By-Band-Sleeve trial: RYGB betters LSG on weight loss and quality-of-life outcomes

Outcomes from the By-Band-Sleeve trial has confirmed the safety and effectiveness of bariatric and metabolic surgery (BMS) and that Roux-en-Y gastric bypass and sleeve gastrectomy are more effective than adjustable gastric banding. Interestingly, sleeve gastrectomy had inferior weight loss compared with Roux-en-Y gastric bypass, and sleeve gastrectomy was less clinically effective for quality-of-life compared with Roux-en-Y gastric bypass.

The findings, reported by the By-Band-Sleeve Collaborative Group, were featured in the paper, 'Roux-en-Y gastric bypass, adjustable gastric banding, or sleeve gastrectomy for severe obesity (By-Band-Sleeve): a multicentre, open label, three-group, randomised controlled trial', published in The Lancet Diabetes & Endocrinology. The paper’s authors said that future research is required to understand longer term outcomes and to compare BMS with obesity management medications is needed to guide evidence-based practice in this rapidly evolving field.

“Based on the trial findings, we recommend patients electing to have bariatric and metabolic surgery are advised to have Bypass. Sleeve should be a secondary option when Bypass is not possible. Our evidence does not support Band as standard treatment for people living with severe obesity” said Jane Blazeby, Professor of Surgery at the University of Bristol, who led the trial. "The hard work undertaken by all the study participants, surgeons, nurses and dieticians means we now have reliable information to inform NHS practice. The next challenge is for surgical teams to work with researchers and physicians to conduct a new study that compares surgery to weight loss drugs to create evidence to understand they compare surgery in terms of weight loss, quality of life and costs.”

The By-Band-Sleeve is a multi-centre, open-label, randomised controlled trial conducted in 12 hospitals in the UK. Eligible participants were adults (aged ≥18 years) meeting national criteria for metabolic and bariatric surgery. Initially, a 2-group trial (Roux-en-Y gastric bypass versus adjustable gastric banding) became a 3-group trial to include sleeve gastrectomy at 2·6 years from study opening, when it became widely used in the UK. Co-primary endpoints were weight (proportion achieving ≥50% excess weight loss) and quality-of-life (EQ-5D utility score) at 3 years. If the proportion achieving at least 50% excess weight loss was non-inferior (<12% difference between groups) and quality-of-life was superior, sleeve gastrectomy and Roux-en-Y gastric bypass were considered more effective than adjustable gastric banding, and sleeve gastrectomy more effective than Roux-en-Y gastric bypass. The RCT also compared the cost-effectiveness of the procedures.

Trial design

Participating centres were mandated to have a specialist multidisciplinary bariatric team, to perform a minimum of 50 bariatric operations annually and to have a minimum of two surgeons involved in the By-Band-Sleeve trial. Individual surgeons had to have performed at least 50 adjustable gastric banding, 100 Roux-en-Y gastric bypass, and 50 sleeve gastrectomy procedures, be willing to offer participation in the trial to patients, and carry out the surgery according to the randomised allocation and pre-agreed surgical protocols.

After discharge participants attended hospital for surgical or dietetic, or both, follow-up except during the COVID-19 pandemic, which necessitated telephone appointments. Research appointments were designed to coincide with standard NHS care at 4 weeks post-surgery and month 6, 12, and 24 after randomisation, with one additional follow-up at 3 years after randomisation. Follow-up consultations for participants with an adjustable gastric banding in the first 24 postsurgical months were undertaken according to participant need. Postoperative vitamin and mineral supplementation were prescribed in accordance with national guidelines.

The co-primary outcomes were (1) loss of greater than or equal to 50% excess weight (defined as 100 × [BMI at 3 years – BMI at randomisation]/[BMI at randomisation – 25], and (2) the EQ-5D-5L utility score at 3 years after randomisation. Secondary outcomes included percentage total weight loss and BMI, disease-specific and other generic quality-of-life measures (Short-Form 12 (SF-12), Impact of Weight on Quality of life [IWQOL-Lite], Gastro-intestinal Quality of Life Index [GIQLI], Hospital and Anxiety Depression Scale [HADs]), dietary intake (assessed by interview), and binge eating behaviour (assessed via questionnaire),13 sleepiness (Epworth sleepiness scale), and resource use.14 Blood measurements were used to assess metabolic control (HbA1c, fasting glucose, triglycerides, total cholesterol and HDL-cholesterol, plus blood pressure); safety (haemoglobin, 25-hydroxyvitamin D, calcium, ferritin, folate, parathyroid hormone, serum iron, and vitamin B12); and liver and kidney function (alkaline phosphatase, alanine transaminase, and creatinine). Liver fibrosis was assessed using the ELF test measured at baseline and three years.

Outcomes

Out of 6,961 patients screened, 4,140 were eligible and were offered participation. In total, 1,346 participants were included in this report; 462 were randomly assigned to Roux-en-Y gastric bypass, 464 to adjustable gastric banding, and 420 to sleeve gastrectomy, with 1159 (86%) of 1346 remaining in follow-up at three years.

The participants' mean age was 47.3 years (SD 10·6), 1020 (76%) of 1,344 were women and 324 (24%) of 1,344 were men, mean weight was 129.7 kg (23·6) and mean BMI was 46.4 (6.9) kg/m2. Overall, of 1,344 participants, 413 (31%) had diabetes; 92 (7%) self-identified as Black or African or Caribbean or Black British, 46 (3%) as Asian or Asian British, 33 (2%) as mixed or of multiple ethnicity, 33 (2%) as other ethnic group, and 1,140 (85%) as White.

In total, 163 (12%) of 1346 participants did not undergo surgery within three years of randomisation, mostly due to patient choice. Additionally, 115 (10%) of 1,183 participants did not receive their randomised allocation with crossovers occurring in 43 (11%) of 401 in the Roux-en-Y gastric bypass group, 67 (16%) of 417 in the adjustable gastric banding group, and five (1%) of 365 participants in the sleeve gastrectomy group. Overall, of 1,346 participants, 1,321 (98%) provided data on weight and 1,284 (95%) that on EQ-5D-5L at least once during follow-up.

In total, 276 (68%) of 405 participants randomised to Roux-en-Y gastric bypass achieved at least 50% excess weight loss at three years, compared with 97 (25%) of 383 in the adjustable gastric banding group and 141 (41%) of 342 in the sleeve gastrectomy group. Comparing the risk between groups in relation to the predefined 12% non-inferiority margin, both Roux-en-Y gastric bypass and sleeve gastrectomy were non-inferior (and superior) to adjustable gastric banding, and sleeve gastrectomy was inferior to Roux-en-Y gastric bypass for weight loss (Figure 1A).

QoL scores

The EQ-5D-5L utility score shows a similar pattern (Figure 1B). At three years, the mean utility score was significantly higher in the Roux-en-Y gastric bypass group compared with adjustable gastric banding (mean difference 0.079, 98% CI 0.040 to 0.117) and in the sleeve gastrectomy group compared to adjustable gastric banding (0.045, 98% CI 0.006 to 0.085). The difference between sleeve gastrectomy and Roux-en-Y gastric bypass was not significant (–0.033, 98% CI –0.072 to 0.006), although this exceeded the minimally important clinical difference for this measure. There was no statistical evidence to suggest that the treatment effects differed by diabetes status (p=0.90) or weight category at recruitment (p=0.30).

Figure 1: Primary outcomes - results at three years after randomisation (all participants)

Sensitivity analyses for EQ-5D-5L utility score were similarly consistent with the primary analysis with an increased difference between sleeve gastrectomy and Roux-en-Y gastric bypass in favour of Roux-en-Y gastric bypass when excluding participants who did not have surgery. All sensitivity analyses favoured sleeve gastrectomy over adjustable gastric banding, with some being significant at the 2% level. No subgroup differences were found (diabetes p=0.79, baseline BMI p=0.31). Analyses of outcomes by whether participants were recruited into the two or three group phase of the trial did not alter the findings.

Mean percentage total weight loss at three years was –26·8 (SD 13.5) for the Roux-en-Y gastric bypass, –14·0 (13.5) for adjustable gastric banding, and –19·4 (13.1) for the sleeve gastrectomy. In addition, 364 (90%) of 405 participants in the Roux-en-Y gastric bypass group, 217 (57%) of 383 participants in the adjustable gastric banding group, and 269 (79%) of 342 participants in the sleeve gastrectomy group achieved at least 10% total weight loss at three years.

HbA1c of less than 48 mmol/mol at three years was reached by 293 (91%) of 323 participants in the Roux-en-Y gastric bypass group, 253 (82%) of 308 participants in the adjustable gastric banding group, and 233 (88%) of 266 participants in the sleeve gastrectomy group; of those known to have diabetes at baseline the corresponding percentages were 76% (90 of 118), 50% (51 of 103), and 62% (50 of 81).

Mean triglycerides were significantly lower for Roux-en-Y gastric bypass compared with adjustable gastric banding (mean difference 0·82, 98% CI 0.76–0.88) and for Roux-en-Y gastric bypass compared with sleeve gastrectomy (1.15, 1.07–1.24) at three years. The proportions of participants in each group with a total cholesterol of 5 mmol/L or less at three years were 73% (215 of 293) for Roux-en-Y gastric bypass, 58% (166 of 284) for adjustable gastric banding, and 52% (129 of 249) for sleeve gastrectomy.

The EQ-5D-5L visual analogue scale and the SF-12 physical function scores showed the same improvements as the EQ-5D-5L utility scale over time and between groups. Exceptions were the mental health component score of the SF-12 and the anxiety score of the HADs which showed no differences between groups at three years although adjustable gastric banding participants had significantly higher scores (worse depression) than participants receiving Roux-en-Y gastric bypass. The disease-specific quality-of-life measures mostly mirrored the primary outcome (Roux-en-Y gastric bypass and sleeve gastrectomy had better quality-of-life than adjustable gastric banding and differences observed between Roux-en-Y gastric bypass and sleeve gastrectomy favoured Roux-en-Y gastric bypass) with some showing significant differences between groups.

Complications

Hospital stay was a median of 2 days and 1099 (93%) of 1,181 participants had a post-operative recovery classified as normal (ie, Clavien–Dindo grade 0). Rates of any adverse events were lowest for sleeve gastrectomy, and this was statistically lower compared with the adjustable gastric banding group (incidence rate ratio 0.78, 98% CI 0.62–0.98). The adverse event rate in the period from 30-days post-surgery to 3 years was significantly lower following sleeve gastrectomy compared with Roux-en-Y gastric bypass and adjustable gastric banding surgery.

The proportion of serious adverse events was similar between the sleeve gastrectomy and Roux-en-Y gastric bypass groups. Complications related to technical aspects of surgery included internal hernia repairs after Roux-en-Y gastric bypass (in 15 [4%] of 389), leaks from the staple line following sleeve gastrectomy (in three [1%] of 429), and revision operations comprising correction, removal, or conversion to another procedure following adjustable gastric banding (in 52 [14%] of 363.

One death occurred during admission for BMS in the adjustable gastric banding group because of peritonitis and sepsis related to a leak from gastric sutures. Rates of all adverse events from surgery to 30 days were similar in the three groups. There were ten further deaths from randomisation to three years, four each in the Roux-en-Y gastric bypass and adjustable gastric banding groups and two in the sleeve gastrectomy group, none of which were attributable to surgery.

Costs

The mean costs per participant over the three years, including the costs of surgery, were highest for Roux-en-Y gastric bypass and lowest for adjustable gastric banding. Participants randomised to Roux-en-Y gastric bypass accrued on average more QALYs over the three years than those allocated to adjustable gastric banding and sleeve gastrectomy (2.02 [95% CI 1.95–2.09], 1.82 [1.75–1.90] and 1.95 [1.88–2.03], respectively). Combining the costs and QALYs, Roux-en-Y gastric bypass was the most cost-effective option at the cost-utility thresholds applied by NICE, with low probabilities that sleeve gastrectomy (<0.30) or adjustable gastric banding (<0.02) are the most cost-effective option. Results were similar across subgroups and a wide range of sensitivity analyses, including instrumental variable analyses estimating the effect of performing the surgeries at start of follow-up.

“The results of this pragmatic RCT confirm the safety and effectiveness of metabolic and bariatric surgery and that Roux-en-Y gastric bypass and sleeve gastrectomy are more effective than adjustable gastric banding. Sleeve gastrectomy had inferior weight loss compared with Roux-en-Y gastric bypass, and sleeve gastrectomy was less clinically effective for quality-of-life compared with Roux-en-Y gastric bypass,” the authors concluded. “Future research to understand longer term outcomes and to compare metabolic and bariatric surgery with obesity management medications is needed to guide evidence-based practice in this rapidly evolving field.”

"Obesity is one of the biggest challenges facing us globally and in the UK. This treatment will help patients save pounds on their waistline while saving the NHS pounds on their budgets,” said Professor Danny McAuley, Scientific Director for NIHR Programmes. “NIHR funded trial results provide vital evidence that gastric bypass improves the quality of life for UK patients and help clinicians and those providing NHS and care services choose the best treatments. The findings demonstrate how our life changing NIHR research, funded by the public, has impact to support the health and care sector to achieve more effective and efficient use of resources.”

This trial was studied by the National Institute for Health and Care Research Health Technology Assessment Programme.

To access this paper, please click here