Chapter 27. Morbid Obesity and Its Surgical Treatment

Obesity looms as the single largest threat to world health in the next few decades. In countries such as the United States, its consequences pose the very real likelihood that the next generation may not live longer or be more healthier than the previous one.1 This would reverse a trend that has been present for centuries. The rate of obesity is rising worldwide, not just in countries that enjoy privileged economic status. The rise in obesity over the past 25 years in the United States has been dramatic. Currently nearly one-third of adults in the United States are obese, defined as having a body mass index (BMI, calculated as being weight in kg divided by height in meters squared) of 30 kg/m2 or greater.2 More concerning yet, the rate of obesity in young adolescents and teenagers is approaching or exceeding the adult rate in many geographic areas. Because obese adolescents have a very high likelihood of being obese adults, this predicts that the problem will continue to grow in terms of its consequences on the health of the population.

Obesity was not a major health problem in many areas of the world, such as Asia and Africa, until the past decade. Now even those countries, where the problem was previously rare, are experiencing a significant increase in its prevalence. It is likely that wider access to high-calorie fast-food meals and other higher-calorie foods from Western countries, combined with the decreased need for physical labor and activity with increasing mechanization present in these countries, are significant contributing factors. The flattening of the world has also led to its fattening.

Obesity also remains the only major characteristic or attribute for which discrimination is not illegal. Laws exist to prevent discrimination on the basis of gender, sexual preference, race, religion, or handicapped status. However, there are no laws to prevent the current and prevalent discrimination against obesity in the workplace, in travel, in accommodations, and in other areas of life that are often overlooked by nonobese individuals. Most damaging, however, is the persistent belief by the majority of the public that obesity stems from laziness and gluttony, rather than being a disease. Even more sadly, there are still medical care providers who hold such opinions. The Centers for Medicare and Medicaid Services (CMS) has officially recognized obesity as a disease and sanctioned its treatment as appropriate for recipients of federal insurance.3 Unfortunately, the insurance industry overall, including those companies assigned to administer services to federally insured patients and the federal administrators as well in certain situations, has consistently raised barriers and made it difficult for patients with obesity, and particularly those with severe obesity, to obtain optimal treatment for their disease.4 While such short-sighted behavior may save money for their balance sheets in the short term, it will not make the problem go away. The next generation worldwide will be required to pay the price of addressing the needs of the population with this disease.

One feasible reason for the aversion of insurance carriers toward funding surgical therapy for severe obesity is increased costs. The cost of obesity is now a huge part of the health care budget and rising more rapidly than the overall high rate of increasing health care costs. It is estimated that in 2007 the direct cost in dollars for treating obesity was $93 billion or over 9% of all direct health care costs.5 That cost is likely higher today. In addition, such a number is based only on direct hospital or physician charges for diagnosis codes of obesity and related conditions. They do not include the increased cost that obesity confers to many other situations. An example would be a severely obese individual with a large incisional hernia. That person has a much higher risk of the recurrence of the hernia, the need for further operative care, and even for more severe complications such as bowel injury, fistula, or obstruction. Yet these costs are not at all calculated in to the cost of obesity, because incisional hernia is the problem. Similarly, the huge cost annually to the population of diets and other nonmedical-related costs of obesity are also not included in this figure.

This text focuses more on the surgical therapy of obesity than medical treatment options. As such, it also focuses largely on the individuals of the obese population that are surgical candidates. Table 27-1 defines the categories of obesity that we follow in this chapter. The term severe obesity will be used in preference to morbid obesity, as recommended by the National Institutes of Health (NIH) Consensus Conference that also established the still adhered to guidelines for the indications for surgical therapy for obesity.6 Arguments exist that, probably correctly, call for a reexamination of the data regarding bariatric surgery and revision of these quite old NIH guidelines, established long before laparoscopic surgery, and when vertical banded gastroplasty was still a commonly performed operation in the United States.7

Similarly, arguments have been voiced that BMI is not an adequately accurate predictor of health deficits related to obesity.8 A highly muscular individual could have a high BMI but very little body fat, for example. Thus the risk to his or her health would not be nearly as great as the individual with the same BMI but high-body-fat content. While both such arguments have validity, and this is hereby recognized, barring the creation of newer NIH guidelines or the creation of a better and more easily calculated measure of obesity than BMI, this text accepts current use of current NIH guidelines and BMI as the gold standards for indications for bariatric surgery and measurement of obesity respectively.

Obesity is a poorly understood disease. Currently much investigation is ongoing to try and determine the aberrancies that occur both at a cellular level and at the level of the intact individual to determine its manifestations. It is hoped that such research will provide insights as to treatments for the metabolic consequences of obesity, as well as treatments for the metabolic causes of it. Until such time, however, individuals who develop obesity are highly susceptible to also developing one or more of the medical illnesses associated with obesity. These are listed in Table 27-2. It is these comorbid medical problems that jeopardize the length and quality of the lives of individuals with obesity and especially severe obesity. It is estimated that a male who is severely obese from childhood will live an average of 12 years less than his nonobese counterpart, while for a woman the difference is 9 years.9 There is some variability by race and sex as to the consequences of obesity on health. The consequence of being a male patient with severe obesity and its limitation of lifespan is shown in Table 27-3, which gives the percentage of the population in the United States with severe obesity based on age and sex. Note that old males do not represent as high a percentage of the population as females, largely because they are already dead of their comorbid medical diseases. Flum et al10 showed that for males with Medicare insurance (and hence disabled if under age 65) undergoing bariatric surgery, the average mortality during the year after surgery was 6.4% for males age 18–65.

In addition to medical illnesses commonly associated with obesity, there is an increased incidence of certain types of malignancy in obese individuals. Cancers of the uterus, breast, prostate, and pancreas are all increased in this patient population.

Finally, among the major reasons for individuals to seek surgical therapy for severe obesity, lifestyle issues are often more important than medical problems. The loss of ability to perform common daily activities, the inability to participate fully as a parent, the lack of ability to perform on the job, and the social discrimination against obesity are all often cited by patients as the primary reason they seek surgical therapy for this condition.

Deciphering the likely multifactorial causes of the pathophysiology of obesity has given rise to several lines of investigation. Alterations of metabolism at the cellular level, genetic predispositions and patterns, and environmental influences all likely are active in contributing to the disease process etiology and mechanisms. The focus of this text is not on that subject, but a few observations from many years of clinical practice are offered. Certainly the alteration in satiety has to be forefront among the abnormalities in individuals with severe obesity. Appetite is often insatiable in these individuals, despite high-calorie intake daily. Body fat distribution is known to affect the incidence of comorbid problems. For example, the incidence of the metabolic syndrome is much higher in individuals with central obesity than those with pear-shaped body habitus. Higher amounts of organ fat and omentum are associated with conditions such as metabolic syndrome and diabetes. Genetic analysis of the different properties of adipose tissue is ongoing. Considerable investigation has been focused in the past 5 years on the mechanisms of rapid improvement of type 2 diabetes after Roux-en-Y gastric bypass. Such mechanisms almost certainly involve alterations in glucose metabolism by peripheral tissues based in turn on the altered pathway of food through the upper gastrointestinal tract. This gut influence on glucose metabolism is now under intense scrutiny. Hopefully one day the mechanisms of appetite regulation, satiety, and metabolism of adipose tissue will be better understood. When they are, perhaps we will have a better understanding of the complex disease process that is obesity.

All individuals with the problem of obesity need to make modifications in their eating habits, lifestyle, and exercise habits as a lifelong commitment to overcoming the disease. For individuals with class 1 obesity, such modifications have the possibility of altering weight and in turn altering the health risk of comorbid medical problems enough to make a difference in the toll that comorbid medical problems take on their life and health. However, as the amount of obesity increases, modest weight loss from medical therapy is less likely to make a profound difference in health. It is also less likely to be successfully sustained. The figure generally given for the likelihood of a severely obese individual successfully losing enough weight by dieting to become only obese or overweight is approximately 3%. In short, medical therapy is highly unlikely to be successful in reversing the problems of severe obesity. Nevertheless, because medical therapy has little risk involved (except the time lost from the benefit of potentially more effective therapy), it is uniformly agreed that a trial of medical therapy should occur first in any individual with morbid obesity prior to the use of surgical therapy. Unfortunately, this practice is not data-driven, but driven instead by the practicality of attempting the less risky and invasive therapy first prior to one that is more invasive. In reality, there are extraordinarily few patients who seek bariatric surgery who have not had multiple attempts at dieting during their lifetime. The motivation of those few individuals who have never dieted must be questioned, as they often perceive of surgery as a “magic bullet” which will eliminate obesity but not require any other changes in eating habits or lifestyle to produce or sustain these effects.

The role of medical therapy in the sense of dieting, exercising, and otherwise attempting to control weight by nonsurgical means is often not addressed further in texts on bariatric surgery beyond the simple statement that it is effective therapy in a minority of cases. However, it is our opinion that no surgical therapy can hope to have long-term effects without an accompanying change in eating and exercise habits by the recipient of the operation. Even the most effective bariatric operation can have its weight loss benefits mitigated over time by the patient who will not adapt better eating and exercise habits. Fortunately, the physical and mental metamorphosis that accompanies the postoperative period after bariatric surgery is usually profound enough in most patients to reinforce the need to make such changes to preserve this alteration in body habitus and health. The typical mindset of the long-term successful postbariatric surgery patient is that he or she will never allow themselves to go back to being severely obese again, now that they have been given the chance to be relatively normal in weight, had their medical diseases go into remission, and enjoy a lifestyle free of the burdens of obesity. Such a mindset is usually successful in maintaining the benefits of bariatric surgery. When patients do not have enough metamorphosis or if they regress to poor eating habits and stop exercising, erosion of the benefits of the operation is inevitable and they eventually join the estimated 30% of patients who regain weight and ultimately are considered to have failed bariatric surgery. The exact formula for bariatric surgery success is not well defined, but it includes selection of a motivated patient, success of the operation to produce significant physical, mental and medical changes, and the persistence of the patient in maintaining exercise and diet habits to preserve these changes. Thus surgical therapy is truly only part, though the pivotal part, of the long-term success of therapy for the patient with severe obesity. Attention to the follow-up of patients, maintaining motivation to sustain appropriate exercise and diet habits, and any other such supportive measures that can be done postoperatively can all help ensure long-term success of bariatric surgery. In this sense, the “medical” treatment of obesity is quite important as an adjunct to maintaining the benefits achieved by surgical weight loss.

The basic guidelines generally adhered to for performing bariatric surgery were defined by the NIH Consensus Conference in 1991.6 Eligible patients must have a BMI of greater than or equal to 40 kg/m2 or a BMI of greater than or equal to 35 kg/m2 and the presence of a disease caused by or exacerbated by obesity such as diabetes, hypertension, or other comorbidities. Beyond these basic criteria, most centers advocate that the patient have the ability to understand the planned procedure and the major changes it will cause in eating and the required changes in diet and exercise to optimize results. The patient should be appropriately motivated. Areas less well defined, but which usually have some limitations from center to center, include age, upper limit of weight, substance abuse, psychiatric history and problems, compliance problems, ambulatory status, and severity of comorbid medical conditions.

While all patients should be given information on the types of available operations, some operations may be more appropriate or effective or feasible as the procedure of choice for a patient, depending on the individual circumstances. The expertise and ability of the bariatric surgeon is another factor that affects operative choice for many patients. The relative advantages and disadvantages of the various operative procedures are discussed within their following respective sections.

Unfortunately, often the operative procedure that a patient undergoes for weight reduction is governed by the procedures that his or her insurance company will cover. Access to surgery for patients is currently limited by numerous factors. Many insurance companies have set a variety of preoperative requirements for patients otherwise medically qualified for bariatric surgery. These include the need to follow a medically supervised diet for 3–12 months, the need for a psychological evaluation, and other measures. Available evidence suggests that these requirements add no benefit to operative outcomes, and if anything delay potentially helpful surgical intervention.4,11 Their presence does serve as a deterrent for patients who may not be fully motivated or otherwise do not have the ability to meet these additional requirements. Some insurance companies will require a psychological evaluation, for example, but not cover the cost of such an evaluation.

Information about bariatric surgery and the availability of bariatric surgeons to the public as well as to the referring physician is now much more easily obtained than even a decade ago, and certainly much more available than two decades ago. Internet websites of bariatric societies, physician provider networks, hospital providers, and others all offer information on available surgeons. Use of search engines will usually produce surgeons in a prospective patient's geographic area.12

Access to care is still an issue for some patients with certain forms of insurance coverage. Some patients have policies that exclude bariatric surgery. Others have policies that include it only if the patient or their employer pays a significant fee for a rider to the policy. Even if a patient has coverage, some insurance policies, such as Medicare, reimburse the surgeon at such a low rate that only surgeons serving on the staff of public health care institutions will offer surgical care to such patients.

Once a patient is seen by a bariatric surgeon who will offer surgical services, the choice of operation is usually determined by a combination of any insurance restrictions, procedures offered by that surgeon, and patient interest. Medical comorbidities and conditions can affect operative choice as well. Limitations of indications of the various bariatric operations in terms of efficacy and overall outcomes and effectiveness are discussed with each individual operation in the following text.

Patients who are preparing for bariatric surgery need preparation for surgery in two major areas:

1. Specific knowledge about the planned bariatric operation and its expected outcomes, course, and potential complications and side effects

2. General preparation for a major surgical operation, including maximization of treatment of existing comorbid medical problems

Specific knowledge of the planned bariatric operation allows the patient to prepare for the changes in diet, lifestyle, daily activities, exercise patterns, and body image that will occur after the operation. No bariatric operation will produce optimal long-term results without significant changes in diet, exercise, and lifestyle by the patient. Hopefully, the combination of the power and durability of the operation to force an effective change in eating patterns, a decrease in appetite, an increase in satiety with eating, and a change in exercise patterns by the patient will serve to create an adequate metamorphosis of the patient and his or her lifestyle to promote a mindset that is strong enough to sustain these changes and resist any intrinsic genetic or behavioral tendency for recidivism to previous eating patterns that resulted in the preoperative morbidly obese condition. This alteration of the patient's physical and mental state of being is inherently necessary for the long-term success of bariatric surgery. Fortunately, such an alteration is achieved and sustained in the majority of patients who undergo bariatric operations. A frequently heard quotation from patients who are years out from bariatric surgery and who have successfully maintained weight loss and reversal of their comorbid medical conditions associated with obesity is, “I will never go back to being the way I was before.”

While not essential, personal interaction by prospective patients with other patients who have successfully undergone the same bariatric operation they are planning to undergo can be very helpful in terms of reassurance for likely success, as well as information exchange. The more the prospective bariatric patient is educated and familiar with the expected postoperative course of events, the more likely they will be to have less preoperative anxiety and more compliance with recommended postoperative treatment plans. Such interactions can be facilitated by support groups, Internet chat groups, and personal relationships. The surgeon and his or her staff should always inquire as to the support network available to the patient after surgery and their potential interaction with other individuals who have undergone the same operation. In our practice, we become concerned when a patient is pursuing bariatric surgery in opposition to the wishes of their family and/or spouse. While such patients may be strong-willed enough to do well postoperatively, if any complications result, there is often a significant adverse reaction by the family in these circumstances. Certainly the patient's wishes should always prevail, but a word of caution in such situations is in order.

Laparoscopic Adjustable Gastric Banding

Procedure-specific preparation prior to surgery is always indicated. For patients planning to undergo Lap-Band (Allergan, Inc., Irvine, CA) procedures, we have found the following list of criteria tend to produce the optimal results after this operation:

1. Patient is able and willing to undergo the recommended schedule of band adjustments. This includes the coverage for such adjustments by their insurance carrier. Lack of such coverage inevitably leads to decreased compliance with suggested follow-up, and often less optimal results.

2. Patient is able and willing to undergo a regular exercise regimen. Our experience shows a strong correlation between exercise and postoperative weight loss for patients undergoing laparoscopic adjustable gastric banding (LAGB).

3. Patients expected and optimal weight loss is in the 100 lb or less range. We have found, and the literature supports, the fact that patients whose weight is over approximately 350 lb, and who have a BMI over 50–55, have a less optimal outcome after LAGB than those who have lower preoperative BMI and expected weight loss.

4. Patient can maintain a “dieter's mentality”. This is usually documented by the fact that the patient has had at least one previous episode in their life of successful dieting, sustained over greater than 6 months. While the obvious fact is that such a diet did not have long-term effectiveness due to the patient now planning LAGB, the fact that LAGB is a relatively, but not overwhelmingly, powerful suppressant of appetite in patients makes this requirement important. The patient who can diet successfully without any appetite suppression is likely to be able to maintain an appropriate eating pattern after LAGB, which suppresses but does not eliminate appetite in most patients.

5. The Lap-Band is an excellent operation for the patient whose BMI is 50 or less, and who has no severe comorbid condition better treated by an alternative operation (such as severe diabetes better treated with a gastric bypass). Patients who have a BMI over 50 may do well with a Lap-Band, provided they do exercise and the band provides adequate suppression of appetite to alter eating habits. Most series in the literature reporting LAGB outcomes have either a limited number of patients with BMI over 50 or have a decreased success rate of the operation in that patient population.

Laparoscopic Roux-En-Y Gastric Bypass

1. Patients who choose this as their bariatric operation of choice must be prepared to undergo a significant change in their eating pattern and ability. Laparoscopic Roux-en-Y gastric bypass (LRYGB) is a very powerful operation, usually eliminating appetite in most patients for a period of at least several months and altering the ability to intake food volume dramatically in most cases. Patients who rely on food as a significant psychological crutch may find the altered eating pattern forced on them to be a particular problem. However, most often the patient is delighted by the elimination of appetite and the rapidity of weight loss with concurrent resolution of comorbid medical conditions.

2. Patients planning LRYGB must understand that the operation will intrinsically limit their ability to absorb the critical nutrients of iron and calcium.

3. LRYGB is now the most commonly performed bariatric operation in the United States, and for good reason. The operation has the longest track record of continuous use, having first been performed in the late 1960s in slightly different form and of course using an open surgery approach in those days. Modifications to the operation have improved it, but not changed the basic essence of a largely restrictive operation with reduction in the gastric reservoir capacity to a minimal volume. In addition, the gastrojejunostomy of the proximal gastric pouch eliminates the pyloric sphincter's regulation of slowly delivering food to the small intestine, resulting in variably profound dumping syndrome in these patients. While the dumping is often ameliorated with time, its initial presence usually is instrumental in forcing a behavioral modification in eating habits by patients, which involves the avoidance of highly concentrated sweets in their diet.

4. Other benefits generated by the anatomic alteration of the gastric bypass include a profound improvement in insulin sensitivity in patients with type 2 diabetes due to elimination of food substances passing through the duodenum. Gastroesophageal reflux disease (GERD) is also resolved in over 90% of patients after gastric bypass, because the volume of available gastric acid or secretions from the diminutive proximal gastric pouch is minimal. The operation thus is particularly indicated or effective for severely obese patients who suffer from GERD or type 2 diabetes.

5. It is an effective operation for the patient who needs to lose a significant amount of weight and is not a candidate for a malabsorptive operation. Overall, gastric bypass avoids the majority of the metabolic complications inherent in malabsorptive operations, yet is powerful and durable enough to provide good weight loss even in superobese individuals.

6. Gastric bypass has few contraindications. It does limit iron absorption, so it may not be a good choice for the patient with preexisting significant iron deficiency anemia. In populations where there is an appreciable incidence of gastric cancer, it may also not be a good choice as it eliminates the ability to monitor the majority of the stomach for potential development of that disease.

7. Performing gastric bypass using a laparoscopic approach has resulted in drastic reduction in the most common postoperative complication after open Roux-en-Y gastric bypass (RYGB)—that of incisional hernia. Wound infections are less severe. Benefits established for laparoscopic versus open operations of other types also apply to LRYGB, such as quicker return to normal activities, decreased pain, and so forth.

Laparoscopic Sleeve Gastrectomy

1. The laparoscopic sleeve gastrectomy (LSG) is the most recent addition to the armamentarium of operations used by bariatric surgeons. Its adoption as a commonly performed operation is only evolving. Few centers had a large experience with the operation more than 3 years ago. The longest published follow-up data in the literature show only a small number of cases followed out to or beyond 5 years. The operation is, as with all primary bariatric operations, optimally performed laparoscopically, hence LSG.

2. LSG is a restrictive operation. It is technically easier to perform, in general, than LRYGB and avoids the risk of two anastomoses. It is more invasive than LAGB, as it involves resection of the majority of the stomach. Data to date show that the operation has a risk profile between that of LAGB (safest) and LRYGB for frequency of potentially severe complications.

3. Not all insurance companies currently cover LSG. Federal insurance and some other major private insurance companies still do not provide coverage for this operation.

4. The LSG originally had its birth as the first of a staged procedure to perform a laparoscopic duodenal switch (LDS) operation with less morbidity. Original experience with LDS showed a higher mortality rate than expected. LSG was then adopted as the first stage of the operation, to be followed by the malabsorptive component after initial weight loss.13 The success of the LSG caused many patients to decline the second step and led to the adoption of LSG as a primary procedure.

5. Concerns still remain among experienced bariatric surgeons that this operation bears a strong resemblance anatomically and functionally to the vertical banded gastroplasty, which was a very popular bariatric operation two to three decades ago. It has been largely abandoned for poor long-term weight loss results.

6. Candidates for LSG would be patients appropriate for a restrictive operation of almost any weight class above or including class 2 obesity. Because its initial use was in higher-weight individuals felt to be candidates for the DS procedure, LSG is technically feasible on very large patients, a positive factor from the bariatric surgeon's viewpoint. Published data show LSG is also effective as a primary operation for patients with BMI less than 50.14,15

Laparoscopic Duodenal Switch

1. This operation has been the least frequently performed of the commonly approved and accepted bariatric operations. It is technically the most challenging of the bariatric operations and is associated with both the highest durable weight loss as well as the highest frequency of complications, especially when metabolic complications are considered as well.

2. Laparoscopic duodenal switch (LDS) causes relative malabsorption of fats and proteins. Fat-soluble vitamins as well as the elements iron and calcium are not well absorbed. Supplements to replace these vitamins and minerals are indicated postoperatively. These supplements can be expensive.

3. Patients undergoing malabsorptive operations require close and long-term follow-up, due to the high potential for metabolic deficiencies and problems that may arise after such procedures.

4. Another malabsorptive operation, the biliopancreatic diversion (BPD), is now less frequently performed than the duodenal switch (DS), due to its higher incidence of marginal ulcers.16 Both operations involve comparable rearrangement of the alimentary tract distal to the duodenum, with the gastric resection for BPD being distal, while for DS it involves the greater curvature.

5. Most patients who are felt to be candidates for LDS are those with higher weights initially, who are felt to likely not achieve adequate weight loss with a restrictive operation alone. Patients with preoperative BMI over 60 certainly should be considered for such procedures if the patient is able to be closely followed and is willing and able to take required nutritional supplements. Patients interested in the most effective operation for large amounts of weight loss who are willing to accept the increased risk profile are candidates for LDS.

6. The technical difficulty of this operation as well as the concern for follow-up requirements has limited its performance to only certain centers specializing in this procedure or centers where it is offered as a choice among other bariatric operations.

Laparoscopic Adjustable Gastric Banding

Laparoscopic adjustable gastric banding (LAGB) for the treatment of morbid obesity was first described independently by Kuzmak in the United States and Halberg and Forsell in Sweden.17,18 The band is an inflatable silicone balloon placed around the proximal stomach connected by thin tubing to a reservoir implanted subcutaneously usually on the abdomen. The reservoir is accessed via a noncoring needle to adjust the diameter of the balloon. In 2001 the Food and Drug Administration approved for use the one band currently marketed by Allergan called the Lap-Band, and subsequently the Swedish band was added in 2007, which is now marketed by Ethicon and called the Realize band. The bands can be placed either laparoscopically or open although most authors will argue for the laparoscopic route. Recently the FDA also approved the use of the Lap-Band for use in patients with a BMI 30–35 with comorbid medical problems or a BMI greater than 35 without comorbid medical problems.

Operative Procedure

The preoperative evaluation and indications are similar for any bariatric operation and are described previously in the chapter. The patient should receive appropriate perioperative antibiotics as well as deep venous thrombosis prophylaxis. The patient will need to be positioned comfortably on the operating table with care taken to ensure the patient cannot slip when placed into steep reverse Trendelenburg's position.

Port number and position vary among surgeons. We describe what works in our practice. The pneumoperitoneum is established in the left upper quadrant, at the midclavicular line, one hand breath below the xiphoid. We prefer to use a Veress needle through an incision just large enough to admit a 15-mm trocar. We elevate the fascia with a tracheostomy hook that facilitates insertion of the needle into the appropriate space. Alternatively, access can be gained with an optical viewing trocar; this is again dependent on surgeon experience and preference. Once pneumoperitoneum is established, the abdomen is inspected for any evidence of injury. Additional trocars are placed. We place a 12-mm port for the camera in the midline approximately 15 cm from the xyphoid. A 5-mm port is placed in the right upper quadrant for the surgeon's left hand. A 5-mm port in the left upper quadrant is placed for the assistant and a port in the epigastrium is placed to assist in retracting the left lobe of the liver (Fig. 27-1).

The patient is placed in steep reverse Trendelenburg and dissection is begun with either the hook cautery or a harmonic scalpel at the angle of His. The peritoneum over the left crus and the gastrophrenic ligament is opened, and the dissection is carried down into the retrogastric fat. The gastrohepatic omentum is then opened in its avascular area, the pars flaccida. This allows exposure of the base of the right crus. If a hiatal hernia is appreciated, it should be repaired at this point and a standard posterior esophageal dissection is performed with suture closure of the crura. If no hernia is found, the peritoneum over the right crus is incised just wide enough to allow a grasper (Lap-Band) or the gold finger (Realize band) to pass (Fig. 27-2). It is important to note that these instruments should pass into the left upper quadrant in the previously dissected space with ease; any resistance indicates a wrong dissection plane. The grasper is passed just cephalad to the attachments of the proximal stomach to the retroperitoneum, thus remaining outside the lesser sac. Positioning the band in this space has decreased posterior slippage or prolapse of the gastric band. Once the instrument is visualized in the left upper quadrant, the band is placed into the abdomen through the 15-mm port. The tubing or suture is grasped and pulled through behind the stomach, pulling the band into place posteriorly (Fig. 27-3). The band is then buckled into its ring configuration (Fig. 27-4). Gastrogastric plication sutures secure the anterior fundus to the small portion of the stomach (optimally about 1–2 cm) above the band (Fig. 27-5). Usually two to three sutures are sufficient. It is important to not cover the buckle of the band as this may lead to band erosion. The tubing is removed from the abdomen through the epigastric or 15-mm port, depending on the surgeon's preference, and attached to the reservoir port. The reservoir port is secured to the abdominal wall fascia (Fig. 27-6). It is important to place the port in as thin a portion of the abdominal wall as possible to facilitate future access. The band system is accessed percutaneously with a Huber needle (Piwania Technologies Pvt. Ltd., New Delhi, India) to add and withdraw to confirm reservoir capacity and rule out leakage. All fluid is then removed from the reservoir at the end of the procedure, to be added later in adjustments postoperatively. It is recommended to begin with the band system empty so as not to have too much restriction initially for the patient.

Postoperative Care

The LAGB is now primarily performed as an outpatient procedure, unless medical or insurance issues require an overnight stay. Preoperatively and prior to discharge, the patient as well as family members should receive instructions on diet, activity, and pain medications. Instructions should be given on when and whom to call in case of emergencies. The patient is discharged on a liquid diet for 2–3 weeks. We usually see our patients back at this time to check the wounds and advance the diet. The diet is advanced to a soft diet and medications are checked as well as comorbidities. A multivitamin is recommended as sufficient supplementation as LAGB does not cause nutritional deficiencies seen in other bariatric procedures. Patients are often ready to return to work, if they have not done so already.

The success of LAGB depends on adjustments, adherence to dietary changes, and exercise. The timing of adjustments varies among surgeons; however, there is wide agreement that the goal rate of weight loss is 1–2 lb/wk. Less weight loss, later satiety, or larger meal size are indications that an adjustment may be in order. This is done after careful evaluation of the patients' weight, dietary history, and exercise.

Band adjustments are typically office-based minor procedures. On occasion fluoroscopic assistance may be required to access the port or evaluate the restriction of the band. There are several adjustment algorithms, with the amount of fluid added based on hunger, weight loss, ability to eat bread, and type of band.19 This is an area that is frequently more art than science.

Outcomes

Outcomes after LAGB are generally good in terms of weight loss and resolution of comorbidities versus significant adverse problems. The average weight loss for the larger series in the literature is usually reported for % excess weight loss (%EWL) at 1, 2, or multiple years after surgery. Generally speaking, patients undergoing LAGB will have a slower weight loss curve and take longer to achieve maximum weight loss than patients with either RYGB or LSG or DS. Figure 27-7 illustrates the weight loss curve for one such series.20 The %EWL at two years after surgery has been reported in the 45–55% range from various authors.21–23 Resolution of medical comorbidities is certainly seen after LAGB. Perhaps the most heralded study on this topic is the prospective randomized trial by Dixon et al24 in which the group of patients with type 2 diabetes showed as 73% remission rate of their disease 2 years after undergoing LAGB, as opposed to the control group that had a very low resolution rate. Other comorbidities are improved as well after LAGB and generally correlate with successful weight loss.

Optimal outcomes for patients with LAGB seem to be correlated with frequent band adjustments, participation in support groups, and regular physical exercise.25

Complications

Complications following gastric banding include prolapse or slippage, erosion, port or tubing complications, over-filling of the band, esophageal dilation, and weight loss failure.

Prolapse or slippage is the most common complication requiring reoperation. The mechanics of the process are that too much gastric tissue from below the band lumen pushes up through the band circumference (prolapse) or, similarly, the band slips down on the stomach further than desired, resulting in too much stomach above the circumference of the band (slippage). The effect of both is similar: the excess tissue causes almost immediate complete food intolerance if severe, or heartburn and moderate food intolerance if not severe. New onset of GERD symptoms in an LAGB patient strongly suggests prolapse, which should be ruled out.

Diagnosis of prolapse begins with the above clinical picture. A plain radiograph will usually show the band in an abnormally horizontal position (Fig. 27-8A). Barium swallow will show a significantly greater amount of stomach above the band than would be expected, confirming the prolapse (Fig. 27-8B). The prolapse can be anterior or posterior.

Initial treatment of the prolapse is withdrawal of all fluid from the band. This will often allow the prolapse to spontaneously resolve. Radiographic confirmation of this can be performed. If symptoms persist and radiographic evidence shows persistence of the prolapse, operative intervention to reduce the prolapse is indicated. This can usually be done laparoscopically. Often, the band must be unbuckled to allow full reduction of the prolapse. Repositioning the band and resuturing the fundic plication to maintain its position complete the operation.

Band erosion is an uncommon problem, occurring in 1% or less of most large series.26 Band removal and repair of the erosion, with appropriate antibiotic and supportive care, are indicated.

Esophageal dilation is perhaps one of the most severe complications that may result from LAGB. This complication arises when the band position is too high, restricting the distal esophagus instead of the proximal stomach. The incidence is in the 1–2% range in most series. Reflux, dysphagia, pain, and food intolerance may be presenting symptoms. Resultant dilation of the esophagus occurs. Esophageal motor dysfunction may occur if the condition becomes longstanding. Treatment for the problem, once discovered, is to immediately remove all fluid from the band, minimizing the restriction and obstruction. Hopefully this will reverse the dilation of the esophagus, and restore function. Band repositioning may be needed to prevent recurrence of the problem.

Port and tubing problems occur from 2–5% in most series.26 These are usually issues that can be repaired with procedures under local anesthesia or limited intervention under general anesthesia. Repositioning the reservoir that tilts to a position that does not allow access is a not uncommon problem. Care in attaching the reservoir to the fascia at the index operation is the best prevention of this problem.

Band overfilling is usually an easily correctable problem. When the adjustment is done, the patient may not have immediate symptoms. However, dysphagia ensues within the next day, and persists. Removal of all or most of the fluid added from the last adjustment will usually promptly resolve the problem.

Poor weight loss is, unfortunately, a problem with all bariatric operations. However, some centers have experienced an unusually high incidence of poor weight loss in their LAGB patients. We have found that attention to the criteria listed previously for selection has decreased but not eliminated the incidence of individuals with poor weight loss. While our institutional band removal rate is well under 5%, there are well over 15% of our patients who have had poor weight loss (as defined by <25% EBW) after LAGB. Some centers in Europe, which have had over a 15-year experience in LAGB at this time, are reporting an increasing incidence of abandoning the use of LAGB because of poor long-term efficacy.27 On the other hand, in Australia, LAGB is by far the main bariatric operation used, and the outcomes from that continent are generally the best published.28

Nutritional complications after LAGB are rare and are solely based on poor intake. LAGB does not alter the digestive process whatsoever, and hence there is no malabsorption of any nutrients. A standard multivitamin supplement is all that is necessary for patients following LAGB.

Laparoscopic Sleeve Gastrectomy

Laparoscopic sleeve gastrectomy (LSG) is the most recently recognized standard bariatric operation performed. The American Society for Metabolic and Bariatric Surgery recognized the procedure as being an appropriate standard operation for the surgical production of weight loss in 2009.29 The operation is now rapidly increasing in popularity in the United States and is currently the third most common procedure performed (after LRYGB and LAGB). Insurance reimbursement is still not uniform for all carriers, which has likely limited its rise in popularity to some extent. It is predicted that recognition and reimbursement by all major insurers in the near future will lead to its increasing performance over the next several years. Midterm data are available in a few instances, but most of the reported results thus far for the procedure are short-term results. As with all bariatric operations, long-term data will solidify or nullify this operation's position as a standard bariatric operation in the years ahead.

Patient selection for LSG, described previously, varies widely based on surgeon experience and alternative operations available to the patient by the treating surgeon, as well as patient preference and insurance reimbursement capacity.

Operative Procedure

The patient is positioned supine, with adequate support of the feet and legs such that reverse Trendelenburg's position is safely possible. The operating table must have capacity to safely maneuver with the largest of patients on it; hence hydraulic control is essential. The surgeon stands on the patient's right side, the assistant on the patient's left side, and the camera operator adjacent to the surgeon, on his or her right. A suggested port configuration is given in Fig. 27-9. Port placement can be varied, however. There are reports now in the literature of single-port performance of this procedure, with three instruments being placed through an enlarged single umbilical port.30

The LSG begins by dividing the blood supply along the greater curvature of the stomach, beginning at a point approximately 5 cm proximal to the pylorus on the greater curvature of the stomach. The gastroepiploic vessels are divided as they come off the greater curvature of the stomach, proceeding from distal to proximal along the greater curvature. An ultrasonic scalpel or other specialized energy device is used; electrocautery is insufficient to secure hemostasis of these vessels (Fig. 27-10). The division of vessels continues with the short gastric vessels, until the top of the greater curvature of the stomach is reached and a complete devascularization of the greater curvature above the distal antrum has been achieved.

A bougie, dilator, or comparable space-occupying device (some surgeons prefer a flexible endoscope) is positioned along the lesser curvature of the stomach. This bougie may be in the 32–40F range, based on surgeon preference and experience. The smaller the bougie, the better the postoperative weight loss, but also the greater the potential for a stricture of the gastric channel. A linear stapler is now used to begin dividing the stomach. In the antrum area, the height of the staples used should be longer than in the upper stomach. Division of the stomach is begun from the area where devascularization was initiated. The stomach is divided adjacent to the bougie or endoscope, leaving only a relatively narrow tube of lesser curvature stomach to serve as the passageway for ingested food (Fig. 27-11). Care should be taken not to divide the proximal fundus portion of the stomach too close to the gastroesophageal junction and the angle of His. Devascularization of this narrow segment of tissue may produce an ischemic leak postoperatively, which is a difficult and persistent problem to heal. Some surgeons feel that staple line reinforcement material offers the advantage of decreasing bleeding and leakage, while others feel this is not established and the material is too costly. In either case, it is appropriate to take particular care to avoid the complications of staple line bleeding or leakage after surgery during the stapled division of the stomach. Similarly, stenosis must be avoided as well. The completed division of the stomach and thus the operation is pictured in Fig. 27-12. The devascularized piece of greater curvature stomach is removed through the largest of the ports in a laparoscopic bag. Intraoperative performance of a leak test is not uniform but often done by surgeons at the completion of the operation.

Postoperative Care

LSG is normally performed as an inpatient procedure, though the stay in the hospital may be as short as 24 hours. Postoperative length of stay is often determined by the patient's medical comorbidities. Pain control is initially achieved with appropriate parenteral medications. Once oral medications are begun, usually within 24 hours, a liquid form of narcotic medication is preferred initially for pain control. Routine radiographic study on the first postoperative day is performed by many surgeons. Whether this becomes standard or optional is still controversial, as it is with LRYGB, which has a much longer history. Once the patient takes adequate liquid intake, has adequate pain control, and shows no signs of leakage, hemorrhage, or stenosis from the operation, they are discharged from the hospital.

Following LSG, patients must adhere to a liquid diet until they become accustomed to the restriction of the long and narrow gastric lumen. The length of such a liquid diet is variable, but usually in the 2- to 3-week range. Thereafter, initiation of soft followed by well-chewed solid food over the next few weeks ensues. Patients have a limited appetite due to the anatomic arrangement of the operation. This facilitates the diet and slow progression to solid food.

Follow-up for the first year should be frequently enough to detect problems of long-term stenosis, and occasional nutritional issues that may arise. Protein intake must be encouraged, and liquid protein supplements as well as dairy-related protein foods often serve as the initial largest component of protein intake. Later, once a larger volume of solid food is consumed, standard protein sources in the diet serve to meet protein needs. Vitamin B12 needs supplementation for most patients in the long term, and levels should be checked beginning a few months after surgery. Iron intake, due to low intake of iron-rich foods, may need supplementation. A multivitamin is a standard recommendation for daily intake by patients. LSG is still in its early phases of follow-up, and thus far no other major nutritional deficiencies have been identified after the procedure.

Outcomes

LSG has produced excellent weight loss as a primary bariatric operation. Reports in the literature show 1-year follow-up of 50–70% EWL.31,32 Resolution of comorbidities has also been reported, and is excellent and mirrors the percentage of excess weight lost. Patient satisfaction with the operation has been reported to be very high as well.

LSG is currently in its “honeymoon” period as a bariatric operation. The short-term results have been excellent, overall, in most reported series. However, the operation does not have a long enough track record to determine what will be the long-term problems seen with the operation. Particularly, the incidence of weight regain or recidivism is not yet reported after this procedure. An anatomically similar operation, the vertical banded gastroplasty, relied on a shorter lesser curvature tunnel of stomach with a band to constrict outflow.33 It too enjoyed immense initial popularity in the 1980s and, like LSG, was a technically easier operation to perform than a procedure like RYGB. During the1980s, the vertical banded gastroplasty was the most commonly performed bariatric operation in the United States. However, long-term follow-up showed a high incidence of patients changing their diet to accommodate for the restriction, and eating a high-calorie liquid diet. Weight regain slowly resulted, and after a decade one institution reported the number of patients with still successful weight loss after vertical banded gastroplasty was under 25%.34

Complications

Mortality for the operation has been under 1% in all major series and generally in the 0.2–0.3% range. Complications include bleeding, stenosis, and staple line leakage as problems arising soon after surgery, with overall short-term complication incidences reported in the range of 2–5%.29,35,36 Stenosis, food intolerance, and reflux are the most commonly cited problems after the immediate postoperative period.

Treatment of postoperative hemorrhage may be difficult endoscopically, due to the tightness of the lumen of the gastric tube. Operative treatment may be needed if initial conservative therapy with transfusions fails or hemodynamic instability occurs.

Leaks from the staple line are probably best treated initially with operative intervention to repair them. Depending on the circumstances, tissue quality, degree of peritonitis and soiling, a jejunal feeding tube may be appropriately placed for a safe site for enteral nutrition. Drainage of the repaired area is always indicated. Recurrent leakage may be amenable to endoscopic stent placement as treatment, depending on the lumen of the gastric pouch and the location of the leakage.

Stenosis of the gastric lumen after LSG is a difficult problem. Balloon endoscopic or fluoroscopic dilation is indicated as the initial treatment of choice if feasible. Few results have been published in the literature. The role of temporary endoscopic stents in treating this problem is also not established but potentially may prove effective.37

Longer-term problems of reflux and food intolerance have been reported, but the number of reports is still small and the relative frequency and severity of these problems have yet to be accurately determined.

Nutritional complications have not been reported with any significant frequency as of yet after LSG. Because most of the stomach is removed, it is anticipated that low vitamin B12 levels and potential megaloblastic anemia and vitamin B12–related neuropathy could result from LSG. Other nutritional deficiencies would, based on the anatomy of the operation, need to arise from inadequate intake of nutrients, because there is no malabsorption associated with LSG.

Laparoscopic Roux-En-Y Gastric Bypass

Laparoscopic Roux-en-Y gastric bypass is the most frequently performed bariatric operation in the United States. While there are many variations on the theme, there are certain defined characteristics and components of the operation that are common to all procedures bearing this name. The operative description below will favor our own approach at the University of Virginia, although the two coauthors of this chapter also have differences in their own techniques. Wherever possible, mention of common variations on each step of the operation will be included.

Patient Positioning and Preparation in the Operating Room

The operation is performed with the patient in the supine position, with the legs together. The surgeon gains little from operating between the patient's legs, and allowing the legs to be together and supported directly in line on the operating table decreases the potential for neural injuries to the legs if they were to be placed in any spread or supported position that would allow the surgeon to work between them. We also have found that the use of a footboard, large sponge cushion blocks to surround the feet, and taping those blocks securely to the operating table all allow the patient to be more easily placed into the reverse Trendelenburg position, which is advantageous for the gastric portion of the operation. The supine position also likely results in less venous thromboembolism (VTE) than if the legs are apart and supported.

The patient is positioned initially supine for the intestinal part of the operation but later in steep reverse Trendelenburg's position for the gastric portion. Both arms are normally out to the sides for vascular access, and supported appropriately. It is key to place padding under the axillary/upper arm areas of extremely obese individuals, as the body is so massive that the arms are not supported adequately in the supine position.

Preoperative antibiotics, of an appropriate dose, are indicated. Normally a first-generation cephalosporin will suffice to cover the proximal gut floras that are potential pathogens in this operation. The severity of wound infections has dramatically decreased with laparoscopic surgery as compared to open surgery in the past.

Prophylaxis against VTE is more controversial. Pulmonary embolism is one of the leading causes of death after bariatric surgery, and many surgeons use all measures feasible to decrease its incidence. This includes early ambulation after surgery. However, measures prior to surgery usually are at least foot or leg sequential compression devices or some form of heparin therapy. At our institution, after having had 1 year with a high incidence of VTE in the past, we use low-molecular-weight heparin subcutaneously, given just prior to the start of surgery, as well as sequential compression boots. Many papers have been written on this subject in the literature, without a clear consensus.38 However, it is now the practice of most bariatric surgeons to use both mechanical and chemical prophylaxis against VTE. Very high-risk patients, such as those with a hypercoagulable state or who have a history of VTE, may be candidates for use of a temporary prophylactic inferior vena cava (IVC) filter as the ultimate measure of protection against pulmonary embolism.

Skin preparation is with a standard chlorhexidine or iodine-based solution, with coverage of the entire abdominal wall up to a level 2 in above the xyphoid.

Operative Procedure

Pneumoperitoneum and Port Placement.

The pneumoperitoneum in the severely obese patient is best established, in our experience, using a Veress needle. A Hasson trocar has almost no place in bariatric surgery: it is difficult to make a deep incision to place it while maintaining tissue security around it to seal the pneumoperitoneum. Instead, using a standard tracheostomy hook to elevate the fascia produces an excellent countertraction of the abdominal wall that allows the Veress needle to penetrate the peritoneal cavity despite the thickness of the abdominal wall. The tracheostomy hook must continue to hold up the abdominal wall during the initial phase of insufflation, until there is an adequate pneumoperitoneum such that the tip of the Veress needle no longer touches any tissue or organs when this traction is released. Our preferred site of creation of the pneumoperitoneum is the left upper quadrant, near the costal margin, where the assistant will have his or her right hand 12-mm trocar port. This quadrant of the abdomen has, on average, the least amount of adhesions from previous surgery. If there has been surgery in this quadrant but not on the right, we will use a site in the right upper quadrant, where the surgeon's left-hand port is to be positioned. In super obese patients, a pneumoperitoneum pressure of 18 is often required for adequate visualization. Extremely thick abdominal walls may require extra long ports.

While there are a variety of port positions used for the performance of LRYGB, the configuration in Fig. 27-13 is our preference. The camera is placed through a port above the umbilicus, usually at the maximum of the “dome” created by the pneumoperitoneum. The ports are 12 mm for the surgeon's right and left hands and the assistant's right hand. These positions are chosen because of the ease and advantage of the angles created by these positions for firing the stapler during various steps of the operation. The assistant's left hand is a 5-mm trocar. At times, in extremely obese individuals, the addition of an extra port or two is needed. The intestinal portion of the operation may be difficult if the camera port is too close to the operative field. If this proves to be the case, we simply insert an additional port lower in the midline. The gastric stapling portion may also prove difficult, and the supraumbilical camera port may be very far from the operative field. In this situation, we do not hesitate to move the camera to the assistant's right-hand port, and add another 5-mm trocar higher up along the costal margin on the left for the assistant's right hand.

Liver retraction may be performed using one of several liver retraction devices. The T-Boone (Haynes and Boone, Dallas, TX) retractor, a simple tubular-shaped slightly curved piece of metal with a short cross bar, holds the left lobe of the liver out of the way adequately for most patients. Larger livers require the Nathanson retractor. The liver retractor is optimally placed in the epigastric region, high enough to be above the liver edge and hence hold the liver up when the retractor is inserted directly downward into the peritoneal cavity.

Intestinal Portion: Creating the Roux Limb.

This portion of the operation is begun by identifying the ligament of Treitz. This is only possible by having the omentum free from adhesions to other abdominal structures, at least for the midportion and left half of the omentum. Adhesions to the right lateral side wall can be usually ignored. Once the omentum is free, is it placed above the level of the transverse colon, and the transverse colon mesentery then grasped and elevated to help expose the ligament of Treitz. Once identified, the proximal jejunum is then divided between 30 and 50 cm distal to the ligament. The shorter the distance, the better iron and calcium absorption occur. However, longer distances are needed for larger patients where there is a greater distance to the proximal stomach and the Roux limb must therefore be able to reach more proximally. Generally, dividing the bowel at the 50 cm level provides a proximal end of the Roux limb that can reach the proximal stomach in even very large patients.

The jejunum is divided using one or two firings of the GIA-type stapler, using a white staple cartridge load. It is placed through the surgeon's right hand port, which gives a good angle for dividing the bowel (Fig. 27-14). The second staple load may often extend into the mesentery. This is helpful to begin increasing the mobility of the Roux limb. However, the division of the mesentery must continue in a direction directly downward on the mesentery, equally dividing the mesentery between the two segments of divided bowel. Straying to either side will cause ischemia to one of the bowel segments, requiring further resection back to viable tissue. The ultrasonic scalpel is used to continue the division of the mesentery below the staple load, carrying this division down to the root of the mesentery (Fig. 27-15). Care should be taken to apply slow application of energy in several adjacent points when dividing the major crossing vessels of the jejunal mesentery. Once the base of the leaflet of mesentery is reached, no further division should be attempted due to the risk of encountering major hemorrhage from vessels not easily controlled with laparoscopic energy devices.

The ends of the jejunum are assessed for viability. Any ischemic portion is resected back. Resected pieces of intestine are either removed directly now or placed in a specimen bag for later removal, depending on their size. If the end of the Roux limb is quite healthy, a 0.25-in Penrose drain is sewn to its end immediately (Fig. 27-16). If it needs resection, the drain is placed as soon as resection is performed. The end of the Roux limb should always be held by a grasper until the drain is attached. This is done to prevent any possibility of confusing the Roux limb with the biliopancreatic limb.

The proximal jejunum is now oriented adjacent to the ligament of Treitz, with its mesentery straight and pointed caudally. This leaves the stapled end of the biliopancreatic limb (the proximal jejunal segment) facing the camera and on the left of the operative field. The Roux limb is now measured for length. With experience this can be done by visual estimate, but in the beginning the surgeon may wish to use an instrument or some marker that helps estimate intestinal length. As the Roux limb is measured, it is pulled up and to the right on the screen, or to the patient's left upper quadrant. In this way the mesentery of the Roux limb has a continuous bend in a counter-clockwise direction as the limb is being measured. We generally make our Roux limb lengths approximately 100 cm for the patients with a BMI of 40–50, 125–130 cm for patients with a BMI of 50–55, and 150 cm for a BMI over 55. Once the appropriate length of Roux limb is measured, that point is sutured to the biliopancreatic limb with a single suture on the antimesenteric side of the Roux limb connecting to the antimesenteric surface of the biliopancreatic limb about 6 cm proximal to its end.

Enteroenterostomy.

The harmonic scalpel is now used to create enterotomies 1 cm distal to the suture holding the two segments of bowel together, on the antimesenteric sides of each segment of bowel. These enterotomies should be adjacent to each other. The white load 45-mm GIA stapler legs are now placed into each enterotomy from the surgeon's left-hand port, and the stapler is fully inserted into the bowel lumen, closed and fired (Fig. 27-17). A second white load is placed into the enterotomy site from the assistant's right hand, which usually is in a good position for the easy placement of the stapler in the opposite direction. The stapler is placed fully into the bowel lumen, which usually is just long enough to accommodate the upper leg in the short segment of the distal biliopancreatic limb. The stapler is closed and fired. We have found that this double-firing technique essentially eliminates the occasional issue of stenosis of this distal anastomosis, a complication that can prove fatal. The enterotomy is now sutured closed, beginning at the alimentary side of the opening and closing it upward toward the biliopancreatic limb (Fig. 27-18). Finally, the mesenteric defect of the enteroenterostomy is closed with a running permanent suture, beginning at the base of the mesenteric defect and completing the suture by sutures between the end of the biliopancreatic limb and the side of the Roux limb a few centimeters beyond the enteroenterostomy (Fig. 27-19).

Some surgeons choose to create the enteroenterostomy using a single-stapled technique. If this is done, it is highly advisable to suture the stapler defect, as trying to close it with a stapler may cause stenosis. The double staple technique is more amenable to stapling closed the stapler defect. Some surgeons also prefer to create this anastomosis with a hand-sutured technique.

Passing the Roux Limb.

Our preference is to perform a retrocolic retrogastric pathway for the Roux limb. The authors prefer this configuration because it allows the Roux limb to pass via the shortest distance to the proximal stomach, minimizing the risk of tension on the anastomosis. It also minimizes the risk for Petersen's hernia, based on the configuration of the Roux limb and biliopancreatic limb. Proponents for an antecolic Roux limb argue that this method is quicker and has little risk of anastomotic problems due to tension. They also argue there is a decreased incidence of internal hernias without a defect in the transverse colon mesentery.39

The biliopancreatic limb is now traced back to find the ligament of Treitz again. Then the transverse colon mesentery just to the left of the ligament of Treitz is held up and stretched so that it presents a solid flat surface. We find that the area just to the left and above the ligament of Treitz usually serves as a safe location to create an opening in the transverse colon mesentery with the ultrasonic scalpel. A longitudinal opening several centimeters in length is made, and the mesentery is then carefully divided with the ultrasonic scalpel. Usually larger vessels are not encountered, but caution must be taken in case of aberrant vascular distribution. Once an opening is made into the lesser sac, the stomach is usually readily visible and can be grasped and pulled up to the opening in the mesentery. Enlarging the mesenteric opening allows enough room to pass the Roux limb into the retrogastric space. This is facilitated by locating the Penrose drain and passing it into the space first, followed by the first 2–3 in of the Roux limb (Fig. 27-20). Care must be taken that the mesentery of the Roux limb is not twisted but instead is oriented straight up and down below the bowel as it is being passed through the transverse colon mesentery.

Difficulties in passing the Roux limb can be overcome with several tricks that have been learned over the years. The large and bulky mesentery presents more difficulty. In this situation the gastrocolic ligament often needs to be opened with the ultrasonic scalpel over a 4- to 6-in area. This allows a wide visualization of the superior surface of the transverse colon mesentery. The assistant then grasps the end of the Penrose drain, pushes it into the transverse colon mesentery through the standard incision in the inferior surface of the mesentery. The distention of the superior surface of the mesentery usually allows identification of the grasper and the Penrose drain just under the mesenteric surface. The surface can then be pierced with the grasper under direct vision, and the opening in the mesentery then also enlarged under direct vision to allow the Roux limb to pass through. A careful check of the mesentery is also needed in this situation as well.

Creating the Proximal Gastric Pouch.

The proximal gastric pouch is best constructed from the upper lesser curvature of the stomach, with only a minimal amount of fundus included. In order to better expose the stomach, we place the patient in relatively steep reverse Trendelenburg's position, after first placing the liver retractor (which is not needed until this point). The ultrasonic scalpel is then used to create an opening in the lesser curvature mesentery adjacent to the lesser curvature of the stomach. If the patient is larger, the pouch is best made a bit longer, starting within 1 or 2 cm above the incisura, to ensure that the Roux limb will reach the pouch. More often, the pouch is created starting several centimeters above the incisura. Once an opening is made, the blue load (or green load for thicker stomach) GIA stapler is fired directly across the stomach, creating a divided cut into the stomach from the lesser curvature side (Fig. 27-21). The surgeon must double-confirm with the anesthesiologist that any nasogastric tubes, temperature probes, or other tubes that could possibly be in the lumen of the stomach have been removed prior to firing the stapler.

We have found the next best step for guiding the size of the pouch is to have the anesthesiologist pass an Ewald tube, which is a gastric lavage tube used to evacuate the stomach of large clot or particles, measuring 32F in diameter. This tube is then positioned along the lesser curvature of the stomach, and it serves as a guide for creating the pouch size. The pouch is created just slightly larger than the size of the tube. The GIA stapler is then fired several times adjacent to the side of the tube until the top of the stomach is reached in the area of the angle of His (Fig. 27-22). An opening through the mesentery underneath the edge of the top of the fundus should be created to facilitate the final firing of the stapler and allow complete division of the stomach and thus complete separation of the proximal gastric pouch from the distal stomach.

Some surgeons prefer to use a flexible endoscope, esophageal dilator, or other space-occupying tube to help as a guide to create the proximal gastric pouch.

Creating the Gastrojejunostomy.

Once the gastric pouch is created, the Penrose drain is usually visible behind the lower stomach. If not, the gastrocolic ligament may need to be opened to locate it and facilitate passage of the Roux limb up to the proximal gastric pouch. Retrogastric adhesions are a problem that must be recognized if they prevent easy passage of the Roux limb. They can act as a barrier for passage of the Roux limb or, worse, act as a bowstring across the mesentery of the proximal Roux limb, rendering it ischemic. Any limitation to the easy passage of the Roux limb should be investigated by opening the gastrocolic ligament and assessing the posterior gastric surface.

Once the Penrose drain is located, it is used to gently pull the Roux limb up past the lower stomach. The proximal end of the Roux limb is placed adjacent to the lowermost end of the proximal gastric pouch, and the more distal Roux limb is gently teased up to approximate the entire length of the proximal gastric pouch. Once the two organs are thus aligned, a running suture is used to connect the side of the Roux limb to the staple line of the proximal gastric pouch. The suture is begun at the top end of the gastric staple line, and, when completed and tied, one end is left long for later use (Fig. 27-23). The Ewald tube is now used as a backstop to create a gastrotomy in the distal stomach pouch, about 1 cm from the end. An adjacent enterotomy is made in the Roux limb, using the ultrasonic scalpel for both. The anesthesiologist is asked to withdraw the Ewald tube 6–8 cm, so as to avoid stapling it. Visual confirmation of this is mandatory. The blue load of the GIA stapler is now inserted into the two lumens, one leg in each, and is best passed from the surgeon's left-hand port. This aligns it with the natural direction of the organs. Once the stapler is inserted to its full length, it is fired (Fig. 27-24). The enterotomy remaining is now closed using a running layer of absorbable suture, starting at the inferior apex of the opening near the knot of the running suture used to approximate the organs. This enterotomy closure is then reinforced with an outer layer of absorbable suture, after first having the anesthesiologist advance the Ewald tube so that it is just across the anastomosis, and thus stenting it open to prevent any stenosis by the second layer of closure. We then test the anastomosis for leaks by forcefully injecting methylene blue into the Ewald tube while holding pressure on the Roux limb just beyond the anastomosis. Distention without leakage must be accomplished.

Many surgeons use the circular stapler to create the gastrojejunostomy. The first author also used this technique for many years but has abandoned it in favor of the linear stapler. Using the linear stapler, without trying to limit anastomotic size, we have experienced that the incidence of anastomotic stenosis is under 2% (it was 10–14% with the circular stapler) and the amount of weight loss is comparable to that seen with the circular-stapled anastomosis.40 The linear stapler is also technically easier than the circular one for the surgeon to use if he or she does not have a highly skilled first assistant.

Closure of the Mesenteric Defects.

The retrocolic Roux limb must be secured to a reliable structure to keep it from telescoping up behind the stomach. Should it do so, the Roux limb will assume the shape of an accordion, and multiple partial obstructions or a single dominant point of obstruction will often then result. We prevent this by tacking the Roux limb to the adjacent proximal portion of the biliopancreatic limb, just distal to the ligament of Treitz, with several nonabsorbable sutures (Fig. 27-25). We also suture the Roux limb to the mesenteric defect both above and laterally to the patient's left side, to prevent herniation in either of these areas. Although our incidence of internal hernias is low with this technique, it is not zero, as sometimes sutures will fail or spaces develop between sutures that allow an internal herniation.

Completion of the Operation.

The 12-mm port sites are closed at the fascial/peritoneal level using a suture passer to pass absorbable 0 weight sutures. Then all ports are removed and the pneumoperitoneum decompressed. Skin closure is with subcuticular absorbable suture, and the skin sites are dressed with Dermabond (cyanoacrylate). Bupivacaine (0.25%) with epinephrine is infiltrated at all port sites, as it was when they were created.

Open Roux-En-Y Gastric Bypass

Operative Procedure

The open approach to performing Roux-en-Y gastric bypass (RYGB) is a time-tested bariatric operation that is the only procedure with considerable longevity. However, the benefits of doing the same operation laparoscopically cannot be denied, because there is a clear decrease in wound complications, incisional hernias, and also a more rapid return to normal function after surgery. The bariatric surgeon must, however, be able to perform an open operation if needed, as certain patients will require this approach. The most common reasons that a laparoscopic approach cannot be performed include excessive intra-abdominal scarring (almost always present at revision operations), massively thick abdominal wall precluding adequate mobility of the ports, a massive liver unable to be retracted using laparoscopic retractors, and intraoperative complications requiring conversion to an open incision for optimal treatment.

The open RYGB that we now perform is modeled on the laparoscopic operation, because we now have so much more experience with the latter approach and use it as our routine or default operation. There are a few differences, summarized in the following text.

Incision and Exposure.

The open incision must be made high enough to expose the diaphragm. Usually the skin incision must extend above the xyphoid by about 1–2 cm, and the division of the fascia and muscle should continue alongside the left side of the xyphoid process. This will allow the surgeon to look down and visualize the diaphragm. If the diaphragm is not visible, the exposure is not high enough to allow a safe procedure. Use of mechanical retractors is mandatory to help maintain adequate exposure. Human retractors are not adequate or appropriate. The length of the incision should be long enough to allow good exposure of the transverse colon and a few inches below it, where the enteroenterostomy is created.

Liver retraction using the open approach may be different than the laparoscopic one. Because there is no telescope to look up at the stomach from a lower plane, simple elevation of the left lobe of the liver with a retractor under it often does not move it adequately out of the field of vision of the operation. Instead, division of the triangular ligament of the left lobe of the liver and folding the liver inferiorly and medially exposes the gastroesophageal junction area much more adequately. This liver retraction should only be maintained during the gastric portion of the operation to prevent excessive compression of the liver over a longer time.

Enteroenterostomy.

We now begin the open operation with this part of the procedure and perform it in an identical manner as the laparoscopic approach. The Roux limb is marked with a suture instead of a Penrose drain. It is not passed until the gastric pouch is created, however.

Creation of the Gastric Pouch.

The proximal gastric pouch is created in a similar manner, except that the open approach allows the easy division of the gastrocolic ligament, and thus the placement of the surgeon's finger behind the stomach when creating the mesenteric defect along the lesser curvature. Palpation and upward pressure of the mesentery from behind the stomach facilitates creating the opening in the mesentery adjacent to the stomach. The pouch is then created in a similar manner with the stapler. The green load is more often used with the open approach, because the individuals having open surgery are often larger with thicker stomachs.

Passage of the Roux Limb.

The opening in the gastrocolic ligament allows the surgeon to palpate the superior surface of the transverse colon mesentery, exposing the inferior surface of it for division in an optimal place where palpation ensures a thin layer of tissue. The opening is usually made in about the same location as the laparoscopic procedure, but the manual palpation assists its creation in this setting. The Roux limb is now gently passed both retrocolic and retrogastric manually, positioning it adjacent to the proximal gastric pouch. Confirmation that the mesentery is not twisted is again mandatory.

The remainder of the operation is done essentially the same as the laparoscopic approach, with of course the exception of closing the incision. We favor the use of a looped no. 1 PDS-type suture to close fascia and peritoneum in a single layer. The subcutaneous tissues are copiously irrigated before the skin is stapled closed and dressed with sterile gauze.

Drains, Oversewing, and Gastrostomies.

In general, it is the authors' contention that a gastrointestinal anastomosis should be closed, not drained. However, if the intraoperative methylene blue test has shown a leak that needed repair, or if the quality of the tissue or the anastomosis is at all in doubt, or for those extremely technically challenging operations where visualization was just barely optimal, then in those situations the patient is treated in a manner to prophylactically anticipate a possible leak. If there is any concern about the stapling during the creation of the proximal gastric pouch, the staple line is oversewn with a running absorbable suture. If the gastrojejunostomy is of concern, a closed-suction drain is left adjacent to it, placed just inferior to it and coursing behind the spleen and out through the left flank. It is usually removed before discharge, or at latest on the first clinic visit. A distal gastrostomy is also placed during difficult open operations or converted operations where complications may have arisen. A standard Stamm-type gastrostomy with a 28- to 32-size tube is created to access and drain the distal stomach.

Postoperative Care

Postoperative care for LRYGB has now been formulated into a protocol at our institution that serves to generally avoid lapses in major postoperative desired treatments or orders, facilitates nursing care by following a routine, and promotes earlier discharge while maintaining attention to important postoperative needs of the patient. Major aspects of this protocol are as follows:

Intravenous isotonic fluids are given at a rate of 250 mL/h for the first 12–24 hours and adjusted as needed based on urine output. Some bariatric patients have been on diuretics for many years. In these individuals, care must be taken not to excessively bolus them postoperatively with multiple liters of fluid to treat oliguria, If no evidence of bleeding or other signs of fluid loss are present, we will give several liters of volume at most, then give a dose of intravenous diuretic, which usually produces appropriate urine output.

Early ambulation is a key component to the prevention of VTE. Patients are expected to ambulate within a few hours of their operation. Getting out of bed frequently is encouraged and stressed. Sequential compression foot devices are to be worn by the patient when in bed. Low-molecular-weight subcutaneous heparin is used for VTE prophylaxis until discharge in average risk patients. In high-risk patients, it is continued at home for 3 more weeks on a twice daily subcutaneous injection dosing.

Oral intake is limited to ice chips the night after surgery. An upper GI study with water-soluble contrast is done the following day. If there are no problems on the study, a clear liquid diet is begun. This is advanced to a blenderized diet the next day. While many authorities have written that such a postoperative study is inaccurate and cost-ineffective,41 we still use it to detect any potential problems of obstruction distal to the anastomosis and to document gastric pouch size.

Pain control is achieved through a combination of intravenous medications graduated to oral medications by the first postoperative day. Patient-controlled analgesia (PCA) pumps are often used in the first 24 hours and then stopped as oral medications are introduced.

Intravenous antibiotics are stopped after a postoperative dose in addition to the preoperative dose.

Wound care is simplified by the Dermabond, which allows wounds to be exposed to water if needed. No special care is required.

Oxygen is supplied the first 24 hours and then removed as appropriate based on oxygen saturation levels. As per our policy, patients with obstructive sleep apnea are required to bring their continuous positive airway pressure (CPAP) masks and use them while in the hospital. High-risk pulmonary patients all have a mandatory arterial blood gas done preoperatively to determine their “baseline” status. This is important in case postoperative ventilatory support is needed, which is rare. If it is, the pre-op ABG serves as a guideline for extubation. Otherwise, the surgeon is often confronted with an intensive care unit (ICU) team who wants to see “normal” blood gases prior to any extubation attempt.

Our normal protocol for patients undergoing LRYGB is for discharge at noon on the second postoperative day. Most patients achieve this timeframe for discharge, while occasional patients are detained an extra day for issues, including hypoxia, urinary retention, pain control, other medical problems, or social issues. Patients undergoing open RYGB undergo the same postoperative protocol, except they often are not able to wean off intravenous narcotic pain medications or are not adequately ambulatory until the third postoperative day.

Follow-Up

Patients undergoing LRYGB are seen back for follow-up clinic visits at approximately 3 weeks, 3 months, 6 months, and 1 year after surgery, then annually thereafter. More frequent visits are scheduled as needed. The 3-week visit is focused on adjustment issues to new eating habits, recovery from postoperative problems, advancement of diet, initiation of an exercise program, and adjustment of any medications (always done in coordination with the patient's primary care physician). The 3-month visit confirms the effectiveness of the diet and exercise plans, as well as improvements in comorbid medical problems and further medication adjustments as indicated. The 6-month visit assesses for any potential diet or nutrient deficiencies and reemphasizes the need for a consistent exercise program. Medications are again reviewed. Medication that is being taken to prevent gallstone formation with rapid weight loss (ursodiol 400 mg twice daily) is discontinued. The 1-year visit assesses changes in comorbid medical conditions, reviews the improvements that have resulted from the operation, and emphasizes the need to continue the diet and exercise changes that the patient has established during the year. The patients are cautioned that weight regain is a major issue if such diet and exercise habits are abandoned, because adaptation to the operation will allow greater oral intake than before (though still limited) and often is accompanied by a return of appetite as well.

It cannot be emphasized enough that any and all bariatric operations will not produce the durable weight loss sought and the long-term improvements in health and comorbid medical problems desired unless the adjustments to eating, exercise habits, and lifestyle produced by the operation are maintained long term. Regaining weight, or recidivism, is the single greatest long-term problem facing the patient who undergoes bariatric surgery.

Outcomes

Laparoscopic RYGB produces excellent weight loss, with 1 year percentage of excess body weight loss (%EBWL) being reported as between 60 and 75% in most series.42–44 The laparoscopic approach has only been commonly used for a decade, and it did not achieve a position as being more frequently done than the open RYGB until probably 2003. Therefore, most of the long-term outcomes of RYGB have been published about patients who underwent open RYGB. For this approach, the weight loss was very comparable to that of LRYGB, although there were few reports of %EBWL of over 70%. Long-term maintenance of weight loss has been reported by Pories et al45 and the Swedish Obesity Study46 that included RYGB as well as gastroplasty for its operative procedures. The laparoscopic approach has probably been associated with slightly higher weight loss in many series because of two factors. The first is the average size of patients undergoing RYGB since the laparoscopic era has decreased. Prior to having a laparoscopic option for the operation, not nearly as many patients with BMI in the 40–50 range wished to undergo surgery, especially when it entailed an open incision. The popularity of RYGB was immeasurably enhanced by having the laparoscopic option, as evidenced by the rapid rise in popularity of RYGB being performed in the United States. between 1999 and 2003, during which time the number of procedures went from approximately 25,000 to 120,000 per year (Fig. 27-26). While other factors, such as Internet communication and publicity in national media probably also contributed to the rise in popularity of the procedure, they were most likely a secondary by-product of the sudden increase in demand and popularity of the operation. The costs of performing RYGB have been shown to be less, over a short follow-up period of 5 years, than the costs of medical care for patients who qualified for but do not undergo surgery.47

RYGB is known to have had many beneficial effects for patients who have undergone the procedure. It has been shown to improve longevity.48–50 It has been shown to reduce medical comorbidities, the results of which are summarized in Table 27-4.51–53 Certain specific medical problems merit more detail in terms of the effect of RYGB on them.

Diabetes

RYGB has been shown to be a highly effective treatment for type 2 adult-onset diabetes. Patients suffering from this disease will often experience remission of symptoms of the disease after only a few weeks of time have passed since surgery. The amount of weight lost during such a period of time does not alone explain this rapid amelioration of the disease. This clinical observation had been made by many surgeons performing RYGB and was best summarized in a study by Pories et al.54 Over the past decade, a significant amount of research work on the mechanism of how RYGB improves diabetes has been performed. Studies in obese mice done by Rubino et al55 demonstrated that the diversion of the food stream from the proximal intestinal tract resulted in remission of the animal's obesity and diabetes, and reversal of the operation resulted in reappearance of both conditions. The metabolic aspects of how RYGB may change the enteroinsular axis are still being debated.56 It appears that glucagon-like peptide-1 (GLP-1) is important in the process, but other gut hormones have also been implicated as potentially having a role in the metabolic changes seen after RYGB. Entire conferences and symposia have been held since 2007 on this topic, and the subject continues to be one of intense interest among scientists studying the metabolic effect of RYGB on diabetes, the metabolic syndrome, hyperlipidemia, and other metabolic conditions.

Cholelithiasis

Gallstones may form with rapid weight loss for any reason, be it surgical or diet-induced. This was first observed after subjects following diet programs in the 1980s were noted to have high incidences of gallstones. Subsequent studies have shown that the incidence of sludge or stone formation after rapid weight loss to be in the 30% range.57 This condition must be therefore considered in the planning of LRYGB, LSG, and DS for patients. Fortunately, it has also been shown that the consumption of prophylactic bile acids, specifically ursodiol at a dose of 300 mg twice daily for 6 months following RYGB surgery, can decrease the incidence of gallstone formation to 4%.58

Performance of synchronous therapeutic cholecystectomy is generally recognized as appropriate for patients with symptomatic cholelithiasis. There are some surgeons, however, that do not direct preoperative attention to the biliary system or feel such attention is unwarranted. Historical data do not support such a position, but some recent short-term follow-up studies have suggested a low incidence of biliary complications within the first year or two after surgery using such an approach.59,60 However, studies with longer follow-up are more likely to document the not insignificant incidence of biliary complications that result from untreated cholelithiasis long term.61–63

More controversial is the question of whether to treat asymptomatic cholelithiasis if discovered prior to elective bariatric surgery. While some studies have concluded that the length of hospital stay or the complications associated with performing simultaneous cholecystectomy do not warrant its performance,64 our experience has shown that synchronous prophylactic cholecystectomy, when performed with RYGB, either open or laparoscopic, has little effect on outcomes other than to lengthen operating time slightly.65

During the era of open bariatric surgery, performing a synchronous cholecystectomy was commonplace among many practices, including our own. The rationale was that the incidence of postoperative development of cholelithiasis and the need for subsequent operation outweighed the risk of performing a synchronous prophylactic procedure, provided the latter was done under good operative conditions following the successful completion of the bariatric operation. Now with most bariatric operations being done laparoscopically, the ability to perform a laparoscopic cholecystectomy as a second procedure is very high, and hence the need for the prophylactic cholecystectomy for the patient with a normal gallbladder has greatly diminished.

Our current recommendation for this issue is synchronous cholecystectomy for any patient with biliary pathology undergoing RYGB, strong consideration for synchronous cholecystectomy for all patients undergoing malabsorptive operations where bile salt pool will be depleted, and use of oral chemoprophylaxis in the form of ursodiol 400 mg BID for 6 months following performance of an operation that will produce rapid (>50 lb/3 mo) weight loss. Because most patients who undergo LAGB will not lose weight as rapidly as with malabsorptive or RYGB operations, the need for prophylactic cholecystectomy is minimal and not indicated. Nor is the routine use of ursodiol. However, for patients with cholelithiasis, it should be the judgment by the surgeon as to whether synchronous cholecystectomy at the time of LAGB is indicated, based on comfort of not spilling bile during the procedure as well as the patient's preoperative symptoms.

Gastroesophageal Reflux Disease

This comorbid medical problem also warrants some special consideration in the recommendation of operative bariatric choices for patients. Patients with gastroesophageal reflux disease (GERD) will have significantly greater improvement in symptoms after RYGB66 than after LAGB67 or LSG.32 Patients who undergo RYGB have as higher a symptomatic relief from preoperative GERD than those who undergo specific surgical fundoplication for GERD.66,68 In fact, patients who are referred for surgical fundoplication with a BMI of greater than 35 should be offered the option of having a LRGYB as an alternative to a laparoscopic fundoplication. The former will treat the patient's reflux symptoms as well as improve their physical and medical issues related to severe obesity.69

Complications

Mortality from LRYGB has now decreased considerably from the figure of approximately 1% reported in the literature a decade ago. Recent reports from very large databases have given the incidence as 0.13–0.18%.70,71 Increased experience of surgeons, an overall less ill and less large patient population undergoing surgery, improvements in care through centers of excellence programs, and the decreased burden of an open incision have all likely contributed to this decrease in mortality. LRYGB is now safer than almost all intra-abdominal operations with perhaps the exception of cholecystectomy.

Morbidity from LRYGB has also decreased as well with improving experience over the past decade. Major operative morbidity is now given as under 2% in most series and total 30-day comprehensive morbidity as reported by our NSQIP (National Surgical Quality Improvement Program) database and for the University Health System Consortium (UHC) has been 15 and 14% respectively.72 NSQIP outcomes data for our institution and the UHC consortium are summarized in Table 27-5.

Specific Complications and Their Treatment

Complications after LRYGB and open RYGB are comparable except the latter has a higher incidence of wound complications and incisional hernias. Intraoperative complications are generally low in incidence, in the 2% range. These include hemorrhage, organ injury, twisting the Roux limb during performance of anastomosis, and anastomotic leakage as documented by intraoperative testing. Postoperative complications that are commonly described in the early postoperative period include intra-abdominal and anastomotic hemorrhage73; mechanical bowel obstruction due to technical error, severe edema, or intraluminal hematoma74; and anastomotic leakage.75,76 Other nonbariatric specific complications also certainly occur, including VTE, cardiac arrhythmias, pulmonary atelectasis, and pneumonia.

Of these immediate postoperative complications, anastomotic leak is the one that is most feared by bariatric surgeons, because of its potentially fatal consequences. The condition may be difficult to diagnose clinically, with many documented cases of patients having isolated tachycardia as the only presenting symptom. Other common symptoms and signs suggesting a leak include fever, abdominal pain, tachypnea, a sense of impending doom, oliguria, and hypotension. In general, after RYGB, a patient who becomes ill in the first week after surgery has a leak until proven otherwise. Radiographic testing may not always diagnose the problem, because of the combination of lack of 100% sensitivity as well as the fact that leaks may occur in the distal gastric staple line or at the enteroenterostomy, where little or no contrast may be present on a fluoroscopic study.77 If clinical symptoms are concerning for a leak, lack of radiographic confirmation should not keep the surgeon from reexploring the patient for the problem. While very small leaks that already are drained may be considered for conservative treatment,78 in general the appropriate treatment of a leak is operative.79 Treatment includes primary repair, placement of adequate drains in case of releakage, provision of an enteral feeding route (usually a gastrostomy in the lower defunctionalized stomach), and, for persistent or severe cases, consideration of use of an endoscopic stent.80

Vomiting within the first week after LRYGB or RYGB should be considered a very worrisome sign of potential early bowel obstruction. Stenosis, mechanical or technical issues, edema, and hematoma at the enteroenterostomy can all produce a relative or near-complete obstruction of the alimentary tract in this location. While the patient may vomit to clear the Roux limb, the biliopancreatic limb and distal stomach cannot be decompressed and are subject to massive distention and staple line rupture, with potentially fatal consequence (Fig. 27-27). Thus vomiting must be considered a sign of obstruction requiring vigorous investigation and reoperation if at all in doubt about the patient's condition.81

Long-term complications of RYGB are less numerous in terms of type, but several conditions are common and must be recognized to prevent life-threatening situations. By far the most dangerous and life-threatening is that of small bowel obstruction from an internal hernia.82,83 Loss of a significant portion of the small bowel from ischemic necrosis may result if this condition is not promptly appreciated, diagnosed, and treated when it occurs. Any patient who has had a LRYGB and presents with abdominal pain and signs of a bowel obstruction should be considered to have a bowel obstruction with internal herniation until proven otherwise. The surgeon should NEVER treat this condition with a nasogastric tube and observation if there is any doubt this condition may exist. Operative treatment is the only appropriate approach. This diagnosis requires a high index of suspicion and can be compared to the diagnosis of ischemic bowel in the nonbariatric patient population. Symptoms usually are present in greater severity than physical, radiographic, or laboratory findings. Sometimes there may be a “swirling” pattern on the CT scan (Fig. 27-28). The authors have had several patients who presented with intermittent pain after eating but without confirmatory laboratory or radiographic signs. In these cases, much of the small bowel had herniated through an internal hernia and fortunately had not yet developed strangulation. Reduction of the hernia is facilitated by beginning at the terminal ileum in these cases, as this will reliably identify the more distal bowel and allow reduction in the correct direction. Many of these cases may be treated laparoscopically, but distention preventing visualization and edema to the bowel precluding safe laparoscopic manipulation are the limiting factors.

Stenosis of the gastrojejunostomy following LRYGB presents as progressive food intolerance in the 4–12 week period after surgery. The symptoms should prompt the performance of a flexible upper endoscopy, which may then be both diagnostic for the condition as well as therapeutic. The opening of the anastomosis may be very stenotic. Fortunately, however, the use of an endoscopic balloon can usually produce significant, if not complete, remission of the obstruction. If one endoscopic balloon dilation does not produce relief of symptoms, either another endoscopic or a subsequent fluoroscopic dilation with a larger-size balloon usually treats this condition quite adequately.84,85

Reoperation for stenosis is usually not needed unless the condition has been allowed to persist or is secondary to the presence of a marginal ulcer, which then causes further scarring and stenosis.86 We have found by experience that the use of a linear stapler dramatically reduces the incidence of stenosis of the gastrojejunostomy after LRYGB when compared to the use of a circular stapler for this anastomosis.40

Marginal ulcer is the other major long-term problem after LRYGB or RYGB that deserves special discussion as well. The etiology of marginal ulcers is still controversial and felt to likely be multifactorial.87 The incidence is increased in the setting of mechanical stapling or permanent suture to create the anastomosis.88 The presence of Helicobacter pylori colonization leads to a higher incidence of marginal ulcer postoperatively than if the condition is treated preoperatively.89 Ischemia is also felt to play a role in the formation of some marginal ulcers. Cigarette smoking is a risk factor for persistence, recurrence, and nonhealing of a marginal ulcer.90 Many bariatric surgeons will not offer an LRYGB to a patient who is a smoker for this reason.

Persistent burning epigastric pain, relatively unchanged by eating, though possibly slightly relieved in some cases, is the hallmark for marginal ulcer. A high index of suspicion and performance of endoscopy will yield the diagnosis. The incidence of the problem is between 2 and 14%.91,92 This variability is likely based on aggressiveness of diagnosis as well as risk factors. Treatment of the problem is medical, which is effective in the vast majority of cases. Nonhealing of a marginal ulcer should precipitate performance of a Gastrografin (diatrizoate meglumine) contrast study to rule out the possibility of the ulcer having penetrated into the lower stomach. If that has happened, surgical intervention is needed as the gastrogastric fistula tract is unlikely to close spontaneously.93 Similarly, if the patient had an RYGB without a divided stomach and the gastric staple line has broken down to allow communication from lower to upper stomach, surgery is needed to separate the lower stomach from the proximal gastric pouch and thereby eliminate the backwash of acid onto the anastomosis and ulcer.

Nutritional Issues and Complications

Because LRYGB diverts food only from the majority of the stomach and the duodenum and up to 50 cm of proximal jejunum, the incidence of major malnutrition issues or nutrient deficiencies is limited to those nutrients specifically absorbed in those areas or those nutrients taken at a significantly lower volume after surgery. Protein malnutrition is rare without the presence of depression or other illness essentially minimizing intake. Liver disease can contribute to hypoalbuminemia. Much more common nutritional problems arise from the lack of ability to absorb iron and calcium due to the diversion of the alimentary stream from the proximal small bowel. Similarly, lack of gastric nutrient presence decreases the secretion of vitamin B12 because of lack of gastric production of intrinsic factor required to absorb the nutrient. These three elements should be measured for all patients undergoing RYGB. Folate can, on occasion, also be deficient in these patients. Another uncommon but problematic vitamin deficiency is thiamine deficiency. This may arise in the setting of progressive and significant vomiting and can, if untreated, result in permanent neurologic sequelae such as Korsakoff's syndrome.94 Vitamin D has been shown to be deficient in a high percentage of the obese population and therefore is, as expected, also deficient in a high percentage of postoperative patients.95 While the incidence of significant osteoporosis is not known after RYGB, the prudent approach to vitamin D deficiency is to treat it with supplementation.96

Summarizing the vitamin and nutrient needs of patients undergoing LRYGB, we follow what is likely a common set of recommendations for postoperative vitamin supplementation. This includes the following:

1. A daily multivitamin with iron (ferrous gluconate or fumarate)

2. Ferrous gluconate 300 mg BID for women of reproductive age or with documented iron deficiency

3. Calcium citrate 400–600 mg with vitamin D 300–400 IU BID for most individuals who have had RYGB and especially for those who have documented low levels of vitamin D3

4. Vitamin B12 intramuscular injection (usually 1000 μg) or sublingual tablets (1000 μg combined with folate, vitamin B6, and biotin, one or two daily) as needed for biochemically low levels of vitamin B12 on testing

Probably the most common clinically significant vitamin deficiency is iron deficiency anemia. This is usually adequately treated with the appropriate oral supplement. Ferrous sulfate requires an acid medium for optimal breakdown and absorption and should not be prescribed in this patient population. Ferrous gluconate is appropriate instead. Low biochemical levels of vitamin D3 are common, and, if significant, we treat those with additional vitamin D beyond the amount in the usual vitamin plus calcium supplements. Clinically significant vitamin B12 deficiency after RYGB is rare, but despite this, patients with biochemically low levels do receive parenteral or sublingual supplementation until normal levels are achieved.

Biliopancreatic Diversion and Duodenal Switch

Biliopancreatic diversion (BPD) was first described by Scopinaro et al in 1976.97 The modification of the BPD to a duodenal switch (DS) was performed by Hess and Hess in 1988.98 These are some of the most difficult bariatric operations to perform and have the highest postoperative complications. They also produce the most durable weight loss, especially in superobese patients. BPD and DS procedures together used to represent less than 5% of all bariatric cases performed in the United States99 and now represent perhaps fewer than 1% of cases.70 These operations are best suited to the superobese, for those patients who have failed a restrictive operation, or for those patients who value the ability to continue to eat larger portions of food as a most important aspect of life after bariatric surgery. Patients must be prepared to take a significant number of vitamin supplements after these operations.

Operative Procedure

The BPD and DS have been performed using both laparoscopic and open approaches. The technical details are similar other than the access to the abdomen, which is similar to all other laparoscopic bariatric operations. Port positions for doing the operation may vary among surgeons, but one example of port placement for laparoscopic DS (LDS) is given in Fig. 27-29.

In performing the BPD, a subtotal horizontal distal gastrectomy is performed, leaving a 200- to 400-mL gastric remnant. The ileocecal valve is identified and the bowel is measured back 250 cm and divided. The distal end of the divided bowel is anastomosed to the proximal gastric pouch. Either a linear or circular stapler may be used, with the linear the most common choice. The proximal end of the divided ileum is then anastomosed side to side using a linear stapler to the terminal ileum at a point 100 cm from the ileocecal valve. The completed operation is shown in Fig. 27-30. Previously surgeons performed this anastomosis 50 cm proximal to the valve, but this produced higher complications with protein malabsorption. Even Scopinaro and associates described the need to make this “common channel” longer in patients from southern Italy who ate a less protein-rich diet than those from northern Italy.100 A prophylactic cholecystectomy should be performed due to the high risk of postoperative cholelithiasis.

The DS differs from BPD in the type of gastrectomy. It was developed to reduce the incidence of dumping syndrome and marginal ulceration. In a DS a vertical or sleeve gastrectomy based on the lesser curve of the stomach is performed (Fig. 27-31). The sleeve is calibrated over a bougie typically 32–40F leaving a gastric volume of approximately 100–200 mL. The first portion of the duodenum is then divided approximately 2 cm from the pylorus. The end of the distal ileum and the duodenal cuff are then anastomosed. Usually this anastomosis is best done using a hand suturing technique, whether open or laparoscopic. The distal small bowel anastomosis proceeds as previously mentioned, and this is most often performed with a standard linear stapling device. The completed DS is shown in Fig. 27-32. A cholecystectomy should also be performed.

Postoperative Care

Both BPD and DS operations are comparable in terms of most of the issues that must be addressed both in postoperative care and follow-up. The DS is now an accepted procedure in the United States and reimbursed by most third-party payers. The BPD is rarely done in the United States. Thus the remaining discussion pertains to the DS rather than the BPD, unless specified. It should be reemphasized that the BPD poses virtually the same risk and complication spectrum.

Following DS, patients are at risk for anastomotic leak, gastrointestinal bleeding, pulmonary embolism, bowel obstruction, and stenosis. The complications present with a similar clinical picture to their counterparts after LRYGB. Treatment is also usually similar and outlined in the section under complications later.

The BPD and DS procedures are primarily malabsorptive. Patients must be educated as to the consequences of the operation. They should be instructed to expect diarrhea after any large meal. This diarrhea serves as a powerful behavioral modification to eating. DS patients eat, on average, more calories per day than LRYGB patients but surprisingly only a small amount more.101 The prolific diarrhea that follows overeating precludes significant excessive food intake. Patients also will learn which foods, primarily those high in fat content, produce the worst diarrhea. Patients should be cautioned about the potential perianal problems that can follow prolific diarrhea, as well as eating patterns that will improve the situation.

Nutritional deficiencies are inherently likely after malabsorptive operations. Education as to proper intake of nutrients and supplements is essential. These patients are at risk for protein-calorie malnutrition, fat-soluble vitamin,, and micronutrient deficiency. Because the duodenum and proximal jejunum are diverted from the food stream, they are as at risk for iron and calcium deficiency as patients undergoing RYGB. Vitamin B12 must also be supplemented because of the lack of much of the stomach. Thus one of the main issues regarding postoperative care in DS patients is to ensure that they understand the need for, and have obtained and are taking the proper vitamin and mineral supplements. These may be costly, and that cost should have been part of the preoperative discussion with the patient prior to undertaking the procedure. Magnesium deficiency may also result from short gut syndrome and malabsorption, which was more commonly seen after intestinal bypass operations. It is less common but still possible after DS.

1. A typical recommended supplementation scheme for patients undergoing DS is as follows:

2. Parenteral fat-soluble vitamin injections to include vitamins A, D, and K

3. Oral ferrous gluconate supplementation and oral vitamin D and calcium supplementation

4. Parenteral or sublingual vitamin B12 supplements monthly or weekly, respectively

5. Standard multivitamin supplementation orally daily

Follow-Up

Following DS, patients must be closely followed by the surgical team, as other care providers may not fully appreciate the problems related to malabsorption. This is particularly true of longer-term protein malnutrition. The primary care physician may mistake the protein-calorie malnutrition seen after DS with manifestations of congestive heart failure or liver disease, which can produce peripheral edema and hypoalbuminemia. If an exacerbation of protein-calorie malnutrition occurs, hospitalization and parenteral nutrition may be needed. Multiple such admissions are an indication to revise the operation and make the common channel longer. The potential for vitamin or mineral deficiency or protein malnutrition never is eliminated for patients after DS; it is a lifetime condition. Thus semiannual blood testing is recommended for such patients even years after surgery.

Outcomes

DS produces the best weight loss and the most durable weight loss of any of the commonly recognized and reimbursed bariatric operations.40 Weight loss in the 75% of EWL is expected after this operation. Furthermore, the weight loss after DS is usually preserved, and recidivism is lower than after either LAGB or LRYGB.102 Patient satisfaction is generally given as high after the operation.

Resolution of comorbidities after DS is also remarkably good. Conditions that are especially well treated after this procedure include hyperlipidemia, diabetes, and metabolic syndrome.40,103

Complications

Immediate postoperative complications that are seen after LRYGB are also seen after DS. These complications present in a similar manner to those described previously after LRYGB, with the exception of stenosis. This may occur at the proximal anastomosis but may also occur along the length of the sleeve gastrectomy. Such a stenosis may be amenable to balloon dilation, but success is not as likely as with the stenosis of the gastrojejunostomy after LRYGB. Options to treat the ischemic stricture or a severely stenotic stricture of the gastric sleeve are limited to local revision versus a gastrojejunostomy to a Roux limb brought above the stenosis, effectively converting the operation to a RYGB. Bowel obstruction from an internal hernia is also possible after DS, although just at the enteroenterostomy from an internal hernia. Leak at the duodenoileostomy is the most common site of leakage after DS and occurs in the 2–4% range. Treatment of this problem may require a period of parenteral nutritional support before eating is possible.

Fat-soluble vitamin deficiencies may result as a consequence of DS. Vitamin A deficiency, which clinically presents as night blindness, was present in over 70% of patients in one series.104 In that same series, vitamin D was found deficient in 63% of patients after 4 years. Vitamin K deficiency, manifested as coagulopathy, is more uncommon. Vitamin D deficiency is very prevalent in the population as a whole, and so supplementation after DS is always indicated and low levels are not uncommon. Osteoporosis may result from chronic vitamin D deficiency. The condition is further exacerbated by the fact that calcium is poorly absorbed after DS as well. Iron deficiency anemia will invariably also arise if no supplementation occurs. Although there is no consensus, generally it is recommended to supplement all the above elements as needed, with careful monitoring by serum blood tests serving as the ultimate guide for each patient.105

The most feared nutritional complication after DS is protein-calorie malnutrition. Clinical manifestations of this problem include edema, weight loss, skin and nail problems, hair loss, and general malaise. Laboratory tests reveal low albumin and serum protein levels. Increasing oral intake of high-quality proteins may help, but, if the condition is more advanced, parenteral nutrition is often needed. When parenteral nutrition is consistently required, reoperation to lengthen the distance of the “common channel” or intestine below the level of the enteroenterostomy is indicated. Most surgeons make that initial channel about 100 cm. If revision is required, there is no exact formula as to the appropriate length of the revised common channel, but most surgeons would consider adding at least 50 cm if not more to the length to prevent recurrence of protein-calorie malnutrition. Long-term follow-up studies of DS show the incidence of reoperation to be in the 3–5% range.106

The BPD has a higher rate of marginal ulcers than DS, with the incidence being over 12% in the original report by Scopinaro et al.100

Revisional Surgery

Revisional surgery is a highly controversial area of bariatric surgery. Wide variability exists in the philosophies of bariatric surgeons as to the appropriateness as well as their enthusiasm for performing revisional surgery. A few generalized statements regarding revisional surgery are as follows:

1. The revisional operation will usually produce less of a major change in the patient's weight or reduction in comorbid medical problems than did the index operation.

2. The complications of the revisional operations are higher than the index operation.

3. There should be even greater scrutiny and assessment of patients undergoing revisional surgery than initial operations.

4. Care should be taken to avoid severe nutritional complications that may accompany the combination of too much restriction and malabsorption with revisional surgery.

In general, revisional operations fall under two broad categories:

1. Those done to correct technical shortcomings of the index operation, or complications developing as a result of it

2. Those done as a consequence of poor weight loss from the index operation

Because only RYGB has a track record of durable success of more than two decades among all the restrictive operations, the revision of the others has been a frequent issue. Malabsorptive operations have also enjoyed two decades of success, though their popularity has been significantly lower than restrictive operations. Few patients who have had either a BPD or DS have had the operation reversed, or at least there are few publications reporting this process. The jejunoileal bypass, commonly performed 35 years ago, resulted in an extremely high incidence of revisional surgery for consequences of its malabsorption.107 Most of these revisional operations were performed years ago. Reversal of the operation was done, and those patients who did not have an associated new bariatric operation performed often regained weight and suffered from recurrent severe obesity.108

The vertical banded gastroplasty (VBG) was performed with high frequency in the 1980s but was found to have poor long-term weight loss preservation. As a result, many of the patients who underwent VBG have subsequently undergone revisional surgery. There are numerous accounts in the literature of conversion of VBG to RYGB.109–111 Most of these accounts include a mixture of patients who had poor weight loss as well as other complications or shortcomings, such as the reported 17% incidence of stenosis after VBG,112 as the reason for reoperation. Thus the indications for reoperation for VBG have been both as a consequence of both poor weight loss and technical complications. Most of the series of VBG revised to RYGB have had acceptable results in terms of weight loss and resolution of comorbidities. However, these revisional surgeries have also been accomplished with a higher incidence of morbidity than index operations of RYGB.113

During the 1970s and 1980s there were a number of gastric stapling operations performed to provide weight loss for patients.114 Unfortunately, these had uniformly poor long-term outcomes in terms of durable weight loss. Many patients undergoing those procedures were then converted to operations with a better results record, including VBG and RYGB. This text does not belabor the numerous accounts of these revisions, all of which proved no more effective than an index operation of either RYGB or VGB, and often less effective as a revisional procedure.

Current candidates for revisional surgery include patients who have failed any of the current commonly performed operations. There are increasing reports of poor long-term weight loss outcomes of LAGB from some centers in Europe.115 Increasing percentages of patients who have requested band removal for poor results are seen in more recent reports regarding LAGB. For patients wishing revision after LAGB, the options would include conversion to LRYGB or to LSG or to DS. Reports of successful conversion to both exist in the literature, but the complication rate is still higher than for the index operation.116,117

Revision of RYGB has focused on several different types of problems with the index operation and thus several alternative treatments. Prior to the laparoscopic era, many surgeons performing open RYGB did not actually divide the stomach, but stapled across it with multiple staple rows to create the proximal gastric pouch. Long-term follow-up studies have shown that the risk of dividing the stomach at the index operation proved to be less than the risk of that staple line subsequently breaking down and allowing loss of restriction of the gastric pouch with weight regain, marginal ulcer, or both resulting in the need for reoperation.118

Another theory of failure of RYGB is that if the anastomotic opening of the RYGB is too large, it has resulted in the weight regain. From that theory follows the recommendation that an operation or endoscopic procedure to narrow the anastomosis in patients who have regained weight after LRYGB or RYGB will be effective in reversing the weight gain and produce new weight loss. As a result, over the past several years, there have been several reports in the literature about endoscopic119,120 or other operative procedures to restrict the anastomosis of the RYGB.121 Unfortunately, the long-term follow-up of those procedures to produce durable weight loss has shown that they are likely to accomplish limited short-term weight loss but not significant durable long-term weight loss.122,123 The fact that increasing the anastomotic size after LRYGB was associated with a decrease in stomal stenosis after surgery, while still producing comparable weight loss, should have been ample evidence to demonstrate that anastomotic size has only minimal effect on the overall weight loss for patients after LRYGB. This observation is likely true for other operations as well but has not been nearly as well documented. The failure of VBG, however, which depended on the restriction of a small outflow and maintained a fixed small gastric outfow, should have also provided evidence that anastomotic size is not an important determinant of weight loss after bariatric surgery.

Revisional bariatric surgery could encompass quite a long list of relatively small series of revisional procedures of all existing operations. The success of any of these series has been limited. Clearly if one overwhelmingly successful bariatric operation had been discovered among these perturbations and variations on the theme, it is likely it would have been greeted with considerable enthusiasm and publicity. This has not occurred. Therefore, this text does not delve into further details regarding revisional bariatric surgery until data exist supporting its use for improved outcomes over any of the existing index bariatric operations. To date no such data exist.

It also should be freely admitted that the lead author follows the philosophy noted previously, as to whether the operation or the patient has failed as being the determinant of whether revisional surgery is indicated.

Special Situations

Pregnancy

It has been well established that the desire to bear children does not preclude a woman from having bariatric surgery. Normal pregnancy and childbirth without risk to the fetus is the norm for patients who have had LAGB.124 The ability to loosen the band by removing fluid from the reservoir is a feature that is desirable for pregnancy, especially if the pregnancy produces any limited ability to eat and drink itself. It has also been well established that patients who have recovered from LRYGB are at minimal to no risk for any problems during pregnancy and childbirth due to the operation.125 It is generally recommended that patients undergoing LRYGB not become pregnant soon after surgery, during the rapid weight loss phase of the postoperative period. While even this probably has only small risks to the pregnancy, it is best to avoid pregnancy during this time, when weight loss is inevitable and the change in body hormone composition is ongoing due to the operation. Pregnancy during this rapid weight loss phase is a more challenging problem to ensure that the mother has adequate nutrition for the fetus.

Age Limit

There is no distinct guideline for age limit for bariatric surgery. On the younger side, performing bariatric operations for teenagers is well established.126,127 The youngest appropriate age for a bariatric operation is controversial and still unknown. Limiting further growth potential is a concern when performing bariatric surgery on the adolescent population. This must be weighed against the risk of a high likelihood of lifelong of severe obesity if the adolescent does not lose weight and become closer to normal weight by the time adulthood has arrived. Currently in the United States, the laparoscopic adjustable band is not FDA approved for individuals younger than 18 years of age. This is somewhat ironic in that many bariatric surgeons who perform bariatric surgery for the pediatric age group feel that LAGB offers the best option for this age group, providing weight loss with the least amount of nutrient malabsorption and the most ability to reverse or revise the operation in the future. Indications for performing bariatric surgery itself for adolescent individuals involve all the same steps and precautions as for adults. Parental consent is obviously needed, and preoperative education, planning, and counseling for a bariatric operation must by necessity be a family affair in the pediatric and adolescent age group. It is controversial as to which operation is best performed for individuals younger than 18 years of age, but currently in the United States LRYGB seems to be the procedure of choice for severely obese adolescents.

Upper age limit for bariatric surgery is similarly controversial. Data do show that in experienced surgeon's hands, individuals who are older than 65 will still have comparable outcomes to their younger counterparts after bariatric surgery.128,129 However, there must be some limitation of age after which the natural aging process and the toll of organ dysfunction at that age make performance of bariatric surgery an unwise decision. This age is likely more a functional than a chronologic one, and it should be the judgment of the bariatric surgeon to determine age appropriateness for patients for surgery. Individuals who present as candidates for surgery and have been severely obese all their lives are less likely to have good organ function than those prospective patients who became severely obese later in life. The lead author's personal philosophy on this topic is that individuals who have been severely obese all their lives tend to have significant organ dysfunction over age 60 and more so over age 65, and become less optimal candidates for any bariatric operation at that age. However, there are always exceptions to this general guideline. It should also be remembered that the goal of bariatric surgery is to produce improved function and quality of life for a significant period of time in the foreseeable future to warrant the risk of the operation. When life expectancy is limited by age, the potential gain from the operation is thus limited as well.

Weight Limits

While the NIH guidelines are clear as to when bariatric surgery should be performed for individuals in terms of being heavy enough to warrant surgery, there are no guidelines as to the patient who presents at the upper echelons of weight. Certainly it has been shown that patients with a BMI exceeding 50 have increased risk of mortality from any complication after bariatric surgery.130 There are few reports in the literature about performing bariatric surgery for patients with BMI over 70 on a regular basis. We have recently reviewed our own institutional experience and found that patients with a BMI over 60 have had a surprisingly low mortality of under 2% after LRYGB and RYGB. Based on the follow-up study of Christou et al,50 this patient population would have had a much higher mortality than observed without bariatric surgery. Surgery on individuals with exceedingly high weights is complicated by the logistics of providing care for them as well as the ability to safely complete any bariatric operation. Assuming the latter itself could be overcome, the former still poses a major problem for hospitals in terms of imaging capacity, nursing care and patient hygiene, transportation, and the capacity of hospital clothing, instruments such as blood pressure cuffs, and other aspects of patient care. The Centers of Excellence model of the ACS and ASMBS have required institutions to provide such special availability of equipment for individuals undergoing bariatric surgery, but even those institutions are hard pressed to have any capacity to routinely accommodate the patient with a BMI over 80 or a weight over 500 lb for the above list of needs and equipment. No clear guidelines exist as to what an appropriate upper weight or BMI limit should be for the performance of bariatric surgery at this time.

Endoscopic Surgery

The ideal operation to produce weight loss in the bariatric population would be one that could be done without any incisions, but endoscopically. Currently no procedure has been established as having the ability to endoscopically convey durable weight loss for patients. Recent trials of such endoscopic procedures have taken place, only to fail to produce durable weight loss. Short-term nondurable weight loss has been reported for space-occupying devices such as intragastric balloons. However, these procedures are to be viewed with great skepticism, as they can only produce short-term weight loss that must then be followed by performance of a more established bariatric operation if the benefits of any weight loss are to be preserved or enhanced. Performing a single procedure for patients is always more desirable.13

It is likely that more trials of endoscopic procedures will continue as the ability of technology to perform endoscopic suturing, tissue approximation, and other key steps in the performance of surgical operations evolves. Perhaps at the next edition of this text, such an operation will be included in a chapter on this subject.