当前位置: 首页 > 期刊 > 《新英格兰医药杂志》 > 2005年第3期 > 正文
编号:11325554
Gastric Bypass and Nesidioblastosis — Too Much of a Good Thing for Islets?
http://www.100md.com 《新英格兰医药杂志》
     As the obesity pandemic continues to worsen and medical interventions remain only moderately effective, bariatric surgery is at present the only method that reliably results in major, long-term weight loss.1,2 The most successful procedures durably reduce body weight by about one third and ameliorate nearly all obesity-related complications, most notably — almost miraculously — type 2 diabetes mellitus.2 Not surprisingly, bariatrics is the fastest-growing surgical subspecialty. The number of bariatric surgeons registered with the American Society for Bariatric Surgery increased by almost 50 percent per year during the past several years, and the number of bariatric operations nearly doubles annually.3 Still, questions remain with regard to the physiological mechanisms and pathophysiological consequences of bariatric operations, especially Roux-en-Y gastric bypass, the most common bariatric procedure performed in the United States. In this issue of the Journal, Service et al. describe a novel complication of Roux-en-Y gastric bypass: pathologic overgrowth of pancreatic beta cells (nesidioblastosis), resulting in life-threatening hyperinsulinemic hypoglycemia.4 These findings identify a new adverse effect to consider in an increasing patient population and may also shed light on one mechanism by which diabetes resolves after bariatric surgery.

    Service et al., experts in the field of hypoglycemic disorders, report that in the past five years, 40 percent of their confirmed cases of nesidioblastosis occurred in persons who had undergone Roux-en-Y gastric bypass surgery, whereas less than 0.1 percent of the general population has had this operation.3 Their patients presented with repeated episodes of symptoms of profound postprandial neuroglycopenia associated with endogenous hyperinsulinemic hypoglycemia. Nesidioblastosis was definitively confirmed by selective arterial calcium-stimulation testing, alleviation of hypoglycemic symptoms after partial pancreatectomy, and demonstration of diffuse beta-cell hypertrophy and hyperplasia in resected pancreatic tissue. Insulinoma was ruled out with triple-phase spiral computed tomography and intraoperative ultrasonography in all but one patient, who had nesidioblastosis as well as multiple insulinomas. Although a causal link between Roux-en-Y gastric bypass and nesidioblastosis was not established, these observations strongly suggest that gastric bypass can occasionally result in pathologic beta-cell overgrowth and hypoglycemia. This possibility is supported by other case reports of nesidioblastosis5 and insulinoma6 after Roux-en-Y gastric bypass.

    What mechanisms could cause nesidioblastosis after a procedure that diverts food from 95 percent of the stomach and a few feet of proximal small intestine? One theoretical possibility is that in obese persons with insulin resistance, adaptive beta-cell hypertrophy develops and causes hypoglycemia after insulin sensitivity is improved by the surgically induced weight loss. Service et al. argue against this as a mechanism by showing that the islets in obese control subjects without gastric bypass are of normal size.4 Moreover, there is no association between non–surgically induced weight loss and endogenous hypoglycemia or nesidioblastosis.

    Another plausible explanation, with broader implications, is that nesidioblastosis occasionally arises after Roux-en-Y gastric bypass because of long-term stimulation of beta-cell growth and activity by gut hormones that are perturbed as a result of the altered gastrointestinal transit. A prime candidate to mediate this effect is the incretin hormone glucagon-like peptide 1 (GLP-1). Produced by L cells primarily in the distal intestine, GLP-1 is secreted in response to the rapid passage of food from the stomach into the intestine and direct contact of the hindgut with nutrient chyme.7 GLP-1 potently increases insulin secretion and, possibly, insulin sensitivity.8 Moreover, at least in rodents, GLP-1 triggers beta-cell neogenesis and proliferation while inhibiting apoptosis.9

    Theoretically, operations that expedite the delivery of nutrients to the hindgut should increase GLP-1 secretion, thereby enhancing insulin-mediated glucose disposal. Indeed, large (up to 10-fold) and durable (up to 20-year) elevations of GLP-1 or other nutrient-stimulated L-cell hormones, including peptide YY and enteroglucagon, have been documented after Roux-en-Y gastric bypass, biliopancreatic diversion, and jejunoileal bypass surgery.10 These bariatric operations create shortcuts to the hindgut for ingested nutrients. They also powerfully reverse diabetes, which completely resolves in well over 80 percent of cases.1,2,11,12 Remarkably, diabetes typically remits within days to weeks after these operations — too early for the remission to be explained by weight loss alone, suggesting that other mechanisms, such as modulation of gut hormones, may play a role.

    If increased secretion of GLP-1 and possibly other gastrointestinal hormones after gastric bypass provides continuing beta-cell stimulation, could this phenomenon occasionally result in nesidioblastosis? The observations of Service et al. support such a possibility. Accordingly, the authors caution against ascribing postprandial vasomotor symptoms to the dumping syndrome in patients who have undergone gastric bypass without considering organic hyperinsulinism.4 However, since the Mayo Clinic, a hypoglycemia referral center, observed only six cases of apparent Roux-en-Y gastric bypass–induced nesidioblastosis in five years,4 this complication hardly represents a public health crisis.

    Perhaps more important, the findings of Service et al. hint at a possible risk resulting from long-term medicinal stimulation of GLP-1 signaling, the newest strategy used to treat type 2 diabetes.13 The GLP-1–receptor agonist exenatide recently received approval by the Food and Drug Administration, and three related compounds — liraglutide, CJC-1131, and ZP10 — are in clinical trials. Agents are also in development that increase endogenous GLP-1 signaling by inhibiting dipeptidyl peptidase IV (DPP-IV), an enzyme that degrades GLP-1. Three such compounds, vildagliptin, sitagliptin, and saxagliptin, are in advanced clinical trials, and many others are in the pipeline. GLP-1–receptor agonists reduce levels of glycosylated hemoglobin as effectively as existing oral agents do while promoting weight loss or preventing weight gain. Moreover, because these compounds may increase beta-cell mass, they might slow or reverse the progressive islet-cell deterioration characteristic of diabetes. Since GLP-1–like agents acutely stimulate insulin secretion only in hyperglycemia, they are heralded as posing little independent risk of hypoglycemia. The findings of Service et al. would suggest that the last of these assertions may not always be true, if long-term overstimulation of GLP-1 signaling can cause pathologic beta-cell hypertrophy and hyperactivity, culminating in hypoglycemia.

    Before one laments this potential drawback of medicines that increase GLP-1 signaling, however, some caveats should be considered. First, DPP-IV inhibitors, which hinder degradation of the endogenous hormone, do not stimulate GLP-1 signaling to the extent that high-dose GLP-1–receptor agonists do. Consequently, DPP-IV inhibitors probably decrease glucose somewhat less effectively, but they are also theoretically less prone to cause adverse effects such as nausea and, possibly, nesidioblastosis. Second, even doses of GLP-1–receptor agonists that substantially exceed peak endogenous GLP-1 activity improve diabetes far less than either Roux-en-Y gastric bypass or biliopancreatic diversion. Thus, these bariatric operations probably increase glucose disposal through mechanisms other than just increasing GLP-1. Because postsurgical antidiabetic effects occur before substantial weight loss, alterations in other gut hormones may play a role. For example, the orexigenic foregut hormone ghrelin can exert prodiabetic effects by suppressing insulin secretion, stimulating counter-regulatory hormones, and directly opposing insulin action.14 Roux-en-Y gastric bypass usually impairs ghrelin secretion,15 possibly increasing glucose tolerance. Other gut hormones that could theoretically mediate some of the effects of bariatric surgery include peptide YY, oxyntomodulin, and factors yet to be discovered. In short, bariatric surgery may increase glucose disposal through multiple mechanisms, including increased GLP-1 secretion. Thus, the risk of hypoglycemia after gastric bypass should exceed that of the use of medicines that selectively increase GLP-1 signaling.

    Finally, should reports of Roux-en-Y gastric bypass–associated nesidioblastosis be considered worrisome or promising? Nesidioblastosis probably represents the pathologic extreme of a phenomenon that would benefit the vast majority of obese patients with diabetes. The possibility that Roux-en-Y gastric bypass stimulates beta-cell–trophic factors should spur research to identify these entities, GLP-1 or otherwise, so that their physiological effects can be harnessed and used pharmacologically to treat diabetes.

    Source Information

    From the Department of Medicine, Division of Metabolism, Endocrinology, and Nutrition, University of Washington, and Veterans Affairs Puget Sound Health Care System, Seattle.

    References

    Sj?str?m L, Lindroos A-K, Peltonen M, et al. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med 2004;351:2683-2693.

    Buchwald H, Avidor Y, Braunwald E, et al. Bariatric surgery: a systematic review and meta-analysis. JAMA 2004;292:1724-1737.

    Steinbrook R. Surgery for severe obesity. N Engl J Med 2004;350:1075-1079.

    Service GJ, Thompson GB, Service FJ, Andrews JC, Collazo-Clavell ML, Lloyd R. Hyperinsulinemic hypoglycemia with nesidioblastosis after gastric-bypass surgery. N Engl J Med 2005;353:249-254.(David E. Cummings, M.D.)