Evidence summary

Population and studies description

This interventional procedures overview is based on over 600,000 people from 1 RCT, 4 systematic reviews and meta-analyses, 2 cohort studies and 1 retrospective non-randomised comparative study with propensity score matched analysis. There is a significant overlap of primary studies on ESG included in 3 meta-analyses (Singh 2020, Marincola 2021 and de Miranda Neto 2020). About 12,000 people had an ESG procedure. Most of the included people had SG, about 7,000 people had LSG, about 3,000 had IGB and 110 people had only lifestyle modification. This is a rapid review of the literature, and a flow chart of the complete selection process is shown in figure 2. This overview presents 8 studies as the key evidence in table 2 and table 3, and lists the other 71 relevant studies in table 5.

Two meta-analysis (Marincola 2021, Singh 2020) pooled data from heterogenous observational studies and done an indirect comparison between ESG and LSG or IGB. One meta-analysis done a direct comparison between ESG and LSG (Beran 2022). Another meta-analysis assessed midterm outcomes of ESG alone (de Miranda Neto 2020). Primary studies on ESG included in these analyses were mainly observational studies and are prone to risk of bias and confounding. The quality of evidence was assessed using the objective criteria from GRADE in 1 meta-analysis (de Miranda Neto 2020) and evidence was graded as low certainty (confidence in the effect estimate is limited).

One RCT compared ESG plus a lifestyle modification programme with lifestyle modification alone and people in the control group crossed over at 52 weeks to ESG. ESG procedures were done by experienced gastroenterologists or bariatric surgeons. Double blinding was not possible in the RCT. The COVID-19 pandemic affected both primary and extended follow-up periods in ESG and control groups and limited lifestyle activities and clinical visits. Crossover ESG procedures were delayed by 1 to 6 months and the loss to follow-up rate was 16%. The study was mainly funded by the device company.

Two cohort studies (Sharaiha 2021, Bhandari 2023) reported long-term outcomes on ESG. Overall follow up ranged from 1 month to 5 years.

In a large retrospective analysis of ESG, propensity score matching was done with SG cases to mitigate potential confounding bias. Data on weight loss and adverse events are limited by short-term follow up (Gudur 2023).

Most of the people (70% to 85%) included in studies were female and between a mean age range of 33 years to 47 years. Studies included people with class 1 to class 3 obesity and comorbidities, with BMI ranging between 30 kg/m² and 40 kg/m² (mean BMI ranged between 33 kg/m² and 39 kg/m²).

The primary outcome reported was weight loss after ESG. Weight loss was assessed as per the ASGE task force recommended threshold in studies. Studies also assessed durability of these procedures. Adverse events were classified according to the ASGE in 2 meta-analyses (Marincola 2021, de Miranda Neto 2020) and complications were graded according to Clavien-Dindo classification system in the RCT (Abu Dayyeh 2022).

Primary studies included in the meta-analyses were done in the US, Spain, Saudi Arabia, Brazil, Australia, France and India. None of the studies were done in the UK.

Table 2 presents study details.

Figure 2 Flow chart of study selection

Table 2 Study details

Study no.	First author, date, country	Patients (male: female)	Age	Study design	Inclusion criteria	Intervention	Follow up
1	de Miranda Neto et al. (2020) Brazil	11 studies N=2,170 people with obesity or overweight. Average BMI pre-ESG was 35.78 kg/m² Average weight was 98.43 kg (95% CI 94.73 to 102.13) and 393 (18.11%) were males.	Mean age 42.3 years (95% CI 39.94 to 44.76)	Systematic review and meta-analysis	RCTs or observational studies, abstracts in English or Spanish, studies with at least 15 participants who underwent ESG with a minimum follow up of at least 1 month.	ESG alone. Procedure is done with endoscopic suturing system (OverStitch). Variations of procedure technique reported (the number of sutures used and the suturing patterns were described as 'Z' 'U', and triangular or rectangular). A layer of reinforcement sutures was reported in many studies.	Range 1 month to 18 months.
2	Abu Dayyeh BK et al. (2022) (MERIT trial) US	N=187 people with class 1 and class 2 obesity. ESG (n=77) versus control (n=110)	ESG mean age 47.3 years. Control mean age 45.7 years. ESG 88% female Control 84% female	RCT	People aged 21 to 65 years with class 1 or class 2 obesity who agreed to comply with lifelong dietary restrictions required by the procedure, with a BMI between 30 kg/m² and less than 40 kg/m², with a history of failure with non-surgical weight loss methods.	ESG (OverStitch device) plus moderate intensity lifestyle modifications (ESG group) versus lifestyle modifications alone (control group). People in the control group who did not reach more than 25% EWL and completed follow-up (n=72) crossed over and had ESG after 52 weeks and were followed up for an additional 52 weeks. Lifestyle modifications included low-calorie diet plan and physical activity counselling, which was customised.	52 weeks: ESG n=68 Control n=89 Follow up of primary ESG group extended to 104 weeks to evaluate the durability of the original procedure or the effect of suture reinforcement in 5 individuals who had it at investigators discretion at 52 weeks. 104 weeks follow up: ESG n= 50 Control n=59
3	Marincola et al. (2021) Italy	16 studies N=2,188 people with obesity. LSG: 1,429; ESG: 759 Mean BMI 34.34 kg/m² and 34.72 kg/m² for LSG and ESG LSG: 1 RCT and 7 observational studies ESG: 8 observational studies 79.6% female	Mean age LSG 35.5 years, ESG 38.5 years	Systematic review and meta-analysis (pooled data from non-comparative studies, heterogenous study designs)	Studies with people with obesity who have a baseline BMI between 30 and 40 kg/m² with a minimum of 12 months of follow up and a rate of complications reported	ESG (with OverStitch device) versus LSG	12 months
4	Beran et al. (2022) US	7 studies N=6,775 people with obesity 3,413 ESG versus 3,362 LSG (5 retrospective cohort studies and 2 prospective cohort studies) Mean baseline BMI was 33.7 kg/m² (SD 4.8)	Mean age was 34.9 years (SD 10.2) 87% female	Systematic review and meta-analysis	Studies that performed a direct comparison between ESG and LSG in people with obesity	ESG (with OverStitch device) versus LSG	Range 6 months to 36 months
5	Singh et al. (2020) US	28 studies N=5,004 people with obesity ESG patients n=1,979 (mean BMI 36.1 kg/m²) IGB patients n=3,025 (mean BMI 41.7 kg/m²)	Not reported	Systematic review and meta-analysis	RCTs or observational studies in which people had IGB or ESG alone with or without lifestyle modification for obesity, studies reporting %TWL or %EWL with at least 12 months of follow up were included	IGB and ESG for the treatment of obesity 1 study compared ESG to IGB, 9 observational studies evaluated ESG alone, while 18 studies (4 RCTs and 14 observational studies) evaluated IGB	12 months
6	Bhandari et al. (2023) India	N= 612 people with obesity Mean BMI 34.30 kg/m² (SD 5.05) 69.3% (494 out of 612) female	Mean age 40.70 years (SD 12.66)	Prospective cohort study (single centre)	People with a BMI over 30 kg/m² (or over 27 kg/m² with comorbidities).	ESG for treatment of obesity; done by single surgeon Liquid diet for 2 weeks, followed by modified bariatric diet for 4 weeks	4 years 570 (93.1%), 1 year 552 (90.2%), 2 years 466 (81.7%) 3 years 254 (81.9%) 4 years
7	Sharaiha et al. (2021) USA	N=216 people with obesity 68% (146 out of 216) female Mean BMI of 39.6 kg/m²	Mean age 46 years (SD 13)	Prospective cohort study	People with a BMI of over 30 kg/m² (or over 27 kg/m² with comorbidities), and failure of previous non-invasive weight loss measures including pharmacotherapy (if no change in weight for at least 3 months or if they were gaining weight) to achieve TWL of at least 5% People with a BMI of more than 40 kg/m² who refused bariatric surgery or were deemed to be high-risk surgical candidates	ESG for treatment of obesity in a single centre done by a single surgeon People restricted to a full-liquid diet for the first 2 weeks, then advanced to a modified bariatric diet for an additional 4 weeks Adjunct anti-obesity pharmacotherapy was given if no change in weight for at least 3 months or if they were gaining weight (before ESG n=78; after ESG n=58).	Up to 5 years 203, 96, and 68 people were eligible for 1-year, 3-year and 5-year follow up, respectively Data available 1 year 70% (142), 3 years 71% (68), and 5 years 82% (56)
8	Gudur (2021) US	N=603,517 (6,054 people had ESG (6,053 in matched cohort) and 597,463 people had SG (30,270 in matched cohort) Most people were female (ESG 84.5% versus SG 79.8%; matched 82.3%). Unmatched analysis: mean BMI ESG 40.5 kg/m² versus SG 44.9 kg/m²; matched: 42.8 kg/m²	Unmatched analysis ESG 47.5 years versus SG 44.2 years. Matched 44.9 years.	Retrospective non-randomised comparative study and propensity score matched analysis	Data on ESG and SG from 2016 to 2020 the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) database	ESG versus surgical SG.	30 days

Table 3 Study outcomes

First author, date	Efficacy outcomes	Safety outcomes
de Miranda Neto et al. (2020)	Pooled mean %TWL 1 month: 8.56 (95% CI 7.94 to 9.18, I²=0.3%, 5 studies, n=2,538) 3 months: 11.65 (95% CI 10.76 to 12.53, I²=0%, 5 studies, n=2,296) 6 months: 15.32 (95% CI 14.54 to 16.10, I²=15.3%, 9 studies, n=2,256) 9 months: 16.15 (95% CI 14.94 to 17.37, I²=0%, 3 studies, n=948) 12 months: 17.33 (95% CI 16.30 to 18.36, I²=10.8%, 9 studies, n=1,706) 18 months: 16.80 (95% CI 13.02 to 20.56, I²=0%, 2 studies, n=252) Certainty of evidence was low for all time periods Pooled mean %EWL 1 month: 31.08 (95% CI 20.79 to 41.36, I²=0%, 3 studies, n=2,100) 3 months: 46.13 (95% CI 38.79 to 53.47, I²=0%, 3 studies, n=1,838) 6 months: 55.80 (95% CI 50.61 to 60.99, I²=15.09%, 6 studies, n=1,816) 9 months: 66.20 (95% CI 57.54 to 74.86, I²=8.52%, 3 studies, n=912) 12 months: 60.07 (95% CI 53.39 to 66.74, I²=18.09%, 6 studies, n=1,148) 18 months: 73.04 (95% CI 58.94 to 87.14, I²=0%, 2 studies, n=252) Certainty of evidence was low for all time periods Pooled mean %AWL 1 month: 7.73 (95% CI 7.06 to 8.40, I²=16.82%, 3 studies, n=2,020) 3 months: 10.23 (95% CI 8.44 to 12.03, I²=0%, 3 studies, n=1,768) 6 months: 14.88 (95% CI 13.33 to 16.42, I²=0%, 6 studies, n=1,730) 9 months: 15.44 (95% CI 12.70 to 18.17, I²=0%, 2 studies, n=878) 12 months: 17.32 (95% CI 15.65 to 18.99, I²=0%, 7 studies, n=1,218) 18 months: 15.95 (95% CI 10.95 to 20.95, I²=0%, 2 studies, n=252) Certainty of evidence was low for all time periods	Procedure related mortality = 0 Overall adverse events (graded according to the ASGE lexicon as mild, moderate and severe) = 2.3% (95% CI 1.2 to 4.1, I²=24.08%, 7 studies, n=38 events). Mild: 1.5% (95% CI 0.5 to 4.3, I²=0%, 2 studies, n=12 events) Moderate: 1.7% (95% CI 0.9 to 3.1, I²=8.16%, 6 studies, n=22 events) Severe: 0.8% (95% CI 0.3 to 2.0, I²=0%, 3 studies, n=4 events) Gastrointestinal bleeding (n=13) and perigastric collections (n=10) were the most common major adverse events reported. Other events included severe abdominal pain (n=8), fever (n=5), deep vein thrombosis (n=1) and pneumothorax (n=1). Most events were managed conservatively, but 2 people with gastrointestinal bleeding needed sclerotherapy and 3 people with perigastric collections needed surgical interventions (including closure of a gastric fistula and 1 reversal of ESG). Nausea, vomiting, and mild abdominal pain were not considered as adverse events.
Abu Dayyeh et al. (2022)	Mean %EWL at 52 weeks in primary ESG and control groups ESG group 49.2% (SD 32.0) and 3.2% (SD 18.6) for control group-lifestyle modification programme alone (p<0.0001). Crossover group (n=72) achieved mean 44.1% (SD 35.7) EWL at 52 weeks from crossover. Mean EWL of all people who had the ESG procedure (both primary ESG and crossovers) at week 52 post procedure was 46.7% (SD 33.8). 81% (55 out of 68) of people reached the primary endpoint (25% or more EWL) at 52 weeks, compared with 72% (46 out of 64) in the crossover ESG group with 52 weeks follow up (p=0.21). Mean %TWL at 52 weeks ESG group 13.6% (SD 8.0) and 0.8% (SD 5.0) for control group (p<0.0001). After adjusting for age, sex, type 2 diabetes, hypertension, and baseline BMI in a modified intention to treat analysis with mixed-effects models, people in the ESG group had a MD of 44.7% (95% CI 37.5 to 51.9) EWL and 12.6% (95% CI 10.7 to 14.5) TWL, compared with the control group at 52 weeks (p <0.0001 using last observation carried forward and p<0.0001 using mixed-model imputations for missing data). Proportion of people with 25% or more EWL at 52 weeks ESG group 77% (59 out of 77) versus control group 12% (13 out of 110; p<0.0001). Proportion of people with 25% or more EWL at 104 weeks (in ESG group only) 68% (41 out of 60); 41% EWL, 11.4% TWL. Change in at least 1 obesity comorbidity at 52 weeks ESG group: improved in 80% (41 out of 51) and worsened in 12% (6 out of 51) Control group: improved in 45% (28 out of 62) and worsened in 50% (31 out of 62) Diabetes Improved: ESG 92% (12 out of 13), control 15% (4 out of 27), MD –77.5 (10.1; 95% CI –91.4 to –47.4) p<0.0001. Worsened: ESG 0% (0 out of 13), control 44% (12 out of 27), MD 44.4 (9.6; 95% CI 16.1 to 60.2) p=0.0041. Hypertension Improved: ESG 67% (24 out of 36) control 40% (19 out of 48), MD –27.1 (10.6; 95% CI –46.1 to 5.5) p=0.014 Worsened: ESG 6% (2 out of 36), control 23% (11 out of 48), MD 17.4 (7.2; 95% CI 1.5 to 30.7) p=0.029 Metabolic syndrome Improved: ESG 83% (24 out of 29), control 35% (10 out of 29), MD –48.3 (11.3; 95% CI –67.0 to –23.3) p=0.0002 Worsened: ESG 0% (0 out of 29), control 38% (11 out of 29), MD 37.9 (9.0; 95% CI 17.2 to 53.7) p=0.0002 Hyperlipidaemia Improved: ESG 40% (6 out of 15), control 32% (8 out of 25), MD –8.0 (15.7; 95% CI –37 to –22) p=0.61 Worsened: ESG 27% (4 out of 15), control 28% (7 out of 25), MD 1.3 (14.9; 95% CI –28 to 28) p=0.93 Authors state that quality of life, eating behaviours, improvement in depression, and patients' satisfaction were all superior in the ESG group compared with the control group.	Primary safety endpoint at 104 weeks 5% or less device or procedure related serious adverse events (in primary and crossover ESG groups). Total adverse events: n=927 events in 92% (138 out of 150) Serious adverse events:2% (3 out of 131) Abdominal abscess, grade 3 managed endoscopically (n=1); upper gastrointestinal bleed, managed conservatively without transfusion (n=1); malnutrition requiring endoscopic reversal of the ESG (n=1). ESG reversal (requested by patient) n=1 Minor adverse events, 66% (612 out of 927) Gastrointestinal symptoms including pain, heartburn, nausea, and vomiting (symptoms resolved within 1 week) Hospital admission for management of accommodative symptoms: 4% (6 out of 150).
Marincola et al. (2021)	Pooled mean % EWL at 12 months ESG (n=759): 62.2% (95% CI 57.8 to 66.6; I²=65.52, Tau²=24.68; Cochran's Q test p=0.005). LSG (n=1,429): 80.32% (95% CI 68.1 to 92.5; I²=98.88, Tau²=56.62; Cochran's Q test p=0.001). Absolute difference=18.1% (p=0.0001)	Pooled mean peri-procedural major or minor adverse events ESG (n=1,778): 0.15% (Cochran's Q test p=0.0001, I²=42.81). LSG (n=1,929): 0.30% (Cochran's Q test p=0.0001, I²=62.26). Difference in mean rate of major or minor adverse events was 0.19% (χ²=1.602, p=0.2056).
Beran et al. (2022)	Pooled %TWL (meta-analysis of 7 studies) ESG: 6 months: 15.2 (SD 6.3); 12 months 19.1 (SD 7.9); 24 months 16.4 (SD 10.1) LSG: 6 months 18.8 (SD 7.5); 12 months 28.9 (SD 8.2); 24 months 22.3 (SD 8.3) Pooled %EWL ESG: 6 months 66.7% (SD 28.7) and 12 months 71.04% (SD 24.6) LSG 6 months 76.6% (SD 31.3) and 12 months 94.9% (SD 20.6) %TWL 6 months: MD −7.48 (95% CI −10.44 to −4.52; p<0.0001, I² = 94%), 7 studies (n=5,516), ESG n=2,882 versus LSG n=2,634 12 months: MD −9.90; (95% CI −10.59 to −9.22; p<0.00001, I²=9%), 4 studies (n=5,113), ESG n=2,542 versus LSG n=2,571 24 months: MD −7.63 (95% CI −11.31 to −3.94; p<0.0001, I² = 85%), 2 studies (n=5,260), ESG n=2,641 versus LSG n=2,619 %EWL 6 months: MD −10.23 (95% CI −11.90 to −8.56; p<0.00001, I²=0%), 3 studies (n=4,884), ESG n=2,526 versus LSG n=2,358 12 months: MD −23.99 (95% CI −25.30 to −22.68; p<0.00001, I²=0%), 2 studies (n=4,642), ESG n=2,252 versus LSG n=2,390 Improvement or remission of diabetes mellitus 2 studies, LSG 81.9% (73 out of 114) versus ESG 64% (289 out of 353), RR 0.78, 95% CI 0.68 to 0.91, p=0.001, I²=0% Improvement or remission of hypertension 2 studies,ESG 51% (53 out of 104) versus LSG 45.6% (57 out of 125), RR 1.12, 95% CI 0.86 to 1.47, p=0.39, I²=0%	Overall adverse events Pooled analysis of 7 studies ESG 0.7% (24 out of 3,250) versus LSG 1.7% (52 out of 3,104); RR 0.51; 95% CI 0.23 to 1.11; p=0.09, I²=50%. Pooled analysis of 6 studies (Excluding 1 large study Alqahtani 2019, which caused significant between study heterogeneity) ESG 2.5% (10 out of 395) versus LSG 12.2% (42 out of 344); RR 0.39 (95% CI 0.18 to 0.83), p=0.01, I²=23% New onset GERD (2 studies) ESG 1.3% (1 out of 77) versus LSG 17.9% (19 out of 106); RR 0.10, (95% CI 0.02–0.53), p=0.006, I²=0%
Singh et al. (2020)	ESG Mean %TWL 6 months: 15.34 (95% CI 14.33 to 16.35, I²=92.23, 9 studies) 12 months: 17.51 (95 % CI 16.44 to 18.58, I²=88.35, p=0.004, 9 studies) 18 to 24 months: 17.85 (95% CI 15.85 to 19.86, I²=69.57, p=0.025, 4 studies) Mean %EWL 6 months: 55.61 (95% CI 50.28 to 60.95, I²=83.38, 6 studies) 12 months: 60.51 (95% CI 54.39 to 66.64, I²=66.67, p=0.22, 6 studies) 18 to 24 months: 66.77 (95% CI 57.54 to 76.00, I²=67.72, p=0.047, 4 studies) IGB Mean %TWL (4 RCTs and 5 observational studies) 6 months: 12.16 (95% CI 10.37 to 13.95, I²=91.32%, 9 studies) 12 months: 10.35 (95% CI 8.38 to 12.32, I²=89.80%, p=0.13, 9 studies) 18 to 24 months: 6.89 (95% CI 3.78 to 10.01, I²=96.50%, p=0.003, 3 studies) Mean %EWL (2 RCTs and 13 observational studies) 6 months: 34.83 (95% CI 30.97 to 38.69, I² = 97.71%, 15 studies) 12 months: 29.65 (95% CI 25.40 to 33.91, I²=97.51%, p=0.10, 13 studies) 18 to 24 months: 23.88 (95% CI 17.41 to 30.33, I²=87.05%, p=0.001, 5 studies) Comparative analysis ESG achieved statistically significantly higher %TWL and %EWL than IGB. The difference in mean %TWL between ESG and IGB was 3.07 at 6 months (95% CI 1.46 to 4.67, p=0.002), 7.33 at 12 months (95% CI 5.22 to 9.44, p=0.0001), and 11.51 at 18 to 24 months (95% CI 5.33 to 17.69, p=0.0003). The difference in mean %EWL between ESG and IGB was 20.80 at 6 months (95% CI 12.50 to 29.10, p=0.0001), 30.99 at 12 months (95% CI 22.81 to 39.16, p=0.0001), and 43.78 at 18 to 24 months (95% CI 35.98 to 51.58, p=0.0001).	ESG Overall adverse events 1.52% Severe abdominal pain (2.2%) Mild to moderate abdominal pain (50.65%) Nausea, managed with medications (32.31%) Gastrointestinal bleeding (0.61%) Perigastric fluid collection (0.45%) Perforation (0.10%) Post-procedure fever (0.25%) Pulmonary embolism and DVT (0.10%) Reversal of ESG (because of persistent symptoms; 0.15%, n=3). Mortality=0 IGB Overall adverse events in 4% of people. Abdominal pain (32.51%) Nausea (55.09%) Balloon hyperinflation (0.03%) Balloon resting in antrum (0.10%) Severe dehydration (0.77%) Esophagitis (2.33%) Gastrointestinal bleeding (0.21%) Obstruction (0.10%) Perforation (0.10%) Ulcers (0.24%) Severe GERD (0.17%) Early removal of IGB because of intolerance (6%) Mortality (3 people; 2 due to acute gastric perforation, 1 due to cardiac arrest at 4 weeks).
Bhandari et al. (2023)	Mean %TWL 6 months: 12.6% (95% CI 9.28 to 19.06, p<0.001) with 94% (575 out of 612) of people achieving 5% or more TWL 1 year: 21.20% (95% CI 20.81 to 21.59, p<0.001) with 98% (558) of people achieving 5% or more TWL 2 years: 20.05% (95% CI 19.61 to 20.48, p<0.001) with 93% (513) of people achieving 5% or more TWL 3 years: 18.74% (95% CI 18.31 to 19.12, p<0.001) with 91% (425) of people maintaining 5% or more TWL 4 years: 18.19% (95% CI 17.72 to 18.57, p<0.001) with 90% (229) of people maintaining 5% or more TWL People's %TWL at their nadir weight after ESG had a mean of 18.9% (95% CI 18.5 to 19.3). People's mean weight gain after nadir was 3.5 kg until the end of the follow-up period (95% CI 3.1 to 3.3) Mean %EWL 1-year 56.9% (95% CI 56.51 to 57.30, p<0.001) with 90% (513) of people achieved 25% or more EWL 2 years, 54.4% (95% CI 54.03 to 54.80, p<0.001) with 442 and 80% of people maintained 25% or more EWL 3 years, 50.1% (95% CI 49.71 to 50.08, p<0.001) with 327 and 70% (327) of people maintained 25% or more EWL 4 years, 49.3% (95% CI 48.91 to 49.68, p<0.001) with 70% (177) of people maintained 25% or more EWL People's %EWL at their nadir weight after ESG had a mean of 57.2% (95% CI 56.8 to 57.5) Mean duration of surgery was 61.96 minutes (SD 2.1; range 45.2 to 121.1) and hospital length of stay was 3 days (range 2 to 4 days) Resolution or improvement of comorbidities within 90 days Diabetes 51.2% (121 out of 236) Hypertension 65.8% (216 out of 328) Dyslipidaemia 73.6% (302 out of 410) Obstructive sleep apnoea 89.9% (401 out of 446)	Adverse events % (n) Post-operative complications included: Nausea 35.45% (217) Vomiting 17.6% (108) Bloating 12.25% (75) Abdominal pain 46.6% (284) Generalised weakness 2.6% (16) Revision or redo surgery 0.3% (2) of people underwent revision to SG after 12 months of primary ESG, due to weight regain. 0.4% (3) of people had reversal redo-ESG
Sharaiha 2021	Mean %TWL 1 year: 15.6% (95% CI 14.1 to 17.1; p<0.001) 3 years: 14.9% (95% CI 12.1 to 17.7; p<0.001) 5 years: 15.9% (95% CI 11.7 to 20.5; p<0.001) At nadir weight: 16.7% (95% CI 15.6 to 17.7, p<0.0001) Percentage of people achieving 5% or more TWL 1 year: 89% (n=118) 3 years: 85% (n=50) 5 years: 90% (n=28) At nadir weight: 96% (207 out of 216) Percentage of people achieving 10% or more TWL 1 year: 77% (n=103) 3 years: 63% (n=37) 5 years: 61% (n=19) At nadir weight: 80% (172 out of 216) Mean weight gain after ESG 2.9 kg from nadir until the end of the follow-up period (95% CI 2.3 to 3.7) Mean EWL 1 year: 47.9% (95% CI 42.4 to 53.3; p<0.001) 3 years: 45.1% (95% CI 34.9 to 55.2; p<0.001) 5 years: 45.3% (95% CI 32.9 to 57.7; p<0.001) At nadir weight: 53.5% (95% CI, 49.1 to 57.9, p value not reported). Percentage of people achieving 25% EWL 1 year: 80% 3 years: 68% 5 years: 74% Adjunct pharmacotherapy (in 58 cases with TWL of 5% or more and weight regain at a median 5 months) Adjunct pharmacotherapy was not associated with a significant difference in mean TWL compared with people who did not have it. Weight stabilised but no additional weight loss noted. Repeat ESG in those with weight regain (n=13 with mean 2.6 kg weight regain) Average TBWL of 21.5% before second ESG was stabilized at 24.2% at 1 year after the second ESG. Predictors of TWL The amount of weight loss at 1 month after the procedure, patient's compliance with follow up, and endoscopist's experience are independent predictors of weight loss.	Mild adverse events in 32% of people. Heartburn (up to 3 weeks) 25% (n=54) Nausea or vomiting (managed with medications) 25% (n=43) Epigastric pain (beyond 24 hours, managed with medications) 31% (n=65) Constipation (managed with laxatives) 29% (n=63) Superficial oesophageal tear (from the over tube with the device, managed endoscopically) n=1 Asymmetric paraesthesia n=1 Spinal white-matter plaques and low serum thiamine (vitamin B1) n=1 Moderate adverse events 1.3% (n=3) Pain, n=1 (in left upper quadrant 18 months after ESG, scan showed sutures with bridging fibrosis bands. ESG sutures were released leading to increased gastric volume and improvement in pain) Perigastric leak n=2 (after dietary indiscretion, managed with antibiotics and percutaneous drainage in 1) Bariatric surgery or SG (for inadequate weight loss) 1% (n=2). Revision procedures Repeat ESG (for mean 2.6 kg weight gain) 6% (13 out of 216). Average TWL after 1 year was 24.2%. Laparoscopic SG (due to inadequate weight loss after ESG) 1% (n=2)
Gudur 2023	Mean% TWL within 30 days follow up (propensity score matched cohort) ESG 4.0% (SD 6.7) versus SG 5.4% (SD 4.3); p<0.001. Mean change in BMI from pre-operative to post-operative ESG –1.77 kg/m² (SD 2.89) versus SG –2.36 kg/m² (SD 1.78); p< 0.001 Procedure time ESG 62.9 minutes versus SG 72.4 minutes, p<0.001. Length of stay ESG 0.87 days versus SG 1.45 days, p<0.001.	Overall adverse events SG 1.1% (340 out of 30,270) versus ESG 1.4% (86 out of 6,053); p=0.058 Factors impacting adverse events Propensity score matched analysis demonstrated that black ethnicity associated with a higher risk of adverse events in SG compared with ESG (OR, 1.23; 95% CI, 1.13 to 1.35). Multivariate regression noted that people with a higher BMI were less likely to have an adverse event after ESG. In the ESG cohort, albumin, BMI, renal insufficiency, age, therapeutic anticoagulation, non-insulin-dependent diabetes, chronic steroid use, and female gender were statistically significant patient factors. These variables were also associated with adverse events in the SG cohort. Additional variables associated with adverse events in the SG cohort were GERD, insulin-dependent diabetes, previous surgery, hypertension, black ethnicity, history of pulmonary embolism, previous cardiac surgery, independent preoperative functional status, sleep apnoea, smoking, history of myocardial infarction, chronic obstructive pulmonary disease, dialysis, and hyperlipidaemia Readmission within 30 days ESG 3.8% (231 out of 6,053) versus SG 2.6% (794 out of 30,270); p<0.001 Causes of readmission for ESG Nausea, vomiting, fluid or electrolyte disturbance 26% (n=72) Abdominal pain 17% (n=47) Gastrointestinal leak 6% (n=16) Gastrointestinal bleeding 5% (n=15) Causes of readmission for SG Nausea, vomiting, fluid or electrolyte disturbance 31% (n=5,686) Abdominal pain 12% (n=2,097) Gastrointestinal leak 6% (n=1,063) Reoperation within 30 days ESG 1.4% (86 out of 6,053) versus SG 0.8% (238 out of 30,270), p<0.001 Main causes of reoperation for ESG Abdominal re-exploration 13% (n=10), Obstruction or GI perforation or pain 10% Main causes of reoperation for SG Abdominal re-exploration 27% (n=1,315) Gastrointestinal bleeding 25% (n=1,217) Staple line leak 15% (n=1,217) Reintervention within 30 days ESG 2.8% (171 out of 6,053) versus SG 0.7% (209 out of 30,270); p<0.001. Common endoscopic interventions after ESG Therapeutic endoscopy (was mainly for stent placement or dilatation) ESG 48% (n=52) versus SG 32% (n=1,714) Treatment for dehydration ESG 2.4% (147 out of 6,053) versus 3.3% (993 out of 30,270); p=0.001 Emergency visit with no admission ESG 4.9% (294 out of 6,053) versus 5.9% (1,786 out of 30,270); p=0.002.

First author, date

Efficacy outcomes

Safety outcomes

de Miranda Neto et al. (2020)

Pooled mean %TWL

1 month: 8.56 (95% CI 7.94 to 9.18, I²=0.3%, 5 studies, n=2,538)

3 months: 11.65 (95% CI 10.76 to 12.53, I²=0%, 5 studies, n=2,296)

6 months: 15.32 (95% CI 14.54 to 16.10, I²=15.3%, 9 studies, n=2,256)

9 months: 16.15 (95% CI 14.94 to 17.37, I²=0%, 3 studies, n=948)

12 months: 17.33 (95% CI 16.30 to 18.36, I²=10.8%, 9 studies, n=1,706)

18 months: 16.80 (95% CI 13.02 to 20.56, I²=0%, 2 studies, n=252)

Certainty of evidence was low for all time periods

Pooled mean %EWL

1 month: 31.08 (95% CI 20.79 to 41.36, I²=0%, 3 studies, n=2,100)

3 months: 46.13 (95% CI 38.79 to 53.47, I²=0%, 3 studies, n=1,838)

6 months: 55.80 (95% CI 50.61 to 60.99, I²=15.09%, 6 studies, n=1,816)

9 months: 66.20 (95% CI 57.54 to 74.86, I²=8.52%, 3 studies, n=912)

12 months: 60.07 (95% CI 53.39 to 66.74, I²=18.09%, 6 studies, n=1,148)

18 months: 73.04 (95% CI 58.94 to 87.14, I²=0%, 2 studies, n=252)

Certainty of evidence was low for all time periods

Pooled mean %AWL

1 month: 7.73 (95% CI 7.06 to 8.40, I²=16.82%, 3 studies, n=2,020)

3 months: 10.23 (95% CI 8.44 to 12.03, I²=0%, 3 studies, n=1,768)

6 months: 14.88 (95% CI 13.33 to 16.42, I²=0%, 6 studies, n=1,730)

9 months: 15.44 (95% CI 12.70 to 18.17, I²=0%, 2 studies, n=878)

12 months: 17.32 (95% CI 15.65 to 18.99, I²=0%, 7 studies, n=1,218)

18 months: 15.95 (95% CI 10.95 to 20.95, I²=0%, 2 studies, n=252)

Certainty of evidence was low for all time periods

Procedure related mortality = 0

Overall adverse events (graded according to the ASGE lexicon as mild, moderate and severe) = 2.3% (95% CI 1.2 to 4.1, I²=24.08%, 7 studies, n=38 events).

Mild: 1.5% (95% CI 0.5 to 4.3, I²=0%, 2 studies, n=12 events)
Moderate: 1.7% (95% CI 0.9 to 3.1, I²=8.16%, 6 studies, n=22 events)
Severe: 0.8% (95% CI 0.3 to 2.0, I²=0%, 3 studies, n=4 events)

Gastrointestinal bleeding (n=13) and perigastric collections (n=10) were the most common major adverse events reported. Other events included severe abdominal pain (n=8), fever (n=5), deep vein thrombosis (n=1) and pneumothorax (n=1).

Most events were managed conservatively, but 2 people with gastrointestinal bleeding needed sclerotherapy and 3 people with perigastric collections needed surgical interventions (including closure of a gastric fistula and 1 reversal of ESG).

Nausea, vomiting, and mild abdominal pain were not considered as adverse events.

Abu Dayyeh et al. (2022)

Mean %EWL at 52 weeks in primary ESG and control groups

ESG group 49.2% (SD 32.0) and 3.2% (SD 18.6) for control group-lifestyle modification programme alone (p<0.0001).

Crossover group (n=72) achieved mean 44.1% (SD 35.7) EWL at 52 weeks from crossover.

Mean EWL of all people who had the ESG procedure (both primary ESG and crossovers) at week 52 post procedure was 46.7% (SD 33.8).

81% (55 out of 68) of people reached the primary endpoint (25% or more EWL) at 52 weeks, compared with 72% (46 out of 64) in the crossover ESG group with 52 weeks follow up (p=0.21).

Mean %TWL at 52 weeks

ESG group 13.6% (SD 8.0) and 0.8% (SD 5.0) for control group (p<0.0001).

After adjusting for age, sex, type 2 diabetes, hypertension, and baseline BMI in a modified intention to treat analysis with mixed-effects models, people in the ESG group had a MD of 44.7% (95% CI 37.5 to 51.9) EWL and 12.6% (95% CI 10.7 to 14.5) TWL, compared with the control group at 52 weeks (p <0.0001 using last observation carried forward and p<0.0001 using mixed-model imputations for missing data).

Proportion of people with 25% or more EWL at 52 weeks

ESG group 77% (59 out of 77) versus control group 12% (13 out of 110; p<0.0001).

Proportion of people with 25% or more EWL at 104 weeks (in ESG group only) 68% (41 out of 60); 41% EWL, 11.4% TWL.

Change in at least 1 obesity comorbidity at 52 weeks

ESG group: improved in 80% (41 out of 51) and worsened in 12% (6 out of 51)

Control group: improved in 45% (28 out of 62) and worsened in 50% (31 out of 62)

Diabetes

Improved: ESG 92% (12 out of 13), control 15% (4 out of 27), MD –77.5 (10.1; 95% CI –91.4 to –47.4) p<0.0001.

Worsened: ESG 0% (0 out of 13), control 44% (12 out of 27), MD 44.4 (9.6; 95% CI 16.1 to 60.2) p=0.0041.

Hypertension

Improved: ESG 67% (24 out of 36) control 40% (19 out of 48), MD –27.1 (10.6; 95% CI –46.1 to 5.5) p=0.014

Worsened: ESG 6% (2 out of 36), control 23% (11 out of 48), MD 17.4 (7.2; 95% CI 1.5 to 30.7) p=0.029

Metabolic syndrome

Improved: ESG 83% (24 out of 29), control 35% (10 out of 29), MD –48.3 (11.3; 95% CI –67.0 to –23.3) p=0.0002

Worsened: ESG 0% (0 out of 29), control 38% (11 out of 29), MD 37.9 (9.0; 95% CI 17.2 to 53.7) p=0.0002

Hyperlipidaemia

Improved: ESG 40% (6 out of 15), control 32% (8 out of 25), MD –8.0 (15.7; 95% CI –37 to –22) p=0.61

Worsened: ESG 27% (4 out of 15), control 28% (7 out of 25), MD 1.3 (14.9; 95% CI –28 to 28) p=0.93

Authors state that quality of life, eating behaviours, improvement in depression, and patients' satisfaction were all superior in the ESG group compared with the control group.

Primary safety endpoint at 104 weeks

5% or less device or procedure related serious adverse events (in primary and crossover ESG groups).

Total adverse events: n=927 events in 92% (138 out of 150)

Serious adverse events:2% (3 out of 131)

Abdominal abscess, grade 3 managed endoscopically (n=1); upper gastrointestinal bleed, managed conservatively without transfusion (n=1); malnutrition requiring endoscopic reversal of the ESG (n=1).

ESG reversal (requested by patient) n=1

Minor adverse events, 66% (612 out of 927)

Gastrointestinal symptoms including pain, heartburn, nausea, and vomiting (symptoms resolved within 1 week)

Hospital admission for management of accommodative symptoms: 4% (6 out of 150).

Marincola et al. (2021)

Pooled mean % EWL at 12 months

ESG (n=759): 62.2% (95% CI 57.8 to 66.6; I²=65.52, Tau²=24.68; Cochran's Q test p=0.005).

LSG (n=1,429): 80.32% (95% CI 68.1 to 92.5; I²=98.88, Tau²=56.62; Cochran's Q test p=0.001).

Absolute difference=18.1% (p=0.0001)

Pooled mean peri-procedural major or minor adverse events

ESG (n=1,778): 0.15% (Cochran's Q test p=0.0001, I²=42.81).

LSG (n=1,929): 0.30% (Cochran's Q test p=0.0001, I²=62.26).

Difference in mean rate of major or minor adverse events was 0.19% (χ²=1.602, p=0.2056).

Beran et al. (2022)

Pooled %TWL (meta-analysis of 7 studies)

ESG: 6 months: 15.2 (SD 6.3); 12 months 19.1 (SD 7.9); 24 months 16.4 (SD 10.1)

LSG: 6 months 18.8 (SD 7.5); 12 months 28.9 (SD 8.2); 24 months 22.3 (SD 8.3)

Pooled %EWL

ESG: 6 months 66.7% (SD 28.7) and 12 months 71.04% (SD 24.6)

LSG 6 months 76.6% (SD 31.3) and 12 months 94.9% (SD 20.6)

%TWL

6 months: MD −7.48 (95% CI −10.44 to −4.52; p<0.0001, I² = 94%), 7 studies (n=5,516), ESG n=2,882 versus LSG n=2,634

12 months: MD −9.90; (95% CI −10.59 to −9.22; p<0.00001, I²=9%), 4 studies (n=5,113), ESG n=2,542 versus LSG n=2,571

24 months: MD −7.63 (95% CI −11.31 to −3.94; p<0.0001, I² = 85%), 2 studies (n=5,260), ESG n=2,641 versus LSG n=2,619

%EWL

6 months: MD −10.23 (95% CI −11.90 to −8.56; p<0.00001, I²=0%), 3 studies (n=4,884), ESG n=2,526 versus LSG n=2,358

12 months: MD −23.99 (95% CI −25.30 to −22.68; p<0.00001, I²=0%), 2 studies (n=4,642), ESG n=2,252 versus LSG n=2,390

Improvement or remission of diabetes mellitus

2 studies, LSG 81.9% (73 out of 114) versus ESG 64% (289 out of 353), RR 0.78, 95% CI 0.68 to 0.91, p=0.001, I²=0%

Improvement or remission of hypertension

2 studies,ESG 51% (53 out of 104) versus LSG 45.6% (57 out of 125), RR 1.12, 95% CI 0.86 to 1.47, p=0.39, I²=0%

Overall adverse events

Pooled analysis of 7 studies

ESG 0.7% (24 out of 3,250) versus LSG 1.7% (52 out of 3,104); RR 0.51; 95% CI 0.23 to 1.11; p=0.09, I²=50%.

Pooled analysis of 6 studies (Excluding 1 large study Alqahtani 2019, which caused significant between study heterogeneity)

ESG 2.5% (10 out of 395) versus LSG 12.2% (42 out of 344); RR 0.39 (95% CI 0.18 to 0.83), p=0.01, I²=23%

New onset GERD (2 studies)

ESG 1.3% (1 out of 77) versus LSG 17.9% (19 out of 106); RR 0.10, (95% CI 0.02–0.53), p=0.006, I²=0%

Singh et al. (2020)

ESG

Mean %TWL

6 months: 15.34 (95% CI 14.33 to 16.35, I²=92.23, 9 studies)

12 months: 17.51 (95 % CI 16.44 to 18.58, I²=88.35, p=0.004, 9 studies)

18 to 24 months: 17.85 (95% CI 15.85 to 19.86, I²=69.57, p=0.025, 4 studies)

Mean %EWL

6 months: 55.61 (95% CI 50.28 to 60.95, I²=83.38, 6 studies)

12 months: 60.51 (95% CI 54.39 to 66.64, I²=66.67, p=0.22, 6 studies)

18 to 24 months: 66.77 (95% CI 57.54 to 76.00, I²=67.72, p=0.047, 4 studies)

IGB

Mean %TWL (4 RCTs and 5 observational studies)

6 months: 12.16 (95% CI 10.37 to 13.95, I²=91.32%, 9 studies)

12 months: 10.35 (95% CI 8.38 to 12.32, I²=89.80%, p=0.13, 9 studies)

18 to 24 months: 6.89 (95% CI 3.78 to 10.01, I²=96.50%, p=0.003, 3 studies)

Mean %EWL (2 RCTs and 13 observational studies)

6 months: 34.83 (95% CI 30.97 to 38.69, I² = 97.71%, 15 studies)

12 months: 29.65 (95% CI 25.40 to 33.91, I²=97.51%, p=0.10, 13 studies)

18 to 24 months: 23.88 (95% CI 17.41 to 30.33, I²=87.05%, p=0.001, 5 studies)

Comparative analysis

ESG achieved statistically significantly higher %TWL and %EWL than IGB.

The difference in mean %TWL between ESG and IGB was 3.07 at 6 months (95% CI 1.46 to 4.67, p=0.002), 7.33 at 12 months (95% CI 5.22 to 9.44, p=0.0001), and 11.51 at 18 to 24 months (95% CI 5.33 to 17.69, p=0.0003).

The difference in mean %EWL between ESG and IGB was 20.80 at 6 months (95% CI 12.50 to 29.10, p=0.0001), 30.99 at 12 months (95% CI 22.81 to 39.16, p=0.0001), and 43.78 at 18 to 24 months (95% CI 35.98 to 51.58, p=0.0001).

ESG

Overall adverse events 1.52%

Severe abdominal pain (2.2%)

Mild to moderate abdominal pain (50.65%)

Nausea, managed with medications (32.31%)

Gastrointestinal bleeding (0.61%) Perigastric fluid collection (0.45%)

Perforation (0.10%)

Post-procedure fever (0.25%)

Pulmonary embolism and DVT (0.10%)

Reversal of ESG (because of persistent symptoms; 0.15%, n=3).

Mortality=0

IGB

Overall adverse events in 4% of people.

Abdominal pain (32.51%)

Nausea (55.09%)

Balloon hyperinflation (0.03%)

Balloon resting in antrum (0.10%)

Severe dehydration (0.77%)

Esophagitis (2.33%)

Gastrointestinal bleeding (0.21%)

Obstruction (0.10%)

Perforation (0.10%)

Ulcers (0.24%)

Severe GERD (0.17%)

Early removal of IGB because of intolerance (6%)

Mortality (3 people; 2 due to acute gastric perforation, 1 due to cardiac arrest at 4 weeks).

Bhandari et al. (2023)

Mean %TWL

6 months: 12.6% (95% CI 9.28 to 19.06, p<0.001) with 94% (575 out of 612) of people achieving 5% or more TWL

1 year: 21.20% (95% CI 20.81 to 21.59, p<0.001) with 98% (558) of people achieving 5% or more TWL

2 years: 20.05% (95% CI 19.61 to 20.48, p<0.001) with 93% (513) of people achieving 5% or more TWL

3 years: 18.74% (95% CI 18.31 to 19.12, p<0.001) with 91% (425) of people maintaining 5% or more TWL

4 years: 18.19% (95% CI 17.72 to 18.57, p<0.001) with 90% (229) of people maintaining 5% or more TWL

People's %TWL at their nadir weight after ESG had a mean of 18.9% (95% CI 18.5 to 19.3). People's mean weight gain after nadir was 3.5 kg until the end of the follow-up period (95% CI 3.1 to 3.3)

Mean %EWL

1-year 56.9% (95% CI 56.51 to 57.30, p<0.001) with 90% (513) of people achieved 25% or more EWL

2 years, 54.4% (95% CI 54.03 to 54.80, p<0.001) with 442 and 80% of people maintained 25% or more EWL

3 years, 50.1% (95% CI 49.71 to 50.08, p<0.001) with 327 and 70% (327) of people maintained 25% or more EWL

4 years, 49.3% (95% CI 48.91 to 49.68, p<0.001) with 70% (177) of people maintained 25% or more EWL

People's %EWL at their nadir weight after ESG had a mean of 57.2% (95% CI 56.8 to 57.5)

Mean duration of surgery was 61.96 minutes (SD 2.1; range 45.2 to 121.1) and hospital length of stay was 3 days (range 2 to 4 days)

Resolution or improvement of comorbidities within 90 days

Diabetes 51.2% (121 out of 236)

Hypertension 65.8% (216 out of 328)

Dyslipidaemia 73.6% (302 out of 410)

Obstructive sleep apnoea 89.9% (401 out of 446)

Adverse events % (n)

Post-operative complications included: Nausea 35.45% (217)

Vomiting 17.6% (108)

Bloating 12.25% (75)

Abdominal pain 46.6% (284) Generalised weakness 2.6% (16)

Revision or redo surgery

0.3% (2) of people underwent revision to SG after 12 months of primary ESG, due to weight regain.

0.4% (3) of people had reversal redo-ESG

Sharaiha 2021

Mean %TWL

1 year: 15.6% (95% CI 14.1 to 17.1; p<0.001)

3 years: 14.9% (95% CI 12.1 to 17.7; p<0.001)

5 years: 15.9% (95% CI 11.7 to 20.5; p<0.001)

At nadir weight: 16.7% (95% CI 15.6 to 17.7, p<0.0001)

Percentage of people achieving 5% or more TWL

1 year: 89% (n=118)

3 years: 85% (n=50)

5 years: 90% (n=28)

At nadir weight: 96% (207 out of 216)

Percentage of people achieving 10% or more TWL

1 year: 77% (n=103)

3 years: 63% (n=37)

5 years: 61% (n=19)

At nadir weight: 80% (172 out of 216)

Mean weight gain after ESG

2.9 kg from nadir until the end of the follow-up period (95% CI 2.3 to 3.7)

Mean EWL

1 year: 47.9% (95% CI 42.4 to 53.3; p<0.001)

3 years: 45.1% (95% CI 34.9 to 55.2; p<0.001)

5 years: 45.3% (95% CI 32.9 to 57.7; p<0.001)

At nadir weight: 53.5% (95% CI, 49.1 to 57.9, p value not reported).

Percentage of people achieving 25% EWL

1 year: 80%

3 years: 68%

5 years: 74%

Adjunct pharmacotherapy (in 58 cases with TWL of 5% or more and weight regain at a median 5 months)

Adjunct pharmacotherapy was not associated with a significant difference in mean TWL compared with people who did not have it. Weight stabilised but no additional weight loss noted.

Repeat ESG in those with weight regain (n=13 with mean 2.6 kg weight regain)

Average TBWL of 21.5% before second ESG was stabilized at 24.2% at 1 year after the second ESG.

Predictors of TWL

The amount of weight loss at 1 month after the procedure, patient's compliance with follow up, and endoscopist's experience are independent predictors of weight loss.

Mild adverse events in 32% of people.

Heartburn (up to 3 weeks) 25% (n=54)

Nausea or vomiting (managed with medications) 25% (n=43)

Epigastric pain (beyond 24 hours, managed with medications) 31% (n=65)

Constipation (managed with laxatives) 29% (n=63)

Superficial oesophageal tear (from the over tube with the device, managed endoscopically) n=1

Asymmetric paraesthesia n=1

Spinal white-matter plaques and low serum thiamine (vitamin B1) n=1

Moderate adverse events 1.3% (n=3)

Pain, n=1 (in left upper quadrant 18 months after ESG, scan showed sutures with bridging fibrosis bands. ESG sutures were released leading to increased gastric volume and improvement in pain)

Perigastric leak n=2

(after dietary indiscretion, managed with antibiotics and percutaneous drainage in 1)

Bariatric surgery or SG (for inadequate weight loss) 1% (n=2).

Revision procedures

Repeat ESG (for mean 2.6 kg weight gain) 6% (13 out of 216).

Average TWL after 1 year was 24.2%.

Laparoscopic SG (due to inadequate weight loss after ESG) 1% (n=2)

Gudur 2023

Mean% TWL within 30 days follow up (propensity score matched cohort)

ESG 4.0% (SD 6.7) versus SG 5.4% (SD 4.3); p<0.001.

Mean change in BMI from pre-operative to post-operative

ESG –1.77 kg/m² (SD 2.89) versus SG –2.36 kg/m² (SD 1.78); p< 0.001

Procedure time

ESG 62.9 minutes versus SG 72.4 minutes, p<0.001.

Length of stay

ESG 0.87 days versus SG 1.45 days, p<0.001.

Overall adverse events

SG 1.1% (340 out of 30,270) versus ESG 1.4% (86 out of 6,053); p=0.058

Factors impacting adverse events

Propensity score matched analysis demonstrated that black ethnicity associated with a higher risk of adverse events in SG compared with ESG (OR, 1.23; 95% CI, 1.13 to 1.35). Multivariate regression noted that people with a higher BMI were less likely to have an adverse event after ESG.

In the ESG cohort, albumin, BMI, renal insufficiency, age, therapeutic anticoagulation, non-insulin-dependent diabetes, chronic steroid use, and female gender were statistically significant patient factors. These variables were also associated with adverse events in the SG cohort. Additional variables associated with adverse events in the SG cohort were GERD, insulin-dependent diabetes, previous surgery, hypertension, black ethnicity, history of pulmonary embolism, previous cardiac surgery, independent preoperative functional status, sleep apnoea, smoking, history of myocardial infarction, chronic obstructive pulmonary disease, dialysis, and hyperlipidaemia

Readmission within 30 days

ESG 3.8% (231 out of 6,053) versus SG 2.6% (794 out of 30,270); p<0.001

Causes of readmission for ESG

Nausea, vomiting, fluid or electrolyte disturbance 26% (n=72)

Abdominal pain 17% (n=47)

Gastrointestinal leak 6% (n=16)

Gastrointestinal bleeding 5% (n=15)

Causes of readmission for SG

Nausea, vomiting, fluid or electrolyte disturbance 31% (n=5,686)

Abdominal pain 12% (n=2,097)

Gastrointestinal leak 6% (n=1,063)

Reoperation within 30 days

ESG 1.4% (86 out of 6,053) versus SG 0.8% (238 out of 30,270), p<0.001

Main causes of reoperation for ESG Abdominal re-exploration 13% (n=10),

Obstruction or GI perforation or pain 10%

Main causes of reoperation for SG

Abdominal re-exploration 27% (n=1,315)

Gastrointestinal bleeding 25% (n=1,217)

Staple line leak 15% (n=1,217)

Reintervention within 30 days

ESG 2.8% (171 out of 6,053) versus SG 0.7% (209 out of 30,270); p<0.001.

Common endoscopic interventions after ESG

Therapeutic endoscopy (was mainly for stent placement or dilatation) ESG 48% (n=52) versus SG 32% (n=1,714)

Treatment for dehydration

ESG 2.4% (147 out of 6,053) versus 3.3% (993 out of 30,270); p=0.001

Emergency visit with no admission

ESG 4.9% (294 out of 6,053) versus 5.9% (1,786 out of 30,270); p=0.002.

Procedure technique

ESG procedures were done with an endoscopic suturing system (OverStitch) and authors of the meta-analyses reported that the procedure technique varied across included primary studies. The differences reported were the variable number of sutures used, the suturing patterns ('Z', 'U,' or triangular, rectangular or square patterns, in running or interrupted fashion) and a layer of reinforcement sutures.

Efficacy

Weight loss

Percentage total weight loss

ESG alone

In a systematic review and meta-analysis of 11 studies (with 2,170 people who had ESG), the pooled mean %TWL at 1 month was 8.56 (95% CI 7.94 to 9.18, I²=0.3%, 5 studies), at 3 months was 11.65 (95% CI 10.76 to 12.53, I²=0%, 5 studies), at 6 months was 15.32 (95% CI 14.54 to 16.10, I²=15.3%, 9 studies), at 9 months was 16.15 (95% CI 14.94 to 17.37, I²=0%, 3 studies), at 12 months was 17.33 (95% CI 16.30 to 18.36, I²=10.8%, 9 studies), and at 18 months was 16.80 (95% CI 13.02 to 20.56, I²=0%, 2 studies). The certainty of evidence was low (de Miranda Neto 2020).

In a systematic review and meta-analysis of 28 studies on endobariatric therapies with a follow up of at least 12 months, meta-analysis of 9 studies on ESG alone (with a mean BMI of 36.1 kg/m²) reported that the pooled mean %TWL at 6 months, 12 months and between 18 months and 24 months follow-up was 15.34, 17.51 and 17.85. Pooled analysis of IGB alone studies showed that mean %TWL at 12 months was 10.35. This was significantly decreased at between 18 months and 24 months (6.89) indicating weight regain after IGB removal. An indirect comparison of ESG to IGB (non-comparative studies) showed that ESG achieved significantly superior weight loss (difference in mean %TWL was 3.07 at 6 months, 7.33 at 12 months [p=0.0001], and 11.51 at between 18 months and 24 months; Singh 2020).

In a prospective cohort study of 612 people who had ESG for treatment of obesity, the mean %TWL was 18.19 (95% CI 17.72 to 18.57) with 90% of participants maintaining a percentage of total weight loss of 5% or more (Bhandari 2023).

In a prospective cohort study of 216 people with mean BMI of 39.6 kg/m², who had ESG, mean %TWL was 15.6 (95% CI 14.1 to 17.1; p<0.001) at 1 year with 89% and 77% of people achieving 5% or more and 10% or more TWL, respectively. At 3 years mean TWL was 14.9% (95% CI 12.1 to 17.7; p<0.001) and 85% and 63% of people maintained 5% or more and 10% or more TWL, respectively. At 5 years mean TWL was 15.9% (95% CI 11.7 to 20.5; p<0.001) and 90% and 61% of people maintained 5% or more and 10% or more TWL, respectively. Mean %TWL at nadir weight was 16.7 (95% CI 15.6 to 17.7, p<0.0001; Sharaiha 2023).

ESG plus lifestyle modification versus lifestyle modification alone

In a multicentre RCT (MERIT trial) of 187 people with class 1 or 2 obesity comparing ESG plus intensive lifestyle modification (n=77) with intensive lifestyle modification alone (n=110) at 52 weeks, the %TWL was 13.6 for the ESG group and 0.8 for the control group (p<0.0001; Abu Dayyeh 2022).

ESG versus LSG

A meta-analysis of 7 studies (6,775 people) that directly compared ESG (n=3,413) with LSG (n=3,362) reported that there were significant differences in %TWL at 6 months ( MD –7.48; 95% CI –10.44 to –4.52; p<0.00001), 12 months (MD –9.90; 95% CI –10.59 to –9.22; p<0.00001), and 24 months (MD –7.63; 95% CI –11.31 to –3.94; p<0.0001) showing superiority of LSG over ESG (Beran 2022).

ESG versus SG

In a propensity score matched analysis of 6,054 people who had ESG and 30,270 people who had SG, mean %TWL was higher in the SG group compared with ESG group (5.4 versus 4.0, p<0.001). A greater mean reduction in BMI within the initial 30 days after the procedure was also reported in the SG group compared with ESG group, (–2.36 versus –1.77, p<0.001) (Gudur 2023).

Percentage excess weight loss

ESG alone

In the systematic review and meta-analysis of 11 studies (with 2,170 people who had ESG), the pooled mean %EWL at 1 month was 31.08 (95% CI 20.79 to 41.36, I²=0%, 3 studies), at 3 months was 46.13 (95% CI 38.79 to 53.47, I²=0%, 3 studies), at 6 months was 55.80 (95% CI 50.61 to 60.99, I²=15.09%, 6 studies), at 9 months was 66.20 (95% CI 57.54 to 74.86, I²=8.52%, 3 studies), at 12 months was 60.07 (95% CI 53.39 to 66.74, I²=18.09%, 6 studies), and at 18 months was 73.04 (95% CI 58.94 to 87.14, I²=0%, 2 studies). The certainty of evidence was low (de Miranda Neto 2020).

In a prospective cohort study of 612 people who had ESG for treatment of obesity (mean BMI 34.30 kg/m²), mean %EWL was 49.30 (95% CI 48.91 to 49.68) with 70% of people maintaining an EWL of 25% or more at 4 years (Bhandari 2023).

In the prospective study of 216 people who had ESG, at 1 year mean EWL was 47.9% (95% CI 42.4 to 53.3; p<0.001) with 80% of people achieving 25% EWL. At 3 years, mean EWL was 45.1% (95% CI 34.9 to 55.2; p<0.001) and 68% of people maintained 25% EWL. At 5 years, mean EWL was 45.3% (95% CI 32.9 to 57.7; p<0.001) and 74% of people maintained 25% EWL. Mean EWL at nadir weight was 53.5% (95% CI 49.1 to 57.9, p value not reported; Sharaiha 2023).

ESG plus lifestyle modification versus lifestyle modification alone

In the multicentre RCT (MERIT trial) of 187 people with class 1 or 2 obesity comparing ESG plus intensive lifestyle modification (n=77) with intensive lifestyle modification alone (n=110) at 52 weeks, the mean %EWL (primary endpoint) was 49.2 for the ESG group and 3.2 for the control group (p<0.0001). Additionally, 77% (59 out of 77) of people in the ESG group achieved 25% or more %EWL at 52 weeks compared with 12% (13 out of 110) in the control group (p<0.0001). The crossover ESG group (n=72) achieved a mean 44.1% EWL at 52 weeks from crossover. The mean %EWL of all participants who underwent the ESG procedure (both primary ESG and crossovers) at week 52 post procedure was 46.7%. A %EWL of 25% or more was maintained in 68% (41 out of 60) people in the ESG group at 104 weeks. In the primary ESG group, 81% (55 out of 68) of people reached the primary endpoint (25% or more EWL) at 52 weeks, compared with 72% (46 out of 64) in the crossover ESG group with a similar follow-up period (p=0.21, no statistically significant difference; Abu Dayyeh 2022).

ESG versus LSG

In a systematic review and meta-analysis of 16 studies comparing ESG with LSG, the mean %EWL was 80.3 (95% CI 68.1 to 92.5; p=0.001; I²=98.9%, Tau²=56.62) for the LSG group and 62.2 ( CI 57.8 to 66.6; p=0.005; I²=65.52, Tau²=24.68) for the ESG group, corresponding to an absolute difference of 18.1% (p=0.0001; Marincola 2021).

A meta-analysis of 7 studies (with 6,775 people) that directly compared ESG (n=3,413) with LSG (n=3,362) reported that there significant differences in EWL% at 6 months (MD –10.23; 95% CI –11.90 to –8.56; p<0.00001; I²=0%, 3 studies) and at 12 months (MD –23.99; 95% CI –25.30 to –22.68; p<0.00001; I²=0%, 2 studies) showing superiority of LSG over ESG (Beran 2022).

ESG versus IGB

In the systematic review of 28 studies on endobariatric therapies with at least 12 months follow up, overall pooled mean %EWL after ESG alone (in 9 studies) was 55.6 at 6 months, 60.5 at 12 months, and 66.8 between 18 months to 24 months. Weight loss slightly increased at 12 months and 24 months compared with 6 months. Pooled analysis of IGB alone studies showed that mean %EWL was 34.8 at 6 months (15 studies) and 29.7 at 12 months (13 studies). This significantly decreased at 18 months or 24 months (%EWL 23.9, p=0.001, 5 studies) indicating weight regain after IGB removal. An indirect comparison of ESG to IGB, showed that ESG achieved significantly superior weight loss (the difference in mean %EWL at 6 months, 12 months, and 18 months to 24 months was 20.8, 31.0, and 43.8, respectively; Singh 2020).

Percentage absolute weight loss

ESG alone

In the systematic review and meta-analysis of 11 studies (with 2,170 people who had ESG), the pooled mean AWL at 1 month was 7.73 kg (95% CI 7.06 to 8.40, I²=16.82%, 3 studies), at 3 months was 10.23 kg (95% CI 8.44 to 12.03, I²=0%, 3 studies), at 6 months was 14.88 kg (95% CI 13.33 to 16.42, I²=0%, 6 studies), at 9 months was 15.44 kg (95% CI 12.70 to 18.17, I²=0%, 2 studies), at 12 months was 17.32 kg (95% CI 15.65 to 18.99, I²=0%, 7 studies), and at 18 months was 15.95 kg (95% CI 10.95 to 20.95, I²=0%, 2 studies). The certainty of evidence was low (de Miranda Neto 2020).

Improvement in comorbidities

ESG alone

In a prospective cohort study of 612 people who had ESG for treatment of obesity, resolution or improvement of comorbidities was reported in 51% cases of diabetes, 66% cases of hypertension, 74% cases of dyslipidaemia and 90% of people with obstructive sleep apnoea (Bhandari 2023).

ESG plus lifestyle modification versus lifestyle modification alone

In the multicentre RCT (MERIT trial) of 187 people with class 1 or 2 obesity comparing ESG plus intensive lifestyle modification (n=77) with intensive lifestyle modification alone (n=110) at 52 weeks, 80% (41 out of 51) of people in the ESG group had improvement in 1 or more metabolic comorbidities, while 12% (6 out of 51) worsened. 45% (28 out of 62) of people in the control group showed improvement, while 50% (31 out of 62) worsened (Abu Dayyeh 2022).

ESG versus LSG

A meta-analysis of 7 studies (with 6,775 people) that directly compared ESG (n=3,413) with LSG (n=3,362) reported that the improvement or remission of diabetes mellitus was significantly higher with LSG compared with ESG (82% versus 64% respectively; RR 0.78, 95% CI 0.68 to 0.91, p=0.001, I²=0%). The improvement or remission of hypertension was similar between the ESG and LSG groups (51% versus 46% respectively; RR 1.12, 95% CI 0.86 to 1.47, p=0.39, I²=0%; Beran 2022).

Safety

Adverse events

ESG alone

In a systematic review and meta-analysis of 11 studies, the overall pooled rate of adverse events was 2.3% (95% CI 1.2 to 4.1, I²=24%, 7 studies). Of these, a rate of 1.5% (95% CI 0.5 to 4.3, I²=0%, 2 studies) for mild, 1.7% (95% CI 0.9 to 3.1, I²=8.16%, 6 studies) for moderate, and 0.8% (95% CI 0.3 to 2.0, I²=0%, 3 studies) for severe adverse events was observed. No procedure-related mortality was reported in any of the included studies. The most common major adverse events were gastrointestinal bleeding (n=13) and perigastric fluid collection (n=10). Most events were managed conservatively, but 2 of the gastrointestinal bleeding cases needed sclerotherapy, and 3 of the cases with perigastric fluid collection needed surgical interventions. These included 1 person who developed a gastric fistula, which needed closure and reversal of the ESG. Other adverse events included severe abdominal pain (n=8), fever (n=5), deep vein thrombosis (n=1) and pneumothorax (n=1) (de Miranda Neto 2020).

The prospective cohort study of 612 people reported post-operative complications such as nausea in 36% (217) of people, vomiting in 18% (108), bloating in 12% (75), abdominal pain in 47% (284) and generalised weakness in 3% (16) of people (Bhandari 2023).

The prospective study of 216 people reported moderate adverse event rate of 1% (n=3). These events included pain at sutures after 18 months (managed by releasing suture lines to increase gastric volume); 2 cases of perigastric leak (after dietary indiscretion, managed with antibiotics, and percutaneous drainage in 1 case). Minor adverse events reported include nausea or vomiting (managed with medications) in 25% (n=43), epigastric pain (beyond 24 hours, managed with medicines) in 31% (n=65), constipation (managed with laxatives) in 29% (n=63), superficial oesophageal tear in 1 (from the device; managed endoscopically), asymmetric paraesthesia in 1, and thiamine deficiency in 1. SG (for inadequate weight loss) was reported in 1% (n=2) of people (Sharaiha 2023).

ESG plus lifestyle modification versus lifestyle modification alone

A multicentre RCT (MERIT trial) of 187 people with class 1 or 2 obesity comparing ESG plus intensive lifestyle modification (n=77) with intensive lifestyle modification alone (n=110) and a crossover to ESG at 12 months for those people not losing 25% of EBW (n=72) reported serious adverse events in 2% (3 out of 131) of people at 104 weeks follow up. These include abdominal abscess, managed endoscopically; upper gastrointestinal bleed, managed conservatively without transfusion; and a case of malnutrition requiring endoscopic reversal of the ESG. 927 events were reported in 93% (138 out of 150) of people in the primary and crossover ESG group. 66% (612 out of 927) of reported adverse events were gastrointestinal symptoms, including pain, heartburn, nausea, and vomiting. Most of these symptoms resolved within 1 week (Abu Dayyeh 2022).

ESG versus LSG

In a systematic review and meta-analysis of 16 studies comparing ESG with LSG, the pooled mean periprocedural complication rate in the ESG group was 0.15% (Cochran's Q test p=0.0001), with a moderate grade of heterogeneity (I²=42.81). The pooled mean periprocedural complication rate in the LSG group was 0.30% (Cochran's Q test p=0.0001), with a moderate grade of heterogeneity (I²=62.26). The difference in mean rate of adverse events was 0.19 % (between study heterogeneity χ²=1.602; p=0.2056; Marincola 2021).

In the meta-analysis of 7 studies (with 6,775 people) that directly compared ESG (n=3,413) with LSG (n=3,362), there was a lower rate of adverse events with ESG compared with LSG, but this was not statistically significant (RR 0.51, 95% CI 0.23 to 1.11, p=0.09). The rate of new-onset GERD was significantly lower after ESG compared with LSG, 1.3% versus 17.9% respectively (RR 0.10, 95% CI 0.02 to 0.53, p=0.006; Beran 2022).

ESG versus IGB

In a systematic review of 28 studies on endobariatric therapies with at least 12 months of follow up, 9 studies on ESG reported severe abdominal pain in 2.2% of people, mild to moderate abdominal pain in 51% of people, and nausea in 32% of people. Reversal of ESG because of persistent symptoms was needed in 0.15% (3) of people. Serious adverse events included gastrointestinal bleeding (0.61%), perigastric fluid collection (0.45%), perforation (0.10%), post-procedure fever (0.25%), and pulmonary embolism and DVT (0.10%). Overall, these adverse events were seen in 1.5% of people. No deaths associated with the procedure were reported (Singh 2020).

In studies with IGB, adverse events were reported in 4% of people. These include abdominal pain (32.5%), nausea (55.1%), balloon hyperinflation (0.03%), balloon resting in antrum (0.10%), severe dehydration (0.77%), esophagitis (2.3%), gastrointestinal bleeding (0.2%), obstruction (0.1%), perforation (0.1%), ulcers (0.2%), and severe GERD (0.2%). Early removal of IGB because of intolerance was reported in 6% of people. Mortality was reported in 3 people. Two deaths were due to acute gastric perforation, and 1 due to cardiac arrest at 4 weeks (Singh 2020).

ESG versus SG

A propensity score matched analysis of 6,054 people who had ESG and 30,270 people who had SG reported that adverse events were low and there was no significant difference in major adverse events between the groups within 30 days follow up (1.4% versus 1.1%, p=0.058). People in the ESG group had more readmissions (3.8% versus 2.6%), reoperations (1.4% versus 0.8%), and reinterventions (2.8% versus 0.7%) within 30 days compared with the SG group (p<0.001). In the ESG group, readmission was mainly because of nausea, vomiting, fluid or electrolyte disturbance (26%), abdominal pain (17%), gastrointestinal leak (6%), or gastrointestinal bleeding (5%). In the SG group, readmission was due to nausea, vomiting, fluid or electrolyte disturbance (31%), abdominal pain (12%), or gastrointestinal leak (6%). The most commonly performed reoperation after both ESG and SG was abdominal re-exploration. The most common reason for reoperation after ESG was suspicion for obstruction or gastrointestinal perforation, whereas gastrointestinal bleeding and staple line leak were the most common reasons after SG. Therapeutic endoscopy was the most common reintervention after ESG (48%) and SG (32%), with stent placement or dilatation representing the most common endoscopic interventions. Black ethnicity was associated with a higher risk of adverse events in SG compared with ESG (OR 1.23; 95% CI 1.13 to 1.35; Gudur 2023).

Anecdotal and theoretical adverse events

Expert advice was sought from consultants who have been nominated or ratified by their professional society or Royal College. They were asked if they knew of any other adverse events for this procedure that they had heard about (anecdotal), which were not reported in the literature. They were also asked if they thought there were other adverse events that might occur, even if they had never happened (theoretical).

They listed the following anecdotal adverse events:

gallbladder problems (full thickness suture placed into gallbladder) requiring successful cholecystectomy
abdominal abscess.

They listed future bariatric surgery being more complex as a theoretical adverse event.

Six professional expert questionnaires for this procedure were submitted. Find full details of what the professional experts said about the procedure in the specialist advice questionnaires for this procedure.

Validity and generalisability

In this overview we only considered studies on ESG (in which stomach capacity is reduced by making full-thickness sutures along the greater curvature using the OverStitch endoscopic suturing system) as a primary obesity procedure.
All other ESG techniques (Primary obesity surgery endoluminal [POSE™],Endoscopic sutured gastroplasty [E-ESG, Endomina system], Endoluminal vertical gastroplasty [EVG], Transoral gastric volume reduction [TGVR- sequel of EVG], Transoral vertical gastroplasty [TOGA], Articulating circular endoscopic stapler [ACE], Endozip) are not considered in this evidence summary as 2 of these procedures (POSE and E-ESG) are not undertaken in the NHS and the remaining systems are not currently available for commercial use in the UK.
Evidence on ESG performed after other treatments is also not considered in this overview.
Most of the studies included people with obesity (a BMI between 30 to 39.9 kg/m²), those with a history of failure with non-surgical weight loss treatments, those refused to undergo bariatric surgery or not eligible or considered to be at high risk of surgery. Therefore, the indication in the title has been amended to 'obesity'.
Studies were heterogenous, in terms of study designs, sample sizes, procedure technique, length of follow-up, and outcome measures.
Postoperative rehabilitation (including guidance on water or food intake and exercise) varied between the studies.
The RCT comparing ESG plus lifestyle modification with lifestyle modification alone was funded by the device company presenting some possibility of bias.
Two cohort studies reported follow-up from 4 to 5 years.
The technique of ESG is evolving. One small retrospective study (Glaysher 2019) done in the UK reported a modified gastroplasty suture pattern (longitudinal compression sutures). This has been added to the appendix.
Ongoing trials:
1. NCT04060368: Efficacy and safety of endoscopic sleeve gastroplasty (using OverStitch Endoscopic Suture System) versus laparoscopic SG in obese subjects with NASH (Non-Alcoholic Steatohepatitis). RCT, n=30; ESG plus lifestyle modifications versus LSG plus lifestyle modifications; primary outcome: proportion of subjects undergoing ESG relative to LSG achieving resolution of NASH without worsening of fibrosis, Proportion of subjects undergoing ESG relative to LSG with cardiovascular and liver-related death events; study location: Spain; study completion June 2023.
2. NCT03705416: Long term outcomes of bariatric patients treated with surgery (surgical reduction or bypass) or endoscopy (ESG). Observational prospective cohort study, n=250, ESG versus surgery (vertical SG or a Roux-en-Y gastric bypass); Primary outcome: percentage of participants with GERD based on symptoms, and abnormal acid exposure time and/or reflux esophagitis; location USA; study completion March 2028.
3. Jamie Kelly, Vinod Menon, Frank O'Neill et al. UK cost-effectiveness analysis of endoscopic sleeve gastroplasty versus lifestyle modification alone for adults with class 2 obesity (BMI 35.0-39.9 kg/m²) based on results from the MERIT RCT (pre-print) https://doi.org/10.21203/rs.3.rs-2616584/v1
A UK cost-effectiveness study sponsored by Apollo Endosurgery and under review as a full manuscript at the International Journal of Obesity. According to the authors, the cost-utility analysis was done in line with the NICE reference case and methodologies and is the first cost-effectiveness of ESG versus LM alone in adults with class 2 obesity (BMI 35.0-39.9 kg/m²) from a national healthcare system perspective in England. The model was informed by patient level data from the MERIT study and demonstrates that ESG (in addition to lifestyle modification) is highly cost-effective compared with lifestyle modification alone. According to the authors, the base-case ICER was £2,453/QALY and ESG remained cost-effective at the NICE willingness to pay threshold in all sensitivity analyses.
1. Wilson E, O'Neill F et al. Short term cost savings with endoscopic sleeve gastroplasty; a 30 day US cost consequence analysis (presentation at DDW 2023 conference)
A 30-day comparison of LSG and ESG from the US perspective. According to the authors, this demonstrated significant cost-savings with ESG. Savings were driven by a reduced OR time ($871) and length of stay ($2,776) between the procedures. After 30 days there was an increased incidence of GERD (6.6%) in the LSG compared with ESG (0.4)(p<0.01). All other adverse events contributed to a cost saving with ESG after 30 days.

How are you taking part in this consultation?

Interventional procedure overview of endoscopic sleeve gastroplasty for obesity