Fat stores increase when energy intake exceeds energy utilization; they decrease when utilization exceeds intake. Although that is invariant physics, the reasons why some people form more fat cells or more efficiently fill their fat stores is more obscure. Irrespective of why a variation in fat storage occurs, excess storage is a disease with high comorbidity and wide prevalence in the industrialized world.

Seventy per cent of obesity may be hereditary. The genetic contribution to obesity is great enough to enable infants who burn energy sources efficiently at 3 weeks of age to be discriminated from those who do not and who will be overweight at 3 months of age. In a study of adopted Danish children the major correlate of body weight was the weight of biological parents. In Swedish children familial obesity and the degree of overweight in puberty were the best predictors of adult overweight and excess mortality. Southwestern American Indians (the Pimas) are genetically predisposed to converting ingested calories to fat and to have a genetic inability to generate heat after a meal or during exercise. Studies of twins and families in Canada have conclusively demonstrated that the capacity for thermogenesis is inherited.

Although a fat person seeking to lose weight is always fighting his genes, he is not unable to reduce his weight. Historically, the masses of people were thin: European peasants and artisans of the 18th century lived on the edge of starvation, eating bread and grains as their chief foods. Surprisingly few English and Scots ate cheese, milk, and eggs, even when these were available. In the 18th century, the average consumption of meat was about 240 g per year, chiefly on feast days, a notable exception in the 16th century being a cannibalistic family of Scottish brigands who subsisted on travellers to the point of hanging their flesh to cure.

Obesity in times past must have been more often a consequence of Falstaffian indulgence in alcohol than of overeating. Shakespeare recognized that the grave did ‘gape thrice wider’, for Falstaff ‘than for ordinary men’: an increased mortality rate for the grossly obese could naturally explain the leanness of the masses. Social stigmatization and ostracism, then as now, probably contributed to underestimation of the prevalence of obesity.

When food became more easily available, Victorian embonpoint was the reaction. Curves were desired; angles were anathema. Today the main reason why ‘overweight’ people want to lose weight is to conform to the angular fashion of the time. Although women are more concerned with that goal than men, for both sexes appearance is a more important stimulus for regulating body weight than is the potentiation of cancer, cardiovascular disease, non-insulin dependent diabetes, renal disease, disorders of respiration and sleep, pseudotumor cerebri, joint disease, masking of intra-abdominal disease, and interference with the activities of daily living caused by obesity (Fig. 1) 1010. Indeed, ‘life changes’ —meaning stressful events initiating the search for a new start—often precipitate searching for a surgical treatment for obesity.

Obesity inflicts handicaps other than disease and appearance. Obese men in Denmark have lower social standing, and, hence, lower income, than men of normal weight. As a determining variable, obesity is independent of parental social class, intelligence, and education: only 30 per cent of obese Danish men reach the social class of a semiskilled labourer or lorry driver, compared with 51 per cent of normal-weight controls. Several reasons for this situation have been proposed: a negative attitude of normal weight people toward the obese, of personal conflicts in the normal-weight people who would be fat in the absence of coercion to be thin, and the public's stereotype that fat people have lost control of themselves. Negative attitudes toward obesity, documented as early as kindergarten, continue into adult life.

Life insurance statistics unequivocally demonstrate increased mortality in those who are overweight as well as in those who areunderweight. The criticism that such studies are flawed by being limited to people who can afford insurance has been amply met by large national population studies in Norway, the United Kingdom, and the United States. Obesity is an independent risk factor for cardiovascular disease, diabetes, respiratory insufficiency, some forms of cancer, stroke, and sudden death. The risks of comorbidity increase exponentially with increasing weight in all groups studied: young men aged 23 to 34 years who were twice or more ideal weight had a 12-fold increased incidence of death in a Veterans' Administration study in Los Angeles.

Data emerging during recent years demonstrate that the distribution of fat is of greater importance than is the magnitude of obesity. People whose deposits of fat are in the trunk or abdomen (android obesity) have greater morbidity and mortality rates than those with a more typically female distribution on the hips and buttocks. In fact, the distribution of adipose tissue, which is hereditary, is a determinant of dietary thermogenesis and resting energy requirements. Sex steroids and hepatic metabolism are strongly implicated in the morbidity associated with abdominal distribution of fat.

People become so desperate to look as they wish they did—or as others think they should—that they will resort to operations to lose weight or to lose specific areas of fat. This point is reached after lesser remedies have been exhausted, including, sometimes, supervised diets of 400 to 600 calories daily for many months in a hospital losing weight, only to regain and overshoot their former weight in a short period of time.

The logical surgical extension of the semistarvation approach is to keep the fat person's jaws closed with interdental wires to prevent eating. This method generally works, but only so long as the wires remain in place. Once the wires are removed, the rate of recidivism is over 90 per cent, as it is in every other form of self-controlled dietary programme or after use of intragastric balloons. As well as the potential problems of dental damage there is also non-compliance despite mandibular wiring: consumption of high calorie liquids is not impeded by dental fixation alone.

Other surgical remedies fall into two main categories: procedures causing malabsorption, and those restricting intake of food, though the common gastric bypass procedures do both.

For a patient who has failed all other treatment under the directions of a skilled family practitioner or internist, criteria appropriate for being accepted into a program of bariatric surgery (baros = weight [Greek]) are, firstly, morbid obesity—a patient at least 45 kg above the weight for height specified by charts of insurance companies, or the presence of severe obesity-related morbidity. The heaviest patients seeking surgical treatment, the ‘superobese’, weigh around 200 to 300 kg; typical mean weights are around 130 kg in women.

Secondly, patients should preferably be less than 50 years old: the number of complications increases with age, and the amount of weight lost is inversely proportional to age. Established habits of eating are difficult to break after the age of about 35, and levels of physical activity decrease. On the other hand, lack of sufficient maturity and understanding usually excludes patients in or below the early twenties.

Thirdly a normal psychological balance, and ability to co-operate and understand the importance of follow-up care is also important. Psychiatric consultation should be obtained when there is doubt. A history of psychiatric hospitalization, remarkably, is a predictor of medical complications of the surgery. Alcoholism is a contraindication. Fourthly, the patient must have a commitment to adhere to regular follow-up visits to his physician and surgeon, ideally for his lifetime. A surgical setting in which patients are part of a long-term investigative programme and in which surgeons perform at least two such operations a month is the ideal, as is a hospital with the multidisciplinary resources to manage the perioperative and postoperative needs of the morbidly obese.

Operations that cause malabsorption or speed the passage of food, or both, are intestinal bypass and, to a certain extent, gastric bypass. Operations limiting the amount of solid food that can be ingested or passed through the upper intestinal tract are the various methods of stapling the stomach and of restricting the size of the gastric orifice through which food must pass before it can reach the small intestine. Gastric bypass does both.

However well these operations are performed, failures are inevitable. The walls of the stomach are smooth muscle; hence they expand with time; compensatory hypertrophy and hyperplasia of the intestine increase absorptive capacity technical failures of staple lines and stomal reinforcement occur and, probably most important, (mal)adaptive eating behaviour by the patient defeats the purpose of the operation. There is, therefore, less correlation between weight loss and the size of stomas and pouches than there might be if only laws of physics prevailed.

Obese patients are at increased risk of perioperative complications such as wound infections, herniae, pneumonia, and thromboembolism. Development of techniques for preventing these complications have appreciably reduced these risks in centres specializing in this type of surgery (Table 1) 325, and these techniques are applicable to any operation on obese patients.

The processes listed above have variable effects, and similar patients given apparently identical operations can have disparate results. Long-term assessments can be difficult because patients tend to move away and to take up new lives, whether fat or not. Thus, for intelligent assessment and application of bariatric surgery, most of it should be done by surgeons committed to well-controlled trials of all the variables.

The original bariatric operations were inspired by observations of weight loss in patients with extensive resections for vascular compromise, inflammatory bowel disease, or cancer. Unwillingness to perform irreversible procedures for obesity led to abandonment and development of a host of bypass operations. The severe debility and hepatic dysfunction following jejunocolostomy, for example, quickly made this procedure obsolete.

A short-bowel syndrome can be created by joining the end of about 35 cm of proximal jejunum either to the side of the distal ileum about 10 cm proximal to the ileocolic junction or to the end of the transected ileum, attaching the blind stump of bypassed intestine to the colon. Despite this prescription and adoption of other formulae of length and mode of bypass, none gives consistent results. Although almost every patient loses appreciable weight, and about 15 per cent reach ideal weight, the complications of hepatic dysfunction (presumably from nutritional cirrhosis, portal bacteraemia, or both); hypokalaemia, hypocalcaemia, and hypomagnesaemia; renal lithiasis from fat malabsorption, oxaluria, and dehydration; cholelithiasis; a peculiar form of enteroarthritis; pneumatosis intestinalis; and prevalent diarrhoea are considered to be too high a price to pay, though patients uniformly are more satisfied than physicians. Weight loss following these operations isprobably most dependent upon aversive conditioning with concomitant decreased food intake—every episode of gorging being followed by diarrhoea, cramping, and flatulence.

With knowledge of management and prevention of complications and technical improvements, there is possibly still a minor role for jejunoileal bypass in selected patients: for example, a 300-kg patient who ideally requires a bariatric operation on the stomach might be brought to a more manageable weight for gastric surgery by preliminary jejunoileal bypass.

These operations have not been widely accepted. They aim at producing malabsorption by diverting the flow of bile or pancreatic juice, or both to a more distal site.

As originally developed, this operation consisted of resection of at least two-thirds of the stomach, emptying being provided by Roux-en-Y gastroileostomy. This short circuit bypasses half of the small intestine, which is drained into the distal ileum 50 cm before the ileocaecal valve. It is aimed at producing maldigestion and malabsorption; indeed, it predisposes to frank protein–calorie malnutrition unless it is meticulously managed. The biliopancreatic bypass has been modified by excluding most of the stomach without resecting it, as in gastric bypass (below). Biliopancreatic bypass procedures must be reserved for very highly selected patients in research centres.

Using an upper midline incision, a small gastric pouch is created by firing a surgical stapling apparatus across the stomach, immediately below the oesophagogastric junction. The gastric pouch is drained by a Roux-en-Y gastrojejunostomy.

A large self-retaining retractor system fixed to the operating table provides exposure. A surgical stapler puts four parallel rows of staples across in a single firing of the device. Because there has to be a means of egress of food and secretions from the stomach, a Roux-en-Y limb of jejunum is joined to the pouch. Deficiencies of iron, calcium, thiamine, vitamin B&sub1;&sub2;, and of folic acid must be prevented after gastric bypass. Marginal (stomal) ulcers are rare complications.

This operation aims to create both a small gastric pouch and an outlet that does not dilate. These goals are accomplished by placing four rows of staples parallel to the lesser curvature of the stomach, creating a gastric tube with a capacity of 15 ml or less. A 32 F Ewald orogastric tube passed by the anaesthetist serves as a template for creating the pouch. Banding of the outlet is done by stitching a 15-mm wide strip of polypropylene or nylon mesh to itself through a hole stapled through both stomach walls using an entero-entero-anastomotic type stapler (EEA® ) with a diameter of 25 or 28 mm. In the latest modification of the vertical banded gastroplasty, the length of the strip is 4.75 cm. A 4.5-cm circumference adds complications, but does not produce superior weight loss.

The Silastic®ring vertical gastroplasty is an improvement over the vertical banded gastroplasty. It avoids the creation of a stapled hole through the stomach, which has occasionally been the site of fatal postoperative leaks. A nylon suture passed through a silicone tube is stitched through both walls of the stomach adjacent to the lowest portion of the vertical staple line. Because of the risk of erosion of the foreign material in the band or ring into the stomach, we band the outlet with strips of linea alba fascia harvested before entering the peritoneal cavity.

Staple-line disruption, the other technical complication of the procedure, has been vastly diminished since the introduction of broad, four-row staplers. The most serious complications are perforations of the oesophagus and stomach, either from lacerations during freeing of the oesophagogastric junction or from perforations or leaks caused by the stapling. Mortality rates are generally below 1 per cent in centres specializing in this type of surgery.

Failure of weight loss after gastric surgery for obesity is troubling. Patients who eat sweets habitually lose less weight after gastroplasty than after gastric bypass (55 per cent versus 70 per cent of excess weight). Presumably the aversive conditioning from the dumping syndrome after gastric bypass helps reduce intake of calorically dense sweet liquids. Comparisons between purely gastric restrictive operations and gastric bypass have consistently revealed greater weight loss with gastric bypass. The bypass operations take longer, entail more perioperative complications, and cause more long-term deficiencies. Reoperations for failure of weight loss after gastric restriction are hazardous. The 40 to 50 per cent of gastroplasty patients who fail to lose weight over the long term and who require reoperation should have an operation with a component of malabsorption, such as gastric bypass or even an intestinal bypass.

The gallbladder is often removed during any bariatric operation because of the risk of developing cholecystitis or forming stones during weight loss. Treatment with ursodeoxycholic acid can prevent gallstones from forming during weight loss. The gastric restrictive procedures in themselves, as well as the weight loss they achieve, are effective against gastro-oesophageal reflux disease in hiatus hernia. Some surgeons remove a fibroid uterus, or one that is otherwise abnormal or perform tubal ligation, or do both, at the time of bariatric surgery. Patients with a large panniculus adiposus (‘apron’) can benefit from lipectomy and abdominoplasty at the time of their bariatric procedure. Many patients require reduction of redundant skin and adipose tissue after weight loss.

The goal of bariatric surgery, to reduce weight, is fulfilled at least temporarily in the majority of patients. Around 30 per cent of preoperative body weight is lost during the first 1 to 2 years after most types of procedures. Long-term maintenance, however, is elusive, as many patients disappear from follow-up examination.

From a medical standpoint, however, the surgery is a resounding success. Not only is it vastly superior to diet, exercise, drugs, hypnosis, behaviour modification, or any non-operative treatment, but it dramatically improves diabetes, hypertension, respiratory insufficiency, and the host of comorbid conditions associated with obesity. Well-controlled studies are needed to determine the effects on longevity. Patients attest to the fact that quality of life is better.

Broomfield PH, et al. Effects of ursodeoxycholic acid and aspirin on the formation of lithogenic bile and gallstones during loss of weight. N Engl J Med 1989; 319: 1567–78.

Carey LC, Martin EW, Mojzisik C. The surgical treatment of morbid obesity. Curr Probl Surg 1984; 78

Eckhout GV, Willbanks OL, Moore JT. Vertical-ring gastroplasty for morbid obesity. Five year experience with 1463 patients. Am J Surg 1986; 152: 713.

Hall JC, et al. Gastric surgery for morbid obesity. Ann Surg 1990; 211: 419–27.

Harrison GG. Height-weight tables. Ann Intern Med 1985; 103: 989–94.

Hirsch J, Leibel RL. New light on obesity. N Engl J Med 1988; 318: 509–10.

Ismail T, Kirby RM, Crowson MC, Baddeley RM. Vertical silastic ring gastroplasty: a 6-year experience. Br J Surg 1990; 77: 80–2.

Knapp VJ. Major dietary changes in nineteenth-century Europe. Persp Biol Med 1988; 31: 188–93.

Kral JG. Morbid obesity and related health risks. Ann Intern Med 1985; 103: 1043–7.

Kral JG. Surgical treatment of obesity. Med Clin N Am 1989; 73: 251–64.

MacLean LD, Rhode BM, Forse RA. Late results of vertical banded gastroplasty for morbid and superobesity. Surgery 1990; 107: 20–7.

Mason EE. Morbid obesity: Use of vertical banded gastroplasty. Surg Clin N Am 1987; 67: 521–37.

Mossberg H-O. 40-year follow-up of overweight children. Lancet 1989; ii: 491–493.

Naslund I. The size of the gastric outlet and the outcome of surgery for obesity. Acta Chir Scand 1986; 152: 205–10.

Pasulka PS, et al. The risks of surgery in obese patients. Ann Intern Med 1986; 104: 540.

Roberts SB, Savage J, Coward WA, Chew B, Lucas A. Energy expenditure and intake in infants born to lean and overweight mothers. N Engl J Med 1988; 318: 461–6.

Sonne-Holm S, Sorensen TIA. Prospective attainment of social class of severely obese subjects in relation to parental social class, intelligence, and education. Br Med J 1986; 292: 586–9.

Stricker EM, Verbalis JG. Control of appetite and satiety: insights from biological and behavioral studies. Nutr Rev 1990; 48: 49–56.

Stunkard AJ, et al. An adoption study of human obesity. N Engl J Med 1986; 314: 193–8.

Sugerman HJ, et al. Weight loss with vertical banded gastroplasty and Roux-Y gastric bypass for morbid obesity with selective versus random assignment. Am J Surg 1989; 157: 93–102.

Vasudeva R, Holt S, Taylor TV. Obesity: medical and surgical management. Curr Opin Gastroenterol 1989; 5: 865–9.

Weststrate JA, et al. Resting energy expenditure in women: Impact of obesity and body-fat distribution. Metabolism 1990; 39: 11–17.