May 1, 2019
From 1978 to 1991, the National Institutes of Health (NIH) hosted consensus conferences organized for the purpose of summarizing the state of knowledge in the field of metabolic and bariatric surgery (MBS) as determined by an evolving panel of experts who reviewed available published scientific literature and presentations. This article summarizes the three MBS consensus conferences and discusses the development of the 1998 NIH guideline that was the culmination of these conferences. The most recent product of these efforts, a guideline published by the American Heart Association (AHA), the American College of Cardiology (ACC), and The Obesity Society (TOS) in 2013, is also reviewed in this article. Previously, NIH scientists appointed panel members and oversaw the process in releasing guideline statements. However, during the most recent conference in 2013, the guideline process and efforts were transferred to the aforementioned organizations for release of the latest statement. This latter report indicates a departure of the NIH in the organization of MBS guideline statements.
This article also summarizes NIH-supported research and funding of clinical trials examining both medical and surgical weight-loss interventions, highlighting the Look AHEAD (Action for Health in Diabetes) research study and the Longitudinal Assessment of Bariatric Surgery (LABS) studies and their effect on this field. Finally, a description of future research priorities for evaluating the state of the evidence and defining priorities for MBS research is offered.
The next frontier in MBS research will be providing research support to strengthen personalized care among populations with obesity and enhancing the capability to predict which patient populations may experience the greatest benefit from MBS as compared with nonsurgical interventions. Continued NIH support is vital to the ongoing development of safe and effective interventions that address obesity and its comorbidities.
Summary reports of the proceedings and recommendations of the expert panel were published as NIH consensus conference statements. This process played a major role in the early development of metabolic/bariatric abdominal surgical procedures as treatment for severe obesity and related metabolic diseases. The first conference devoted to this issue took place in 1978. The evidence base was a series of cases on the surgical treatment of obesity with intestinal (jejunoileal) bypass operations. The report noted the treatment was effective in reducing weight but had a number of undesirable complications.1
A second NIH Consensus Conference on MBS convened in 1985 and focused on the health implications of obesity. The third and most recent NIH Consensus Conference took place in 1991. A panel of experts reviewed published literature as well as oral presentations and responded to questions from the audience in order to construct a consensus statement.2 By the third conference, vertical banded gastroplasty and Roux-en-Y gastric bypass (RYGB) replaced jejunoileal bypass as the preferred surgical treatment. The panel made the following recommendations:2
The 1991 consensus conference and its subsequent statement were seminal events in the development and acceptance of bariatric surgery as an appropriate treatment for severe obesity and its related diseases. These basic criteria for selection of patients have persisted. Notably, the studies that led to these recommendations did not include the laparoscopic approach to the procedures, which is known to decrease the incidence of complications; the implementation of national accreditation; or results of high-level evidence describing the procedures’ effects on specific treatment groups. In 2013, the NIH formally retired the Consensus Development Program and concluded the organization of consensus conferences of any type. As a result, the 1991 Conference Statement has not been updated by the NIH to include consideration of MBS in patients with less severe obesity (BMI 30–35 kg/m2) with associated comorbid conditions, particularly type 2 diabetes mellitus (T2DM).
Although the NIH consensus conferences produced statements based on the recommendations of the expert panels, they were not official policy statements of the NIH; however, the consensus conference statements did lead to the development and publication of NIH guidelines, which are official NIH documents. In 1998, a panel of experts in obesity and health policy examined emerging criteria for construction of evidence-based guidelines. The panelists, none of them surgeons, recognized a preference for clinical evidence based on randomized control trials (RCTs).3-7 One RCT comparing gastric bypass to nonoperative/medical controls was identified,8 and several other RCTs regarding specific aspects of conduct of the operations and perioperative care were identified. In addition, observational data were considered, including the Swedish Obese Subjects (SOS) study, a trial that consisted of observational data from surgical patients as well as matched patients treated with usual care.9 The panel found no basis to alter the conclusions of the 1991 consensus panel and issued guidelines that mirrored recommendations from the 1991 consensus panel.10
In 2007, the NIH appointed a new expert panel, including one surgeon, to update the 1998 guidelines. Criteria for selection of research papers to comprise the evidence base were refined to include the requirement of 80 percent retention. Because few evidence-based RCTs had been completed, reports from the SOS trial as well as the LABS study were included among the papers comprising the evidence base.11 Following a prolonged, five-year process, the panel issued NIH’s Systematic Evidence Review from the Obesity Expert Panel, 2013 and referred the review and publication of any additional guidelines to the AHA, the ACC, and TOS.12 Hence, the NIH has no official medical or surgical position, consensus statement, or guideline regarding the treatment of obesity at present.
This guideline includes the strongest evidence-based recommendation of support for the surgical treatment of obesity among several guidelines in stating that physicians should “be proactive in identifying patients who would benefit” when “referring them to a surgeon.”13 Specifically, the 2013 AHA/ACC/TOS guideline states that patients with a BMI >40 kg/m2 or BMI >35 kg/m2 with an obesity-related comorbid condition who have failed behavioral and dietary modification with or without pharmacotherapy may be appropriate MBS candidates, and physicians should offer referral to an experienced bariatric surgeon for consultation and evaluation.14 In 2013, the evidence was judged to be insufficient to either endorse or discourage surgical intervention for patients with BMI <35 kg/m2, because the evidence base at the time of the systematic literature review omitted the multiple RCTs that addressed metabolic/surgical intervention for patients with T2DM and BMI 30–35 kg/m2. There has been no indication to date that the AHA, ACC, or TOS intend to update these guidelines.
The NIH conducts and supports investigators in the conduct of basic science, physiology, and treatment of obesity and related diseases. These investigations have made significant contributions to the health care professions’ understanding of obesity and the effects of treatment (see Table 1). Study of patients who underwent bariatric surgical procedures contributed to greater identification and understanding of the biology of obesity and its relation to gastrointestinal tract structure and function.
Table 1. NIH-funded basic research on the biology of obesity
The NIH has funded clinical trials of both medical and surgical weight-loss interventions for more than 50 years. A large number of these clinical trials have examined aspects of obesity and the response to interventions with a range of outcomes. Two examples of these trials are the Look AHEAD research study and the LABS consortium.
The Look AHEAD trial tested the hypothesis that intense lifestyle intervention (ILI) to accomplish weight loss among adults with obesity and T2DM, in comparison with usual care, would reduce all-cause mortality.15 A total of 5,145 patients were randomized at 16 clinical research sites to either the ILI or usual-care groups. The toolbox made available to investigators carrying out the ILI included dietary and physical activity instruction, decreased caloric intake, and weekly structured visits, among other interventions. Weight loss for the intervention group at one year was 8.5 percent total weight loss (TWL). A weight loss of 4.7 percent TWL persisted at the four- and eight-year intervals. Weight loss was highly variable, with 26.9 percent of patients achieving and maintaining a 10 percent weight loss over eight years.16 Definite and persistent improvement of markers or mediators of cardiovascular disease were demonstrated in addition to weight loss. However, observed mortality was similar between the two groups.17
The National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) established a multicenter research consortium known as LABS, which used an observational cohort design to address multiple clinical, epidemiological, and behavioral hypotheses.11 These studies comprise four trial groups: LABS-1, LABS-2, LABS-3, and Teen-LABS.
LABS-1 focused on the evaluation of short-term safety of MBS and related issues, which was the first priority of bariatric surgical clinical research. A total of 4,776 subjects were recruited with a 30-day retention rate of 100 percent. The all-cause 30-day mortality was 0.3 percent with a 4.1 percent major complication rate, 25 percent of which were laparoscopic adjustable gastric banding (LAGB). RYGB was performed in 71 percent of the subjects.18 Patient factors predictive of a serious complication were a history of deep vein thrombosis or pulmonary embolus, obstructive sleep apnea, or impaired functional status. Extreme BMI values also were associated with increased risk, though age, sex, race/ethnicity, and comorbid conditions were not. Reoperations were found to have similar perioperative mortality but with a fourfold increase in serious complications.19 Low-surgeon volume also was found to be an important predictor of adverse outcomes.20
LABS-2 comprised a portion of the cohort studied in LABS-1 (2,458 adults) and was a longer-term observational study designed to further examine the safety and efficacy of MBS. The following detailed information regarding multiple outcomes and domains was collected: weight loss and body composition, T2DM and insulin resistance, cardiovascular disease, pulmonary disease, renal disease, liver function, behavior/psychological factors, musculoskeletal and functional status, and gender and economic impact.11 Both validated instruments as well as instruments created by LABS were used for self-reported data collection (see Table 2).
Table 2. LABS patient-reported outcomes
The use of self-reported weights within this clinical trial was validated.21,22 Variable data were available on 92 percent of study participants, including body weight in 83 percent. Multiple peer-reviewed publications have been developed reporting outcomes of many of the domains (61 as of press time) that the LABS consortium has studied. Some key findings to date are as follows:
LABS-3 is a detailed study of a subset of the LABS-2 participants with or without T2DM who underwent frequent testing for intravenous glucose tolerance and meal-stimulated gut hormone response. At both six and 24 months following surgery, substantial improvement of the disposition index (DI) in both those participants with and without T2DM was demonstrated. Although the DI improved among the participants with T2DM, it remained in the fifth percentile of normal, providing an explanation for the recurrence of diabetes that has been reported by LABS and other investigators. LABS-3 is presently ongoing with eight–nine years of follow-up for all participants, and these long-term outcomes are under analysis.
Recruitment for Teen-LABS, a study of adolescents with obesity, was completed in 2007 and the trial is active with funding through 2021. The study design is a prospective observational cohort design with more than 200 participants. Preliminary publications have addressed perioperative outcomes, cardiovascular risk factors, quality of life, candidate characteristics, and safety, among others.28-32 Outcomes that are being reviewed involve psychosocial status and cognitive function, micronutrient deficiencies, and risk-taking behaviors, as well as continued data collection on pregnancies, additional abdominal procedures, and mortalities within the study population. The standardization of definitions, validated metrics, and shared data collection protocols have ensured the study team’s ability to produce reliable and accurate data for evidence-based recommendations.
The NIDDK funded several ancillary LABS studies, using the LABS database and bio-samples to explore additional domains. A subset of LABS participants underwent detailed studies of cognitive function both before surgery and at various time points, including two years following surgery. A modest but clinically important improvement in the impairment of cognition associated with severe obesity was demonstrated.33 Further analyses of the LABS outcomes are pending upon completion of the genomic and metabolomic analyses of LABS biospecimens. Other LABS ancillary studies include detailed assessment of physical activity/energy expenditure, an in-depth assessment of eating behavior and food choice, and analysis of gonadal steroids. The NIDDK terminated funding of the LABS Consortium after seven years of follow-up. The LABS database and limited biospecimens (serum, plasma, and deoxyribonucleic acid) are available from the NIDDK repository. Multiple analyses of the extensive LABS database and bioassays are in progress.
Additional clinical trials funded wholly or in part by the NIDDK include the SOS study. Though conducted in Sweden and primarily funded by Swedish sources, NIH support is listed as contributing to a recent analysis of the effect of nutrient intake on weight loss and other outcomes.34 Multiple RCTs of subjects with variable severity of obesity and T2DM have been or are being conducted to compare surgical intervention with nonoperative medical intervention.25,35 These and other trials—funded by industry, foundations, and the institutional sites—have consistently shown that surgery leads to substantially greater remission of T2DM as well as improved control of glycemia. For example, the STAMPEDE (Systemic Therapy for Advanced or Metastatic Prostate cancer: Evaluation of Drug Efficacy) trial has reported five-year results with persistent improvement in T2DM among RYGB and sleeve gastrectomy patients.35 Bond and colleagues have reported on physical activity before and after MBS. As with the LABS data, a marked impact on weight loss resulting from changes in physical activity remains unconfirmed.36,37 NIH-funded basic science research studies relative to obesity and its treatment continue to be numerous. One example is the Small Animal Metabolic Surgery Resource Core, from which, to date, 14 publications have been developed.38
In 1975, the National Heart, Lung, and Blood Institute funded the first trial using metabolic surgery as the intervention modality. The Program on the Surgical Control of the Hyperlipidemias was not intended to focus on bariatric surgery. Nonetheless, it provided the first statistically significant determination that the marked cholesterol lowering achieved by the partial ileal bypass operation reduced the dual endpoint of recurrent myocardial infarction or atherosclerotic death, the incidence of peripheral vascular disease, and the need for coronary bypass or stenting, as well as increased life expectancy. The clinical findings were accompanied by serial arteriographic evidence of decreased atherosclerotic progression and actual plaque regression.39
The NIH/NIDDK has been a valuable partner in developing our understanding of obesity, related diseases, and treatment outcomes. Together with the clinical consensus and guidelines process, the NIH has played a crucial role in evaluating the evidence and in defining priorities for ongoing MBS research. In the future, the NIH may provide research support in investigating and strengthening personalized care, as most MBS studies to date have focused on reporting large, homogeneous populations. The ability to predict more accurately which patients will achieve specific benefits from MBS would greatly focus these interventions. In addition, the capacity to predict which patients will benefit most from MBS relative to the flexible endoscopic and nonsurgical interventions in development will reduce risk and costs to patients. Continued NIH support into researching the etiology and mechanisms of obesity and diabetes is vital to ongoing development of safe and effective MBS, pharmacotherapy and lifestyle, and novel synergies of therapy.
This work was supported by the American College of Surgeons (ACS). The authors declare that they have no relevant conflict of interest.
We are grateful to the ACS for their generous sponsorship of the Metabolic Surgery Symposium and associated journal publication development. We thank Jane N. Buchwald, Chief Scientific Research Writer, Medwrite Medical Communications, Maiden Rock, WI, for manuscript editing and publication coordination. And we thank Patrick Beebe and Donna Coulombe, ACS Executive Services, for their expert organization of the Metabolic Surgery Symposium.
References