Operations on the stomach for weight control are designed to avoid ulceration where stomach acid empties. Mechanisms of weight control and choice of operation are reviewed. The patient must make an informed choice regarding the operation and must therefore be educated as to the short and long term effects. Simple restriction of food intake does not produce as great a weight loss as more complex operations that include malabsorption. The more malabsorptive operations require more medical supervision and longterm care. Each operation, simple or complex, requires a close adherence to the technique developed during a long experience with that operation. Malnutrition must be watched for and prevented or treated before injury occurs. Laparoscopic surgery, with an adjustable gastric band, is in widespread use except in the United States, where the device awaits approval. Slippage between the band and stomach is a remaining problem. Whether recent modifications in the technique of band placement will prevent slippage of lower stomach wall into the pouch will require additional time to determine. This review covers only a sample of the extensive literature which needs to be studied by any surgeon caring for the severely obese. The reader will find references in these recent citations to an older literature that remains of value. Evidence requires a well defined, stable operative technique plus years of follow-up. Ten year followups are beginning to appear. Watch with a critical eye, especially for the descriptions of operative technique.
Gastric surgery for severe obesity began in 1966 (Mason 1967) with a form of gastric bypass that was patterned after Billroth II gastrectomy, an operation known to cause weight loss. Like intestinal bypass, gastric bypass was reversible should it not be tolerated. Reversibility has been one of the tenets of obesity surgery from the beginning. As experience accumulated there were efforts to reduce risk by use of gastroplasty instead of gastric bypass. Other surgeons continued with gastric bypass and some included bypass of more intestine in order to increase weight loss (Mason, Tang 1997). In order to determine the outcome of operations used for weight control, it is mandatory that the operative technique be completely described and not change during the time under study. There is disagreement about the relative importance of weight control and avoidance of undesirable side effects of the operations. The one variable most predictive of weight loss is the initial weight of the patient. Patients who are the heaviest lose the most absolute weight but have the lowest relative weight loss. This makes it difficult to define success and failure in a way that is applicable to all patients.
Weight control mechanisms
Operations control body weight in at least three ways, (1) restriction of intake by creation of a small gastric pouch, (2) malabsorption through bypass of stomach, duodenum and a variable extent of small bowel and (3) by stimulation of the terminal ileum to release GLP-1 when there is undigested food in this segment. This last variable has been overlooked by clinicians since it was first studied in obese rats by Koopmans in the early 1980s. At that time release of enterogastrone was discovered to cause weight loss when the terminal ileum was transposed to near the duodenum. It is now apparent that the hormone responsible for the weight loss is glucagone-like peptide-1 (GLP-1) released when chyme is in the terminal ileum. This hormone may be of even more importance in improving carbohydrate metabolism than in the control of body weight. It turns out that there are high levels of GLP-1 in the serum of patients 20 years after intestinal bypass (Naslund 1998). This literature was reviewed recently (Mason 1999).
Much of the learning curve over the last 30 years in the use of gastric restriction operations relates to establishing and maintaining a small enough upper stomach segment without causing obstruction and reflux. For over a decade now it has been known from rates of reoperation with gastric bypass and gastroplasty that the pouch must be small, less than 20 ml in volume in vertical banded gastroplasty, at time of the primary operation. The stretching tension in the pouch wall is related to the radius of the pouch. In consequence, the larger the initial pouch, the more it will stretch over time. When the pouch reaches a certain size, it no longer empties properly and esophageal reflux and vomiting may occur. Weight gain may also occur if the storage capacity is sufficient, even though there is impaired emptying.
For malabsorption operations the gastric storage capacity must be somewhat larger to allow an adequate intake of meat and other high protein foods. The digestive tract evolved to provide an efficient and safe sequence of digestion and absorption of food. To redesign this complex system to control body weight at a desired level for patients who are unable to control their food intake has required a great effort.
HCl and stomal ulceration
William Beaumont in 1833 reported that the stomach secreted hydrochloric acid, an astonishing finding. Two ulcerogenic operations were identified early in this century, (1) antral exclusion and (2) gastroenterostomy. Before gastric bypass could be used for treatment of obesity in humans it was necessary to determine whether the operation would cause ulcers. Animal experiments showed that when most of the acid secreting portion of the stomach was excluded along with the antrum, there was inhibition of gastrin secretion and suppression of HCl secretion (Mason and Ito, 1967). As a result stomal ulcer and duodenal ulcer were unlikely to result from gastric bypass as long as the division of the stomach created a small upper segment.
When the staple line breaks down, there is produced the same situation as if a gastroenterostomy had been performed with an intact stomach. A recent study of 128 patients with disrupted gastric partitions in a series of 499 patients with gastric bypass showed 36 who developed stomal ulcer (MacLean 1997). All patients who developed stomal ulceration had disruption of the staple line. Acid production was found to be increased by the disruption. Closure of the communication decreased acid secretion and allowed healing. Division of the stomach at the primary operation was observed to decrease the risk of disruption of the staple line but did not eliminate it. Interposition of a well vascularized organ between the two segments of divided stomach was suggested as a way to prevent development of a gastro-gastric fistula.
The informed patient and operative choice
The patient must live with the operation chosen and therefore must know enough to make an informed choice. This places a major responsibility upon the surgeon in education of the patient about the advantages and disadvantages of the operation recommended (Mason and Hesson 1998). One of the simple operations is vertical banded gastroplasty (VBG). Gastroplasty has the advantage of maintaining a normal sequence of digestion and absorption and does not have any of the potential complications and side effects peculiar to bypass operations. VBG has a measured pouch of less than 20 ml and an outlet stabilized with a 5 cm circumference collar of Marlex mesh. The mesh is rapidly incorporated in fibrous tissue that keeps the stomach wall from slipping through the collar.
Serious perioperative complications occur in one percent of patients with VBG and consist of leak with peritonitis in 0.6% and thromboembolism in 0.4%. Because of the introduction of a 4-row stapler in 1986, we are just beginning to obtain sufficient ten year follow-up to know the long term results with the operative technique now in use (Mason, Doherty 1998). For morbidly obese patients, operative weight averaged 200% of ideal and 10 year weight 159%. For super obese patients operative weight averaged 262% and the 10 year weight, 197%. The reoperation rate during this decade has been less than one percent per year. A recent report recommended VBG with a divided partition for prevention of disruption of the pouch and the laparoscopic approach (Toppino 1999). This included an experience with several types of collar and concluded that Marlex mesh was the material of choice. A well defined technique that evolved over many years has been described for the open approach (Mason, Doherty 1997). Every change in technique may effect the results and needs at least 10 years of evaluation before we can be assured of an advantage.
Gastric bypass produces a better early weight loss than VBG (Pories 1995, MacLean 1997) but not enough better to satisfy many surgeons who use bypass operations. The trend has therefore been to use longer lengths of bypass of small bowel. The greatest and most stable weight reduction comes from biliopancreatic diversion, in which most of the small bowel is bypassed (Scopinaro 1998). There is also a variation of BPD called duodenal switch (Marceau 1998). These operations require a great deal closer monitoring, more medical care, laboratory work, medications and nutritional supplementation.
Scopinaro resects distal stomach rather than exclude it. Because BPD requires a larger pouch there is an increased risk of stomal ulcer even with resection of the gastrin producing antrum. Scopinaro has been able to reduce the incidence of stomal ulcer to 3.2% during the first postoperative year and 4.4% during the second year by administration of H2-blockers. Marceau removes most of the parietal cell mass by resecting the greater curvature aspect of the stomach, which requires a long staple line and creates a tube of the lesser curvature aspect of the stomach. Any resection is of course irreversible although the bypass could be reversed in each of these operations, should the patient need a return to more normal anatomy.
The SOS study of gastric surgery for severe obesity is designed to compare 2000 surgical patients with 2000 matched controls. An analysis of changes in health related quality of life is now available for a two year follow-up of 487 patients in each group (Karlsson 1998). A comprehensive battery of self-assessment measures revealed marked positive effects of surgery at six to 12 months followed by moderate reduction at two years. We can anticipate further reports regarding longer term effects as well as preoperative outcome predictors in this study. Also, there should soon be very valuable studies of other variables. This is a patient based study that is non-randomized but with a matched control group. It is a model of ethical prospective design, and complete collection of data. Patients are entered and followed throughout Sweden according to the study center directions. Follow-up at two years in this particular study is 98% for the surgical patients and 84% of the matched, control group. The latter are receiving whatever treatment they elect or no treatment. When this study reaches the 10 year follow-up for all 4000 patients it should provide most of the evidence regarding gastric surgery for obesity. More important, it should demonstrate how much can be accomplished by control of the obesity epidemic by whatever effective measure can be developed. With so many surgeons involved there could be variation in measurement and size of pouches, collar size and material and other variables. However, with so many patients involved, it should be possible to obtain additional evidence regarding the effect of variations in operative technique.
Malnutrition avoidance; emesis no, anamnesis yes
The goal in obesity surgery is weight loss. With intake restriction, vomiting may occur if a patient eats too rapidly, or too much for a small pouch, or does not chew food well enough. Even with malabsorption operations, in which there is less dependence upon a small pouch, there are adjustments that the digestive tract must make before the patient is able to eat without symptoms. In other words there is expectation of rapid weight loss and a tendency to dismiss warning symptoms that would otherwise raise questions about possible obstruction or the patient’s nutrition. It is therefore most important, when a patient complains of vomiting, that a sufficient history and investigation be conducted to assure that there is not uncontrolled vomiting because of obstruction or even intolerance to the operation. This must not be postponed since the risk is malnutrition that can be fatal or lead to irreversible changes (Mason, 1998) There are reports in the literature of sudden death from protein malnutrition, death from the refeeding syndrome and death or irreversible neurologic changes from thiamin deficiency (Wernicke-Korsakoff syndrome). The surgeon who performs operations for the treatment of obesity should know how to recognize, anticipate and prevent such complications. Patients must be taught to return for study and treatment if they develop vomiting that they cannot control by appropriate modification of their eating.
Placement of a band around the upper stomach would seem to be an even simpler operation than VBG since neither window nor stapling are required. There are in use adjustable bands with a bladder on the inner surface, connected to an injection port placed on the upper, anterior abdominal wall and beneath the skin. This devise has not been approved for general use in the United States but is in widespread use in Europe. Like VBG, this operation requires a very precise technique. The juxtaesophageal, stomach pouch must be small, as in VBG, or it will balloon and require reoperation or removal of the band. To prevent slipping, the stomach is sutured over the band. In back, such suturing is incomplete. It is in this area where slippage is most likely to occur, with stomach wall herniating cephalad. The result is obstruction and need for reoperation. As a means of preventing posterior herniation, it is recommended that the posterior dissection and band placement be above the curvature of the balloon that is inflated in the lumen of the upper stomach, and above the peritoneal reflection of the bursa omentalis (Favretti 1995, Belachew 1998). The potential advantage of the adjustable band rests upon the importance of adapting the diameter of the outlet of the small pouch to the patient. An excellent study of the early effect of outlet diameter has shown that it is possible for patients to eat more solid food and protein with almost no vomiting and without impairment of weight control if the outlet is allowed to be larger during the first month and then gradually adjusted as needed to provide sufficient control of intake (Busetto 1997). More patientyears of follow-up are needed to determine whether sufficient stability of the pouch, absence of esophageal reflux, weight control and freedom from an excessive reoperation rate have been achieved by the operative technique(s) now in use with adjustable banding.
There is an extensive older literature on rings, collars and bands which surgeons who use bands should be familiar with. A choice must be made between a band that does not adhere to the stomach (and can be easily removed) and a mesh that is infiltrated with connective tissue and does not slip, allow herniation of the stomach wall or tilting of collar (as was observed with the silastic ring). In summary there is a great deal of evidence that has accumulated during the last three decades about surgical treatment of obesity that came from retrospective analysis and correction of technical problems as they arose. Randomized prospective studies have been difficult to design and none of these studies were double blind. That does not mean that we cannot learn from past experience.
- Belachew M, Legrand M, Vincent V, Lismonde M, Le Docte N, Deschamps V. Laparoscopic adjustable gastric banding. World J Surg. (1998);22:955–963. [PubMed: 9717421]
- Busetto L, Pisent C, Segato G, De Marchi F, Favretti F, Lise M, Enzi G. The influence of a new timing strategy of band adjustment on the vomiting frequency and food consumption of obese women operated with laparo- scopic adjustable silisone gastric banding (LAP-Band) Obesity Surg. (1997);7:505–512. [PubMed: 9730509]
- Favretti F, Cadiere G B, Segato G, Bruyns G, DeMarchi F, Himpens J, Belluco C, Lise M. Laparoscopic placement of adjustable silicone gastric banding: early experience. Obesity Surg. (1995);5:71–73. [PubMed: 10733797]
- Karlsson J, Sjostrom L, Sullivan M. Swedish obese subjects (SOS) - an intervention study of obesity. Two-year follow-up of health related quality of life (HRQL) and eating behavior after gastric surgery for severe obesity. Internat J Obesity & Related Metab Disorders. (1998);22:113–126. [PubMed: 9504319]
- MacLean L D, Rhode B M, Nohr C, Katz S, McLean A P. Stomal ulcer after gastric bypass. J Am Coll Surg. (1997);185:1–7. [PubMed: 9208953]
- Marceau P, Hould F -S, Simard S, Lebel S, Bourque R -A, Potvin M, Biron S. Biliopancreatic diversion with duodenal switch. World J Surg. (1998);22:947–954. [PubMed: 9717420]
- Mason E E. Starvation injury after gastric reduction for obesity. World J Surg. (1998);22:1002–1007. [PubMed: 9717429]
- Mason EE (1999) Ilial transposition and enteroglucagon/GLP1 in obesity (and diabetic?) surgery. Obesity Surg 9: in press . [PubMed: 10484306]
- Mason E E, Doherty C. Vertical banded gastroplasty “How I do it” Digestive Surg. (1997);14:335–360.
- Mason E E, Doherty C, Cullen J J, Scott D, Rodriguez E M, Maher J W. Vertical gastroplasty: Evolution of vertical banded gastroplasty. World J Surg. (1998);22:919–924. [PubMed: 9717417]
- Mason E E, Hesson W W. Informed consent for obesity surgery. Obesity Surg. (1998);8:419–428. [PubMed: 9731677]
- Mason EE, Ito C (1967) Gastric bypass in obesity. Surg Clinics N Am 47:1345–1351; reprinted (1996) as a classical article in Obesity Res 4: 316-319 .
- Mason E E, Tang S, Renquist K E, Barnes D T, Cullen J J, Doherty C, Maher J W. NBSR Contributors. A decade of change in obesity surgery. Obesity Surg. (1997);7:189–197. [PubMed: 9730547]
- Naslund E, Backman L, Holst J J, Theodorsson E, Hellstrom P M. Importance of small bowel peptides for the improved glucose metabolism 20 years after jejunoileal bypass for obesity. Obesity Surg. (1998);8:253–260. [PubMed: 9678192]
- Pories W J, Swanson M S, MacDonald K G, Long S B, Morris P G, Brown B M, Barakat H A, deRamon R A, Israel G, Dolezal J M, Dohm L. Who would have thought it? An operation proves to be the most effective therapy for adult-onset diabetes mellitus. Ann Surg. (1995);222:339–352. [PMC free article: PMC1234815] [PubMed: 7677463]
- Scopinaro N, Adami G F, Marinari G M, Gianetta E, Traverso E, Friedman D, Camerini G, Baschieri G, Simonelli A. Biliopancreatic diversion. Word J Surg. (1998);22:936–946. [PubMed: 9717419]
- Toppino M, Morino M, Capuzzi P, Mistrangelo M, Carrera M, Morino F. Outcome of vertical banded gastroplasty. Obesity Surg. (1999);9:51–54. [PubMed: 10065583]
Edward E Mason, M.D.
Mason EE. Gastric surgery for morbid obesity. In: Holzheimer RG, Mannick JA, editors. Surgical Treatment: Evidence-Based and Problem-Oriented. Munich: Zuckschwerdt; 2001.