Key Points
Question
What are the long-term outcomes of laparoscopic sleeve gastrectomy (SG) vs Roux-en-Y gastric bypass (RYGB) for patient weight loss in the SM-BOSS trial?
Finding
The SM-BOSS randomized clinical trial found no significant difference in weight loss between SG and RYGB in the intention-to-treat analysis; however, the per-protocol analysis showed greater weight loss with RYGB. Both procedures improved comorbidities, but RYGB provided better control of gastroesophageal reflux disease, with similar complication rates and quality of life in both groups.
Meaning
Over more than 10 years of follow-up, RYGB demonstrated superiority over SG for patient excess body mass index loss.
Abstract
Importance
Reports on long-term outcomes from randomized clinical trials comparing laparoscopic sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB) are scarce.
Objective
To compare long-term weight and metabolic outcomes, reoperation rates, and quality of life for patients undergoing SG vs RYGB at 10 years and beyond.
Design, Setting, and Participants
The SM-BOSS (Swiss Multicenter Bypass or Sleeve Study) randomized clinical trial was conducted from January 2007 to November 2011 at 4 bariatric centers in Switzerland. (The last follow-up was obtained in July 2023.) A total of 3971 patients with severe obesity were assessed, and 217 patients were enrolled and randomized to undergo SG or RYGB.
Interventions
Laparoscopic SG or RYGB.
Main Outcomes and Measures
The primary outcome of the SM-BOSS trial was the percentage excess body mass index loss (%EBMIL) at 5 years. The present study reports on the long-term weight and metabolic outcomes at 10 years and beyond, including changes in weight and obesity-related diseases, reoperation rates, and quality of life.
Results
Of 217 patients randomized to undergo SG or RYGB, mean (SD) age was 42.5 (11.1) years, mean (SD) baseline BMI was 43.9 (5.3), and 156 patients (71.9%) were female. Of 217 patients, 110 patients were randomized to RYGB and 107 to SG. Complete 10-year follow-up is available for 65.4% of patients. In the intention-to-treat population, mean (SD) %EBMIL was 60.6% (25.9) after SG and 65.2% (26.0) after RYGB (P = .29). Patients who underwent SG had significantly higher conversion rates because of insufficient weight reduction or reflux compared to RYGB (29.9% vs 5.5%; P < .001). Patients undergoing RYGB had significantly higher mean (SD) %EBMIL compared to SG after 10 years in the per-protocol (PP) population (65.9% [26.3] vs 56.1% [25.2]; P = .048). However, mean (SD) percentage total weight loss was not significantly different between groups (RYGB: 27.7% [10.8]; SG: 25.5% [15.1]; P = .37). SG patients had significantly more de novo gastroesophageal reflux (GERD) compared with RYGB (P = .02).
Conclusions and Relevance
In the SM-BOSS randomized clinical trial, RYGB led to significantly higher %EBMIL in the PP population compared with SG beyond 10 years of follow-up, with better results for GERD. Patients undergoing SG experienced a significantly higher number of conversions to different anatomy compared with RYGB.
Trial Registration
ClinicalTrials.gov Identifier NCT00356213
The SM-BOSS (Swiss Multicenter Bypass or Sleeve Study) randomized clinical trial compares the long-term weight and metabolic outcomes, reoperation rates, and quality of life among patients undergoing laparoscopic sleeve gastrectomy or Roux-en-Y gastric bypass at 10 years and beyond.
Introduction
Bariatric metabolic surgery is the treatment of choice for patients with severe obesity and associated comorbidities.1,2,3,4 Sleeve gastrectomy (SG) became the most commonly performed bariatric metabolic procedure worldwide in 2014, leaving the Roux-en-Y gastric bypass (RYGB) in second place.5 Due to technical advances and standardization in the past 2 decades, both procedures provide an excellent safety profile in large databases analyses.6,7 Despite SG being faster and easier to learn and perform than RYGB, recently published long-term data (>10 years of follow-up) show a considerable rate of recurrent weight gain, de novo gastroesophageal reflux disease (GERD), and reoperations after SG.8,9,10 The increased prevalence of pathologic acid exposure and esophagitis after SG is of particular interest considering the high numbers of patients undergoing SG worldwide.11
The first long-term results of a randomized clinical trial (RCT) directly comparing both procedures show a favorable outcome for RYGB, yet do not reach the arbitrarily prespecified equivalence margins.12
The 5-year results of the SM-BOSS trial published in 2018 did not show a significant difference in excess body mass index (BMI, calculated as weight in kilograms divided by height in meters squared) loss between SG and RYGB and in reoperation rates.13 The current article on the same cohort provides further long-term results of SG vs RYGB and addresses weight loss, the evolution of obesity-associated comorbidities, reoperation rates, and quality of life.
Methods
This study was conducted in accordance with the principles of the Declaration of Helsinki, approved by each local ethical committee (EKNZ 2005/00272 and EKNZ 2018/00356), and registered at the clinical trials registry of the US National Institutes of Health (NCT00356213).14 All patients gave written informed consent.
Study Design, Participants, Randomization, and Interventions
The trial protocol and further details can be found in the original publication.13 In brief, the trial was conducted at 4 bariatric centers in Switzerland from January 2007 until November 2011, and a total of 217 participants were included and randomized to undergo either laparoscopic SG or proximal RYGB. Inclusion criteria were BMI higher than 40 or BMI higher than 35 with at least 1 obesity-related comorbidity and an age between 18 and 65 years. The main exclusion criteria were previous bariatric surgery, severe GERD (defined by presence of GERD symptoms despite medication), large hiatal hernia, and inflammatory bowel disease. Block randomization was computer-based and performed centrally; there was no masking procedure. All patients underwent preoperative multidisciplinary assessment and esophago-gastro-duodenoscopy. The 2 interventions were highly standardized. Briefly, SG was performed using a 35F catheter calibration bougie along the lesser curvature, starting 3 to 6 cm orally of the pylorus. RYGB was performed with a 150-cm antecolic Roux limb and a 50-cm biliopancreatic limb.13
Outcomes
Inclusion, randomization, and follow-up of the cohort were reported according to the Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines (Figure 1).
Figure 1. CONSORT Flow Diagram of the SM-BOSS Trial Cohort After 5 and 10 Years.
The primary end point of the trial was percentage excess BMI loss (%EBMIL, calculated as 100 × [baseline BMI − follow-up BMI]/[baseline BMI − 25]) at 5 years and was displayed thoroughly in the original work. The primary outcome in the current report is the weight loss expressed as %EBMIL at 10 years and beyond. Briefly, suboptimal weight loss was defined as a percentage total weight loss (%TWL) of less than 20% or recurrent weight gain of more than 30% from nadir.
Secondary end points entailed %TWL (calculated as [weight loss in kg/baseline weight in kg] × 100) and change in BMI (baseline BMI − follow-up BMI) at 10 years and beyond, as well as the evolution of obesity-related diseases, such as type 2 diabetes (T2D, including fasting plasma glucose values and hemoglobin A1C at baseline and last follow-up), dyslipidemia (including lipid profile at baseline and last follow-up), GERD, arterial hypertension (AHT), and obstructive sleep apnea (OSA).
Complications related to the operative procedures were reported using the validated comprehensive complication index (CCI).15 Further exploratory outcomes included the reoperation or conversion rate for any reason or therapeutic endoscopies. Conversion was defined as a change of bariatric anatomy (ie, conversion from SG to RYGB or modification of limb length in RYGB).
Quality of life at the latest follow-up was reported using the validated Quality of Life Bariatric Analysis and Reporting Outcome System (BAROS), and the overall postoperative outcome was analyzed using the Swiss-Finnish Bariatric Metabolic Outcome score (SF-BARI score).16,17
Statistical Analysis
The power estimation for %EBMIL was based on the assumptions of equal variances in both treatment groups, a pooled standard deviation of 20%, a base effect of excess BMI loss at 5 years of 50% in the control group and a minimum detectable difference of an additional 10%, and an α = .05. A sample size of 100 patients per group was estimated to provide a power of 94% to reject the null hypothesis of equal means using a 2-sided, 2-sample equal-variance t test.18,19 According to a meta-analysis, a difference of 10% in %EBMIL resulted in a superior T2D remission rate when comparing gastric banding, RYGB, and biliopancreatic diversion.20 Therefore, a minimal difference of 10% %EBMIL was considered clinically relevant. Statistical differences were tested by open-source software R version 4.3.2 (The R Foundation). The analysis was performed in 2 study populations. The intention-to-treat (ITT) population included all randomized patients, regardless of whether they completed the study per protocol, received their assigned treatment, or deviated from the intervention plan, and included those who underwent a reoperation. (In this population, missing values were imputed using the last observation carried forward method.) The per-protocol (PP) population included only patients who completed the weight-related assessments 10 years after the operation without treatment conversion before the 10-year assessment. Longitudinal data reflecting the repeated measurement over the study period were analyzed by linear mixed-effect modeling using the R lme4 package (The R Foundation) using initially center, duration, treatment (and their interaction), basal BMI, sex, and age as fixed factors and participant as random factors.18 Post hoc tests were calculated by the emmeans module to perform pairwise comparisons with a Tukey test in order to compare groups with their control. Binary data were compared by χ2 test, and absolute differences in means with 95% confidence intervals were calculated with the prop.test module (The R Foundation). Survival analysis was performed with the nonparametric Kaplan-Meier analysis using a log-rank test to compare survival curves.21 Level of significance was P = .05.
To better visualize the dynamic development of the study effect in this 10-year longitudinal study, the study effect was further analyzed on a PP basis in each study period. This additional graphical analysis demonstrates the dynamics of the treatment effects and the number of patients retained at various time points. Notably, at the 10-year mark, the analysis corresponds to a classical PP analysis, wherein only those subjects who completed the entire study period are considered, effectively illustrating the longitudinal retention and impact over the decade.
Results
Patient Demographics and Follow-Up
The flowchart is shown in Figure 1. Baseline characteristics of the initial 217 trial patients were previously reported13 and can be found in eTable 1 in Supplement 2. Of the 217 patients randomized to undergo SG or RYGB, mean (SD) age was 42.5 (11.1) years, mean (SD) baseline BMI was 43.9 (5.3), and 156 patients (71.9%) were female. In total, 107 patients underwent SG (77 females [72.0%]; mean [SD] age, 43.0 [11.1] years; mean [SD] baseline BMI, 43.6 [5.3]) and 110 patients underwent RYGB (79 females [71.8%]; mean [SD] age, 42.1 [11.2] years; mean [SD] baseline BMI, 44.2 [5.3]) within the study period. At 10 or more years, 142 patients of the initially available 217 patients (65.4%)were available for follow-up—69 patients (64.5%) in the SG group and 73 (66.4%) in the RYGB group, respectively.
Weight Loss
The %EBMIL trajectories at the predefined time points during the study follow-up are shown in Figure 2. Compared to baseline, mean (SD) BMI was significantly lower at 10 years for both SG (33.5 [5.5]; P < .001) and RYGB (32.0 [6.3]; P < .001). At 10 years, the mean (SD) %EBMIL for the ITT population was 60.6% (25.9) after SG and 65.2% (26.0) after RYGB (P = .29). The mean (SD) %TWL for the ITT population at 10 years was 26.3% (13.6) for SG and 27.5% (10.6) for RYGB. However, the PP analyses of %EBMIL showed significantly higher differences between the 2 intervention groups (P = .048), with a mean (SD) %EBMIL of 56.1% (25.2) for SG and 65.9% (26.3) for RYGB.
Figure 2. Weight Loss Outcomes After Sleeve Gastrectomy and Roux-en-Y Gastric Bypass for the Intention-to-Treat (ITT) and Per-Protocol (PP) Groups at 10 Years or Longer.
A-D, Percentage excess body mass index (BMI) loss after sleeve gastrectomy or Roux-en-Y gastric bypass over the study period in the ITT and PP populations. Time course of superimposed cumulative individual values of percentage excess BMI loss over the period after surgery. Circles indicate mean values at 1, 3, 5, and ≥10 years after surgery. Missing values were imputed for number of participants at each time point.
BMI, %EBMIL, and %TWL data are reported in detail from baseline to 10 years in Table 1 and in the eFigure in Supplement 2. At 10 years or more, a %TWL of 20% or more was achieved in 46 of 69 patients after SG (66.7%) and in 53 of 73 patients after RYGB (72.6%) (P = .47). A %TWL of less than 5% was present in 3 patients after SG (4.3%) and in 2 patients after RYGB (2.7%) (P = .67).
Table 1. Weight Loss Outcomes of the Study Population.
| Parameter or unit | Mean (SD) | P valuea | |
|---|---|---|---|
| Sleeve gastrectomy | Roux-en-Y gastric bypass | ||
| Intention to treat b | |||
| Participants, No. | 69 | 73 | NA |
| Weight, kg | |||
| Baseline | 123.6 (19.4) | 124.8 (19.8) | >.99 |
| Year 10 | 92.5 (18.9) | 91.7 (21.6) | >.99 |
| P valuec | <.001 | <.001 | |
| BMId | |||
| Baseline | 43.6 (5.3) | 44.2 (5.3) | >.99 |
| Year 10 | 32.7 (5.6) | 32.3 (6.3) | >.99 |
| P valuec | <.001 | <.001 | |
| %EBMIL, % | |||
| Year 10 | 60.6 (25.9) | 65.2 (26.0) | .29 |
| %TWL, % | |||
| Year 10 | 26.3 (13.6) | 27.5 (10.6) | .56 |
| Per protocol b | |||
| Participants, No. | 48 | 66 | NA |
| Weight, kg | |||
| Baseline | 126.0 (20.4) | 125.3 (20.1) | >.99 |
| Year 10 | 95.9 (18.9) | 90.7 (20.7) | >.99 |
| P valuec | <.001 | <.001 | |
| BMId | |||
| Baseline | 44.1 (5.9) | 44.3 (5.6) | >.99 |
| Year 10 | 33.5 (5.5) | 32.0 (6.3) | >.99 |
| P valuec | <.001 | <.001 | |
| %EBMIL, % | |||
| Year 10 | 56.1 (25.2) | 65.9 (26.3) | .048 |
| %TWL, % | |||
| Year 10 | 25.5 (15.1) | 27.7 (10.8) | .37 |
Abbreviations: %EBMIL, percentage excess body mass index loss; %TWL, percentage total weight loss; BMI, body mass index; NA, not applicable.
Within the group (sleeve gastrectomy or Roux-en-Y gastric bypass) from baseline to 10 years.
Weight loss outcomes for the intention-to-treat and per-protocol (PP) populations. In the PP population, all patients were included who completed the weight-related assessments after 10 years without prior treatment conversion.
Comparing sleeve gastrectomy or Roux-en-Y gastric bypass at a specific time point.
Calculated as weight in kilograms divided by height in meters squared.
Obesity-Related Diseases
Of those patients with complete follow-up, 18 patients in the SG group (26.1%) and 14 patients in the RYGB group (19.2%) had T2D at baseline. Ten or more years after surgery, complete remission of T2D was seen in 11 patients after SG (61.1%) and in 10 patients after RYGB (71.4%) (absolute difference, 10.3%; 95% CI, −49.7 to 34.3%; P = .71; Table 2). Marked amelioration of glycemic control with a decrease in mean (SD) fasting plasma glucose values was seen beyond 10 years compared with baseline for both procedures (SG, baseline: 5.7 [2.3] mmol/L vs follow-up: 5.0 [1.3] mmol/L; RYGB, baseline: 5.5 [1.6] mmol/L vs follow-up: 5.3 [1.3] mmol/L). No significant differences between the treatment groups in fasting glucose or hemoglobin A1c were found (eTable 2 in Supplement 2).
Table 2. Outcomes of Obesity-Related Medical Problems 10 Years After Initial Surgery.
| Obesity-related medical problem | No. | Absolute difference (95% CI), % | P value | ||
|---|---|---|---|---|---|
| SG (n = 69) | RYGB (n = 73) | Unadjusted | Adjusted | ||
| Type 2 diabetes | |||||
| Prevalent at baseline | 18 | 14 | 11.3 (−9.8 to 12.57) | .20 | >.99 |
| Remission | 11 | 10 | 10.3 (−49.7 to 34.3) | .71 | >.99 |
| Improved | 4 | 4 | 6.35 (−45.6 to 33.5) | .70 | >.99 |
| Unchanged | 1 | 0 | 5.56 (−20.5 to 25.8) | >.99 | >.99 |
| Worsened | 2 | 0 | 11.1 (−18.4 to 32.8) | .49 | >.99 |
| Not prevalent at baseline | 40 | 57 | NA | NA | NA |
| Not prevalent | 36 | 52 | 1.23 (−19.2 to 15.0) | .87 | >.99 |
| De novo symptoms | 1 | 1 | 0.75 (−8.85 to 117.6) | >.99 | >.99 |
| Hypertension | |||||
| Prevalent at baseline | 40 | 41 | 7.0 (−16.3 to 29.4) | .52 | >.99 |
| Remission | 21 | 20 | 3.7 (−26.0 to 33.8) | .91 | >.99 |
| Improved | 8 | 14 | 14.1 (−39.0 to 13.2) | .24 | >.99 |
| Unchanged | 9 | 6 | 7.9 (−16.1 to 30.6) | .53 | >.99 |
| Worsened | 1 | 0 | 2.5 (−8.12 to 12.88) | .49 | >.99 |
| Not prevalent at baseline | 21 | 29 | NA | NA | NA |
| Not present | 19 | 25 | 13.5 (−41.4 to 16.8) | .99 | >.99 |
| De novo symptoms | 0 | 1 | 3.5 (−17.2 to 14.9) | .91 | >.99 |
| Dyslipidemia | |||||
| Prevalent at baseline | 42 | 31 | 17.8 (−5.48 to 39.26) | .05 | .05 |
| Remission | 13 | 6 | 11.6 (−17.2 to 36.8) | .72 | >.99 |
| Improved | 21 | 20 | 14.5 (−43.4 to 17.5) | .98 | >.99 |
| Unchanged | 4 | 3 | 0.2 (−21.1 to 18.7) | >.99 | >.99 |
| Worsened | 2 | 0 | 4.8 (−9.71 to 15.79) | .51 | >.99 |
| Not prevalent at baseline | 27 | 41 | NA | NA | NA |
| Not present | 23 | 38 | 7.5 (−30.0 to 13.5) | .42 | >.99 |
| De novo symptoms | 3 | 1 | 8.7 (−8.74 to 27.63) | .29 | >.99 |
| Obstructive sleep apnea | |||||
| Prevalent at baseline | 32 | 35 | 4.9 (−19.0 to 28.3) | .70 | >.99 |
| Remission | 30 | 29 | 10.9 (−12.0 to 31.0) | .26 | >.99 |
| Improved | 0 | 0 | 0.0 (−9.89 to 10.72) | >.99 | >.99 |
| Unchanged | 0 | 2 | 5.7 (−18.61 to 9.14) | .49 | >.99 |
| Worsened | 1 | 0 | 3.1 (−9.34 to 15.74) | .48 | >.99 |
| Not prevalent at baseline | 27 | 36 | NA | NA | NA |
| Not present | 24 | 28 | 11.1 (−16.3 to 34.2) | .33 | >.99 |
| De novo symptoms | 0 | 1 | 2.8 (−14.2 to 12.0) | >.99 | >.99 |
| Gastroesophageal reflux disease | |||||
| Prevalent at baseline | 25 | 32 | 3.3 (−26.7 to 20.5) | .84 | >.99 |
| Remission | 17 | 23 | 3.9 (−36.0 to 28.2) | .98 | >.99 |
| Improved | 3 | 1 | 8.9 (−11.6 to 29.4) | .31 | >.99 |
| Unchanged | 0 | 2 | 6.3 (−20.1 to 11.6) | .499 | >.99 |
| Worsened | 4 | 1 | 12.9 (−9.34 to 34.10) | .16 | >.99 |
| Not prevalent at baseline | 34 | 38 | NA | NA | NA |
| Not present | 21 | 32 | 22.4 (−47.52 to 6.52) | .06 | .06 |
| De novo symptoms | 11 | 3 | 24.5 (−1.67 to 46.44) | .02 | .03 |
| Missing | 2 | 3 | NA | NA | NA |
Abbreviations: NA, not applicable; RYGB, Roux-en-Y gastric bypass; SG, sleeve gastrectomy.
Before surgery, 42 patients in the SG group (60.9%) and 31 patients in the RYGB group (42.5%) had dyslipidemia. Complete remission was seen in 13 of 42 patients in the SG group (31.0%) vs 6 of 31 patients in the RYGB group (19.4%) 10 years or more after surgery (absolute difference, 11.6; 95% CI, −17.2 to 36.8; P = .72; Table 2).
Significant amelioration was seen beyond 10 years for total and high-density lipoprotein cholesterol, the ratio of total cholesterol to high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and triglycerides in both groups compared to baseline, without any differences between the groups (eTable 2 in Supplement 2).
At the time of surgery, 25 of 69 patients in the SG group (36.2%) and 32 of 73 patients in the RYGB group (43.8%) experienced some degree of GERD. Beyond 10 years, remission of reflux symptoms was seen in 17 of 25 patients in the SG group (68.0%) and 23 of 32 patients in the RYGB group (71.9%). In addition, 11 of 34 patients who had no GERD at baseline (32.3%) reported de novo reflux symptoms 10 years after SG, whereas this was only the case in 3 of 38 patients who underwent RYGB (7.9%) (P = .02; Table 2).
At baseline, 40 of 69 patients in the SG group (58.0%) and 41 of 73 patients in the RYGB group (56.2%) had AHT. There was no statistically significant difference between the groups regarding AHT remission—21 patients of 40 in the SG group (52.5%) had a complete remission vs 20 of 41 patients in the RYGB group (48.8%) (P = .91).
At baseline, 32 of 69 patients in the SG group (46.4%) and 35 of 73 patients in the RYGB group (47.9%) had OSA. There was no statistically significant difference between the SG and RYGB groups regarding the OSA remission rate. Beyond 10 years, remission or improvement was seen in 30 of 32 patients in the SG group (93.8%) and in 29 of 35 patients in the RYGB group (71.9%).
Complications and Reoperations
Adverse events related to surgery have been reported in detail previously.13,18,22 The overall complications were not significantly different between the groups, with mean (SD) CCI scores of 13.0 (19.3) for the SG group and 12.0 (31.3) for the RYGB group. The CCI score was greater than 0 in 32 of 107 patients in the SG group (29.9%) and in 21 of 110 patients in the RYGB group (19.1%) (P = .08). For those patients with a CCI score greater than 0, the severity of complications was not significantly different between groups (mean [SD] severity, SG: 38.1 [20.5] vs RYGB: 38.2 [20.4]; P = .99). However, patients after SG had significantly higher conversion rates because of suboptimal clinical response or GERD compared to RYGB (SG: 32 of 107 patients [29.9%] vs RYGB: 6 of 110 patients [5.5%]; P < .001; Figure 3). Reasons for conversion in the SG group were reflux (15 of 32 [46.9%]), suboptimal weight loss (9 [28.1%]), a combination of both factors (6 [18.8%]), or strictures (2 [6.3%]). Of 32 SG patients receiving reoperation, 1 patient (3.1%) received a resleeve and 5 patients (15.6%) were converted to biliopancreatic diversion with duodenal switch because of suboptimal weight loss. Conversion from SG to RYGB was performed in 26 patients (81.3%): 17 of 26 patients received a short alimentary limb RYGB (15 because of reflux, 2 because of strictures), 7 patients received a long biliopancreatic limb RYGB (6 because of reflux and suboptimal weight loss, 1 because of suboptimal weight loss alone), and 2 patients received a long alimentary limb RYGB (both because of suboptimal weight loss).
Figure 3. Time to Conversion Over the Study Period Up to 14 Years.
In the sleeve gastrectomy group, 64% of patients (n = 32) underwent a conversion operation, whereas 7.5% of patients from the Roux-en-Y group (n = 6) underwent a conversion.
Reasons for conversion or revision in the RYGB group were all because of suboptimal weight loss (6 patients [100%]). One patient (16.7%) had a conversion to a long biliopancreatic limb RYGB and 2 patients (33.3%) received a pouch resizing, while 3 patients (50.0%) underwent pouch resizing and Fobi ring placement.
All-cause mortality was 1 of 107 patients in the SG group (0.9%) and 5 of 110 patients in the RYGB group (4.5%) (P = .21). Procedure-related mortality was 0 in the SG group and 1 of 110 in the RYGB group (0.9%).
Quality of Life and SF-BARI Score
The mean (SD) Moorehead-Ardelt BAROS quality of life score was increased in both groups after 10 years compared to baseline (SG: 0.1 [1.2] vs 2.2 [2.1]; RYGB: 0.2 [1.3] vs 1.9 [1.9]; P < .001), but without differences between the groups. The mean (SD) SF-BARI score beyond 10 years was 84.0 points (30.1) for SG and 88.7 points (36.0) for RYGB (P = .40), which corresponds to a categorization as “good” (eTable 3 in Supplement 2).17
Discussion
The SM-BOSS RCT randomized 217 patients living with severe obesity to undergo either laparoscopic SG or RYGB. For the ITT analysis, no significant difference in %EBMIL was found after 10 years comparing SG and RYGB. However, the PP analysis revealed superior weight loss in patients who underwent RYGB compared to SG. Both procedures could significantly and comparably reduce obesity-related diseases, including T2D, dyslipidemia, and OSA. More patients reported de novo GERD after SG compared to RYGB. The 2 groups did not show significant differences in complications; however, patients undergoing SG had a higher conversion rate to a different bariatric procedure over time. Both interventions increased quality of life after 10 years, with no significant difference between the 2 groups.
The SM-BOSS RCT showed no significant difference in %EBMIL in the ITT analysis 10 years after SG compared to RYGB. In contrast, the recently published 10-year results of a large Finnish SG vs RYGB trial (SLEEVEPASS RCT) showed superior percentage excess weight loss (%EWL) for RYGB, but the prespecified equivalence margins (plus or minus %EWL) were not reached.12 Both the SM-BOSS and SLEEVEPASS RCTs showed differences in weight loss for their 5-year mark favoring RGYB; however, differences did not reach statistical significance.13,19 The merged 5-year data of both RCTs demonstrate improved %EBMIL for patients undergoing RYGB.20 Also, a large cohort study including almost 10 000 patients showed that RYGB leads to 8% more %TWL than SG after 5 years.23 Overall, RYGB resulted in 4.6% more %EBMIL than SG in the ITT analysis and 9.8% more %EBMIL in the PP analysis. While larger studies may show statistical significance with smaller weight loss differences, our study reached statistical significance only in the PP analysis. However, 4.6% and 9.8% differences in %EBMIL in the ITT and PP analyses, respectively, may both be relevant for the clinical setting. More patients after SG needed a conversion operation compared to RYGB. This is further underlined by the PP weight loss analysis, where %EBMIL was more favorable for RYGB compared to SG when patients retained their primary bariatric procedure. The chances of achieving favorable weight loss with a single operation is higher with RYGB compared to SG.
One main driver for survival benefits after bariatric and metabolic surgery seems to originate from control and remission of T2D.24 In this trial, 23% of patients had T2D at baseline. After 10 years, the remission rate of T2D was 61% after SG and 71% after RYGB; the between-group difference was not significantly different. Most RCTs are underpowered to detect differences in T2D remission between SG and RYGB.13,19,25,26 Some trials suggest a possible reduction of antidiabetic drugs in patients after RYGB compared to SG.1 A recently published secondary analysis of a matched-cohort study with 1400 RYGB and 700 SG patients revealed that in patients with obesity and T2D, RYGB was associated with greater weight loss, better diabetes control, and lower risk of major adverse cardiovascular events compared to SG.27
Significant improvements in dyslipidemia and AHT were seen after 10 years in this trial, but no difference between the groups was found. In contrast, our 5-year merged results (SM-BOSS and SLEEVEPASS) and other trials showed better outcomes for dyslipidemia and AHT for RYGB.20,28,29 This again may be explained by high conversion rates from SG to RYGB and a smaller sample size.
In the last decade, high rates of Barrett esophagus were observed after SG and RYGB (15% and 17%, respectively).8,30 In the present RCT, we observed a good GERD remission rate of around 70% for both SG and RYGB after 10 years. This stands in contrast to our 5-year results and the merged results, where RYGB was significantly superior regarding GERD remission compared to SG.18,20 However, in the present work, many patients in the SG group were converted to a different bariatric procedure due to severe reflux symptoms before reaching the 10-year end point. Additionally, significantly more patients without GERD at baseline reported de novo symptoms 10 years after SG compared to RYGB.
The overall complication by the total CCI score was not significantly different between SG and RYGB after 10 years. The high conversion rate after SG was described previously in other long-term follow-up studies.8,10 The reasons for conversion from SG to RYGB in the present study were GERD, suboptimal weight loss, or a combination of both. For GERD, the conversion chosen was usually an RYGB. According to recent evidence, RYGB is an appropriate choice for patients suffering from severe GERD after SG.10,31,32 Also, patients with suboptimal clinical response or recurrent weight gain were converted from SG to RYGB. According to recent publications,33,34 these patients may require more effective hypoabsorptive procedures or drug therapy to achieve significant additional weight loss after SG.
Limitations
The study is subject to several limitations that have been discussed in detail previously. The most relevant limitations previously noted in 2017 and 2018 included insufficient power to distinguish outcomes related to T2D remission, the inclusion of patients with a BMI over 60, and the absence of masking for patients and staff regarding procedure type.13,18 New limitations may arise specifically after 10 years. The trial was designed and powered as an RCT with 5-year follow-up—therefore, the analysis at 10 years may result in insufficient statistical power due to loss of participants and loss of follow-up. Furthermore, in the SG group, many patients have an RYGB as their active bariatric and metabolic procedure. Nevertheless, the high conversion rate corresponds to other long-term follow-up studies and displays the real-life scenario.8
Conclusions
In this randomized clinical trial, RYGB led to significantly higher %EBMIL in the PP population compared to SG beyond 10 years’ follow-up. Improvement of comorbidities was similar except for GERD, which seems to be better controlled by RYGB. Patients undergoing SG experienced a significantly higher number of conversions to different anatomy compared to RYGB.
Statistical Analysis Plan
eFigure. Weight Loss and Glycemic Outcomes Over the Study Period
eTable 1. Baseline Characteristics
eTable 2. Biochemical Outcomes After Sleeve Gastrectomy and Roux-en-Y Gastric Bypass
eTable 3. Quality of Life, Complication and Overall Outcomes After SG and RYGB
Data Sharing Statement
References
- 1.Schauer PR, Bhatt DL, Kirwan JP, et al. ; STAMPEDE Investigators . Bariatric surgery versus intensive medical therapy for diabetes - 5-year outcomes. N Engl J Med. 2017;376(7):641-651. doi: 10.1056/NEJMoa1600869 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Verrastro O, Panunzi S, Castagneto-Gissey L, et al. Bariatric-metabolic surgery versus lifestyle intervention plus best medical care in non-alcoholic steatohepatitis (BRAVES): a multicentre, open-label, randomised trial. Lancet. 2023;401(10390):1786-1797. doi: 10.1016/S0140-6736(23)00634-7 [DOI] [PubMed] [Google Scholar]
- 3.Järvholm K, Janson A, Peltonen M, et al. Metabolic and bariatric surgery versus intensive non-surgical treatment for adolescents with severe obesity (AMOS2): a multicentre, randomised, controlled trial in Sweden. Lancet Child Adolesc Health. 2023;7(4):249-260. doi: 10.1016/S2352-4642(22)00373-X [DOI] [PubMed] [Google Scholar]
- 4.Adams TD, Davidson LE, Litwin SE, et al. Weight and metabolic outcomes 12 years after gastric bypass. N Engl J Med. 2017;377(12):1143-1155. doi: 10.1056/NEJMoa1700459 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Angrisani L, Santonicola A, Iovino P, Ramos A, Shikora S, Kow L. Bariatric Surgery Survey 2018: similarities and disparities among the 5 IFSO chapters. Obes Surg. 2021;31(5):1937-1948. doi: 10.1007/s11695-020-05207-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Chhabra KR, Telem DA, Chao GF, et al. Comparative safety of sleeve gastrectomy and gastric bypass: an instrumental variables approach. Ann Surg. 2022;275(3):539-545. doi: 10.1097/SLA.0000000000004297 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Arterburn DE, Telem DA, Kushner RF, Courcoulas AP. Benefits and risks of bariatric surgery in adults: a review. JAMA. 2020;324(9):879-887. doi: 10.1001/jama.2020.12567 [DOI] [PubMed] [Google Scholar]
- 8.Felsenreich DM, Artemiou E, Steinlechner K, et al. Fifteen years after sleeve gastrectomy: weight loss, remission of associated medical problems, quality of life, and conversions to Roux-en-Y gastric bypass-long-term follow-up in a multicenter study. Obes Surg. 2021;31(8):3453-3461. doi: 10.1007/s11695-021-05475-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sakran N, Soifer K, Hod K, et al. Long-term reported outcomes following primary laparoscopic sleeve gastrectomy. Obes Surg. 2023;33(1):117-128. doi: 10.1007/s11695-022-06365-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kraljević M, Cordasco V, Schneider R, et al. Correction to: long-term effects of laparoscopic sleeve gastrectomy: what are the results beyond 10 years? Obes Surg. 2021;31(8):3434-3435. doi: 10.1007/s11695-021-05460-4 [DOI] [PubMed] [Google Scholar]
- 11.Wölnerhanssen BK, Meyer-Gerspach AC, Nussbaumer R, et al. Prospective clinical cohort study: low incidence of Barrett esophagus but high rate of reflux disease at 5-year follow-up after sleeve gastrectomy versus Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2023;19(7):707-715. doi: 10.1016/j.soard.2023.02.012 [DOI] [PubMed] [Google Scholar]
- 12.Salminen P, Grönroos S, Helmiö M, et al. Effect of laparoscopic sleeve gastrectomy vs Roux-en-Y gastric bypass on weight loss, comorbidities, and reflux at 10 years in adult patients with obesity: the SLEEVEPASS randomized clinical trial. JAMA Surg. 2022;157(8):656-666. doi: 10.1001/jamasurg.2022.2229 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Peterli R, Wölnerhanssen BK, Peters T, et al. Effect of laparoscopic sleeve gastrectomy vs laparoscopic roux-en-y gastric bypass onweight loss in patients with morbid obesity the SM-BOSS randomized clinical trial. JAMA. 2018;319(3):255-265. doi: 10.1001/jama.2017.20897 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.World Medical Association . World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013;310(20):2191-2194. doi: 10.1001/jama.2013.281053 [DOI] [PubMed] [Google Scholar]
- 15.Slankamenac K, Graf R, Barkun J, Puhan MA, Clavien PA. The comprehensive complication index: a novel continuous scale to measure surgical morbidity. Ann Surg. 2013;258(1):1-7. doi: 10.1097/SLA.0b013e318296c732 [DOI] [PubMed] [Google Scholar]
- 16.Oria HEMM, Moorehead MK. Bariatric analysis and reporting outcome system (BAROS). Obes Surg. 1998;8(5):487-499. doi: 10.1381/096089298765554043 [DOI] [PubMed] [Google Scholar]
- 17.Peterli R, Hurme S, Bueter M, Grönroos S, Helmiö M, Salminen P. Standardized assessment of metabolic bariatric surgery outcomes: secondary analysis of 2 randomized clinical trials. JAMA Surg. 2024;159(3):306-314. doi: 10.1001/jamasurg.2023.6254 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Peterli R, Wölnerhanssen BK, Vetter D, et al. Laparoscopic sleeve gastrectomy versus Roux-Y-Gastric Bypass for morbid obesity-3-year outcomes of the prospective randomized Swiss Multicenter Bypass Or Sleeve Study (SM-BOSS). Ann Surg. 2017;265(3):466-473. doi: 10.1097/SLA.0000000000001929 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Salminen P, Helmiö M, Ovaska J, et al. Effect of laparoscopic sleeve gastrectomy vs laparoscopic Roux-en-Y gastric bypass on weight loss at 5 years among patients with morbid obesity: the SLEEVEPASS randomized clinical trial. JAMA. 2018;319(3):241-254. doi: 10.1001/jama.2017.20313 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wölnerhanssen BK, Peterli R, Hurme S, et al. Laparoscopic Roux-en-Y gastric bypass versus laparoscopic sleeve gastrectomy: 5-year outcomes of merged data from two randomized clinical trials (SLEEVEPASS and SM-BOSS). Br J Surg. 2021;108(1):49-57. doi: 10.1093/bjs/znaa011 [DOI] [PubMed] [Google Scholar]
- 21.Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53(282):457-481. doi: 10.1080/01621459.1958.10501452 [DOI] [Google Scholar]
- 22.Peterli R, Borbély Y, Kern B, et al. Early results of the Swiss Multicentre Bypass or Sleeve Study (SM-BOSS): a prospective randomized trial comparing laparoscopic sleeve gastrectomy and Roux-en-Y gastric bypass. Ann Surg. 2013;258(5):690-694. doi: 10.1097/SLA.0b013e3182a67426 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.McTigue KM, Wellman R, Nauman E, et al. ; PCORnet Bariatric Study Collaborative . Comparing the 5-Year diabetes outcomes of sleeve gastrectomy and gastric bypass: the National Patient-Centered Clinical Research Network (PCORNet) Bariatric Study. JAMA Surg. 2020;155(5):e200087. doi: 10.1001/jamasurg.2020.0087 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Syn NL, Cummings DE, Wang LZ, et al. Association of metabolic-bariatric surgery with long-term survival in adults with and without diabetes: a one-stage meta-analysis of matched cohort and prospective controlled studies with 174 772 participants. Lancet. 2021;397(10287):1830-1841. doi: 10.1016/S0140-6736(21)00591-2 [DOI] [PubMed] [Google Scholar]
- 25.Keidar A, Hershkop KJ, Marko L, et al. Roux-en-Y gastric bypass vs sleeve gastrectomy for obese patients with type 2 diabetes: a randomised trial. Diabetologia. 2013;56(9):1914-1918. doi: 10.1007/s00125-013-2965-2 [DOI] [PubMed] [Google Scholar]
- 26.Yang J, Wang C, Cao G, et al. Long-term effects of laparoscopic sleeve gastrectomy versus Roux-en-Y gastric bypass for the treatment of Chinese type 2 diabetes mellitus patients with body mass index 28-35 kg/m(2). BMC Surg. 2015;15(88):88. doi: 10.1186/s12893-015-0074-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Aminian A, Wilson R, Zajichek A, et al. Cardiovascular outcomes in patients with type 2 diabetes and obesity: comparison of gastric bypass, sleeve gastrectomy, and usual care. Diabetes Care. 2021;44(11):2552-2563. doi: 10.2337/dc20-3023 [DOI] [PubMed] [Google Scholar]
- 28.Climent E, Benaiges D, Flores-Le Roux JA, Ramón JM, Pedro-Botet J, Goday A. Changes in the lipid profile 5 years after bariatric surgery: laparoscopic Roux-en-Y gastric bypass versus laparoscopic sleeve gastrectomy. Surg Obes Relat Dis. 2018;14(8):1099-1105. doi: 10.1016/j.soard.2018.05.006 [DOI] [PubMed] [Google Scholar]
- 29.Du X, Fu XH, Peng BQ, Luo R, Hu JK, Cheng Z. Resolution of metabolic syndrome and related metabolic disorders after bariatric surgery: comparison of sleeve gastrectomy and gastric bypass. Surg Obes Relat Dis. 2018;14(9):1348-1356. doi: 10.1016/j.soard.2018.05.016 [DOI] [PubMed] [Google Scholar]
- 30.Genco A, Soricelli E, Casella G, et al. Gastroesophageal reflux disease and Barrett’s esophagus after laparoscopic sleeve gastrectomy: a possible, underestimated long-term complication. Surg Obes Relat Dis. 2017;13(4):568-574. doi: 10.1016/j.soard.2016.11.029 [DOI] [PubMed] [Google Scholar]
- 31.Lyo V, Stroud A, Wood S, et al. Reoperations after sleeve gastrectomy: a dual academic institutional experience. Surg Obes Relat Dis. 2022;18(5):641-649. doi: 10.1016/j.soard.2022.01.015 [DOI] [PubMed] [Google Scholar]
- 32.Casillas RA, Um SS, Zelada Getty JL, Sachs S, Kim BB. Revision of primary sleeve gastrectomy to Roux-en-Y gastric bypass: indications and outcomes from a high-volume center. Surg Obes Relat Dis. 2016;12(10):1817-1825. doi: 10.1016/j.soard.2016.09.038 [DOI] [PubMed] [Google Scholar]
- 33.Roach E, Laplante S, Stogryn S, Maeda A, Jackson T, Okrainec A. Weight loss outcomes for patients undergoing conversion to Roux-en-Y-gastric bypass after sleeve gastrectomy. Surg Endosc. 2023;37(4):3208-3214. doi: 10.1007/s00464-022-09506-1 [DOI] [PubMed] [Google Scholar]
- 34.Felsenreich DM, Steinlechner K, Langer FB, et al. Outcome of sleeve gastrectomy converted to Roux-en-Y gastric bypass and one-anastomosis gastric bypass. Obes Surg. 2022;32(3):643-651. doi: 10.1007/s11695-021-05866-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Statistical Analysis Plan
eFigure. Weight Loss and Glycemic Outcomes Over the Study Period
eTable 1. Baseline Characteristics
eTable 2. Biochemical Outcomes After Sleeve Gastrectomy and Roux-en-Y Gastric Bypass
eTable 3. Quality of Life, Complication and Overall Outcomes After SG and RYGB
Data Sharing Statement
