2896fb

Corporate Medical Policy

Title: Bariatric Surgery

Policy # 7.01.47.0

 

Last Review Date: February 2018

Next Review Date: February 2019

 

 

Policy DISCLAIMER

Current medical policy is to be used in determining a Member's contract benefits on the date that services are rendered.  Contract language, including definitions and specific inclusions/ exclusions, as well as state and federal law, must be considered in determining eligibility for coverage.  Members must consult their applicable benefit plans or contact a Member Services representative for specific coverage information. Likewise, medical policy, which addresses the issue(s) in any specific case, should be considered before utilizing medical opinion in adjudication.  Medical technology is constantly evolving and the Company reserves the right to review and update medical policy periodically.

Bariatric Surgery

Populations

Interventions

Comparators

Outcomes

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Gastric bypass

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Laparoscopic adjustable gastric banding

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Sleeve gastrectomy

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity


Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Biliopancreatic diversion with duodenal switch

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Biliopancreatic diversion without duodenal switch

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Vertical-banded gastroplasty

 

 

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Two-stage bariatric surgery procedures

 

Comparators of interest are:

·  One-stage bariatric surgery procedure

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Laparoscopic gastric plication

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Single anastomosis duodenoileal bypass with sleeve gastrectomy

 

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Duodenojejunal sleeve

 

 

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Intragastric balloon devices

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity

 

Interventions of interest are:

·  Aspiration therapy device

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adults with morbid obesity and failed bariatric surgery

 

Interventions of interest are:

·  Revision bariatric surgery

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are diabetic and not morbidly obese

Interventions of interest are:

·  Gastric bypass

·  Sleeve gastrectomy

·  Biliopancreatic diversion

·  Adjustable gastric banding

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are not diabetic and not morbidly obese

Interventions of interest are:

·  Any bariatric surgery procedure

 

 

Comparators of interest are:

·  Standard medical care

 

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adolescent children with morbid obesity

 

Interventions of interest are:

·  Gastric bypass, laparoscopic adjustable gastric banding, or sleeve gastrectomy

Comparators of interest are:

·  Standard medical care

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are adolescent children with morbid obesity

 

Interventions of interest are:

·  Bariatric surgery other than gastric bypass, laparoscopic adjustable gastric banding, or sleeve gastrectomy

Comparators of interest are:

·  Standard medical care

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  Who are preadolescent children with morbid obesity

 

Interventions of interest are:

·  Bariatric surgery

 

Comparators of interest are:

·  Standard medical care

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Individuals:

·  With morbid obesity and a preoperative diagnosis of hiatal hernia

Interventions of interest are:

·  Hiatal hernia repair with bariatric surgery

Comparators of interest are:

·  Bariatric surgery alone

 

Relevant outcomes include:

·  Overall survival

·  Change in disease status

·  Functional outcomes

·  Health status measures

·  Quality of life

·  Treatment-related mortality

·  Treatment-related morbidity

Summary

Bariatric surgery is a treatment for morbid obesity in patients who fail to lose weight with conservative measures. There are numerous gastric and intestinal surgical techniques available. While these techniques have heterogeneous mechanisms of action, the result is a smaller gastric pouch that leads to restricted eating. However, these surgeries may lead to malabsorption of nutrients or eventually to metabolic changes.

Adults With Morbid Obesity

For individuals who are adults with morbid obesity who receive gastric bypass, the evidence includes randomized controlled trials (RCTs), observational studies, and systematic reviews. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. TEC Assessments and other systematic reviews of RCTs and observational studies found that gastric bypass improves health outcomes, including weight loss and remission of type 2 diabetes. A TEC Assessment found similar weight loss with open and laparoscopic gastric bypass. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who are adults with morbid obesity who receive laparoscopic adjustable gastric banding (LAGB), the evidence includes RCTs, observational studies, and systematic reviews. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. Systematic reviews of RCTs and observational studies have found that LAGB is a reasonable alternative to gastric bypass. There is less weight loss with LAGB than with gastric bypass, but LAGB is less invasive and is associated with fewer serious adverse events. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who are adults with morbid obesity who receive sleeve gastrectomy (SG), the evidence includes RCTs, observational studies (evaluating SG alone and comparing SG with gastric bypass), as well as systematic reviews. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. Systematic reviews of RCTs and observational studies have found that SG results in substantial weight loss and that this weight loss is durable for at least 5 years. A meta-analysis found that short-term weight loss was similar after SG compared with gastric bypass. Long-term weight loss was greater after gastric bypass, but SG is associated with fewer adverse events. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who are adults with morbid obesity who receive biliopancreatic diversion (BPD) with duodenal switch, the evidence includes nonrandomized comparative studies, observational studies, and a systematic review. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. Nonrandomized comparative studies have found significantly higher weight loss after BPD with duodenal switch compared with gastric bypass at 1 year. A large case series found sustained weight loss after 7 years. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

For individuals who are adults with morbid obesity who receive BPD without duodenal switch, the evidence includes observational studies and systematic reviews. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. A TEC Assessment reviewed the available observational studies and concluded that weight loss was similar after BPD without a duodenal switch or gastric bypass. However, concerns have been raised about complications associated with BPD without duodenal switch, especially long-term nutritional and vitamin deficiencies. The evidence is insufficient to determine the effects of the technology on health outcomes.

For individuals who are adults with morbid obesity who receive vertical-banded gastroplasty, the evidence includes observational studies and systematic reviews. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. A TEC Assessment identified 8 nonrandomized comparative studies evaluating vertical-banded gastroplasty, and these studies found that weight loss was significantly greater with open gastric bypass. Moreover, vertical-banded gastroplasty has relatively high rates of complications, revisions, and reoperations. The evidence is insufficient to determine the effects of the technology on health outcomes.

For individuals who are adults with morbid obesity who receive 2-stage bariatric surgery procedures, the evidence includes a small RCT and observational studies. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. There is a lack of evidence that 2-stage bariatric procedures improve outcomes compared with 1-stage procedures. The small RCT compared intragastric balloon (IGB) plus gastric bypass with the standard of care plus gastric bypass and did not detect a difference in weight loss at 6 months postsurgery. Case series have shown relatively high complication rates in 2-stage procedures, and patients are at risk of complications in both stages. The evidence is insufficient to determine the effects of the technology on health outcomes.

For individuals who are adults with morbid obesity who receive laparoscopic gastric plication, the evidence includes 2 RCTs, observational studies, and systematic reviews. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. A 2014 systematic review only identified a small nonrandomized comparative study comparing laparoscopic gastric plication with other bariatric surgery procedures. Since the systematic review, 2 RCTs have been published, one comparing laparoscopic gastric plication with a sham procedure and another comparing laparoscopic gastric plication with SG. Laparoscopic gastric plication was more effective than sham at 1-year follow-up and equally effective as SG at 2-year follow-up. Additional comparative studies and RCTs with longer follow-up are needed to permit conclusions about the safety and efficacy of laparoscopic gastric plication. The evidence is insufficient to determine the effects of the technology on health outcomes.

For individuals who are adults with morbid obesity who receive single anastomosis duodenoileal bypass with SG (SADI-S), the evidence includes observational studies. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. No controlled trials have evaluated SADI-S. There are a few case series, the largest of which had fewer than 100 patients. A retrospective chart review of patients receiving gastric bypass, BPD, and SADI-S, reported that among patients without diabetes, SADI-S was more effective in weight loss and cholesterol outcomes than gastric bypass. Among patients with diabetes, SADI-S and BDP had higher remission rates than gastric bypass. Comparative studies and especially RCTs are needed to permit conclusions about the safety and efficacy of SADI-S. The evidence is insufficient to determine the effects of the technology on health outcomes.

For individuals who are adults with morbid obesity who receive duodenojejunal sleeve, the evidence includes RCTs and systematic reviews. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. A systematic review of duodenojejunal sleeves included 5 RCTs and found significantly greater short-term weight loss (12-24 weeks) with the sleeves compared with medical therapy. There was no significant difference in symptoms associated with diabetes. All RCTs were small and judged by systematic reviewers to be at high-risk of bias. High-quality comparative studies are needed to permit conclusions on the safety and efficacy of the procedure. The evidence is insufficient to determine the effects of the technology on health outcomes.

For individuals who are adults with morbid obesity who receive IGB devices, the evidence includes RCTs, systematic reviews, and case series. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. RCTs assessing the 2 IGB devices approved by the Food and Drug Administration have found significantly greater weight loss with IGB than with sham treatment or lifestyle therapy alone after 6 months (maximum length of device use). Some adverse events were reported, mainly related to accommodation of the balloon in the stomach; in a minority of cases, these adverse events were severe. One RCT followed patients for an additional 6 months after IGB removal and found sustained weight loss. There are limited data on the durability of weight loss in the long-term. Comparative data are lacking. A large case series found that patients gradually regained weight over time. Moreover, it is unclear how 6 months of IGB use would fit into a long-term weight loss and maintenance intervention. The evidence is insufficient to determine the effects of the technology on health outcomes.

For individuals who are adults with morbid obesity who receive an aspiration therapy device, the evidence includes an RCT and case series. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. The RCT found significantly greater weight loss with aspiration therapy than lifestyle therapy at 1 year. One small case series reported on 15 patients at 2 years. The total amount of data on aspiration therapy remains limited and additional studies are needed before conclusions can be drawn about the effects of treatment on weight loss, metabolism and nutrition and long-term durability of treatment. The evidence is insufficient to determine the effects of the technology on health outcomes.

Revision Bariatric Surgery

For individuals who are adults with morbid obesity and failed bariatric surgery who receive revision bariatric surgery, the evidence includes case series and registry data. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. Case series have shown that patients receiving revision bariatric surgery experienced satisfactory weight loss. Data from a multinational bariatric surgery database has found that corrective procedures following primary bariatric surgery are relatively uncommon but generally safe and efficacious. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

Adults With Type 2 Diabetes

For individuals who are diabetic and not morbidly obese who receive gastric bypass, SG, BPD, or LAGB, the evidence includes RCTs, nonrandomized comparative studies, and case series. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. Systematic reviews of RCTs and observational studies have found that certain types of bariatric surgery are more efficacious than medical therapy as a treatment for type 2 diabetes in obese patients, including those with a body mass index (BMI) between 30 and 34.9 kg/m2. The greatest amount of evidence is on gastric bypass. Systematic reviews have found significantly greater remission rates of diabetes, decrease in hemoglobin A1c levels, and decrease in BMI with bariatric surgery than with nonsurgical treatment. The efficacy of surgery is balanced against the short-term risks of the surgical procedure. Most RCTs in this population have 1 to 3 years of follow-up; 1 RCT that included patients with BMI between 30 and 34.9 kg/m2 had 5-year follow-up data. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

There are clinical concerns about durability and long-term outcomes at 5 to 10 years as well as potential variation in observed outcomes in community practice vs clinical trials. As a result, bariatric surgery for individuals who are diabetic and not morbidly obese is considered not medically necessary.

Nondiabetic and Nonobese Adults

For individuals who are not diabetic and not morbidly obese who receive any bariatric surgery procedure, the evidence includes RCTs, nonrandomized comparative studies, and case series. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. There is limited evidence for bariatric surgery in patients who are not diabetic or morbidly obese. A few small RCTs and case series have reported a loss of weight and improvements in comorbidities for this population. However, the evidence does not permit conclusions on the long-term risk-benefit ratio of bariatric surgery in this population. The evidence is insufficient to determine the effects of the technology on health outcomes.

Adolescent Children With Morbid Obesity

Gastric Bypass, LAGB, or SG

For individuals who are adolescent children with morbid obesity who receive gastric bypass, or LAGB, or SG, the evidence includes RCTs, observational studies, and systematic reviews. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. Systematic reviews of studies on bariatric surgery in adolescents, who mainly received gastric bypass or LAGB or SG, found significant weight loss and reductions in comorbidity outcomes with bariatric surgery. For bariatric surgery in the adolescent population, although data are limited on some procedures, studies have generally reported that weight loss and reduction in risk factors for adolescents are similar to that for adults. Most experts and clinical practice guidelines have recommended that bariatric surgery in adolescents be reserved for individuals with severe comorbidities, or for individuals with a BMI greater than 50 kg/m2. Also, greater consideration should be placed on the patient developmental stage, on the psychosocial aspects of obesity and surgery, and on ensuring that the patient can provide fully informed consent. The evidence is sufficient to determine that the technology results in a meaningful improvement in the net health outcome.

Bariatric Surgery Other Than Gastric Bypass, LAGB, or SG

For individuals who are adolescent children with morbid obesity who receive bariatric surgery other than gastric bypass, or LAGB, or SG, the evidence includes systematic reviews and a cohort study. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. Studies using bariatric surgery other than gastric bypass, LAGB, or SG, have small sample sizes. Results from a meta-analysis including patients using other procedures have shown significant improvements in BMI reduction, fasting blood insulin, and total cholesterol, although the estimates have wide confidence intervals, limiting interpretation. The evidence is insufficient to determine the effects of the technology on health outcomes.

Preadolescent Children With Morbid Obesity

For individuals who are preadolescent children with morbid obesity who receive bariatric surgery, the evidence includes no studies focused on this population. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. Several studies of bariatric surgery in adolescents have also included children younger than 12 years old, but findings were not reported separately for preadolescent children. Moreover, clinical practice guidelines have recommended against bariatric surgery for preadolescent children. The evidence is insufficient to determine the effects of the technology on health outcomes.

Hiatal Hernia Repair with Bariatric Surgery

For individuals with morbid obesity and a preoperative diagnosis of a hiatal hernia who receive hiatal hernia repair with bariatric surgery, the evidence includes cohort studies and case series. Relevant outcomes are overall survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. Results from the cohort studies and case series have shown that, when a preoperative diagnosis of a hiatal hernia has been present, repairing the hiatal hernia during bariatric surgery resulted in fewer complications. However, the results are limited to individuals with a preoperative diagnosis. There was no evidence on the use of hiatal hernia repair when the hiatal hernia diagnosis is incidental. The evidence is sufficient to determine that the technology results in an improvement in the net health outcome.

Objective

The objective of this evidence review is to evaluate whether various bariatric surgery procedures improve the net health outcome in adults who are obese, in adolescents and preadolescents who are obese, and in those with diabetes who are obese.

Policy

Bariatric Surgery in Adults With Morbid Obesity

The following bariatric surgery procedures may be considered medically necessary for the treatment of morbid obesity (see Policy Guidelines section for patient selection criteria) in adults who have failed weight loss by conservative measures:

·         Open gastric bypass using a Roux-en-Y anastomosis

·         Laparoscopic gastric bypass using a Roux-en-Y anastomosis

·         Laparoscopic adjustable gastric banding

·         Sleeve gastrectomy

·         Open or laparoscopic biliopancreatic bypass (ie, Scopinaro procedure) with duodenal switch.

Bariatric surgery should be performed in appropriately selected patients, by surgeons who are adequately trained and experienced in the specific techniques used, and in institutions that support a comprehensive bariatric surgery program, including long-term monitoring and follow-up postsurgery.

The following bariatric surgery procedures are considered investigational for the treatment of morbid obesity in adults who have failed weight loss by conservative measures:

·         Vertical-banded gastroplasty

·         Gastric bypass using a Billroth II type of anastomosis (mini-gastric bypass)

·         Biliopancreatic diversion without duodenal switch

·         Long-limb gastric bypass procedure (ie, >150 cm)

·         Two-stage bariatric surgery procedures (eg, sleeve gastrectomy as initial procedure followed by biliopancreatic diversion at a later time)

·         Laparoscopic gastric plication

·         Single anastomosis duodenoileal bypass with sleeve gastrectomy.

The following endoscopic procedures are investigational as a primary bariatric procedure or as a revision procedure (ie, to treat weight gain after bariatric surgery to remedy large gastric stoma or large gastric pouches):

·         Insertion of the StomaphyX™ device

·         Endoscopic gastroplasty

·         Use of an endoscopically placed duodenojejunal sleeve

·         Intragastric balloons

·         Aspiration therapy device.

Bariatric Surgery in Patients With a Body Mass Index Less Than 35 kg/m2

Bariatric surgery is considered not medically necessary for patients with a body mass index less than 35 kg/m2.

Revision Bariatric Surgery

Revision surgery to address perioperative or late complications of a bariatric procedure is considered medically necessary. They include but are not limited to, staple line failure, obstruction, stricture, nonabsorption resulting in hypoglycemia or malnutrition, weight loss of 20% or more below ideal body weight, and band slippage that cannot be corrected with manipulation or adjustment (see Policy Guidelines section).

Revision of a primary bariatric procedure that has failed due to dilation of the gastric pouch or dilation proximal to an adjustable gastric band (documented by upper gastrointestinal examination or endoscopy) is considered medically necessary if the initial procedure was successful in inducing weight loss prior to pouch dilation, and the patient has been compliant with a prescribed nutrition and exercise program.

Bariatric Surgery in Adolescents

Bariatric surgery in adolescents may be considered medically necessary according to the same weight-based criteria used for adults, but greater consideration should be given to psychosocial and informed consent issues (see Policy Guidelines section). In addition, any devices used for bariatric surgery must be used in accordance with the U.S. Food and Drug Administration‒approved indications.

Bariatric Surgery in PreAdolescent Children

Bariatric surgery is considered investigational for the treatment of morbid obesity in preadolescent children.

Concomitant Hiatal Hernia Repair With Bariatric Surgery

Repair of a hiatal hernia at the time of bariatric surgery may be considered medically necessary for patients who have a preoperatively diagnosed hiatal hernia with indications for surgical repair (see Policy Guidelines section).

Repair of a hiatal hernia that is diagnosed at the time of bariatric surgery, or repair of a preoperatively diagnosed hiatal hernia in patients who do not have indications for surgical repair, is considered investigational.

Policy Guidelines

Patient Selection Criteria

Morbid obesity is defined as a body mass index (BMI) 40 kg/m2 or more or a BMI 35 kg/m2 or more with at least 1 clinically significant obesity-related disease such as diabetes, obstructive sleep apnea, coronary artery disease, or hypertension for which these complications or diseases are not controlled by best practice medical management.

While there is limited evidence on which to assess the long-term impacts of bariatric surgery for patients younger than age 18 years, very severely obese (BMI ≥40 kg/m2) adolescents with serious obesity-related comorbidities that are poorly controlled, or who have a BMI of 50 kg/m2 or greater with less severe comorbidities may be considered for bariatric surgery. U.S. Food and Drug Administration (FDA) premarket approval for the LAP-BAND® System indicates it is intended for severely obese adults. (The clinical study submitted to FDA for approval of the LAP-BAND was restricted to adults ages 18-55 years.)

Patients should have documented failure to respond to conservative measures for weight reduction prior to consideration of bariatric surgery, and these attempts should be reviewed by the practitioner prior to seeking approval for the surgical procedure. As a result, some centers require active participation in a formal weight reduction program that includes frequent documentation of weight, dietary regimen, and exercise. However, there is a lack of evidence on the optimal timing, intensity, and duration of nonsurgical attempts at weight loss, and whether a medical weight loss program immediately preceding surgery improves outcomes.

Patients with a BMI of 50 kg/m2 or more need a bariatric procedure to achieve greater weight loss. Thus, use of adjustable gastric banding, which results in less weight loss, should be most useful as a procedure for patients with a BMI less than 50 kg/m2. Malabsorptive procedures, although they produce more dramatic weight loss, potentially result in nutritional complications, and the risks and benefits of these procedures must be carefully weighed in light of the treatment goals for each patient.

Patients who undergo adjustable gastric banding and fail to achieve adequate weight loss must show evidence of postoperative compliance with diet and regular bariatric visits prior to consideration of a second bariatric procedure.

Considerations for Bariatric Surgery in Adolescents

Guidelines for bariatric surgery in adolescents are not uniform, with variability in weight-based criteria, ranging from a BMI of 35 kg/m2 with comorbidities to a BMI of 50 kg/m2. Most guidelines use weight-based criteria that parallel those for adults.

In addition to the weight-based criteria, there is greater emphasis on issues of developmental maturity, psychosocial status, and informed consent for adolescent patients. All guidelines mention these issues, but recommendations are not uniform for addressing them. The following are examples from U.S. guidelines published since 2008 that address issues of maturity and psychosocial status.

Endocrine Society

·         The child has attained Tanner 4 or 5 pubertal development and final or near-final adult height.

·         Psychological evaluation confirms the stability and competence of the family unit.

·         The patient demonstrates the ability to adhere to the principles of healthy dietary and activity habits (August et al, 2008).

Institute for Clinical Systems Improvement

The Institute for Clinical Systems Improvement’s 2013 obesity guidelines have indicated that bariatric surgery should only be considered in the pediatric population under the following conditions (Fitch et al, 2013).

·         “The child has a BMI > 40 kg/m2 or has BMI above 35 kg/m2 with a significant, severe comorbidities such as type 2 diabetes mellitus, obstructive sleep apnea, or pseudotumor cerebri.”

·         “The child has attained Tanner 4 or 5 pubertal development or has a bone age ≥13 years in girls or ≥15 years in boys.”

·         “Failure of ≥6 months of organized attempts at weight management….”

·         “The adolescent should have decisional capacity and also demonstrate commitment to comprehensive medical and psychological evaluation before and after surgery.”

·         “A supportive family environment….”

The choice of procedure in adolescents may also differ from adults, but there is a lack consensus in guidelines or expert opinion as to the preferred procedure(s) for adolescents. The following factors should be considered in the choice of bariatric surgery in adolescents (Aikenhead A, Lobstein T, Knai C. Review of current guidelines on adolescent bariatric surgery. Clin Obes. Feb 2011;1(1):3-11. PMID 25586970):

·         As in adults, laparoscopic gastric bypass is the most common procedure in adolescents.

·         Devices used for laparoscopic adjustable gastric banding do not have FDA approval in the United States for individuals younger than age 18 years.

·         Some guidelines for bariatric surgery in adolescents do not recommend biliopancreatic diversions because of the greater frequency of nutritional deficiencies on long-term follow-up, but other guidelines do not specify that biliopancreatic diversion not be done in adolescents.

Hiatal Hernia Repair Guidelines

The Society of American Gastrointestinal and Endoscopic Surgeons issued evidence-based guidelines for the management of hiatal hernia (Kohn et al, 2013). The Society noted that the general methodologic quality of available studies is low. Recommendations for indications for repair are as follows:

·         “Repair of a type I hernia [sliding hiatal hernias, where the gastroesophageal junction migrates above the diaphragm] in the absence of reflux disease is not necessary” (moderate quality evidence, strong recommendation).

·         “All symptomatic paraesophageal hiatal hernias should be repaired [high-quality evidence, strong recommendation], particularly those with acute obstructive symptoms or which have undergone volvulus.”

·         “Routine elective repair of completely asymptomatic paraesophageal hernias may not always be indicated. Consideration for surgery should include the patient’s age and co-morbidities” (moderate quality evidence, weak recommendation).

Coding

Hiatal hernia repair performed at the time of bariatric surgery would not be reported with the hiatal hernia repair code. There is no code for this specific surgery; therefore, it should be reported with code 43289 (Unlisted laparoscopy procedure, esophagus).

Benefit Application

BlueCard/National Account Issues

State mandates and contractual exclusions may apply to coverage eligibility of bariatric surgery in general.

State or federal mandates (eg, Federal Employee Program) may dictate that certain U.S. Food and Drug Administration‒approved devices, drugs, or biologics may not be considered investigational, and thus these devices may be assessed only by their medical necessity.

Background

Bariatric Surgery

Bariatric surgery is performed to treat morbid (clinically severe) obesity. Morbid obesity is defined as a body mass index (BMI) greater than 40 kg/m2 or a BMI greater than 35 kg/m2 with associated complications including, but not limited to, diabetes, hypertension, or obstructive sleep apnea. Morbid obesity results in a very high-risk for weight-related complications, such as diabetes, hypertension, obstructive sleep apnea, and various types of cancers (for men: colon, rectal, prostate; for women: breast, uterine, ovarian), and a shortened lifespan. A morbidly obese man at age 20 can expect to live 13 fewer years than his counterpart with a normal BMI, which equates to a 22% reduction in life expectancy.

The first treatment of morbid obesity is dietary and lifestyle changes. Although this strategy may be effective in some patients, only a few morbidly obese individuals can reduce and control weight through diet and exercise. Most patients find it difficult to comply with these lifestyle modifications on a long-term basis.

When conservative measures fail, some patients may consider surgical approaches. A 1991 National Institutes of Health Consensus Conference defined surgical candidates as “those patients with a BMI of greater than 40 kg/m2, or greater than 35 kg/m2 in conjunction with severe comorbidities such as cardiopulmonary complications or severe diabetes.”1

Resolution (cure) or improvement of type 2 diabetes after bariatric surgery and observations that glycemic control may improve immediately after surgery before a significant amount of weight is lost have promoted interest in a surgical approach to the treatment of type 2 diabetes. The various surgical procedures have different effects, and gastrointestinal rearrangement seems to confer additional antidiabetic benefits independent of weight loss and caloric restriction. The precise mechanisms are not clear, and multiple mechanisms may be involved. Gastrointestinal peptides, eg, glucagon-like peptide-1, glucose-dependent insulinotropic peptide, and peptide YY, are secreted in response to contact with unabsorbed nutrients and by vagally mediated parasympathetic neural mechanisms. Glucagon-like peptide-1 is secreted by the L cells of the distal ileum in response to ingested nutrients and acts on pancreatic islets to augment glucose-dependent insulin secretion. It also slows gastric emptying, which delays digestion, blunts postprandial glycemia, and acts on the central nervous system to induce satiety and decrease food intake. Other effects may improve insulin sensitivity. Glucose-dependent insulinotropic peptide acts on pancreatic beta cells to increase insulin secretion through the same mechanisms as glucagon-like peptide-1, although it is less potent. Peptide YY is also secreted by the L cells of the distal intestine and increases satiety and delays gastric emptying.

Types of Bariatric Surgery Procedures

The following summarizes the most common bariatric surgery procedures.

Open Gastric Bypass

The original gastric bypass surgeries were based on the observation that postgastrectomy patients tended to lose weight. The current procedure (CPT code 43846) involves both a restrictive and a malabsorptive component, with the horizontal or vertical partition of the stomach performed in association with a Roux-en-Y procedure (ie, a gastrojejunal anastomosis). Thus, the flow of food bypasses the duodenum and proximal small bowel. The procedure may also be associated with an unpleasant “dumping syndrome,” in which a large osmotic load delivered directly to the jejunum from the stomach produces abdominal pain and/or vomiting. The dumping syndrome may further reduce intake, particularly in “sweets eaters.” Surgical complications include leakage and operative margin ulceration at the anastomotic site. Because the normal flow of food is disrupted, there are more metabolic complications than with other gastric restrictive procedures, including iron deficiency anemia, vitamin B12 deficiency, and hypocalcemia, all of which can be corrected by oral supplementation. Another concern is the ability to evaluate the “blind” bypassed portion of the stomach. Gastric bypass may be performed with either an open or laparoscopic technique.

Note: In 2005, CPT code 43846 was revised to indicate that the short limb must be 150 cm or less, compared with the previous 100 cm. This change reflects the common practice in which the alimentary (ie, jejunal limb) of a gastric bypass has been lengthened to 150 cm. This length also serves to distinguish a standard gastric bypass with a very long, or very, very long gastric bypass, as discussed further here.

Laparoscopic Gastric Bypass

CPT code 43644 was introduced in 2005 and described the same procedure as open gastric bypass (CPT code 43846), but performed laparoscopically.

Adjustable Gastric Banding

Adjustable gastric banding (CPT code 43770) involves placing a gastric band around the exterior of the stomach. The band is attached to a reservoir implanted subcutaneously in the rectus sheath. Injecting the reservoir with saline will alter the diameter of the gastric band; therefore, the rate-limiting stoma in the stomach can be progressively narrowed to induce greater weight loss, or expanded if complications develop. Because the stomach is not entered, the surgery and any revisions, if necessary, are relatively simple.

Complications include slippage of the external band or band erosion through the gastric wall. Adjustable gastric banding has been widely used in Europe. Two banding devices are approved by the Food and Drug Administration (FDA) for marketing in the United States. The first to receive FDA approval was the LAP-BAND (original applicant, Allergan, BioEnterics, Carpinteria, CA; now Apollo Endosurgery, Austin, TX). The labeled indications for this device are as follows:

"The LAP-BAND® system is indicated for use in weight reduction for severely obese patients with a body mass index (BMI) of at least 40 or a BMI of at least 35 with one or more severe comorbid conditions, or those who are 100 lb or more over their estimated ideal weight according to the 1983 Metropolitan Life Insurance Tables (use the midpoint for medium frame). It is indicated for use only in severely obese adult patients who have failed more conservative weight-reduction alternatives, such as supervised diet, exercise and behavior modification programs. Patients who elect to have this surgery must make the commitment to accept significant changes in their eating habits for the rest of their lives."

In 2011, FDA-labeled indications for the LAP-BAND were expanded to include patients with a BMI from 30 to 34 kg/m2 with at least 1 obesity-related comorbid condition.

The second adjustable gastric banding device approved by FDA through the premarket approval process is the REALIZE® model (Ethicon Endo-Surgery, Cincinnati, OH). Labeled indications for this device are:

“Th[e REALIZE] device is indicated for weight reduction for morbidly obese patients and is indicated for individuals with a Body Mass Index of at least 40 kg/m2, or a BMI of at least 35 kg/m2 with one or more comorbid conditions. The Band is indicated for use only in morbidly obese adult patients who have failed more conservative weight-reduction alternatives, such as supervised diet, exercise, and behavior modification programs.”

Sleeve Gastrectomy

A sleeve gastrectomy (SG; CPT code 43775) is an alternative approach to gastrectomy that can be performed on its own or in combination with malabsorptive procedures (most commonly biliopancreatic diversion [BPD] with duodenal switch). In this procedure, the greater curvature of the stomach is resected from the angle of His to the distal antrum, resulting in a stomach remnant shaped like a tube or sleeve. The pyloric sphincter is preserved, resulting in a more physiologic transit of food from the stomach to the duodenum and avoiding the dumping syndrome (overly rapid transport of food through the stomach into intestines) seen with distal gastrectomy. This procedure is relatively simple to perform and can be done as an open or laparoscopic procedure. Some surgeons have proposed the SG as the first in a 2-stage procedure for very high-risk patients. Weight loss following SG may improve a patient’s overall medical status and, thus, reduce the risk of a subsequent more extensive malabsorptive procedure (eg, BPD).

Biliopancreatic Diversion

The BPD procedure (also known as the Scopinaro procedure; CPT code 43847), developed and used extensively in Italy, was designed to address drawbacks of the original intestinal bypass procedures that have been abandoned due to unacceptable metabolic complications. Many complications were thought to be related to bacterial overgrowth and toxin production in the blind, bypassed segment. In contrast, BPD consists of a subtotal gastrectomy and diversion of the biliopancreatic juices into the distal ileum by a long Roux-en-Y procedure. The procedure consists of the following components:

a.     A distal gastrectomy induces temporary early satiety and/or the dumping syndrome in the early postoperative period, both of which limit food intake.

b.     A 200-cm long “alimentary tract” consists of 200 cm of ileum connecting the stomach to a common distal segment.

c.     A 300- to 400-cm “biliary tract” connects the duodenum, jejunum, and remaining ileum to the common distal segment.

d.     A 50- to 100-cm “common tract” is where food from the alimentary tract mixes with biliopancreatic juices from the biliary tract. Food digestion and absorption, particularly of fats and starches, are therefore limited to this small segment of bowel, creating selective malabsorption. The length of the common segment will influence the degree of malabsorption.

e.     Because of the high incidence of cholelithiasis associated with the procedure, patients typically undergo an associated cholecystectomy.

Many potential metabolic complications are related to BPD, including, most prominently, iron deficiency anemia, protein malnutrition, hypocalcemia, and bone demineralization. Protein malnutrition may require treatment with total parenteral nutrition. Also, several case reports have noted liver failure resulting in death or liver transplant.

BPD With Duodenal Switch

CPT code 43845, which specifically identifies the duodenal switch procedure, was introduced in 2005. The duodenal switch procedure is a variant of the BPD previously described. In this procedure, instead of performing a distal gastrectomy, a SG is performed along the vertical axis of the stomach. This approach preserves the pylorus and initial segment of the duodenum, which is then anastomosed to a segment of the ileum, similar to the BPD, to create the alimentary limb. Preservation of the pyloric sphincter is intended to ameliorate the dumping syndrome and decrease the incidence of ulcers at the duodenoileal anastomosis by providing a more physiologic transfer of stomach contents to the duodenum. The SG also decreases the volume of the stomach and decreases the parietal cell mass. However, the basic principle of the procedure is similar to that of the BPD, ie, producing selective malabsorption by limiting the food digestion and absorption to a short common ileal segment.

Vertical-Banded Gastroplasty

Vertical-banded gastroplasty (CPT code 43842) was formerly one of the most common gastric restrictive procedures performed in the United States but has now been replaced by other restrictive procedures due to high rates of revisions and reoperations. In this procedure, the stomach is segmented along its vertical axis. In order to create a durable reinforced and rate-limiting stoma at the distal end of the pouch, a plug of the stomach is removed, and a propylene collar is placed through this hole and then stapled to itself. Because the normal flow of food is preserved, metabolic complications are uncommon. Complications include esophageal reflux, dilation, or obstruction of the stoma, with the latter two requiring reoperation. Dilation of the stoma is a common reason for weight regain. Vertical-banded gastroplasty may be performed using an open or laparoscopic approach.

Long-Limb Gastric Bypass (ie, >150 cm)

Variations of gastric bypass procedures have been described, consisting primarily of long-limb Roux-en-Y procedures (CPT code 43847), which vary in the length of the alimentary and common limbs. For example, the stomach may be divided with a long segment of the jejunum (instead of ileum) anastomosed to the proximal gastric stump, creating the alimentary limb. The remaining pancreaticobiliary limb, consisting of stomach remnant, duodenum, and length of proximal jejunum, is then anastomosed to the ileum, creating a common limb of variable length in which the ingested food mixes with the pancreaticobiliary juices. While the long alimentary limb permits absorption of most nutrients, the short common limb primarily limits absorption of fats. The stomach may be bypassed in a variety of ways (eg, resection or stapling along the horizontal or vertical axis). Unlike the traditional gastric bypass, which is a gastric restrictive procedure, these very long-limb Roux-en-Y gastric bypasses combine gastric restriction with some element of malabsorptive procedure, depending on the location of the anastomoses. Note that CPT code for gastric bypass (43846) explicitly describes a short limb (<150 cm) Roux-en-Y gastroenterostomy, and thus would not apply to long-limb gastric bypass.

Laparoscopic Malabsorptive Procedure

CPT code 43645 was introduced in 2005, to specifically describe a laparoscopic malabsorptive procedure. However, the code does not specifically describe any specific malabsorptive procedure.

Weight Loss Outcomes

There is no uniform standard for reporting results of weight loss or for describing a successful procedure. Common methods of reporting the amount of body weight loss are the percent of ideal body weight achieved or percent of excess body weight (EBW) loss, with the latter most commonly reported. EBW is defined as actual weight minus “ideal weight” and “ideal weight” and is based on 1983 Metropolitan Life Insurance height-weight tables for “medium frame.”

These 2 reporting methods are generally preferred over the absolute amount of weight loss because they reflect the ultimate goal of surgery: to reduce weight to a range that minimizes obesity-related morbidity. Obviously, an increasing degree of obesity will require a greater amount of weight loss to achieve these target goals. There are different definitions of successful outcomes, but a successful procedure is often considered one in which at least 50% of EBW is lost, or when the patient returns to within 30% of ideal body weight. The results may also be expressed as the percentage of patients losing at least 50% of EBW. Table 1 summarizes the variations in reporting weight loss outcomes.

Table 1. Weight Loss Outcomes

Outcome Measure

Definition

Clinical Significance

Decrease in weight

Absolute difference in weight pre- and posttreatment

Unclear relation to outcomes, especially in morbidly obese

Decrease in BMI

Absolute difference in BMI pre- and posttreatment

May be clinically significant if change in BMI clearly leads to change in risk category

Percent EBW loss

Amount of weight loss divided by EBW

Has anchor to help frame clinical significance; unclear threshold for clinical significance

Percent patients losing >50% of EBW

No. patients losing >50% EBW divided by total patients

Additional advantage of framing on per patient basis. Threshold for significance (>50%) arbitrary.

Percent ideal body weight

Final weight divided by ideal body weight

Has anchor to help frame clinical significance; unclear threshold for clinical significance

BMI: body mass index; EBW: excess body weight.

Durability of Weight Loss

Weight change (ie, gain or loss) at yearly intervals is often reported. Weight loss at 1 year is considered the minimum length of time for evaluating these procedures; weight loss at 3 to 5 years is considered an intermediate time period for evaluating weight loss; and weight loss at 5 to 10 years or more is considered to represent long-term weight loss following bariatric surgery.

Short-Term Complications (Operative and Perioperative Complications <30 Days)

In general, the incidence of operative and perioperative complications is increased in obese patients, particularly in thromboembolism and wound healing. Other perioperative complications include anastomotic leaks, bleeding, bowel obstruction, and cardiopulmonary complications (eg, pneumonia, myocardial infarction).

Reoperation Rate

Reoperation may be required to “take down” or revise the original procedure. Reoperation may be particularly common in vertical-banded gastroplasty due to pouch dilation.

Long-Term Complications (Metabolic Adverse Events, Nutritional Deficiencies)

Metabolic adverse events are of particular concern in malabsorptive procedures. Other long-term complications include anastomotic ulcers, esophagitis, and procedure-specific complications such as band erosion or migration for gastric banding surgeries.

Improved Health Outcomes in Terms of Weight-Related Comorbidities

Aside from psychosocial concerns, which may be considerable, one motivation for bariatric surgery is to decrease the incidence of complications of obesity, such as diabetes, cardiovascular risk factors (ie, increased cholesterol, hypertension), obstructive sleep apnea, or arthritis. Unfortunately, these final health outcomes are not consistently reported.

Regulatory Status

Forms of bariatric surgery performed without specific implantable devices are surgical procedures and, as such, are not subject to regulation by the FDA.

Table 2 shows forms of bariatric surgery with implantable devices approved by the FDA through the premarket approval process.

Table 2. FDA-Approved Bariatric Surgery Devices

Device

Manufacturer

PMA Date

Labeled Indications

AspireAssist SystemÒ

Aspire Bariatrics

Jun 2016

For long-term use in conjunction with lifestyle therapy and continuous medical monitoring in obese adults >22 y, with a BMI of 35.0 to 55.0 kg/m2 and no contraindications to the procedure who have failed to achieve and maintain weight loss with nonsurgical weight loss therapy

ORBERAÒ intragastric balloon system

Apollo Endosurgery

Aug 2015

For use in obese adults (BMI, 30-40 kg/m2) who have failed weight reduction with diet and exercise, and have no contraindications. Maximum placement time is 6 mo. Balloon placed endoscopically and inflated with saline.

ReShapeÒ Integrated Dual Balloon System

ReShape Medical

Jul 2015

For use in obese adults (BMI, 30-40 kg/m2) and 1 comorbid conditions who have failed weight reduction with diet and exercise, and have no contraindications. Maximum placement time is 6 mo. Balloon delivered transorally and inflated with saline.

LAP-BANDÒ Adjustable Gastric Banding System

Apollo Endosurgery (original applicant: Allergan)

Apr 2010

For use in weight reduction for severely obese adults with BMI of at least 40 kg/m2 or a BMI of at least 30 kg/m2 with 1 severe comorbid conditions who have failed more conservative weight-reduction alternatives (eg, supervised diet, exercise, behavior modification programs).

REALIZEÒ Adjustable Gastric Band

 

Ethicon Endosurgery

Nov 2007

For use in weight reduction for morbidly obese patients and for individuals with BMI of at least 40 kg/m2, or a BMI of at least 35 kg/m2 with 1 comorbid conditions, or those who are 45.4 kg over their estimated ideal weight. Indicated for use only in morbidly obese adults who have failed more conservative weight-reduction alternatives (eg, supervised diet, exercise, behavior modification programs).

BMI: body mass index: FDA: Food and Drug Administration; PMA: premarket approval.

In February 2017, the FDA issued a letter to health care providers discussing the potential risks with liquid-filled intragastric balloons in response to reports of 2 types of adverse events related to the balloons. Several dozen reports concerned spontaneous overinflation of the balloons, which caused pain, swelling, and vomiting. The second set of adverse event reports indicated that acute pancreatitis developed in several patients due to compression of gastrointestinal structures. These reports involved both ReShape and ORBERA brands. The adverse events may require premature removal of the balloons.

In August 2017, the FDA issued a second letter to health care providers informing them of 5 unanticipated deaths occurring from 2016 through the time of the letter, due to intragastric balloons. The FDA recommended close monitoring of patients receiving these devices.

Rationale

This evidence review was created in July 1996 and has been updated regularly with searches of the MEDLINE database. The most recent literature update was performed through December 11, 2017.

Evidence reviews assess the clinical evidence to determine whether the use of a technology improves the net health outcome. Broadly defined, health outcomes are length of life, quality of life, and ability to function¾including benefits and harms. Every clinical condition has specific outcomes that are important to patients and to managing the course of that condition. Validated outcome measures are necessary to ascertain whether a condition improves or worsens; and whether the magnitude of that change is clinically significant. The net health outcome is a balance of benefits and harms.

To assess whether the evidence is sufficient to draw conclusions about the net health outcome of a technology, 2 domains are examined: the relevance and the quality and credibility. To be relevant, studies must represent one or more intended clinical use of the technology in the intended population and compare an effective and appropriate alternative at a comparable intensity. For some conditions, the alternative will be supportive care or surveillance. The quality and credibility of the evidence depend on study design and conduct, minimizing bias and confounding that can generate incorrect findings. The randomized controlled trial (RCT) is preferred to assess efficacy; however, in some circumstances, nonrandomized studies may be adequate. RCTs are rarely large enough or long enough to capture less common adverse events and long-term effects. Other types of studies can be used for these purposes and to assess generalizability to broader clinical populations and settings of clinical practice. The following is a summary of the key literature to date.

Overview: Bariatric Surgery in Adults With Morbid Obesity

There is a vast literature on bariatric surgery for adults with morbid obesity. This literature is characterized by a preponderance of single-arm clinical series from individual institutions. These types of studies can be used to determine the amount of weight loss expected from surgery, the durability of the weight loss, and the rate of adverse events. However, these studies are not adequate for determining the comparative efficacy of bariatric surgery vs conservative treatment, or the comparative efficacy of different bariatric surgery techniques. Some comparative trials, including randomized and nonrandomized designs, compare bariatric surgery with conservative therapy and/or compare outcomes of different bariatric surgery procedures. RCTs of bariatric surgery have been performed but are limited and insufficient to draw conclusions about comparisons of bariatric surgery and conservative treatments for weight loss.2 RCTs are difficult in bariatric surgery because many experts consider it inappropriate or unethical to randomize patients to bariatric surgery. Also, most patients and clinicians have strong preferences for treatment, which result in a select population that might agree to randomization and, therefore, limited generalizability. As a result, the emphasis for this evidence review is on comparative nonrandomized trials of bariatric surgery and nonsurgical therapy or of different types of bariatric surgery procedures.

Swedish Obese Subjects Trial

The Swedish Obese Subjects (SOS) trial is the most influential study of bariatric surgery vs conservative treatment. The SOS trial started in 1987 with a registry containing a detailed questionnaire and clinical data on obese patients with a body mass index (BMI) greater than 34 kg/m2 at 480 primary health care centers in Sweden. From this registry, patients who met eligibility criteria were recruited and offered bariatric surgery. Thus, SOS patients self-selected into treatment, and there were baseline differences between groups, primarily reflecting more excess weight and a higher incidence of comorbidities in the surgery group. A total of 2010 people chose surgery, and 2037 people chose conservative care. Each surgical patient was matched on 18 clinical variables with a patient from the registry who received nonsurgical treatment (usual care). Each surgeon chose the surgical procedure offered. Most procedures were vertical-banded gastroplasty (VBG; >70%), with gastric bypass (6%) and gastric banding (23%) procedures performed as well. Usual care in the SOS trial was the local practice of the primary care center and usually did not include pharmacologic treatment. Patients were followed at regular intervals with repeat questionnaires and physical examinations for at least 10 years.

Many publications from this trial have reported on methods, weight loss, and clinical outcomes.3-7 The following general conclusions can be drawn from the SOS study:

·         Weight loss was greater with bariatric surgery than with conservative treatment. At 10 years of follow-up, weight loss in the surgery group was 16% of total body weight compared with a weight gain of 1.6% in the conservative treatment group.

·         There was significant improvement in glucose control for diabetics and reduced incidence of new cases of diabetes.

·         The effect on other cardiovascular risk factors (eg, hypertension, lipidemia) was also positive, but less marked than that seen for diabetes.

·         Mortality was reduced by 29% after a mean follow-up of 10.9 years.

·         Quality of life improved in the 2- to 10-year follow-up period, with the degree of improvement in quality of life correlating with the amount of weight loss.

Longitudinal Assessment of Bariatric Surgery Consortium

The Longitudinal Assessment of Bariatric Surgery Consortium study is a large prospective, longitudinal, noncomparative study of patients who underwent Roux-en-Y gastric bypass (RYGB) or laparoscopic adjustable gastric banding (LAGB) with follow-up through three years postprocedure.8 The study enrolled 2458 subjects, with median a BMI 45.9 kg/m2 (interquartile range [IQR], 41.7-51.5 kg/m2). For their first bariatric surgical procedure, 1738 participants underwent RYGB, 610 LAGB, and 110 other procedures. At 3-year follow-up, for 1533 Roux-en-Y patients with available data, the percentage of baseline weight lost was 31.5% (IQR, 24.6%-38.4%). For the 439 LAGB patients with available data at 3 years, the percentage of baseline weight loss was 15.9% (IQR, 7.9%-23.0%). At 3 years postsurgery, 67.5% and 28.5% of RYGB and LAGB patients, respectively, had at least partial diabetes remission. Dyslipidemia was in remission in 61.9% and 27.1% of RYGB and LAGB patients, respectively. Subsequent bariatric procedures (revision or reversal) were required in 0.3% (95% confidence interval [CI], 0.1% to 0.9%) of the RYGB patients and in 17.5% (95% CI, 13.8% to 21.9%) of LAGB patients.

Systematic Reviews

Numerous systematic reviews have compared the efficacy of bariatric surgery with conservative therapy or compared different types of bariatric surgery techniques, some of which are older and do not extend across the full range of available studies.9,10

Kang et al (2017) conducted a systematic review with a network meta-analysis that compared the 3 most common types of bariatric surgery techniques: RYGB, sleeve gastrectomy (SG), and LAGB.11 The literature search, conducted through July 2016, identified 11 RCTs for inclusion (8 RYGB vs SG; 2 RYGB vs LAGB; 1 SG vs LAGB). Quality of the trials was assessed using the Jadad score, based on allocation concealment, blinding, intention-to-treat analysis, power calculation, and funding. Most trials had a Jadad score of 3 (scale range, 1-5). A meta-analysis for the outcome of BMI reduction (6 trials) showed that there was no difference between SG and RYGB (0.7; 95% CI, -1.6 to 3.1). A meta-analysis of RYGB and LAGB (2 trials) and a single trial of SG and LAGB showed that LAGB was not as effective as RYGB or SG (5.8; 95% CI, 2.3 to 9.1; and 5.1; 95% CI, 0.9 to 8.9; respectively). Meta-analyses for the outcome of percent excess weight loss (EWL) showed the same pattern, no difference comparing SG and RYGB (5 trials; -4.0; 95% CI, -14.0 to 8.2), and both SG and RYGB more effective than LAGB (2 trials; 22.0; 95% CI, 6.5 to 34.0; 1 trial; 26.0; 95% CI, 6.4 to 41.0; respectively).

Colquitt et al (2014) updated 2003 and 2009 Cochrane reviews of bariatric surgery for obesity.12 They identified 22 randomized trials that compared bariatric surgery with nonsurgical obesity management or that compared different bariatric surgery procedures (total N=1798 participants; sample size range, 15-250 participants). All 7 RCTs comparing surgery with nonsurgical interventions found benefits of surgery on measures of weight change at 1- to 2-year follow-ups. However, reviewers noted that adverse event rates and reoperation rates were poorly reported across trials, and long-term follow-up (beyond 1-2 years) was limited.

Gloy et al (2013) conducted a systematic review and meta-analysis of RCTs comparing current bariatric surgery techniques with nonsurgical treatment for patients with BMI of 30 kg/m2 or more.13 Eleven studies (total N=796 patients) were included. Overall, patients after bariatric surgery lost more body weight than patients after nonsurgical treatment (mean difference [MD], -26 kg; 95% CI, -31 to -21 kg; p<0.001). Remission of type 2 diabetes (T2D) was more likely for bariatric surgery patients than for nonsurgical patients (relative risk of T2D remission, 22.1; 95% CI, 3.2 to 154.3; p<0.000); similarly, remission of metabolic syndrome was more likely for bariatric surgery patients (relative risk, 2.4; 95% CI, 1.6 to 3.6; p<0.001). After bariatric surgery, 21 (8%) of 261 patients required reoperations (5/124 after LAGB, 4/69 after RYGB, 1/49 after SG, 1/19 after biliopancreatic diversion [BPD]). Similar to the Colquitt meta-analysis, no studies reported longer term follow-up (>2 years) and heterogeneity between studies were high.

Chang et al (2014) published a systematic review and meta-analysis of RCTs and observational studies to evaluate the effectiveness and risks of bariatric surgery.14 Reviewers included 164 studies (37 RCTs, 127 observational studies), with a total of 161,756 patients. Mean presurgery BMI was 45.62 kg/m2 and, among the studies that provided information about obesity-related comorbidities, 26% of patients had T2D, 47% had hypertension, 28% had dyslipidemia, 7% had cardiovascular disease, and 25% had obstructive sleep apnea (OSA). Perioperative complications were relatively low, with a perioperative mortality rate in RCTs of 0.08% (95% CI, 0.01% to 0.24%) and in observational studies of 0.22% (95% CI, 0.14% to 0.31%). Complication rates were 17% (95% CI, 11% to 23%) for RCTs and 10% for observational studies (95% CI, 7% to 13%). At 1-year follow-up, mean change in BMI was -13.53 kg/m2 (95% CI, -15.51 to -11.55 kg/m2) in RCTs and -11.79 kg/m2 (95% CI, -13.89 to -9.69 kg/m2) in observational studies. Decreases in BMI were generally sustained over 2 to 4 years of follow-up among studies reporting this outcome.

Many systematic reviews have reported improvements in specific obesity-related comorbidities following bariatric surgery. These reviews have relied primarily on the results of observational studies and included the outcomes of hypertension, T2D, hyperlipidemia, cardiovascular events, quality of life, cancer, knee pain, and liver disease.15-30

Puzziferri et al (2014) conducted a systematic review of studies of bariatric surgery reporting follow-up beyond 2 years, which included 29 studies (total N=7971 patients).31 At follow-up, which ranged from 2 to 5 years postprocedure, the mean sample size‒weighted percentage of EWL was higher for gastric bypass (65.7%) than for gastric banding (45.0%). Reviewers noted that few studies reported sufficient long-term results to minimize bias.

Section Summary: Bariatric Surgery in Adults With Morbid Obesity

There is a lack of large-scale RCTs with long-term follow-up comparing bariatric surgery with nonsurgical treatment for the general population of patients with morbid obesity. Evidence from nonrandomized comparative studies and case series and meta-analyses of existing RCTs, has consistently reported that bariatric surgery results in substantially greater weight loss than nonsurgical therapy. Data from the largest comparative study (the SOS study) has reported that bariatric surgery is associated with improvements in mortality, diabetes, cardiovascular risk factors, and quality of life.

Evidence for Specific Types of Bariatric Surgery Procedures

Gastric Bypass

The body of literature on improved weight loss has been instrumental in establishing gastric bypass as the reference procedure to which other procedures are compared. Practice patterns in the United States have indicated surgeons have adopted this approach, with gastric bypass now comprising most of the bariatric procedures performed.

Comparative trials summarized in the 2003 TEC Assessment32 consistently reported favorable outcomes for open gastric bypass compared with VBG, including 2 RCTs. Some nonrandomized trials that compared open gastric bypass with procedures other than VBG were also summarized in the TEC Assessment. While there were fewer trials for these other procedures, comparisons of open gastric bypass to gastric banding, horizontal gastroplasty, and silastic ring gastroplasty all reported that weight loss was superior with open gastric bypass. Metabolic abnormalities were seen more frequently in gastric bypass patients than in those receiving a VBG. Anemia, iron deficiency, vitamin B12 deficiency, and red blood cell folate-deficiency were commonly seen. Marginal ulcerations were also seen in gastric bypasses, particularly in those whose gastric pouches were too large and included acid-secreting parietal cells.

A 2005 TEC Assessment focused on laparoscopic gastric bypass, which intends to reproduce the open procedure via minimally invasive techniques.33 This technically complex surgery requires a dedicated team and a relatively high degree of skill and experience in laparoscopic technique. This Assessment reviewed 7 comparative trials of the open gastric bypass and laparoscopic gastric bypass, including 3 RCTs. Also, 18 large clinical series of laparoscopic gastric bypass were included. The Assessment concluded that weight loss at 1 year was similar for laparoscopic and open gastric bypass approaches.33 Longer follow-up periods were less well-reported but appeared to be similar for both approaches. While comparisons of complication rates were less certain, some patterns were evident and consistent across the data examined. The profile of adverse events differed between the 2 approaches, with each having advantages and disadvantages. Laparoscopic gastric bypass offered a less invasive procedure associated with decreased hospital stay and earlier return to usual activities. Mortality might be lower with the laparoscopic approach, although both procedures had mortality rates less than 1%. Postoperative wound infections and incisional hernias were also less frequent with laparoscopic gastric bypass. However, anastomotic problems, gastrointestinal tract bleeding, and bowel obstruction appeared to be higher with the laparoscopic approach, though not markedly higher. Given these data, the overall benefit-risk profile for these 2 approaches appeared to be similar.

Yan et al (2016) published a systematic review of RCTs comparing gastric bypass with medical treatment in obese patients (ie, BMI ≥30 kg/m2) who had T2D.34 The primary study outcome was remission of T2D, which was reported in 5 of the 6 studies. A pooled analysis found a significantly higher remission rate after gastric bypass than after medical treatment (odds ratio [OR], 76.37; 95% CI, 20.70 to 271.73; p<0.001). Also, a pooled analysis found a significantly lower final BMI in the gastric bypass group than in the medical treatment group (MD = -6.54 kg/m2; 95% CI, -9.28 to -3.80 kg/m2; p<0.001).

Section Summary: Gastric Bypass

Gastric bypass has been extensively studied. TEC Assessments and other systematic reviews have found that gastric bypass improved health outcomes, including weight loss and remission of T2D. A TEC Assessment also found similar weight loss with open and laparoscopic gastric bypass.

Laparoscopic Adjustable Gastric Banding

A 2006 TEC Assessment updated the evidence on LAGB and compared outcomes with gastric bypass.35 This Assessment concluded that, for patients considering bariatric surgery, there was sufficient evidence to permit an informed choice between gastric bypass and LAGB. An informed patient might reasonably choose open gastric bypass or LAGB as the preferred procedure. Preoperative counseling should include education on the comparative risks and benefits (eg, extent of weight loss and frequency and timing of potential complications) of the 2 procedures to optimize choice based on preferences and shared decision making.

Weight loss outcomes from the studies reviewed in the Assessment confirmed that weight loss at 1 year was lower for LAGB than for open gastric bypass. The percentage of EWL at 1 year was approximately 40%, compared with 60% or higher for open gastric bypass. At time points beyond 1 year, some comparative studies have reported that the difference in weight loss between LAGB and open gastric bypass narrows, but other studies did not. Weight loss outcomes from the 9 single-arm series with the most complete follow-up did not support the hypothesis that the difference in weight loss shrinks after 1 to 2 years of follow-up. It appears more likely from the current data that attrition bias might have accounted for the diminution of the difference in weight loss over time, particularly when patients with bands removed or deflated were excluded from analysis.

These studies also confirmed that short-term (perioperative) complications are very low with LAGB and lower than with open gastric bypass or LAGB. Death was extremely rare, and serious perioperative complications probably occurred at rates less than 1%.

The reported rates of long-term adverse events vary considerably. In the comparative trials, reoperations were reported in approximately 25% of patients, while, in the single-arm studies, the composite rate for reoperations were approximately 50% lower (11.9%). The rates of other long-term complications were also highly variable; eg, the range of rates for band slippage was 1% to 36%, and the range for port access problems was 2% to 20%. These data on long-term complications remain suboptimal. The reporting of long-term complications in these trials was not systematic or consistent. While impossible to determine the precise rates of long-term complications from these data, it is likely that complications have been underreported in many studies due to incomplete follow-up and lack of systematic surveillance. A recent publication by Ibrahim et al (2017) reviewed 25,042 Medicare beneficiaries who underwent LAGB surgery; 18.5% (n=4636) patients underwent one or more reoperation(s). Reoperation was prompted by the need for band removal (41.8%), band and port replacement (28.6%), and other requirements.36 The rates of long-term complications reported in some studies raise concern about the impact of these events on the overall benefit-risk profile for LAGB.

In comparing LAGB with open gastric bypass, there are tradeoffs in terms of risks and benefits. LAGB is a less invasive procedure associated with fewer procedural complications, decreased hospital stay and earlier return to usual activities. However, benefits defined by the amount of weight lost are lower for LAGB. The patterns of long-term complications also differ between the 2 procedures. For LAGB, longer term adverse events related to the presence of a foreign body in the abdomen will occur and will result in reoperations and removal of the band in a minority of patients. Patients who have their bands removed can later be offered an alternative bariatric surgery procedure, such as gastric bypass.

A systematic review by Chakravarty et al (2012)37 comparing LAGB with other bariatric surgery procedures drew conclusions similar to the TEC Assessment. Reviewers included 5 RCTs. The RCTs found that patients using LAGB lost weight, but less weight than with other procedures (eg, gastric bypass or SG). However, the short-term complication rate was lower with LAGB, and no difference was found in           quality of life after LAGB vs other procedures.

Section Summary: Laparoscopic Adjustable Gastric Banding

Systematic reviews of the literature have concluded that LAGB is a reasonable alternative to gastric bypass. There is less weight loss with LAGB; however, the procedure is associated with fewer serious adverse events.

Sleeve Gastrectomy

Systematic Reviews

SG may be performed as a stand-alone procedure or in combination with a malabsorptive procedure, such as the BPD with duodenal switch (BPD-DS). It has also been proposed as the first step in a 2-stage procedure, with gastric bypass or BPD as the second stage.

Osland et al (2017) published a systematic review and meta-analysis of RCTs comparing laparoscopic vertical SG with RYGB (see Table 3).38 The literature search, conducted from 2000 to November 2015, identified 9 RCTs for inclusion (total N=865 patients). Four trials were included in meta-analyses comparing percent EWL between the 2 groups. Results at both 6- and 12-month follow-ups showed that the procedures are comparable (see Table 4).

A systematic review by Juodeikis and Brimas (2017) summarized evidence on long-term results after SG (see Table 3).39 Reviewers included an RCT and 19 retrospective studies, with a total of 2713 patients who received SG. Mean preoperative BMI was 46.9 kg/m2. Mean duration of follow-up ranged from 5 to 11 years, and mean proportion of patients followed for 5 years was 68.5%. Seventeen studies (n=1501 patients) reported 5-year follow-up data. At 5 years, resolution of T2D, arterial hypertension, dyslipidemia, OSA, gastroesophageal reflux disease (GERD), and degenerative joint diseases also improved in most patients (see Table 4). Two studies reported weight loss after 7 and 8 years; percent EWL rates were 56.6% and 54.8%, respectively.

In a meta-analysis of 21 randomized and nonrandomized studies (total N=18,766 patients) comparing SG with laparoscopic RYGB for morbid obesity, Zhang et al (2015) reported no significant difference in percent EWL from 0.5- to 1.5-year follow-ups (see Tables 3 and 4).40 However, after 1.5 years, RYGB was associated with higher percent EWL (2-year MD=5.77; 95% CI, 4.29 to 7.25; p<0.05). Adverse events were more frequent following RYGB (OR for major complication, 1.29; 95% CI, 1.22 to 3.22; p<0.01).

Trastulli et al (2013) conducted a systematic review of 15 RCTs (total N=1191 patients) that compared SG with other bariatric procedures (see Table 3).41 Summary statistics were provided; meta-analyses were not conducted (see Table 4). Reviewers reported mean complication rates with SG of 12.1% (range, 10%-13.2%) compared with 20.9% with LAGB (range, 10%-26.4%). Percent EWL ranged from 49% to 81% with SG and from 62.1% to 94.4% with LAGB.

Brethauer et al (2009) reviewed 36 studies (total N=2570 patients) in a systematic review of SG as a staged and primary procedure, the largest trials coming from European centers (see Table 3).42 Thirteen studies (n=821 patients) reported on high-risk patients having a staged approach and 24 studies (n=1749 patients) on SG as primary procedure. Mean percent EWL, reported in 24 studies (n=1662 patients), was 55.4% overall. Mean postoperative BMI, reported in 26 studies (n=1940 patients), decreased from a baseline of 51.2 to 37.1 kg/m2. Other studies reported weight loss in terms of BMI decrease, the percentage of BMI lost, or percentage of total weight lost; all had significant reductions from baseline. Rates of major postoperative complications ranged from 0% to 23.8% for all studies and from 0% to 15.3% in studies with more than 100 patients. Leaks (2.2%), bleeding episodes requiring reoperation (1.2%), and postoperative strictures requiring endoscopic or surgical intervention (0.6%) were reported in the 33 studies (n=2570 patients). All extracted studies reported mortality data, with 5 deaths within 30 days of surgery (overall mortality rate, 0.19%; 2 in the high-risk/staged group, 3 in the primary procedure group).

Table 3. Systematic Review Characteristics for SG

Study

Dates

Studies

Participants

Design

Duration

Osland et al (2017)38

2000-Nov 2017

9

·   SG=437

·   RYGB=428

RCTs

3 mo to 5 y

Juodeikis et al (2017)39

Through May 2016

20

1626

·   1 RCT

·   19 retrospective

5 to 11 y

Zhang et al (2015)40

Through Oct 2013

21

18,766

·   8 RCTs

·   13 nonrandomized comparative

1 to 5 y

Trastulli et al (2013)41

Through Nov 2012

15

1191

RCTs

6 mo to 3 y

Brethauer et al (2009)42

1996 to 2009

36

2570

·   2 RCTs

·   1 cohort

·   33 case series

3 mo to 5 y

NR: not reported; RCT: randomized controlled trial; RYGB: Roux-en-Y gastric bypass; SG: sleeve gastrectomy.

Table 4. Systematic Review Results for Sleeve Gastrectomy

Study

            Percent EWL (95% CI)           

Comorbidities (95% CI)

Osland et al (2017)38

Mean difference, SG and RYGB:

·   6 mo (3 trials): 0.5 (-5.0 to 6.0)

·   12 mo (2 trials): 7.6 (-0.1 to 15.3)

Not reported

Juodeikis et al (2016)39

Mean rates for SG:

·   5 y (17 trials): 58.4%

·   7 y (2 trials): 56.6%

·   11 y (1 trial): 62.5%

Remission/improvement:

·   Type 2 diabetes: 77.8%

·   Hypertension: 68.0%

·   Dyslipidemia: 65.9%

·   Sleep apnea: 75.8%

Zhang et al (2015)40

Mean difference, RYGB and SG:

·   6 mo (9 studies): 0.2 (-2.5 to 2.9)

·   12 mo (15 studies): 2.9 (-0.2 to 6.0)

·   4 y (3 studies): 2.7 (0.2 to 5.2)

Mean difference resolution, RYGB and SG:

·   Type 2 diabetes (10 studies): 3.3 (2.0 to 5.5)

·   Hypertension (10 studies): 1.3 (0.7 to 2.4)

·   Dyslipidemia (5 studies): 1.1 (0.3 to 1.3)

·   Sleep apnea (7 studies): 1.5 (0.8 to 2.6)

Trastulli et al (2013)41

Mean by procedure:

·   SG: 49% to 81%

·   LGB: 62% to 94%

·   LAGB: 29% to 48%

Type 2 diabetes:

·   SG, 67% to 100%

·   LGB, 80% to 100%

Brethauer et al (2009)42

Mean rate overall for SG:

·   55% (range, 33%-85%)

 

Remission/improvement

·   Type 2 diabetes: >70%

·   Significant reductions also seen in hypertension, hyperlipidemia, and sleep apnea

CI: confidence interval; EWL: excess body weight loss; LAGB: laparoscopic adjustable gastric banding; LGB: laparoscopic gastric bypass; RYGB: Roux-en-Y gastric bypass; SG: sleeve gastrectomy.

Randomized Controlled Trials

An RCT comparing short-term outcomes of laparoscopic SG with gastric bypass was published in 2012.43 Trialists compared 30-day outcomes for 117 patients randomized to gastric bypass with 121 patients randomized to laparoscopic SG. The rate of major complications (no deaths in either group) was 9.4% in the gastric bypass group compared with 5.8% in the LSG group (p=0.29). Minor complications were more common in the gastric bypass group than in the laparoscopic SG group (17.1% vs 7.4%, p=0.02), as were combined major and minor complications (26.5% vs 13.2%, p=0.01).

Karamanakos et al (2008) carried out a double-blind RCT comparing outcomes of laparoscopic RYGB and laparoscopic SG on body weight, appetite, fasting, and postprandial ghrelin and peptide YY (levels at 1, 3, 6, and 12 months after surgery).44 Thirty-two patients were randomized, half to each procedure. The decrease in body weight and BMI were marked and comparable in each group. EWL was greater after LSG than laparoscopic RYGB at 6 months (55.5% vs 50.2%; p=0.04) and 12 months (69.7% vs 60.5%; p=0.05), all respectively. Fasting peptide YY levels increased after both surgical procedures. Appetite decreased in both groups but decreased more after laparoscopic SG.

Himpens et al (2006) reported on a randomized trial comparing LAGB with laparoscopic isolated SG in 80 patients and reported 3-year follow-up.45 Median baseline BMI was 37 kg/m2 (range, 30-47 kg/m2) in the LAGB group and 39 kg/m2 (range, 30-53 kg/m2) in the SG group. Outcomes of weight loss, feeling of hunger, sweet-eating, GERD, complications, and reoperations were recorded at 1- and 3-year follow-ups. Median decrease in BMI in the gastric bypass group was 15.5 kg/m2 (range, 5-39 kg/m2) after 1 year and 18 kg/m2 (range, 0-39 kg/m2) at 3 years after LAGB. One year after SG, decrease in BMI was 25 kg/m2 (range, 0-45 kg/m2) and 27.5 kg/m2 (range, 0-48 kg/m2) after 3 years. Median EWL in the LAGB group was 41.4% after 1 year and 48% at 3 years. Median EWL after SG was 58% and 66% at 1 and 3 years, respectively. More patients having SG than LAGB reported a loss of craving for sweets, but the difference was not statistically significant; GERD appeared de novo in more SG than LAGB patients at 1 year, and the relation reversed at 3 years; between-group differences were not statistically significant at either time point. Two SG patients required reoperation for complications. Seven late complications required reoperation after LAGB, including pouch dilations treated by band removal (n=2) or conversion to RYGB (n=1), 1 gastric erosion treated by conversion to RYGB, and 3 system disconnections that required reconnection. Four patients had reoperations for lack of efficacy (2 LAGB patients underwent conversion to RYGB, 2 SG patients underwent conversion to duodenal switch). The trialists noted that the number of reoperations was significant in both groups and that the severity of complications was greater in the SG group.

Section Summary: Sleeve Gastrectomy

Systematic reviews of RCTs and observational studies, evaluating SG alone and comparing SG with RYGB, have found that SG results in substantial weight loss, comparable to RYGB and that this weight loss is durable for at least 5 years. A meta-analysis found that short-term weight loss was similar after SG or gastric bypass. Long-term weight loss was greater after gastric bypass, but SG is associated with fewer adverse events.

BPD With Duodenal Switch

BPD may be performed with or without the duodenal switch procedure. In the BPD-DS, an SG is performed, preserving the pyloric sphincter. Preservation of the pyloric sphincter is intended to ameliorate the dumping syndrome and to decrease the incidence of ulcers at the duodenoileal anastomosis by providing a more physiologic transfer of stomach contents to the duodenum.

Systematic Review

In an evidence-based review of literature, Farrell et al (2009) summarized data on BPD with or without duodenal switch, RYGB (proximal), and LAGB, and reported that at a mean 1-year follow-up, EWL for BPD with or without duodenal switch (outcomes with and without duodenal switch not reported separately) was 72% (4 studies; n=896 patients), 67% for RYGB (7 studies; n=1627 patients), and 42% for LAGB (11 studies; n=4456 patients).46 At mean follow-up of 5 years, EWL for BPD with or without duodenal switch was 73% (3 studies; n=174 patients), 58% for RYGB (3 studies; n=176 patients), and 55% for LAGB (5 studies; n=640 patients). Reviewers noted that “given the marked paucity of prospectively collected comparative data among the different bariatric operations, it remains impossible to make definitive recommendations for one procedure over another.”

Nonrandomized Comparative Studies

Skogar et al (2017) published results from a retrospective mail survey of patients undergoing BPD-DS (n=113) or RYGB (n=98) (see Table 5).