Efficacy and safety of the gastric bypass stent system in Chinese adults with obesity: A multicenter, randomized controlled trial

To the Editor: Obesity is a widespread chronic disease with a simultaneous increase globally. Endoscopic bariatric methods include intragastric balloons, endoscopic sleeve gastroplasty, and duodenal-jejunal bypass liner (DJBL). DJBL contains a stent anchored at the duodenal bulb and an endoscopically delivered 60-cm impermeable sleeve that covers the surface of the duodenum and the proximal jejunum. The most common DJBL is EndoBarrier (GI Dynamics, Inc., Watertown, Massachusetts), but adverse effects, especially hepatic abscess, limit its use.1 In this study, we used a novel endoscopic DJBL-gastric bypass stent system (GBSS; Hangzhou Tangee Medical Technology Co., Ltd., China) for weight control. Compared with EndoBarrier, GBSS Supplementary Figure 1, https://links.lww.com/CM9/C841 has made several technical innovations, particularly the use of antibacterial polymer composite materials for sleeves, which might help reduce the occurrence of hepatic abscess. This was a multicenter, randomized, parallel controlled, open-label trial that enrolled patients from seven medical centers in China Supplementary File 1, https://links.lww.com/CM9/C902. The study was conducted in accordance with the principles of the Declaration of Helsinki. The trial was approved by the ethics committe of Beijing Friendship Hospital (No. 2019-P1-028-02). All patients signed informed consents. The study was registered at www.clinicaltrials.gov (NCT05938231). Eligible individuals were adults aged 18–60 years with a body mass index (BMI) of ≥30 kg/m2. The key exclusion criteria were as follows: weight loss of >4.5 kg in the past 3 months; use of bariatric medication; severe organ dysfunction of the heart, lung, liver, or kidney; type 1 diabetes mellitus; anticoagulant or antiplatelet drug use; severe coagulopathy; iron deficiency; history of peptic ulcer, gastrointestinal bleeding, ileus, pancreatitis, or hepatic abscess; cholecystitis; symptomatic gallstones; gallstones larger than 20 mm in diameter; and pregnancy/intent to become pregnant. The study lasted for 9 months. After screening, all patients underwent gastroscopy and were randomly assigned at a 1:1 ratio to the GBSS group (GBSS with dietary control for the first 3 months and then dietary control for another 6 months) or the control group (only dietary control for 9 months). GBSS was both implanted and removed endoscopically under general anesthesia. Neither the research staff nor the participants were blinded to the intervention after randomization. All the participants received a liquid/semiliquid diet for the first week after gastroscopy, followed by a daily caloric intake of less than 25 kcal/ideal body weight (weight kg at which BMI = 25 kg/m2). All the subjects were scheduled for follow-up at months 1, 3, 4, 6, and 9, and those in the GBSS group had an extra follow-up within 7 days of implantation to monitor adverse events (AEs). Data collection included demographic information, medical history, symptoms of discomfort, laboratory results (complete blood count, liver function, blood glucose, fasting insulin, blood amylase, blood iron tests, and fecal occult blood), electrocardiography, and abdominal ultrasonography. The primary outcome was the percentage of excess weight loss (EWL%) at month 3. EWL% = (Baseline weight – Weight at follow-up)/(Baseline weight – Weight at a BMI of 25 kg/m2) × 100%. The secondary outcomes included EWL% at months 1, 4, 6, and 9; the percentage of total weight loss (TWL%); changes in body weight, BMI, and insulin resistance at each visit; and device safety. Insulin resistance was assessed by the homeostasis model assessment of insulin resistance (HOMA-IR). HOMA-IR was calculated as fasting blood glucose (mmol/L) × fasting insulin (μU/mL)/22.5, and a value ≥2.69 was considered insulin resistance. Statistical analyses were performed with SAS version 9.4 (SAS Institute Inc, Cary, NC). A superiority test was used for the primary outcome, with a significance level of α = 0.025 (one-sided) and a confidence level of 1 – β = 0.8. For the other tests, the significance level was α = 0.05 (two-sided), and the confidence level was 1 – β = 0.8. A P value of 5.0% were greater at months 1, 3, 4, and 6 in the GBSS group, and the proportions of TWL% >10.0% were greater at months 1, 4, and 9 in the GBSS group Supplementary Figure 4, https://links.lww.com/CM9/C841. Table 1 - Body weight-related changes over time of patients with BMI ≥30 kg/m2 undergoing gastroscopy in the FAS and PPS analysis. Variables FAS PPS GBSS group (n = 49) Control group (n = 49) P value GBSS group (n = 42) Control group (n = 39) P value Month 1 N 43 47 40 39 EWL% 19.6 (13.8, 26.4) 9.4 (6.0, 18.0) <0.001 19.1 (13.7, 26.2) 10.9 (7.0, 18.7) <0.001 TWL% 5.6 (3.5, 7.1) 2.9 (1.5, 4.2) <0.001 5.4 (3.3, 7.1) 3.4 (1.7, 4.3) <0.001 Body weight decrease (kg) 4.8 (3.3, 7.3) 2.9 (1.2, 4.6) <0.001 4.8 (3.2, 7.2) 3.2 (1.5, 5.1) 0.001 BMI decrease (kg/m2) 1.7 (1.2, 2.6) 1.1 (0.5, 1.5) <0.001 1.7 (1.2, 2.5) 1.1 ( 0.6, 1.7) <0.001 Month 3 N 45 40 42 39 EWL% 28.1 (15.2, 46.8) 10.5 (7.2, 22.0) <0.001 31.3 (17.3, 51.4) 11.4 (8.8, 29.2) <0.001 TWL% 8.4 (4.9, 12.5) 3.5 (2.2, 7.1) <0.001 9.1 (5.0, 12.8) 3.5 (2.2, 6.5) <0.001 Body weight decrease (kg) 8.9 (4.4, 11.6) 4.3 (1.9, 6.5) <0.001 9.1 (4.8, 11.9) 4.3 (1.9, 6.0) <0.001 BMI decrease (kg/m2) 3.1 (1.7, 4) 1.4 (0.7, 2.3) <0.001 3.3 (1.8, 4.1) 1.4 (0.7, 2.3) <0.001 Month 4 N 43 38 39 36 EWL% 28.9 (15.4, 51.3) 12.8 (6.2, 29.0) 0.001 31.9 (25.7, 53.7) 12.8 (7.1, 29.1) <0.001 TWL% 8.3 (4.6, 14.3) 3.5 (1.9, 6.4) <0.001 9.0 (5.0, 14.8) 3.5 (2.2, 6.1) <0.001 Body weight decrease (kg) 8.7 (4.0, 13.1) 3.2 (1.6, 7.5) <0.001 9.2 (4.9, 14.4) 3.1 (1.9, 7.4) <0.001 BMI decrease (kg/m2) 3.0 (1.5, 4.7) 1.2 (0.7, 2.4) <0.001 3.3 (1.6, 4.8) 1.2 (0.7, 2.4) <0.001 Month 6 N 43 40 40 36 EWL% 28.5 (10.0, 50.8) 10.1 (2.2, 33.2) 0.008 29.2 (13.8, 51.6) 10.1 (2.2, 33.2) 0.005 TWL% 7.2 (3.4, 13.8) 3.0 (0.7, 7.6) 0.003 7.4 (3.5, 14.0) 3.0 (0.7, 7.6) 0.002 Body weight decrease (kg) 7.0 (3.3, 11.9) 2.8 (0.7, 6.4) 0.002 7.3 (3.4, 13.0) 2.9 (0.8, 6.4) 0.002 BMI decrease (kg/m2) 2.3 (1.2, 4.6) 1.0 (0.3, 2.2) 0.002 2.5 (1.3, 4.6) 1.0 (0.3, 2.2) 0.002 Month 9 N 45 43 42 37 EWL% 23.9 (2.0, 49.1) 9.6 (–1.8, 31.5) 0.044 24.8 (4.8, 50.9) 6.8 (–4.6, 31.8) 0.018 TWL% 5.9 (0.7, 12.8) 2.7 (–0.7, 7.1) 0.032 6.1 (1.7, 13.5) 1.9 (–1.1, 7.8) 0.015 Body weight decrease (kg) 7.0 (0.7, 12.0) 2.4 (–0.9, 7.7) 0.036 7.0 (1.7, 12.8) 1.7 (–1, 8.2) 0.018 BMI decrease (kg/m2) 2.2 (0.4, 4.1) 0.9 (–0.3, 2.8) 0.034 2.3 (0.7, 4.2) 0.6 (–0.4, 2.8) 0.015 All data were expressed as median (Q1, Q3). Body weight decrease was calculated by body weight at baseline minus body weight at each follow-up, and so was BMI decrease; P is the value for the Wilcoxon rank-sum test. BMI: Body mass index; EWL: Excess weight loss; FAS: Full analysis set; GBSS: Gastric bypass stent system; n: Number of participants; N: Number of participants who actually had the data; PPS: Per protocol set; TWL: Total weight loss. Insulin resistance improvement, as calculated by HOMA-IR decrease, was also evaluated. The decreases in HOMA-IR were significantly greater at months 4, 6, and 9 in the GBSS group than in the control group Supplementary Table 2, https://links.lww.com/CM9/C841. There were 245 AEs reported in 48 patients in the GBSS group, whereas 101 AEs were reported in 35 patients in the control group (P <0.001) Supplementary Table 3, https://links.lww.com/CM9/C841. Duodenal ulcer, with a rate of 42.9% (21/49), was the most common in the GBSS group. The percentage of positive fecal occult blood was 28.6% (14/49) in the GBSS group, which was higher than 2.0% (1/49) in the control group. There was no hematemesis, melena, or hemoglobin decrease in either group, and no one with positive fecal occult blood needed repeated gastroscopy or early removal. Gastrointestinal symptoms, including abdominal pain, nausea, and vomiting, were also more common in the GBSS group (P <0.05). The majority of gastrointestinal symptoms occurred only in the first week after implantation, were mild to moderate in severity, and disappeared after drug therapy. However, six participants could not tolerate gastrointestinal symptoms, and they chose early explantation. There were six serious adverse events (SAEs) in four subjects (8.2%) in the GBSS group and two SAEs in two subjects (4.1%) in the control group; this difference was not significant (P = 0.678). Two SAEs were considered relevant to the device or the operation: one was esophageal mucosal laceration during explantation, which was successfully treated with endoscopic clips and subsequent intravenous proton pump inhibitor; the other was stent migration to the descending duodenum, which was removed by gastroscopy on the 74th day after implantation. There were no cases of acute pancreatitis, gastrointestinal perforation, sleeve obstruction, or hepatic abscess in either group. In this study, GBSS had good short-term effects on weight control and improved insulin resistance. The bariatric efficacy of GBSS was similar to that of intragastric balloons, which had a 6-month TWL% of (7.6–10.2)%.2,3 However, its bariatric effect was not as good as that of endoscopic sleeve gastrectomy, which had a 6-month TWL% of 15.2%.4 The safety of GBSS is a great concern. The high ratios of duodenal ulcer and positive fecal occult blood prompted us to use proton pump inhibitor regularly after implantation. The occurrence of esophageal mucosal laceration was related to the incorrect selection of the retrieval line, which resulted in an incomplete stent retraction. The rate of device migration seemed lower than that of EndoBarrier, which was reported to be 6%.5 Furthermore, GBSS caused a greater decrease in serum iron but no iron deficiency anemia because the sleeve affected the absorption of iron by the duodenum and jejunum. Acknowledgements We thank the study participants and all the collaborating medical centers involved in this study. Funding This work was supported by the Beijing Hospitals Authority “Dengfeng” talent training plan (No. DFL20220101). Conflicts of interest The equipment of this trial was funded by Hangzhou Tangee Medical Technology Co., Ltd., but no payments were made to the authors or to the institutions. The research design, data collection, interpretation of clinical results, and submission of this manuscript were independently completed by the researchers and were not influenced by the company.