstruating women. During pregnancy supplements Roux-en-Y gastric bypass must be increased and liaison between the obstetrician and bariatric surgeon is

15 necessary. The absorption of ethanol is gastric resection [60]. Eleven patients, fast and results in higher peak out of 4300, had perforated peptic concentrations after RYGBP [56]. This duodenal or gastric ulcer [61] and two is noticed by several patients as an cases of gastric cancer has been enhanced feeling of inebriation. This reported [62, 63]. rapid absorption, due to the absence of normal stomach emptying might also The state of the gastric mucosa can now have implications for the be assessed by a combination of pharmacological effect of other drugs. serological tests: S-pepsinogen I, This phenomenon could exist also in antibodies against H. pylori and H,K- patients after total . ATPase. This analysis has been developed as a screening test for The mucosa in the excluded stomach selecting patients for endoscopy [64]. The unknown fate of the bypassed Serum pepsinogen levels have also segments is a major drawback. The been correlated to pernicious anaemia excluded stomach and entire duodenum [65, 66], ulcers [67] and Helicobacter are not readily available for pylori infection[68]. In Japan, radiographic or endoscopic measurements of serum pepsinogens examinations The fundus pouch has have been used as a screening test to been studied after the earlier loop detect subjects with extensive atrophic gastric bypass by endoscopy. Frequent gastritis, who have a high risk of findings were chronic gastritis, bile developing gastric cancer [69-71]. The reflux, and occasional intestinal accuracy is as good as the routine metaplasia [48, 57-59]. There are only [72]. sparse reports on the excluded stomach after RYGBP, probably due to its inaccessibility. Eight patients, in a series of 3000, had bleeding from the bypassed segment that did not respond to conservative treatment and required

16 Vertical Banded Gastroplasty, The pouch was made by stapling a hole VBG near the lesser curvature and from there stapling the front and back wall of the In the mid-1970s, Mason and Printen stomach together up to the angel of performed the first horizontal Hiss. The narrow channel was gastroplasties [73]. They failed to calibrated by a 28- to 32-Fr boogie and control weight because the pouch was a collar of Marlex mesh reinforced the too large and the outlet on the greater outlet. VBG had a high frequency of curvature not stabilized. In 1980, vertical staple line rupture leading to Gomez refined the procedure by passage of food into the main stomach enforcing the outlet by a and subsequent weight regain. This was nonabsorbable, seromuscular suture overcome by dividing the stomach, [74], but the fundus distended easily. instead of just stapling it together. The stomach was therefore partitioned vertically, as mentioned above, and the VBG has been used for several years; it outlet reinforced by a band, creating the is reputable and rather straightforward vertical banded gastroplasty, VBG to perform. The attraction of VBG is [49]. the maintained passage through the duodenum, which preserves iron, phosphate and calcium absorption, and normal metabolism [75]. However, many patients suffer frequent reflux and vomiting, due to the restrictive outlet. VBG is still used today, but has lost its supremacy to RYGBP [42].

Vertical banded gastroplasty

17 Bilio-Pancreatic Diversion, BPD anastomosed to the gastric remnant. The bilio-pancreatic diversion was The bilio-pancreatic limb (from the designed in the late 1970s by Nicola duodenum) is anastomosed 50 cm Scopinaro [76] to overcome the proximal to the ileo-cecal valve, to bacterial overgrowth in the blind loop form a 200-cm alimentary limb and a of the jejuno-ileal bypass, JIB. 50-cm common limb. The digestion with juices from the upper gastrointestinal tract can only occur in these most distal 50 cm of ileum. is done to prevent gallstone problems.

The weight loss is related to restriction of food intake and maintained by malabsorption of starches and fats in the common limb. BPD provides the greatest weight loss (80% excess weight loss) but requires close long- term follow-up since severe nutritional deficiencies may develop.

BPD with The original BPD by Scopinaro has Bileo-pancreatic diversion been modified by Hess and Marceau [13, 77] into the BPD with duodenal The distal part of the stomach is switch. resected to avoid marginal ulceration and restricts food intake initially. The The greater curvature of the stomach is small bowel is divided 250 cm resected, instead of the distal part, proximal to the ileo-cecal valve and leaving the intact. The

18 alimentary limb is anastomosed to the The metabolic drawbacks are similar to divided duodenum (and stomach) those of the original BPD. Super obese proximal to the entry of the common patients (BMI>60 kg/m2) benefit . The biliopancreatic limb is particularly from this procedure, since connected 100 cm proximal to the ileo- it produces massive weight loss [78]. cecal valve, creating a common limb The BPD with duodenal switch can also for absorption. Dumping and marginal be used when correcting severe ulceration is reduced by the intact complications after adjustable gastric pylorus and , which banding, since no pouch or also produces initial weight loss. has to be created at the scarred proximal stomach [79].

Gastric Banding In the 1970s, a restricting band of fixed length was placed around the upper part of the stomach to achieve gastric partitioning without stapling. However, it was difficult to achieve the optimal stoma diameter at surgery. If the band was too loose, weight reduction was minimal or absent and if the stoma was too tight, the proximal pouch would dilate and the patient would experience gastro-oesophageal reflux and frequent vomiting. Band slippage or migration was other problems. A few surgeons reported good results in large series [80, 81].

BPD with Duodenal Switch

19 Adjustable Gastric Banding, or removing fluid postoperatively AGB changes the inner diameter of the stoma. Most bariatric surgeons had disappointing experiences with the The weight loss is inferior to that fixed band. This stimulated the obtained by the more complex methods, development of adjustable inflatable but the main advantage of ABG is devices in the late 1980s [82, 83]. simplicity: laparoscopic placement and no alteration of the normal anatomy. There are two large manufacturers and more than 80,000 bands have been placed, mostly in Europe. There are concerns about reflux-oesophagitis, vomiting due to band slippage, pouch dilatation and band erosions into the stomach. Long-term follow-up has Adjustable gastric banding shown that the AGB method is far from ideal. The cumulative re-operation rate A non-elastic band equipped with an has been 10 to 58% [84-87]. Revisional inflatable silicon balloon on the gastric operations after band erosions are side is placed around the stomach challenging from a technical point of creating the small proximal reservoir. view and are associated with increased The balloon is connected to a morbidity. subcutaneous injection port, and adding

20 Summary of bariatric procedures used today:

AGB VBG RYGBP BPD with DS

Type of procedure Restrictive Restrictive Restrictive/ Malabsoptive malabsorp?

Excess weight loss (%) 40-50 40-50 50-70 60-80

Relief of co-morbidities Good Good Very good Excellent

Advantages No changed metabolism Well Very durable of ingested foods. tolerated results.

Disadvantages Band Staple line B12 and iron Protein and erosion. breakdown deficiency calorie malnutrition Inadequate weight Dumping Diarrhoea, gas, loss and frequent and kidney stones vomiting.

21 Laparoscopy major contribution to the development of laparoscopy in general surgery [90]. Laparoscopy has become a major part In 1985, Muhe performed the first of general surgery with the introduction laparoscopic cholecystectomy, but was of laparoscopic cholecystectomy in the poorly received [91]. Finally, the late 1980s. This new operative technique spread throughout the world technique has started a worldwide and has become a stimulus for revolution of minimal invasive surgery expanding the possibilities of minimal for abdominal surgeons and the invasive procedures in general surgery development of new instruments and [92]. operative procedures. Today, there is a strong demand of laparoscopy from The presented benefits of laparoscopy both patients and surgeons, even in are earlier mobilisation, reduced bariatric surgery. postoperative pain, shorter hospital stay and shorter sick leave in comparison Laparoscopy is an old technique of with the corresponding open procedure. examining the abdominal cavity and its The risk of incisional hernias, as well as contents. In 1901, Kelling introduced a intra-abdominal adhesions is also cystoscope in the abdominal cavity of a diminished [93]. Laparoscopic dog during high-pressure insufflation of , inguinal repair, air and Jacobeus reported on and hiatal hernia fundoplication have laparoscopy in humans ten years later. become common surgical procedures. In 1937, Veres described the use of a All types of gastrointestinal surgery, needle for the creation of and even autopsy can be performed pneumoperitoneum [88] and Hasson laparoscopically [94]. later developed a blunt trocar to eliminate the blind puncture of the abdominal wall [89]. Semm, a gynaecologist, performed the first laparoscopic appendectomy in 1983 - a

22 Laparoscopy in bariatric RYGBP, where extensive dissection is surgery necessary and multiple anastomoses are constructed is a major challenge for the Laparoscopy requires special skills of laparoscopic surgeon. The procedure is the surgeon and morbid obesity poses recognised to be one of the most an added challenge. The establishment difficult laparoscopic procedures in of pneumoperitoneum, exposure and general surgery [95]. The learning intra-abdominal dissection is more curve has been shown to be very long. demanding due to the large quantity of Besides, to obtain good results one fat. Laparoscopy in morbidly obese is must perform a large number of these also associated with other problems, procedures on a regular basis, which is such as impaired respiratory function, why lap-RYGBP is performed mainly high intra-abdominal pressure and liver in large specialised centres [12, 96, 97]. steatosis contributing to the lack of A survey of the American Society for exposure. Bariatric Surgery in 1999 revealed that only 3% of all bariatric cases in the The laparoscopic approach in bariatric USA were done by laparoscopy, but surgery began in 1991 with the this number will probably rise in the development of the Lap-band¡, for near future. adjustable gastric banding [82]. In 1994, the band became available and Nguyen has compared lap-RYGBP to has engendered heated discussions open procedures in several randomised among bariatric surgeons. Nevertheless, studies. Patients have significantly less placement of an adjustable band is the impairment of pulmonary function after least invasive operative procedure that lap-RYGBP than open on the first three can be offered to patients with morbid days [98]. The systemic stress was obesity. In 1993, Wittgrove and Clark similar, but concentrations of ACTH, started to perform laparoscopic Roux- C-reactive protein and IL-6 were lower en-Y gastric bypass (lap-RYGBP) [51]. after lap-RYGBP. These findings may A complex bariatric procedure like lap- suggest less operative trauma [99]. The

23 postoperative length of stay was shorter 103]. In addition, hernias never after laparoscopy, three versus four encountered in open surgery have days. Furthermore, wound-related resulted, such as the Petersen’s hernia complications, such as infection (1.3 vs. between the Roux limb and 10.5%) and (0 vs. mesocolon.[104, 105] 7.9%) were less frequent after laparoscopy [16]. Hand-assisted laparoscopy Lack of tactile gnosis is a problem On the other hand, the operative time is inherent in most types of laparoscopic prolonged and side-effects peculiar to procedures and is especially the laparoscopic approach must be unfavourable in technically demanding avoided. Urinary output is lower procedures. In 1995, Cuschieri and throughout the operation, probably due Shapiro described an extracorporeal to the increased intra-abdominal pneumoperitoneum access bubble [106] pressure [100]. The prolonged and different hand-assisted devices pneumoperitoneum does not impair have subsequently been developed. The cardiac function if proper attention is hand-assisted device is applied in the taken to intravascular volume abdominal wall and seals airtight resuscitation and normal acid-base against the surgeon’s wrist, allowing status, at least in patients without him or her a helping hand in the compromised cardiac function [101]. with maintained The trocars must be placed in the pneumoperitoneum. direction of the operative field through the rich subcutaneous fat; otherwise, The development of hand-assisted too much force is needed to obtain the laparoscopic techniques for advanced correct angle making the handling of procedures in other fields, such as the instruments less sensitive. The splenectomy [107], colo-rectal surgery defects in the fascia are difficult to [108] and living-donor nefrectomy close and hernias have been described, [109] has been promising. More than at all trocar sites and of all sizes [102, 100 hand-assisted procedures have been

24 now described in the literature, Forstieri et al found that hand- including pancreatico-duodenotomy assistance facilitated the placement of [110] and abdominal aortic aneurysm the and in VBG repair [111]. The reports on hand- the technique has been used by Watson assistance in morbid obese subjects [113], Gerhart [114] and Vasallo [115], include adjustable gastric banding, both for correction or first choice VBG, RYGBP and BPD with duodenal operation. The main advantage was switch and are reviewed in Seminars of increased control, here in positioning Laparoscopic Surgery 2001 by the circular stapler for the gastric hole Sundbom and Gustavsson [112]. and subsequent mesh.

25 Aims of the Thesis

This thesis deals with two different issues. First, we wanted to evaluate the use of hand-assistance, proposed to be valuable in other demanding laparoscopic procedures. Second, we wanted to establish an access route to the bypassed stomach and design a variety of techniques to investigate the changes in gastrointestinal physiology and to evaluate the gastric mucosa over time.

The specific aims of the different studies were:

I to develop a new laparoscopic operative technique by hand-assistance to reduce our high conversion and re-operation rate in laparoscopic Roux-en- Y gastric bypass;

II to compare our hand-assisted laparoscopic technique for RYGBP to the conventional open procedure in a prospective, randomised study;

III to obtain access to the excluded stomach after RYGBP by means of interventional radiology, and to perform endoscopy with and measure the production of hydrochloric acid;

IV to study the occurrence of duodeno-gastric bile reflux to the excluded stomach by HIDA-scintigraphy (99mTc-labelled mebrofenin);

V to evaluate the state of the gastric mucosa before and after RYGBP by means of serology (pepsinogen I, antibodies towards H. pylori and K,H- ATPase)

26 Material and Methods 4-6 weeks postoperatively and in complying patients a gastroscopy was Paper I: performed to evaluate the gastro- Thirteen consecutive women with . median age of 38 (26-53) years and BMI of 45 (38-50) kg/m2, not Paper II: previously operated on for morbid Fifty patients (45 women and 5 men) obesity, accepted to have their Roux- with a median age of 38 years (19-54) en-Y gastric bypass (RYGBP) and a median BMI of 44 kg/m2 (34-54) performed by hand-assisted scheduled for Roux-en-Y gastric bypass laparoscopy. All patients had were randomised to either hand-assisted undergone a thorough medical work-up laparoscopic or open operation. The and were motivated to undergo surgical RYGBP technique was identical for treatment. Five patients had previously both procedures (a small totally undergone open cholecystectomy and transsected pouch, a >50-cm retrocolic six patients had had lower abdominal and retrogastric Roux limb, and a operations. The hand-assisted device gastrojejunostomy by a 21-mm circular was introduced through a right stapler) and is described in detail subcostal incision after merely below. The randomisation was done separating the muscle fibres. In one after the induction of anaesthesia. patient, we placed the hand-assisted Dressings were applied for both device in the upper part of a previous techniques and the selected operative midline-incision. The operative technique was blinded for the patient technique developed for the hand- and staff at the ward. The amount of assisted laparoscopic RYGBP is morphine taken by the patients described in detail below. Patient themselves (PCA) was recorded daily. controlled analgesia (PCA) with C-reactive protein (CRP) was measured intravenous morphine was used. Per- as an indicator of inflammatory and postoperative data ware collected. response and systemic stress. Per- and All patients were seen at a clinical visit postoperative data were collected

27 continuously. At discharge, a two-week muscle separated in the fibre direction, sick leave was given and subsequently usually without cutting muscle tissue. prolonged, if needed. The patients were The hand-assisted device was applied examined at 1, 6 and 12 twelve months. and the surgeon introduced his left hand into the abdomen. He could protect the Technical description of the hand- intra-abdominal contents while the first assisted laparoscopic Roux-en-Y 10-mm trocar was introduced in the gastric bypass, used in paper I and II supraumbilical region. The 45-degree laparoscope was inserted, pneumoperitoneum applied and four additional trocars introduced under direct vision in the direction of the operative field.

The enlarged left lobe of the liver was kept aside with a steel retractor. The gastrophrenic ligaments were exposed and incised with cautery and by gentle finger dissection the lesser sac was Incisions for the hand and 5 trocars entered from above. Next the flaccid part of the lesser omentum was opened The patient was placed supine in a and the lesser sac entered below the slight anti-Trendelenburg position. The coronary vein. The intra-abdominal surgeon stood on the patient’s right side hand could easily remove adhesions and the assistant on the left. We used an between the back wall of the stomach 8-cm right subcostal incision for the and the retroperitoneum creating hand-assisted device, and divided the pathways for the coming EndoGia blue subcutaneous fat by blunt fraction. The 45-mm linear staplers (Tyco rectus fascia was incised and the Healthcare).

28 The anvil is introduced through a gastrotomy and a small gastric pouch is constructed. The gastro-jejunostomy is made by a circular stapler (#21).

At a distance of 4 cm from the angle of remaining stomach by two cutting Hiss, a third opening into the lesser sac linear staplers directed towards the was made by finger dissection, very angle of Hiss. close to the lesser curvature of the stomach. The gastric pouch was The ligament of Treitz was identified divided horizontally by linear stapler after retracting the transverse colon and and next a gastrotomy was made on the the major omentum cranially. The small anterior corpus by cautery. The helping bowel was transsected with a linear hand was withdrawn and re-entered stapler 20 to 30 cm further distally with the anvil of the 21-mm circular without dividing the small bowel stapler (Ethicon Endo-surgery). It was mesentery. passed through the gastrotomy by a 10- mm Anvil grasper (Ethicon Endo- The Roux limb was made more than 50 surgery) and taken out near the cm long and the enteroanastomosis was horizontal gastric staple line at the stapled side-to-side using linear stapler. intended position of the gastro-jejunal The intra-abdominal hand was used to anastomosis. A linear stapler closed the manipulate the small bowel, totally gastrotomy. The proximal pouch was avoiding maltreatment by laparoscopic completely separated from the instruments. The mesocolon was

29 Martin Lidholt The entero-enterostomy is made side-to-side by linear staplers. opened near the ligament of Treitz, through a jejunotomy into the Roux where the distance to the retrogastric limb as shown in the earlier figure. The bursa is short. A purse-string suture circular stapler created the end-to-end was placed in-between the lower part of gastrojejunostomy and was left the mesocolon, the ligament of Treitz obturating the porthole. A linear stapler and the root of the small bowel closed the jejunotomy and the Roux mesentery, for future closure of this limb was straightened by hand from defect to prevent entrapment of the below the mesocolon. The purse-string small bowel. The surgeon could suture was tied and three interrupted digitally check that the retrocolic sutures were placed between the serosa window was wide enough and placed of the Roux limb and the mesocolon to the Roux limb beside the proximal prevent internal hernias. The circular gastric pouch without tension or stapler was removed after placing rotation. absorbable sutures in the fascia laparoscopically. A running suture and The circular stapler was introduced PDS loop closed the incision for the through the lower left porthole, after hand-assisted device. All skin incisions having removed the trocar. Guided by were closed by sutures. the surgeon’s hand, it was introduced

30 Open RYGBP stomach and duodenum were examined An identical procedure was performed, by percutaneous endoscopy with using laparoscopic instruments, through multiple histological . a short upper midline incision always Conventional barium studies were also sparing the periumbilical region. The done. The pigtail catheter could also be incision was closed by a PDS loop and used for subsequent determination of skin sutures. . We measured the basal acid output (BAO) and maximal acid Paper III: output (MAO), after pentagastrin Three men, 50 to 64 years old, stimulation (0.5 mg sc). Serum gastrin experienced obscure gastrointestinal and serology for H. pylori were studied. bleeding six months to seven years after Four women, 38 to 63 years old, with RYGBP. They had tarry stools and catheters after complicated required blood transfusions repeatedly. RYGBP, served as controls in the acid A thorough investigation with repeated determinations. upper and could not find the bleeding lesion. In Paper IV: this clinical situation, we needed to Twenty-two patients with a median age examine the excluded stomach to rule of 44 (30-58) years and a BMI of 29 out serious disease. We therefore (17-38) kg/m2 were studied 18 (13-33) developed a technique for percutaneous months after RYGBP by HIDA- access to the excluded stomach, which scintigraphy to evaluate the rate of was first identified by a CT scan. Under duodeno-gastric bile reflux (DGBR) to local anaesthesia and light sedation, a the excluded stomach. Nine patients guide wire was introduced into the had had previous cholecystectomy. stomach using ultrasound guidance. A 99mTc-labelled mebrofenin was given pigtail catheter (Ch 14) was introduced intravenously and the fate of the tracer with Seldinger technique. One to two was recorded every 2 minutes for 90 weeks later, the gastrostomy channel minutes in a dual-head gamma camera. was dilated to 18 Ch. The excluded The bile flow was enhanced by

31 intravenous cholecystokinin (CCK, 20 (10 VBG and 20 adjustable gastric dog units) given 20 minutes after the banding) not altering the normal mebrofenin injection in the first 15 passage of food through the stomach patients. Two of these patients had a and duodenum. Sera were obtained at repeated examination without CCK. the study intervals and kept at –70° C. The excluded stomach was visualised Clinical data, especially abdominal pain by 99mTc-pertechnetate in the last 12 and the use of PPI, were collected in all patients. Eight patients with a median patients. The serological probe had age of 48 (35-67) years and of BMI 24 been constructed and evaluated against (23-25) kg/m2 who had not undergone endoscopic biopsies at the University of upper gastrointestinal surgery served as Linköping, Sweden. All our tests were controls. They were examined due to analysed in duplicates at that laboratory  suspected liver disease during this in the same run: PGI by Gastroset and period at the same laboratory. In all the antibodies by in-house ELISA. patients both anterior and posterior images were recorded, allowing Statistics subsequent computer analysis. The results are presented as median and range in paper I, II, III, and IV. In paper Paper V: V mean, standard deviation and range We evaluated the gastric mucosa in 64 are used. Comparisons between groups patients before and one to four years were conducted using t-tests and the after surgery for morbid obesity by a chi-squared test. P values of less than combination of serological tests: S- 0.05 were defined as significant. In pepsinogen I (PGI), antibodies against paper V univariate and multivariate H. pylori and H,K-ATPase. Thirty-four associations were evaluated. Details are patients (mean age 39 years and mean given in each paper. BMI 44 kg/m2) had a RYGBP excluding the stomach and duodenum. Thirty patients (42 years and BMI 44 kg/m2) had simple gastric restriction

32 Results and Discussion Discussion Paper I: Laparoscopic RYGBP is considered Results one of the most technically demanding The hand-assisted device could be laparoscopic procedures, since it successfully placed and allowed good requires creation of two anastomoses working conditions in all patients. and dissection in two operative fields, Duration of surgery (including i.e. above and below the transverse learning-curve time) was 205 minutes colon and greater omentum. The large (135-280). One patient (8%) was amounts of intra-abdominal fat in the converted to full for a safe morbid obese contribute to a major lack closure of a small perforation of the of exposure, which makes the lack of proximal gastric pouch caused by the tactile gnosis especially unfavourable. anvil of the circular stapler. In another patient, a small accidental gastric The hand-assisted technique was perforation caused by a Babcock clamp feasible. In comparison with our earlier was closed by laparoscopically placed study by Westling et at [116], the sutures. The amount of morphine taken conversion rate was reduced (23% to by the patients (PCA) during 8%). In retrospect, we could even have postoperative day 1-3 was in median completed the converted operation by 45, 32 and 18 mg, respectively. All suturing the small rupture at the patients made an uneventful recovery horizontal staple-line laparoscopically, and the postoperative hospital stay was as we did for the instrumental 5 days (4-7). No sign of leakage, perforation. No patient developed any pneumonia or deep sign of Roux-limb obstruction, which was seen. At the clinical visit after 4-6 occurred in as many as 20% in our weeks, weight loss was 14 (12-23) kg. previous total lap-RYGBP. To focus on Two patients needed endoscopic the new technique, all 13 operations dilatation of a relative stricture at the were performed during three months by gastro-jejunostomy. two surgeons, alternating as surgeon

33 and assistant. We decided to use the 21- organ, often with a delayed diagnosis mm circular stapler to facilitate the [123]. Vascular injuries are a major creation of the gastro-jejunostomy. We cause of death from laparoscopy, chose to enter the anvil through a small second only to anaesthesia, with a gastrotomy, as described by de la Torre mortality rate of 15% [124]. Studying [117] and Hedenbro [118], since the injury-based data, Chandler found 506 transoral route is controversial [119] injuries, resulting in 65 deaths (13%). and possibly dangerous [119-121]. Most injuries were severe (bowel or retroperitoneal vascular injuries in We believe that hand-assistance has 76%) and done when establishing four main advantages: the blind primary entry access [125]. Aortic abdominal puncture with the Veres´ injuries have been reported even with needle is avoided, the hand can be used the use of a Hasson trocar [126]. The for dissection, for handling the small complication rates for urologic bowel and checking a newly laparoscopic surgery on massively constructed anastomosis, and for obese patients (BMI>30), were higher exposure of the surgical field. than in the general population undergoing laparoscopic surgery, 21% In a thick abdominal wall, even the and 0.3%, respectively [127]. ordinary blind puncture with the Veres´ canula can be difficult. The use of an The surgeon’s hand can be used for open approach to introduce a Hassan dissection of the tissue as an alternative trocar is not so easy through the thick to conventional laparoscopic subcutaneous fat. By first establishing instruments. The tactile gnosis was an the hand-assisted device we could invaluable complement to pure vision, introduce one hand into the abdomen especially in opening the retrogastric and guide the introduction of the first bursa and creating the mesocoloic trocar. Entry access injuries occur in tunnel for the Roux limb. The total laparoscopy [122] and the small laparoscopic dissectors are at least 8 bowel is the most commonly injured times less sensitive than the human

34 hand [128] and not optimal [129], with In bariatric surgery, large amounts of a low percentage of successful grasping intra-abdominal fat contribute to a [130]. For a surgeon, the performance severe lack of exposure. The left lobe of the hand is superior to any of the enlarged fatty liver has to be held mechanical device. aside when working at the lesser curvature. The steel retractors One can handle delicate structures, such constructed for normal weight patients as the small bowel, more safely than are sometimes not sufficient and can with a Babcock clamp. High pressures tear the fragile capsule and hepatic are generated at the tip of laparoscopic parenchyma, resulting in annoying dissectors locally on the tissue [131], bleeding. The mere handling of the resulting in damage or even perforation, fatty transverse colon and thick as in our patient. In addition, a small omentum can be difficult. The intra- accidental perforation can pass abdominal hand can reliably handle the unnoticed. The patency of a newly adjacent organs, while total laparoscopy constructed anastomosis can be of course requires mechanical checked by the thumb and index finger. retractors. If the entero-enterostomy is too narrow In addition, the hand is also in place there is an increased risk both for should sudden major haemorrhage leakage at the gastro-jejunostomy and occur. Moreover, a difficult bowel gastric distension with subsequent anastomosis or concomitant blow-out. After having fired the gastro- cholecystectomy can be performed jejunostomy, the circular stapler is not through the incision for the hand, easy to remove, since it has to pass should this be necessary. Undoubtedly, through the inverted tissue at the fragile hand assistance provides more control anastomosis. The countertraction by the during a difficult operation. intra-abdominal hand increases the surgeon’s comfort level during this The drawback of the hand-assisted disengagement. technique is of course the 8-cm incision for the hand, which could omit the

35 benefits of laparoscopy. In addition, the abdominal wall for the creation of hand-assisted technique places the pneumoperitoneum. The HandPort intra-abdominal hand in awkward System is secured by an intra- positions not encountered during open abdominal ring with an inflatable outer surgery, which can result in fatigue and portion. After the device is placed, the ache. The hand-assisted device itself surgeon can guide the introduction of provides wrist and forearm constriction the first trocar (for subsequent that can result in numbness [132]. pneumoperitoneum) by the intra- These difficulties are related to the abdominal hand. Both these two surgeon’s experience and length of the devices require an additional sleeve. procedure. The cost of the hand- The Intromitt has a taut valve and is assisted device is probably justified by secured to the abdominal wall by an the reduction in theatre time. adhesive flange. An insufflation port on the device is used to establish Four devices are now available: the pneumoperitoneum and the hand can Dexterity Pneumo Sleeve® (Dexterity then be introduced without the use of an Surgical Inc), the HandPort System® extra sleeve. The LapDisc consists of (Smith and Nephew Endoscopy), the two rings to create an iris effect. The Intromitt® (MedTech), and the device is placed around the surgeons LapDisc® (Ethicon Endosurgery). They hand and inserted into the abdomen. have similar working mechanism, but The seal around the gloved intra- specific advantages and disadvantages. abdominal hand is tightened by turning The Dexterity Pneumo Sleeve is the rings and pneumoperitoneum applied by an adhesive base to the skin applied by a separate trocar. In after pneumoperitoneum has been addition, instruments can be placed obtained. Failure to maintain through this device by tightening the pneumoperitoneum and skin reactions iris valve even more. from the adhesive have been described. In addition, the device does not Schweitzer et al [133] have reported eliminate the blind puncture of the another approach to the hand-assisted

36 technique for RYGBP, by using the incisional hernia or small bowel Dexterity Sleeve. They had to start the obstruction. Obstruction due to operation by applying adhesions is rare, but RYGBP can be pneumoperitoneum with a Visiport, followed by a specific and potentially since, as already mentioned, the dangerous complication, namely adhesive layer of this device requires internal herniation of the small bowel the abdomen to be distended for an [104, 134]. Three internal hernias are airtight seal. The incision for the sleeve created: the Petersen hernia (behind the was placed in the midline, near the mesentery of the Roux limb), beside the umbilicus, to be able to elongate the Roux limb in the mesocolon [135], and incision should a full laparotomy at the entero-enterostomy [136]. We become necessary. Furthermore, the believed that the hand-assistance proximal small bowel was eviscerated facilitated the closure of these defects at for the construction of the entero- the primary operation. Aggressive enteroanastomosis. We have by using operative treatment is warranted in the HandPort System been able to RYGBP patients with severe cramping protect the intra-abdominal contents abdominal pain whose symptoms have while introducing the first trocar. Our no other obvious explanation [105]. operative technique has been totally intra-abdominal, withdrawing the hand At that time, we were very satisfied only for the introduction of the 21-mm with our hand-assisted technique for anvil. laparoscopic RYGBP. In 2000, we started study II; a prospective, At one year, the weight loss was 42 kg randomised trial between hand-assisted and the BMI 28 kg/m2, i.e. fairly laparoscopic and open RYGBP to similar to conventional open RYGBP. evaluate the new technique. One patient, who had endoscopic balloon-dilatation due to stenosis, has later been re-operated due to recurring strictures. No patient has developed

37 Paper II: days in both groups. Six patients (12%) Results needed endoscopic dilatation of a Duration of surgery in the hand-assisted relative stricture in the gastro- laparoscopic and open group was 150 jejunostomy, two after hand-assisted (110-265) and 85 (60-150) minutes, laparoscopy and four after open respectively, (p<0.001). The median surgery. There was no difference in peroperative bleeding was 250 ml in wound or pulmonary complications. both groups. One patient in each group The sick leave could be evaluated in 40 required two units of blood, due to patients and was in the hand-assisted postoperative anaemia. The median and open group, 30 and 37 days, amount of morphine taken by the respectively. The median weight loss patients (PCA) for postoperative day 1- was 13 kg 4-6 weeks postoperatively in 3 was for the hand-assisted group 48, both groups. It was at one year, 39 kg 36, and 28 mg, respectively. The after hand-assisted and 41 kg after open corresponding values for the open operation, resulting in a BMI of 29 and 2 group was 32, 30, and 25 mg, 30 kg/m , respectively. Accordingly, respectively. The total amount of BMI had been reduced by 14 and 15 morphine required after hand-assisted units. At one year, one symptomatic and open operation during the three incisional hernia was seen in the open first postoperative days was in median group. 98 and 66 mg, respectively. The systemic inflammatory response, as Discussion measured by C-reactive protein, did not The hand-assisted laparoscopic differ between the groups. One patient procedure felt very stable and there was in the hand-assisted group was re- no need for conversions to open operated due to leakage from the surgery. The operative time was gastro-jejunostomy. She was treated by acceptable, 150 minutes. It even large bore drains and intravenous decreased by 55 minutes, from 205 antibiotics, and made a full recovery. minutes in the pilot study (paper I). The median postoperative stay was 6 Our earlier problem in total

38 laparoscopy, obstruction of the Roux- . The new stapler with limb in the mesocolon was still absent. roticulating head can be used in However, the hand-assisted technique especially demanding situations. The had no benefits, compared to the open laparoscopic ultrasonic coagulating procedure, in terms of reduced shears can also simplify tissue morphine consumption, postoperative dissection and division of small vessels, hospital stay or sick leave. but has not been used in this study, except to fenestrate a liver cyst in one In comparing the two techniques, one patient. The routine use of laparoscopic must remember that our open group instruments in open RYGBP has was favoured by the use of recently been advocated by Harold laparoscopic instruments. The [137]. advancement in laparoscopic surgery has supplied new surgical instruments The upper-abdominal incision could, with many attributes (jaw design, long using laparoscopic instruments be short, shafts and handles) which make them 12-15 cm, and always spare the especially valuable in the deep, obese periumbilical region. This could abdomen. The standard GIA or TA facilitate postoperative mobilisation and stapler can be technically difficult to reduce the increased risk of incisional position across the stomach in patients hernia. The open procedures performed with a large, immobile left hepatic lobe. in this study might therefore not be The long shaft of the laparoscopic representative of ordinary open EndoGIA stapler allows the working RYGBP performed in clinical routine. handpiece of the instrument to remain However, in planning the present study outside the abdomen when constructing we thought it was correct, from a the gastric pouch, which improves scientific point of view, to make all visualisation. The division of the efforts to achieve the best outcome in proximal stomach can be made under both groups. better control reducing the risk of leakage and remaining gastro-gastric

39 The most crucial part of the procedure, peroperative endoscopy [143], often done at the end of a demanding reinforcing the anastomosis by tissue laparoscopy, is construction of the glue or bio-membrane [144, 145], and gastro-jejunostomy. We decided to routine use of drains [146]. Most use the circular stapling technique for centres report leakage in 0.5 to 3% of this anastomosis, since it is completely primary RYGBP [134, 142, 147-150]. automatic and obviates the need for We have a rate of 1% without using any hand suture. Leakage from the gastro- of these techniques routinely. Even if a jejunostomy is one of the most feared leakage is diagnosed early and treated complications after RYGBP. The adequately, patients may die. clinical symptoms can be vague, tachycardia, general discomfort and We used the smallest stapler (21 mm), impaired oxygenation [138], as in our according to Wittgrove et al [12], to patient and justify an immediate achieve a stoma of standardised width Gastrograffin swallow [139]. At with restrictive properties. However, laparotomy, the defect in the the incidence of strictures in the gastro- anastomosis can often not be visualised jejunostomy requiring endoscopic and surgically repaired, but large bore balloon dilatation was rather high drains are most important. A few (12%). In a recent report of Gonzalez leakages are contained within the [151], the stricture rate was 31% for the retrogastric bursa with no free contrast 21-mm circular stapler, compared to 3 seen in the abdomen. These can % for a hand sewn anastomosis, which perhaps, in selected cases, be treated by they now favour. According to our percutaneous drains and intravenous recent experience a lower incidence of antibiotics, not laparotomy. Several strictures can be obtained by using the methods have been tried to reduce 25-mm circular stapler. An added leakage: oversewing the staple lines advantage of this stapler is the tilting [140], invaginating the anastomosis, anvil head, which facilitates its removal methylene-blue test [141], air after completing the anastomosis. insufflation under water [142],

40 The length of postoperative stay as midline incision extended to the reported in the present study (6 days in umbilicus by either interrupted figure- median) might seem long. As stated in eight technique by non-absorbable the paper, the in-hospital stay depends sutures or continuous double-stranded on many factors and varies between PDS. They report incisional hernias in centres using the same operative 18 and 10%, respectively, during the 30 technique. Our patients are not months follow-up [18]. Jones [153] motivated to an early discharge for performs open RYGBP through a left economic reasons, due to the socialised subcostal incision and reports few medical system. However, this should hernias. have been affected both groups equally and not biased our comparison. The main finding in this prospective, randomised study was that the patients We found a low incidence of do not appear to derive significant symptomatic incisional hernias, only benefit from having their RYGBP one (4%) in the open group. This is performed by the hand-assisted probably a reward for leaving the laparoscopic technique in comparison periumbilical region intact. The hand- to open surgery. Initially, when the assisted device was placed in a right- hand-assisted approach was introduced subcostal incision and the midline into our surgical practice, we were incisions were always short. DeMaira et rather enthusiastic about the results of al place the hand-assisted device this new technique [154]. However, supraumbilical and report a 20% rate of now that we have appropriate controls hernias [152]. The same group has in an it appears that we have lost the benefits earlier study of 968 patients found of laparoscopy. Also, DeMaria et al greater risk of incisional hernia in [152] have found that hand-assistance morbid obese than in steroid-dependent does not improve patient outcome and patients, 20% and 4%, respectively increases costs when compared to the [17]. Brolin et al have, in a randomised open procedure. Many arguments in study of 229 patients, closed the favour of the hand-assisted technique

41 (tactile gnosis, eased dissection and showed any ulcers, post-ulcerous orientation) are in fact arguments for deformities or tumours in the excluded open surgery. stomach or duodenum. Mucosal biopsies revealed chronic antral I am personally disappointed by our gastritis in all patients and intestinal results. The hand-assisted laparoscopic metaplasia in one. CLO test, histology technique has been valuable for our and culture for H. pylori were all climb on the learning curve, but to negative. The basal acid output (BAO) obtain all the proposed benefits the was 0-2.6 mmol/h and the maximal acid operation has to be done by total output (MAO), after pentagastrin laparoscopy [16, 116]. The learning stimulation, was 0.6 to 9.0 mmol/h. curve for total laparoscopy is long, at These values are lower than the least 75 to 100 cases [96, 155] for an reference value for normal persons. individual surgeon. In Sweden, the Serum gastrin was normal and serology annual number of potential for H. pylori was negative in all laparoscopic cases would be about 30 patients. to 40 at the largest hospitals. It is therefore questionable whether one can Discussion acquire enough experience to perform The excluded stomach is a potential this technically demanding procedure hazard. The fact that the excluded with good results in these conditions. stomach and duodenum cannot be studied by conventional endoscopy or Paper III: barium studies may lead to a delay in Results diagnosis of a serious disease. It is an Access to the excluded stomach was accepted clinical rule that patients with obtained without complications. We alarm symptoms from the upper could perform endoscopy, barium gastrointestinal tract should undergo studies and measurements of the gastric endoscopy to exclude malignancy. acid output through the gastrostomy. After RYGBP, an exploratory Neither endoscopy nor barium studies laparotomy with gastrotomy may be the

42 only alternative to confidently exclude We were able to exclude serious serious gastric disease. disease in the stomach and duodenum as the cause for the bleeding episodes in In our experience access to the our patients. However, the actual cause excluded stomach has been needed in of the bleeding remains obscure [162]. less than one percent of the cases; A temporal relation with cessation of hence the routine placement of a PPI therapy indicates that overlooked gastrostomy tube in the bypassed stomal ulcers could be the cause in spite stomach [156] or a radio-opaque of the repeated normal upper marker [157] seems unnecessary. endoscopies. We were also able to Retrograde endoscopic intubation of the confirm that the mucosa in the excluded excluded stomach and duodenum has stomach showed histological changes been used in the earlier gastric bypass of chronic gastritis and intestinal procedure with a loop gastro- metaplasia. The future clinical jejunostomy [48, 57]. The currently significance of these mucosal changes favoured retrogastric, retrocolic Roux- is not known, but still worrying. The en-Y technique makes endoscopic excluded stomach could harbour quiet intubation of the afferent loop almost dysplastic progression. impossible, due to the sharp angle at the entero-enterostomy. CT-guided catheter The amount of acid produced in the placement [158] and direct Chiba excluded stomach was less than needle punction [159] have been used expected when compared to the in selected cases. We believe that reference values for normal persons ultrasonographic guidance is [163]. Mason [164, 165] and Printen technically easier and safer to perform. [165] has earlier described this in the Virtual CT endoscopy is now being loop gastric bypass. The increase in introduced for examination of the colon acid production after stimulation by [160] and also demonstrated for lesions pentagastrin proves that the parietal in the excluded stomach [161]. cells in the excluded stomach still react to stimuli.

43 Most gastrostomies have bile-stained DGBR occurred equally often in fluid, implying duodeno-gastric bile patients with or without scintigraphic reflux. Bile reflux is known to cause evidence of a functioning gallbladder. chronic gastritis after subtotal A semi-quantitative analysis showed gastrectomy, which can promote an that most of the tracer was transported increased risk of malignancy in the in an aboral direction and the amount long-term [166-169]. We therefore found in the excluded stomach did not wanted to investigate the occurrence of exceed 20 % in any patient. However, bile reflux to the excluded stomach the radioactivity remained in the after RYGBP. excluded stomach throughout the study period of 90 minutes. Paper IV: Results Repeat examinations were performed The examination was easily tolerated without cholecystokinin (CCK) in two by all patients. Eight (36 %) of the RYGBP patients and confirmed the RYGBP patients had scintigraphic presence of DGBR. The tracer appeared evidence of duodenogastic bile reflux in the excluded stomach at about the (DGBR). The tracer appeared in the same time in the first examination (with excluded stomach after a median of 28 CCK), as in the second (without CCK). (21-45) min after the mebrofenin In addition, three patients without a injection history of cholecystectomy were found to have a non-functioning gallbladder.

No scintigraphic evidence of DGBR was found in any of the eight controls. Three were found to have a non- functioning gallbladder. The suspected liver disease did not disturb the tracer’s Tracer in the excluded stomach conjugation to the bile, giving an examination of good quality in which

44 the distribution of the tracer was easy to DGBR in individuals with an intact follow. upper gastrointestinal tract is unusual. None of our controls had any Discussion scintigraphic evidence of DGBR. We performed this exploratory study to Although the gastric bypass procedure investigate whether, and to what extent, does not involve manipulation of the DGBR occurred after RYGBP. To our antroduodenal area, the physiology of surprise, DGBR occurred in as much as the upper gastrointestinal tract is 36 % of the patients. In addition, the profoundly changed after RYGBP. Bile bile-refluxate was not emptied from the flow is co-ordinated from the excluded stomach during the study gastroduodenal area by nervous and time. This leads to a prolonged and hormonal pathways and the absence of enhanced interference of bile with the passing food could disturb this tuning gastric mucosa, since it cannot be system leading to bile reflux. diluted by ingested food and normal gastric contents after RYGBP. HIDA-scintigraphy has been a reliable method to determine enterogastric DGBR has been postulated to reflux after partial gastrectomy [175- contribute to the development of 177]. The method is non-invasive and gastritis and the increased risk of determines the fate of labelled bile malignancy after partial gastrectomy without disturbing tubes or endoscopes, [167, 168, 170-172]. The risk is simulating physiological conditions. estimated to be increased by a factor of Moreover, all images are recorded and two or three when evaluated 15-20 allow later assessment. Special regions years postoperatively [166]. Bile reflux of interest can be identified and specific is also a component in oesophagitis indexes calculated, based on the [173] and a potential component in the amount of measured activity [178]. development of Barrett’s oesophagus [174].

45 This high incidence of bile reflux One year postoperatively, all RYGBP warranted additional evaluation of the patients, except one, had developed a mucosa in the excluded stomach. reduced PGI titre. The reduction was on average 18.7 µg/l. In addition, eight Paper V: patients (24%) showed abnormally low Results values, i.e. below the reference value of Serum pepsinogen I 28 µg/l (not shown in the figure). Before surgery, serum Pepsinogen I (PGI) was 69.3 µg/l and 65.1 µg/l in the RYGBP and control group, respectively, not significantly different.

RYGBP, change in PGI one year postop

30 20 10 0 -10 -20

micrograms/l -30 -40 -50 Patients

46 Controls, change in PGI one year postop

30 20 10 0 -10 -20

micrograms/l -30 -40 -50 Patients

In the control group, the mean statistical significant (p<0.0001). reduction in pepsinogen I was 3.7 µg/l One to four years postoperatively, the at one year. No clear pattern was seen change in mean PGI for the RYGBP among the individual patients, as patients (-13.7, -12.8, -16.6, and -12.5 illustrated in the figure. The difference µg/l) remained throughout the study between RYGBP and controls, -18.7 and was highly significant compared to µg/l and -3.7 µg/l, respectively, was the controls.

Change in Pepsinogen I after RYGBP

10 5 0 -5 -10 -15

micrograms/l -20 -25 -30 Postop year 1 2 3 4

Controls RYGBP

47 The reduction for the RYGBP patients superficial gastritis, duodenal ulcer did not significantly correlate to age, [181, 182] or H. pylori infection [68]. gender, BMI, PPI treatment, , The excluded stomach after RYGBP is status in H. pylori or H,K-ATPase. hence not characterised by acute mucosal inflammation or ulcers. Three RYGBP and six control patients had increased titres of antibodies How can the low PGI values be against H. pylori. Six patients in both interpreted and are there clinical groups had increased titres of implications for patients who have had antibodies against H,K-ATPase. These or are thinking of undergoing a gastric abnormalities remained unchanged bypass? during the study. No additional patient developed increased titres of antibodies On the one hand, the environment in against H. pylori or H,K-ATPase. the excluded stomach could favour the development of chronic atrophic Discussion gastritis. These epithelial changes could Pepsinogen is produced in the gastric have pre-malignant potential, similar to mucosa as an inactive precursor of the the situation in the gastric remnant after protein-digestive enzyme pepsin. A subtotal Billroth II gastrectomy, where small amount of pepsinogen leaks into an increased risk of adenocarcinoma the blood and can be measured as an has been proposed24. If this is true, an indicator of the histological state of the increased awareness of the state of the gastric mucosa [64, 179, 180]. Serum gastric mucosa and readiness for pepsinogen is low in pernicious invasive investigation would be [64, 65] due to the severe atrophy of the warranted for the safety of the RYGBP mucosa. Chronic gastritis, intestinal patients. metaplasia, and gastric cancer are also characterised by low levels (REF). Second, the absence of food stimulation High levels of serum pepsinogen are in the excluded stomach and duodenum associated with mucosal inflammation; contributes to a substantial change in

48 upper gastrointestinal physiology. The No additional patient showed an normal regulatory system is affected, increased titre of antibodies against H. which could set the mucosa in the pylori during the four-year period as a excluded stomach in a resting state. sign of an acquired H. pylori infection. Pepsinogen secretion is known to be Unfortunately, our sample is too small reduced in hibernating animals, after to evaluate whether patients with and with diminished blood gastric and duodenal bypass, for supply10. The fact that the PGI values example after RYGBP, are protected are fully reduced as early as one year against H. pylori infection. In addition, afterwards strengthens this no patients showed serological signs of interpretation, whereas a dysplastic being cured from the H. pylori process would be expected to give a infection. We do not routinely perform more progressive year-by-year pattern. endoscopy [183, 184] or H. pylori eradication preoperatively, as done in In our study, the BMI is similar in both other centres. groups, but it is well known from prospective, randomised studies that The typical RYGBP patient is young RYGBP is followed by a better weight and with the weight loss induced by the loss than gastric restriction. That operation, life expectancy is long. similarity simplified the comparisons of Therefore, all potential adverse effects PGI in the study and was probably due in the excluded stomach have to be to selection bias. We converted patients taken seriously. A permanent solution with poor weight loss after gastric is of course the resectional gastric restriction into RYGBP, leaving only bypass, described by Curry et al [185]. the successful cases to be evaluated in Instead of excluding the main stomach, the control group. It would be they resect it and leave only a 30- to 50- interesting to measure the PGI values in cc gastric pouch, which is anastomosed VBG patients with a subsequent late to a 50-cm retrocolic Roux limb. The staple line rupture. Tacoma group perform the resectional bypass due to the potential problems in

49 the in situ defunctionalized stomach, patient, as it can be technically lacking easy diagnostic access. They demanding. Leakage from the divided believe that the preservation of the duodenum is infrequent, but more stomach is unnecessary from three common than leakage from the staple standpoints: First, in practice, full lines of the excluded stomach, and reversal of a gastric bypass is requires subsequent drainage. infrequently performed. Second, Moreover, a gastrectomy obviates all obesity is a chronic disease and a major production of gastric acid and intrinsic weight regain is known to follow factor. The gastric acid provides reversal. Third, there is extensive protection of the natural bacterial experience with long-term nutritional environment in the small bowel. The outcome, both after total gastrectomy risk for bacterial overgrowth in the performed for reasons other than Roux limb is increased after RYGBP, obesity and in traditional gastric bypass when evaluated by culture and the (with the main stomach excluded). In activity in the Paneth cells, as a marker addition, they report improved exposure of the innate immune system for the gastro-jejunal anastomosis by (Sundbom, unpubl data). A resectional the absence of the main stomach, bypass is of course totally irreversible. especially in revisional procedures. Most centres have reversed patients due Armstrong et al, share the worry about to postoperative complications, such as the potential problems in the excluded small bowel gangrene or severe stomach and also perform a 95% anorexia, thus preventing severe gastrectomy, as well as a simultaneous malnutrition or death. cholecystectomy and appendectomy .

However, a gastrectomy implies an additional risk to the morbid obese

50 Conclusions Paper I We were able to develop a hand-assisted laparoscopic technique for Roux-en-Y gastric bypass that reduced our earlier high conversion and re-operation rate in total laparoscopy. The hand-assisted laparoscopic technique for RYGBP is feasible.

Paper II In a prospective, randomised study we did not find hand-assisted laparoscopy to be superior to conventional open RYGBP. The operative time was longer and no postoperative benefits were seen. The hand-assisted technique cannot replace total laparoscopic RYGBP.

Paper III We could obtain access to the excluded stomach after RYGBP. Endoscopic biopsies showed chronic gastritis and intestinal metaplasia and we could exclude serious disease. The basal, as well as maximal, production of hydrochloric acid was lower than normal, indicating that the excluded stomach is not ulcerogenous.

Paper IV Thirty-six percent of our patients have duodenogastric bile reflux to the excluded stomach after RYGBP. This frequent exposure to bile can imply long-term risks, similar to those after subtotal gastric resections where the risk of malignancy is increased.

Paper V The state of the gastric mucosa in the excluded stomach can be assessed by serology. The pepsinogen I values are reduced and persistently low after RYGBP. The mucosa is probably in a resting state, even if genuine atrophy cannot be excluded.

51 Svensk sammanfattning (Summary in Swedish) Roux-en-Y gastric bypass (RYGBP) är en operation vid sjuklig övervikt (BMI>40 kg/m2) som ger kraftig viktminskning, men lämnar magsäcken urkopplad och svårundersökt för resten av livet. RYGBP är möjlig att utföra med laparoskopi (titthålskirurgi) men tekniskt krävande. Inom andra fält har handassisterad laparoskopi, en hybrid där man kan ha en hjälpande hand inne i buken varit lovande. Målet med avhandlingen har varit att studera om handassisterad laparoskopi ger fördelar vid RYGBP och att utveckla tekniker för att undersöka den urkopplade magsäcken. Vi utvecklade en handassisterad operationsteknik i en pilotserie på 13 konsekutiva patienter. Handporten fungerade bra och endast en patient behövde konverteras till öppen operation. Operationstiden var 205 minuter och alla patienter återhämtade sig väl. Vi randomiserade sedan 50 patienter med median BMI på 46 kg/m2 till handassisterad eller öppen operation. Ingen handassisterad operation behövde konverteras, men operationstiden var längre, 150 mot 85 minuter jämfört med öppen operation. Det var ingen skillnad i patienternas återhämtning (vårdtid 6 dagar) eller viktnedgång (40 kg) efter 1 år. Hos patienter med oklara blödningar efter RYGBP kunde vi punktera den urkopplade magsäcken med ledning av ultraljud. Gastroskopi genom denna kanal visade inga sår eller tumörer men kronisk magkatarr. Saltsyraproduktionen i den urkopplade magsäcken var lägre än normalt. 22 patienter undersöktes i gamma-kamera och 36 % hade reflux av galla till magsäcken: Magsäcksslemhinnan undersöktes hos 64 opererade patienter med en nyutvecklad analys (mätning av pepsinogen, antikroppar mot H pylori samt H,K-ATPas i blod). Hos alla som genomgått RYGBP hade pepsinogenvärdena sjunkit, talande för slemhinneatrofi eller inaktivitet. Arbetet har varit inriktat på säkerhet vid RYGBP. Handassisterad laparoskopi var genomförbar men gav inte patienterna några fördelar. Den urkopplade magsäcken kännetecknas av kronisk magkatarr, låg saltsyraproduktion samt backflöde av galla. Magsäcksslemhinnan förefaller vara i ett vilostadium, men dessa fynd skulle kunna innebära ge ökad cancerrisk på lång sikt.

52 Acknowledgements I would like to express my sincere gratitude to everyone involved in these studies, for all their encouragement and assistance. My special thanks are due to:

Sven Gustavsson, my supervisor, for introducing me into the fields of bariatric surgery and research, daily encouragement and outstanding support. I am especially grateful for your unselfish sharing of operative skills and for placing a “however” in every paragraph in all my manuscripts.

Ulf Haglund, head of the Department of Surgery and Lars Wiklund, head of the Department of Surgical Sciences, for providing the facilities for me to perform these studies simultaneously with my surgical work.

My co-authors, Margareta Öhrvall, Hans Hedenström, Rikard Nyman, Sven Mårdh and Erik Mårdh for contributing with your expertise and good advice.

Friends and colleagues in the upper gastrointestinal team, Britt-Marie Karlsson, Rikard Henriksson, Ib Rasmussen, Rune Sandbu, Bengt Wallner and Agneta Westling, for daily support and encouragement

Friends and colleagues at the Department of Surgery, Uppsala

Anita Ohlin, for a genuine interest and never ending care of the obese patients

The staff of 70AII, for creating a pleasant atmosphere at the ward for upper gastrointestinal surgery

The staff at the operating theatre, for invaluable help and patience

Martin Lidholt, for transforming my sketches into understandable, high class illustrations and generous assistance

Steve Scott-Robson, for friendship and fast, skilful linguistic revision

Former colleagues at the Department of Surgery, Eskilstuna, for giving me a sound start in general surgery

My wife Ann and our children Sofia and Johan, for support, immense understanding and just being who you are

My mother Gunnel, father Lennart, and brothers Stefan and Håkan with family, for giving me a good start in life and constant concern

My parents-in-law Cecilia and Gustaf, for always helping out when needed

53 References

1. Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser, 2000. 894: p. i-xii, 1-253. 2. Gardner, G. and B. Halweil, Hunger, escaping excess. World Watch, 2000. 13(4): p. 25-35. 3. Drenick, E.J., et al., Excessive mortality and causes of death in morbidly obese men. Jama, 1980. 243(5): p. 443-5. 4. O'Brien, P.E. and J.B. Dixon, The extent of the problem of obesity. Am J Surg, 2002. 184(6B): p. 4S-8S. 5. Prentice, A.M. and S.A. Jebb, Obesity in Britain: gluttony or sloth? Bmj, 1995. 311(7002): p. 437-9. 6. Must, A., et al., The disease burden associated with overweight and obesity. Jama, 1999. 282(16): p. 1523-9. 7. Allison, D.B. and S.E. Saunders, Obesity in North America. An overview. Med Clin North Am, 2000. 84(2): p. 305-32, v. 8. Larsson, U., J. Karlsson, and M. Sullivan, Impact of overweight and obesity on health- related quality of life--a Swedish population study. Int J Obes Relat Metab Disord, 2002. 26(3): p. 417-24. 9. Kral, J.G., L.V. Sjostrom, and M.B. Sullivan, Assessment of quality of life before and after surgery for severe obesity. Am J Clin Nutr, 1992. 55(2 Suppl): p. 611S-614S. 10. Fisher, B.L. and A.E. Barber, Gastric bypass procedures. Eur J Gastroenterol Hepatol, 1999. 11(2): p. 93-7. 11. Schauer, P.R., et al., Outcomes after laparoscopic Roux-en-Y gastric bypass for morbid obesity. Ann Surg, 2000. 232(4): p. 515-29. 12. Wittgrove, A.C. and G.W. Clark, Laparoscopic gastric bypass, Roux-en-Y- 500 patients: technique and results, with 3-60 month follow-up. Obes Surg, 2000. 10(3): p. 233-9. 13. Hess, D.S. and D.W. Hess, Biliopancreatic diversion with a duodenal switch. Obes Surg, 1998. 8(3): p. 267-82. 14. Brolin, R.E., Bariatric surgery and long-term control of morbid obesity. Jama, 2002. 288(22): p. 2793-6. 15. Sjostrom, L., Surgical intervention as a strategy for treatment of obesity. Endocrine, 2000. 13(2): p. 213-30. 16. Nguyen, N.T., et al., Laparoscopic versus open gastric bypass: a randomized study of outcomes, quality of life, and costs. Ann Surg, 2001. 234(3): p. 279-89; discussion 289-91. 17. Sugerman, H.J., et al., Greater risk of incisional hernia with morbidly obese than steroid-dependent patients and low recurrence with prefascial polypropylene mesh. Am J Surg, 1996. 171(1): p. 80-4. 18. Brolin, R.E., Prospective, randomized evaluation of midline fascial closure in gastric bariatric operations. Am J Surg, 1996. 172(4): p. 328-31. 19. Anand, P.K., A. Ralph-Edwards, and M. Deitel, Pulmonary Complications in Obesity Surgery. Obes Surg, 1992. 2(4): p. 327-331. 20. Sugerman, H.J., et al., Risks and benefits of gastric bypass in morbidly obese patients with severe venous stasis disease. Ann Surg, 2001. 234(1): p. 41-6. 21. Sugerman, H.J., et al., Gastric bypass for treating severe obesity. Am J Clin Nutr, 1992. 55(2 Suppl): p. 560S-566S.

54 22. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee. World Health Organ Tech Rep Ser, 1995. 854: p. 1-452. 23. Lean, M.E., T.S. Han, and C.E. Morrison, Waist circumference as a measure for indicating need for weight management. Bmj, 1995. 311(6998): p. 158-61. 24. Lean, M.E., Pathophysiology of obesity. Proc Nutr Soc, 2000. 59(3): p. 331-6. 25. 1983 metropolitan height and weight tables. Stat Bull Metrop Life Found, 1983. 64(1): p. 3-9. 26. Flegal, K.M., et al., Overweight and obesity in the United States: prevalence and trends, 1960-1994. Int J Obes Relat Metab Disord, 1998. 22(1): p. 39-47. 27. Bhatti, W.A., et al., Familial Incidence and Influence on the Development of Obesity. Obes Surg, 1996. 6(2): p. 138-142. 28. Gortmaker, S.L., et al., Television viewing as a cause of increasing obesity among children in the United States, 1986-1990. Arch Pediatr Adolesc Med, 1996. 150(4): p. 356-62. 29. Linne, Y., et al., Vision and eating behavior. Obes Res, 2002. 10(2): p. 92-5. 30. Perusse, L., Y.C. Chagnon, and C. Bouchard, Etiology of massive obesity: role of genetic factors. World J Surg, 1998. 22(9): p. 907-12. 31. Steinbeck, K., Obesity: the science behind the management. Intern Med J, 2002. 32(5- 6): p. 237-41. 32. Stevens, J., et al., The effect of age on the association between body-mass index and mortality. N Engl J Med, 1998. 338(1): p. 1-7. 33. Cowan, G.S., Jr. and C.K. Buffington, Significant changes in blood pressure, glucose, and lipids with . World J Surg, 1998. 22(9): p. 987-92. 34. Pories, W.J., et al., Is type II diabetes mellitus (NIDDM) a surgical disease? Ann Surg, 1992. 215(6): p. 633-42; discussion 643. 35. Choban, P.S., et al., A health status assessment of the impact of weight loss following Roux-en-Y gastric bypass for clinically severe obesity. J Am Coll Surg, 1999. 188(5): p. 491-7. 36. Rossner, S., et al., Weight loss, weight maintenance, and improved cardiovascular risk factors after 2 years treatment with orlistat for obesity. European Orlistat Obesity Study Group. Obes Res, 2000. 8(1): p. 49-61. 37. Finer, N., et al., One-year treatment of obesity: a randomized, double-blind, placebo- controlled, multicentre study of orlistat, a gastrointestinal lipase inhibitor. Int J Obes Relat Metab Disord, 2000. 24(3): p. 306-13. 38. Lean, M.E., How does sibutramine work? Int J Obes Relat Metab Disord, 2001. 25 Suppl 4: p. S8-11. 39. Fisher, B.L. and P. Schauer, Medical and surgical options in the treatment of severe obesity. Am J Surg, 2002. 184(6B): p. 9S-16S. 40. Livingston, E.H., Obesity and its surgical management. Am J Surg, 2002. 184(2): p. 103-13. 41. Kral, J.G., et al., Research considerations in obesity surgery. Obes Res, 2002. 10(1): p. 63-4. 42. Pope, G.D., J.D. Birkmeyer, and S.R. Finlayson, National trends in utilization and in- hospital outcomes of bariatric surgery. J Gastrointest Surg, 2002. 6(6): p. 855-61. 43. NIH conference. Gastrointestinal surgery for severe obesity. Consensus Development Conference Panel. Ann Intern Med, 1991. 115(12): p. 956-61. 44. International Federation for the Surgery of Obesity. Statement on morbid obesity and its treatment. Obes Surg, 1997. 7(1): p. 40-1.

55 45. Sugerman, H.J., et al., Bariatric surgery for severely obese adolescents. J Gastrointest Surg, 2003. 7(1): p. 102-8. 46. Stanford, A., et al., Laparoscopic Roux-en-Y gastric bypass in morbidly obese adolescents. J Pediatr Surg, 2003. 38(3): p. 430-3. 47. Mason, E.E. and C. Ito, Gastric bypass in obesity. 1967. Obes Res, 1996. 4(3): p. 316- 9. 48. McCarthy, H.B., et al., Gastritis after gastric bypass surgery. Surgery, 1985. 98(1): p. 68-71. 49. Mason, E.E., Vertical banded gastroplasty for obesity. Arch Surg, 1982. 117(5): p. 701-6. 50. Torres, J.C., C.F. Oca, and R.N. Garrison, Gastric bypass: Roux-en-Y gastrojejunostomy from the lesser curvature. South Med J, 1983. 76(10): p. 1217-21. 51. Wittgrove, A.C., G.W. Clark, and L.J. Tremblay, Laparoscopic Gastric Bypass, Roux- en-Y: Preliminary Report of Five Cases. Obes Surg, 1994. 4(4): p. 353-357. 52. Rhode, B.M., et al., Iron absorption and therapy after gastric bypass. Obes Surg, 1999. 9(1): p. 17-21. 53. Rhode, B.M., et al., Treatment of after Gastric Surgery for Severe Obesity. Obes Surg, 1995. 5(2): p. 154-158. 54. Behrns, K.E., C.D. Smith, and M.G. Sarr, Prospective evaluation of gastric acid secretion and cobalamin absorption following gastric bypass for clinically severe obesity. Dig Dis Sci, 1994. 39(2): p. 315-20. 55. Marcuard, S.P., et al., Absence of luminal intrinsic factor after gastric bypass surgery for morbid obesity. Dig Dis Sci, 1989. 34(8): p. 1238-42. 56. Klockhoff, H., I. Naslund, and A.W. Jones, Faster absorption of ethanol and higher peak concentration in women after gastric bypass surgery. Br J Clin Pharmacol, 2002. 54(6): p. 587-91. 57. Sinar, D.R., et al., Retrograde endoscopy of the bypassed stomach segment after gastric bypass surgery: unexpected lesions. South Med J, 1985. 78(3): p. 255-8. 58. Strodel, W.E., J.A. Knol, and F.E. Eckhauser, Endoscopy of the partitioned stomach. Ann Surg, 1984. 200(5): p. 582-6. 59. Park, H.K., et al., Histologic and endoscopic studies before and after gastric bypass surgery. Arch Pathol Lab Med, 1986. 110(12): p. 1164-7. 60. Printen, K.J., J. LeFavre, and J. Alden, Bleeding from the bypassed stomach following gastric bypass. Surg Gynecol Obstet, 1983. 156(1): p. 65-6. 61. Macgregor, A.M., N.E. Pickens, and E.K. Thoburn, Perforated peptic ulcer following gastric bypass for obesity. Am Surg, 1999. 65(3): p. 222-5. 62. Lord, R.V., P.D. Edwards, and M.J. Coleman, Gastric cancer in the bypassed segment after operation for morbid obesity. Aust N Z J Surg, 1997. 67(8): p. 580-2. 63. Raijman, I., S.V. Strother, and W.L. Donegan, Gastric cancer after gastric bypass for obesity. Case report. J Clin Gastroenterol, 1991. 13(2): p. 191-4. 64. Mardh, E., et al., Diagnosis of gastritis by means of a combination of serological analyses. Clin Chim Acta, 2002. 320(1-2): p. 17-27. 65. Samloff, I.M., et al., Relationships among serum pepsinogen I, serum pepsinogen II, and gastric mucosal histology. A study in relatives of patients with pernicious anemia. Gastroenterology, 1982. 83(1 Pt 2): p. 204-9. 66. Lindgren, A., et al., Serum antibodies to H+,K+-ATPase, serum pepsinogen A and Helicobacter pylori in relation to gastric mucosa morphology in patients with low or

56 low-normal concentrations of serum cobalamins. Eur J Gastroenterol Hepatol, 1998. 10(7): p. 583-8. 67. Samloff, I.M., et al., Elevated serum pepsinogen I and II levels differ as risk factors for duodenal ulcer and gastric ulcer. Gastroenterology, 1986. 90(3): p. 570-6. 68. Ohkusa, T., et al., Changes in serum pepsinogen, gastrin, and immunoglobulin G antibody titers in helicobacter pylori-positive gastric ulcer after eradication of infection. J Clin Gastroenterol, 1997. 25(1): p. 317-22. 69. Kodoi, A., et al., Serum pepsinogen in screening for gastric cancer. J Gastroenterol, 1995. 30(4): p. 452-60. 70. Yoshihara, M., et al., Correlation of ratio of serum pepsinogen I and II with prevalence of gastric cancer and adenoma in Japanese subjects. Am J Gastroenterol, 1998. 93(7): p. 1090-6. 71. Kitahara, F., et al., Accuracy of screening for gastric cancer using serum pepsinogen concentrations. Gut, 1999. 44(5): p. 693-7. 72. Yoshihara, M., et al., The usefulness of gastric mass screening using serum pepsinogen levels compared with photofluorography. Hiroshima J Med Sci, 1997. 46(2): p. 81-6. 73. Printen, K.J. and E.E. Mason, Gastric surgery for relief of morbid obesity. Arch Surg, 1973. 106(4): p. 428-31. 74. Gomez, C.A., Gastroplasty in the surgical treatment of morbid obesity. Am J Clin Nutr, 1980. 33(2 Suppl): p. 406-15. 75. Marcinowska-Suchowierska, E.B., et al., Calcium/phosphate/vitamin D homeostasis and bone mass in patients after gastrectomy, vagotomy, and cholecystectomy. World J Surg, 1995. 19(4): p. 597-601; discussion 601-2. 76. Scopinaro, N., et al., Bilio-pancreatic bypass for obesity: II. Initial experience in man. Br J Surg, 1979. 66(9): p. 618-20. 77. Marceau, P., et al., Biliopancreatic diversion with duodenal switch. World J Surg, 1998. 22(9): p. 947-54. 78. Feng, J.J. and M. Gagner, Laparoscopic biliopancreatic diversion with duodenal switch. Semin Laparosc Surg, 2002. 9(2): p. 125-9. 79. de Csepel, J., et al., Conversion to a laparoscopic biliopancreatic diversion with a duodenal switch for failed laparoscopic adjustable silicone gastric banding. J Laparoendosc Adv Surg Tech A, 2002. 12(4): p. 237-40. 80. Fried, M., M. Peskova, and M. Kasalicky, Assessment of the outcome of laparoscopic nonadjustable gastric banding and stoma adjustable gastric banding: surgeon's and patient's view. Obes Surg, 1998. 8(1): p. 45-8. 81. Fried, M. and M. Peskova, Gastric banding in the treatment of morbid obesity. Hepatogastroenterology, 1997. 44(14): p. 582-7. 82. Kuzmak, L.I., A Review of Seven Years' Experience with Silicone Gastric Banding. Obes Surg, 1991. 1(4): p. 403-408. 83. Belachew, M., et al., Laparoscopic adjustable silicone gastric banding in the treatment of morbid obesity. A preliminary report. Surg Endosc, 1994. 8(11): p. 1354- 6. 84. Morino, M., M. Toppino, and C. Garrone, Disappointing long-term results of laparoscopic adjustable silicone gastric banding. Br J Surg, 1997. 84(6): p. 868-9. 85. Doherty, C., J.W. Maher, and D.S. Heitshusen, Prospective investigation of complications, reoperations, and sustained weight loss with an adjustable gastric

57 banding device for treatment of morbid obesity. J Gastrointest Surg, 1998. 2(1): p. 102-8. 86. Westling, A., et al., Silicone-adjustable gastric banding: disappointing results. Obes Surg, 1998. 8(4): p. 467-74. 87. Gustavsson, S. and A. Westling, Laparoscopic adjustable gastric banding: complications and side effects responsible for the poor long-term outcome. Semin Laparosc Surg, 2002. 9(2): p. 115-24. 88. Sandor, J., et al., A needle-puncture that helped to change the world of surgery. Homage to Janos Veres. Surg Endosc, 2000. 14(2): p. 201-2. 89. Hasson, H.M., A modified instrument and method for laparoscopy. Am J Obstet Gynecol, 1971. 110(6): p. 886-7. 90. Semm, K., Endoscopic appendectomy. Endoscopy, 1983. 15(2): p. 59-64. 91. Litynski, G.S., Erich Muhe and the rejection of laparoscopic cholecystectomy (1985): a surgeon ahead of his time. Jsls, 1998. 2(4): p. 341-6. 92. Himal, H.S., Minimally invasive (laparoscopic) surgery. Surg Endosc, 2002. 16(12): p. 1647-52. 93. Schippers, E., et al., Laparoscopy versus laparotomy: comparison of - formation after in a canine model. Dig Surg, 1998. 15(2): p. 145-7. 94. Catheline, J.M., et al., Autopsy can be performed laparoscopically. Surg Endosc, 1999. 13(11): p. 1163-4. 95. Schauer, P.R. and S. Ikramuddin, Laparoscopic surgery for morbid obesity. Surg Clin North Am, 2001. 81(5): p. 1145-79. 96. Schauer, P., et al., The learning curve for laparoscopic Roux-en-Y gastric bypass is 100 cases. Surg Endosc, 2003. 17(2): p. 212-5. 97. Higa, K.D., K.B. Boone, and T. Ho, Complications of the laparoscopic Roux-en-Y gastric bypass: 1,040 patients--what have we learned? Obes Surg, 2000. 10(6): p. 509-13. 98. Nguyen, N.T., et al., Comparison of pulmonary function and postoperative pain after laparoscopic versus open gastric bypass: a randomized trial. J Am Coll Surg, 2001. 192(4): p. 469-76; discussion 476-7. 99. Nguyen, N.T., et al., Systemic stress response after laparoscopic and open gastric bypass. J Am Coll Surg, 2002. 194(5): p. 557-66; discussion 566-7. 100. Nguyen, N.T., et al., Effect of prolonged pneumoperitoneum on intraoperative urine output during laparoscopic gastric bypass. J Am Coll Surg, 2002. 195(4): p. 476-83. 101. Nguyen, N.T., et al., Cardiac function during laparoscopic vs open gastric bypass. Surg Endosc, 2002. 16(1): p. 78-83. 102. Lajer, H., S. Widecrantz, and L. Heisterberg, Hernias in trocar ports following abdominal laparoscopy. A review. Acta Obstet Gynecol Scand, 1997. 76(5): p. 389- 93. 103. Matthews, B.D., B.T. Heniford, and R.F. Sing, Preperitoneal Richter hernia after a laparoscopic gastric bypass. Surg Laparosc Endosc Percutan Tech, 2001. 11(1): p. 47- 9. 104. Filip, J.E., et al., Internal hernia formation after laparoscopic Roux-en-Y gastric bypass for morbid obesity. Am Surg, 2002. 68(7): p. 640-3. 105. Schweitzer, M.A., et al., Laparoscopic closure of mesenteric defects after Roux-en-Y gastric bypass. J Laparoendosc Adv Surg Tech A, 2000. 10(3): p. 173-5. 106. Cuschieri, A. and S. Shapiro, Extracorporeal pneumoperitoneum access bubble for endoscopic surgery. Am J Surg, 1995. 170(4): p. 391-4.

58 107. Bemelman, W.A., et al., Hand-assisted laparoscopic splenectomy. Surg Endosc, 2000. 14(11): p. 997-8. 108. Darzi, A., Hand-assisted laparoscopic colorectal surgery. Surg Endosc, 2000. 14(11): p. 999-1004. 109. Bemelman, W.A., et al., Hand-assisted laparoscopic donor nephrectomy. Ascending the learning curve. Surg Endosc, 2001. 15(5): p. 442-4. 110. Naitoh, T. and M. Gagner, Laparoscopically assisted gastric surgery using Dexterity Pneumo Sleeve. Surg Endosc, 1997. 11(8): p. 830-3. 111. Kolvenbach, R., Hand-assisted laparoscopic abdominal aortic aneurysm repair. Semin Laparosc Surg, 2001. 8(2): p. 168-77. 112. Sundbom, M. and S. Gustavsson, Hand-assisted laparoscopic bariatric surgery. Semin Laparosc Surg, 2001. 8(2): p. 145-52. 113. Watson, D.I. and P.A. Game, Hand-assisted laparoscopic vertical banded gastroplasty. Initial report. Surg Endosc, 1997. 11(12): p. 1218-20. 114. Gerhart, C.D., Hand-assisted laparoscopic vertical banded gastroplasty: report of a series. Arch Surg, 2000. 135(7): p. 795-8. 115. Vassallo, C., et al., Divided vertical banded gastroplasty either for correction or as a first-choice operation. Obes Surg, 1999. 9(2): p. 177-9. 116. Westling, A. and S. Gustavsson, Laparoscopic vs open Roux-en-Y gastric bypass: a prospective, randomized trial. Obes Surg, 2001. 11(3): p. 284-92. 117. de la Torre, R.A. and J.S. Scott, Laparoscopic Roux-en-Y gastric bypass: a totally intra-abdominal approach--technique and preliminary report. Obes Surg, 1999. 9(5): p. 492-8. 118. Hedenbro, J.L. and S.G. Frederiksen, Fully stapled gastric bypass with isolated pouch and terminal anastomosis: 1-3 year results. Obes Surg, 2002. 12(4): p. 546-50. 119. Wittgrove, A.C. and G.W. Clark, Combined laparoscopic/endoscopic anvil placement for the performance of the . Obes Surg, 2001. 11(5): p. 565-9. 120. Nguyen, N.T. and B.M. Wolfe, Hypopharyngeal perforation during laparoscopic Roux-en-Y gastric bypass. Obes Surg, 2000. 10(1): p. 64-7. 121. Scott, D.J., D.A. Provost, and D.B. Jones, Laparoscopic Roux-en-Y gastric bypass: transoral or transgastric anvil placement? Obes Surg, 2000. 10(4): p. 361-5. 122. Champault, G., F. Cazacu, and N. Taffinder, Serious trocar accidents in laparoscopic surgery: a French survey of 103,852 operations. Surg Laparosc Endosc, 1996. 6(5): p. 367-70. 123. Philips, P.A. and J.F. Amaral, Abdominal access complications in laparoscopic surgery. J Am Coll Surg, 2001. 192(4): p. 525-36. 124. Nordestgaard, A.G., et al., Major vascular injuries during laparoscopic procedures. Am J Surg, 1995. 169(5): p. 543-5. 125. Chandler, J.G., S.L. Corson, and L.W. Way, Three spectra of laparoscopic entry access injuries. J Am Coll Surg, 2001. 192(4): p. 478-90; discussion 490-1. 126. Hanney, R.M., et al., Use of the Hasson cannula producing major vascular injury at laparoscopy. Surg Endosc, 1999. 13(12): p. 1238-40. 127. Mendoza, D., et al., Laparoscopic complications in markedly obese urologic patients (a multi-institutional review). Urology, 1996. 48(4): p. 562-7. 128. den Boer, K.T., et al., Sensitivity of laparoscopic dissectors. What can you feel? Surg Endosc, 1999. 13(9): p. 869-73. 129. den Boer, K.T., et al., Problems with laparoscopic instruments: opinions of experts. J Laparoendosc Adv Surg Tech A, 2001. 11(3): p. 149-55.

59 130. Heijnsdijk, E.A., J. Dankelman, and D.J. Gouma, Effectiveness of grasping and duration of clamping using laparoscopic graspers. Surg Endosc, 2002. 16(9): p. 1329- 31. 131. Cartmill, J.A., et al., High pressures are generated at the tip of laparoscopic graspers. Aust N Z J Surg, 1999. 69(2): p. 127-30. 132. Hanna, G.B., M. Elamass, and A. Cuschieri, Ergonomics of hand-assisted laparoscopic surgery. Semin Laparosc Surg, 2001. 8(2): p. 92-5. 133. Schweitzer, M.A., et al., Laparoscopic-assisted Roux-en-Y gastric bypass. J Laparoendosc Adv Surg Tech A, 1999. 9(5): p. 449-53. 134. Blachar, A., et al., Gastrointestinal complications of laparoscopic Roux-en-Y gastric bypass surgery: clinical and imaging findings. Radiology, 2002. 223(3): p. 625-32. 135. Serra, C., et al., Internal hernias and gastric perforation after a laparoscopic gastric bypass. Obes Surg, 1999. 9(6): p. 546-9. 136. Brolin, R.E., The antiobstruction stitch in stapled Roux-en-Y enteroenterostomy. Am J Surg, 1995. 169(3): p. 355-7. 137. Harold, K.L., et al., Laparoscopic approach to open gastric bypass. Am J Surg, 2002. 184(1): p. 61-2. 138. Hamilton, E.C., et al., Clinical predictors of leak after laparoscopic Roux-en-Y gastric bypass for morbid obesity. Surg Endosc, 2003. 139. Arteaga, J.R., S. Huerta, and E.H. Livingston, Management of gastrojejunal anastomotic leaks after Roux-en-Y gastric bypass. Am Surg, 2002. 68(12): p. 1061-5. 140. Reddy, R.M., et al., Open Roux-en-Y gastric bypass for the morbidly obese in the era of laparoscopy. Am J Surg, 2002. 184(6): p. 611-5; discussion 615-6. 141. Lujan, J.A., et al., Laparoscopic gastric bypass in the treatment of morbid obesity. Preliminary results of a new technique. Surg Endosc, 2002. 16(12): p. 1658-62. 142. DeMaria, E.J., et al., Results of 281 consecutive total laparoscopic Roux-en-Y gastric bypasses to treat morbid obesity. Ann Surg, 2002. 235(5): p. 640-5; discussion 645-7. 143. Champion, J.K., T. Hunt, and N. DeLisle, Role of routine intraoperative endoscopy in laparoscopic bariatric surgery. Surg Endosc, 2002. 16(12): p. 1663-5. 144. Shikora, S.A., J.J. Kim, and M.E. Tarnoff, Reinforcing gastric staple-lines with bovine pericardial strips may decrease the likelihood of gastric leak after laparoscopic Roux- en-Y gastric bypass. Obes Surg, 2003. 13(1): p. 37-44. 145. Kini, S., et al., A biodegradeable membrane from porcine intestinal submucosa to reinforce the gastrojejunostomy in laparoscopic Roux-en-Y gastric bypass: preliminary report. Obes Surg, 2001. 11(4): p. 469-73. 146. Serafini, F., et al., The utility of contrast studies and drains in the management of patients after Roux-en-Y gastric bypass. Obes Surg, 2002. 12(1): p. 34-8. 147. Gould, J.C., et al., Evolution of minimally invasive bariatric surgery. Surgery, 2002. 132(4): p. 565-71; discussion 571-2. 148. Matthews, B.D., et al., Initial results with a stapled gastrojejunostomy for the laparoscopic isolated roux-en-Y gastric bypass. Am J Surg, 2000. 179(6): p. 476-81. 149. Kirkpatrick, J.R. and J.L. Zapas, Divided gastric bypass: a fifteen-year experience. Am Surg, 1998. 64(1): p. 62-5; discussion 65-6. 150. Fobi, M.A., et al., Gastric bypass operation for obesity. World J Surg, 1998. 22(9): p. 925-35. 151. Gonzalez, R., et al., Gastrojejunostomy during laparoscopic gastric bypass: analysis of 3 techniques. Arch Surg, 2003. 138(2): p. 181-4.

60 152. DeMaria, E.J., et al., 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, 2002. 16(10): p. 1452-5. 153. Jones, K.B., Jr., The left subcostal incision revisited. Obes Surg, 1998. 8(2): p. 225-8. 154. Sundbom, M. and S. Gustavsson, Hand-assisted laparoscopic Roux-en-y gastric bypass: aspects of surgical technique and early results. Obes Surg, 2000. 10(5): p. 420-7. 155. Oliak, D., et al., Laparoscopic Roux-en-Y gastric bypass. Surg Endosc, 2003. 17(3): p. 405-8. 156. Wood, M.F., et al., Micropouch gastric bypass: indications for gastrostomy tube placement in the bypassed stomach. Obes Surg, 2000. 10(5): p. 413-9. 157. Fobi, M.A., K. Chicola, and H. Lee, Access to the bypassed stomach after gastric bypass. Obes Surg, 1998. 8(3): p. 289-95. 158. Barmeir, E.P., et al., Radiologic assessment of the distal stomach and duodenum after gastric bypass: percutaneous CT-guided transcatheter technique. Gastrointest Radiol, 1984. 9(3): p. 203-5. 159. Rankin, R.N. and D.M. Grace, Examination of the excluded distal stomach after gastric bypass for obesity. J Can Assoc Radiol, 1985. 36(2): p. 146-7. 160. Gollub, M.J., . Lancet, 2002. 360(9338): p. 964. 161. Silecchia, G., et al., Virtual gastroduodenoscopy: a new look at the bypassed stomach and duodenum after laparoscopic Roux-en-Y gastric bypass for morbid obesity. Obes Surg, 2002. 12(1): p. 39-48. 162. Braley, S.C., N.T. Nguyen, and B.M. Wolfe, Late gastrointestinal hemorrhage after gastric bypass. Obes Surg, 2002. 12(3): p. 404-7. 163. Blackman, A.H., et al., Computed normal values for peak acid output based on age, sex and body weight. Am J Dig Dis, 1970. 15(9): p. 783-9. 164. Mason, E.E., et al., Effect of gastric bypass on gastric secretion. Am J Surg, 1976. 131(2): p. 162-8. 165. Printen, K.J. and M. Owensby, Vagal innervation of the bypassed stomach following gastric bypass. Surgery, 1978. 84(4): p. 455-6. 166. Lundegardh, G., et al., Stomach cancer after partial gastrectomy for benign ulcer disease. N Engl J Med, 1988. 319(4): p. 195-200. 167. Domellof, L., Gastric carcinoma promoted by alkaline reflux gastritis -- with special reference to bile and other surfactants as promoters of postoperative gastric cancer. Med Hypotheses, 1979. 5(4): p. 463-76. 168. Lorusso, D., et al., Duodenogastric reflux and gastric mucosal polyamines in the non- operated stomach and in the gastric remnant after Billroth II gastric resection. A role in gastric carcinogenesis? Anticancer Res, 2000. 20(3B): p. 2197-201. 169. Black, R.B. and J. Rhodes, Bile damage to gastric mucosa--the influence of concentration and pH. Br J Surg, 1971. 58(4): p. 297. 170. von Holstein, C.S., Long-term prognosis after partial gastrectomy for gastroduodenal ulcer. World J Surg, 2000. 24(3): p. 307-14. 171. Gustavsson, S., et al., Cholecystectomy as a risk factor for gastric cancer. A cohort study. Dig Dis Sci, 1984. 29(2): p. 116-20. 172. Nakamura, M., et al., Duodenogastric reflux is associated with antral metaplastic gastritis. Gastrointest Endosc, 2001. 53(1): p. 53-9. 173. Dixon, M.F., et al., Bile reflux gastritis and intestinal metaplasia at the cardia. Gut, 2002. 51(3): p. 351-5.

61 174. Triadafilopoulos, G., Acid and bile reflux in Barrett's : a tale of two evils. Gastroenterology, 2001. 121(6): p. 1502-6. 175. Tolin, R.D., et al., Enterogastric reflux in normal subjects and patients with Bilroth II gastroenterostomy. Measurement of enterogastric reflux. Gastroenterology, 1979. 77(5): p. 1027-33. 176. Gustavsson, S., et al., Scintigraphic assessment of biliary reflux into the residual stomach after subtotal gastrectomy and gastrojejunostomy. Acta Radiol Diagn (Stockh), 1980. 21(5): p. 639-43. 177. Xynos, E., et al., Enterogastric reflux after various types of antiulcer gastric surgery: quantitation by 99mTc-HIDA scintigraphy. Gastroenterology, 1991. 101(4): p. 991-8. 178. Zoras, O., et al., A composite score of enterogastric reflux quantitation on 99mTc- HIDA scintigraphy. Hepatogastroenterology, 1995. 42(6): p. 847-50. 179. Oksanen, A., et al., Atrophic gastritis and Helicobacter pylori infection in outpatients referred for gastroscopy. Gut, 2000. 46(4): p. 460-3. 180. Ley, C., et al., Screening markers for chronic atrophic gastritis in Chiapas, Mexico. Cancer Epidemiol Biomarkers Prev, 2001. 10(2): p. 107-12. 181. Mardh, S., et al., Occurrence of autoantibodies against intrinsic factor, H,K-ATPase, and pepsinogen in atrophic gastritis and rheumatoid arthritis. Scand J Gastroenterol, 1991. 26(10): p. 1089-96. 182. Gritti, I., G. Banfi, and G.S. Roi, Pepsinogens: physiology, pharmacology pathophysiology and exercise. Pharmacol Res, 2000. 41(3): p. 265-81. 183. Renshaw, A.A., et al., Helicobacter pylori infection in patients undergoing gastric bypass surgery for morbid obesity. Obes Surg, 2001. 11(3): p. 281-3. 184. Schirmer, B., C. Erenoglu, and A. Miller, Flexible endoscopy in the management of patients undergoing Roux-en-Y gastric bypass. Obes Surg, 2002. 12(5): p. 634-8. 185. Curry, T.K., et al., Resectional gastric bypass is a new alternative in morbid obesity. Am J Surg, 1998. 175(5): p. 367-70.

62