13.4.5 Metabolic surgery in the treatment of type 2 diabetes mellitus

Faced with the dual pandemics of obesity and type 2 diabetes mellitus, heath care providers require a broad array of treatment options. Diet, exercise, and medications remain the cornerstones of type 2 diabetes therapy, but long-term results with lifestyle modifications can be disappointing, and, despite an ever-increasing armamentarium of pharmacotherapeutics, adequate glycaemic control often remains elusive. Moreover, most diabetes medications promote weight gain, and using them to achieve tight glycaemic control introduces a proportionate risk of hypoglycaemia.

In cases where behavioural/pharmacological strategies prove insufficient, gastrointestinal surgery offers powerful alternatives for obesity and type 2 diabetes treatment (Fig. 13.4.5.1). Among severely obese patients, bariatric operations cause profound, sustained weight loss, ameliorating obesity-related comorbidities and reducing long-term mortality (1–4). Operations involving intestinal bypasses exert particularly dramatic antidiabetes effects. For example, approximately 84% of obese patients with type 2 diabetes experience diabetes remission after a Roux-en-Y gastric bypass (RYGB), maintaining euglycaemia off diabetes medications for at least 14 years (1, 5–8). Mounting evidence indicates that these effects result not only from weight loss, but also from weight-independent antidiabetic mechanisms (9).

Whereas diabetes is traditionally viewed as a relentless disease in which delay of end-organ complications is the major treatment goal, gastrointestinal surgery offers a novel endpoint: complete disease remission. Consequently, conventional bariatric procedures and experimental gastrointestinal manipulations are being used worldwide to treat type 2 diabetes in association with obesity, and, increasingly, among less obese or merely overweight patients (8). Gastrointestinal surgery also offers valuable research opportunities to improve knowledge of diabetes pathogenesis and help develop less invasive procedures and novel pharmaceuticals.

This chapter discusses the effects of gastrointestinal operations on type 2 diabetes, and focuses on potential antidiabetic mechanisms that mediate those effects.

Tradition asserts that bariatric operations promote weight loss through gastric restriction, intestinal malabsorption, or both (1, 2) (Fig. 13.4.5.1). Restrictive procedures, such as adjustable gastric banding (AGB), reduce gastric capacity and retard emptying. Malabsorptive procedures, such as biliopancreatic diversion (BPD), leave the stomach largely intact, but divert food from the stomach to the ileum, compromising nutrient absorption. The typical proximal RYGB combines gastric restriction with nutrient bypass of most of the stomach and a short segment of proximal intestine; the remaining small bowel is sufficient to prevent clinically significant malabsorption. This operation causes weight loss through gastric restriction and additional endocrine/metabolic mechanisms (see below). Of bariatric procedures, RYGB and BPD produce the highest and most rapid rates of type 2 diabetes remission (7).

Although bariatric surgery was designed to facilitate weight loss, anecdotes of rapid type 2 diabetes remission after RYGB emerged more than 30 years ago. In 1995, Pories et al. published results from a study of 608 obese patients who underwent gastric bypass, with 93% follow-up over 14 years (5). Among those with type 2 diabetes, 83% experienced complete, durable remission. Several other large studies also found rates of 80% or higher for post-RYGB diabetes remission—defined as normal blood glucose and haemoglobin A_1c (HbA_1c) values without diabetes medications—and a meta-analysis of 136 bariatric surgery studies including 22 094 individuals confirmed an 84% remission rate (1). In a subsequent meta-analysis of 621 studies involving 135 246 patients, diabetes remission was observed in 80% of patients following RYGB (7). The Swedish Obese Subjects study, a prospective, contemporaneously matched, multicentre trial of bariatric surgery versus medical care for severe obesity, reported outcomes after 2 years, in 4047 patients, and 10 years, in 1703 patients (2). At 2 years, no postsurgical patients had developed type 2 diabetes, whereas 5% of the patients in receipt of medical care had. This protection persisted to 10 years, and development of type 2 diabetes was more than 3 times lower for surgical patients, while type 2 diabetes remission was more than 3 times more common. In these studies, the few RYGB patients who remained diabetic after surgery had longer disease duration, suggesting that their β cell mass was irreparably compromised (6). Nevertheless, almost all of these patients experienced improved glycaemic control and reduced medication dependence.

Rates of surgical morbidity and mortality after commonly performed bariatric operations, e.g. the RYGB and the laparoscopic adjustable gastric band, are low and decreasing with developments in laparoscopic techniques, centres of excellence, quality-of-care control systems, and multidisciplinary approaches (8, 10). Operative mortality from bariatric operations is now approximately 0.3%—lower than that following cholecystectomy.

At least in patients obese enough to qualify for bariatric operations by existing standards (11), the benefits of surgery outweigh the risks, and numerous reports show decreased overall long-term mortality after bariatric surgery, including a remarkable 92% decrease in diabetes-related deaths after RYGB (3, 4).

Currently, bariatric surgery in the USA is National Institutes of Health (NIH)-approved for individuals with a body mass index (BMI) of more than 40 kg/m², or more than 35 kg/m² with comorbidities, such as type 2 diabetes (11). The UK’s National Institutes of Clinical Excellence (NICE) makes similar recommendations. However, the remarkable impact of these operations on type 2 diabetes raises the possibility of surgery for less obese patients with diabetes. Trials of RYGB in people with a BMI of less than 35 kg/m² have reported similar or higher diabetes remission rates than those conducted in severely obese patients (8). Given that leaner individuals lose less weight after RYGB than do more obese persons, the similar type 2 diabetes response suggests that the operation exerts weight-independent antidiabetic effects. Likewise, experimental variations of RYGB improve type 2 diabetes in obese and nonobese animals (12–14).

Despite growing interest in using RYGB specifically to target type 2 diabetes, mechanisms mediating its metabolic effects remain enigmatic. Evidence that gastrointestinal surgery eliminates type 2 diabetes more effectively than do nonsurgical therapies, and the rapid increase in bariatric surgery, impel research to elucidate mechanisms mediating these effects. Such insights could help to optimize surgical design and lead to novel pharmaceuticals.

Although the antidiabetic mechanisms of gastrointestinal surgery are not fully determined (9), most human studies of RYGB report increases in insulin sensitivity, with proportionate elevations in high-molecular-weight adiponectin, an insulin-sensitizing hormone. In muscle, insulin-receptor concentration and insulin signalling increase, as does expression of the mitochondrial transcription co-factor PGC-1 (peroxisome-proliferator-activated receptor-gamma coactivator-1) and of its target mitofusin-2 (MFN2). Insulin signalling and PGC-1 activity stimulate fatty acid metabolism, decreasing intracellular lipids in muscle and liver after RYGB, increasing insulin sensitivity.

Complementary observations pertain to the malabsorptive procedure BPD (15). Increases occur in muscle glucose uptake, expression of genes controlling glucose and fatty acid metabolism, and insulin-induced glucose oxidation and nonoxidative glucose disposal. The resulting increase in insulin sensitivity occurs by 6 postoperative months, peaking at 2 years, with no subsequent changes.

A key limitation of these observations is that they were made months to years following surgery, after profound weight loss had occurred. Because these findings are expected consequences of weight loss, they do not prove direct antidiabetic effects of gastrointestinal operations per se.

The following evidence indicates that RYGB improves glycaemic control through mechanisms beyond weight loss and reduced caloric intake.

Type 2 diabetes often resolves within days to weeks following RYGB, before substantial weight loss has occurred (5, 6). In a study of 240 patients with diabetes and undergoing RYGB, 30% were discharged from their surgical hospitalization with normal glucose concentrations off diabetes medicines—after an average inpatient stay of only 2.8 days (6). Most of the remainder discontinued diabetes treatments within a few weeks, and the eventual remission rate was 83%. Similar rapid resolution of type 2 diabetes following RYGB has been observed by many investigators, with in-hospital remission rates (within a few postoperative days) up to 89% (16). In contrast, purely gastric-restrictive operations, such as AGB (and which, like RYGB, involve perioperative caloric restriction followed by long-term weight loss), ameliorate diabetes only after promoting substantial weight loss (17). One study reported that insulin sensitivity increased significantly by 6 days after RYGB, before any meaningful weight loss (16). The improved glucose tolerance resulted largely from reduced fasting glucose, suggesting enhanced hepatic insulin sensitivity. Other investigators, however, argue that RYGB ameliorates diabetes primarily through increased insulin secretion (18).

In summary, glucose homeostasis can improve within a few days after RYGB, suggesting weight loss-independent mechanisms. Whether these rapid changes primarily involve increased insulin secretion, sensitivity, or both, is unclear.

Although weight loss typically ameliorates diabetes, recent studies have demonstrated that glucose control improves more after RYGB than with equivalent weight loss from other means, indicating weight-independent antidiabetic effects of RYGB. Laferrère et al. studied patients with type 2 diabetes who underwent either RYGB or dietary weight loss (19). The groups were exquisitely matched for preoperative severity and duration of diabetes, homoeostatic model assessment values, glucose tolerance, magnitude of incretin effect, BMI, age, and gender. Repeat glucose-homoeostasis measurements on subjects with an equivalent amount of weight loss (c.9.5 kg) revealed that the surgery increased glucose tolerance and the incretin effect substantially more than dieting. Similar findings apply in rats. In a study of RYGB versus AGB, Pattou et al. found analogous results (20). Fifty obese patients with type 2 diabetes, matched for preoperative BMI and glucose tolerance results, were studied postoperatively after 10% body weight reduction. Despite equivalent weight loss, the post-RYGB group showed markedly better oral glucose tolerance than did the post-AGB group. Similarly, LeRoux et al. observed better glucose tolerance in RYGB patients matched for weight loss with AGB patients, and the insulin sensitivity of post-RYGB subjects was equivalent to that of lean counterparts, despite significantly higher BMIs in the former group (21). Studies comparing RYGB versus purely restrictive vertical-banded gastroplasty also showed that, in patients with similar post-operative body weight and caloric intake, glucose concentrations decreased further and faster after RYGB (22). Finally, Lee et al. randomized patients for moderate obesity but poorly controlled type 2 diabetes to either a sleeve gastrectomy or a sleeved version of an RYGB, with a similar in-continuity gastric pouch after both operations (23). Although the RYGB ultimately causes greater, more durable weight loss than the sleeve gastrectomy, the two procedures caused equivalent weight loss after 6 months in this study. Nevertheless, HbA_1c decreased much further after RYGB than a sleeve gastrectomy.

In summary, glycaemic control improves more after RYGB than after equivalent weight loss from dieting or purely restrictive bariatric operations, indicating a weight-independent antidiabetic effect of RYGB, likely related to intestinal bypass.

Hints that an RYGB can exert weight-independent effects on β cell mass and/or function come from reports of patients with life-threatening hyperinsulinaemic hypoglycaemia developing long after surgery. In 2005, Service et al. described post-operative cases of apparent adult-onset nesidioblastosis severe enough to necessitate pancreatectomy (24, 25). Following similar reports by others (26), a meeting was convened to discuss approximately 135 known cases of severe, late-onset, post-RYGB hyperinsulinaemic hypoglycaemia, many with intractable neuroglycopenic episodes necessitating pancreatectomy.

An intuitive explanation for this phenomenon is that longstanding insulin resistance in obese patients produces hypertrophy and/or hyperplasia of β cells, which regress too slowly after RYGB to match weight-loss-induced improvements in insulin sensitivity. This hypothesis predicts that hypoglycaemia should occur early after surgery, when β cell mass is maximal and insulin sensitivity is peaking, i.e. either during dynamic weight loss or near the body-weight nadir (c.1 year). In actuality, known cases of hyperinsulinaemic neuroglycopenia have developed very late, 1–9 years after RYGB (typically, 2–4 years). At this time, β cell adaptation to the postoperative milieu, plus renewed insulin resistance from incipient weight regain, should prevent excessive insulin secretion. The late onset of hyperinsulinaemia in these cases suggests that RYGB might cause long-lasting stimulation of β cell mass and/or function—a likely benefit to most patients with diabetes, but a serious complication for some.

Reports of increased β cell mass in samples from hypoglycaemic post-RYGB patients who required pancreatectomy fuelled enthusiasm that RYGB might activate potentially novel β cell trophic factors (24, 26). Other investigators, however, found no changes in β cell mass or proliferation beyond those expected from obesity and therefore claimed that the syndrome must arise from overactive β cell function (27). Because pancreas samples from appropriately matched obese controls are rare, the controversy regarding whether RYGB stimulates β cell growth is difficult to address in humans. Nonetheless, failure to reduce β cell output to match improved insulin sensitivity never occurs with nonsurgical weight loss; thus, these observations constitute additional evidence favouring weight-independent glucose-lowering effects of RYGB.

Several plausible hypotheses can be articulated to explain the rapid, weight-independent glycaemic effects of RYGB. None of these preclude the others, so any combination may apply.

One theory asserts that glycaemia improves soon after RYGB simply because patients eat very little soon after surgery, thus their β cells are minimally challenged. Transiently improved glycaemia in type 2 diabetes with acute caloric restriction is well described, and, according to this model, by the time patients return to ad libitum eating, they begin to experience the insulin-sensitizing effects of weight loss from the operation.

Although this hypothesis is reasonable, it fails to explain many observations about post-RYGB glycaemic control. First, if acute caloric restriction were the major effector of improved insulin sensitivity, rapid diabetes remission would occur following all bariatric procedures, because all involve perioperative food restriction followed by weight loss. After AGB, however, diabetes remits in only 48% of cases, compared with 84% of cases after RYGB (1). More importantly, the pace of glycaemic improvement is far slower after AGB than RYGB, and is strongly linked to the timing and degree of weight loss. In a randomized trial of AGB versus best medical care for type 2 diabetes, no patients experienced diabetes remission within 6 months of surgery, even though after 2 years the disease resolved in 73% of these individuals (who had very modest degrees of hyperglycaemia) (17). In contrast, most type 2 diabetes patients who undergo RYGB experience disease remission within several days to a few weeks (6).

In many types of general surgery, especially involving the gastrointestinal tract, patients are subjected to perioperative caloric restriction. However, rapid remission of type 2 diabetes is not observed in these settings; if anything, glycaemic control worsens post-operatively due to inflammation and up-regulation of counterregulatory stress hormones such as cortisol and catecholamines. The remarkable phenomenon of rapid post-operative type 2 diabetes dissipation is unique to RYGB and other intestinal-bypass operations, such as BPD. Furthermore, the starvation-followed-by-weight-loss hypothesis fails to explain the superior glycaemic control achieved after RYGB versus equivalent weight loss from dieting or restrictive bariatric operations. Neither can this model account for the severe hyperinsulinaemic hypoglycaemia that occasionally develops late after RYGB. Although the starvation theory has merit, it cannot explain the full antidiabetic impact of RYGB.

Our group provided the first evidence suggesting that compromised secretion of the orexigenic, prodiabetic, upper gastrointestinal hormone ghrelin might contribute to the anorexic and antidiabetic effects of RYGB (28). We found that human 24-h ghrelin profiles displayed marked preprandial surges followed by postprandial suppression, and levels increased proportionately to dietary weight loss. These and other observations implicated ghrelin in both mealtime hunger and the adaptive increase of hunger that resists non-surgical weight loss. Because more than 90% of ghrelin is produced by the stomach and duodenum—tissues altered by RYGB—we hypothesized that ghrelin regulation might be disturbed following this operation. Indeed, we found that 24-h ghrelin profiles were extremely low in post-RYGB patients, a paradoxical response to profound weight loss. Since then, eight other groups have shown prospectively that ghrelin levels fall after RYGB, and four cross-sectional studies have confirmed abnormally low levels in post-RYGB patients compared with appropriate controls (29). Ghrelin also decreases after RYGB in rodents. Three other groups found no significant change in ghrelin levels after RYGB, but interpreted this as an impairment of normal ghrelin stimulation by weight loss. In contrast, four groups reported normal ghrelin elevations with RYGB-induced weight loss. These heterogeneous findings suggest that differences in surgical techniques, possibly involving the vagus nerve (30), might account for disrupted ghrelin secretion in most but not all cohorts.

Beyond contributing to decreased food intake following RYGB, compromised ghrelin secretion might also improve glucose tolerance. Ghrelin can stimulate insulin counterregulatory hormones, suppress the insulin-sensitizing hormone adiponectin, block hepatic insulin signalling, and inhibit insulin secretion (31). All of these actions acutely elevate blood glucose levels, as do ghrelin’s chronic effects to increase food intake, gastrointestinal motility, and body weight. Thus, part of the glycaemic improvement after RYGB may arise from reduced ghrelin secretion.

An intuitive potential mechanism for improved glucose homeostasis after some bariatric operations involves expedited delivery of ingested nutrients to the lower bowel, due to an intestinal bypass. An increase in unabsorbed nutrients in the distal gut should accentuate secretion of glucagon-like peptide 1 (GLP-1), thereby improving glucose homeostasis. GLP-1 is an incretin peptide that increases glucose tolerance by enhancing insulin secretion, suppressing glucagon secretion, inhibiting gastric emptying, increasing β cell mass (at least in animals), and, possibly, improving insulin sensitivity. The peptide—along with peptide YY (PYY) and oxyntomodulin—is produced primarily in the ileum and colon by nutrient-stimulated L cells. All three peptides can reduce food intake, and are, therefore, implicated as possible contributors to the anorectic effects of some bariatric operations.

Consistent with the lower intestinal hypothesis, the bariatric operations most noted for rapid, high-frequency type 2 diabetes remission—RYGB and BPD—both create gastrointestinal shortcuts for food to access the distal bowel. After BPD, which conducts food directly from the stomach to the ileum, postprandial GLP-1 excursions are unquestionably increased. It is less obvious that this would occur with RYGB, because its intestinal bypass is far less extensive. Moreover, GLP-1 secretion is stimulated not only by direct nutrient contact with distal intestinal L cells, but also by proximal nutrient-related signals transmitted neurally from the duodenum to the distal bowel (32). Since RYGB diverts nutrients away from the duodenum, the operation might, theoretically, lower postprandial GLP-1 levels. Nevertheless, recent studies demonstrate that meal-stimulated secretion of GLP-1 and other L-cell peptides is indeed substantially and durably increased after RYGB, but not after AGB (21, 33, 34). Consistent with elevated postprandial GLP-1 secretion, post-RYGB patients display an increased incretin effect (19, 34). GLP-1 enhances insulin secretion, but may also increase proliferation and decrease apoptosis of β cells, at least in rodents (35). Thus, increased GLP-1 secretion may mediate the expansion of β cell mass claimed by some investigators to accompany post-RYGB hyperinsulinaemic hypoglycaemia (24, 26).

Further support for the lower intestinal hypothesis comes from experiments involving ileal interposition, wherein a segment of the L-cell-rich ileum is transplanted into the upper intestine, increasing its exposure to ingested nutrients and enhancing postprandial GLP-1 and PYY surges. This operation, with no gastric restriction or malabsorption, improves glycaemic control—with or without weight loss, depending on the rodent model or humans studied (36–38). It remains unclear whether the procedure primarily enhances insulin secretion, as predicted from increases in the incretin GLP-1, or improves insulin sensitivity; findings from different experiments support both possibilities.

This theory posits that exclusion of a short segment of proximal small intestine (primarily, the duodenum) from contact with ingested nutrients exerts direct antidiabetes effects, presumably via one or more unidentified glucoregulatory duodenal factors or processes.

Francesco Rubino was the first to provide strong evidence supporting this model. He developed the duodenal–jejunal bypass (DJB), a gastric-sparing variant of RYGB involving an intact stomach but a modest proximal intestinal bypass, similar to that in a standard RYGB (Fig. 13.4.5.2) (12). In Goto–Kakizaki rats, a nonobese model of polygenic type 2 diabetes, DJB improved diabetes—rapidly, durably, and impressively—without altering food intake or body weight, compared with sham-operated controls. Similar independent observations have subsequently been made in nonobese diabetic Goto–Kakizaki rats (14, 39) and obese diabetic Zucker rats (13). Likewise, several small, ongoing, human studies of DJB show improved glycaemic control in obese and nonobese patients, with little or no weight loss (40–42; J Arguelles-Sarmiento and H Bernal-Valazquez, personal communication, 2006 and M Lakdawala, personal communication, 2007).

More recent work has demonstrated that the antidiabetic effect of DJB in rats stems from nutrient exclusion of the proximal small intestine (43). A variation of DJB—in which ingested food passes from the stomach not only into the proximal jejunum (as in a regular DJB), but also into the proximal duodenum through the pylorus—had no impact on glycaemia. Goto–Kakizaki rats subjected to DJB with duodenal exclusion followed by DJB without duodenal exclusion, or vice versa, experienced reversible remission and reconstitution of type 2 diabetes. Diabetes was eliminated or restored based on the absence or presence, respectively, of duodenal nutrient passage, with an unchanging minor shortcut for nutrients to reach the lower bowel (44). These studies indicate that enhanced delivery of nutrients to the distal intestine is unlikely fully to explain early diabetes improvement following upper intestinal bypass. The work strongly implicates a role for the excluded proximal intestine per se, identifying a fundamentally novel physiological effect of RYGB and, possibly, shedding new light on diabetes pathogenesis.

Additional support for the upper intestinal hypothesis comes from experiments in which a flexible plastic sleeve is implanted in the upper intestine, conducting food from the pylorus to the beginning of the jejunum, avoiding duodenal mucosal exposure. Such endoluminal duodenal sleeves, which have been tested in rats, pigs, and humans, promote little or no weight loss, but markedly improve glucose tolerance (45–48). In humans with type 2 diabetes, the device substantially improved fasting and postprandial area-under-the-curve glucose concentrations, starting as early as 1 week, long before any weight loss had occurred (47, 49). By 6 months, although minor weight loss (3.8 kg) had begun, glycaemia was disproportionately improved, with a remarkable fall of HbA_1c levels from 9.0% to 6.1% (75 to 43 mmol/mol)—a result better than would be expected with any type 2 diabetes medication except insulin.

A very intriguing, unexpected feature of both DJB and upper intestinal sleeves is that they reduce fasting and postprandial blood glucose levels to approximately the same degree, and, consistent with this continuous reduction in glycaemia, they have a major impact on HbA_1c levels (12, 40–43, 45, 47, 48). Thus, although these interventions simply re-route food through the gastrointestinal tract after meals, they exert salutary effects on glycaemia that persist between meals.

Recent studies have demonstrated that intestinal nutrient sensing and metabolism influence insulin sensitivity, complementing the known insulin-secretory effects of intestinal incretins. Wang et al. showed in rats that calorically insignificant intraduodenal lipid infusions activate a novel intestine–brain–liver neurocircuit to increase hepatic insulin sensitivity (49). This pathway involves intestinal sensing of fatty acyl-CoA (co-enzyme A) molecules, generating signals that are transmitted up the afferent vagus nerve to the hindbrain, then down the efferent vagus to the liver, reducing hepatic glucose output. Thus, the intestine acts as an early responder to ingested food, heralding the conversion from a non-fed to fed status and preventing profligate mobilization of endogenous fuel stores following meals. This circuit operates in concert with established postprandial actions of the intestine to generate incretins that facilitate insulin secretion, as well as satiation factors that promote meal termination, thereby limiting both internal and external fuel influxes to minimize postprandial glycaemic perturbations. The intestine thus emerges as a neuroendocrine organ regulating food intake, insulin secretion, and insulin action to improve glucose tolerance and help orchestrate a seamless postprandial transition from fuel catabolism to storage.

It is not clear how this neurocircuitry might be affected by gastrointestinal surgery to help mediate antidiabetes effects. On face value, the observation that proximal intestinal bypass operations, e.g., RYGB and DJB, improve glucose tolerance seems paradoxical to these new observations, in which intraduodenal nutrient administration increased insulin sensitivity. This apparent contradiction might be explained if the relevant site of nutrient sensing to activate the circuit is downstream of the duodenal lipid infusion site used by Wang et al. (49); e.g. in the jejunum. RYGB expedites jejunal delivery of nutrients, including fatty acids (which are converted to fatty acyl-CoAs), and, because these arrive in the jejunum unconjugated with bile acids, they could efficiently activate a fatty acid-stimulated insulin-sensitizing pathway originating at that site. Alternatively, if the relevant nutrient sensor is in the duodenum, it might be hyperactivated after proximal intestinal bypass operations by fat-rich bile, which is secreted postoperatively into the duodenum without being diluted by, or conjugated with, food. Interestingly, serum bile acids are elevated in post-RYGB patients, correlating with adiponectin and GLP-1 levels (50). These and other speculative hypotheses offer compelling opportunities for further research.

Another recently described mechanism for gut regulation of insulin sensitivity involves intestinal carbohydrate metabolism. Mithieux et al. showed that the key enzymes for gluconeogenesis are expressed in the small intestine, where they are induced in energy-deficient states (51–53). The resulting intestinal glucose production apparently activates portal-vein glucose sensors to engage a neurocircuit that increases hepatic insulin sensitivity, decreasing hepatic glucose output (analogous to the aforementioned lipid-sensing circuit), while also inhibiting food intake. A stomach-sparing proximal intestinal bypass operation (a DJB variant) increased expression of gluconeogenesis enzymes in the distal bowel and enhanced intestinal glucose output into the portal vein (53). The authors linked this to a postoperative increase in hepatic insulin sensitivity and overall glucose tolerance, effects that were independent of GLP-1 and only partly explained by weight loss. They concluded that the beneficial glycaemic effects of proximal intestinal bypass involve enhanced intestinal gluconeogenesis and its detection by portal-vein glucose sensors that increase hepatic insulin sensitivity via a neural pathway.

Several features of this model are enigmatic. For example, why would intestinal gluconeogenesis, which is induced in energy-deficient states (fasting, uncontrolled diabetes, and a high-protein diet) (51–53), generate signals that decrease food intake and hepatic glucose output? These are maladaptive responses to energy insufficiency. Moreover, if intestinal gluconeogenesis is normally stimulated by nutrient deficiency, why would it increase in the distal bowel after intestinal bypasses that enhance nutrient delivery to that region?

Although many questions remain regarding the lipid- and carbohydrate-based intestinal neurocircuits that influence hepatic glucose production, the novel concept that the gut regulates insulin sensitivity in response to ingested nutrients opens new arenas of research into how the re-routeing of nutrient flow by gastrointestinal operations might influence these systems to affect glucose tolerance.

RYGB typically causes remission of type 2 diabetes, and mounting evidence indicates that this remarkable phenomenon results from effects beyond just reduced food intake and body weight. Weight-independent anti-diabetes actions of RYGB are apparent from the rapid resolution of type 2 diabetes before weight loss occurs, the greater improvement of glucose homeostasis after RYGB than after equivalent weight loss by other means, and the occasional development of very late-onset, β cell hyper-function. Several mechanisms may mediate the direct anti-diabetes impact of RYGB, including enhanced nutrient stimulation of distal intestinal GLP-1, intriguing but unidentified phenomena related to exclusion of the upper intestine from ingested nutrients, compromised ghrelin secretion, possible alterations of intestinal nutrient sensing and metabolism that affect insulin sensitivity, and, probably, other effects. It is increasingly clear that the gut plays a major role in glucose homeostasis, regulating insulin secretion and sensitivity (35, 49, 53, 54), and RYGB likely influences several gastrointestinal pathways in complementary ways to improve glucose control. Further characterizing the antidiabetes mechanisms of gastrointestinal surgery is a compelling objective that promises not only to guide surgical design, but also to reveal novel targets for diabetes pharmacotherapeutics.