- Email: [email protected]
Laparoscopic sleeve gastrectomy: Effect on long-term remission for morbidly obese patients with type 2 diabetes at 5-year follow up
ARTICLE IN PRESS
Laparoscopic sleeve gastrectomy: Effect on long-term remission for morbidly obese patients with type 2 diabetes at 5-year follow up Marius Nedelcu, MD,a,b,c Marcelo Loureiro, MD,a,b,d Mehdi Skalli, MD,a,b Florence Galtier, MD,a Audrey Jaussent, MD,a Melanie Deloze, MD,a Michel Gagner, MD, FRCSC, FACS, FASMBS,e Jean Michel Fabre, MD, PhD,a,b and David Nocca, MD, PhD,a,b Montpellier and Toulon, France, Curitiba, Brazil, and Montre al, Canada
Background. In the short-term, laparoscopic sleeve gastrectomy has been shown to be effective for the treatment of the type 2 diabetes in patients with severe obesity. There are few data with greater follow-up. Our aim was to evaluate the results of laparoscopic sleeve gastrectomy on the control of type 2 diabetes in patients with severe obesity at 5 years at the University Hospital, France. Methods. From a total of 355 patients with severe obesity operated between January 2006 and June 2010, 52 (15%) had a diagnosis of type 2 diabetes before undergoing laparoscopic sleeve gastrectomy. Results. There were 31 females (60%) and 21 males (40%), with a mean age of 51 ± 10 years (range 27–67) with a mean body mass index of 48 ± 10 kg/m2 (range 35–82). The mean duration of type 2 diabetes was 10.8 ± 10.8 years before bariatric operation. The preoperative glycated hemoglobin was 8 ± 2% (range 5.9–12.8) in 45 patients; 17 patients (38%) had levels of glycated hemoglobin $9%. Three patients (6%) required insulin alone, 4 (8%) were taking oral antidiabetic medicine and insulin, and the remaining 45 patients (87%) were taking only oral antidiabetic medicines. The complete data regarding weight loss at 5-year follow-up were obtained for 46 patients, yielding an overall followup rate of 89%. The prolonged remission of type 2 diabetes achieved at 1 year that persisted at 5 years of follow-up was present in 9 patients (17%). No patient with complete remission of their type 2 diabetes required insulin preoperatively. Improvement of type 2 diabetes was observed in 30 patients (58%) at 1 year, which was maintained for 27 patients (52%) at 5-year follow-up. Conclusion. Laparoscopic sleeve gastrectomy has demonstrated a moderate efficacy in the treatment morbidly obese patients with type 2 diabetes. Markedly increased preoperative glycated hemoglobin levels, older age, and preoperative need for insulin treatment may be the factors predicting failure of complete remission of type 2 diabetes after laparoscopic sleeve gastrectomy. (Surgery 2017;j:j-j.) From the Centre Hospitalier Universitaire Montpellier,a Montpellier, France; Universite de Montpellier 1,b Montpellier, France; Centre Chirurgical de l’Obesite,c Clinique Saint Michel, Toulon, France; Universidade Positivo,d Curitiba, Brazil; H^o pital Du Sacre Cœur,e Montre al, Canada
Drs Nedelcu, Loureiro, Skalli, Galtier, Jaussent, and Fabre have no conflicts of interest or financial ties to disclose. Dr Nocca has honorarium for speaking engagements from Ethicon Endosurgery, MID, Gore, and MSD. Dr Gagner has honorarium for speaking engagements from Ethicon Endosurgery, Covidien, Olympus, MID, Transenterix, Gore, and Boehringer Labs. For the remaining authors none were declared. Accepted for publication February 20, 2017. Reprint requests: Marius Nedelcu, MD, Service de Chirurgie Digestive A, P^ ole Digestif, H^ opital Saint Eloi, Universite de Montpellier, 80, Avenue Augustin Fliche, Montpellier 34295 France. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2017 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2017.02.018
THE MOST RECENT GLOBAL PREDICTIONS by the International Diabetes Federation suggest there are 285 million people currently with diabetes worldwide. This prevalence of diabetes is thought to increase to 438 million by 2030,1 with another half billion people at high risk. Diabetes is one of the greatest public health threats of the 21st century. The relationship between obesity and type 2 diabetes (T2D) is well known, and up to one-third of patients presenting for bariatric operation are known to have diabetes.2 In France, about 20% of patients who have undergone bariatric procedures to treat morbid obesity have T2D.3 The risk of T2D increases with body mass index (BMI) from 2% among those with of BMI 25 to SURGERY 1
ARTICLE IN PRESS 2 Nedelcu et al
29.9 kg/m2 to 13% if BMI is >35 kg/m2. Patients with a BMI >35 kg/m2 have a risk of diabetes about 40 times greater than people with BMI <23 kg/m2.4 Increasing evidence suggests that health of obese persons with T2D, including metabolic control of diabetes and its associated risk factors, can benefit substantially from bariatric operation.5,6 The goals of bariatric operation evolved originally around achieving a sustained weight loss. In reality, weight loss is only one of the outcomes of bariatric operation. Bariatric operation can be associated with substantial other health benefits, including improvement or normalization of hyperglycemia, hyperlipidemia, blood pressure, obstructive sleep apnea, and improved quality of life.7 In view of the broad benefits of weight loss and the growing evidence that some bariatric procedures provide metabolic benefits that cannot be explained completely by their effects on body weight alone,8 the name “bariatric–metabolic surgery” is emerging as a more appropriate name. In the short-term, laparoscopic sleeve gastrectomy (LSG) is effective for the treatment of the T2D in patients with severe obesity.9 It decreases glycemia shortly after operation by changing gastric emptying, modifying circulating levels of several gut hormones, and also by altering feeding habits and caloric consumption. One year after operation, LSG seems to be as effective as laparoscopic Rouxen-Y gastric bypass (RYGB) for the management of T2D in severely obese patients.10,11 There were few data with greater durations of follow-up. The aim of our present study is to evaluate the results of LSG on the control of T2D in patients with severe obesity at 5-year follow-up. METHODS A retrospective, observational, monocentric study was conducted at Montpellier University Hospital (CHU de Montpellier, France). Since 2005, a database with all bariatric operation information including weight loss, comorbidities, and laboratory information, was developed and utilized prospectively. All consecutive, morbidly obese, patients with diabetes admitted to our bariatric operation unit for LSG between January 2006 and June 2010, were included in this study. Indications for LSG were severe obesity (BMI >35 kg/m2) associated with $1 comorbidities or morbid obesity (BMI >40 kg/m2). For all patients, the LSG was considered as a primary procedure with no plan for a secondary stage. From a total of 355 bariatric patients operated during this period, 52 (14.6%) had a diagnosis of T2D before undergoing LSG. T2D diagnosis was defined when
Surgery j 2017
patients were being treated with oral antiglycemic therapy and/or insulin. All patients underwent standard preoperative evaluation by a multidisciplinary team 4 to 8 months prior to the LSG and signed a written informed consent. Blood tests were performed in order to measure the level of glycated hemoglobin (A1C) and to evaluate the efficacy of pharmacologic treatment. Primary outcome was the efficacy on T2D, by recording HbA1c levels at 1 and 5 years after LSG and comparing these values to preoperative data. Secondary outcomes included operative complications and percentage of excess weight loss (%EWL). Complete remission of T2D was defined as a A1C value of #5.6% and partial remission as an A1C value of #6.5%12 for both without the use of hypoglycemic oral medication or insulin. Additionally, prolonged remission was defined when these conditions were present for at least 5 years. For those who did not achieve remission, improvement of diabetes was defined as the use of less medication (ie, #1 less antidiabetic medication) with A1C of <7% or the use of the same preoperative medication associated with a decrease in A1C level. To meet the criteria for “improved” using only the medication data, the laboratory values had to be no worse than the preoperative levels. We used the following definitions for comorbidities: hypertension (systolic blood pressure $140 and/or diastolic blood pressure $90 mm Hg, or antihypertensive drug therapy), obstructive sleep apnea (repeated upper airway occlusions during sleep with or without sleepiness and high apnea/ hypopnea index and need for continuous positive airway pressure during sleep). The criteria for remission included medication and laboratory data, but the criteria for improvement included medication or laboratory data. Operative technique. All operations were performed laparoscopically under general anesthesia using the French position (the surgeon standing between the patient’s legs). Each procedure required 5 trocars. Pneumoperitoneum was established by an open trocar insertion at the umbilicus (Hasson technique) and maintained at a pressure of 15 mm Hg. Dissection began on the greater curvature, 6 cm from the pylorus. The gastrocolic ligament along the greater curvature of the stomach was opened using an impedance coagulator Ultracision (Ethicon Endo-surgery, Johnson & Johnson Inc, Cincinnati, OH) and was freed as far as the cardio esophageal junction at the root of the left pillar of the hiatus. The short gastric vessels close to the spleen were carefully coagulated
ARTICLE IN PRESS Nedelcu et al 3
Surgery Volume j, Number j
Table I. Preoperative demographic data Sex Male Female Age (y; mean [± SD]) BMI (kg/m2; mean [± SD]) Physical activity Yes No Smoking Yes No Hypertension Yes No Sleep Apnea Yes No Dyslipidemia Yes No
n
%
21 31 50.9 (±9.6) 48.4 (±9.7)
40 60
10 42
19 81
14 36
28 72
32 19
63 37
38 13
74 26
25 27
48 52
SD, Standard deviation.
separately. A calibration tube using a 37-F, Midsleeve (MID, Medical Innovation Development, Dardilly, France) was then inserted transorally into the stomach by the anesthesiologist and directed toward the antrum; then 50 mL of 154 mM NaCl was injected into distal balloon to define the beginning of the staple line, 6 cm from the pylorus. A laparoscopic linear stapler was used to divide the stomach parallel to the orogastric tube along the lesser curvature. The instrument was fired, reloaded, and the procedure was repeated; a maximum 7 gold cartridges were used to staple the antrum, the body, and fundus of the stomach. Statistical analysis. Characteristics of the population were described by using proportions for categorical variables and mean with standard deviation (SD) for quantitative variables. Distributions were tested with the Shapiro-Wilk statistic. In order to evaluate the potential predictive factors of remission at 1 and 5 years after LSG, univariate logistic regressions were performed. Odds ratio (OR) with 95% confidence interval (CI) and P value of Wald were calculated. Quantitative variables were categorized using terciles of distributions. The association between excess weight loss at 5 years and remission at 5 years was tested using Student t test because distributions were normally distributed. Significance was set at P < .05. All analyses were carried out with the SAS software, version 9 (SAS Institute, Cary, NC).
RESULTS Demographic data, physical activity, smoking habits, and comorbidities before LSG are summarized in Table I. There were 31 females (60%) and 21 males (40%), with a mean age of 51 ± 10 years (range 27–67) with a mean BMI of 49 ± 19 kg/m2 (range 35–82). Only 19% of patients participated in a regular schedule of physical exercise, and 28% were active smokers. The associated comorbidities are also included in Table I, 63% of patients were hypertensives, 48% dyslipidemics, and 75% had sleep apnea. The duration of the diagnosis of diabetes was analyzed for 44 patients was a mean of 11 ± 11 years before the LSG (range 0–39 years). The patients were subdivided into 3 equal groups: those with diabetes for <4 years, 14 patients, between 4 and 13 years; 15 patients, and 13 years or more in 15 patients. The preoperative A1C was 8.3% ± 1.7 (range 5.9–12.8) in 45 patients; 17 patients (38%) had levels of A1C $8.9%. Three patients (6%) required insulin alone, 4 (8%) were taking oral antidiabetic medicine (OAD) and insulin, and the remaining 45 patients (87%) were taking only OADs. Forty-eight patients (92%) had a primary LSG and 4 patients underwent revisional LSG at a minimum of 3 months after removal of a prior gastric banding. Mean hospital stay was 5 days. There were no conversions to open operation, and the mortality was nil. Two patients needed future operative intervention during the follow-up, one for symptomatic cholecystolithiasis after 20 months follow-up, and the other one for incisional hernia at 26 months after the initial intervention. The complete data regarding weight loss at 5-year follow-up were obtained for 46 patients, yielding an overall follow-up rate of 89% (Table II). The prolonged remission of T2D achieved at 1 year persisting up to 5 years of follow-up was present in 9 patients (17%). The improvement of T2D was observed in 30 patients (58%) at 1 year and maintained for 27 patients (52%) at 5-year follow-up. We analyzed preoperative actors that were predictive for complete remission of T2D at 1-year follow-up (12 patients, 23%) and respectively, at 5 years (10 patients, 19%) after LSG, as well as for the prolonged remission. Age, sex, pre- and postoperative BMI, preoperative duration of T2D, pre- and postoperative levels of A1C and insulin use were analyzed as predictive factors. At both 1 and 5 years after LSG, no statistically significant factor was found to be associated with T2D remission. Despite the statistical analysis, which is
ARTICLE IN PRESS 4 Nedelcu et al
Surgery j 2017
Table II. Results 1 and 5 years postoperative 1 y postoperative
2
BMI (kg/m ) EWL (%) A1C (%) Complete remission Yes No Partial remission Yes No
5 y postoperative
n
Mean (± SD) or %
n
Mean (± SD) or %
50 50 52
34 (±7) 54 (±17) 6.4 (±1.0)
46 46 51
35 (±35) 49 (±18) 6.6 (±0.9)
12 40
23 77
10 43
19 81
26 26
50 50
17 34
33 67
SD, Standard deviation.
limited because of the low number of total patients with complete remission (n = 9), it is worth noting that no patient who achieved complete remission of T2D had been treated preoperatively with insulin. Finally, 3 patients (6%) had a recurrence of their T2D at 5 years. One patient achieved remission only in 5 years, but not at 1-year follow-up. Substantial improvement occurred in the status of the co-morbidities after LSG after 5 years. For hypertension, 17 patients (53%) had complete remission and 11 patients (34%) improved their medication status. For dyslipidemia, complete remission was recorded in 15 patients (60%) and partial remission in 4 patients (16%). The sleep apnea resolved in 25 of 38 patients (66%). DISCUSSION A recent meta-analysis performed on the few retrospective, randomized, and prospective studies confirmed that LSG has a similar effect on T2D as RYGB.13,14 The literature about remission of T2D after LSG is controversial. Many studies have reported a resolution rate of 60% to 90% for diabetes after LSG in the short term,15,16 while other studies have reported limited efficacy.17-19 It is uncertain if the widely varied remission rates between the studies of LSG are related to patient characteristics (ethnicity), or to the absence of a clear definition of resolution of T2D. Another explanation could be that these studies used different definitions of “improvement,” “resolution,” “remission,” or “cure” of diabetes. Basso et al15 reported a resolution rate of 80% after a mean follow-up period of 3 years. The mean postoperative time at discontinuation of medication was 3.3 months for LSG compared with 12.6 months after gastric banding. Todkar et al16 reported a resolution rate of 72% at 3 years after LSG. Schauer et al17 reported a
randomized prospective trial of 150 obese patients with T2D to receive either intensive medical therapy alone or intensive medical therapy plus RYGB or LSG. After 36 months with a follow-up of 91%, complete remission was achieved by 5% of patients in the medical therapy group compared with 38% of those in the RYGB group (P < .001) and 24% of those in the LSG group (P = .01). Metabolic and weight-loss outcomes generally were similar in the 2 operative groups at 1 year, although some advantages of RYGB over SLG have emerged with greater follow-up, including a greater likelihood of reaching a A1C #7.0% (a therapeutic goal of the American Diabetes Association) with no need for diabetes medications, a decreased requirement for diabetes and cardiovascular medications, greater sustained decreases in weight and BMI, and greater improvement in quality of life. Within the same clinical trial, Schauer et al19 reported that 42% of patients after RYGB, 37% after LSG, and 12% with medical therapy alone achieved the primary endpoint of A1C #6% after 1 year. Complete remission rates in our study of 19% of diabetes at 5-year follow-up after LSG in our study are inferior to those presented previously. A different population, as well as a stricter definition of remission, could be the factors that can explain the discrepancies with other reports.15,16 Up to 40% of our studied population were men, in contrast with our overall bariatric operative cohort (21% men). The high men/women ratio in series of operative treatment of T2D is becoming frequent. The reason could be that men consider bariatric operation when they are sicker. The improvement of T2D was observed in 30 patients (58%) at 1 year and it was maintained for 27 patients (52%) at 5-year follow-up. One of the main limitations of our study is that prior to our study
ARTICLE IN PRESS Nedelcu et al 5
Surgery Volume j, Number j
period (2006–2010) a different definition was considered by the endocrinologist for complete remission of T2D. During that period, every patient with diabetes and HbA1C <6.5 in the first year after operation was considered in complete remission and diabetic treatment was stopped. Currently, these patients cannot be considered in complete remission with the new definition12 at 1-year follow-up. There is strong evidence that patients with a shorter history of diabetes may benefit more than those with a longer one, which is concordant with some studies that showed a direct correlation between decreased b-cell volume, trend to apoptosis, and death with a greater duration of T2D.20 Still, in our experience, even if this evidence has been noticed, no statistical difference was obtained at 1-year or 5-year follow-up, most likely secondary to our small cohort. Lee et al21 have proposed a specific diabetes surgery score (ABCD score) to help with patient selection. The score consists of 4 independent predictors of T2D remission: patient age, BMI, Cpeptide level, and duration of diabetes, a 4-point score (except age which accounts for 2 points). The score was validated on 157 patients operated between 2006 and 2013. Patients who had remission of T2D after bariatric operation had a greater ABCD score than those who did not achieve remission (7 ± 2 versus 5 ± 2, P < .05). Patients with a greater ABCD score were also at a greater likelihood of success in remission of T2D (from 0% in score 0 to 100% in score 10). The calculation of the ABCD score was not possible in our population due to lack of data, because our investigations did not evaluate preoperative C peptide levels. C peptide levels would have been interesting and useful, because several studies have since defined the presence of greater preoperative C peptide levels (C peptide >3 ng/mL) to be an important predictor of resolution of diabetes, thereby helping to define the best candidates for bariatric treatment of diabetes.22 Our study has several limitations. First, the follow-up was not continuous for all patients and, some follow-up data were available at only 1 time. Even though our sample size (46 patients) at 5-year follow-up (87%) are better than the existing data in the literature (39 of 56 patients; 70% follow-up reported by Keidar et al),23 we cannot make a robust conclusion regarding the changes in T2D. Another limitation is the retrospective character of our study, with no possibility to evaluate the C-peptide level and with no selection criteria for patient with T2D. Our study showed that some
severely obese patients with T2D may benefit less from the LSG, but hopefully in the future, using certain selection criteria, we hope to be better able to predict those patients who will achieve complete remission of their T2D. In our experience, although the LSG is not a definitive treatment of T2D, as also with the other bariatric procedures, LSG definitely offers significant (>70%) and durable (5 years) improvement of T2D. High preoperative A1C levels, older age, and a preoperative therapy consisting of insulin may be the factors predicting failure of resolution of T2D. REFERENCES 1. Shaw JE, Sicree RA, Zimmet PZ. Global estimates of the prevalence of diabetes for 2010 and 2030. Diabetes Res Clin Pract 2010;87:4-14. 2. Residori L, Garcia-Lorda P, Flancbaum L, Pi-Sunyer FX, Laferrere B. Prevalence of co-morbidities in obese patients before bariatric surgery: effect of race. Obes Surg 2003;13: 333-40. 3. Haute autorite de sante. Obesite: Rapport sur la prise en charge chirurgicale chez l’adulte. Saint-Denis La Plaine. HAS 2009. 4. Chan JM, Rimm EB, Colditz GA, Stampfer MJ, Willett WC. Obesity, fat distribution, and weight gain as risk factors for clinical diabetes in men. Diabetes Care 1994;17:961-9. 5. Sjostrom L, Narbro K, Sjostrom CD, et al. Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med 2007;357:741-52. 6. Buchwald H, Estok R, Fahrbach K, et al. Weight and type 2 diabetes after bariatric surgery: systematic review and metaanalysis. Am J Med 2009;122:248-56. 7. Colquitt JL, Picot J, Loveman E, Clegg AJ. Surgery for obesity. Cochrane Database Syst Rev 2009:CD003641. 8. Pories WJ, Albrecht RJ. Etiology of type II diabetes mellitus: role of the foregut. World J Surg 2001;25:527-31. 9. Shah PS, Todkar JS, Shah SS. Effectiveness of laparoscopic sleeve gastrectomy on glycemic control in obese Indians with type 2 diabetes mellitus. Surg Obes Relat Dis 2009;6: 138-41. 10. Lakdawala MA, Bhasker A, Mulchandani D, Goel S, Jain S. Comparison between the results of laparoscopic sleeve gastrectomy and laparoscopic Roux-en-Y gastric bypass in the Indian population: a retrospective 1 year study. Obes Surg 2010;20:1-6. 11. Nocca D, Guillaume F, Noel P, et al. Impact of laparoscopic sleeve gastrectomy and laparoscopic gastric bypass on HbA1c blood level and pharmacological treatment of type 2 diabetes mellitus in severe or morbidly obese patients. Results of a multicenter prospective study at 1 year. Obes Surg 2011;21:738-43. 12. Buse JB, Caprio S, Cefalu WT, et al. How do we define cure of diabetes? Diabetes Care 2009;32:2133-5. 13. Yip S, Plank LD, Murphy R. Gastric bypass and sleeve gastrectomy for type 2 diabetes: a systematic review and metaanalysis of outcomes. Obes Surg 2013;23:1994-2003. 14. Cho J-M, Kim H-J, Menzo EL, Park S, Szomstein S, Rosenthal RJ. Effect of sleeve gastrectomy on type 2 diabetes as an alternative treatment modality to Roux-en-Y gastric bypass: systemic review and meta-analysis. Surg Obes Relat Dis 2015;11:1273-80.
ARTICLE IN PRESS 6 Nedelcu et al
15. Abbatini F, Rizzello M, Casella G, et al. Long-term effects of laparoscopic sleeve gastrectomy, gastric bypass, and adjustable gastric banding on type 2 diabetes. Surg Endosc 2010;24:1005-10. 16. Todkar JS, Shah SS, Shah PS, Gangwani J. Long-term effects of laparoscopic sleeve gastrectomy in morbidly obese subjects with type 2 diabetes mellitus. Surg Obes Relat Dis 2010;6:142-5. 17. Schauer PR, Bhatt DL, Kirwan JP, et al. Bariatric surgery versus intensive medical therapy for diabetes---3-year outcomes. N Engl J Med 2014;370:2002-13. 18. Rosenthal R, Li X, Samuel S, Martinez P, Zheng C. Effects of sleeve gastrectomy on patients with diabetes mellitus. Surg Obes Relat Dis 2009;5:429-34. 19. Schauer PR, Kashyap SR, Wolski K, et al. Bariatric surgery versus intensive medical therapy in obese patients with diabetes. N Engl J Med 2012;366:1567-76.
Surgery j 2017
20. Klein S, Fontana L, Young VL, et al. Absence of an effect of liposuction on insulin action and risk factors for coronary heart disease. N Engl J Med 2004;350: 2549-57. 21. Lee WJ, Almulaifi A, Tsou JJ, Ser KH, Lee YC, Chen SC. Laparoscopic sleeve gastrectomy for type 2 diabetes mellitus: predicting the success by ABCD score. Surg Obes Relat Dis 2015;11:991-6. 22. Lee WJ, Ser KH, Chong K, et al. Laparoscopic sleeve gastrectomy for diabetes treatment in nonmorbidly obese patients: efficacy and change of insulin secretion. Surgery 2010;147: 664-9. 23. Golomb I, Ben David M, Glass A, Kolitz T, Keidar A. Longterm metabolic effects of laparoscopic sleeve gastrectomy. JAMA Surg 2015;150:1051-7.
Laparoscopic sleeve gastrectomy: Effect on long-term remission for morbidly obese patients with type 2 diabetes at 5-year follow up Marius Nedelcu, MD,a,b,c Marcelo Loureiro, MD,a,b,d Mehdi Skalli, MD,a,b Florence Galtier, MD,a Audrey Jaussent, MD,a Melanie Deloze, MD,a Michel Gagner, MD, FRCSC, FACS, FASMBS,e Jean Michel Fabre, MD, PhD,a,b and David Nocca, MD, PhD,a,b Montpellier and Toulon, France, Curitiba, Brazil, and Montre al, Canada
Background. In the short-term, laparoscopic sleeve gastrectomy has been shown to be effective for the treatment of the type 2 diabetes in patients with severe obesity. There are few data with greater follow-up. Our aim was to evaluate the results of laparoscopic sleeve gastrectomy on the control of type 2 diabetes in patients with severe obesity at 5 years at the University Hospital, France. Methods. From a total of 355 patients with severe obesity operated between January 2006 and June 2010, 52 (15%) had a diagnosis of type 2 diabetes before undergoing laparoscopic sleeve gastrectomy. Results. There were 31 females (60%) and 21 males (40%), with a mean age of 51 ± 10 years (range 27–67) with a mean body mass index of 48 ± 10 kg/m2 (range 35–82). The mean duration of type 2 diabetes was 10.8 ± 10.8 years before bariatric operation. The preoperative glycated hemoglobin was 8 ± 2% (range 5.9–12.8) in 45 patients; 17 patients (38%) had levels of glycated hemoglobin $9%. Three patients (6%) required insulin alone, 4 (8%) were taking oral antidiabetic medicine and insulin, and the remaining 45 patients (87%) were taking only oral antidiabetic medicines. The complete data regarding weight loss at 5-year follow-up were obtained for 46 patients, yielding an overall followup rate of 89%. The prolonged remission of type 2 diabetes achieved at 1 year that persisted at 5 years of follow-up was present in 9 patients (17%). No patient with complete remission of their type 2 diabetes required insulin preoperatively. Improvement of type 2 diabetes was observed in 30 patients (58%) at 1 year, which was maintained for 27 patients (52%) at 5-year follow-up. Conclusion. Laparoscopic sleeve gastrectomy has demonstrated a moderate efficacy in the treatment morbidly obese patients with type 2 diabetes. Markedly increased preoperative glycated hemoglobin levels, older age, and preoperative need for insulin treatment may be the factors predicting failure of complete remission of type 2 diabetes after laparoscopic sleeve gastrectomy. (Surgery 2017;j:j-j.) From the Centre Hospitalier Universitaire Montpellier,a Montpellier, France; Universite de Montpellier 1,b Montpellier, France; Centre Chirurgical de l’Obesite,c Clinique Saint Michel, Toulon, France; Universidade Positivo,d Curitiba, Brazil; H^o pital Du Sacre Cœur,e Montre al, Canada
Drs Nedelcu, Loureiro, Skalli, Galtier, Jaussent, and Fabre have no conflicts of interest or financial ties to disclose. Dr Nocca has honorarium for speaking engagements from Ethicon Endosurgery, MID, Gore, and MSD. Dr Gagner has honorarium for speaking engagements from Ethicon Endosurgery, Covidien, Olympus, MID, Transenterix, Gore, and Boehringer Labs. For the remaining authors none were declared. Accepted for publication February 20, 2017. Reprint requests: Marius Nedelcu, MD, Service de Chirurgie Digestive A, P^ ole Digestif, H^ opital Saint Eloi, Universite de Montpellier, 80, Avenue Augustin Fliche, Montpellier 34295 France. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2017 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2017.02.018
THE MOST RECENT GLOBAL PREDICTIONS by the International Diabetes Federation suggest there are 285 million people currently with diabetes worldwide. This prevalence of diabetes is thought to increase to 438 million by 2030,1 with another half billion people at high risk. Diabetes is one of the greatest public health threats of the 21st century. The relationship between obesity and type 2 diabetes (T2D) is well known, and up to one-third of patients presenting for bariatric operation are known to have diabetes.2 In France, about 20% of patients who have undergone bariatric procedures to treat morbid obesity have T2D.3 The risk of T2D increases with body mass index (BMI) from 2% among those with of BMI 25 to SURGERY 1
ARTICLE IN PRESS 2 Nedelcu et al
29.9 kg/m2 to 13% if BMI is >35 kg/m2. Patients with a BMI >35 kg/m2 have a risk of diabetes about 40 times greater than people with BMI <23 kg/m2.4 Increasing evidence suggests that health of obese persons with T2D, including metabolic control of diabetes and its associated risk factors, can benefit substantially from bariatric operation.5,6 The goals of bariatric operation evolved originally around achieving a sustained weight loss. In reality, weight loss is only one of the outcomes of bariatric operation. Bariatric operation can be associated with substantial other health benefits, including improvement or normalization of hyperglycemia, hyperlipidemia, blood pressure, obstructive sleep apnea, and improved quality of life.7 In view of the broad benefits of weight loss and the growing evidence that some bariatric procedures provide metabolic benefits that cannot be explained completely by their effects on body weight alone,8 the name “bariatric–metabolic surgery” is emerging as a more appropriate name. In the short-term, laparoscopic sleeve gastrectomy (LSG) is effective for the treatment of the T2D in patients with severe obesity.9 It decreases glycemia shortly after operation by changing gastric emptying, modifying circulating levels of several gut hormones, and also by altering feeding habits and caloric consumption. One year after operation, LSG seems to be as effective as laparoscopic Rouxen-Y gastric bypass (RYGB) for the management of T2D in severely obese patients.10,11 There were few data with greater durations of follow-up. The aim of our present study is to evaluate the results of LSG on the control of T2D in patients with severe obesity at 5-year follow-up. METHODS A retrospective, observational, monocentric study was conducted at Montpellier University Hospital (CHU de Montpellier, France). Since 2005, a database with all bariatric operation information including weight loss, comorbidities, and laboratory information, was developed and utilized prospectively. All consecutive, morbidly obese, patients with diabetes admitted to our bariatric operation unit for LSG between January 2006 and June 2010, were included in this study. Indications for LSG were severe obesity (BMI >35 kg/m2) associated with $1 comorbidities or morbid obesity (BMI >40 kg/m2). For all patients, the LSG was considered as a primary procedure with no plan for a secondary stage. From a total of 355 bariatric patients operated during this period, 52 (14.6%) had a diagnosis of T2D before undergoing LSG. T2D diagnosis was defined when
Surgery j 2017
patients were being treated with oral antiglycemic therapy and/or insulin. All patients underwent standard preoperative evaluation by a multidisciplinary team 4 to 8 months prior to the LSG and signed a written informed consent. Blood tests were performed in order to measure the level of glycated hemoglobin (A1C) and to evaluate the efficacy of pharmacologic treatment. Primary outcome was the efficacy on T2D, by recording HbA1c levels at 1 and 5 years after LSG and comparing these values to preoperative data. Secondary outcomes included operative complications and percentage of excess weight loss (%EWL). Complete remission of T2D was defined as a A1C value of #5.6% and partial remission as an A1C value of #6.5%12 for both without the use of hypoglycemic oral medication or insulin. Additionally, prolonged remission was defined when these conditions were present for at least 5 years. For those who did not achieve remission, improvement of diabetes was defined as the use of less medication (ie, #1 less antidiabetic medication) with A1C of <7% or the use of the same preoperative medication associated with a decrease in A1C level. To meet the criteria for “improved” using only the medication data, the laboratory values had to be no worse than the preoperative levels. We used the following definitions for comorbidities: hypertension (systolic blood pressure $140 and/or diastolic blood pressure $90 mm Hg, or antihypertensive drug therapy), obstructive sleep apnea (repeated upper airway occlusions during sleep with or without sleepiness and high apnea/ hypopnea index and need for continuous positive airway pressure during sleep). The criteria for remission included medication and laboratory data, but the criteria for improvement included medication or laboratory data. Operative technique. All operations were performed laparoscopically under general anesthesia using the French position (the surgeon standing between the patient’s legs). Each procedure required 5 trocars. Pneumoperitoneum was established by an open trocar insertion at the umbilicus (Hasson technique) and maintained at a pressure of 15 mm Hg. Dissection began on the greater curvature, 6 cm from the pylorus. The gastrocolic ligament along the greater curvature of the stomach was opened using an impedance coagulator Ultracision (Ethicon Endo-surgery, Johnson & Johnson Inc, Cincinnati, OH) and was freed as far as the cardio esophageal junction at the root of the left pillar of the hiatus. The short gastric vessels close to the spleen were carefully coagulated
ARTICLE IN PRESS Nedelcu et al 3
Surgery Volume j, Number j
Table I. Preoperative demographic data Sex Male Female Age (y; mean [± SD]) BMI (kg/m2; mean [± SD]) Physical activity Yes No Smoking Yes No Hypertension Yes No Sleep Apnea Yes No Dyslipidemia Yes No
n
%
21 31 50.9 (±9.6) 48.4 (±9.7)
40 60
10 42
19 81
14 36
28 72
32 19
63 37
38 13
74 26
25 27
48 52
SD, Standard deviation.
separately. A calibration tube using a 37-F, Midsleeve (MID, Medical Innovation Development, Dardilly, France) was then inserted transorally into the stomach by the anesthesiologist and directed toward the antrum; then 50 mL of 154 mM NaCl was injected into distal balloon to define the beginning of the staple line, 6 cm from the pylorus. A laparoscopic linear stapler was used to divide the stomach parallel to the orogastric tube along the lesser curvature. The instrument was fired, reloaded, and the procedure was repeated; a maximum 7 gold cartridges were used to staple the antrum, the body, and fundus of the stomach. Statistical analysis. Characteristics of the population were described by using proportions for categorical variables and mean with standard deviation (SD) for quantitative variables. Distributions were tested with the Shapiro-Wilk statistic. In order to evaluate the potential predictive factors of remission at 1 and 5 years after LSG, univariate logistic regressions were performed. Odds ratio (OR) with 95% confidence interval (CI) and P value of Wald were calculated. Quantitative variables were categorized using terciles of distributions. The association between excess weight loss at 5 years and remission at 5 years was tested using Student t test because distributions were normally distributed. Significance was set at P < .05. All analyses were carried out with the SAS software, version 9 (SAS Institute, Cary, NC).
RESULTS Demographic data, physical activity, smoking habits, and comorbidities before LSG are summarized in Table I. There were 31 females (60%) and 21 males (40%), with a mean age of 51 ± 10 years (range 27–67) with a mean BMI of 49 ± 19 kg/m2 (range 35–82). Only 19% of patients participated in a regular schedule of physical exercise, and 28% were active smokers. The associated comorbidities are also included in Table I, 63% of patients were hypertensives, 48% dyslipidemics, and 75% had sleep apnea. The duration of the diagnosis of diabetes was analyzed for 44 patients was a mean of 11 ± 11 years before the LSG (range 0–39 years). The patients were subdivided into 3 equal groups: those with diabetes for <4 years, 14 patients, between 4 and 13 years; 15 patients, and 13 years or more in 15 patients. The preoperative A1C was 8.3% ± 1.7 (range 5.9–12.8) in 45 patients; 17 patients (38%) had levels of A1C $8.9%. Three patients (6%) required insulin alone, 4 (8%) were taking oral antidiabetic medicine (OAD) and insulin, and the remaining 45 patients (87%) were taking only OADs. Forty-eight patients (92%) had a primary LSG and 4 patients underwent revisional LSG at a minimum of 3 months after removal of a prior gastric banding. Mean hospital stay was 5 days. There were no conversions to open operation, and the mortality was nil. Two patients needed future operative intervention during the follow-up, one for symptomatic cholecystolithiasis after 20 months follow-up, and the other one for incisional hernia at 26 months after the initial intervention. The complete data regarding weight loss at 5-year follow-up were obtained for 46 patients, yielding an overall follow-up rate of 89% (Table II). The prolonged remission of T2D achieved at 1 year persisting up to 5 years of follow-up was present in 9 patients (17%). The improvement of T2D was observed in 30 patients (58%) at 1 year and maintained for 27 patients (52%) at 5-year follow-up. We analyzed preoperative actors that were predictive for complete remission of T2D at 1-year follow-up (12 patients, 23%) and respectively, at 5 years (10 patients, 19%) after LSG, as well as for the prolonged remission. Age, sex, pre- and postoperative BMI, preoperative duration of T2D, pre- and postoperative levels of A1C and insulin use were analyzed as predictive factors. At both 1 and 5 years after LSG, no statistically significant factor was found to be associated with T2D remission. Despite the statistical analysis, which is
ARTICLE IN PRESS 4 Nedelcu et al
Surgery j 2017
Table II. Results 1 and 5 years postoperative 1 y postoperative
2
BMI (kg/m ) EWL (%) A1C (%) Complete remission Yes No Partial remission Yes No
5 y postoperative
n
Mean (± SD) or %
n
Mean (± SD) or %
50 50 52
34 (±7) 54 (±17) 6.4 (±1.0)
46 46 51
35 (±35) 49 (±18) 6.6 (±0.9)
12 40
23 77
10 43
19 81
26 26
50 50
17 34
33 67
SD, Standard deviation.
limited because of the low number of total patients with complete remission (n = 9), it is worth noting that no patient who achieved complete remission of T2D had been treated preoperatively with insulin. Finally, 3 patients (6%) had a recurrence of their T2D at 5 years. One patient achieved remission only in 5 years, but not at 1-year follow-up. Substantial improvement occurred in the status of the co-morbidities after LSG after 5 years. For hypertension, 17 patients (53%) had complete remission and 11 patients (34%) improved their medication status. For dyslipidemia, complete remission was recorded in 15 patients (60%) and partial remission in 4 patients (16%). The sleep apnea resolved in 25 of 38 patients (66%). DISCUSSION A recent meta-analysis performed on the few retrospective, randomized, and prospective studies confirmed that LSG has a similar effect on T2D as RYGB.13,14 The literature about remission of T2D after LSG is controversial. Many studies have reported a resolution rate of 60% to 90% for diabetes after LSG in the short term,15,16 while other studies have reported limited efficacy.17-19 It is uncertain if the widely varied remission rates between the studies of LSG are related to patient characteristics (ethnicity), or to the absence of a clear definition of resolution of T2D. Another explanation could be that these studies used different definitions of “improvement,” “resolution,” “remission,” or “cure” of diabetes. Basso et al15 reported a resolution rate of 80% after a mean follow-up period of 3 years. The mean postoperative time at discontinuation of medication was 3.3 months for LSG compared with 12.6 months after gastric banding. Todkar et al16 reported a resolution rate of 72% at 3 years after LSG. Schauer et al17 reported a
randomized prospective trial of 150 obese patients with T2D to receive either intensive medical therapy alone or intensive medical therapy plus RYGB or LSG. After 36 months with a follow-up of 91%, complete remission was achieved by 5% of patients in the medical therapy group compared with 38% of those in the RYGB group (P < .001) and 24% of those in the LSG group (P = .01). Metabolic and weight-loss outcomes generally were similar in the 2 operative groups at 1 year, although some advantages of RYGB over SLG have emerged with greater follow-up, including a greater likelihood of reaching a A1C #7.0% (a therapeutic goal of the American Diabetes Association) with no need for diabetes medications, a decreased requirement for diabetes and cardiovascular medications, greater sustained decreases in weight and BMI, and greater improvement in quality of life. Within the same clinical trial, Schauer et al19 reported that 42% of patients after RYGB, 37% after LSG, and 12% with medical therapy alone achieved the primary endpoint of A1C #6% after 1 year. Complete remission rates in our study of 19% of diabetes at 5-year follow-up after LSG in our study are inferior to those presented previously. A different population, as well as a stricter definition of remission, could be the factors that can explain the discrepancies with other reports.15,16 Up to 40% of our studied population were men, in contrast with our overall bariatric operative cohort (21% men). The high men/women ratio in series of operative treatment of T2D is becoming frequent. The reason could be that men consider bariatric operation when they are sicker. The improvement of T2D was observed in 30 patients (58%) at 1 year and it was maintained for 27 patients (52%) at 5-year follow-up. One of the main limitations of our study is that prior to our study
ARTICLE IN PRESS Nedelcu et al 5
Surgery Volume j, Number j
period (2006–2010) a different definition was considered by the endocrinologist for complete remission of T2D. During that period, every patient with diabetes and HbA1C <6.5 in the first year after operation was considered in complete remission and diabetic treatment was stopped. Currently, these patients cannot be considered in complete remission with the new definition12 at 1-year follow-up. There is strong evidence that patients with a shorter history of diabetes may benefit more than those with a longer one, which is concordant with some studies that showed a direct correlation between decreased b-cell volume, trend to apoptosis, and death with a greater duration of T2D.20 Still, in our experience, even if this evidence has been noticed, no statistical difference was obtained at 1-year or 5-year follow-up, most likely secondary to our small cohort. Lee et al21 have proposed a specific diabetes surgery score (ABCD score) to help with patient selection. The score consists of 4 independent predictors of T2D remission: patient age, BMI, Cpeptide level, and duration of diabetes, a 4-point score (except age which accounts for 2 points). The score was validated on 157 patients operated between 2006 and 2013. Patients who had remission of T2D after bariatric operation had a greater ABCD score than those who did not achieve remission (7 ± 2 versus 5 ± 2, P < .05). Patients with a greater ABCD score were also at a greater likelihood of success in remission of T2D (from 0% in score 0 to 100% in score 10). The calculation of the ABCD score was not possible in our population due to lack of data, because our investigations did not evaluate preoperative C peptide levels. C peptide levels would have been interesting and useful, because several studies have since defined the presence of greater preoperative C peptide levels (C peptide >3 ng/mL) to be an important predictor of resolution of diabetes, thereby helping to define the best candidates for bariatric treatment of diabetes.22 Our study has several limitations. First, the follow-up was not continuous for all patients and, some follow-up data were available at only 1 time. Even though our sample size (46 patients) at 5-year follow-up (87%) are better than the existing data in the literature (39 of 56 patients; 70% follow-up reported by Keidar et al),23 we cannot make a robust conclusion regarding the changes in T2D. Another limitation is the retrospective character of our study, with no possibility to evaluate the C-peptide level and with no selection criteria for patient with T2D. Our study showed that some
severely obese patients with T2D may benefit less from the LSG, but hopefully in the future, using certain selection criteria, we hope to be better able to predict those patients who will achieve complete remission of their T2D. In our experience, although the LSG is not a definitive treatment of T2D, as also with the other bariatric procedures, LSG definitely offers significant (>70%) and durable (5 years) improvement of T2D. High preoperative A1C levels, older age, and a preoperative therapy consisting of insulin may be the factors predicting failure of resolution of T2D. REFERENCES 1. Shaw JE, Sicree RA, Zimmet PZ. Global estimates of the prevalence of diabetes for 2010 and 2030. Diabetes Res Clin Pract 2010;87:4-14. 2. Residori L, Garcia-Lorda P, Flancbaum L, Pi-Sunyer FX, Laferrere B. Prevalence of co-morbidities in obese patients before bariatric surgery: effect of race. Obes Surg 2003;13: 333-40. 3. Haute autorite de sante. Obesite: Rapport sur la prise en charge chirurgicale chez l’adulte. Saint-Denis La Plaine. HAS 2009. 4. Chan JM, Rimm EB, Colditz GA, Stampfer MJ, Willett WC. Obesity, fat distribution, and weight gain as risk factors for clinical diabetes in men. Diabetes Care 1994;17:961-9. 5. Sjostrom L, Narbro K, Sjostrom CD, et al. Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med 2007;357:741-52. 6. Buchwald H, Estok R, Fahrbach K, et al. Weight and type 2 diabetes after bariatric surgery: systematic review and metaanalysis. Am J Med 2009;122:248-56. 7. Colquitt JL, Picot J, Loveman E, Clegg AJ. Surgery for obesity. Cochrane Database Syst Rev 2009:CD003641. 8. Pories WJ, Albrecht RJ. Etiology of type II diabetes mellitus: role of the foregut. World J Surg 2001;25:527-31. 9. Shah PS, Todkar JS, Shah SS. Effectiveness of laparoscopic sleeve gastrectomy on glycemic control in obese Indians with type 2 diabetes mellitus. Surg Obes Relat Dis 2009;6: 138-41. 10. Lakdawala MA, Bhasker A, Mulchandani D, Goel S, Jain S. Comparison between the results of laparoscopic sleeve gastrectomy and laparoscopic Roux-en-Y gastric bypass in the Indian population: a retrospective 1 year study. Obes Surg 2010;20:1-6. 11. Nocca D, Guillaume F, Noel P, et al. Impact of laparoscopic sleeve gastrectomy and laparoscopic gastric bypass on HbA1c blood level and pharmacological treatment of type 2 diabetes mellitus in severe or morbidly obese patients. Results of a multicenter prospective study at 1 year. Obes Surg 2011;21:738-43. 12. Buse JB, Caprio S, Cefalu WT, et al. How do we define cure of diabetes? Diabetes Care 2009;32:2133-5. 13. Yip S, Plank LD, Murphy R. Gastric bypass and sleeve gastrectomy for type 2 diabetes: a systematic review and metaanalysis of outcomes. Obes Surg 2013;23:1994-2003. 14. Cho J-M, Kim H-J, Menzo EL, Park S, Szomstein S, Rosenthal RJ. Effect of sleeve gastrectomy on type 2 diabetes as an alternative treatment modality to Roux-en-Y gastric bypass: systemic review and meta-analysis. Surg Obes Relat Dis 2015;11:1273-80.
ARTICLE IN PRESS 6 Nedelcu et al
15. Abbatini F, Rizzello M, Casella G, et al. Long-term effects of laparoscopic sleeve gastrectomy, gastric bypass, and adjustable gastric banding on type 2 diabetes. Surg Endosc 2010;24:1005-10. 16. Todkar JS, Shah SS, Shah PS, Gangwani J. Long-term effects of laparoscopic sleeve gastrectomy in morbidly obese subjects with type 2 diabetes mellitus. Surg Obes Relat Dis 2010;6:142-5. 17. Schauer PR, Bhatt DL, Kirwan JP, et al. Bariatric surgery versus intensive medical therapy for diabetes---3-year outcomes. N Engl J Med 2014;370:2002-13. 18. Rosenthal R, Li X, Samuel S, Martinez P, Zheng C. Effects of sleeve gastrectomy on patients with diabetes mellitus. Surg Obes Relat Dis 2009;5:429-34. 19. Schauer PR, Kashyap SR, Wolski K, et al. Bariatric surgery versus intensive medical therapy in obese patients with diabetes. N Engl J Med 2012;366:1567-76.
Surgery j 2017
20. Klein S, Fontana L, Young VL, et al. Absence of an effect of liposuction on insulin action and risk factors for coronary heart disease. N Engl J Med 2004;350: 2549-57. 21. Lee WJ, Almulaifi A, Tsou JJ, Ser KH, Lee YC, Chen SC. Laparoscopic sleeve gastrectomy for type 2 diabetes mellitus: predicting the success by ABCD score. Surg Obes Relat Dis 2015;11:991-6. 22. Lee WJ, Ser KH, Chong K, et al. Laparoscopic sleeve gastrectomy for diabetes treatment in nonmorbidly obese patients: efficacy and change of insulin secretion. Surgery 2010;147: 664-9. 23. Golomb I, Ben David M, Glass A, Kolitz T, Keidar A. Longterm metabolic effects of laparoscopic sleeve gastrectomy. JAMA Surg 2015;150:1051-7.