Surgery for Obesity and Related Diseases 7 (2011) 691– 696 Original article Predictors of remission of type 2 diabetes mellitus after laparoscopic gastric banding and bypass Numan Hamza, M.D.a,b, Muhammad Hasan Abbas, M.Phil.b,c, Ammar Darwishb,c, Zainab Shafeekb,c, John Newa, Basil J. Ammori, M.B.Ch.B., M.D.a,b,c,* a Salford Royal Hospital, Salford, United Kingdom University of Manchester, Manchester, United Kingdom c Manchester Royal Infirmary, Manchester, United Kingdom Received October 29, 2009; revised January 23, 2010; accepted March 10, 2010 b Abstract Background: Type 2 diabetes mellitus (T2DM) is associated with obesity and results in considerable morbidity and mortality. Our objectives were to evaluate the effect of laparoscopic bariatric surgery on the control of T2DM in morbidly obese patients in a U.K. population and to determine the predictors of T2DM remission after bariatric surgery. The study was performed at teaching university hospitals and affiliated private hospitals. Methods: Of 487 patients who underwent laparoscopic bariatric procedures from 2002 to 2007, 74 patients (15.2%) had established T2DM. The results are presented as the mean values. Multivariate analysis was used to identify the factors predictive of remission of T2DM after bariatric surgery. Results: The body mass index before laparoscopic gastric bypass (LGB; n ⫽ 48) and laparoscopic adjustable gastric banding (LAGB; n ⫽ 26) were comparable (52 versus 51 kg/m2, P ⫽ .508). At a mean follow-up of 16.9 months, 41% had remission and 59% had experienced improvement in T2DM. Although the duration of follow-up was significantly longer for the patients who had undergone LAGB than for those who had undergone LGB (23 versus 13.4 months, P ⫽ .001), the percentage of excess weight loss (%EWL) was significantly greater after LGB than after LAGB (59.4% versus 48.8%, P ⫽ .031), with an associated greater remission rate of T2DM (50% versus 24%, P ⫽ .034). Multivariate analysis revealed a greater %EWL and younger age to be independent predictors of postoperative remission of T2DM, and LGB, longer follow-up, and female gender were independent predictors of a greater %EWL. Conclusion: The %EWL was the only predictor of remission of T2DM that was influenced by the choice of bariatric procedure. In our study, LGB offered greater weight loss and a chance of remission of T2DM compared with LAGB and within 2 years of surgery. (Surg Obes Relat Dis 2011;7:691– 696.) © 2011 American Society for Metabolic and Bariatric Surgery. All rights reserved. Keywords: Laparoscopic gastric bypass; Gastric banding; Obesity; Diabetes Obesity and type 2 diabetes mellitus (T2DM) are currently 2 of the most common chronic debilitating diseases of Western societies, and both have experienced epidemic This study was presented at the Annual Meeting of the European Association for Endoscopic Surgery, Stockholm, Sweden, June 11–14, 2008. *Reprint requests: Basil J. Ammori, M.B.Ch.B., M.D., University of Manchester, Salford Royal Hospital, Stott Lane, Manchester M6 8HD United Kingdom. E-mail: bammori@btinternet.com growth in the past few decades [1]. An estimated 2.35 million people have diabetes in England (4.8% of the population) and 90% of these have T2DM [2]. The incidence of T2DM among patients undergoing bariatric surgery has been 15–23% [1,3–5]. Bariatric surgery is vastly superior to traditional weight loss therapies in promoting weight reduction in morbidly obese subjects [6] and has been more effective in achieving remission of T2DM than conventional therapies [7]. The aims of the present study were to evaluate the effect 1550-7289/11/$ – see front matter © 2011 American Society for Metabolic and Bariatric Surgery. All rights reserved. doi:10.1016/j.soard.2010.03.292 692 N. Hamza et al. / Surgery for Obesity and Related Diseases 7 (2011) 691– 696 of laparoscopic bariatric surgery on the control of T2DM in a U.K. population and to study the factors predictive of disease remission after bariatric surgery. Methods free fluids the following morning and a liquid diet at lunchtime of the first postoperative day. The patients were discharged from hospital when adequately mobile and tolerating a liquid diet. We did not routinely perform oral Gastrografin studies postoperatively. Patients Follow-up Patients with morbid obesity (body mass index [BMI] ⱖ40 kg/m2 or those with a BMI of ⱖ35 kg/m2 with an associated co-morbidity) qualified for bariatric surgery. We extended surgery to include patients with T2DM with a BMI as low as 32 kg/m2. The choice of surgical procedure was the patient’s ultimate informed decision. The patients who had undergone LGB were followed up at our outpatient clinic at 6 weeks postoperatively and then at 3-month intervals during the first year and 6-month thereafter. Those who had undergone LAGB were examined more frequently. The first band inflation was performed with radiologic guidance and was done for the large majority of patients at 6 weeks postoperatively. Subsequent band inflation was guided by the patients’ dietary intake and weight loss progression as assessed at the 3-month follow-up visits. The patients had easy access to the clinics, dietician, and bariatric nurse and surgeon for advice and arrangement of additional band adjustments. Telephone conversations, electronic mail, and a postal MooreheadArdelt Quality of Life Questionnaire [8] were an integral part of our follow-up protocol. Patients with T2DM were regularly followed up by their diabetologist at the local hospital or at the diabetic clinic of their general practitioner, and their diabetic medication was titrated accordingly. Operative technique All procedures were performed by 1 of us (B.J.A.) or under his direct supervision, with the patient under general anesthesia. All patients received a single dose of a broadspectrum antibiotic for coverage and deep vein thrombosis prophylaxis. A 10-mm 30° laparoscope was used. Laparoscopic gastric bypass. For laparoscopic gastric bypass (LGB), a 30-cm3 gastric pouch was fashioned using an EndoGIA stapler with 45– 60-mm blue (3.5 mm thickness) staples (Covidien, Norwalk, CT). The bypass was fashioned using an antecolic antegastric Roux-en-Y technique (n ⫽ 29) with the alimentary limb and the total length of the bypass made to measure 100 –200 cm and 150 –300 cm respectively. A minigastric bypass (n ⫽ 19) was created, bringing up an antecolic antegastric loop of 150 –250 cm. An approximately 1-cm gastrojejunostomy was created to just accommodate a 32F orogastric tube. It has been our practice to construct all anastomoses using continuous Vicryl 3-0 suture (Johnson & Johnson, St. Stevens-Woluws, Belgium). The gastrojejunostomy was tested for leakage using methylene blue diluted in saline. If a leak was detected, it was oversewn and retested to confirm a watertight anastomosis and watertight staple lines. No abdominal drains or nasogastric tubes were placed. Laparoscopic adjustable gastric banding. For laparoscopic adjustable gastric banding (LAGB), the pars flaccida approach was used to deploy the AMI gastric band (AMI Agency for Medical Innovations, Feldkirch, Austria). An anterior gastric wrap over the band was fashioned using interrupted Ethibond sutures (Johnson & Johnson). The connecting tube was tunneled subcutaneously from the left mid-clavicular wound to the port that had been placed extrafascially at the mid-epigastric wound. The port was fixed to the fascia with 3 Ethibond sutures. Postoperative band inflation was performed with radiologic guidance. Definitions Patients with T2DM were those with an elevated fasting blood glucose level ⬎7 mmol/L (1 mmol/L equals 18 mg/ dL) or a 2-hour plasma glucose level of ⱖ11.1 mmol/L during an oral glucose tolerance test as defined by the American Diabetes Association. Remission of T2DM was defined as the discontinuation of all diabetic drug therapy and the return of blood glucose levels to normal (normal fasting blood glucose value 2.8 –5.8 mmol/L and normal glycosylated hemoglobin value 4.1– 6.5%]. Improvement was defined as the reduction in the number and/or dose of diabetic drug therapy with a reduction or maintenance of the blood glucose levels. Excess weight reflected weight greater than the upper limit of the normal BMI (i.e., BMI ⬎24.9 kg/m2). The percentage of excess weight loss (%EWL) was calculated as 100 ⫻ (total postoperative weight loss in kilograms/excess weight in kilograms). Data management The data were collected prospectively using specifically designed audit sheets and were entered into a computer database and updated regularly with the follow-up information. Postoperative care Statistical analysis After LGB, the patients were instructed to drink 60 mL/hr water orally on the evening of surgery and received The results are presented as the mean ⫾ standard deviation or as ranges. Changes within the same group were N. Hamza et al. / Surgery for Obesity and Related Diseases 7 (2011) 691– 696 compared using the nonparametric Wilcoxon-matched pairs test for non-normally distributed parameters and the paired samples t test for normally distributed parameters. Comparisons between the groups and univariate analysis were performed using the Mann-Whitney U test for the non-normally distributed parameters, the independent samples t test for normally distributed parameters, and the chi-square test and Fisher’s exact test for categorical variables. A 2-tailed P ⱕ.05 was considered statistically significant. Cox multiple stepwise regression analysis was applied to determine the factors independently predictive of remission of T2DM after bariatric surgery and of the factors predictive of the %EWL, reporting their relative risk and 95% confidence intervals. Results Patients From 2002 to 2007, 487 patients (401 women, 82.3%) with a mean age of 41.6 ⫾ 10.0 years underwent laparoscopic bariatric surgery (LGB, n ⫽ 289 and LAGB, n ⫽ 198). Of these patients, 74 (15.2%) had T2DM (49 women; age 47.9 ⫾ 9.1 years; preoperative BMI 51.6 ⫾ 8.2 kg/m2). Of these 74 patients, 48 had undergone LGB and 26 had undergone LAGB; 20 (27%) required insulin, 52 (70.3%) required oral hypoglycemic agents (OHAs), and 2 (2.7%) controlled their T2DM with diet. The mean age for the LGB and LAGB groups was 46 ⫾ 8 years and 52 ⫾ 10 years, respectively (P ⫽ .008). Patients with impaired glucose tolerance test results were excluded from the study. Operative outcomes All procedures were completed laparoscopically. The postoperative hospital stay was 2.1 ⫾ .95 days and 1.7 ⫾ 1.14 days in the LGB and LAGB groups, respectively (P ⫽ .034). Complications developed in 4 patients (8.3%) and 1 693 patient (3.9%) after LGB and LAGB, respectively (P ⫽ .452). In the LGB group, 1 patient died of aspiration pneumonia at reoperation and laparoscopic release of small bowel obstruction within a paraumbilical hernia (mortality rate .3% of 289 LGB patients), 1 patient developed a left subphrenic abscess due to an infected hematoma that was drained laparoscopically, and 1 patient developed purulent peritonitis due to a gastrojejunal anastomotic leak and recovered after laparotomy and drainage (leak rate .7% of 289 LGB patients). One patient in each group developed a wound infection that resolved with antibiotics. Metabolic outcomes and effect of bariatric procedure The follow-up data for 2 patients in the LGB group and 1 patient in the LAGB group were not available. The patients were followed up for 16.9 ⫾ 11.6 months, during which their BMI decreased from 52 ⫾ 8 to 37 ⫾ 7 kg/m2. The %EWL of the patients with T2DM remission (41% of the patients) was significantly greater than that of the patients with only T2DM improvement (59% of the patients) after bariatric surgery (62.5% ⫾ 19.5% versus 51.4% ⫾ 15.9%, respectively; P ⫽ .009). The patients who underwent LGB and LAGB had a comparable preoperative BMI (52 ⫾ 8 versus 51 ⫾ 8 kg/m2, respectively; P ⫽ .508). Although the LAGB patients had a significantly longer follow-up period (23 versus 13.4 months, P ⫽ .001), the %EWL (59.4% ⫾ 16.2% versus 48.8% ⫾ 20.2%, P ⫽ .031) and the frequency of remission of T2DM (50% versus 24%, P ⫽ .034) were significantly greater after LGB. Figure 1 shows that a greater percentage of patients experienced remission of T2DM after LGB than after LAGB at various follow-up points. The mean interval between surgery and the discontinuation or reduction in antidiabetic medication was 6.6 ⫾ 6.8 and 16.4 ⫾ 10.5 months in the LGB and LAGB groups, respectively (P ⬍ .001). The remission of T2DM was more likely for the patients Fig. 1. Graph demonstrating accumulative percentage of patients with T2DM remission after LGB and LAGB at various follow-up points. ⴱSignificantly different. 694 N. Hamza et al. / Surgery for Obesity and Related Diseases 7 (2011) 691– 696 Table 1 Univariate analysis of factors that might influence T2DM remission rate after bariatric surgery Factor Age (y) ⱕ45 (n ⫽ 32) ⬎45 (n ⫽ 39) Gender Male (n ⫽ 22) Female (n ⫽ 49) T2DM duration (y) ⬍5 (n ⫽ 34) ⱖ5 (n ⫽ 30) Preoperative BMI (kg/m2) ⬍50 (n ⫽ 31) ⱖ50 (n ⫽ 40) Preoperative medication Insulin (n ⫽ 21) OHA (n ⫽ 48) %EWL ⬍55% (n ⫽ 34) ⱖ55% (n ⫽ 36) Follow-up duration (mo) ⱕ15 (n ⫽ 35) ⬎15 (n ⫽ 36) Procedure LGB (n ⫽ 46) LAGB (n ⫽ 25) T2DM remission (n) Yes No 18 (56.3) 12 (30.8) 14 (43.7) 27 (69.2) P value to be independent predictors of postoperative T2DM remission (Table 2), and LGB, longer follow-up, and female gender were independent predictors of a greater %EWL (Table 2). .058 Discussion .994 9 (40.9) 20 (40.8) 13 (59.1) 29 (59.2) 14 (41.2) 14 (46.7) 20 (58.8) 16 (53.3) 14 (45.2) 15 (38) 17 (54.8) 25 (62) 5 (24) 23 (48) 16 (76) 25 (52) 8 (23.5) 21 (58.3) 26 (76.5) 15 (41.7) 11 (31.4) 18 (50) 24 (68.6) 18 (50) 23 (50) 6 (24) 23 (50) 19 (76) .665 .522 .060 .082 .003 .115 .034 T2DM ⫽ type 2 diabetes mellitus; BMI ⫽ body mass index; OHA ⫽ oral hypoglycemic agent; %EWL ⫽ percentage of excess weight loss; LGB ⫽ laparoscopic gastric bypass; LAGB ⫽ laparoscopic adjustable gastric banding. Data in parentheses are percentages. treated preoperatively with OHAs than for the insulintreated patients (48% versus 24%, respectively; P ⫽ .082). LGB was associated with greater rates of remission of insulin-treated T2DM (33% versus 0%, respectively; P ⫽ .262) and OHA-treated T2DM (60% versus 28%, respectively; P ⫽ .040) than LAGB. The fasting blood glucose level decreased significantly after bariatric surgery (10.2 ⫾ 4.8 versus 6.0 ⫾ 2.0 mmol/L, P ⫽ .001), regardless of the procedure (LGB, 10.4 ⫾ 5.5 versus 5.6 ⫾ 1.6 mmol/L, P ⫽ .008; LAGB 9.9 ⫾ 3.5 versus 6.9 ⫾ 2.6 mmol/L, P ⫽ .043). Also, the glycosylated hemoglobin level decreased significantly after bariatric surgery (7.8% ⫾ 1.8% versus 6.1% ⫾ 1.0%, P ⬍ .001), regardless of the procedure (LGB 7.9% ⫾ 1.9% versus 6.2% ⫾ 1.2%, P ⬍ .001; LAGB 7.5% ⫾ 1.8% versus 6.5% ⫾ 1.5%, P ⫽ .003). Factors influencing T2DM remission after bariatric surgery Univariate analysis identified a greater %EWL and LGB as factors significantly associated with a greater likelihood of T2DM remission after bariatric surgery (Table 1). Multivariate analysis revealed a greater %EWL and younger age The results of the present series have demonstrated the advantages of laparoscopic bariatric surgery and the superiority of LGB compared with LAGB in achieving the control and remission of T2DM. The mean fasting blood glucose and glycosylated hemoglobin levels decreased to within the normal range postoperatively, irrespective of the bariatric procedure used, and T2DM underwent remission or improvement in 100% of the patients after LGB and LAGB. Despite the significantly longer follow-up period after LAGB, the %EWL was significantly greater after LGB, with an associated significantly greater rate of T2DM remission. The %EWL for the patients with T2DM remission was significantly greater than that for those in whom T2DM did not undergo remission. Younger age and a greater %EWL were independent predictors of T2DM remission. Each additional 12 years of age reduced the chance of T2DM remission by 20%, and each additional 10% EWL increased the chance of T2DM remission by an additional 8%. Multivariate analysis identified LGB, longer follow-up duration, and female gender as independent predictors of a greater %EWL. It has been recognized that the rate of T2DM remission is closely related to the %EWL achieved after bariatric surgery, whether bypass [1,3] or adjustable gastric banding [7,9,10]. The results from the present series have confirmed this observation and found %EWL to be an independent predictor of T2DM remission. Previous studies have shown that the %EWL is predictive of a major increase in insulin sensitivity among diabetic and nondiabetic patients after both LGB and LAGB [9,11] and that it was strongly associated with improvement in pancreatic ␤-cell function among those with T2DM [9,12]. Table 2 Multivariate analysis of predictive factors of T2DM remission and of %EWL after bariatric surgery Predictor T2DM remission %EWL Age %EWL LGB Follow-up duration Male gender† Relative risk* .008/%EWL ⫺.017/y 13.717 .587/mo ⫺17.283 95% CI .002 to .014 ⫺.029 to ⫺.005 5.696 to 21.737 .259 to .915 ⫺24.929 to ⫺9.637 P value .007 .005 .001 .001 ⬍.001 CI ⫽ confidence interval; other abbreviations as in Table 1. * Indicates effect of “1-unit” difference in independent variable on dependent variable. † Male patients were less likely to lose weight. N. Hamza et al. / Surgery for Obesity and Related Diseases 7 (2011) 691– 696 In our series, the LAGB patients had the advantage of a significantly longer duration of postoperative follow-up compared with the LGB patients. We identified a longer follow-up duration as an independent predictor of a greater %EWL after bariatric surgery. Nonetheless, LAGB achieved an inferior rate of T2DM remission compared with LGB. The significantly older age of our LAGB patients might have contributed to this finding, because we found younger age to be an independent predictor of postoperative T2DM remission. Dixon and O’Brien [13] found older age to be a predictor of inferior %EWL after LAGB, and Sugerman et al. [3] observed that younger patients had better glycemic control after LGB. Not withstanding the influence of age on the rate of T2DM remission, our multivariate analysis identified LGB, and not LAGB, as an independent factor influencing the only postoperative predictor of T2DM remission (i.e., %EWL). The LGB patients were 14 times more likely to achieve an additional %EWL than the LAGB patients. In their study of the factors that predicted for the %EWL after LAGB, Dixon and O’Brien [13] found T2DM and insulin resistance to be independent predictors of inferior results. Also, our multivariate analysis revealed that male patients were 17 times less likely to achieve an additional %EWL compared with female patients. Other factors, such as the duration and severity of the T2DM, might play a role in influencing the rate of disease remission after bariatric surgery. Schauer et al. [1] noted a greater rate of immediate medication independence after LGB among patients with a history of T2DM of ⬍5 years compared with those who had had T2DM for ⬎10 years (34% versus 18%, P ⬍ .003) and in those patients taking OHAs alone compared with those requiring insulin (57% versus 23%, P ⬍ .001). Our univariate analysis found neither the T2DM duration nor the type of preoperative diabetic therapy of significant influence on the rate of T2DM remission after surgery. Sample size calculations, however, revealed that a type II error might have contributed to the former (106 patients would have been needed to show a statistically significant difference regarding the duration of T2DM) but not the latter, for which the sample size was more than sufficient. A meta-analysis [14] and a more recent systematic review of the published data [15] reported a greater remission rate of T2DM with gastric bypass than with gastric banding (83.8% versus 62.7% and 78% versus 50% at 1 year, respectively). In their recent review, Favretti et al. [16] acknowledged that gastric bypass resulted in greater weight loss than LAGB in the first 2 years. However, although they claimed a less marked difference between the 2 procedures at 3 years postoperatively and beyond, they did not support this statement with outcomes from a methodologic analysis. In contrast, Cottam et al. [17] reported superior weight loss and T2DM remission (78% versus 50%, P ⫽ .010) at 3 years of follow-up when LGB was compared with LAGB in a case-controlled matched-pair cohort study, and Angrisani 695 et al. [18] reported a significantly greater %EWL at 5 years of follow-up from a randomized trial that compared LGB and LAGB. We believe that the relatively lower remission rate of T2DM in our series compared with these reports, despite the similar %EWL achieved in the patients with T2DM was multifactorial and might reflect, in part, the relatively short follow-up, the older age of our patients compared with those reported by some others [3,10,19], and that the diabetic management was performed by a wide range of general practitioners and diabetologists not with the bariatric surgery service, some of whom might not have actively pursued justifiable discontinuation of antidiabetic drug therapy, a prerequisite that was inherent to the definition of remission. The interval between surgery and the discontinuation or reduction of antidiabetic therapy seemed to vary considerably with the type of bariatric procedure undertaken. Approximately 39% of our patients had their insulin and/or OHAs discontinued or reduced by their diabetologist or general practitioner within the initial 3 months of LGB, well before considerable weight loss was achieved. In contrast, only 4.6% of patients who underwent LAGB had their medication discontinued or reduced. These finding are in line with those observed by Schauer et al. [1] who reported discontinuation of all diabetic medications for 30% of patients after LGB immediately after hospital discharge. The postoperative changes in gut hormones such as decreased suppression of plasma ghrelin (a gastroenteric appetitestimulating peptide hormone), increased secretion of incretins (gut peptides that stimulate insulin secretion postprandially) such as gastric inhibitory peptide, glucagon-like peptide-1 secretion, and polypeptide YY, and greater insulin sensitivity and a reduction in calorie intake are factors that might explain this early diabetic response to gastric bypass [20 –22]. In a comparative study, Lin et al. [23] found that LGB, but not LAGB, ameliorated T2DM remission within 6 months of surgery, and this was achieved by early augmentation of ␤-cell function at 1 month and attenuation of peripheral insulin resistance that was induced by weight loss at 6 months. The observed remarkable remission T2DM rates after bariatric surgery have led a number of centers to lower the BMI threshold for surgery to ⬍35 kg/m2, with good results. In a randomized controlled trial of LAGB versus conventional medical therapy in patients with recently diagnosed T2DM that included patients with a BMI of 30 –35 kg/m2, Dixon et al. [7] reported a 76% T2DM remission rate at 2 years, with a mean %EWL of 62.5%. Lee et al. [24] reported T2DM remission rate of 76.5% in patients with a BMI of ⬍35 kg/m2 at 1 year of follow-up after laparoscopic minigastric bypass. We have extended the application of LAGB to include patients with T2DM and a BMI as low as 32 kg/m2. 696 N. Hamza et al. / Surgery for Obesity and Related Diseases 7 (2011) 691– 696 Conclusion The results of our study have shown that laparoscopic bariatric surgery is effective in achieving remission or improvement of T2DM. However, LGB appears to offer a greater %EWL and chance of T2DM remission than LAGB and within 2 years of surgery. [12] [13] [14] References [1] Schauer PR, Burguera B, Ikramuddin S, et al. Effect of laparoscopic Roux-en Y gastric bypass on type 2 diabetes mellitus. Ann Surg 2003;238:467– 85. [2] Roberts S. Working together for better diabetes care: clinical case for change. London, UK: NHS Department of Health; 2007. [3] Sugerman HJ, Wolfe LG, Sica DA, Clore JN. Diabetes and hypertension in severe obesity and effects of gastric bypass-induced weight loss. Ann Surg 2003;237:751– 8. [4] Ballantyne GH, Belsley S, Stephens D, et al. Bariatric surgery: low mortality at a high-volume centre. Obes Surg 2008;18:660 –7. [5] White S, Brooks E, Jurikova L, Stubbs RS. Long-term outcomes after gastric bypass. Obes Surg 2005;15:155– 63. [6] Schernthaner G, Morton JM. Bariatric surgery in patients with morbid obesity and type 2 diabetes. Diabetes Care 2008;31(Suppl 2):S297– 302. [7] Dixon JB, O’Brien PE, Playfair J, et al. Adjustable gastric banding and conventional therapy for type 2 diabetes: a randomized controlled trial. JAMA 2008;299:316 –23. [8] Moorehead MK, Ardelt-Gattinger E, Lechner H, Oria HE. The validation of the Moorehead-Ardelt quality of life questionnaire II. Obes Surg 2003;13:684 –92. [9] Dixon JB, Dixon AF, O’Brien PE. Improvements in insulin sensitivity and beta-cell function (HOMA) with weight loss in the severely obese: homeostatic model assessment. Diabet Med 2003;20:127–34. [10] Ponce J, Haynes B, Paynter S, et al. Effect of lap-band-induced weight loss on type 2 diabetes mellitus and hypertension. Obes Surg 2004;14:1335– 42. [11] Ballantyne GH, Wasielewski A, Saunders JK. The surgical treatment of type II diabetes mellitus: changes in HOMA insulin resistance in the first year following laparoscopic Roux-en-Y gastric bypass [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] (LRYGB) and laparoscopic adjustable gastric banding (LAGB). Obes Surg 2009;19:1297–303. Gumbiner B, Polonsky KS, Beltz WF, et al. Effects of weight loss and reduced hyperglycemia on the kinetics of insulin secretion in obese non-insulin dependent diabetes mellitus. J Clin Endocrinol Metab 1990;70:1594 – 602. Dixon JB, O’Brien PE. Selecting the optimal patient for LAP-BAND placement. Am J Surg 2002;184:17S–20S. Buchwald H, Avidor Y, Braunwald E, et al. Bariatric Surgery: a systematic review and meta-analysis. JAMA 2004;292:1724 –37. Tice JA, Karliner L, Walsh J, Petersen AJ, Feldman MD. Gastric banding or bypass? A systematic review comparing the two most popular bariatric procedures. Am J Med 2008;121:885–93. Favretti F, Ashton D, Busetto L, Segato G, De Luca M. The gastric band: first-choice procedure for obesity surgery. World J Surg 2009; 33:2039 – 48. Cottam DR, Atkinson J, Anderson, Grace B, Fisher B. A casecontrolled matched-pair cohort study of laparoscopic Roux-en-Y gastric bypass and lap-band patients in a single US center with three-year follow-up. Obes Surg 2006;16:534 – 40. Angrisani L, Lorenzo M, Borrelli V. Laparoscopic adjustable gastric banding versus Roux-en-Y gastric bypass: 5-year results of a prospective randomized trial. Surg Obes Relat Dis 2007;3:127–33. Mottin CC, Vontobel Padoin A, Schroer CE, Barancelli FT, Glock L, Repetto G. Behavior of type 2 diabetes mellitus in morbid obese patients submitted to gastric bypass. Obes Surg 2008;18:179 – 81. Bose M, Oliván B, Teixeira J, Pi-Sunyer FX, Laferrère B. Do incretins play a role in the remission of type 2 diabetes after gastric bypass surgery: what are [stet] the evidence? Obes Surg 2009;19:217–29. Oliván B, Teixeira J, Bose M, et al. Effect of weight loss by diet or gastric bypass surgery on peptide YY3-36 levels. Ann Surg 2009; 249:948 –53. Vetter ML, Cardillo S, Rickels MR, Iqbal N. Narrative review: effect of bariatric surgery on type 2 diabetes mellitus. Ann Intern Med 2009;150:94 –103. Lin E, Davis SS, Srinivasan J, et al. Dual mechanism for type 2 diabetes resolution after Roux-en-Y gastric bypass. Am Surg 2009; 75:498 –502. Lee WJ, Wang W, Lee YC, Huang MT, Ser KH, Chen JC. Effect of laparoscopic mini-gastric bypass for type 2 diabetes mellitus: comparison of BMI ⬎35 and ⬍35 kg/m2. J Gastrointest Surg 2008;12: 945–52.