Hormone changes and diabetes resolution after biliopancreatic diversion and laparoscopic sleeve gastrectomy: a comparative prospective study

Surgery for Obesity and Related Diseases 9 (2013) 667–678

Original article

Hormone changes and diabetes resolution after biliopancreatic diversion and laparoscopic sleeve gastrectomy: a comparative prospective study Marina Tsoli, M.D.c, Aikaterini Chronaiou, M.D.c, Ioannis Kehagias, M.D.a, Fotis Kalfarentzos, M.D., F.A.C.S.a,b, Theodore K. Alexandrides, M.D.c,* b

a Department of Surgery, School of Medicine, University of Patras, Greece Nutrition Support and Morbid Obesity Unit, Department of Surgery, School of Medicine, University of Patras, Greece c Department of Internal Medicine, Division of Endocrinology, School of Medicine, University of Patras, Greece Received May 30, 2012; accepted December 9, 2012

Abstract

Background: Biliopancreatic diversion (BPD) is the most effective bariatric procedure in terms of weight loss and remission of diabetes type 2 (T2DM), but it is accompanied by nutrient deficiencies. Sleeve gastrectomy (SG) is a relatively new operation that has shown promising results concerning T2DM resolution and weight loss. The objective of this study was to evaluate and compare prospectively the effects of BPD long limb (BPD) and laparoscopic SG on fasting, and glucosestimulated insulin, glucagon, ghrelin, peptide YY (PYY), and glucagon-like peptide-1 (GLP-1) secretion and also on remission of T2DM, hypertension, and dyslipidemia in morbidly obese patients with T2DM. Methods: Twelve patients (body mass index [BMI] 57.6 ⫾ 9.9 kg/m2) underwent BPD and 12 (BMI 43.7 ⫾ 2.1 kg/m2) underwent SG. All patients had T2DM and underwent an oral glucose tolerance test (OGTT) before and 1, 3, and 12 months after surgery. Results: BMI decreased more after BPD, but percent excess weight loss (%EWL) was similar in both groups (P ¼ .8) and T2DM resolved in all patients at 12 months. Insulin sensitivity improved more after BPD than after SG (P ¼ .003). Blood pressure, total and LDL cholesterol decreased only after BPD (P o .001). Triglycerides decreased after either operation, but HDL increased only after SG (P o .001). Fasting ghrelin did not change after BPD (P ¼ .2), but decreased markedly after SG (P o .001). GLP-1 and PYY responses during OGTT were dramatically enhanced after either procedure (P ¼ .001). Conclusions: SG was comparable to BPD in T2DM resolution but inferior in improving dyslipidemia and blood pressure. SG and BPD enhanced markedly PYY and GLP-1 responses but only SG suppressed ghrelin levels. (Surg Obes Relat Dis 2013;9:667–678.) r 2013 American Society for Metabolic and Bariatric Surgery. All rights reserved.

Keywords:

Biliopancreatic diversion; Sleeve gastrectomy; Weight loss; Diabetes mellitus; Hypertension; Dyslipidemia; Ghrelin; PYY; GLP-1; Insulin; Glucagon

Biliopancreatic diversion (BPD) is considered the most effective surgical procedure in terms of weight loss, resolution of the components of metabolic syndrome and *

Correspondence: Theodore K. Alexandrides, M.D., Division of Endocrinology, Department of Internal Medicine, School of Medicine, University of Patras, Patras University Hospital, Patras 26500, Greece. E-mail: [email protected]

remission of diabetes type 2 in obese patients, resulting in normalization of glucose tolerance and insulin sensitivity and restoration of normal b-cell response to glucose [1–3]. It has been proposed that the improvement in glycemic control after BPD may be due to changes in gastrointestinal hormones [4,5]. However, iron, calcium, and vitamin deficiencies may occur after BPD, and regular follow-up and administration of supplements are mandatory [6].

1550-7289/13/$ – see front matter r 2013 American Society for Metabolic and Bariatric Surgery. All rights reserved. http://dx.doi.org/10.1016/j.soard.2012.12.006

668

M. Tsoli et al. / Surgery for Obesity and Related Diseases 9 (2013) 667–678

Laparoscopic sleeve gastrectomy (SG) has been applied in patients with body mass index (BMI) r50 kg/m2, and the weight loss after SG is comparable to Roux-en-Y (RYGB) with low complication and mortality rates [7,8]. Furthermore, sleeve gastrectomy results in a high rate of remission of diabetes and metabolic syndrome, comparable to that of RYGB [8–10]. The aim of this study was to compare directly and prospectively the effectiveness of SG and BPD long limb (BPD), a variant of BPD frequently performed in our Morbid Obesity Unit [11], on glucose stimulated insulin, glucagon, ghrelin, peptide-YY (PYY), and glucagon-like peptide-1 (GLP-1) secretion and also on weight loss, glucose homeostasis, dyslipidemia, and hypertension in morbidly obese patients with diabetes type 2.

Methods Human studies All human studies were performed according to the Declaration of Helsinki. The study was approved by the Research and Ethics Committee of Patras University Hospital. All patients were informed of the risks and benefits of each operation and gave a written consent. The study was registered in the National Institutes of Health web site (www.clincaltrials.gov). The Clinical Trial Registration Number is NCT 01481675. Twenty-four morbidly obese patients with diabetes mellitus type 2 were included in the study. The exclusion criteria included pregnancy, substance abuse, chronic medical or psychiatric illness, and previous gastrointestinal surgery. The diagnosis of diabetes was established when fasting serum glucose concentration was Z126 mg/dL on 2 occasions or the 2-hour glucose concentration was Z200 mg/dL after a 75-g oral glucose tolerance test. Eighteen of the patients were treated with oral hypoglycaemic agents. None of the patients was on insulin therapy. The mean preoperative duration of diabetes was 17.7 ⫾ 6.8 months, and the mean hemoglobin A1c (HbA1c) level was 6.8% ⫾ 1%. Twelve patients (5 males and 7 females), 42.3 ⫾ 11.9 (range 31–51) years old, with BMI 57.6 ⫾ 9.9 (range 47.1– 83.7) kg/m2 underwent BPD, with open surgery, and 12 (4 males and 8 females), 40.3 ⫾ 8.5 (range 20–57) years old, with BMI 43.7 ⫾ 2.1 (range 38.7–47) kg/m2 underwent laparoscopic SG. In the Morbid Obesity Unit of our institution, the assignment of a patient to BPD is based on preoperative BMI 450 kg/m2, and the assignment of a patient to SG is based on preoperative BMI r50 kg/m2. However, 2 patients with BMI 450 kg/m2 at the first evaluation, presented with BMI 47.1 and 47.6 kg/m2, respectively, on the day of admission for the operation. In our institution, the waiting list is long, and the patient must wait usually more than 8 months for the operation. During

this period, the patient receives dietary advice in an effort to reduce his weight preoperatively. BPD consisted of a 70 ⫾ 10 mL gastric pouch, an alimentary limb composed of 400 cm, a common limb of 100 cm, and a biliopancreatic limb, the remainder of the small intestine [11]. All BPD procedures were carried out with open surgery. In SG, the vascular supply of the greater curvature of the stomach was divided from the left crus of the diaphragm to the pylorus. The dissection of the stomach began with a linear stapler commencing 3 cm from the pylorus, close to a 33F bougie (introduced orally by the anesthetist), up to the incisura angularis. A gastric sleeve tube of 40 to 60 mL in volume remained, and approximately 85% of the stomach was excised [8]. All SG procedures were successfully concluded laparoscopically with no conversion to open surgery. All patients underwent complete evaluation the day before and 1, 3, and 12 months after the operation. The evaluation included clinical and anthropometric parameters, nutritional behavior, medications and blood sampling for glucose, triglycerides, and cholesterol, and other laboratory tests. All patients underwent a 75-g oral glucose tolerance test (OGTT) preoperatively and at every visit during the follow-up. All patients had discontinued the antidiabetic drugs 3 weeks before the preoperative OGTT. None of the patients was taking a thiazolidinedione. Blood samples were collected after an overnight fast and at 30, 60, and 120 minutes after glucose ingestion for the measurement of glucose, insulin, PYY, GLP-1, ghrelin, and glucagon. Weight loss evaluation was based on postoperative BMI and percent of excess weight loss (%EWL). Ideal weight was determined according to the Metropolitan Life Insurance Company 1983 height/ weight tables. Insulin resistance was approximated using the homeostatic model assessment for insulin resistance (HOMA IR). The following formula was used in the calculation: HOMA IR ¼ [fasting glucose (millimoles per liter)  fasting insulin (mU/mL)] / 22.5) [12]. Hormone assays Blood samples were collected in prechilled EDTA containing tubes for ghrelin and GLP-1 and in heparincoated tubes containing 5000 kallikrein inhibitor units of aprotinin* for PYY. Dipeptidyl peptidase-IV inhibitor (10 mL of inhibitor per milliliter of blood) was added immediately in GLP-1 samples to avoid degradation. All samples were immediately centrifuged at 41C and stored at 701C until assayed. Insulin was measured with an electrochemiluminescence immunoassay (Elecsys 2010, Roche Diagnostics, GmbH, Mannheim, Germany). The assay is specific for insulin and does not recognize proinsulin and insulin-like growth factor-1. Intra and interassay coefficients of variation were 2.1% and * Trademark: Trasylols (Bayer HealthCare Pharmaceuticals Inc., Leverkusen, Germany).

Diabetes Resolution and Hormone Changes After BPDLL and LSG / Surgery for Obesity and Related Diseases 9 (2013) 667–678

669

2.8%, respectively, and the sensitivity was .2 mU/mL. Total PYY was assayed using a commercial radioimmunoassay (RIA) kit (Millipore Corporation, MA). It utilizes an antibody that recognizes both the 1–36 and 3–36 forms of human PYY. The sensitivity of the assay was 10 pg/mL, and the coefficient of intraassay variation was o9%. GLP-1 was measured using a commercial RIA kit (Millipore Corporation, MA). The antibody used in this assay binds specifically to the C-terminal portion of GLP-1, both of the amidated and the nonamidated forms. The lower limit of detection was 3 pmoles/liter. Total plasma ghrelin was measured using a commercial RIA (Millipore Corporation, MA), which utilizes an antibody specific for total ghrelin and does not require the presence of the octonyl group on serine 3. The sensitivity of the assay was 93 pg/mL, and the intraassay variation was o10%. Glucagon was assayed using a commercial RIA kit (Linco Research, MI). The lowest limit of detection was 20 pg/mL, and the coefficient of intraassay variation o7%. Assay of the samples of both groups of patients was carried out in the same assay to minimize the between groups variation of measurements.

the 2 groups (P ¼ .33 and P ¼ .48 respectively). Fasting glucose and hormone levels are shown in Table 2. Glucose levels during OGTT were not significantly different between the 2 groups (P ¼ .078, for the interaction) (Fig. 1). The shape of the insulin curve during OGTT was different between the groups (P ¼ .039, for the interaction). Peak insulin response occurred at 60 minutes in the BPD group and at 120 minutes in the SG group (Fig. 1). The BPD group had a trend for higher fasting glucagon levels than the SG group (P ¼ .058) (Table 2) and also a greater glucagon AUC0–120, during OGTT compared with the SG group (Table 3), but at the same time, glucagon levels in response to glucose were partially suppressed (P ¼ .045), while in the SG group the change was not significant (P ¼ .2), (Fig. 1). During OGTT, ghrelin changes were not significant in either group (BPD, P ¼ .64 and SG, P ¼ .77, ANOVA single factor) (Fig. 2) as well as the PYY changes (BPD, P ¼ .38 and SG, P ¼ .27) (Fig. 3). Both groups exhibited a blunted GLP-1 response to glucose ingestion (BPD, P ¼ .01 and SG, P ¼ .06) (Fig. 3).

Appetite assessment

Postoperatively

The patients were interviewed at every visit, focusing on appetite, satiety, and nausea using visual analogue scales (VAS). VAS was based on the Edmonton Symptom Assessment System VAS for appetite, which has been used extensively for the assessment of cancer-induced cachexia and has been shown to be reliable [13]. The VAS consisted of 3 questions assessing appetite and satiety changes and the frequency and intensity of nausea.

Weight and BMI. BMI decreased more in the BPD group (P o .001), but there was no difference in % EWL between the 2 groups at 12 months (P ¼ .8); at 3 months though, % EWL was greater in the SG group (P ¼ .031) (Table 1). Blood pressure, serum lipids and albumin. Total and LDL cholesterol decreased markedly after BPD with no significant changes after SG (Table 1). HDL cholesterol increased after SG (P o .001); after BPD though, it decreased transiently and returned to baseline levels at 12 months (P ¼ .5). Triglycerides decreased markedly after either procedure. The decrease in systolic blood pressure was significant only after BPD, and the decrease in diastolic was not significant in either group (Table 1). A small nonsignificant decrease of albumin was observed in the BPD group (Table 1). One patient of this group had an albumin level of 3 g/dL at 12 months, but all of the other patients had levels higher than 3.9 g/dL. Diabetes mellitus. All patients discontinued the antidiabetic medications postoperatively. At 1 month, 1 patient in the BPD and 3 in the SG group had diabetes, based on OGTT; however, at 12 months, all patients had normal OGTT (Fig. 1), but the BPD group had lower fasting glucose levels than the SG group (Table 2). Insulin and glucagon levels. Fasting insulin and insulin resistance decreased dramatically 1 month after either procedure, however at 12 months, the BPD group had lower insulin levels and HOMA IR than the SG group (Table 2). The AUC0–120, of insulin during OGTT decreased significantly in both groups (Table 3, Fig. 1). Twelve months after the operation peak insulin response, during OGTT, occurred at 30 minutes in the SG and at 60 minutes in the BPD group,

Statistical analysis Data are expressed as mean ⫾ standard deviation, unless otherwise noted. The AUC was calculated using the trapezoidal rule. Comparisons between groups were performed by two-tailed t test or Mann-Whitney U test depending on the distribution of data. Distribution of parameters was tested with the Kolmogorov-Smirnov test. Repeated measures oneway ANOVA was used to evaluate the changes of parameters during follow up and during OGTT for each type of surgery. The two-way mixed ANOVA model was used to examine the interaction between the type of surgery and time and to make between-group comparisons. Statistical significance was set at P o .05. Results Preoperatively The demographic characteristics of the 2 groups are shown in Table 1. The BPD group had higher BMI than the SG group. There was no significant difference in preoperative diabetes duration and HbA1c levels between

670

M. Tsoli et al. / Surgery for Obesity and Related Diseases 9 (2013) 667–678

Table 1 Anthropometric and biochemical parameters and systolic and diastolic blood pressure before and 1, 3 and 12 months after BPD and SG Pre (N ¼ 12) BMI (kg/m2) BPD 57.6 ⫾ 9.9 SG 43.7 ⫾ 2.1 P o.001 % EWL BPD SG P Albumin (g/dL) BPD 4.32 ⫾ 0.35 SG 4.23 ⫾ 0.21 P .409 Cholesterol (mg/dL) BPD 214 ⫾ 30 SG 204 ⫾ 25 P .352 LDL cholesterol (mg/dL) BPD 138 ⫾ 26 SG 122 ⫾ 32 P .190 Triglycerides (mg/dL) BPD 181 ⫾ 73 SG 210 ⫾ 91 P .401 HDL cholesterol (mg/dL) BPD 40.6 ⫾ 8.1 SG 40.5 ⫾ 7.6 P .979 Systolic BP (mm Hg) BPD 142 ⫾ 24 SG 134 ⫾ 17 P .386 Diastolic BP (mm Hg) BPD 83.8 ⫾ 9.3 SG 83.8 ⫾ 13 P .873

1 month (N ¼ 12)

3 months (N ¼ 12)

12 months (N ¼ 12)

P**

Py

51.9 ⫾ 9.2* 39 ⫾ 2* o.001

45.3 ⫾ 8.3* 34.1 ⫾ 2.1* o.001

32.4 ⫾ 4.8* 27.9 ⫾ 3.2* .014

o0.001 o0.001

o0.001

18.6 ⫾ 3.6 22 ⫾ 6.2 .107

37.8 ⫾ 6.7* 44.3 ⫾ 7.1* .031

73.4 ⫾ 9.8* 75 ⫾ 13.6* .802

o0.001 o0.001

.513

4.32 ⫾ 0.29 4.35 ⫾ 0.47 .836

4.08 ⫾ 0.39 4.52 ⫾ 0.35 o.01

4.13 ⫾ 0.40 4.44 ⫾ 0.36 .055

.219 .221

150 ⫾ 20* 178 ⫾ 43 .063

132 ⫾ 17* 191 ⫾ 29 o.0001

122 ⫾ 28* 194 ⫾ 32 o.0001

o0.001 .25

o0.001

81 ⫾ 21* 110 ⫾ 34 .025

76 ⫾ 13* 123 ⫾ 25 o.0001

64 ⫾ 19* 116 ⫾ 29 o.0001

o0.001 .637

o0.001

177 ⫾ 38 148 ⫾ 55* .161

129 ⫾ 23* 122 ⫾ 27* .495

95 ⫾ 18* 110 ⫾ 33* .176

o0.001 o0.001

.118

33.4 ⫾ 5.2* 38.4 ⫾ 8.7 .104

30.7 ⫾ 4.2* 44.3 ⫾ 9.4 o.001

42.5 ⫾ 11.1 57.2 ⫾ 8.6* .0016

o0.001 o0.001

o0.001

120 ⫾ 14* 120 ⫾ 15* .787

120 ⫾ 16* 128 ⫾ 18 .344

121 ⫾ 11* 120 ⫾ 17 .834

.003 .053

.35

78.8 ⫾ 9.6 80.8 ⫾ 9.3 .604

75.8 ⫾ 9.7* 81.7 ⫾ 8.1 .953

78.3 ⫾ 7.2 78 ⫾ 8.1 .150

.197 .541

.945

.457

BMI ¼ Body mass index; BP ¼ blood pressure; BPD ¼ biliopancreatic diversion long limb; EWL ¼ excess weight loss; HDL ¼ high density lipoprotein; LDL ¼ low density lipoprotein; P ¼ t test; Pre ¼ preoperatively; SG ¼ laparoscopic sleeve gastrectomy. Values represent mean ⫾ SD. * P o .05 compared with preoperative values. ** P Repeated measures analysis of variance. y P ¼ for the interaction between the surgery type and time; Pre ¼ preoperatively; SG ¼ laparoscopic sleeve gastrectomy.

although preoperatively the SG group had a more delayed response than the BPD group (Fig. 1). Fasting glucagon levels decreased progressively after SG but increased transiently 1 month after BPD and then declined progressively to levels lower than baseline (Table 2). Glucagon levels increased during OGTT 1 month after surgery in the SG group (P ¼ .013) and in both groups at 3 months (P ¼ .038 and P ¼ .013, respectively) (Table 2 and Fig. 1); however, at 12 months the glucagon changes were not significant during OGTT (P ¼ .08 and P ¼ .13, respectively) (Fig. 1). The AUC0–120, of glucagon increased significantly at 1 month and 3 months and returned to preoperative values at 12 months in both groups (Table 3). Ghrelin. Fasting ghrelin levels did not change significantly after BPD (P ¼ .22), but decreased markedly and

persistently after SG (P o .001) (Table 2). Ghrelin suppression in response to glucose ingestion was not significant in both groups either before or after the operation when assessed with ANOVA single factor (Fig. 2). The AUC0– 120, of ghrelin decreased transiently 1 month after BPD and returned to preoperative values at 3 months and 12 months but decreased significantly and persistently after SG (Table 3). PYY. Fasting PYY levels increased after BPD (P ¼ .001) and decreased after SG (P ¼ .052) (Table 2), so that postoperatively, the BPD group had higher PYY levels than the SG group (Table 2). Postoperatively, in both groups the PYY response to glucose ingestion was dramatically enhanced (Fig. 2), and the PYY AUC0–120, during OGTT increased markedly (Table 3).

Diabetes Resolution and Hormone Changes After BPDLL and LSG / Surgery for Obesity and Related Diseases 9 (2013) 667–678

671

Table 2 Fasting glucose, insulin, PYY, GLP-1, ghrelin, glucagon, and HOMA IR values before and 1, 3 and 12 months after BPD and SG

Glucose (mg/dL) BPD SG P Insulin (mU/mL) BPD SG P HOMA IR BPD SG P PYY (pg/mL) BPD SG P GLP-1 (pM) BPD SG P Ghrelin (pg/mL) BPD SG P Glucagon (pg/mL) BPD SG P

Pre (N ¼ 12)

1 month (N ¼ 12)

3 months (N ¼ 12)

12 months (N ¼ 12)

P**

Py

156 ⫾ 48.8 143 ⫾ 38.6 .343

118 ⫾ 11.4* 104 ⫾ 15.5* .019

96.9 ⫾ 11.8* 96 ⫾ 10* .880

76.8 ⫾ 10.9* 90 ⫾ 9.4* .003

o.001 o.001

.076

34.2 ⫾ 27.4 41.3 ⫾ 19.3 .521

16.4 ⫾ 6.3* 19.4 ⫾ 9* .350

9.6 ⫾ 3.4* 12.1 ⫾ 5* .150

4.7 ⫾ 2.4* 8.3 ⫾ 3.6* .011

o.001 o.001

.768

13 ⫾ 9 14.2 ⫾ 6.1 .708

4.8 ⫾ 2* 4.9 ⫾ 2.2* .919

2.3 ⫾ 1* 2.8 ⫾ 1.1* .217

.9 ⫾ .5* 1.9 ⫾ .9* .003

o.001 o.001

102 ⫾ 44.6 98 ⫾ 38.1 .795

130 ⫾ 51.6 65 ⫾ 20.1* .001

133 ⫾ 29.1* 74 ⫾ 32.6 o.001

176 ⫾ 37.7* 80 ⫾ 21.2 o.001

.001 .003

o.001

59 ⫾ 36.8 64 ⫾ 38.2 .764

87 ⫾ 47.3* 70 ⫾ 68.2 .483

66 ⫾ 38.3 31.2 ⫾ 20.5* .012

63 ⫾ 43.1 24.5 ⫾ 21.8* .014

.74 .03

.054

559 ⫾ 82 639 ⫾ 189 .198

524 ⫾ 115 442 ⫾ 65* .048

529 ⫾ 86 462 ⫾ 53* .034

607 ⫾ 134 467 ⫾ 72* .006

.22 o.001

o.001

69 ⫾ 19.2 55 ⫾ 16.4 .058

83 ⫾ 19.5* 54 ⫾ 19.1 .001

61 ⫾ 16.7* 43 ⫾ 15.3* .013

51 ⫾ 14.3* 38 ⫾ 8.5* .014

.002 .007

.105

.754

BPD ¼ biliopancreatic diversion long limb; GLP-1 ¼ glucagon like peptide-1; HOMA IR ¼ homeostatic model assessment for insulin resistance; Pre ¼ preoperatively; PYY ¼ peptide YY; SG ¼ laparoscopic sleeve gastrectomy. Values represent mean ⫾ SD. * P o .05 compared with preoperative values (paired t test). ** P Repeated measures analysis of variance. y P ¼ the interaction between the surgery type and time. P ¼ t test.

GLP-1. Fasting GLP-1 levels did not change significantly after BPD but decreased after SG (Table 2). GLP-1 response to glucose ingestion was markedly enhanced (Fig. 2), and the GLP-1 AUC0–120, during OGTT increased markedly in the 2 groups postoperatively (Table 3). Appetite and satiety. Appetite was significantly (P o .001) and similarly (P ¼ .23) attenuated in both groups. After the third postoperative month, a partial regain of appetite was observed in both groups (Fig. 4). Satiety was significantly enhanced in both groups postoperatively (P o .001) with no significant difference between the 2 procedures (P ¼ .19) (Fig. 4). Aversion to food increased significantly (P o .05 for either group), but a progressive attenuation was observed with time. No significant nausea was observed postoperatively in either group. Discussion This is the first study to compare directly and prospectively BPD long limbs [11] with SG, in patients with diabetes mellitus type 2. We have adopted this variant of

BPD because of the low incidence of nutrient deficiencies and hypoalbuminemia and the excellent long-term results in regard to weight loss and resolution of diabetes, dyslipidemia, and hypertension [11]. This procedure resulted in 470% %EWL and 497% resolution of diabetes during the 8-year follow up [11]. BPD and SG resulted in similar appetite suppression and induction of satiety. Diabetes resolved in all patients, triglycerides decreased after either procedure, but only BPD resulted in a significant decrease of systolic blood pressure and total and LDL cholesterol. SG on the other hand, resulted in a significant rise of HDL cholesterol. These observations confirm and extend the findings of previous studies [1,3,7–11,14,15]. Malabsorptive procedures are considered the most effective in terms of weight loss and metabolic control, resulting in approximately 95% resolution of diabetes and dyslipidemia [1,3,11]. SG is a relatively new operation that is associated with remission of diabetes and metabolic syndrome at a rate similar to RYGB [10]. In a recent study, SG and RYGB resulted in similar improvement in glucose homeostasis and insulin secretion

672

M. Tsoli et al. / Surgery for Obesity and Related Diseases 9 (2013) 667–678

Fig. 1. Time courses of serum glucose, insulin, and glucagon during OGTT before and 1, 3, and 12 months after BPD and SG. Values represent mean ⫾ SEM. BPD ¼ biliopancreatic diversion long limb; OGTT ¼ oral glucose tolerance test; Pre ¼ preoperatively; SEM ¼ standard error of the mean; SG ¼ laparoscopic sleeve gastrectomy.

after a test meal, 1 week after surgery [9]. It has also been shown that SG results in rapid improvement in insulin sensitivity, independently of EWL [9,16]. In the present study, fasting glucose and insulin concentrations as well as HOMA IR were lower after BPD compared with SG at 12 months (Table 2). Concerning the mechanisms for improved insulin sensitivity, either procedure induces marked reduction in calorie intake, weight loss, and reduction in fat mass; BPD in addition causes significant fat malabsorption, and the reduced free fatty acid oxidation should enhance further insulin sensitivity [17]. The antidiabetic effect of either procedure was obvious from the first postoperative month, when the patients were still very obese (Table 2). Resolution of diabetes (assessed by OGTT) was faster in the BPD group, as 1 month after the operation, 1 patient had diabetes compared with 3 in the SG group, and at 12 months BPD resulted in lower glucose levels and lower insulin resistance compared with SG

(Table 2). However, the BPD group may have benefited more from surgery, because the absolute weight loss was greater in this group. In addition, due to malabsorption, BPD is expected to reduce more b-cell fat toxicity than SG and improve more insulin secretion and glucose homeostasis [17]. On the other hand, the SG group preoperatively had a more delayed peak insulin response during OGTT (Fig. 1) consistent with a more severe disturbance of insulin secretion, yet, 12 months after the operation, this group manifested a faster insulin response during OGTT than the BPD group (Fig. 1). Procedures that bypass the foregut are thought to be associated with faster and greater improvement in glucose homeostasis compared with purely restrictive operations or nonsurgical interventions, implying a specific effect of these procedures in terms of diabetes resolution [4,5,18]. Indeed, there is increasing evidence that alterations in circulating gastrointestinal hormones play an important role in mediating the remission of diabetes [4,5]. In the present study,

Diabetes Resolution and Hormone Changes After BPDLL and LSG / Surgery for Obesity and Related Diseases 9 (2013) 667–678 Table 3 AUC0 -

120

673

of postprandial insulin, PYY, GLP-1, ghrelin and glucagon values before and 1, 3 and 12 months after BPD and SG Pre (N ¼ 12)

Insulin (mU/mL.min) BPD 9659 ⫾ 7287 SG 14539 ⫾ 8145 P .136 PYY (pg/mL.min) BPD 13323 ⫾ 6894 SG 12847 ⫾ 3858 P .837 GLP-1 (pM.min) BPD 7537 ⫾ 3890 SG 8667 ⫾ 5092 P .550 Ghrelin (pg/mL.min) BPD 63061 ⫾ 10280 SG 70778 ⫾ 18705 P .228 Glucagon (pg/mL.min) BPD 7538 ⫾ 1611 SG 5819 ⫾ 1455 P .012

1 month (N ¼ 12)

3 months (N ¼ 12)

12 months (N ¼ 12)

P**

Py

6689 ⫾ 3803 10195 ⫾ 6588* .128

5865 ⫾ 5747* 9848 ⫾ 7505 .159

4630 ⫾ 3134* 6773 ⫾ 2938§ .098

.004 .005

.715

30582 ⫾ 10052þ 30505 ⫾ 12193þ .987

28407 ⫾ 9662z 25578 ⫾ 15052* .591

29163 þ 7705# 24928 þ 15387 .407

o0.00 o0.001

.751

21430 ⫾ 8802z 22579 ⫾ 13420# .807

19492 ⫾ 8719z 19126 ⫾ 14626* .942

15486 ⫾ 5580# 14866 ⫾ 13334 .884

o.001 o0.001

.933

57703 ⫾ 11724* 53546 ⫾ 7809§ .319

59982 ⫾ 11616 54496 ⫾ 5954§ .164

66421 ⫾ 15189 53566 ⫾ 7442§ .018

.004 .006

o0.001

12184 ⫾ 3323z 8529 ⫾ 2000z .004

9492 ⫾ 1845§ 7996 ⫾ 3308* .189

7517 ⫾ 2610 6276 ⫾ 2339 .234

o0.001 .002

.104

AUC0–120 ¼ area under the curve, 0–120 minutes; BPD ¼ biliopancreatic diversion long limb; GLP-1 ¼ glucagon like peptide-1; Pre ¼ preoperatively; PYY ¼ peptide YY; SG ¼ laparoscopic sleeve gastrectomy. * P o 0.05. ** P ¼ Repeated measures analysis of variance. y P ¼ for the interaction between the surgery type and time. P ¼ t test. § P o 0.01. # P o 0.001 (paired t test).

both SG and BPD induced greatly enhanced PYY and GLP1 responses to glucose ingestion from the first postoperative month, but SG induced in addition a marked decrease in ghrelin levels (Fig. 2). These observations are in accordance with previous reports [8,9,19]. Most studies have reported increased basal and postprandial PYY and GLP-1 levels after procedures that bypass part of the small intestine [4]. In a previous study, we reported significantly increased PYY levels 3 months and 12 months after BPD [19]. Ghrelin levels after BPD have been reported to be increased or unchanged one year after the operation [4,19]. Plasma ghrelin levels are low in human obesity and increase after diet or exercise induced weight loss [4]. In the present study, ghrelin levels did not increase significantly 1 year after BPD in spite of the massive weight loss (Table 2). SG was expected to result in significantly lower fasting ghrelin levels due to the resection of the fundus and the major part of the body of the stomach, the main sources of ghrelin production, and this observation is consistent with previous reports [4,8,9]. Enhanced PYY and GLP-1 responses are thought to account also for the decreased appetite and earlier induction of satiety [4,20]. Ghrelin, on the other hand, promotes food intake and attenuates the anorectic effect of PYY and GLP-1 [20,21]. The marked ghrelin reduction after SG, is expected to contribute to appetite attenuation and weight

loss. In the present study, appetite suppression, induction of satiety, and aversion to food were similar in both groups during the first postoperative year (Fig. 4). In a previous study, we also observed similar results comparing SG with RYGB [8]. Responsible for the markedly enhanced postprandial PYY and GLP-1 responses after BPD are considered the increased hindgut L-cell nutrient exposure, as a consequence of reduced foregut nutrient absorption, in combination with expedited nutrient delivery [4,20]. In SG, the foregut is not bypassed, but it has been reported that SG results in accelerated gastric emptying, and this could result in enhanced L-cell nutrient exposure and increased PYY and GLP-1 release [22]. However, other studies on patients undergoing SG reported no acceleration in gastric emptying, and therefore, the controversy remains [23,24]. The markedly reduced ghrelin levels after SG could offer an alternative explanation for the enhanced PYY and GLP-1 responses. We have recently shown that in patients undergoing RYGB, the concomitant resection of the fundus of the excluded part of the stomach not only reduced ghrelin levels substantially but it also markedly enhanced the PYY and GLP-1 responses after glucose ingestion compared with the standard RYGB [25]. Ghrelin has been reported to attenuate the negative effects of GLP-1 and PYY on food intake and gastric emptying [20,21,26], but the observa-

674

M. Tsoli et al. / Surgery for Obesity and Related Diseases 9 (2013) 667–678

Fig. 2. Fasting and nadir ghrelin levels after glucose ingestion before and 1, 3, and 12 months after BPD (upper panel) and SG (lower panel). Values represent mean ⫾ SEM. P values represent comparisons between fasting and nadir ghrelin levels with two-tailed paired t test. BPD ¼ biliopancreatic diversion long limb; Pre ¼ preoperatively; SEM ¼ standard error of the mean; SG ¼ laparoscopic sleeve gastrectomy.

tions of our recently published study suggest that ghrelin in addition to these actions attenuates the postprandial responses of PYY and GLP-1 and also the incretin effect of GLP-1 [25]. This offers a good explanation for the enhanced postprandial PYY, GLP-1, and insulin responses after SG. Moreover, ghrelin exerts prodiabetic effects; it inhibits insulin secretion and hepatic insulin signaling, increases glucose levels and peripheral insulin resistance, stimulates insulin counterregulatory hormones, and suppresses the insulin-sensitizing hormone adiponectin [20,26]. The marked reduction in ghrelin levels after SG is expected to improve insulin sensitivity and glucose homeostasis. Patients with diabetes type 2 display abnormal regulation of glucagon secretion, with high fasting plasma glucagon levels and lack of glucagon suppression or even net secretion after the ingestion of glucose or a meal, and it has been proposed that intestine-derived mechanisms are

crucial for postprandial hyperglucagonemia [27,28]. In the present study, fasting glucagon levels decreased after either procedure, but the decrease was more evident after SG (Table 2). In a recent study performed in patients with diabetes, fasting glucagon levels did not change significantly after RYGB, a procedure that, like BPD, bypasses part of the small intestine [29]. Another study, though, has reported that fasting plasma glucagon is reduced progressively (up to 12 months) after RYGBP [30]. There are no reports concerning the effect of SG on glucagon levels. However, it has been reported that ghrelin directly stimulates glucagon secretion from pancreatic a-cells [31], and the decreased ghrelin levels after SG could account for the reduced fasting glucagon levels. Preoperatively, we observed a mild suppression of glucagon secretion during OGTT of borderline significance (P ¼ .045) in the BPD group and a lack of suppression in the SG group (P ¼ .2) (Fig. 1). The lack of glucagon suppression in the SG group is probably related to the more delayed glucose stimulated insulin response in this group preoperatively compared with the BPD group (Fig. 1). A paradoxical net increase in glucagon secretion during OGTT was observed 1 month and 3 months after surgery; however at 12 months, glucagon responses during OGTT were not significant. To the best of our knowledge, there are no published data regarding the effect of BPD and SG on glucagon response to glucose ingestion, in patients with diabetes, although a paradoxical increase in glucagon levels during OGTT after gastric bypass has been reported [29]. The transient paradoxical glucagon response during OGTT, observed postoperatively in the present study (Fig. 1), occurred in parallel with a markedly increased GLP-1 response, and it is well established that GLP-1 is a powerful inhibitor of glucagon secretion [27]. Other gastrointestinal factors that are clearly stimulated by oral glucose include glucagon-like peptide-2 (GLP-2) and glucose dependent insulinotropic peptide (GIP) [28]. GIP and GLP-2 have been shown to stimulate glucagon secretion [31,32]. A reduction in GIP levels has been detected after BPD [33] and could explain the decrease in fasting glucagon levels after BPD. We could not find published data regarding the effect of SG on glucagon and GIP levels in patients with diabetes. The postoperative changes of GLP-1, GLP-2, and GIP can probably explain the paradoxical postoperative glucagon responses, observed in the present study [28], but this issue needs to be explored further. In the present study, only BPD was associated with significant improvement in total and LDL cholesterol (Table 1). BPD causes fat malabsorption and also interruption of enterohepatic cholesterol and bile salt cycle with the consequent loss of cholesterol and bile salts leading to enhanced hepatic bile acid synthesis at the expense of the cholesterol pool. The intestinal loss of cholesterol can explain the parallel decrease of total, LDL, and HDL cholesterol. In addition, the reduced free cholesterol

Diabetes Resolution and Hormone Changes After BPDLL and LSG / Surgery for Obesity and Related Diseases 9 (2013) 667–678

675

Fig. 3. Time courses of plasma PYY and GLP-1 during OGTT before and 1, 3, and 12 months after BPD and SG. Values represent mean ⫾ SEM. BPD ¼ biliopancreatic diversion long limb; GLP-1 ¼ glucagon like peptide-1; OGTT ¼ oral glucose tolerance test; Pre ¼ preoperatively; PYY ¼ peptide YY; SEM ¼ standard error of the mean; SG ¼ laparoscopic sleeve gastrectomy.

stimulates the synthesis of LDL receptors, leading to an increased removal of LDL from the bloodstream, further enhancing the decrease of LDL cholesterol levels [3,15]. The dramatic decrease of insulin resistance at 12 months after BPD can explain the subsequent, increase of HDL cholesterol, after the initial decrease. It is well established that the low HDL levels in patients with diabetes type 2 or with metabolic syndrome are due to insulin resistance [34]. The amelioration of insulin resistance after SG leads to progressive increase of HDL cholesterol, and this is an

invariable observation after nonmalabsorptive bariatric operations [8,25]. The improvement in hypertension was more evident after BPD, and this is associated with a greater decrease in insulin resistance after BPD [3]. The main limitations of our study are the small number of patients studied and the short follow up, as the long-term results may be different from the present observations. In addition, the 2 groups of patients had per protocol significantly different BMI, making the interpretation of outcomes more difficult. Furthermore, our patients had mild diabetes

Fig. 4. Appetite and satiety scores preoperatively and 1, 3, and 12 months after BPD and SG. BPD ¼ biliopancreatic diversion long limb; SG ¼ laparoscopic sleeve gastrectomy.

676

M. Tsoli et al. / Surgery for Obesity and Related Diseases 9 (2013) 667–678

of short duration, and the results might be different in patients with more severe diabetes of longer duration. Nevertheless, this is the first study to compare directly the effects of BPD and SG on glucose homeostasis, on glucose stimulated insulin, glucagon, PYY, GLP-1, and ghrelin secretion and also on appetite, satiety, and food aversion in morbidly obese patients with diabetes type 2 of similar severity and with similar degree of insulin resistance. Conclusions In conclusion, BPD was more effective compared with SG in terms of blood pressure and HOMA IR reduction and also in correcting dyslipidemia. However, SG during the first postoperative year was equally effective with BPD in regard to appetite suppression, induction of satiety, and diabetes resolution. These effects of SG are associated with markedly enhanced postprandial PYY and GLP-1 responses and markedly reduced ghrelin levels. Suppression of ghrelin could have contributed to the enhanced postprandial PYY and GLP-1 responses and the improved glucose homeostasis. Studies of longer duration and with higher numbers of patients are needed to evaluate the long term effect of SG on weight loss and diabetes resolution. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. References [1] Buchwald H, Estok R, Fahrbach K, et al. Weight and type 2 diabetes after bariatric surgery: systematic review and meta-analysis. Am J Med 2009;122:248–56. [2] Polyzogopoulou EV, Kalfarentzos F, Vagenakis AG, Alexandrides TK. Restoration of euglycemia and normal acute insulin response to glucose in obese subjects with type 2 diabetes following bariatric surgery. Diabetes 2003;52:1098–103. [3] Scopinaro N, Marinari GM, Camerini GB, Papadia FS, Adami GF. Specific effects of biliopancreatic diversion on the major components of metabolic syndrome. A long-term follow-up study. Diabetes Care 2005;28:2406–11. [4] Ashrafian H, le Roux CW. Metabolic surgery and gut hormones—A review of bariatric entero-humoral modulation. Physiol Behav 2009;97:620–31. [5] Thaler JP, Cummings DE. Hormonal and metabolic mechanisms of diabetes remission after gastrointestinal surgery. Endocrinology 2009;150:2518–25. [6] Van Hee RH. Biliopancreatic diversion in the surgical treatment of morbid obesity. World J Surg 2004;28:435–44. [7] Brethauer S, Hammel J, Schauer P. Systematic review of sleeve gastrectomy as staging and primary bariatric procedure. Surg Obes Relat Dis 2009;5:469–75. [8] Karamanakos SN, Vagenas K, Kalfarentzos F, Alexandrides TK. Weight loss, appetite suppression and changes in fasting and postprandial ghrelin and peptide-YY levels after Roux-en-Y gastric bypass and sleeve gastrectomy: a prospective, double blind study. Ann Surg 2008;247:401–7.

[9] Peterli R, W¨olnerhanssen B, Peters T, et al. Improvement in glucose metabolism after bariatric surgery: comparison of laparoscopic Rouxen-Y gastric bypass and laparoscopic sleeve gastrectomy: a prospective randomized trial. Ann Surg 2009;250:234–41. [10] Vidal J, Ibarzabal A, Romero F, et al. Type 2 diabetes mellitus and the metabolic syndrome following sleeve gastrectomy in severely obese subjects. Obes Surg 2008;18:1077–82. [11] Kalfarentzos F, Skroubis G, Karamanakos S, et al. Biliopancreatic diversion with Roux-en-Y gastric bypass and long limbs: Advances in surgical treatment for super-obesity. Obes Surg 2011;21:1849–58. [12] Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care 2004;27:1487–95. [13] Chang VT, Hwang SS, Feuerman M. Validation of the Edmonton symptom assessment scale. Cancer 2000;88:2164–71. [14] Kehagias I, Spyropoulos C, Karamanakos S, Kalfarentzos F. Efficacy of sleeve gastrectomy as sole procedure in patients with clinically severe obesity (BMI r50 kg/m2). Surg Obes Relat Dis Epub 2012 Jan 13. [15] Alexandrides TK, Skroubis G, Kalfarentzos F. Resolution of diabetes mellitus and metabolic syndrome following Roux-en-Y gastric bypass and a variant of biliopancreatic diversion in patients with morbid obesity. Obes Surg 2007;17:176–84. [16] Rizzello M, Abbatini F, Casella G, et al. Early postoperative insulinresistance changes after sleeve gastrectomy. Obes Surg 2010;20:50–5. [17] Unger RH. Lipotoxicity in the pathogenesis of obesity-dependent NIDDM: genetic and clinical implications. Diabetes 1995;44:863–70. [18] Laferrere B, Teixeira J, McGinty J, et al. Effect of weight loss by gastric bypass surgery versus hypocaloric diet on glucose and incretin levels in patients with type 2 diabetes. J Clin Endocrinol Metab 2008;93:2479–84. [19] Stratis C, Alexandrides T, Vagenas K, Kalfarentzos F. Ghrelin and peptide YY levels after a variant of biliopancreatic diversion with Roux-en-Y gastric bypass versus after colectomy: a prospective comparative study. Obes Surg 2006;16:752–8. [20] Karra E, Yousseif A, Batterham R. Mechanisms facilitating weight loss and resolution of type 2 diabetes following bariatric surgery. Trends in Endocrinol Metab 2010;21:337–44. [21] Chelikani PK, Haver AC, Reidelberger RD. Ghrelin attenuates the inhibitory effects of glucagon-like peptide-1 and peptide YY(3–36) on food intake and gastric emptying in rats. Diabetes 2006;55:3038–46. [22] Melissas J, Daskalakis M, Koukouraki S, et al. Sleeve gastrectomy-a ’’food limiting’’ operation. Obes Surg 2008;18:1251–6. [23] Braghetto I, Davanzo C, Korn O, et al. Scintigraphic evaluation of gastric emptying in obese patients submitted to sleeve gastrectomy compared to normal subjects. Obes Surg 2009;19:1515–21. [24] Bernstine H, Tzioni-Yehoshua R, Groshar D, et al. Gastric emptying is not affected by sleeve gastrectomy—scintigraphic evaluation of gastric emptying after sleeve gastrectomy without removal of the gastric antrum. Obes Surg 2009;19:293–8. [25] Chronaiou A, Tsoli M, Kehagias I, Leotsinidis M, Kalfarentzos F, Alexandrides T. Lower ghrelin levels and exaggerated postprandial peptide-YY, glucagon-like peptide-1, and insulin responses, after gastric fundus resection, in patients undergoing Roux-en-Y gastric bypass: a randomized clinical trial. Obes Surg 2012;22:1761–70. [26] Sangiao-Alvarellos S, Cordido F. Effect of ghrelin on glucose-insulin homeostasis: therapeutic implications. Int J Pept 2010;2010:1–25. [27] Knop FK. Resolution of type 2 diabetes following gastric bypass surgery: involvement of gut-derived glucagon and glucagonotropic signalling? Diabetologia 2009;52:2270–6. [28] Holst JJ, Christensen M, Lund A, de Heer J, Svendsen B, Kielgast U, Knop FK. Regulation of glucagon secretion by incretins. Diabetes Obes Metab 2011;13:(Suppl 1):89–94. [29] Umeda LM, Silva EA, Carneiro G, Arasaki CH, Geloneze B, Zanella MT. Early improvement in glycemic control after bariatric surgery

Diabetes Resolution and Hormone Changes After BPDLL and LSG / Surgery for Obesity and Related Diseases 9 (2013) 667–678 and its relationships with insulin, GLP-1, and glucagon secretion in type 2 diabetic patients. Obes Surg 2011;21:896–901. [30] Swarbrick MM, Stanhope KL, Austrheim-Smith IT, et al. Longitudinal changes in pancreatic and adipocyte hormones following Rouxen-Y gastric bypass surgery. Diabetologia 2008;51:1901–11. [31] Chuang J-C, Sakata I, Kohno D, et al. Ghrelin directly stimulates glucagon secretion from pancreatic a-cells. Mol Endocrinol 2011;25:1600–11.

677

[32] Meier JJ, Gallwitz B, Siepmann N, et al. Gastric inhibitory polypeptide (GIP) dose-dependently stimulates glucagon secretion in healthy human subjects at euglycaemia. Diabetologia 2003;46:798–801. [33] Rao RS, Kini S. GIP and Bariatric Surgery. Obes Surg 2011;21:244–52. [34] Kohen Avramoglu R, Basciano H, Adeli K. Lipid and lipoprotein dysregulation in insulin resistant states. Clinica Chimica Acta 2006; 368:1–19.

Editorial comment

Comment on: Hormone changes and diabetes resolution following biliopancreatic diversion and laparoscopic sleeve gastrectomy. A comparative prospective study Received January 2, 2013; accepted January 3, 2013

A few months ago, the authors published a similarly fashioned study comparing gastrointestinal hormone changes after Roux-en-Y gastric bypass (RYGB) with or without fundus resection [1]. As one could expect from the present study (the biliopancreatic diversion [BPD] to which the authors refer is much closer to a distal gastric bypass than a Scopinaro operation) [2], both glucagon-like peptide1 (GLP-1) and peptide-YY (PYY) were postprandially enhanced. These findings are consistent with earlier reports from the literature showing that RYGB is followed within a few days by a postprandial increase of the incretins regardless of the fact that the patients are diabetic or not with sustained changes up to 1 year after the surgery [3–10]. Improvement of glucose homeostasis and insulin sensitivity appears to be directly related to incretins, in particular GLP1 [5,11]. This phenomenon appears within a few days of RYGB. GLP-1 and PYY also increase rapidly after sleeve gastrectomy, as documented by the present study and most of the literature [3,7,8,12], and result in a rapid decrease in insulin resistance [13]. These changes are not seen after medical treatment of obesity or gastric banding (LAGB) irrespective of the weight loss [9,12,14]. Although populations are too small to draw formal conclusions and the SG/LAGB ratio is not mentioned, this is probably why Kashyap et al. [5] did not find any improvement of type 2 diabetes mellitus (T2DM) after restrictive bariatric surgery compared to RYGB. Apart from the malabsorptive effect of the variable magnitude of the small bowel short circuit, which can explain the remission of hyperlipidemia, improvement of insulin resistance appears clearly mediated by an enhanced GLP-1 and PYY secretion. ‘‘Hindgut’’ and ‘‘foregut’’ mechanisms after RYGB have been suggested to explain such hormonal changes [15]: rapid delivery of food into the distal small bowel as well as duodenal exclusion. While the latter does not exist in SG, accelerated gastric

emptying after SG remains controversial [16–19] but might explain in part the increased incretin response to food ingestion. This questions both the theories of hormonal changes after RYGB [8] as well as the mechanisms through which SG produces similar changes to RYGB [10]. The authors have shown in both of their studies that the main hormonal difference between SG (by extent, fundus resection) and RYGB (or their variant BPD) is the decreased ghrelin concentration not found when the fundus remains intact except for a transitory period o3 months. Fasting GLP-1 did not change and remained low irrespective of the resection of the fundus. Some discrepancy exists regarding fasting PYY levels, as they were found to be increased when RYGB was combined with fundus resection [1], whereas the same fundic resection yielded a decrease after SG contrary to BPD in the present study. A remarkable difference between the 2 studies was the presence or not of T2DM. This, as well as the much higher BMI in the BPD group, might explain why fundic resection does translate into the same fasting hormonal variations. Ghrelin secretion appears to be a major difference between SG and RYGBlike procedures. Fasting as well as postprandial ghrelin is markedly decreased after SG, whereas it remains unaffected by RYGB without fundic resection. Although ghrelin, which attenuates the anorectic effect of GLP-1 and PYY, does not seem to alter postprandial satiety and the incretins effects after RYGB, it could promote the antiobesity and antidiabetic effects of SG. It is understandable that decreased ghrelin levels enhance PYY and GLP-1 effects in SG. As a consequence, better control of postprandial glucose levels can be achieved when the fundus is resected [1]. However, there is still no clear explanation why SG has the same consequences on incretins while being a radical departure from RYGB-like procedures. After SG, one possible explanation is that low ghrelin levels ‘‘prevent’’