Efficacy of endoscopic vacuum therapy in esophageal luminal defects: a systematic review and meta-analysis
Article information
Abstract
Background/Aims
Endoscopic vacuum-assisted closure (EVAC) is a novel technique used to repair esophageal perforation and leaks. Varying data have been reported on the overall success rate of EVAC. We aimed to conduct a meta-analysis of the available data on the clinical success rate of EVAC.
Methods
Electronic databases were searched for publications addressing the efficacy of EVAC in esophageal luminal defects. Pooling was conducted using both fixed and random-effects models. The overall clinical success of EVAC therapy was considered the primary outcome, whereas, overall complication rates, need for adjunct therapy, and mortality were considered secondary outcomes.
Results
In total, 366 patients were included in the study. On pooled analysis, the mean age was 66 years with 68.32% of patients being men. Overall pooled clinical success rate of EVAC therapy was 87.95%. Upon subgroup analysis, the pooled clinical success rate of postsurgical anastomotic leak and transmural esophageal perforation were found to be 86.57% and 88.89%, respectively. The all-cause hospital mortality was 14% and 4.2% in patients with esophageal perforation and EVAC, respectively.
Conclusions
This study demonstrates that EVAC therapy has a high overall clinical success rate, with low mortality. EVAC therapy seems to be a promising procedure with excellent outcomes in patients with luminal esophageal defects.
INTRODUCTION
Spontaneous or iatrogenic esophageal luminal defects as well as postoperative leaks in esophagogastrostomy and esophagojejunostomy are considered life-threatening conditions due to the development of mediastinitis and consecutive sepsis.1 The mainstay of treatment is closure of the esophageal luminal defect and drainage of the mediastinum septic focus.1 Flexible covered self-expandable metal stents (SEMS) are frequently used for the management of esophageal luminal defects.1,2 However, SEMS are frequently associated with migration and undrained mediastinal foci of abscess, which requires interventional radiology or surgical drainage.1 Surgical repair is a complicated two-step esophageal reconstruction procedure associated with impairment of patient’s quality of life as well as high morbidity and mortality.1 Thus, endoscopic vacuum therapy (EVAC) has become a promising alternative therapy for esophageal luminal defects.1,2
Vacuum-assisted closure is a well-established technique used for the closure of superficial cutaneous infected wounds based on negative pressure applied to the wound via a vacuum sealed sponge.3 Use of the sponge continuously removes the wound secretions and edema and improves microcirculation.4 Therefore, it results in the accelerated formation of granulation tissue and clean wound closure of the infected wound by primary intention.1 The same principle was used to treat accessible esophageal leakage, where porous polyurethane sponges had been connected to a drainage tube that generates continuous negative pressure, facilitates drainage of the cavity, stimulates formation of granulation tissue while reducing edema, and results in closure of the luminal defect with primary intention.5,6
The primary aim of the study was to evaluate the efficacy and safety of EVAC in esophageal leak due to perforation and anastomotic leak postsurgery.
METHODS
Definitions
Leaks were defined as the postsurgical disruption of the anastomotic site, leading to the accumulation of fluid. Perforation was defined as a spontaneous or iatrogenic transmural full thickness defect of the esophagus. However, multiple studies have used these terms interchangeably.7 Clinical success of EVAC was defined as successful luminal defect closure. All-cause mortality was defined as death resulting from any cause, without being restricted to a specific intervention. Adverse events (AEs) included stenosis, bleeding, sponge migration, and sepsis as a sequel of EVAC therapy.
Search methodology
A literature search was conducted using the electronic database engines Medline through PubMed, Ovid, Cochrane library (Cochrane Central Register of Controlled Trials and Cochrane Database of Meta-Analysis), Embase, American College of Physicians journal club, and Database of Abstracts of Reviews of Effects (DARE) according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines from January 1974 through November 2021 to identify studies addressing the overall results of EVAC therapy. We used the following search keywords: anastomotic leak OR post operative leak OR anastomotic leakage OR post operative leakage OR esophageal leak OR esophageal leakage OR esophageal fistula OR leakage OR fistula OR esophagectomy OR negative pressure OR vacuum assisted closure OR endovac therapy OR endo sponge OR upper GI leaks OR endoscopic vacuum therapy OR esophageal perforation OR anastomosis AND endoscopic vacuum therapy. The studies that were collected underwent scrutiny to eliminate any possible duplicates or data redundancy.
Study selection
During the initial phase of study selection, scrutiny was applied to the titles and abstracts of retrieved articles through keyword searches to eliminate irrelevant ones. Subsequently, the complete texts of all chosen studies underwent screening based on predetermined inclusion and exclusion criteria. The inclusion criteria comprised the following: (1) confirmation of perforation, leak, or fistula within the upper gastrointestinal tract; (2) employment of EVAC as either a primary or salvage intervention; and (3) investigation involving adult individuals aged ≥18 years. Conversely, the exclusion criteria encompassed the following: (1) article formats other than original research articles; (2) case reports involving >10 patients; (3) publications restricted to abstracts only; (4) works not published in English; and (5) studies conducted on animal models, editorials, and comments. In cases where multiple publications addressed the same study cohort, only the most recent iteration was retained for analysis.
Data extraction and quality assessment
The data were independently extracted by two authors (IV and HG) into a standardized form—patient demographics (number of patients enrolled, mean age, gender), study characteristics (including authors of the study, duration of the study, year of publication, and country where the population was studied), study design, intervention specifics (such as the number of EVAC procedures, indications, and sponge details when available), and outcomes (including EVAC-related mortality, overall mortality rate, and complications). In total, 15 articles matched the study criterion and full-text articles were reviewed independently by two authors (IV, HG). Data were extracted, meeting both inclusion and exclusion criteria following review of entire content of each paper. Any differences were resolved by a third investigator (SP) for discussion or revision. The agreement was evaluated using Cohen’s κ.
Outcomes evaluated
The primary outcome of the study was to assess pooled estimates of the overall clinical success rate of EVAC therapy in esophageal luminal defects. Secondary outcomes included AEs such as migration, mortality, dysphagia, or esophageal stenosis/stricture during follow-up. The pooled estimates of the number of times sponges were changed, interval between changing of the sponges, and pooled estimates of feeding tolerability were calculated.
Statistical analysis
The data analysis was performed using Excel 2021 ver. 16.0 (Microsoft).8 The meta-analysis process involved the calculation of pooled proportions. For each individual study, the proportions were converted into a standardized quantity using the Freeman–Tukey variant of the arcsine square-root transformed proportion. The pooled proportion was then determined by reversing these transformation proportions, which was based on the weighted mean of the transformed proportions. The fixed effects model used inverse arcsine variance weights, while the random-effects model employed DerSimonian-Laird weights. Forest plots were used to visually represent the point estimates of each study in relation to the overall combined estimate, with the width of these points indicating the degree of influence of each study in the analysis. To assess the potential impact of publication and selection bias on the summary estimates, both the Egger bias indicator and the Begg–Mazumdar bias indicator were applied.9,10 Furthermore, funnel plots were generated to examine potential publication bias, considering the standard error and diagnostic odds ratio.11
RESULTS
The initial search identified 347 studies, of which 105 relevant articles were reviewed. Data were extracted from 15 studies comprising 366 patients that met the inclusion criteria and were included in the final analysis. In studies that used propensity-matched analysis, the data reported from the propensity-matched cohorts were used for the final analysis. The guidelines provided by the PRISMA were used to outline the specifics of the review process as shown in Figure 1.
Study flow diagram according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines.
Complete full-text articles are accessible for all the studies included in the analysis.1,2,5,12-23 Baseline demographics and study characteristics are shown in Table 1.1,2,5,12-23 Upon pooled analysis, the mean age of the patients was found to be 66 years (standard deviation=15) with 68.32% of the patients being men. The etiology of esophageal defects was as follows: anastomotic leaks 63.2%, spontaneous perforations 17.8%, and nonsurgical iatrogenic 19.1%.
Baseline study characteristics and demographics of included patients
Outcomes
On pooled analysis, overall clinical success rate of EVAC was 87.95% (95% confidence interval [CI], 84.46%–91.05%). The complete restoration of postsurgical anastomotic leak and full thickness perforation was shown in 86.57% (95% CI, 81.94%–90.61%) and 88.89% (95% CI, 83.22%–93.51%) of patients, respectively. A Forest plots demonstrating the individual study estimates and the pooled estimate for clinical success rate of EVAC in esophageal perforation is shown in Figure 2. The pooled estimate for clinical success rate of EVAC in postsurgical anastomotic leak is shown in Figure 3A, and that for full thickness perforation is shown in Figure 3B. No heterogeneity was observed with an I2 score of 31.0% (95% CI, 0%–61.90%) for overall clinical success rate of EVAC, an I2 score of 5% (95% CI, 0%–50.00%) for complete restoration of postsurgical anastomotic leak, and an I2 score of 48.80% (95% CI, 0%–71.00%) for full thickness perforation. No publication bias was noted when calculations were performed using the Egger bias indicator; –0.99 (95% CI, –2.84 to 0.85; p=0.26) or Harbord bias indicator; 1.08 (95% CI, –0.88 to 3.05; p=0.31) for overall clinical success rate of EVAC therapy in the management of esophageal luminal defects. A funnel plot demonstrating no evidence of publication bias is shown in Figure 4.
Forest plots showing overall clinical success of endoscopic vacuum-assisted closure in the management of esophageal perforations.
Forest plots showing clinical success of endoscopic vacuum-assisted closure in complete restoration of the esophageal postsurgical anastomotic leak (A) and the esophageal perforation (B).
Funnel plot showing no publication bias for overall clinical success of endoscopic vacuum-assisted closure in the management of esophageal perforations.
The mean duration of EVAC therapy was 16.21 (95% CI, 12.61–19.92) days, with the mean number of times sponge changed per patient being 4.62 (95% CI, 3.71–5.53) days, and the replacement interval being 3.7 (95% CI, 3.4–4.1) days. The all-cause hospital mortality was 14.13% (95% CI, 10.52%–17.81%) in patients admitted with esophageal perforation. Mortality related to EVAC therapy was 4.21% (95% CI, 2.32%–6.56%). A Forest plots demonstrating the individual study estimates and the pooled estimate for mortality is shown in Figure 5. No heterogeneity was observed with an I2 score of 39.30% (95% CI, 0%–66.50%). The Egger bias indicator was 0.86 (95% CI, 0.19–1.52), indicating no evidence of publication bias as shown in Figure 6.
Forest plots showing endoscopic vacuum-assisted closure-related mortality in the esophageal perforations.
Funnel plot showing no publication bias for endoscopic vacuum-assisted closure-related mortality in patients with esophageal perforations.
The overall AE rate for use of EVAC was 12.56% (95% CI, 10.32%–14.65%). The mean number of sponge migration was 2.62 (95% CI, 0.32–4.91) per patient. The median post hospital follow-up was 210 (95% CI, 150–336) days. During follow-up, 5.51% (95% CI, 2.12%–7.76%) of patients developed stenosis and clinically presented with dysphagia. SEMS was used in 8.52% (95% CI, 7.41%–9.96%) of patients in combination with EVAC therapy. Regarding nutrition, enteral feeding was tolerated in 10.52% (95% CI, 4.62%–15.83%) of patients during the hospital stay. The most common cause of mortality amongst EVAC patients was mediastinitis, sepsis, or multi-organ failure.
DISCUSSION
This meta-analysis and systematic review elucidated the clinical success of EVAC in the treatment of esophageal perforation. Our results demonstrated high rates of clinical success for EVAC, with low overall mortality and AEs. Our results support EVAC therapy as a clinically effective treatment for esophageal luminal defects including postsurgical leaks, as well as iatrogenic and noniatrogenic esophageal perforations.
As newer modalities continue to evolve, they are being assessed for their safety and effectiveness in the management of esophageal perforation.24 The basic principle is to seal the esophageal perforation followed by drainage of fluid collection to prevent secondary mediastinitis, abscess formation, and systemic infection. The current approach includes conservative management, surgery, and endoscopic intervention. Conservative management is used to treat small esophageal leaks without sepsis. Conservative management includes the use of antibiotic therapy, parenteral nutrition, and the placement of a nasogastric tube under direct visualization.24 Surgery is recommended for large esophageal perforation and necrosis, which are refractory to other treatment modalities. Surgical interventions may involve direct or enforced suturing, and in severe cases, pleural patches or flaps may be reserved for use.25,26 Endoscopic approach is an intermediate pathway between two extremes of conservative management and surgery.27 Endoscopic armamentarium includes clips, stents, suturing, or sealants for the management of esophageal perforation. The clips are of two types; through-the-scope clips or over the scope clips (OTSC). For clip closure the adjacent tissue should be healthy to facilitate adhesion of the clip.28 In case of significant friability or necrosis, there is a high risk for technical and clinical failure.28 In a recent metanalysis, the pooled clinical success rate of OTSC was 78.4%.29 SEMS are commonly used for medical management of fistula, leakage, and perforation. In a recent systematic review, the pooled clinical success rate of esophageal stent was 76.8%.30 However, esophageal stents are associated with AEs including migration, bleeding, perforation, stricture, and leaks. In our meta-analysis, pooled clinical success rate of EVAC is 87.95% in successful closure of the esophageal defects, which is significantly higher than that for either stents or OTSC.
In a recent metanalysis, EVAC was compared to SEMS, EVAC therapy was noted to have lower AEs (risk difference, 0.24; 95% CI, 0.13–0.35) and mortality (odds ratio, 0.39; 95% CI, 0.18–0.83).31,32 The rate of lower AEs in EVAC stems from difference in technique. Since sponges need to be changed every 3 to 4 days, wound healing is visually assessed every time a sponge is exchanged.30 We hypothesized that frequent wound evaluation may prevent any AEs before they can progress. This is not feasible with either SEMS or clips. Moreover, in case of esophageal perforation managed with SEMS or clips, extra-esophageal fluid collection and drainage must be performed through an external drain to prevent mediastinitis or systemic infection.33 The EVAC therapy is attached with negative pressure apparatus, which drains fluid accumulations, which is not possible with SEMS. This provides a better safety profile with EVAC therapy compared to SEMS. As EVAC therapy is relatively a novel technique, the variability of negative pressure, duration between sponge changes, occurrence rate of AEs, treatment duration, and length of hospital stay accounts for the heterogeneity observed in our analysis. As the EVAC technique becomes more refined, the rate of AEs will continue to be lower than those reported in current data.
Our study had a few limitations. The included studies had inherent bias, due to the observational nature of the studies, as well as a majority of studies being reported from Germany. The patient population was heterogeneous, which is multifactorial in origin. First, the underlying etiology of transmural defects might vary. The size and site of the esophageal defects, presence of underlying comorbidities, and delay in seeking medical attention may contribute toward heterogeneity of patient population. The EVAC therapy is operator dependent and not standardized and the variations in the extent of negative pressure and the frequency of sponge changes also contributed to the heterogeneity among cases. EVAC therapy is typically carried out in specialized tertiary centers by skilled endoscopists, which could potentially influence its effectiveness. Additionally, there is a possibility of publication bias in our study, as negative findings might not have been reported. As a result, interpreting our results cautiously, considering that they could potentially overstate the clinical effectiveness and success of EVAC therapy, is important. Nevertheless, our study has several strengths. Our study provides pooled clinical success rates based on underlying etiology, rate of AEs, and overall mortality, which did not have significant heterogeneity.
In conclusion, our meta-analysis demonstrates that EVAC therapy is a clinically effective and safe technique in patients with esophageal perforations and leaks. Further studies are needed to directly compare EVAC therapy with other modalities such as OTSC, surgery, and stenting.
Notes
Ethical Statements
Not applicable.
Conflicts of Interest
Dr. Neil R. Sharma is a consultant for Boston Scientific, Medtronic, Steris, and Olympus. The other authors have no potential conflicts of interest.
Funding
None.
Author Contributions
Conceptualization: all authors; Data curation: IV, HG; Formal analysis: all authors; Investigation: IV, HG; Visualization: all authors; Writing–original draft: IV, HG; Writing–review & editing: all authors.