Surgery for Obesity and Related Diseases 11 (2015) 573–577
Original article
Increased risk of OGTT-induced hypoglycemia after gastric bypass in severely obese patients with normal glucose tolerance Marie Pigeyre, MD, Ph.D.a, Charlotte Vaurs, M.D.b, Violeta Raverdy, M.D.c, Hélène Hanaire, M.D.b, Patrick Ritz, M.D., Ph.D.b,d,*, François Pattou, M.D., Ph.D.c a
Inserm U859, Biotherapy Diabetes, Department of Nutrition, University of Lille, CHRU Lille, Lille, France b Cardiovascular and Metabolic Disease Department, CHU de Toulouse, Toulouse, France c Inserm U859, Biotherapy Diabetes, Department of General and Endocrine Surgery, University of Lille, CHRU Lille, Lille, France d Inserm UMR 1027, Université Paul Sabatier, Université de Toulouse 3, Toulouse, France Received August 1, 2014; accepted December 3, 2014
Abstract
Background: Hypoglycemic episodes are described after bariatric surgery. Objective: To report the prevalence of hypoglycemia after a 75 g oral glucose load (OGTT) after Roux-en-Y gastric bypass (RYGB) and adjustable gastric banding (LAGB), and to identify predicting factors. Setting: Bariatric surgery referral center. Methods: Prospective cohort of 351 consecutive patients before and 12 months after bariatric surgery, on whom an OGTT was performed. The main outcome measure was postchallenge hypoglycemia (PCHy), defined as a 120 minute plasma glucose value o2.8 mmol/L (50.4 mg/dL). Results: Only patients with an RYGB presented with PCHy. It occurred in 23 patients or a prevalence of 10.4% after an RYGB. The OR was 25.5 (95% CI 3.4–191; P o .001) compared with before surgery. Patients with PCHy after surgery had a lower glycated hemoglobin (HbA1c), and a lower 2-hour postchallenge value before surgery. Before surgery, patients with normal glucose tolerance had an increased risk of PCHy after surgery (OR 8.6, 95% CI 2.0–37.6; P o .001). Conclusions: The prevalence of OGTT-induced hypoglycemia is increased 25.5 times, 12 months after an RYGB. This is not observed after a gastric banding. (Surg Obes Relat Dis 2015;11:573– 577.) r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Bariatric surgery; Blood glucose; Gastroplasty; Hypoglycemia
The Roux-en-Y gastric bypass (RYGB) procedure of bariatric surgery is recognized as the most efficient longterm weight losing strategy, but it still needs to be evaluated as to the benefit/risk ratio for patients. An impressive decrease in mortality, cardiovascular events and cancer prevalence has been shown [1,2] and a major improvement in metabolic obesity-related co-morbidities has also been reported [3,4].
*
Correspondence: Patrick Ritz, Unité de Nutrition, CHU Larrey, 31059 Toulouse cedex 09, France. E-mail: [email protected]
One of the side effects of RYGB is hypoglycemia. The dumping syndrome is a classical consequence of procedures involving gastrectomy and the anastomosis of the gastric remnant to the jejunum. It has an early after-meal component and a late one, which is a hypoglycemia [5]. The mechanism of postbypass hypoglycemia has been the matter of debate, with some authors considering that it results from pancreatic β-cell hyperplasia, while others refute this argument [6–13]. However, the prevalence of hypoglycemia is not known after surgery [14]. Indeed, the prevalence of severe hypoglycemia (defined as the requirement of a third party to correct the symptoms, or neuroglucopenia) has been estimated to be .36% to o1% [15,16].
http://dx.doi.org/10.1016/j.soard.2014.12.004 1550-7289/r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
574
M. Pigeyre et al. / Surgery for Obesity and Related Diseases 11 (2015) 573–577
However, the patients often report some milder symptoms. There are even asymptomatic low plasma or interstitial glucose values observed with continuous glucose monitoring (CGMS) in real life circumstances [17,18] or after a glucose challenge [14] even after a gastric banding [19]. There is one small study reporting a very high (72%) prevalence of postchallenge hypoglycemia [20]. It is also known that chronic hypoglycemia decreases patient awareness in type 1 diabetic patients, who tend to “tolerate” lower glycemic values, exposing them to a greater risk of severe hypoglycemia [21]. Unless a systematic screening of symptoms or of low blood glucose is carried out after surgery, we may not be able to evaluate the prevalence of hypoglycemia, and to recognize patients at risk. The aim of the present study was to report the prevalence of low plasma glucose values after a 75g oral glucose load (OGTT) after RYGB and adjustable gastric banding, and to look for predicting factors. The present report compares the preoperative and the 12-month results of the OGTT. Methods Patients Patients enrolled in this prospective study were participants of ABOS (Atlas Biologique de l’Obésité Sévère), an ongoing cohort study designed to explore the biological determinants of the outcomes of metabolic surgery (Clinical Gov NCT01129297). Six hundred twenty-nine patients had bariatric surgery from June 2006 to March 2012; 216 patients were excluded from the study because of type 2 diabetes, and 62 OGTT were missing at 1 year postop (follow up of 85%); a total of 351 patients (n ¼ 222 RYGB / n ¼ 129 gastric bandings ; LAGB) were therefore included in the analysis. Ethical permission The local institutional review board approved the study protocol, and each patient signed a written consent form. Inclusion criteria were an age between 18 and 65 years, and indication for bariatric surgery based on the National Institute of Health Consensus Statement criteria. Surgeries The type of surgery was chosen with the patient before enrolment, after a comprehensive multidisciplinary evaluation and individual information about the associated risks and benefits. The 2 operations were performed by laparoscopy in a standardized fashion. Briefly, LAGB consisted of the installation of a silicone band placed with the pars flacida approach, with gastrogastric anterior fixation by 3 nonabsorbable sutures. RYGB was performed with a gastric pouch smaller than 30 mL, a stapled gastro jejunal
anastomosis with an alimentary limb of 150 cm, connected to a biliopancreatic limb of 70 cm with a side-to-side jejuno-jejunal anastomosis. The primary outcomes of the prospective study were fasting plasma glucose (FPG) and body mass index (BMI), measured at baseline and 12 months after surgery. Other outcomes were 2-hour glucose, glycated hemoglobin (HbA1c), and fasting plasma insulin (immunoreactive monoclonal assay), measured at baseline and postoperatively at 12 months. Indices of insulin sensitivity (HOMA-S) and β-cell function (HOMA-BS disposition index) were calculated according to homeostatic model assessment as indicated: (http://www.dtu.ox.ac.uk/homa calculator/index.php). Glucose tolerance was defined at baseline according to the guidelines of the American Diabetes Association [22] based on the preoperative value of FBG, HbA1c, and oral glucose tolerance test (OGTT, 75g). Patients with type III obesity and normal glucose tolerance (NGT), or impaired glucose tolerance (IGT), or impaired fasting glucose (IFG) were considered in the analysis. Patients with type 2 diabetes were excluded to avoid confounding biases introduced by antidiabetic treatments when analyzing hypoglycemia and HOMA indexes. The OGTT: a 75g oral glucose challenge was realized in fasting patients. A venous sample was taken before, and 120 minutes after the challenge in 351 patients. Calculations and statistics Data are reported as mean and SD. PCHy was defined as a plasma glucose value o2.8 mmol/L (50.4 mg/dL) according to Service [6]. The HOMA S and HOMA B were calculated as described in http://www.dtu.ox.ac.uk/homacalculator/index. php. Data were compared with either an ANOVA or a MannWhitney test as appropriate. Calculations were performed using Prism (Prism, version 5, Graphpad Software, San Diego, CA). Results The prevalence of hypoglycemia was 1 in 351 patients before surgery (this person later had a gastric bypass). At 12 months, 23 out of 222 patients with an RYGB, while none of the 129 LAGB group had PCHy (P χ2 o .001). The odds ratio for PCHy after surgery was 25.5 (95% CI 3.4–191; P o .001) for patients having had an RYGB compared with the prevalence before surgery. The prevalence was therefore 10.4%, in the patients who had had an RYGB. In the ABOS cohort, the occurrence of dumping syndrome or hypoglycemia subjective symptoms was available for 70.3% (n ¼ 156) of patients after RYGB in daily life. Among patients with low glucose values at 120 minutes of OGTT, 31.3% described the occurrence of dumping syndrome (versus 29.3% among patients without low glucose values, P ¼ 1.0; Fischer’s exact test) and 18.8%
Postgastric Bypass Hypoglycemia / Surgery for Obesity and Related Diseases 11 (2015) 573–577
575
Table 1 Physical characteristics of the RYGB patients Patients with PCHy
Age (yr) Sex ratio (F:M) Weight before (kg) BMI before (kg/m2) BMI before (range) Waist circumference before (cm) Hip circumference before (cm) Weight at M12 (kg) BMI at M12 (kg/m2) BMI at M12 (range) EWL M12 (%) %weight loss M12
Patients without PCHy
Mean
SD
N
Mean
SD
N
P*
38.1 2.8 134.4 47.3 39.5–60.5 126 144 89.8 31.7 23.1–41.0 71.1 32.8
11.4
25.3 7.7
199 199 199 199
.918 .590† .400 .280
16.3 14.1 17.4 4.65
20 20 23 23
15.5 14 21.6 6.70
184 184 199 199
.068 .336 .374 .301
20.7 8.90
23 23
38.0 3.8 139.5 49.1 35.1–92.5 132 147 95.2 33.5 22.6–70.9 68.2 32.0
10.5
24.9 5.7
23 23 23 23
19.3 7.58
199 199
.489 .318
RYGB ¼ Roux-en-Y gastric bypass; PCHy ¼ postchallenge hypoglycemia, i.e., plasma glucose o2.8 mmol/L at 120 min of the OGTT (oral glucose tolerance test), 1 yr after surgery; SD ¼ standard deviation; BMI ¼ body mass index; EWL ¼ excess weight loss * Analysis of variance or Mann-Whitney as appropriate. † 2 χ.
had hypoglycaemia subjective symptoms (versus 11.6% among patients without low glucose values, P ¼ .44; Fischer’s exact test). No patient had serious side effects during OGTT. The physical characteristics of the patients with PCHy are reported in Table 1. Patients who had an RYGB did not differ in their presurgery physical characteristics (age 38⫾10; initial weight 139⫾25, BMI 49⫾7.5) from those who had a LAGB (age 37⫾11; initial weight 126⫾20.5, BMI 45⫾6.3). Some presurgery characteristics were different in patients who developed PCHy after surgery than in those who did not: they were more likely to be NGT, their HbA1c, and the 2-hour postchallenge values were lower (Table 2). Strictly NGT patients before surgery (no IGT, no IFG [22]) were exposed to an increased risk of PCHy after surgery (OR 8.6, 95% CI 2.0–37.6; P o .001). Among the 23 patients with PCHy after surgery, 21 were NGT and 2 were IGT before surgery. Before surgery, there was no glycemia value at 2 hours that could be used as a predictive threshold of postsurgery PCHy (ROC curve AUC ¼ 0.372). There were significantly more patients with a higher than normal HbA1c (46%) in the patients without PCHy at M12 (86/ 328) than in the patients with PCHy (0/23). Twenty out of the 23 patients with PCHy had a high insulin concentration at 120 minutes (43 microU/mL). The disposition index (HOMA BS) was similar between the 2 groups, as was HOMA S. Patients with and without PCHy did not differ in their weight loss (Table 1).
Discussion This study found that the prevalence of PCHy is multiplied by 25.5 after a gastric bypass, being close to 10% at 12 months postop. This is specific to the gastric bypass;
after a gastric banding none of the patients displayed low glucose values. This study has strengths: it is a prospective and comprehensive evaluation of a cohort of patients operated on with 2 surgical procedures in a single referral center. Patients enrolled in this prospective cohort study were evaluated before, and one year after surgery. The follow-up rate of patients was 85%. The OGTT with a 75 g glucose challenge is a standardized tool [22]. We have excluded the diabetic patients to avoid confusion with iatrogenic hypoglycemia Table 2 Metabolic characteristics of RYGB patients before and after OGTT Patients with PCHy
Patients without PCHy
Mean
SD
N
Mean
SD
N
P*
PG T0 M0 PG T120 M0 Insulin T0 M0 Insulin T120 M0 HbA1 c M0 HOMA B M0 HOMA S M0
5.1 5.9 14.6 71.5 5.6 152 61
0.50 1.4 10.0 63.7 0.3 73.1 31.8
23 23 23 23 22 23 23
5.4 6.6 14.9 71.5 5.9 141 62
0.6 1.7 8.9 64.4 0.5 61.1 40.1
199 199 199 198 198 199 199
.118 .044 .739 .989 .044 .500 .625
PG T0 M12 PG T120 M12 Insulin T0 M12 Insulin T120 M12 HbA1c M12 HOMAB M12 HOMAS M12
4.6 2.3 4.9 7.5 5.4 87 196
0.3 0.4 3.0 4.7 0.30 35.3 128
23 23 22 22 23 22 22
4.7 4.4 5.2 11 5.5 88 185
0.4 1.3 2.9 13.4 0.35 32.6 151
199 199 192 193 192 192 192
.440 .000 .602 .939 .314 .869 .574
PCHy ¼ postchallenge hypoglycemia, i.e., plasma glucose o2.8 mmol/ L (50.4 mg/dL) at 120 min of the OGTT (oral glucose tolerance test); PG ¼ plasma glucose; HbA1c ¼ glycated hemoglobin, HOMA ¼ homeostasis model assessment, which estimates steady-state β-cell function (B) and insulin sensitivity (S); M0 ¼ before the surgery; M12 ¼ 12 months after the surgery; T0, T120 values before and at 120 min of the OGTT. * Analysis of variance or Mann-Whitney as appropriate.
576
M. Pigeyre et al. / Surgery for Obesity and Related Diseases 11 (2015) 573–577
when calculating HOMA values, and retained all other patients. We chose to evaluate the patients after one year since it is the time when the prevalence of symptomatic hypoglycemia rises [10]. There is no other study that has systematically evaluated the prevalence of PCHy except that of Scavini et al. after a LAGB [19]. Roslin et al. [20] reported a much higher prevalence of PCHy (72%) after a 100 g glucose load and with a higher threshold (3.3 mmol/L; 59.4 mg/dL) in 36 patients but in that case both the glucose load and the glycemic threshold used were not standard. Our study has weaknesses: it reports results after gastric banding or bypass but not after a sleeve gastrectomy, which is becoming a frequent procedure [23]. The prevalence shown here is that of biological PCHy, which may not correspond to the prevalence of symptoms reported by the patients in real-life circumstances. It may therefore be an overevaluation of the prevalence. However, Salehi et al. [12] have shown that the nadir of glycemia after a meal tolerance test was lowest in symptomatic hypoglycemic patients after an RYGB. They chose a value of 2.8 mmol/L (50.4 mg/dL) to define hypoglycemia, as we did here. Because bariatric surgery is a powerful tool to reverse various obesity-induced co-morbidities, the number of procedures is increasing, and it tends to be performed in frail patients (such as the elderly patient for example). Hypoglycemic episodes can be associated with adverse events such as falls, injuries, and impaired quality of life. Therefore, it is useful to recognize that at most 10% of the patients that were not diabetic before the study may experience low glucose values after a glucose challenge and that it concerns patients after a gastric bypass. A 75 g oral glucose challenge may appear artificial, since patients with the small gastric pouch of a gastric bypass rarely consume this much carbohydrate. However, this is a standardized and validated tool to diagnose glycemic abnormalities [22]. Meal tests are more physiologic, but are not standardized. We have noted in the literature that their nutrient composition varied widely. Many studies report the characteristics of severe hypoglycemia [10]. Bantle et al. [15] have estimated that .36% of the patients operated on in their practice presented with severe hypoglycemia. From a national Swedish registry, Marsk et al. also report sever hypoglycemia (such as hospital admission, seizures, etc) in operated patients, with a prevalence of o1% [16]. The prevalence of the dumping syndrome has been reported to be between 20–70% [24], but the early component of the syndrome is not always followed by a hypoglycemia [12,13]. Our study does not confirm the data from Scavini et al. [19] who found a high prevalence of PCHy after a LAGB. There is debate about the mechanisms leading to postbypass hypoglycemia. Pancreatic β-cell hyperplasia is demonstrated but gastric bypass as a causative mechanism is discussed [6,7]. Anyhow, there is an inappropriate insulin secretion, since insulin concentrations are too high for the
blood glucose [12,13,25]. At 120 minutes, the majority of the patients studied here (20/23) had insulin concentrations higher than 3 microU/mL, despite a low plasma glucose concentration. It therefore suggests that patients developing a PCHy secrete more insulin but the reasons for this inappropriate secretion cannot be elucidated from this study. We only recently introduced a systematic 30 minute measurement during the OGTT, which concerned only 14 (out of 23) patients with PCHy. Therefore, we cannot discuss the presence of inappropriate early insulin secretion. Furthermore, the fasting disposition index calculated with HOMA BS did not differ between the 2 groups of patients. Salehi et al. [12,13] have suggested that hypoglycemia results from a complex mixture of increased absorbed carbohydrates, increased prehepatic insulin secretion, to some extent related to increased GLP-1 concentrations, and increased glucose disposal. We could identify a phenotype of patients predicting later PCHy. Before surgery, patients with PCHy had lower plasma glucose at 120 minutes, and a lower HbA1c, in relation to a lower proportion of patients with IFG or IGT. This is confirmed by a prevalence of 0% of HbA1c above 6% in the PCHy patients compared with 26% of patients without PCHy. Therefore, the OR of developing a PCHy after surgery was increased in patients who were NGT before surgery, as if patients with impaired glucose metabolism were protected. However, patients with PCHy were not more insulin sensitive either before or after surgery, since the HOMA-S was similar in both groups. It has been found that patients with severe hypoglycemia lose more weight than those who do not have hypoglycemia [10]. This is not confirmed here: patients with PCHy lost a similar amount of weight as those without PCHy. Conclusions The present study shows that the prevalence of OGTTinduced hypoglycemia is increased 25.5 times in some patients, 12 months after an RYGB. This is not observed after a gastric banding. It appears that the 2 groups of patients only differed in their initial glucose metabolism. The high prevalence of hypoglycemia after a RYGB argues in favor of a systematic screening of low plasma glucose values during an OGTT during follow-up for safety and quality improvements of RYGB. Patients with NGT before surgery should be monitored, and educated to recognize hypoglycemic symptoms. All authors made substantial contribution to the design of the study, the acquisition, the analyses and the interpretation of the data. All authors revised the manuscript critically. Disclosures None of the authors have conflict of interest with respect to this manuscript. Guarantor: Patrick Ritz.
Postgastric Bypass Hypoglycemia / Surgery for Obesity and Related Diseases 11 (2015) 573–577
References [1] Adams TD, Gress RE, Smith SC, et al. Long-term mortality after gastric bypass surgery. N Engl J Med 2007;357:753–61. [2] Sjöström L. Review of the key results from the Swedish Obese Subjects (SOS) trial - a prospective controlled intervention study of bariatric surgery. J Intern Med 2013;273:219–34. [3] Sjöström L, Lindroos AK, Peltonen M, et al. Swedish Obese Subjects Study Scientific Group. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med 2004;351: 2683–93. [4] 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. [5] Tack J, Arts J, Caenepeel P, De Wulf D, Bisschops R. Pathophysiology, diagnosis and management of postoperative dumping syndrome. Nat Rev Gastroenterol Hepatol 2009;6:583–90. [6] Service GJ, Thompson GB, Service FJ, Andrews JC, Collazo-Clavell ML, Lloyd RV. Hyperinsulinemic hypoglycemia with nesidioblastosis after gastric-bypass surgery. N Engl J Med 2005;353:249–54. [7] Meier JJ, Butler AE, Galasso R, Butler PC. Hyperinsulinemic hypoglycemia after gastric bypass surgery is not accompanied by islet hyperplasia or increased beta-cell turnover. Diabetes Care 2006;29:1554–9. [8] Vella A, Service FJ. Incretin hypersecretion in post-gastric bypass hypoglycemia—primary problem or red herring? J Clin Endocrinol Metab 2007;92:4563–5 [9] Rumilla KM, Erickson LA, Service FJ, et al. Hyperinsulinemic hypoglycemia with nesidioblastosis: histologic features and growth factor expression. Mod Pathol 2009;22:239–45. [10] Ritz P, Hanaire H. Post-bypass hypoglycaemia: a review of current findings. Diabetes Metab 2011;37:274–81. [11] Khoo CM, Muehlbauer MJ, Stevens RD, et al. Postprandial metabolite profiles reveal differential nutrient handling after bariatric surgery compared with matched caloric restriction. Ann Surg 2014;259:687–93. [12] Salehi M, Gastaldelli A, D'Alessio DA. Altered islet function and insulin clearance cause hyperinsulinemia in gastric bypass patients with symptoms of postprandial hypoglycemia. J Clin Endocrinol Metab 2014. jc20132686. [13] Salehi M, Gastaldelli A, D’Alessio DA. Blockade of glucagon-like peptide 1 receptor corrects postprandial hypoglycemia after gastric bypass. Gastroenterology 2014;146:669–80.
577
[14] Roslin MS, Oren JH, Polan BN, Damani T, Brauner R, Shah PC. Abnormal glucose tolerance testing after gastric bypass. Surg Obes Relat Dis 2013;9:26–31. [15] Bantle JP, Ikramuddin S, Kellogg TA, Buchwald H. Hyperinsulinemic hypoglycemia developing late after gastric bypass. Obes Surg 2007;17:592–4. Erratum in: Obes Surg 2007;17:996. [16] Marsk R, Jonas E, Rasmussen F, Näslund E. Nationwide cohort study of post-gastric bypass hypoglycaemia including 5,040 patients undergoing surgery for obesity in 1986-2006 in Sweden. Diabetologia 2010;53:2307–11. [17] Hanaire H, Bertrand M, Guerci B, Anduze Y, Guillaume E, Ritz P. High glycemic variability assessed by continuous glucose monitoring after surgical treatment of obesity by gastric bypass. Diabetes Technol Ther 2011;13:625–30. [18] Ritz P, Vaurs C, Bertrand M, Anduze Y, Guillaume E, Hanaire H. Usefulness of acarbose and dietary modifications to limit glycemic variability following Roux-en-Y gastric bypass as assessed by continuous glucose monitoring. Diabetes Technol Ther 2012;14: 736–40. [19] Scavini M, Pontiroli AE, Folli F. Asymptomatic hyperinsulinemic hypoglycemia after gastric banding. N Engl J Med 2005;353: 2822–3. [20] Roslin M, Damani T, Oren J, Andrews R, Yatco E, Shah P. Abnormal glucose tolerance testing following gastric bypass demonstrates reactive hypoglycemia. Surg Endosc 2011;25:1926–32. [21] Cox DJ, Gonder-Frederick L, Antoun B, Cryer PE, Clarke WL. Perceived symptoms in the recognition of hypoglycemia. Diabetes Care 1993;16:519–27. [22] American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2014;37(Suppl 1):S81–90. [23] Hutter MM, Schirmer BD, Jones DB, et al. First report from the American College of Surgeons Bariatric Surgery Center Network: laparoscopic sleeve gastrectomy has morbidity and effectiveness positioned between the band and the bypass. Ann Surg 2011;254: 410–20. discussion 420–422. [24] Laurenius A, Olbers T, Näslund I, Karlsson J. Dumping syndrome following gastric bypass: validation of the dumping symptom rating scale. Obes Surg 2013;23:740–55. [25] Goldfine AB, Mun EC, Devine E, et al. Patients with neuroglycopenia after gastric bypass surgery have exaggerated incretin and insulin secretory responses to a mixed meal. J Clin Endocrinol Metab 2007;92:4678–85.
Original article
Increased risk of OGTT-induced hypoglycemia after gastric bypass in severely obese patients with normal glucose tolerance Marie Pigeyre, MD, Ph.D.a, Charlotte Vaurs, M.D.b, Violeta Raverdy, M.D.c, Hélène Hanaire, M.D.b, Patrick Ritz, M.D., Ph.D.b,d,*, François Pattou, M.D., Ph.D.c a
Inserm U859, Biotherapy Diabetes, Department of Nutrition, University of Lille, CHRU Lille, Lille, France b Cardiovascular and Metabolic Disease Department, CHU de Toulouse, Toulouse, France c Inserm U859, Biotherapy Diabetes, Department of General and Endocrine Surgery, University of Lille, CHRU Lille, Lille, France d Inserm UMR 1027, Université Paul Sabatier, Université de Toulouse 3, Toulouse, France Received August 1, 2014; accepted December 3, 2014
Abstract
Background: Hypoglycemic episodes are described after bariatric surgery. Objective: To report the prevalence of hypoglycemia after a 75 g oral glucose load (OGTT) after Roux-en-Y gastric bypass (RYGB) and adjustable gastric banding (LAGB), and to identify predicting factors. Setting: Bariatric surgery referral center. Methods: Prospective cohort of 351 consecutive patients before and 12 months after bariatric surgery, on whom an OGTT was performed. The main outcome measure was postchallenge hypoglycemia (PCHy), defined as a 120 minute plasma glucose value o2.8 mmol/L (50.4 mg/dL). Results: Only patients with an RYGB presented with PCHy. It occurred in 23 patients or a prevalence of 10.4% after an RYGB. The OR was 25.5 (95% CI 3.4–191; P o .001) compared with before surgery. Patients with PCHy after surgery had a lower glycated hemoglobin (HbA1c), and a lower 2-hour postchallenge value before surgery. Before surgery, patients with normal glucose tolerance had an increased risk of PCHy after surgery (OR 8.6, 95% CI 2.0–37.6; P o .001). Conclusions: The prevalence of OGTT-induced hypoglycemia is increased 25.5 times, 12 months after an RYGB. This is not observed after a gastric banding. (Surg Obes Relat Dis 2015;11:573– 577.) r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Bariatric surgery; Blood glucose; Gastroplasty; Hypoglycemia
The Roux-en-Y gastric bypass (RYGB) procedure of bariatric surgery is recognized as the most efficient longterm weight losing strategy, but it still needs to be evaluated as to the benefit/risk ratio for patients. An impressive decrease in mortality, cardiovascular events and cancer prevalence has been shown [1,2] and a major improvement in metabolic obesity-related co-morbidities has also been reported [3,4].
*
Correspondence: Patrick Ritz, Unité de Nutrition, CHU Larrey, 31059 Toulouse cedex 09, France. E-mail: [email protected]
One of the side effects of RYGB is hypoglycemia. The dumping syndrome is a classical consequence of procedures involving gastrectomy and the anastomosis of the gastric remnant to the jejunum. It has an early after-meal component and a late one, which is a hypoglycemia [5]. The mechanism of postbypass hypoglycemia has been the matter of debate, with some authors considering that it results from pancreatic β-cell hyperplasia, while others refute this argument [6–13]. However, the prevalence of hypoglycemia is not known after surgery [14]. Indeed, the prevalence of severe hypoglycemia (defined as the requirement of a third party to correct the symptoms, or neuroglucopenia) has been estimated to be .36% to o1% [15,16].
http://dx.doi.org/10.1016/j.soard.2014.12.004 1550-7289/r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
574
M. Pigeyre et al. / Surgery for Obesity and Related Diseases 11 (2015) 573–577
However, the patients often report some milder symptoms. There are even asymptomatic low plasma or interstitial glucose values observed with continuous glucose monitoring (CGMS) in real life circumstances [17,18] or after a glucose challenge [14] even after a gastric banding [19]. There is one small study reporting a very high (72%) prevalence of postchallenge hypoglycemia [20]. It is also known that chronic hypoglycemia decreases patient awareness in type 1 diabetic patients, who tend to “tolerate” lower glycemic values, exposing them to a greater risk of severe hypoglycemia [21]. Unless a systematic screening of symptoms or of low blood glucose is carried out after surgery, we may not be able to evaluate the prevalence of hypoglycemia, and to recognize patients at risk. The aim of the present study was to report the prevalence of low plasma glucose values after a 75g oral glucose load (OGTT) after RYGB and adjustable gastric banding, and to look for predicting factors. The present report compares the preoperative and the 12-month results of the OGTT. Methods Patients Patients enrolled in this prospective study were participants of ABOS (Atlas Biologique de l’Obésité Sévère), an ongoing cohort study designed to explore the biological determinants of the outcomes of metabolic surgery (Clinical Gov NCT01129297). Six hundred twenty-nine patients had bariatric surgery from June 2006 to March 2012; 216 patients were excluded from the study because of type 2 diabetes, and 62 OGTT were missing at 1 year postop (follow up of 85%); a total of 351 patients (n ¼ 222 RYGB / n ¼ 129 gastric bandings ; LAGB) were therefore included in the analysis. Ethical permission The local institutional review board approved the study protocol, and each patient signed a written consent form. Inclusion criteria were an age between 18 and 65 years, and indication for bariatric surgery based on the National Institute of Health Consensus Statement criteria. Surgeries The type of surgery was chosen with the patient before enrolment, after a comprehensive multidisciplinary evaluation and individual information about the associated risks and benefits. The 2 operations were performed by laparoscopy in a standardized fashion. Briefly, LAGB consisted of the installation of a silicone band placed with the pars flacida approach, with gastrogastric anterior fixation by 3 nonabsorbable sutures. RYGB was performed with a gastric pouch smaller than 30 mL, a stapled gastro jejunal
anastomosis with an alimentary limb of 150 cm, connected to a biliopancreatic limb of 70 cm with a side-to-side jejuno-jejunal anastomosis. The primary outcomes of the prospective study were fasting plasma glucose (FPG) and body mass index (BMI), measured at baseline and 12 months after surgery. Other outcomes were 2-hour glucose, glycated hemoglobin (HbA1c), and fasting plasma insulin (immunoreactive monoclonal assay), measured at baseline and postoperatively at 12 months. Indices of insulin sensitivity (HOMA-S) and β-cell function (HOMA-BS disposition index) were calculated according to homeostatic model assessment as indicated: (http://www.dtu.ox.ac.uk/homa calculator/index.php). Glucose tolerance was defined at baseline according to the guidelines of the American Diabetes Association [22] based on the preoperative value of FBG, HbA1c, and oral glucose tolerance test (OGTT, 75g). Patients with type III obesity and normal glucose tolerance (NGT), or impaired glucose tolerance (IGT), or impaired fasting glucose (IFG) were considered in the analysis. Patients with type 2 diabetes were excluded to avoid confounding biases introduced by antidiabetic treatments when analyzing hypoglycemia and HOMA indexes. The OGTT: a 75g oral glucose challenge was realized in fasting patients. A venous sample was taken before, and 120 minutes after the challenge in 351 patients. Calculations and statistics Data are reported as mean and SD. PCHy was defined as a plasma glucose value o2.8 mmol/L (50.4 mg/dL) according to Service [6]. The HOMA S and HOMA B were calculated as described in http://www.dtu.ox.ac.uk/homacalculator/index. php. Data were compared with either an ANOVA or a MannWhitney test as appropriate. Calculations were performed using Prism (Prism, version 5, Graphpad Software, San Diego, CA). Results The prevalence of hypoglycemia was 1 in 351 patients before surgery (this person later had a gastric bypass). At 12 months, 23 out of 222 patients with an RYGB, while none of the 129 LAGB group had PCHy (P χ2 o .001). The odds ratio for PCHy after surgery was 25.5 (95% CI 3.4–191; P o .001) for patients having had an RYGB compared with the prevalence before surgery. The prevalence was therefore 10.4%, in the patients who had had an RYGB. In the ABOS cohort, the occurrence of dumping syndrome or hypoglycemia subjective symptoms was available for 70.3% (n ¼ 156) of patients after RYGB in daily life. Among patients with low glucose values at 120 minutes of OGTT, 31.3% described the occurrence of dumping syndrome (versus 29.3% among patients without low glucose values, P ¼ 1.0; Fischer’s exact test) and 18.8%
Postgastric Bypass Hypoglycemia / Surgery for Obesity and Related Diseases 11 (2015) 573–577
575
Table 1 Physical characteristics of the RYGB patients Patients with PCHy
Age (yr) Sex ratio (F:M) Weight before (kg) BMI before (kg/m2) BMI before (range) Waist circumference before (cm) Hip circumference before (cm) Weight at M12 (kg) BMI at M12 (kg/m2) BMI at M12 (range) EWL M12 (%) %weight loss M12
Patients without PCHy
Mean
SD
N
Mean
SD
N
P*
38.1 2.8 134.4 47.3 39.5–60.5 126 144 89.8 31.7 23.1–41.0 71.1 32.8
11.4
25.3 7.7
199 199 199 199
.918 .590† .400 .280
16.3 14.1 17.4 4.65
20 20 23 23
15.5 14 21.6 6.70
184 184 199 199
.068 .336 .374 .301
20.7 8.90
23 23
38.0 3.8 139.5 49.1 35.1–92.5 132 147 95.2 33.5 22.6–70.9 68.2 32.0
10.5
24.9 5.7
23 23 23 23
19.3 7.58
199 199
.489 .318
RYGB ¼ Roux-en-Y gastric bypass; PCHy ¼ postchallenge hypoglycemia, i.e., plasma glucose o2.8 mmol/L at 120 min of the OGTT (oral glucose tolerance test), 1 yr after surgery; SD ¼ standard deviation; BMI ¼ body mass index; EWL ¼ excess weight loss * Analysis of variance or Mann-Whitney as appropriate. † 2 χ.
had hypoglycaemia subjective symptoms (versus 11.6% among patients without low glucose values, P ¼ .44; Fischer’s exact test). No patient had serious side effects during OGTT. The physical characteristics of the patients with PCHy are reported in Table 1. Patients who had an RYGB did not differ in their presurgery physical characteristics (age 38⫾10; initial weight 139⫾25, BMI 49⫾7.5) from those who had a LAGB (age 37⫾11; initial weight 126⫾20.5, BMI 45⫾6.3). Some presurgery characteristics were different in patients who developed PCHy after surgery than in those who did not: they were more likely to be NGT, their HbA1c, and the 2-hour postchallenge values were lower (Table 2). Strictly NGT patients before surgery (no IGT, no IFG [22]) were exposed to an increased risk of PCHy after surgery (OR 8.6, 95% CI 2.0–37.6; P o .001). Among the 23 patients with PCHy after surgery, 21 were NGT and 2 were IGT before surgery. Before surgery, there was no glycemia value at 2 hours that could be used as a predictive threshold of postsurgery PCHy (ROC curve AUC ¼ 0.372). There were significantly more patients with a higher than normal HbA1c (46%) in the patients without PCHy at M12 (86/ 328) than in the patients with PCHy (0/23). Twenty out of the 23 patients with PCHy had a high insulin concentration at 120 minutes (43 microU/mL). The disposition index (HOMA BS) was similar between the 2 groups, as was HOMA S. Patients with and without PCHy did not differ in their weight loss (Table 1).
Discussion This study found that the prevalence of PCHy is multiplied by 25.5 after a gastric bypass, being close to 10% at 12 months postop. This is specific to the gastric bypass;
after a gastric banding none of the patients displayed low glucose values. This study has strengths: it is a prospective and comprehensive evaluation of a cohort of patients operated on with 2 surgical procedures in a single referral center. Patients enrolled in this prospective cohort study were evaluated before, and one year after surgery. The follow-up rate of patients was 85%. The OGTT with a 75 g glucose challenge is a standardized tool [22]. We have excluded the diabetic patients to avoid confusion with iatrogenic hypoglycemia Table 2 Metabolic characteristics of RYGB patients before and after OGTT Patients with PCHy
Patients without PCHy
Mean
SD
N
Mean
SD
N
P*
PG T0 M0 PG T120 M0 Insulin T0 M0 Insulin T120 M0 HbA1 c M0 HOMA B M0 HOMA S M0
5.1 5.9 14.6 71.5 5.6 152 61
0.50 1.4 10.0 63.7 0.3 73.1 31.8
23 23 23 23 22 23 23
5.4 6.6 14.9 71.5 5.9 141 62
0.6 1.7 8.9 64.4 0.5 61.1 40.1
199 199 199 198 198 199 199
.118 .044 .739 .989 .044 .500 .625
PG T0 M12 PG T120 M12 Insulin T0 M12 Insulin T120 M12 HbA1c M12 HOMAB M12 HOMAS M12
4.6 2.3 4.9 7.5 5.4 87 196
0.3 0.4 3.0 4.7 0.30 35.3 128
23 23 22 22 23 22 22
4.7 4.4 5.2 11 5.5 88 185
0.4 1.3 2.9 13.4 0.35 32.6 151
199 199 192 193 192 192 192
.440 .000 .602 .939 .314 .869 .574
PCHy ¼ postchallenge hypoglycemia, i.e., plasma glucose o2.8 mmol/ L (50.4 mg/dL) at 120 min of the OGTT (oral glucose tolerance test); PG ¼ plasma glucose; HbA1c ¼ glycated hemoglobin, HOMA ¼ homeostasis model assessment, which estimates steady-state β-cell function (B) and insulin sensitivity (S); M0 ¼ before the surgery; M12 ¼ 12 months after the surgery; T0, T120 values before and at 120 min of the OGTT. * Analysis of variance or Mann-Whitney as appropriate.
576
M. Pigeyre et al. / Surgery for Obesity and Related Diseases 11 (2015) 573–577
when calculating HOMA values, and retained all other patients. We chose to evaluate the patients after one year since it is the time when the prevalence of symptomatic hypoglycemia rises [10]. There is no other study that has systematically evaluated the prevalence of PCHy except that of Scavini et al. after a LAGB [19]. Roslin et al. [20] reported a much higher prevalence of PCHy (72%) after a 100 g glucose load and with a higher threshold (3.3 mmol/L; 59.4 mg/dL) in 36 patients but in that case both the glucose load and the glycemic threshold used were not standard. Our study has weaknesses: it reports results after gastric banding or bypass but not after a sleeve gastrectomy, which is becoming a frequent procedure [23]. The prevalence shown here is that of biological PCHy, which may not correspond to the prevalence of symptoms reported by the patients in real-life circumstances. It may therefore be an overevaluation of the prevalence. However, Salehi et al. [12] have shown that the nadir of glycemia after a meal tolerance test was lowest in symptomatic hypoglycemic patients after an RYGB. They chose a value of 2.8 mmol/L (50.4 mg/dL) to define hypoglycemia, as we did here. Because bariatric surgery is a powerful tool to reverse various obesity-induced co-morbidities, the number of procedures is increasing, and it tends to be performed in frail patients (such as the elderly patient for example). Hypoglycemic episodes can be associated with adverse events such as falls, injuries, and impaired quality of life. Therefore, it is useful to recognize that at most 10% of the patients that were not diabetic before the study may experience low glucose values after a glucose challenge and that it concerns patients after a gastric bypass. A 75 g oral glucose challenge may appear artificial, since patients with the small gastric pouch of a gastric bypass rarely consume this much carbohydrate. However, this is a standardized and validated tool to diagnose glycemic abnormalities [22]. Meal tests are more physiologic, but are not standardized. We have noted in the literature that their nutrient composition varied widely. Many studies report the characteristics of severe hypoglycemia [10]. Bantle et al. [15] have estimated that .36% of the patients operated on in their practice presented with severe hypoglycemia. From a national Swedish registry, Marsk et al. also report sever hypoglycemia (such as hospital admission, seizures, etc) in operated patients, with a prevalence of o1% [16]. The prevalence of the dumping syndrome has been reported to be between 20–70% [24], but the early component of the syndrome is not always followed by a hypoglycemia [12,13]. Our study does not confirm the data from Scavini et al. [19] who found a high prevalence of PCHy after a LAGB. There is debate about the mechanisms leading to postbypass hypoglycemia. Pancreatic β-cell hyperplasia is demonstrated but gastric bypass as a causative mechanism is discussed [6,7]. Anyhow, there is an inappropriate insulin secretion, since insulin concentrations are too high for the
blood glucose [12,13,25]. At 120 minutes, the majority of the patients studied here (20/23) had insulin concentrations higher than 3 microU/mL, despite a low plasma glucose concentration. It therefore suggests that patients developing a PCHy secrete more insulin but the reasons for this inappropriate secretion cannot be elucidated from this study. We only recently introduced a systematic 30 minute measurement during the OGTT, which concerned only 14 (out of 23) patients with PCHy. Therefore, we cannot discuss the presence of inappropriate early insulin secretion. Furthermore, the fasting disposition index calculated with HOMA BS did not differ between the 2 groups of patients. Salehi et al. [12,13] have suggested that hypoglycemia results from a complex mixture of increased absorbed carbohydrates, increased prehepatic insulin secretion, to some extent related to increased GLP-1 concentrations, and increased glucose disposal. We could identify a phenotype of patients predicting later PCHy. Before surgery, patients with PCHy had lower plasma glucose at 120 minutes, and a lower HbA1c, in relation to a lower proportion of patients with IFG or IGT. This is confirmed by a prevalence of 0% of HbA1c above 6% in the PCHy patients compared with 26% of patients without PCHy. Therefore, the OR of developing a PCHy after surgery was increased in patients who were NGT before surgery, as if patients with impaired glucose metabolism were protected. However, patients with PCHy were not more insulin sensitive either before or after surgery, since the HOMA-S was similar in both groups. It has been found that patients with severe hypoglycemia lose more weight than those who do not have hypoglycemia [10]. This is not confirmed here: patients with PCHy lost a similar amount of weight as those without PCHy. Conclusions The present study shows that the prevalence of OGTTinduced hypoglycemia is increased 25.5 times in some patients, 12 months after an RYGB. This is not observed after a gastric banding. It appears that the 2 groups of patients only differed in their initial glucose metabolism. The high prevalence of hypoglycemia after a RYGB argues in favor of a systematic screening of low plasma glucose values during an OGTT during follow-up for safety and quality improvements of RYGB. Patients with NGT before surgery should be monitored, and educated to recognize hypoglycemic symptoms. All authors made substantial contribution to the design of the study, the acquisition, the analyses and the interpretation of the data. All authors revised the manuscript critically. Disclosures None of the authors have conflict of interest with respect to this manuscript. Guarantor: Patrick Ritz.
Postgastric Bypass Hypoglycemia / Surgery for Obesity and Related Diseases 11 (2015) 573–577
References [1] Adams TD, Gress RE, Smith SC, et al. Long-term mortality after gastric bypass surgery. N Engl J Med 2007;357:753–61. [2] Sjöström L. Review of the key results from the Swedish Obese Subjects (SOS) trial - a prospective controlled intervention study of bariatric surgery. J Intern Med 2013;273:219–34. [3] Sjöström L, Lindroos AK, Peltonen M, et al. Swedish Obese Subjects Study Scientific Group. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med 2004;351: 2683–93. [4] 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. [5] Tack J, Arts J, Caenepeel P, De Wulf D, Bisschops R. Pathophysiology, diagnosis and management of postoperative dumping syndrome. Nat Rev Gastroenterol Hepatol 2009;6:583–90. [6] Service GJ, Thompson GB, Service FJ, Andrews JC, Collazo-Clavell ML, Lloyd RV. Hyperinsulinemic hypoglycemia with nesidioblastosis after gastric-bypass surgery. N Engl J Med 2005;353:249–54. [7] Meier JJ, Butler AE, Galasso R, Butler PC. Hyperinsulinemic hypoglycemia after gastric bypass surgery is not accompanied by islet hyperplasia or increased beta-cell turnover. Diabetes Care 2006;29:1554–9. [8] Vella A, Service FJ. Incretin hypersecretion in post-gastric bypass hypoglycemia—primary problem or red herring? J Clin Endocrinol Metab 2007;92:4563–5 [9] Rumilla KM, Erickson LA, Service FJ, et al. Hyperinsulinemic hypoglycemia with nesidioblastosis: histologic features and growth factor expression. Mod Pathol 2009;22:239–45. [10] Ritz P, Hanaire H. Post-bypass hypoglycaemia: a review of current findings. Diabetes Metab 2011;37:274–81. [11] Khoo CM, Muehlbauer MJ, Stevens RD, et al. Postprandial metabolite profiles reveal differential nutrient handling after bariatric surgery compared with matched caloric restriction. Ann Surg 2014;259:687–93. [12] Salehi M, Gastaldelli A, D'Alessio DA. Altered islet function and insulin clearance cause hyperinsulinemia in gastric bypass patients with symptoms of postprandial hypoglycemia. J Clin Endocrinol Metab 2014. jc20132686. [13] Salehi M, Gastaldelli A, D’Alessio DA. Blockade of glucagon-like peptide 1 receptor corrects postprandial hypoglycemia after gastric bypass. Gastroenterology 2014;146:669–80.
577
[14] Roslin MS, Oren JH, Polan BN, Damani T, Brauner R, Shah PC. Abnormal glucose tolerance testing after gastric bypass. Surg Obes Relat Dis 2013;9:26–31. [15] Bantle JP, Ikramuddin S, Kellogg TA, Buchwald H. Hyperinsulinemic hypoglycemia developing late after gastric bypass. Obes Surg 2007;17:592–4. Erratum in: Obes Surg 2007;17:996. [16] Marsk R, Jonas E, Rasmussen F, Näslund E. Nationwide cohort study of post-gastric bypass hypoglycaemia including 5,040 patients undergoing surgery for obesity in 1986-2006 in Sweden. Diabetologia 2010;53:2307–11. [17] Hanaire H, Bertrand M, Guerci B, Anduze Y, Guillaume E, Ritz P. High glycemic variability assessed by continuous glucose monitoring after surgical treatment of obesity by gastric bypass. Diabetes Technol Ther 2011;13:625–30. [18] Ritz P, Vaurs C, Bertrand M, Anduze Y, Guillaume E, Hanaire H. Usefulness of acarbose and dietary modifications to limit glycemic variability following Roux-en-Y gastric bypass as assessed by continuous glucose monitoring. Diabetes Technol Ther 2012;14: 736–40. [19] Scavini M, Pontiroli AE, Folli F. Asymptomatic hyperinsulinemic hypoglycemia after gastric banding. N Engl J Med 2005;353: 2822–3. [20] Roslin M, Damani T, Oren J, Andrews R, Yatco E, Shah P. Abnormal glucose tolerance testing following gastric bypass demonstrates reactive hypoglycemia. Surg Endosc 2011;25:1926–32. [21] Cox DJ, Gonder-Frederick L, Antoun B, Cryer PE, Clarke WL. Perceived symptoms in the recognition of hypoglycemia. Diabetes Care 1993;16:519–27. [22] American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2014;37(Suppl 1):S81–90. [23] Hutter MM, Schirmer BD, Jones DB, et al. First report from the American College of Surgeons Bariatric Surgery Center Network: laparoscopic sleeve gastrectomy has morbidity and effectiveness positioned between the band and the bypass. Ann Surg 2011;254: 410–20. discussion 420–422. [24] Laurenius A, Olbers T, Näslund I, Karlsson J. Dumping syndrome following gastric bypass: validation of the dumping symptom rating scale. Obes Surg 2013;23:740–55. [25] Goldfine AB, Mun EC, Devine E, et al. Patients with neuroglycopenia after gastric bypass surgery have exaggerated incretin and insulin secretory responses to a mixed meal. J Clin Endocrinol Metab 2007;92:4678–85.
