Abstract
Background and Aims:
The long-term efficacy of laparoscopic Roux-en-Y gastric bypass in the treatment of morbid obesity has already been demonstrated. Laparoscopic sleeve gastrectomy has shown promising short-term results, but the long-term efficacy is still unclear. The aim of this prospective randomized multicenter study is to compare the results of Roux-en-Y gastric bypass and sleeve gastrectomy.
Material and Methods:
A total of 240 morbidly obese patients were randomized to undergo either Roux-en-Y gastric bypass or sleeve gastrectomy. The primary end point of study was weight loss, and the secondary end points were resolution of comorbidities and morbidity. The short-term results at 6 months were analyzed.
Results:
The mean excess weight loss at 6 months was 49.2% in the sleeve gastrectomy group and 52.9% in the Roux-en-Y gastric bypass group (p = 0.086). Type 2 diabetes was resolved or improved in 84.3% of patients in the sleeve gastrectomy group and 93.3% in the Roux-en-Y gastric bypass group (p = 0.585). The corresponding results for arterial hypertension were 76.8% and 81.9% (p = 0.707) and for hypercholesterolemia 64.1% and 69.0% (p = 0.485). There was no mortality at 6 months. There was one major complication following sleeve gastrectomy and two after Roux-en-Y gastric bypass (p = 0.531). Eight sleeve gastrectomy patients and 11 Roux-en-Y gastric bypass patients had minor complications (p = 0.403).
Conclusion:
The short-term results of sleeve gastrectomy and Roux-en-Y gastric bypass regarding weight loss, resolution of obesity-related comorbidities and complications were not different at 6 months.
Introduction
Bariatric surgery is the most effective treatment for morbid obesity and the long-term results regarding weight loss and improvement of obesity-related comorbidities have been well established (1, 2). Laparoscopic Roux-en-Y gastric bypass (RYGB) is considered the gold standard in bariatric surgery providing significant and sustained weight loss resulting in resolution or improvement of obesity-related comorbidities at long-term follow-up (3, 4). However, during the last few years, laparoscopic sleeve gastrectomy (SG) has increasingly been performed as a primary procedure. SG was first described in the 1990s by Marceau et al. (5) and Hess and Hess (6) as part of the duodenal switch operation, then introduced as a first step of a two-stage operation for high-risk super obese patients (7). The promising short-term results of SG have somewhat altered the paradigm for SG from a two-stage procedure to a stand-alone definitive bariatric procedure (8–12). As RYGB is the gold standard in bariatric surgery, all other procedures should be compared with RYGB. However, to our knowledge, there are only few reports of randomized prospective studies comparing RYGB and SG with a small number of patients and a short follow-up (13–16). The available long-term results of SG are also scarce (9, 11, 17–19).
SG is perceived less invasive, technically simpler and easier to perform compared with RYGB. The possible long-term benefits of SG include an intact gastrointestinal tract, the absence of internal hernias, and the lack of malabsorption requiring lifelong follow-up of nutritional status. SG could thus become the procedure of choice in treating morbid obesity provided that the long-term results of SG are comparable with RYGB regarding weight loss, the resolution of comorbidities, and improvement in the quality of life. The aim of this prospective randomized multicenter SLEEVEPASS study is to compare RYGB and SG in the treatment of morbid obesity and to report our short-term outcome at 6 months regarding weight loss, resolution of obesity associated comorbidities, and overall morbidity.
Material and Methods
SLEEVEPASS study was carried out at three tertiary referral hospitals in Finland. Between April 2008 and June 2010, a total of 240 patients assigned for surgical treatment for morbid obesity were randomized to undergo either laparoscopic SG or RYGB. The randomization was done by a closed envelope method at the preoperative visit. The study was approved by the Ethics Committee of Turku University Hospital, and a written informed consent was obtained from all patients. The inclusion criteria for the study were (1) body mass index (BMI) ≥ 40 or BMI ≥ 35 with a significant comorbidity associated with morbid obesity (type 2 diabetes, hypertension, obstructive sleep apnea, dyslipidemia, and arthrosis), (2) age 18–60 years, and (3) previous successfully instituted and supervised but failed adequate diet and exercise program. The exclusion criteria included BMI > 60, significant psychiatric disorder, severe eating disorder (binge eating), active alcohol or substance abuse, active gastric ulcer disease, difficult gastroesophageal reflux disease (GERD) with a large hiatal hernia, and previous bariatric surgery. Both study groups were similar regarding age, gender, BMI, and comorbidities. The preoperative evaluation and surgical techniques have been described in detail together with the 30-day preliminary results (20).
The primary end point of the SLEEVEPASS trial is weight loss. The secondary end points assess the associated resolution of obesity-related comorbidities, the improvement of quality of life, and the overall morbidity and mortality of the procedures. According to our study protocol, the patients came to postoperative outpatient control visits at 3 and 6 months. All the data concerning weight loss, state of obesity-related comorbidities, and possible complications were thoroughly recorded. Postoperatively obesity-related comorbidities were classified as “persisting” (medication same as preoperatively), “improved” (reduction in medication), or “resolved” (no more need for medication) after a consulting endocrinologist visit. Postoperative complications were classified as major or minor; morbidity resulting in death or a reoperation, a hospital stay exceeding 7 days, or a need for blood transfusions of four or more units constituted a major complication. All other postoperative problems were evaluated as minor complications.
Statistical Analyses
Sample size calculations were based on a test of mean difference between bypass and sleeve procedures in excessive weight loss (EWL) in 1 year. A mean of 60 and standard deviation (SD) of 20 in the bypass group were assumed and α-level of 0.05 was used in calculations. By using an equivalence design with a margin of equivalence of 15% (−9 to 9), a sample size of 108 patients per group is needed for 90% power. When a 10% dropout is taken into account, 120 patients per group were enrolled in the study.
Categorical variables were characterized using frequencies and percents. For normally distributed continuous variables mean, range, and SD and for non-normally distributed continuous variables median and range were used. Associations between categorical variables were statistically tested using Pearson’s χ2 test, and for small frequencies, Fisher’s exact test was used. Differences between groups in normally distributed continuous variables were tested using independent samples t test, and for non-normally distributed variables, the Mann–Whitney U test was used. The p-values < 0.05 were considered statistically significant. Statistical analyses were performed using SAS System for Windows ver. 9.2 (SAS Institute, Cary, NC, USA).
Results
Altogether 121 patients were operated in the SG group and 117 in the RYGB group. The follow-up results at 6 months were obtained from 119 patients in the SG group (98.3% follow-up rate) and 111 patients in the RYGB group (93.3% follow-up rate). The CONSORT 6-month flowchart is presented in Fig. 1. The mean excess weight loss (EWL) at 6 months was 51.0% (range: 10.7%–94.8%, SD: 16.4%), and there was no statistical difference between the study groups (p = 0.086). The mean postoperative EWL and median postoperative BMI at 3 and 6 months are presented in detail in Figs 2 and 3.
CONSORT flowchart for the trial.
Mean percent EWL at 3 and 6 months. Values are mean (minimum–maximum; SD).
Median BMI at baseline and 3 and 6 months. Values are median (minimum–maximum).
Type 2 diabetes was resolved or improved in 84.3% of patients in the SG group and 93.3% in the RYGB group (p = 0.585). The corresponding results for arterial hypertension were 76.8% and 81.9% (p = 0.707) and for hypercholesterolemia 64.1% and 69.0% (p = 0.485). The detailed data on comorbidity resolution are presented in Figs 4 to 6.
Percent resolution of DM2 at 3 and 6 months.
Percent resolution of arterial hypertension at 3 and 6 months.
Percent resolution of hypercholesterolemia at 3 and 6 months.
There was no mortality at 6 months. All the major and minor postoperative complications between 30 days and 6 months are presented in detail in Table 1. There was one (0.8%) major complication following SG and two (1.8%) after RYGB (p = 0.531). Eight (6.7%) SG patients and 11 (9.9%) RYGB patients had minor complications (p = 0.403). The readmission rate was 4.2% for SG and 2.7% for RYGB. There was one reoperation in the RYGB group between 1 and 6 months postoperatively due to incarcerated incisional hernia at the trocar site successfully treated by relaparoscopy.
Detailed comparison of major and minor postoperative complications after laparoscopic SG and RYGB between 30 days and 6 months.
SG: sleeve gastrectomy; RYGB: Roux-en-Y gastric bypass; GEJ: gastroesophageal junction; GJA: gastojejunal anastomosis.
Values are n (%).
Discussion
RYGB is a safe and effective bariatric procedure with excellent long-term results regarding weight loss and resolution of obesity-related comorbidities (3, 21). However, in recent years, SG has attracted vast surgical interest because of its perceived technical simplicity and feasibility compared with RYGB. In addition to this, SG also has possible long-term advantages over RYGB such as preservation of endoscopic access to the upper gastrointestinal tract, prevention of the dumping syndrome by pylorus preservation, normal intestinal absorption, and avoiding the risk of internal herniation associated with RYGB anastomosis.
Most of the weight loss after bariatric surgery takes place during the first 6 months after the operation. After this, the weight loss gradually slows down continuing up to 2 years postoperatively often followed by a slight regain in weight in the next few years. The short-term follow-up of 6 months may be assessed as a weak point of our study. However, considering this paradigm and schedule of postoperative weight loss, we evaluated the weight loss results and resolution of comorbidities between the study groups to be of importance already at this short-term follow-up especially as randomized studies comparing these bariatric procedures are so scarce. This weight loss tendency can be identified after all bariatric procedures, and despite this slight postoperative weight regain, the long-term weight loss results of bariatric surgery are clearly superior compared to conservative treatment as demonstrated by Sjöström et al. (4) in the Swedish obese subject (SOS) study.
In this prospective randomized study, we have demonstrated that at 6 months postoperatively, the EWL of SG and RYGB is similar. There are reports from nonrandomized trials on tendency for weight regain after SG at 3–5 years following surgery (9, 11). However, this is a general phenomenon following bariatric surgery, and it is not specifically related only to SG. On the contrary, Boza et al. (12) have recently reported excellent results of 1000 consecutive SG procedures with a mean EWL of 84.5% at 3-year follow-up and with minimal weight regain after the first postoperative year. The available data on EWL following SG are largely based on follow-up of only up to 2 years. In a recent systematic review by Fischer et al., a total of 12,129 SG patients were analyzed and 3-year follow-up data were available in only eight studies resulting in a total of 363 patients (3.0%), and 5-year follow-up results were available from only 64 patients (0.5%). The mean EWL of 49.2% in our SG study group at 6 months is comparable with the corresponding 45.2% from 42 articles included in this review article (17).
The technical aspects of SG somewhat lack standardization. The use of a smaller bougie size as calibration during the operation has been reported to associate with a better weight loss and resolution of comorbidities, but on the contrary also with a higher leak rate (22–24). Similarly, the preservation of the antrum and the use of reinforced staple lines have been controversial issues. Recently, an expert panel consensus statement on best practice guidelines for SG was published addressing several of these technical issues as well as indications and contraindications for SG and also evaluating both management and prevention of complications (25). As there is variety in the published data regarding these technical details of SG, we aimed to standardize our procedures, and all the sleeves were created narrow using 33–35 Fr bougie, the distal resection was started 4–6 cm proximal to the pylorus, and all the staple lines were reinforced (20).
From the obesity-related comorbidities type 2 diabetes, arterial hypertension and hypercholesterolemia were analyzed. The state of these comorbid conditions was easy to evaluate in our study as all of the patients underwent regular postoperative endocrinologist evaluation. In this prospective randomized study, we have demonstrated that at 6-month follow-up, the improvement or resolution of these obesity-related comorbidities is equally good between SG and RYGB. Similar results have also been shown in previous randomized or prospective studies only with a smaller number of patients (13, 15, 26).
In Finland as a whole, bariatric surgery has only been performed on a larger scale for the last few years. When our SLEEVEPASS study was initiated in 2008, there were only 430 bariatric procedures performed that year in the whole country, but the number of operations has since been steadily growing. There are no hospitals specialized only in bariatric surgery in Finland, and none of the hospitals performing bariatric surgery can be considered a high-volume center. All of the surgeons performing the procedures were experienced laparoscopists with relative personal experience on bariatric surgery. These factors present a weakness in our study as both the individual and the institutional learning curve effect may have a role in the relatively high complication rates when compared with some previous studies (27–29). For example, the overall morbidity in our SG study group is 20.7% when the early ≤30-day results (20) are also taken into account, and this is somewhat higher than the 0.0%–17.7% morbidity rates presented in the studies included in the review article by Fischer et al. (17). However, the direct and valid comparison of complication rates in different studies is difficult as the definition and reporting of complications varies significantly emphasizing the need for standardized and concise international criteria for grading postoperative complications after bariatric surgery. On the contrary, the reported complication rates in this study also present a strong element of our trial as special emphasis was focused on the thorough prospective recording of all major and minor complications.
A strong advantage of this study is the randomized prospective multicenter setting. The group of patients in our study can be considered to represent average candidates for bariatric surgery and the patient selection was done according to generally accepted international guidelines (30). Due to the multicenter nature, the results may be applied on a larger scale, compared with specialized high-volume bariatric centers.
In conclusion, at 6-month analysis, there were no statistical differences between SG and RYGB regarding weight loss, resolution of obesity-related comorbidities or overall morbidity. Long-term follow-up from prospective randomized trials comparing SG with RYGB is needed to elucidate the status of SG among the surgical treatment options for morbid obesity.
Footnotes
Acknowledgements
The trial was registered at clinicaltrials.gov NCT00793143. The trial was approved by the Ethics Committee of Turku University Hospital DNRO 6/2008 19.2.2008 § 91.
Declaration of Conflicting Interests
None declared.
Funding
Mika Helmiö received grants from Turku University research grant and The Finnish Medical Society Duodecim research grant. Paulina Salminen received a grant from The Mary and Georg C. Ehrnrooth Foundation.