Insurance-mandated preoperative diet and outcomes after bariatric surgery

Insurance-mandated preoperative diet and outcomes after bariatric surgery

Surgery for Obesity and Related Diseases ] (2018) 00–00 Original article Insurance-mandated preoperative diet and outcomes after bariatric surgery C...

266KB Sizes 0 Downloads 67 Views

Surgery for Obesity and Related Diseases ] (2018) 00–00

Original article

Insurance-mandated preoperative diet and outcomes after bariatric surgery Charles J. Keith Jr., M.D., Lauren E. Goss, M.S.P.H, Camille D. Blackledge, M.D., Richard D. Stahl, M.D., Jayleen Grams, M.D., Ph.D.* Department of Surgery, University of Alabama at Birmingham, Birmingham, Alabama Received October 15, 2017; accepted January 14, 2018

Abstract

Background: Despite a lack of demonstrated patient benefit, many insurance providers mandate a physician-supervised diet before financial coverage for bariatric surgery. Objectives: To compare weight loss between patients with versus without insurance mandating a preoperative diet. Setting: University hospital, United States. Methods: Retrospective study of all patients who underwent laparoscopic Roux-en-Y gastric bypass or sleeve gastrectomy over a 5-year period, stratified based on whether an insurancemandated physician-supervised diet was required. Weight loss outcomes at 6, 12, and 24 months postoperation were compared. Linear mixed-models and backward-stepwise selection were used. P o 0.05 was considered significant. Results: Of 284 patients, 225 (79%) were required and 59 (21%) were not required to complete a preoperative diet by their insurance provider. Patients without the requirement had a shorter time to operation from initial consultation (P ¼ .04), were older (P o .01), and were more likely to have government-sponsored insurance (P o .01). There was no difference in preoperative weight or body mass index or co-morbidities. In unadjusted models, percent excess weight loss was superior in the group without an insurance-mandated diet at 12 (P ¼ .050) and 24 (P ¼ .045) months. In adjusted analyses, this group also had greater percent excess weight loss at 6 (P o .001), 12 (P o .001), and 24 (P o .001) months; percent total weight loss at 24 months (P ¼ .004); and change in body mass index at 6 (P ¼ .032) and 24 (P ¼ .007) months. There was no difference in length of stay or complication rates. Conclusions: Insurance-mandated preoperative diets delay treatment and may lead to inferior weight loss. (Surg Obes Relat Dis 2018;]:00–00.) Published by Elsevier Inc. on behalf of American Society for Metabolic and Bariatric Surgery.

Keywords:

Insurance; Diet; Bariatric surgery; Weight loss; Outcomes

Obesity has reached epidemic status over the past several decades, and studies predict the problem to continue growing in the foreseeable future [1,2]. Obesity threatens health in terms of increased risk of diabetes, cardiovascular disease, and certain cancers, and it has been shown to Paper presented at 33rd Annual Meeting of the American Society for Metabolic and Bariatric Surgery, November 4, 2016, New Orleans, LA. * Correspondence: Jayleen Grams, M.D, Ph.D., 1720 2nd Ave South, KB 401, Birmingham, AL 35294. E-mail: [email protected]

increase all-cause mortality [2,3]. It also has tremendous economic impact and has been estimated to account for 49% of all medical expenditures [2]. In fact, each point increase in body mass index (BMI) is associated with an increase in medical and pharmaceutical costs. Bariatric surgery is the best available intervention to achieve and maintain weight loss long term. Additionally, surgery has been shown to improve or effect remission of co-morbidities, including diabetes, hypertension, obstructive sleep apnea, and dyslipidemia. Despite its significant benefits, bariatric surgery is not easily accessible for all

https://doi.org/10.1016/j.soard.2018.01.017 1550-7289/Published by Elsevier Inc. on behalf of American Society for Metabolic and Bariatric Surgery.

2

C.J. Keith Jr. et al. / Surgery for Obesity and Related Diseases ] (2018) 00–00

patients in need. One such barrier to treatment is an insurance-mandated physician-supervised preoperative diet. It is presumable that insurance providers require preoperative diets to improve patients’ weight loss success and reduce the risk of perioperative complications. However, studies have suggested there is no difference in pre- or postoperative weight loss outcomes when comparing patients who had required preoperative diets versus those who did not, and the preoperative diet only delays treatment and leads to increased rates of attrition from bariatric programs [4–7]. Furthermore, patients denied surgery by their insurance provider develop new diagnoses of obesityrelated co-morbidities, suffer from worsening glycemic control with increased need for new diabetic medications, and have decreased overall survival compared with patients who underwent bariatric surgery [8,9]. These results have challenged the practice of insurance-mandated preoperative diets. Recently, the American Society for Metabolic and Bariatric Surgery updated its position statement on the issue, again stating that there is no medical evidence of benefit to patients [10]. The purpose of this study was to compare weight loss outcomes of patients who were required to participate in a physician-supervised preoperative diet by their insurance provider with those who were not subjected to this mandate. Methods The institutional review board approved this study. A retrospective review of all patients who underwent laparoscopic Roux-en-Y gastric bypass or laparoscopic sleeve gastrectomy at a university hospital from January 2009 through December 2013 was performed from a prospectively maintained database. Exclusion criteria included open operations and age o19 years. Patients were stratified into 2 groups based on whether they were required by their insurance provider to complete a physician-supervised preoperative diet before financial coverage. Preoperative characteristics included demographic information, insurance type (private versus government-sponsored), and presence of obesity-related co-morbidities. Preoperative weight and BMI were reported for both the “initial visit” (very first consultation with the surgeon) and the “preoperative visit” (visit closest to the time of operation, usually within 2 wk of operation). Weight changes were determined from the initial visit to the immediate preoperative visit. Postoperative outcomes including length of hospital stay and complications, and weight loss outcomes of percent excess weight loss (%EWL), percent total weight loss (%TWL), and change in BMI (ΔBMI) were compared between the groups. Postoperative weight loss outcomes were calculated using weights from the preoperative visit, defined above. Descriptive statistics were used to compare patient characteristics and insurance-mandated diet status. χ2 and Student’s t tests were used to compare categoric and

continuous variables, respectively. Generalized linear models and mixed-effects multivariate regression with a spline fit adjustment for time since operation was used to model %TWL, %EWL, and ΔBMI, both with and without adjustment for factors independently associated with weight loss outcomes. Data are reported as medians and interquartile range rather than means and standard deviations due to a nonnormal distribution for many of the continuous variables studied. Statistical analyses were conducted using SAS Version 9.4 (SAS Institute Inc., Cary, NC, USA) and R (Version3.13.0, The R Development Core Team, Vienna, Austria); statistical significance was determined by P o .05. Results A total of 284 patients were included in the study. Of these, 225 (79%) were required by insurance providers to participate in a physician-supervised preoperative weight loss program before approval for surgery, while 59 patients (21%) were not held to this mandate. For the entire cohort, median age was 44 years, and the majority was female (78%) and Caucasian (64%). Obesity-related co-morbidities included diabetes (46%), hypertension (72%), obstructive sleep apnea (66%), and gastroesophageal reflux disease (51%; Table 1). Median weights at initial consultation and immediate preoperative visits were 131.5 kg (BMI 47.1 kg/m2) and 134.7 kg (BMI 47.8 kg/m2), respectively (Table 2). At the time of operation, 240 patients (84.5%) underwent Roux-en-Y gastric bypass while 44 (15.5%) had SG. After stratification, patients in the insurance-mandated diet group were younger (median age 43 years versus 51 years; P o .001) and more likely to have private insurance rather than a government-sponsored plan (P o .001). There was no difference in race, sex, or presence of obesity-related co-morbidities (Table 1). At initial consultation, the median weight of those who had an insurance-mandated diet was 131.9 kg (BMI 47.2 kg/m2), which increased to 134.7 kg (BMI 47.9 kg/m2) by the time of their immediate preoperative visit. Patients who did not have the insurance-mandated diet weighed 130.2 kg (BMI 46.5 kg/m2) at initial consultation and 133.8 kg (BMI 46.8 kg/m2) at immediate preoperative visit. There was no significant difference in initial consultation or preoperative weights or BMIs between the groups, or in %EWL or % TWL from initial consultation to immediate preoperative visit. However, time from initial consultation to operation was significantly delayed for the insurance-mandated diet group (218 versus 154 days; P ¼ .037). Operation performed, length of hospital stay, and complication rates were similar (Table 2). In unadjusted comparisons, there was no significant difference in weight, BMI, %TWL, or ΔBMI between the groups at 6 months, 1, or 2 years postoperation (Table 3), but median %EWL was superior at 1 year for the group

Insurance-mandated Preoperative Diet and Outcomes / Surgery for Obesity and Related Diseases ] (2018) 00–00

25%; P ¼ .004), and greater reduction in BMI (−14.9 versus −10.7; P ¼ .007) than patients required to participate in a physician-monitored diet by their insurance provider (Table 4). A comparison of the weight loss trajectories for the groups is displayed in Fig. 1.

Table 1 Patient demographic characteristics

Number Age, yr Sex Male Female Race White Black Co-morbidities Diabetes HTN OSA GERD Insurance type Private Government

Total

InsuranceNo insurance- P value mandated diet mandated diet

284 (100) 44 (36–52)

225 (79) 43 (36–51)

59 (21) 51 (40–58)

64 (22) 220 (78)

49 (22) 176 (78)

15 (25) 44 (75)

.551

181 (64) 103 (36)

145 (64) 80 (36)

36 (61) 23 (39)

.626

132 206 187 144

102 161 142 114

30 45 45 30

(51) (76) (76) (51)

.450 .470 .058 .980

9 (15) 50 (85)

o.001

(46) (72) (66) (51)

225 (79) 59 (21)

(45) (72) (63) (51)

216 (96) 9 (4)

3

o.001

Discussion Bariatric surgery is the best available treatment for obesity and obesity-related co-morbidities. Many patients who are being considered for bariatric surgery have struggled with obesity for years and have tried innumerable weight loss methods, all without longstanding success. Despite this personal history, along with no class I data supporting any benefit of an additional insurance-mandated preoperative diet, many insurance providers continue to require it. The interval of these diets can be 4 to 6 months, and the specific requirements can be quite strict. For example, if each monthly visit does not fall into the prescribed time frame or if the physician clinic note does not contain all prescribed “check boxes,” the entire process may need to be restarted from month 0. Thus, insurancemandated diets may increase attrition or impose significant barriers to treatment that delay care. Our study supports that insurance-mandated diets do not result in preoperative weight loss and in fact delay care. Furthermore, our study suggests that insurance-mandated diets do not decrease the length of stay or improve the safety of the operation and, if anything, may have a negative impact on weight loss outcomes. For our population, the number of patients impacted by this policy is significant, as 79% of our patients were required by their insurance company to have a physician-supervised diet.

HTN ¼ hypertension; OSA ¼ obstructive sleep apnea; GERD ¼ gastroesophageal reflux disease. Data presented as n (%) or median (interquartile range).

who did not have a diet mandated by their insurance provider (67.4% versus 59.4%; P ¼ .050). After adjusting for age, sex, race, operation, and preoperative co-morbidities, patients who did not have a preoperative diet requirement achieved a greater reduction in median %EWL (52.9% versus 49.7%; P o .001) and BMI (−12.2 versus −10.9; P ¼ .032) at 6 months. At 1 year, this group reached a significantly greater median %EWL of 64.9% versus 61.1% in the insurance-mandated diet group (P o .001). At 2 years, they achieved a greater median %EWL (65.6% versus 55.5%; P o .001), greater %TWL (33% versus Table 2 Perioperative data

Initial visit Weight, kg BMI, kg/m2 Time to operation, d Preoperative visit Weight, kg BMI, kg/m2 %EWL from initial visit %TWL from initial visit ΔBMI since initial visit Time to operation, d Operation RYGB SG Postoperative care Length of stay Complications

Total

Insurance-mandated diet

No insurance-mandated diet

P value

131.5 (118.4–149.7) 47.1 (43.3–52.0) 205 (119–317)

131.9 (117–149.7) 47.2 (36.0–51.0) 218 (124–329)

130.2 (121.1–149.2) 46.5 (41.8–52.2) 154 (103–261)

.766 .607 .037

134.7 47.8 0 0 0.2 15

134.7 47.9 0 0 0.3 15

133.8 46.8 0 0 0.0 16

.687 .710 .427 .783 .724 .443

(119.8–152) (43.6–52.8) (−.5 to .4) (−1.9 to 1.0) (−.8 to 1.4) (8–20)

(119.3–152) (43.8–52.7) (−.3 to .8) (−2.1 to .9) (−.7 to 1.4) (5.5–20)

(119.8–152.4) (43.2–53.1) (−.5 to .4) (−1.1 to 1.8) (−.8 to 1.2) (11–22)

240 (84.5) 44 (15.5)

191 (84.9) 34 (15.1)

49 (83.0) 10 (17.0)

.728

2 (2–2) 11 (3.9)

2 (2–2) 9 (4.0)

2 (2–2) 2 (3.4)

.932 .835

BMI ¼ body mass index; %EWL ¼ percent excess weight loss; %TWL ¼ percent total weight loss; ΔBMI ¼ change in BMI; RYGB ¼ roux-en-Y gastric bypass; SG ¼ sleeve gastrectomy. Data presented as median (interquartile range) or n (%).

C.J. Keith Jr. et al. / Surgery for Obesity and Related Diseases ] (2018) 00–00

4

Table 3 Unadjusted comparisons of weight loss outcomes Total

Insurancemandated diet

No insurancemandated diet

6 mo n 281 222 59 Weight, kg 97 (84–111) 97 (83–111) 94 (87–111) BMI 34.4 (31.2–38.6) 34.5 (31.2–39.0) 33.9 (30.8–38.6) %EWL 47.2 (40.6–56.9) 45.9 (39.3–56.7) 52.7 (43–56.9) %TWL 26.2 (22.4–29.9) 26.1 (22.1–29.8) 27.3 (22.8–29.9) ΔBMI 12.4 (10.6–14.4) 12.6 (10.6–14.2) 12.1 (10.6–15.7) 1 yr n 220 174 46 Weight, kg 190 (168–226) 189 (166–226) 195 (174–218) BMI 31.3 (27.6–35.4) 31.6 (28–36.2) 29.9 (27.2–33.4) %EWL 61.2 (49.8–71.9) 59.4 (48.7–71.1) 67.4 (58–75.4) %TWL 33.4 (27.2–38.5) 33.2 (26.6–37.8) 33.7 (28.3–40.6) ΔBMI 15.8 (12.6–19.4) 15.8 (12.8–19.0) 15.5 (12.0–20.1) 2 yr n 127 101 26 Weight, kg 182 (160–222) 188 (160–226) 179 (167–218) BMI 30.7 (27.0–34.4) 31.1 (27.6–35.7) 30.4 (26.0–32.5) %EWL 64.5 (45–72.7) 61.9 (42.2–71.4) 69.1 (61.1–87.3) %TWL 35.3 (24.3–40.7) 31.5 (22.8–39.5) 36.7 (30.0–41.9) ΔBMI 15.7 (12.6–19.4) 15.7 (11.2–19.2) 16.6 (13.8–22.1)

P value

.991 .727 .163 .180 .183

.843 .164 .050 .284 .452

.780 .214 .045 .073 .090

BMI ¼ body mass index; %EWL ¼ percent excess weight loss; %TWL ¼ percent total weight loss; ΔBMI ¼ change in BMI Data presented as median (interquartile range).

The literature contains several studies evaluating the impact of preoperative weight change on outcomes after bariatric surgery. Our institution recently reported that weight loss outcomes 2 years after laparoscopic Roux-en-Y gastric bypass were not associated with preoperative weight change [11]. However, this study did not stratify patients based on whether they had a preoperative diet requirement for coverage by their insurance provider, and there have been relatively few studies that specifically address outcomes based on whether a preoperative diet is mandated by insurance providers. Horwitz et al. [5] analyzed 560 patients and found no difference in weight loss at 1 or 2 years postoperation. Kuwada et al. [4] evaluated pre- and postoperative weight loss in 440 patients stratified by requirement of participation in a medical weight loss program by their insurance provider. They also reported no difference in weight loss outcomes preoperatively or at 6 and 12 months postoperation. The study by Kuwada et al. [4] also demonstrated delayed treatment for patients required by their insurance provider to have a preoperative diet. Our results are consistent with these previous reports and showed that insurance-mandated diets do not result in improved postoperative weight loss outcomes, do not decrease complications, and do result in a delay in treatment. This delay in treatment could have significant impact on patients’ overall health, because the timing of bariatric surgery and its effects on co-morbidities and weight loss have drawn attention in the literature as well. To determine predictive factors of metabolic nonresponse to bariatric surgery in patients with type 2 diabetes, Jurowich et al.

[12] found that longer preoperative duration of diabetes, increasing hemoglobin A1C, reliance on multidrug therapy, and increased age were associated with nonresponse to operation. Hayes et al. [13] agree that patients whose disease has progressed to poorer glycemic control suffer inferior diabetes outcomes after bariatric surgery. These studies, in addition to those by Pories et al. [14] and Schauer et al. [15,16], suggest that earlier operation could improve diabetes outcomes by addressing the disease before it progresses to poor control. The recently updated outcomes of the STAMPEDE trial at 5 years further this argument as goal glycated hemoglobin levels were less frequently met after operation for patients whose disease progressed to poorer control preoperation [16]. Although our study did not directly evaluate co-morbidity–related outcomes, the patients who had insurance providers that mandated a preoperative diet did have a significant delay in treatment of their obesity and, in turn, their obesity-related co-morbidities. This delay in treatment may have contributed to the significant difference in weight loss between the 2 groups at 2 years, with the insurance-mandated preoperative diet group having a delay in treatment and inferior weight loss outcomes. It is presumed that the rationale for insurance-mandated preoperative diets includes reduction of perioperative risks and complications with preoperative weight loss. However, there are several issues with this reasoning. First, patients required to complete a physician-supervised diet by their Table 4 Adjusted comparisons of weight loss outcomes Insurancemandated diet 6 mo n Weight, kg BMI %EWL %TWL ΔBMI 1 yr n Weight, kg BMI %EWL %TWL ΔBMI 2 yr n Weight, kg BMI %EWL %TWL ΔBMI

No insurancemandated diet

P value

214 34.5 49.7 23.2 10.9

222 (182–244) (31.2–39.0) (49.1–50.2) (21.8–24.6) (10.1–11.7)

208 33.9 52.9 24.9 12.2

59 (191–244) (30.8–38.6) (52–53.7) (22.9–26.9) (11.1–13.3)

.991 .727 o.001 .111 .032

189 31.6 61.1 28.7 13.5

174 (166–226) (28.0–36.2) (60.3–61.8) (26.9–30.5) (12.2–14.8)

195 29.9 64.9 31.3 15

46 (174–218) (27.2–33.4) (63.9–66.0) (28.6–34.0) (13.2–16.8)

.843 .165 o.001 .069 .097

188 31.1 55.5 25.0 10.7

101 (160–226) (27.6–35.7) (54.2–56.8) (21.1–28.9) (8.3–13.1)

179 30.4 65.6 33.0 14.9

26 (167–218) (26.0–32.5) (63.9–67.3) (27.9–38.1) (12.0–17.8)

.780 .214 o .001 .004 .007

BMI ¼ body mass index; %EWL ¼ percent excess weight loss; %TWL ¼ percent total weight loss; ΔBMI ¼ change in BMI Data presented as median (interquartile range). Adjusted for age, sex, race, operation, and co-morbidities.

Insurance-mandated Preoperative Diet and Outcomes / Surgery for Obesity and Related Diseases ] (2018) 00–00

5

our study has a relatively small sample size, short-term follow-up, and a large number of patients who were lost to follow-up at 2 years. However, our follow-up rates are consistent with the bariatric literature and there was no difference in follow-up rates between the 2 groups (77% versus 78% follow-up at 12 mo and 49% versus 44% follow-up at 24 mo in the group with the insurancemandated diet versus without, respectively). Conclusions

Fig. 1. Weight loss trajectories by group. P o 0.05 considered significant. %TWL ¼ percent total weight loss.

insurance provider may not achieve greater preoperative weight loss. Our patients who were required to participate in a physician-supervised diet were not required to lose a prescribed amount of weight; they were only mandated to have a prescribed interval of a supervised diet. In fact, our patients did not lose weight in this interval and they tended to gain weight during this time. The patients in the study by Kuwada et al. [4] did not show a difference in preoperative weight loss between groups stratified by whether an insurance-mandated diet was required either. Second, there was no difference in length of stay in the hospital or complication rates postoperation between the groups in our study. It is notable that even if preoperative weight loss is required or achieved, it may not be associated with lower complication rates. In our institution’s previous report comparing outcomes of patients based on preoperative weight change, there was no significant decrease in complication rates for patients with a higher degree of weight loss preoperatively [11]. Alami et al. [17] recently published results of a prospective randomized trial comparing outcomes of patients held to a preoperative weight loss requirement with those of patients not held to this standard. In their study, there was no difference in complication rates or conversions between the groups. Finally, in the absence of demonstrated benefit of an insurance-mandated diet, there may actually be harm in delaying or preventing treatment due to attrition of obesity and obesity-associated medical problems. We acknowledge there are limitations of this study. It is a retrospective analysis of prospectively collected data, so it is subject to errors in chart abstraction or incomplete data. For example, we only had data on those patients who ultimately had a bariatric operation, so we were unable to capture attrition rates between the 2 groups. Also, our study was unable to determine whether patients’ obesity-related co-morbidities improved or worsened during the interval between their initial consultation and the operation. Finally,

Most patients who pursue bariatric surgery have struggled with their weight for years and have attempted innumerable weight loss methods, all without longstanding success. Insurance-mandated preoperative diets delay treatment of obesity and its co-morbidities, do not improve length of stay or complication rates, and do not improve weight loss outcomes. In fact, delayed treatment may result in inferior weight loss outcomes and may be harmful to patients’ overall health. Thus, there is no value to the mandate by insurance providers for a “one last attempt” at a preoperative diet before bariatric surgery. Such mandates only serve as barriers to treatment. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. References [1] Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL. Trends in obesity among adults in the United States, 2005 to 2014. JAMA 2016;315(21):2284–91. [2] Finkelstein EA. How big of a problem is obesity? Surg Obes Relat Dis 2014;10(4):569–70. [3] Flegal KM, Kit BK, Orpana H, Graubard BI. Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA 2013;309(1):71–82. [4] Kuwada TS, Richardson S, El Chaar M, et al. Insurance-mandated medical programs before bariatric surgery: do good things come to those who wait? Surg Obes Relat Dis 2011;7(4):526–30. [5] Horwitz D, Saunders JK, Ude-Welcome A, Parikh M. Insurancemandated medical weight management before bariatric surgery. Surg Obes Relat Dis 2016;12(3):496–9. [6] Jamal M, Demaria E, Johnson J, et al. Insurance-mandated preoperative dietary counseling does not improve outcome and increases dropout rates in patients considering gastric bypass surgery for morbid obesity. Surg Obes Relat Dis 2006;2(2):122–7. [7] Love KM, Mehaffey JH, Safavian D, et al. Bariatric surgery insurance requirements independently predict surgery dropout. Surg Obes Relat Dis 2017;13(5):871–6. [8] Al Harakeh AB, Burkhamer KJ, Kallies KJ, Mathiason MA, Kothari SN. Natural history and metabolic consequences of morbid obesity for patients denied coverage for bariatric surgery. Surg Obes Relat Dis 2010;6(6):591–6. [9] MacDonald KG, Long SD, Swanson MS, et al. The gastric bypass operation reduces the progression and mortality of non-insulindependent diabetes mellitus. J Gastrointest Surg 1997;1(3):213–20.

6

C.J. Keith Jr. et al. / Surgery for Obesity and Related Diseases ] (2018) 00–00

[10] Kim JJ, Rogers AM, Ballem N, Schirmer B. ASMBS updated position statement on insurance mandated preoperative weight loss requirements. Surg Obes Relat Dis 2016;12(5):955–9. [11] Blackledge C, Graham LA, Gullick AA, Richman J, Stahl RD, Grams J. Outcomes associated with preoperative weight loss after laparoscopic Roux-en-Y gastric bypass. Surg Endosc 2016;30 (11):5077–83. [12] Jurowich C, Thalheimer A, Hartmann D, et al. Improvement of type 2 diabetes mellitus after bariatric surgery – who fails in the early postoperative course? Obes Surg 2012;22(10):1521–6. [13] Hayes MT, Hunt LA, Foo J, Tychinskaya Y, Stubbs RS. A model for predicting the resolution of diabetes in severely obese subjects following Roux-en-Y gastric bypass surgery. Obes Surg 2011;21 (7):910–6.

[14] Pories WJ, Swanson MS, MacDonald KG, et al. Who would have thought it? An operation proves to be the most effective therapy for adult-onset diabetes mellitus. Ann Surg 1995;222 (3):339–52. [15] Schauer PR, Burquera B, Ikramuddin S, et al. Effect of laparoscopic Roux-en-Y gastric bypass on type 2 diabetes mellitus. Ann Surg 2003;238(4):467–84. [16] Schauer PR, Bhatt DL, Kirwan JP, et al. Bariatric surgery versus intensive medical therapy for diabetes – 5-year outcomes. N Engl J Med 2017;376(7):641–51. [17] Alami RS, Morton JM, Schuster BR, et al. Is there a benefit to preoperative weight loss in gastric bypass patients? A prospective randomized trial. Surg Obes Relat Dis 2017;3(2):141–5.