Journal of Clinical Medicine Research, ISSN 1918-3003 print, 1918-3011 online, Open Access
Article copyright, the authors; Journal compilation copyright, J Clin Med Res and Elmer Press Inc
Journal website https://jocmr.elmerjournals.com

Original Article

Volume 18, Number 8, August 2026, pages 542-555

Endoscopic Versus Surgical Bariatric Interventions for Obesity: A Systematic Review and Network Meta-Analysis

Figures

↓  Figure 1. PRISMA 2020 flow diagram. PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses.
Figure 1.
↓  Figure 2. Risk of bias assessment using the ROBINS-I tool for the included nonrandomized studies (traffic-light plot of individual study domains). ROBINS-I: Risk of Bias in Nonrandomized Studies of Interventions.
Figure 2.
↓  Figure 3. Risk of bias assessment using a weighted summary bar plot across all domains.
Figure 3.
↓  Figure 4. Pairwise meta-analysis forest plot showing the mean difference (MD) in percentage of total body weight loss (%TBWL) at 12 months (ESG vs. LSG). CI: confidence interval; ESG: endoscopic sleeve gastroplasty; LSG: laparoscopic sleeve gastrectomy; SD: standard deviation.
Figure 4.
↓  Figure 5. Pairwise meta-analysis forest plot showing the risk ratio (RR) of serious adverse events (SAEs, ESG vs. LSG). CI: confidence interval; ESG: endoscopic sleeve gastroplasty; LSG: laparoscopic sleeve gastrectomy.
Figure 5.
↓  Figure 6. Network geometry graph of bariatric interventions. Node size is proportional to the total number of patients, and edge thickness represents the number of direct comparative studies.
Figure 6.
↓  Figure 7. Network meta-analysis forest plot. The network estimates (MD; 95% CI) compared to LSG were as follows: LGCP (–2.18%; –4.93 to 0.56, P = 0.118), ESG (–11.33%; –13.15 to –9.51, P < 0.0001), LAGB (–15.81%; –18.76 to –12.87, P < 0.0001), HIDLT (–17.63%; –21.54 to –13.73, P < 0.0001), and IGB (–19.00%; –23.37 to –14.63, P < 0.0001). ESG: endoscopic sleeve gastroplasty; HIDLT: high-intensity diet and lifestyle therapy; IGB: intragastric balloon; LAGB: laparoscopic adjustable gastric banding; LGCP: laparoscopic greater curve plication; LSG: laparoscopic sleeve gastrectomy; %TBWL: percentage of total body weight loss; MD: mean difference; CI: confidence interval.
Figure 7.
↓  Figure 8. Treatment ranking heatmap (P-scores/SUCRA). A higher score (closer to 1.0) indicates a greater probability of the intervention yielding the highest %TBWL. ESG: endoscopic sleeve gastroplasty; HIDLT: high-intensity diet and lifestyle therapy; IGB: intragastric balloon; LAGB: laparoscopic adjustable gastric banding; LGCP: laparoscopic greater curve plication; LSG: laparoscopic sleeve gastrectomy; %TBWL: percentage of total body weight loss.
Figure 8.
↓  Figure 9. Node-splitting analysis forest plot comparing direct, indirect, and network estimates to assess local inconsistency within the network. ESG: endoscopic sleeve gastroplasty; LAGB: laparoscopic adjustable gastric banding; LGCP: laparoscopic greater curve plication; LSG: laparoscopic sleeve gastrectomy; MD: mean difference; CI: confidence interval.
Figure 9.
↓  Figure 10. Comparison-adjusted funnel plot assessing publication bias and small-study effects across the network. ESG: endoscopic sleeve gastroplasty; HIDLT: high-intensity diet and lifestyle therapy; IGB: intragastric balloon; LAGB: laparoscopic adjustable gastric banding; LGCP: laparoscopic greater curve plication; LSG: laparoscopic sleeve gastrectomy.
Figure 10.

Tables

↓  Table 1. Study Characteristics
 
StudyStudy designTreatment armsNumber
ESG: Endoscopic sleeve gastroplasty; LSG: Laparoscopic sleeve gastrectomy; EIB: endoscopic intragastric balloon; LGCP: Laparoscopic greater curve plication; LAGB: Laparoscopic adjustable gastric banding; EBT: endoscopic bariatric therapy; LBT: laparoscopic bariatric therapy; HIDLT: High-intensity diet and lifestyle therapy; BIB: BioEnterics intragastric balloon; MBSAQIP: Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program.
Gudur et al, 2023 [25]Retrospective database (MBSAQIP)ESG6,054
LSG597,463
Carr et al, 2022 [26]Prospective cohortESG16
LSG45
Joseph et al, 2024 [27]Retrospective cohortESG68
LSG212
Dankar et al, 2025 [28]Retrospective cohortESG70
LSG70
Alqahtani et al, 2022 [29]Propensity-matched cohortESG3,018
LSG3,018
Fayad et al, 2019 [30]Case-matched cohortESG54
LSG83
Tammo et al, 2025 [31]Retrospective cohortEIB36
LSG36
Lopez-Nava et al, 2021 [32]Retrospective cohortESG199
LSG61
LGCP36
Novikov et al, 2018 [33]Retrospective cohortESG91
LSG120
LAGB67
Marshall et al, 2022 [34]Matched cohortESG25
LSG25
Fiorillo et al, 2020 [35]Propensity-matched cohortESG23
LSG23
Mocanu et al, 2025 [36]Retrospective database (MBSAQIP)ESG2,285
LSG504,312
Fayad et al, 2022 [37]Propensity-matched databaseEBT211
LBT9,059
Osman et al, 2025 [38]Propensity-matched cohort (TriNetX)ESG62
LSG62
Ebadinejad et al, 2025 [39]Retrospective database (MBSAQIP)ESG2,171
LSG7,997
Milone et al, 2005 [40]Retrospective cohortBIB57
LSG20
Cheskin et al, 2020 [41]Case-matched cohortESG105
HIDLT281

 

↓  Table 2. GRADE Summary of Findings
 
OutcomeStudies (n)Participants (n)Effect estimateRisk of biasInconsistencyIndirectnessImprecisionPublication biasFinal certainty
GRADE: Grading of Recommendations Assessment, Development, and Evaluation; MD: mean difference; RR: risk ratio; %TBWL: percentage of total body weight loss; CI: confidence interval; GERD: gastroesophageal reflux disease; T2DM: type 2 diabetes mellitus; ESG: endoscopic sleeve gastroplasty; LSG: laparoscopic sleeve gastrectomy.
%TBWL at 12 months (weight-loss efficacy)8 cohortsAbout 620,000MD = –10.5 %TBWL (95% CI, –14.1 to –7.0) ESG vs LSGSerious (–1)Serious (–1)Not seriousNot seriousUndetected⊕◯◯◯ Very low
Serious adverse events (Clavien-Dindo ≥ III)5 (including registries)About 620,000RR = 1.30 (95% CI, 0.97 to 1.73) ESG vs LSGSerious (–1)Not seriousNot seriousSerious (–1)Undetected⊕◯◯◯ Very low
Comorbidity remission (T2DM, GERD)4 cohortsAbout 6,200T2DM remission: ESG 64% vs LSG 82%; New-onset GERD: ESG 4% vs LSG 20%Serious (–1)Serious (–1)Not seriousSerious (–1)Undetected⊕◯◯◯ Very low