January 4, 2019
Editor’s note: The Bulletin is publishing the collected papers from the Metabolic Surgery Symposium, which took place in August 2017 at the American College of Surgeons (ACS) headquarters, Chicago, IL. Following are the first two articles in the series: “Definition and history of metabolic and bariatric surgery” and “Surgery and the Gordian knot of metabolic syndrome.” Be sure to read the February issue for articles on advances in metabolic and bariatric surgery and insulin as an outmoded therapy for type 2 diabetes.
In 333 BC, Alexander the Great marched his army into Gordium, where he found an ancient wagon tied to a post with a knot so tightly entangled that no one could unfasten it. The wagon had once belonged to Gordias, the father of King Midas, and an oracle predicted that the man who could unravel the knot would conquer Asia. Without attempting to untangle the ropes, Alexander freed the cart with one stroke of his sword and conquered much of Asia before his death at age 32.
The comparison between the Gordian knot and metabolic syndrome is apt. For centuries, physicians have pursued the release of the knot of the metabolic syndrome by trying to unravel the complexities of such intertwined diseases as obesity, diabetes, hypertension, dyslipidemias, nonalcoholic steatohepatitis, polycystic ovary syndrome, and even cancer (see Figure 1).1-4 Today, all of these diseases can be prevented or reversed with a single operation. This article provides an overview of the multiple expressions of the metabolic syndrome and offers background on the multiple mechanisms of metabolic and bariatric surgery that decrease these health care conditions.
Figure 1. Effects of metabolic and bariatric surgery
While humanity has, at times, been able to overcome famine and plague, obesity and its comorbidities have become a pandemic, with an estimated 3.4 million deaths in 2010.5 In 2012, approximately 56 million people died throughout the world; 620,000 died as a result of violence (120,000 people died from war-related deaths, and another 500,000 others died from crime-related activity). Additionally, 800,000 individuals committed suicide, and 1.5 million died of diabetes.6
The disease of obesity is understood to be a pathologic process resulting from excess body fat. Its cause is the interaction of inherited genetics with the environment mediated by epigenetic changes in those genes or their expressed proteins.7 Studies show that 67 percent of variability in body mass index (BMI) is attributable to inheritance. Of the total variability, 40 percent is due to genes that control food intake, 12 percent is due to metabolic rate, 5 percent to fat oxidation, and 10 percent to spontaneous physical activity.8
The trend toward increasing weight began around the turn of the 20th century, but the curve markedly shifted between 1976 to 1980, doubling in prevalence from 1999 to 2009; it is now on the way to doubling again by 2019, according to the U.S. Centers for Disease Control and Prevention. Although genes could not have changed that quickly, epigenes can. The growing prevalence of obesity from 1960 to 2014 in women ages 20–74 from 17 percent to 40.4 percent and in men from 11 percent to 35 percent led to the assumption that the increase was a product of a more abundant food environment coupled with more sedentary work environments, as well as a decline in sleep.9 This explanation, while straightforward, was misinterpreted based on the erroneous belief that people were wholly in charge of their own weight and adiposity. A failure of personal responsibility was believed to be the reason people were bigger.
This viewpoint began to change as our understanding of the biology of obesity evolved, and it became apparent that epigenetic changes influenced by the environment could modify the behavior of genes, and those modifications would be passed down to the next generation. Epigenetic gene modulation is a dynamic response to the environment that allows gene modulation to occur continuously, often within minutes. For example, in an experiment with drosophila melanogaster (fruit flies), scientists fed male fruit flies sugar water, silencing the Suvar gene responsible for the metabolism of sugar, resulting in offspring that died when fed sugar water. The offspring inherited a silenced copy of the Suvar gene, demonstrating the contribution of paternal inheritance.10 It is not hard to imagine a young person inheriting a silenced gene and becoming unable to metabolize sugar properly; when drinking high-sugar beverages, he or she may develop early clinically severe obesity at a young age. Epigenetic modulation occurs dynamically and continuously throughout the life of an individual and is influenced by the environment, exercise habits, food, medication, sleep, and stress.
Mathematical modeling is a science that helps to define specific factors that can lead to the amplification of an epidemic. In a recent report that models obesity’s prevalence, obesity continuously increases, reaching 41.03 percent (95 percent confidence intervals CI: 39.28–44.31) in the U.S., and 26.77 percent (95 percent CI: 25.62–28.06) in the U.K., with a stable state of 52.77 percent and 26.84 percent, respectively. If an individual is genetically predisposed, hypothetically, to a “homozygous obesity gene,” he is more susceptible to “socially contagious” weight gain. Socially contagious risk is modifiable and has the biggest impact on prevalence. Several assumptions were made to simplify the model, for instance, eliminating the paternal contribution to genetic inheritance; had it been added, it may have made the predictions even higher.11
Most population-based theories of the evolution of obesity and its reduction center on lifestyle modification strategies, typically a combination of reduced caloric intake, increased physical activity, and behavioral modifications to combat the “toxic” food environment found in most cultures around the world. Depending on their intensity and duration, these approaches can result in weight loss of 5 percent to 10 percent body weight, or more, as well as clinically significant improvements in weight-related comorbidities. Unfortunately, most people are challenged to maintain their weight loss over time and often require another lifestyle modification approach, pharmacotherapy, or bariatric surgery to maintain weight loss. The iterative cascade of these attempts has unknown consequences physiologically. Patients who are unsuccessful may eventually become metabolically inflexible.
Metabolic adaptation is a response to weight loss that is observed to occur in patients with voluntary food restriction. Patients experience weight loss followed by ravenous hunger and then weight regain, often to a new and higher metabolic set point. Recent data from a six-year study of contestants on the television program The Biggest Loser show that when a person loses weight, his or her basal metabolic rate (BMR) will decrease as fat mass decreases but fails to respond to rising weight and, in fact, remains low or goes lower. In this study, the patient developed a lower BMR by 610 kcal per day at 12 weeks. The BMR continued to decrease to 704 kcal per day less than baseline at six-year follow-up.12
When weight regain occurs, it is driven by hunger to eat at three times the baseline amount of food, almost 100 kcal per day for each kg of weight lost. In the face of the marked decline in metabolic rate, it is the normal response of an individual to regain more weight than they lost.13 In stark contrast, patients undergoing Roux-en-Y gastric bypass (RYGB) show a small reduction in their BMR at six months after surgery, but BMR eventually rebounds to its original level at the time of surgery. In the face of a decreasing body weight, the BMR is relatively preserved.14 This finding was corroborated in the ongoing Longitudinal Assessment of Bariatric Surgery (LABS) study, where a decrease in daily energy expenditure was not significantly changed at 24 months—although the decrease and rebound followed a different trajectory from that of the Biggest Loser participants, perhaps because of their high rate of physical activity.15
The brain has taken center stage in determining the pathogenesis of obesity. To understand the effects of metabolic surgery, we need to understand relevant neuroanatomy and signaling to and from the brain to the rest of the body.
The midbrain, specifically the hypothalamus, is cited as the center for input and output, regulating eating behavior and energy metabolism, thus establishing a homeostatic loop. The arcuate nucleus (also known as ARC) of the hypothalamus contains first-order neurons, pro-opiomelanocortin (POMC) that recognizes anorexigenic sympathetic input, and Agouti-related protein (AGRP) and neuropeptide Y that recognize orexigenic, parasympathetic input. The input is integrated in a nearby paraventricular nucleus melanocortin-4 receptor (MC4R) pathway to influence, through downstream signaling, the metabolism of ingested food. Gut hormones, like ghrelin, act on the AGRP neurons to create hunger, increasing food intake and decreasing energy expenditure, whereas the sympathetic system (POMC) receives input from leptin and insulin to decrease food intake and increase energy expenditure. The incretin (GLP-1) also plays a role in both glucose metabolism and weight regulation. GLP-1 spans the gut-brain connection and is present in both the hypothalamus and distal intestine. Although this system represents the core defense of body weight regulation, it appears that other areas of the brain that integrate exercise, stress, exposure to light (circadian rhythms), influence of medications, and other factors also are integral to the overall homeostatic loop.16
Neuronal circuitry has evolved to defend body weight. Important genetic changes in brain-related proteins influence eating and satiation behavior. Food reward pathways override the intake of nutrients to provide for energy needs. Reward pathways influenced by the culture of food and stress mitigation experienced during childhood and adolescence hardwire eating preferences, which then become reinforced by the changes in the microbiome and gut-brain response to meals. Hedonic signals driving behavior derive from the limbic system connected to the nucleus accumbens in the midbrain.
The evidence for defended body weight comes from a mouse model study in which rodents of differing levels of adiposity (mimicking the patient population) underwent RYGB. The rodents were observed for body composition and weight as well as food intake. One group was fed a high-fat diet and accumulated weight and fat mass, followed by caloric restriction and RYGB. Instead of losing weight, the mice gained weight up to the level prior to calorie restriction. In larger studies, mice who weighed more before surgery had a greater loss of weight and body fat than those mice with lower levels of obesity. In addition, the smallest animals actually gained weight. Lean mass was completely conserved by RYGB. In overfeeding experiments, sham-operated animals increased the amount they ate at every meal, whereas RYGB mice ate slower, more frequent meals. At 20 weeks after surgery, these subjects still were eating 35 percent smaller meals; however, the total caloric intake was similar.17 These RYGB data mirror the findings in mice after sleeve gastrectomy (SG).18
Additional data show that individuals who became pregnant or were subjected to food restriction were capable of increasing food intake after surgery to meet their new metabolic demands. They reported feeling less hungry after surgery, and hunger is a major driver of weight regain in voluntary food restriction from dieting. Appetite-regulating gut hormones differ between subjects losing weight by surgery and caloric restriction.19
In humans, an analysis of 16 participants in the LABS study revealed resting metabolic rates (RMR) and total daily energy expenditure (TDEE) of patients at six and 24 months after RYGB. Each participant was admitted for 18 days before surgery and at six and 24 months after surgery, during which the environment was carefully controlled. TDEE and RMR were studied with doubly weighted water using the average of 25-minute periods of indirect calorimetry on days 16 through 18. Body composition, fat-free mass, and fat mass were measured with a dual energy X-ray absorptiometry scan. The sample included patients with RYGB (n = 22), adjustable gastric band (AGB) (n = 2), and biliopancreatic diversion with duodenal switch BPD/DS (n = 1); it demonstrated an average decrease of 379 kcal per day.15
Metabolic surgery exerts weight-dependent (chronic) and weight-independent (acute) effects. These actions are not separate, but rather complementary and synergistic (see Figure 2).20 A primary effect of metabolic bariatric surgery can be summarized as establishing a new and lower body weight set point. The reprogramming that occurs after surgery is best demonstrated by the evidence that humans and rodents return to a new lower set point after periods of over- or under-feeding. In addition, humans and rodents can increase food intake as needed despite surgery, for instance, during pregnancy. The response to the same surgical intervention can be variable and may be based on genetic, epigenetic, environmental, and social factors that limit individual outcomes following intervention.21 This variability is, in fact, one of the strongest arguments against the concept of restriction as a primary mechanism.
Figure 2. Gut-brain adjustments following bariatric surgery
Changes in hormones from adipose tissue (leptin) and the gut, including ghrelin and GLP-1, have been measured and are substantially changed after RYGB or SG. The discovery of leptin, a hormone derived from fat, and the role it plays in feedback and regulation of body weight was a pivotal discovery in defining the relationship between the body and brain in the regulation of weight. The failure of humans with obesity to respond to leptin administration, and the additional understanding of leptin resistance and the breakdown of the regulatory loop to prevent humans from overweight, also were foundational steps in our understanding.
Metabolic procedures produce a decrease in fat mass and concomitant decrease in leptin. Normal leptin signaling is necessary in mice to realize the full effect of RYGB on body weight. Increased leptin sensitivity does not account for the effect. This finding is corroborated in humans and rodents after SG.22 Because leptin signaling is necessary but insufficient for the weight-loss effect of surgery, another molecule may be responsible. A leading contender is fibroblast growth factor 21. MC4R signaling studies have yielded variable results, and their designs have been controversial; however, MC4R signaling within the hypothalamus does not appear to be the critical mechanism of action of RYGB, although more study is needed. MC4R proteins exist on preganglionic neurons and are required for the effect of RYGB.23 Except for a few patients, all participants in studies with MC4R variants achieved the full effect of RYGB.24
The intestine also plays a major role in disease. The intestine is responsible for approximately 25 percent of glucose production in a fasting state. When anatomy is changed through surgery, these processes also are altered and may account for some of the alterations of disease states, like improvement in glucose homeostasis after RYGB.25
The human gut contains roughly 100 trillion bacteria, viruses, and archaea. The microbiome has emerged as a major curator of the signaling from food to the brain and to physiological mediators of overall health. Alternation of fat and sugar in the diet can manipulate the various phyla of microbiota and produce positive or negative health effects.26 For example, in the obese state, bacterial components leak through the gut wall and mediate inflammation implicated in the transition of nonalcoholic liver disease to nonalcoholic steatohepatitis, fibrosis, and type 2 diabetes mellitus (T2DM).27 Alterations in the microbiome are conserved in humans, mice, and rats after RYGB. There is a shift to increase the Firmicutes phylum, specifically Gammaproteobacteria (Escherichia) and Verrucomicrobia (Akkermansia). The changes that occur to the microbiome are early (beginning within one week) and persistent.28 Many preoperative manipulations of diet involved in implementation of enhanced postoperative recovery protocols may positively or negatively affect outcomes from surgery. Recent evidence indicates that manipulation of the microbiome by life events may have a major impact on surgical complications.29
Finally, changes in anatomy affect production of gastric acid and the amount of absorptive surface of the stomach. Both of these alterations may lead to changes in the gut microbiome.30 A patient’s absorption of medications may change, which will affect the primary action of those medications and their effect on the microbiome.31 The anatomic changes of surgical procedures also alter micronutrient absorption and composition in different parts of the gastrointestinal tract.32
Bile acids (BA) are important primarily in the postprandial physiology of eating behavior and energy expenditure. Circulating BA levels may increase more than three times following a meal. Bile acids act through the FX receptor pathway to influence the thermal effect of eating and increase energy expenditure in skeletal muscle and brown fat.33 In comparison with lean individuals, patients with obesity have a postprandial decrease in glycine-conjugated BA.34 In a 40-week follow-up comparison of lean individuals and obese patients who had undergone RYGB, food composition and intake of the surgery patients was roughly equal to their preoperative levels. Total body weight loss was 5.3 percent at one week, 10.6 percent at four weeks, and 33.7 percent at 40 weeks. Fasting circulating BA were not different between lean patients and obese, except the taurine conjugated subset was less in individuals with obesity. The postprandial concentration of BA showed a blunted response (52.4 percent lower) after a meal in individuals with obesity. After the meal, the glycine conjugated subset was more affected. After RYGB, the postprandial blunting was normalized, a change not detected at one or four weeks, but evident at 40 weeks. Because the effect was not observed early after surgery, adaptive changes in the gut might be assumed. Notably, RYGB alters the microbiome that in part regulates bile acid conjugation.35
The human gut is innervated by both sympathetic and parasympathetic nerves, and incorporates chemosensory and stretch receptors in all layers of the gut wall and all parts of the alimentary system. The enteric “brain” has evolved over millions of years to become a highly sophisticated feedback system to the brain. During a meal, the hindgut, specifically the area postrema (AP), is bombarded with constant input detailing the composition and quantity of food ingested. This area integrates hormonal signals from leptin, amylin, CCK, GLP-1, and PYY, among others, that can signal when to stop eating.
The vagus nerve is a key part of the communication pathway between the neural system of the gut and the brain. In rats, disruption of the hepatic branch showed no effect on RYGB; however, disruption of the celiac branches attenuated the effect of RYGB by 20 percent and disruption of the paraesophageal vagus showed a similar effect.36 In some RYGB techniques, the vagus is divided. In SG, both the anterior and posterior trunk are preserved. Study outcomes in human RYGB patients after the vagus has been divided are varied, mirroring the difficulty of establishing effect at the macroscopic level. Although division of the vagus fails to show a short-term weight disadvantage, it may be associated with increased complications as well as changes in sweet taste-receptor stimulation.37,38 Finally, specific nutrients in the diet can influence inflammation of the vagus and brain and interfere with normal regulatory balance.39
Metabolic surgery provides one example in the history of medicine wherein there is profound evidence of this therapy’s efficacy. For example, in a long-term study (≥12 years), 51 percent of participants with T2DM achieved prolonged remission.3 Other examples in medicine of similar minimally understood but effective therapy include the historic use of aspirin-containing birch bark by Native Americans to diminish pain, use of digitalis for heart failure, and use of metformin (from goat’s rue) since the Middle Ages to treat T2DM. The key question remains: Why is bariatric surgery so effective a treatment for weight, diabetes, and other diseases?
Surgical therapy is the most successful treatment option for obesity and a spectrum of related diseases, such as T2DM. The success of metabolic bariatric surgery represents the remarkable achievement of surgeons who have worked to improve its safety and outcomes. Obesity is not a simple matter of calories in and calories out, but rather a pathologic complex disease of neurohormonal and microbiome dysregulation in humans who are genetically predisposed to obesity. As metabolic surgical research moves beyond the original paradigm of mechanical restriction and malabsorption, greater understanding of the master switch of metabolic regulation will be attained, enabling us to provide more effective, personalized, and specific therapy for patients with obesity and associated comorbid disease.
This work was supported by the ACS. The authors declare that they have no relevant conflict of interest.
We are grateful to the ACS for its generous sponsorship of the Metabolic Surgery Symposium and associated journal publication development. We thank Jane N. Buchwald, chief scientific research writer, Medwrite Medical Communications, 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.