Efficacy and safety of endoscopic gastroplasty for treatment of obesity: An overview of comparative meta-analyses

Abstract

BACKGROUND

There is a scarcity of evidence and systematic reviews on endoscopic gastroplasty (EG) compared to other management options for the treatment of obesity.

AIM

To assess the published meta-analyses through a systematic review approach and provide further insight into the current status of available evidence through a critical appraisal.

METHODS

PubMed/MEDLINE, Scopus, Embase and Cochrane Library were searched from inception to November 2022. The meta-analyses that compared the efficacy and safety of EG to other interventions were considered for this overview. The outcomes of interest were total body weight loss (TBWL), excessive weight loss, and average weight loss along with occurrence of adverse effects. Methodological quality, heterogeneity, and limitations were also reviewed.

RESULTS

A total of six meta-analyses out of 364 records were considered for this review with a major contribution from the United States. Overall methodological quality of included studies were moderate to good. EG treatments were significantly better in terms of TBWL, excessive weight loss, and average weight loss. However, there was no significant difference between endoscopic transoral outlet reduction and full-thickness suturing plus argon plasma mucosal coagulation. Lack of comparative studies and randomized trials, lack of long-term follow-up, reporting bias, selection bias, lack of control groups, and considerable level of heterogeneity were the major limitations in the available evidence.

CONCLUSION

Though EG was significantly effective for treatment of obesity, there is limited comparative evidence on this topic. High-quality well-controlled evidence is required to strengthen the current evidence base on EG treatment for obesity.

Key Words: Body weight; Endoscopy; Evidence-based practice; Obesity; Safety

Core Tip: Endoscopic gastroplasty treatments were significantly better in reducing total body weight, excessive weight, and average weight among obese population. No significant difference between endoscopic transoral outlet reduction and full-thickness suturing plus argon plasma mucosal coagulation. Lack of comparative, long-term follow-up and randomized studies, reporting and selection bias, high level of heterogeneity were the major limitations in the currently available meta-analyses.

INTRODUCTION

Obesity and excess weight are a major threat across the world that has grown to a level of epidemic proportion with a contribution of at least 2.8 million deaths annually. As per the global statistics, 13% of adults with obesity, and 39% are overweight. Among children and adolescents, one in five are overweight[1]. Numerous approaches such as lifestyle modifications, dietary interventions, physical activity, behavioral therapy, pharmacological interventions, and surgical procedures have been adopted for the treatment of obesity[2].

Bariatric surgery plays an inevitable role in this area, however the unavoidable morbidity and mortality, limited access to surgery, high surgical risk, cost, and patient preference mean that bariatric surgery is limited to a set amount of patients[3]. Endoscopic gastroplasty (EG) methods are excellent substitutes for conventional surgery for obesity, with the advantage of being suited to high-risk patients, less invasive, and more effective. Three EG techniques are widely used which include endoscopic sleeve gastroplasty (ESG), primary obesity surgery endoluminal procedure, and transoral endoscopic vertical gastroplasty [endoscopic transoral outlet reduction (TORe)][4].

Many primary research reviews have assessed the effectiveness of EG in obesity management, though the findings are inconclusive[5,6]. Meta-analysis employs statistical measures to pool the outcomes from different studies into a single estimate. It is considered to be the highest level of evidence and the findings are easily transferred to clinical practice[7,8]. There are inconclusive findings among the published meta-analyses concerning the effectiveness of EG in the treatment of obesity compared to other management options[9-11]. Hence, this study aimed to perform an overview of meta-analyses to give a clear picture of this topic and provide further insights into efficacy and safety of EG in patients with obesity. This overview also assessed the quality of included studies/risk of bias, heterogeneity, and limitations in the available meta-analyses.

MATERIALS AND METHODS

Participants, intervention, comparator, outcomes, and study design framework to perform this overview and reported the findings as per the PRISMA guidelines to report this systematic review[12]. Ethical approval was not required as this was a review of published literature evidence. The protocol was not registered with any repository; however, this review was conducted as per a pre-defined methodology.

Criteria for considering studies for this review

Study design: Only the meta-analyses with a comprehensive search strategy in at least one literature database and that were considered the primary research were included in this overview. Only studies comparing the effect of EG with other treatments written in English were considered. Systematic reviews without a proper research question, search strategy, defined process of study selection, or pooled estimations were excluded from this overview. Those meta-analyses which did not having a comparative analysis or only had a pooled estimation of single intervention were not considered. Also, research articles, narrative reviews, overviews, study protocols, and duplicates were excluded.

Participants: The studies considered adult patients (≥ 18 years old) with obesity who had undergone surgical treatment were included in this overview. Patients who had weight regain after the initial surgical management were also considered. The diagnosis of obesity and inclusion of participants was in accordance with any international guidelines or as per the author’s discretion.

Intervention: The intervention considered was any of the methods of EG for the management of obesity, excess weight, or weight regain. Studies that did not include any surgical methods other than EG were not considered for inclusion.

Comparator: Any comparator including surgical and non-surgical methods of weight loss were considered in this study. This included bariatric surgery techniques other than EG and non-surgical methods like sham, or diet and lifestyle changes. Meta-analyses that had not performed a comparative analysis were excluded from this review.

Outcomes and outcome measures: The primary outcomes were efficacy measures such as total body weight loss (TBWL), excess weight loss (EWL), absolute weight loss (AWL), or percentage (%) TBWL, EWL, or AWL measured at any time point. The secondary endpoint was safety measures like adverse events or adverse drug reactions (ADR). To be included in this review, the meta-analyses had to report their outcome in terms of mean difference with its 95%CI or mean estimate with SD from its pooled analysis after a follow-up period.

Identification of studies and literature search

Electronic databases including PubMed/MEDLINE, Scopus, Cochrane Library, and Embase were searched using a comprehensive search strategy from inception to November 2022 to identify all the studies. A snowball search on Google, Google Scholar, and Research Gate was also performed to identify any additional studies. The bibliographic lists of all identified articles were screened to check for the addition of any potentially relevant articles. A detailed search strategy in various databases is provided as Supplementary Table 1 and Supplementary Figures 1-3.

Study selection and data extraction

All the studies identified from the databases were retrieved to a Microsoft Excel sheet and their titles and abstracts were screened initially as per the predefined criteria. Consequently, the full texts of the included studies were screened for eligibility using the same criteria. First author’s last name and year of publication were used to identify the studies. All the relevant data such as study characteristics, eligibility criteria, literature search, quality assessment, heterogeneity, and study limitations were extracted to the data extraction grid. Two independent reviewers were involved in the study selection and data extraction and any disagreements between the reviewers were resolved through consensus or discussion with another reviewer.

Narrative synthesis

All the evidence extracted through the systematic process was presented in tabular form and summarised narratively. The evidence from the studies was classified in accordance with the clinical outcomes. A quantitative estimation was not performed as the data used in this review was already from a meta-analysis and another meta-analysis on the same data would exaggerate the actual effect of intervention. Hence, the synthesis was restricted it to the narrative synthesis.

RESULTS

Out of 364 identified studies, a total of 231 studies were screened for their title and abstract. A total of 75 studies were found eligible for full-text screening, and 156 records were excluded. Finally, a total of six meta-analyses[4,9,11,13,14] were included for this overview. A detailed description of the study selection process is provided in Figure 1.

Figure 1 PRISMA Flow diagram. EG: Endoscopic gastroplasty; MA: Meta-analysis.

Characteristics of included studies

The included meta-analyses were published between the years 2017 to 2020 in the United States (n = 3, 50%), Brazil (n = 2, 33.3%) and Australia (n = 1, 16.7%). The included patients were people with obesity or those who had gained weight after undergoing Roux-en-Y Gastric Bypass. The interventions included were ESG, Endoscopic TORe, full-thickness suturing (FTS) plus argon plasma mucosal coagulation (ft-TORe), argon plasma mucosal coagulation alone (APMC-TORe); EG with full-thickness suture or plication devices. These were compared with various techniques like intragastric balloon (IGB), laparoscopic sleeve gastroplasty (LSG), sham, diet, and lifestyle changes. All the studies reported any of the pre-defined outcomes. The included studies in the considered meta-analyses were randomized controlled trials (RCTs), observational studies, case-control studies, and cohort studies. The follow-up and outcome measurement in each study varied from 3 to 24 months. The characteristics of included studies are presented in Table 1.

Table 1 Characteristics of included studies.

No.	Ref.	Country	Eligibility/inclusion criteria	Search characteristics		Number of studies and total number of patients
				Databases searched	Search period and limitations
14	Singh et al[9], 2020	United States	Inclusion criteria: P: Patients underwent EBMT for obesity; I: ESG; C: IGB; O: %TWL, %EWL and adverse events; SD: All RCTs and observational studies with a minimal follow-up of 12 months; studies with more than one treatment arm, patients who underwent ESG or IGB alone with or without lifestyle modification	MEDLINE (PubMed), Scopus, Cochrane Register of Controlled Trials, and Web of Science databases	Inception to August 2019; restricted to observational and RCTs	28 studies; 1 study with direct comparison (n = 58)
			Exclusion criteria: Patients with prior or sequential EBMTs or bariatric surgery; Case reports and study with < 25 patients; studies with endoscopic gastroplasty techniques using devices other than the OverStitch endoscopic suturing system; studies with IGBs not approved by the United States Food and Drug Administration; Letters, editorials, expert opinions, and reviews without original data and studies with overlapping patient cohorts
42	Jaruvongvanich et al[11], 2020	United States	Inclusion criteria: P: Adults aged 18 years or older with obesity; I: ft-TORe or APMC-TORe to manage weight regain after RYGB; C: Any; O: TBWL; SD: case-control studies, cohort studies, RCTs	Ovid MEDLINE and Epub Ahead of Print, In-Process & Other Non-Indexed Citations, and Daily, Ovid Embase, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, and Scopus	Inception to February 10, 2020, limited to the English language and excluding animal studies	16 studies; 1625 participants
43	Mohan et al[13], 2020	United States	Inclusion criteria: P: Patients with moderate to severe obesity; I: ESG; C: LSG on published 2013; O: Pooled and difference in %TWL, %EWL and BMI with ESG at 1, 6, and 12 months; and adverse events; SD: All studies were included irrespective of the study sample-size, inpatient/outpatient setting, and geography as long as they provided data needed for the analysis	PubMed, EMBASE, Google-Scholar, Scopus, and Web of Science databases	Inception to August 2019; restricted to studies in human subjects and published in English language in peer-reviewed journals	15 studies; 3994 patients
			Exclusion criteria: Studies on LSG published until December 31, 2012; studies on LSG published as abstracts; studies on robot-assisted LSG; studies that did not report outcomes on weight loss, in terms of total weight loss and/or excess weight loss and/or BMI; studies that did not report first 12 months’ outcome data; studies done exclusively in elderly and/or geriatric population (age > 60 years); studies done in a pediatric population (age < 18 years); and studies not published in English language
50	Jalal et al[10], 2020	Australia	Inclusion criteria: P: Adult patient study populations greater or equal to 20 with obesity; I: ESG; C: LSG; O: Weight loss potential of ESG, ESG complications, rates of conversion to surgery and cost; SD: English studies as Case-control, cohort and RCTs	PubMed/MEDLINE, Cochrane Library, EMBASE/OVID, and the World Wide Web	NR	2 studies; 348 participants
			Exclusion criteria: Case reports or case series; Articles that did not have extractable ESG data; Same centre studies or studies with cross over patient populations; Patient cohort less than 20
56	Madruga-Neto et al[4], 2018	Brazil	Inclusion criteria: P: Patients with obesity (BMI > 30 kg/m2); I: EG with full-thickness suture or plication devices; C: sham or diet and lifestyle changes; O: AWL, %EWL, responder rate (%TWL ≥ 5%), and potential complications in 6 and 12 months; SD: Only RCTs without restrictions on language or publication year	MEDLINE [PubMed], Embase, Cochrane Central, and LILACS/BIREME	Inception to November 2017	3 studies; 459 patients
			Exclusion criteria: studies with follow-up periods < 1 month, those involving revision endoscopic procedures after bariatric surgery, and those involving patients who were overweight (BMI: 25-30 kg/m2)
57	Brunaldi et al[14], 2018	Brazil	Inclusion criteria: P: Patients with RYGB who presented with weight regain; I: Endoscopic therapy for weight regain following RYGB; C: Any comparator; O: AWL, EWL, and TBWL; SD: RCT, observational cohort studies, and case series and Conference abstracts if they met the eligibility criteria	MEDLINE, Embase, Scopus, Web of Science, Cochrane, OVID, CINAHL/EBSCo, LILACS/Bireme, and gray literature	Inception to October 31, 2016	15 studies; 882 patients
			Exclusion criteria: Reviews, editorials, case-control studies, studies using non-human subjects, articles without English translation, did not describe the endoscopic method clearly; follow-up weight or BMI less than 1 month; with endoscopic treatment for other indications besides weight regain, such as dumping syndrome or fistula closure; did not report baseline BMI; included patients with weight regain who had already undergone other endoscopic or surgical treatment of weight regain

APMC-TORe: Argon plasma mucosal coagulation alone; AWL: Absolute weight loss; BMI: Body mass index; C: Comparator; EBMT: Endoscopic bariatric and metabolic therapies; EG: Endoscopic gastroplasty; ESG: Endoscopic sleeve gastroplasty; EWL: Excess weight loss; I: Intervention; ft-TORe: Full-thickness suturing plus argon plasma mucosal coagulation; IGB: Intragastric balloon; LSG: Laparoscopic sleeve gastroplasty; NR: Not reported; O: Outcome; P: Participants; RCT: Randomized controlled trial; RYGB: Roux-en-Y Gastric Bypass; SD: Study design; TORe: Endoscopic transoral outlet reduction; TWL: Total weight loss.

Characteristics of literature search

The databases searched were MEDLINE (PubMed), Scopus, Cochrane Register of Controlled Trials, Web of Science databases, Google Scholar, World Wide Web, LILACS/ BIREME, CINAHL/EBSCo, and Gray literature. The latest search was on February 10, 2020. The applied search restrictions were English language, RCTs, and human studies. The details are presented in Table 1.

Quality assessment and heterogeneity in the included studies

The various tools such as Cochrane Collaboration risk assessment tool, New Castle Ottawa scale, National Institutes of Health quality assessment scale, GRADE standards, and Joanna Briggs Institute Checklist were used for the quality assessment of included studies. The quality of included studies ranged from very low to high quality. Only four studies[4,9,10,11], reported the heterogeneity in the analysis ranging it from low to high, while two studies did not report the heterogeneity. The quality assessment and heterogeneity characteristics of the included studies are provided in Table 2.

Table 2 Risk of bias assessment, heterogeneity, and limitations in the included studies.

Ref.	Country	Quality assessment/risk of bias	Author reported factors affecting the quality	Heterogeneity in the analysis	Author reported limitations in the review
Singh et al[9], 2020	United States	NOS; Score of 8	NR	Considerable	Limited quality of the included studies, lack of controlled ESG studies, lack of long-term follow-up data, lack of clarify the concomitant use of weight loss medication, considerable heterogeneity, lack of standardized definition for SAE
Jaruvongvanich et al[11], 2020	United States	RCTs: Cochrane Collaboration risk assessment tool; high quality (1), low to medium quality (2); Observational studies: National Institutes of Health quality assessment scale for pre-post Studies; high quality (7) and low to medium quality (6)	Reporting bias, inappropriate patient selection, blinding of personals and outcome assessors, loss to follow-up	High (0%-87%)	Small sample size, limited follow-up duration, small number of studies in each analysis, low to moderate methodological quality of the included studies, the small number of comparative trials between the 2 techniques, substantial heterogeneities, influence of concomitant pharmacotherapy, limiting generalizability of findings, reporting bias
Mohan et al[13], 2020	United States	NOS; 3 were high quality and the rest were medium quality	Included studies were not entirely representative of the general population and community practice, selection bias, inadequate follow-up	NR	Effect of additional confounding factors, included studies were not entirely representative of the general population and community practice, retrospective nature of studies, selection bias, lack of subgroup data based on the presence of comorbidities, lack of data on success/failure of procedure
Jalal et al[10], 2020	Australia	NR: Limited quality	High loss to follow-up rates	Low (19%)	Lack of studies, retrospective nature of cohort studies, lack of control groups, heterogeneity, lack of long-term study data, high loss to follow-up rates
Madruga-Neto et al[4], 2018	Brazil	JADAD quality score: 3; GRADE standards: Very low to moderate	Inappropriate randomization, inappropriate blinding	Moderate (I2: 50% to 68%)	Biases, heterogeneity, inadequate data
Brunaldi et al[14], 2018	Brazil	Joanna Briggs Institute Checklist; very low	Selection bias, unclear reporting of information	NR	Lack of RCT, bias, heterogeneity, low methodological quality, lack of comparative studies

ESG: Endoscopic sleeve gastroplasty; NR: Not reported; RCT: Randomized controlled trial; SAE: Serious adverse event.

Factors affecting the quality of included meta-analyses

Many factors were reported to affect the methodological quality of the included studies. This includes reporting bias, inappropriate patient selection, blinding of outcome assessors, lack of or inadequate follow-up, the inclusion of studies that were not entirely representative of the general population and community practice, and unclear reporting of information. The factors affecting the quality of included meta-analyses are provided in Table 2.

Limitations of the included meta-analyses

The following limitations were reported in the included studies: Limited quality of the included studies, lack of controlled ESG studies, lack of long-term follow-up data, lack of clarification of the confounding factors such as concomitant use of weight loss medication, considerable heterogeneity, lack of standardized definition for serious adverse events, small sample size, small number of studies in each analysis, limiting generalizability of findings, reporting bias, retrospective nature of studies, selection bias, lack of subgroup data based on the presence of comorbidities, lack of data on success/failure of procedure, lack of control groups, lack of follow-up, lack of RCTs and lack of comparative analyses. The author-reported limitations in each of the included studies are provided in Table 2.

Efficacy outcomes

TBWL: (1) At 3 months: A total of 3 studies reported the effect of EG in TBWL outcomes. Among them, two studies reported a better effect in the intervention group, whereas one study reported no significant difference. The meta-analysis by Brunaldi et al[14], from eight studies with 320 participants recorded significantly (P < 0.001) greater weight loss with FTS with argon plasma coagulation (APC) group compared to FTS alone. The study by Jaruvongvanich et al[11] demonstrated significantly greater weight loss with TORe than TORe-gastroplasty from two studies. However, non-significant difference in weight loss was observed between ft-TORe and APMC-TORe[11]. The detailed information is provided in Table 3; (2) At six months: A total of 3 studies with 4 analyses reported the outcomes of EG at six months, of which 2 studies reported on ESG. The study by Singh et al[9] observed significantly (P = 0.002) superior weight loss with ESG than IGB, whereas Jalal et al[10], reported significantly (P < 0.00001) superior weight loss with LSG than ESG. The remaining two studies recorded a non-significant difference between ft-TORe vs APMC-TORe,11 and TORe vs TORe-gastroplasty[11]. The detailed information is provided in Table 3; (3) At 12 months: A total of three studies reported the outcomes of EG at 12 months, out of which two studies reported on ESG. The study by Singh et al[9] observed significantly (P = 0.0001) superior weight loss with ESG than IGB, whereas Mohan et al[13], reported significantly (P = 0.001) superior weight loss with LSG than ESG. The study by Madruga-Neto et al[4], recorded significantly better weight loss with EG than Sham control (P = 0.01). The detailed information is provided in Table 3; and (4) At 18-24 months: Only Singh et al[9] reported the outcomes at 18-24 months, which revealed significantly (P = 0.0003) superior weight loss with ESG compared to IGB. The detailed information is provided in Table 3.

Table 3 The effect of endoscopic gastroplasty vs other interventions on total body weight loss.

Ref.	Country	Number of studies & participants	Intervention and comparator	Effect measure	Effect size (95%CI)	Interpretation with respect to intervention
3 months
Brunaldi et al[14], 2018	Brazil	8 studies; 320 participants	FTS-APC vs FTS-alone	% mean	FTS-APC: 25.0 ± 1.99; FTS-alone: 15.3 ± 9.88	The FTS-APC group and demonstrated significantly a greater weight loss compared to FTS alone (P < 0.001)
Jaruvongvanich et al[11], 2020	United States	2 studies	ft-TORe vs APMC-TORe	% mean difference	-0.1 (-4.6 to 4.4)	There was no significant difference between two treatments
Jaruvongvanich et al[11], 2020	United States	2 studies	TORe vs TORe-gastroplasty	% mean difference	2.6 (0.5 to 4.7)	The TORe group and demonstrated significantly a greater weight loss compared to TORe-gastroplasty
6 months
Singh et al[9], 2020	United States	1 study; 58 participants	ESG vs IGB	% mean difference	3.07 (1.46–4.67)	ESG achieved significantly (P = 0.002) superior weight loss compared to IGB
Jalal et al[10], 2020	Australia	2 studies; 348 participants	ESG vs LSG	% mean difference	8.52 (6.35–0.69)	LSG appeared to have a significantly (P < 0.00001) superior weight loss compared to ESG
Jaruvongvanich et al[11], 2020	United States	3 studies	ft-TORe vs APMC-TORe	% mean difference	0.3 (-5.5 to 6.0)	There was no significant difference between two treatments
Jaruvongvanich et al[11], 2020	United States	2 studies	TORe vs TORe-gastroplasty	% mean difference	0.8 (-2.3 to 3.9)	There was no significant difference between two treatments
12 months
Singh et al[9], 2020	United States	1 study; 58 participants	ESG vs IGB	% mean difference	7.33 (5.22-9.44)	ESG achieved significantly (P = 0.0001) superior weight loss compared to IGB
Mohan et al[13], 2020	United States	9 studies	ESG vs LSG	% mean	ESG: 17.08 (15.05-19.10); LSG: 30.5 (27.4-33.5)	LSG appeared to have a significantly (P = 0.001) superior weight loss compared to ESG
Madruga-Neto et al[4], 2018	Brazil	2 studies; 376 participants	EG vs Sham	% mean difference	4.8 (1.1-8.51)	The difference between the groups was significantly (P = 0.01) higher in the intervention group than in the control group
18-24 months
Singh et al[9], 2020	United States	1 study; 58 participants	ESG vs IGB	% mean difference	11.51 (5.33-17.69)	ESG achieved significantly (P = 0.0003) superior weight loss compared to IGB

APC: Argon plasma coagulation; APMC-TORe: Argon plasma mucosal coagulation alone; ESG: Endoscopic Sleeve Gastroplasty; ft-TORe: Full-thickness suturing plus argon plasma mucosal coagulation; FTS: Full-thickness suturing; IGB: Intragastric balloon; LSG: Laparoscopic sleeve gastroplasty; TORe: Endoscopic transoral outlet reduction.

Excessive weight loss: (1) At six months: Only two studies reported the EWL following EG treatment at six months of which, Singh et al[9], recorded significantly (P = 0.0001) better weight loss with ESG compared to IGB, whereas the study by Madruga-Neto et al[4] reported a non-significant (P = 0.07) difference between EG and conventional treatment. The detailed information is provided in Table 4; (2) At 12 months: A total of four studies recorded EWL at 12 months of which the study by Singh et al[9], recorded significantly (P = 0.0001) superior weight loss with ESG compared to IGB. Similarly, Brunaldi et al[14], demonstrated significantly (P < 0.001) greater weight loss in FTS-APC group compared to FTS alone, whereas another two studies, Mohan et al[13], and Madruga-Neto et al[4], observed a non-significant difference in EWL between LSG vs ESG and EG vs conventional therapy procedures. The detailed information is provided in Table 4; and (3) At 18-24 months: Two studies reported EWL at 18-24 months of which Singh et al[9], reported significantly (P = 0.0001) superior weight loss with ESG compared to IGB, and Brunaldi et al[14], recorded significantly (P < 0.0001) superior weight loss with FTS-APC compared to FTS-alone. The detailed information is provided in Table 4.

Table 4 The effect of endoscopic gastroplasty vs other interventions on excessive weight loss.

Ref.	Country	Number of studies & participants	Intervention and comparator	Effect measure	Effect size (95%CI)	Interpretation with respect to intervention
6 months
Singh et al[9], 2020	United States	1 study; 58 participants	ESG vs IGB	% mean difference	20.80 (12.50-29.10)	ESG achieved significantly (P = 0.0001) superior weight loss compared to IGB
Madruga-Neto et al[4], 2018	Brazil	2 studies; 127 participants	EG vs CT	% mean difference	17.87 (-1.8 to 37.54)	There was no significant (P = 0.07) difference between the groups, although the EG group presented higher effect than the control group
12 months
Singh et al[9], 2020	United States	1 study; 58 participants	ESG vs IGB	% mean difference	30.99 (22.81-39.16)	ESG achieved significantly (P = 0.0001) superior weight loss compared to IGB
Mohan et al[13], 2020	United States	9 studies	ESG vs LSG	% pooled mean	ESG: 62.3 (51.3-74.6); LSG: 69.3 (60.1-78.4)	There was no significant (P = 0.4) difference between LSG and ESG
Madruga-Neto et al[4], 2018	Brazil	2 studies; 127 participants	EG vs CT	% mean difference	16.01 (-1.48 to 33.5)	There was no significant (P = 0.07) difference between the groups, although the EG group presented higher effect than the control group
Brunaldi et al[14], 2018	Brazil	14 studies; 619 participants	FTS-APC vs FTS-alone	% mean difference	FTS-APC: 27.0 ± 2.91; FTS-alone: 17.8 ± 15.3	The FTS-APC group and demonstrated significantly a greater weight loss compared to FTS alone (P < 0.001)
18-24 months
Singh et al[9], 2020	United States	1 study; 58 participants	ESG vs IGB	% mean difference	43.78 (35.98-51.58)	ESG achieved significantly (P = 0.0001) superior weight loss compared to IGB
Brunaldi et al[14], 2018	Brazil	10 studies; 455 participants	FTS-APC vs FTS-alone	% mean difference	FTS-APC: 24.2 ± 0.84; FTS-alone: 11.7 ± 21.6	The FTS-APC group and demonstrated significantly a greater weight loss compared to FTS alone (P < 0.001)

CT: Conservative therapy; EG: Endoscopic gastroplasty; ESG: Endoscopic sleeve gastroplasty; IGB: Intragastric balloon; LSG: Laparoscopic sleeve gastroplasty; FTS: Full-thickness suturing; APC: Argon plasma coagulation.

AWL: (1) At six months: Only the meta-analysis by Madruga-Neto et al[4], reported the AWL at six months which indicated that there was significant (P < 0.0001) weight reduction in the EG group compared to the control group. The detailed information is provided in Table 5; and (2) At 12 months: Only two studies recorded the AWL at 12 months. The meta-analysis by Madruga-Neto et al[4], recorded significant (P = 0.03) weight reduction in the EG group compared to the control group. The analysis by Mohan et al[12], recorded a non-significant (P = 0.65) difference in AWL between LSG and ESG groups. The detailed information is provided in Table 5. No meta-analyses analyzed the outcomes with more than 12-month follow-up. The detailed information is provided in Table 6.

Table 5 The effect of endoscopic gastroplasty vs other interventions on Absolute Weight Loss.

Ref.	Country	Number of studies & participants	Intervention and comparator	Effect measure	Effect size (95%CI)	Interpretation with respect to intervention
6 months
Madruga-Neto et al[4], 2018	Brazil	2 studies; 127 participants	EG vs CT	% mean difference	7.05 (3.83 to 10.28)	There was significant (P < 0.0001) weight reduction in the EG group than the control group
12 months
Mohan et al[13], 2020	United States	11 studies	ESG vs LSG	% pooled mean	ESG: 29.3 (27.6-32.3) LSG: 29.3 (27.1-31.4)	There was no significant (P = 0.65) difference between LSG and ESG
Madruga-Neto et al[4], 2018	Brazil	2 studies; 127 participants	EG vs CT	% mean difference	4.99 (0.52 to 9.46)	There was a significant (P = 0.03) weight reduction in EG group than the control group

CT: Conservative therapy; EG: Endoscopic gastroplasty; ESG: Endoscopic sleeve gastroplasty; LSG: Laparoscopic sleeve gastroplasty.

Table 6 Adverse Events following the treatment endoscopic gastroplasty vs other interventions.

Ref.	Country	Number of studies & participants	Intervention and comparator	Adverse events reported
Singh et al[9], 2020	United States	1 study; 58 participants	ESG vs IGB	3 (2 upper gastrointestinal bleeding, 1 peri gastric fluid collection)
Mohan et al[13], 2020	United States	15 studies	ESG vs LSG	All ADRs: ESG: 2.9 % (1.8-4.4); LSG: 11.8 % (8.4-16.4); P = 0.001; Bleeding: ESG: 1.1 % (0.7-1.8); LSG: 2.6 % (1.9-3.7); P = 0.005; GERD: ESG: 0.4 % (0.1-1.1); LSG: 5.8 % (3.5-9.3, 73); P = 0.001
Madruga-Neto et al[4], 2018	Brazil	3 studies; 459 patients	ESG vs CT	The total rate of adverse events in the EG group was 52.9%-77.8%, of which 5.0%-5.2% of the events were severe
Jaruvongvanich et al[11], 2020	United States	16 studies; 1625 participants	ft-TORe vs APMC-TORe	APMC-TORe: GI bleeding (1); Overall AE rates: ft-TORe: 9.3% (8-17.8); APMC-TORe: 6.4% (1.9-10.9); stricture rates: ft-TORe: 3.3% (1.4-5.3); APMC-TORe: 4.8% (2.3-7.2)

EG: Endoscopic gastroplasty; ESG: Endoscopic sleeve gastroplasty; IGB: Intragastric balloon; LSG: Laparoscopic sleeve gastroplasty; APMC-TORe: Argon plasma mucosal coagulation alone; CT: Conventional therapy; ft-TORe: full-thickness suturing plus argon plasma mucosal coagulation.

Safety outcomes

Adverse effects: Only 4 analyses reported the safety outcomes in their meta-analyses. However, there was no comparative estimation, though some pooled the occurrence of ADRs. The safety outcomes were reported as upper gastrointestinal bleeding, peri gastric fluid collection, all ADRs, bleeding, GERD, rate of adverse events, and stricture rates.

DISCUSSION

Excess weight can impact all domains of human health including physical, mental, spiritual, and social[15]. Obesity can contribute to many comorbidities like hyperglycemia, hypertension, hypercholesterolemia, and metabolic dysfunction-associated steatotic liver disease[16]. It should be managed appropriately through suitable non-surgical and surgical interventional strategies to control the morbidities and mortalities associated with it. Many surgical strategies were employed for obesity management and ESG is an excellent and effective strategy that is minimally invasive[17]. However, there is a paucity of the evidence with respect to the efficacy and safety of the ESG method compared to the other strategies.

This overview revealed that there were six meta-analyses assessing the effectiveness of EG among obese patients and 50% of the included studies were from the United States. These studies reported their outcomes in three to 24 months. All the studies searched the major databases, without many limitations other than English language which is acceptable. However, the lack of primary research raised questions about the validity of conclusions from the meta-analyses.

All the included studies derived their findings of comparative analysis from very few studies. This is due to the lack of comparative analysis available on the efficacy and safety of EG in the treatment of obesity. The quality of included studies varied among the meta-analyses and there was a considerable level of heterogeneity in the analyses. Many factors contributed to the risk of bias in its all aspects and future studies should attempt to reduce the risk of bias. It’s a fact that it may not be possible to blind the participants and study personals in all circumstances, but blinding of outcome assessors will be an effective method to control the bias[18].

The non-randomized nature as well as the retrospective or cohort nature of the studies also contributed to the bias in the included meta-analyses. In some clinical settings, it may be difficult to randomize the surgical intervention and, in some cases, researchers usually prefer to go with the non-randomized design. In those cases, researchers can use methodological quality tools such as ROBINS-I and II tool[19], and Downs and Black[20], checklists for non-interventions. The use of stringent critical appraisal tools like Cochrane Risk of Bias Assessment will significantly reduce the quality of the studies. Unclear reporting as well as undefined uniform criteria for outcomes assessments also contributed to the risk of bias, therefore further meta-analyses should be planned appropriately. In cases of heterogeneous clinical outcomes in the included studies, a subgroup analysis or meta-regression analysis should be planned and performed to understand the influence of potential factors or to understand the effect of intervention on a specific population or a set of characteristics[21].

Though the outcomes in the included meta-analyses had a follow-up of three to 24 months, there was not much data available with a longer follow-up. There was a conflicting finding with respect to the effect of ESG on TBWL compared to the LSG[9,10]. Also, a non-significant difference was observed between ft-TORe vs APMC-TORe and TORe vs TORe-gastroplasty[11]. Altogether, this overview demonstrates the superior effect of EG over other methods which is backed by recent clinical evidence. The new literature evidence confirms the proven efficacy and safety of EG procedure for treating obesity and controlling associated comorbidities[22].

The effect of EG in EWL and AWL is very effective and ESG is shown to be a promising safer option for obesity. Similarly, FTS-APC also appeared to be a better option than FTS-alone for EWL and AWL in patients with obesity or excess weight, and the present literature supports this fact[16,23]. Furthermore, there was a great technical success rate and post-procedure gastric sleeve and suture intactness following ESG[24,25]. The safety concerns were tolerable among patients undergoing EG management for obesity. ADRs such as upper gastrointestinal bleeding, peri gastric fluid collection, bleeding, and GERD were observed among the patients[4,9,11,13]; however, more data is required on this aspect.

The high level of hetrogeneity and the potential risk of bias might have hampered the robustness of the existing findings. Multiple mitigation strategies such as adequate reporting, bias controlled long-term prospective comparative studies, adapting the appropriate randomization technique, controlling the confounding factors, adequate follow-up, adequately powered with a homogenized larger sample sized multi-centric studies, standardized definition to the outcomes, and its reporting, subgroup and sensitivity analyses along with adequate reporting of the findings as per the validated guidelines[8,26-30] will help the future studies to bring a robust and stronger conclusion.

Study limitations

There are some limitations to this study. Firstly, there are very limited data available on the efficacy and safety of EG for obesity, hence more research needs to be conducted to get a clear picture. Secondly, the generalizability of the information globally needs more studies as the major part of derived from specific parts of the world. Hence, more studies need to be conducted across the world to understand the effect of racial and cultural diversity on the effects of intervention. The risk of bias in the included studies ranged from very low to high. Hence, future studies should be planned carefully to ensure higher quality and include adequate participants. The lack of randomized comparative studies are the major concern in this domain. Additionally, outcomes for the obesity treatment should be studies in a long-term manner to fully understand the clinical effectiveness and long-term safety concerns. Hence, future long-term high-quality randomized trials should be planned to understand these concerns and durability of the treatment, which will help decision makers to decide the duration of treatment as well.

CONCLUSION

The limited available evidence indicates that EG was effective and safe in the treatment of obesity compared to other interventional strategies. The quality of evidence was moderate to high. Further good quality well-controlled comparative studies with long-term follow-up is needed to strengthen this evidence. Additionally, the subgroup and sensitivity analyses will increase the generalizability of the findings.