Narrative Review: Effect of Bariatric Surgery on Type 2 Diabetes Mellitus
M. Vetter,S. Cardillo,M. Rickels,N. Iqbal
DOI: https://doi.org/10.7326/0003-4819-150-2-200901200-00007
IF: 39.2
2009-01-20
Annals of Internal Medicine
Abstract:Key Summary Points The rapid improvement in glycemic control after bariatric surgery results from caloric restriction and alterations in the gut hormones that control insulin secretion. The enteroinsular axis includes the incretins glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic peptide (GIP), ghrelin, and peptide YY (PYY) and their subsequent effect on insulin secretion and sensitivity. Restrictive, malabsorptive, and combined bariatric surgery procedures have different effects on the enteroinsular axis. Intestinal bypass procedures increase GLP-1 and PYY levels. In contrast, restrictive procedures do not increase incretin or PYY levels. Familiarity with these changes can help physicians consider the various surgical approaches and develop postoperative treatment regimens for patients. Glycemic control in diabetic patients improves markedly within days of bariatric surgery, which suggests that the procedures alter the hormones that control insulin secretion (1). The enteroinsular axis includes the gut hormones glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic peptide (GIP). These hormones, also known as incretins, are secreted by intestinal L and K cells, respectively, in response to nutrients and directly enhance insulin secretion (2). Restrictive, malabsorptive, and combined bariatric surgery procedures affect the enteroinsular axis differently. The various bariatric procedures also affect the secretion of other gut hormones that affect insulin sensitivity, including ghrelin and peptide YY (PYY). Thus, an altered pattern of gut hormone secretion after bariatric surgery may profoundly affect glucose tolerance. We focus on the short-term pathophysiologic changes in the enteroinsular axis and their subsequent effect on insulin secretion and sensitivity after bariatric surgery. Familiarity with these changes can help clinicians decide among the different surgical approaches and formulate treatment regimens that avoid severe postoperative hypoglycemia. Methods We searched English-language publications in PubMed and reference lists from relevant articles published between 1967 and 2008. Our main search terms were bariatric surgery, Roux-en-Y, gastric bypass, biliopancreatic diversion, gastric banding, laparoscopic adjustable gastric banding, diabetes, enteroinsular axis, incretins, GLP-1, GIP, ghrelin, PYY, insulin, and postoperative management. To determine the rates of diabetes resolution, we included studies that enrolled at least 10 diabetic patients (alone or along with nondiabetic patients) and reported diabetes-related outcomes. We retrieved randomized, controlled trials; cohort studies; and casecontrol studies that reported weight loss, diabetes resolution, and time to restoration of normoglycemia. Because we found few such studies, we also included large case series. We evaluated study quality by using the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) system (3). We used our judgment to identify high-quality studies that described gut hormone levels after bariatric surgery. Types of Bariatric Procedures and Effect on Weight Loss and Diabetes Since its inception in the 1950s, bariatric surgery has become increasingly refined. More recently, it has been touted as a cure for diabetes (4). Several procedures are now available (Figure 1). Bariatric procedures were initially classified as restrictive, malabsorptive, or combined, reflecting the purported mechanism of weight loss (1). Restrictive procedures, such as laparoscopic adjustable gastric banding (LAGB) and vertical banded gastroplasty (VBG), greatly reduce the volume of the stomach to decrease food intake and induce early satiety. Malabsorptive procedures, such as biliopancreatic diversion (BPD), shorten the small intestine to decrease nutrient absorption. Combined procedures, such as the Roux-en-Y gastric bypass (RYGB), incorporate both restrictive and malabsorptive elements. Roux-en-Y gastric bypass surgery is the current gold standard treatment for severe obesity. Both BPD and RYGB alter the secretion of orexigenic and anorexigenic gut peptides, which interact with appetitive centers in the arcuate nucleus of the hypothalamus to decrease appetite (5). Because both BPD and RYGB bypass similar segments of the small bowel, we use the term intestinal bypass procedure to refer to either. Figure 1. Surgical procedures. Reprinted with permission of the American Society for Metabolic and Bariatric Surgery, copyright 2008, all rights reserved. Weight Loss Two recent meta-analyses (6, 7) reported weight loss; operative mortality; and obesity-related comorbid conditions, including diabetes, after bariatric surgery. Buchwald and colleagues (7) reported mean excess weight loss (see Glossary) (710) of 61% across all procedures in 22094 patients; weight loss rates associated with each procedure varied (Table 1). On average, bariatric surgery reduces body mass index by 10 to 15 kg/m2 and weight by 30 to 50 kg (11). Table 1. Results of Different Types of Bariatric Surgery Several relatively poor-quality randomized, controlled trials that compared different bariatric surgery procedures (12) showed that weight loss was greater with gastric bypass than with VBG or LAGB. The SOS (Swedish Obese Subjects) study (13), a landmark observational study that followed more than 4000 obese participants, matched those who selected medical management with those undergoing various bariatric procedures, including RYGB, VBG, or LAGB. At 10 years, RYGB was associated with a 25% reduction in total body weight, whereas VBG and LAGB were associated with 16% and 14% weight loss, respectively. Diabetes Resolution and Improvement Observational evidence suggests that bariatric surgery is associated with a 60% to 80% rate of diabetes resolution (14). In 1 meta-analysis (7), approximately 15% of the patients were diabetic. In studies reporting complete resolution of diabetes (defined as normoglycemia with no diabetes medications), 1417 of 1846 patients (76.8%) met the criteria for resolution. Among studies reporting resolution or improvement of diabetes, 414 of 485 patients (mean, 86.0%) experienced either outcome. Table 1 shows rates of diabetes resolution for individual bariatric procedures. Conclusions about bariatric surgery and diabetes resolution come with an important qualifier: The studies had serious methodological weaknesses. Few are randomized, controlled trials; most surgical outcome studies are uncontrolled case series with considerable missing data (6, 10). In 1 meta-analysis (6), one quarter of the studies did not report enrolling consecutive patients and fewer than 50% reported how many enrolled patients provided follow-up data. Table 2 (4, 1422) includes selected studies that met the minimum GRADE criteria quality standards and reported follow-up rates of at least 80% (3). Paired comparisons of surgical procedures typically favored RYGB or BPD over the restrictive procedures (23). Table 2. Efficacy for Resolution of Diabetes Predictors of Diabetes Resolution Identifying preoperative predictors of diabetes resolution is critical for determining which diabetic patients will obtain the greatest benefit from surgery. In earlier studies of RYGB, longer duration of diabetes (>10 years), poor preoperative glycemic control, and preoperative insulin use reduced the probability of diabetes resolution (18, 19); however, these studies did not adjust for the effects of known confounding factors. More recently, Torquati and colleagues (24) adjusted for body mass index, sex, and preoperative hemoglobin A1c level and found that preoperative treatment with oral antidiabetic agents (as opposed to insulin) and smaller preoperative waist circumference predicted diabetes resolution (18). Shorter duration of diabetes was a weaker, statistically nonsignificant predictor, which supports earlier studies of RYGB (18). The same factors predict diabetes resolution after gastric banding. Dixon and colleagues (25) reported that diabetes for less than 3 years predicted diabetes resolution, after they controlled for age and excess weight loss. Less deterioration in -cell function at the time of surgery may maximize the effect of the surgery-altered secretion of gut peptides that enhance -cell insulin secretion. The Enteroinsular Axis Bayliss and Starling first described the connection between the gut and the pancreas in 1902, when they demonstrated that intestinal mucosa extracts contained a factor, which they called secretin, that acted through the bloodstream to stimulate exocrine secretion by the pancreas (26). Sixty-five years later, Perley and Kipnis (27) demonstrated that ingested nutrients stimulated greater insulin release than intravenously administered glucose. In 1979, Creutzfeldt (28) defined incretins as gastrointestinal hormones that stimulate insulin release after enteral nutrition. This connection between the gut and pancreatic islet cells is called the enteroinsular axis, a term first used by Unger and Eisentraut (29). The Incretins: GLP-1 and GIP By potentiating glucose-dependent insulin secretion, GLP-1 and GIP account for 50% to 60% of nutrient-stimulated insulin release (2). In animal models of diabetes, GLP-1 also increases -cell mass through regulation of proliferation, neogenesis, and apoptosis (30). Glucagon-like peptide-1 is a potent insulin secretatogue that is secreted by the L cells of the distal ileum in response to ingested nutrients and is inactivated by the enzyme dipeptidyl peptidase IV (DPP-IV) (31). By activating adenylate cyclase, GLP-1 acts on pancreatic islets to augment glucose-dependent insulin secretion. The subsequent increase in insulin levels within islets inhibits glucagon secretion, possibly through direct activation of GLP-1 receptors on cells (31). Glucagon-like peptide-1 also slows gastric emptying, which delays digestion and blunts postprandial glycemia (32), and acts on the central nervou