J Gastrointest Surg (2016) 20:237–243 DOI 10.1007/s11605-015-3044-4
2015 SSAT PLENARY PRESENTATION
Surgical Endoscopic Vacuum Therapy for Defects of the Upper Gastrointestinal Tract Florian Kuehn 1 & Leif Schiffmann 1,2 & Florian Janisch 1 & Frank Schwandner 1 & Guido Alsfasser 1 & Michael Gock 1 & Ernst Klar 1
Received: 13 May 2015 / Accepted: 24 November 2015 / Published online: 7 December 2015 # 2015 The Society for Surgery of the Alimentary Tract
Abstract Introduction Intraluminal therapy used in the gastrointestinal (GI) tract was first shown for anastomotic leaks after rectal resection. Since a few years vacuum sponge therapy is increasingly being recognized as a new promising method for repairing upper GI defects of different etiology. The principles of vacuum-assisted closure (VAC) therapy remain the same no matter of localization: Continuous or intermittent suction and drainage decrease bacterial contamination, secretion, and local edema. At the same time, perfusion and granulation is promoted. However, data for endoscopic vacuum therapy (EVT) of the upper intestinal tract are still scarce and consist of only a few case reports and small series with low number of patients. Objectives Here, we present a single center experience of EVT for substantial wall defects in the upper GI tract. Methods Retrospective single-center analysis of EVT for various defects of the upper GI tract over a time period of 4 years (2011–2015) with a mean follow-up of 17 (2–45) months was used. If necessary, initial endoscopic sponge placement was performed in combination with open surgical revision. Results In total, 126 polyurethane sponges were placed in upper gastrointestinal defects of 21 patients with a median age of 72 years (range, 49–80). Most frequent indication for EVT was anastomotic leakage after esophageal or gastric resection (n=11) and iatrogenic esophageal perforation (n=8). The median number of sponge insertions was five (range, 1–14) with a mean changing interval of 3 days (range, 2–4). Median time of therapy was 15 days (range, 3–46). EVT in combination with surgery took place in nine of 21 patients (43 %). A successful vacuum therapy for upper intestinal defects with local control of the septic focus was achieved in 19 of 21 patients (90.5 %). Conclusion EVT is a promising approach for postoperative, iatrogenic, or spontaneous lesions of the upper GI tract. In this series, EVT was combined with operative revision in a relevant proportion of patients. Keywords Anastomotic leakage . Perforation . Esophageal . Endoscopy . Endoscopic vacuum therapy
Florian Kuehn and Leif Schiffmann contributed equally to this work. * Florian Kuehn
[email protected] 1
Department of General, Thoracic, Vascular and Transplantation Surgery, University of Rostock, Schillingallee 35, 18057 Rostock, Germany
2
Protestant Hospital Lippstadt, Wiedenbrücker Str. 33, 59555 Lippstadt, Germany
Introduction Esophageal rupture and anastomotic leakage of the upper gastrointestinal (GI) tract result in significant morbidity and mortality of the patients. Causes for this failure of the esophageal continuity may be iatrogenic injury mainly during endoscopy (30–83 %), spontaneous rupture (5–56 %), trauma (0–14 %), foreign bodies (0–35 %), tumors (0–5 %), surgical procedures (0–7 %), and others in up to 15 %.1–11 Depending on the interval between injury and the start of treatment, mortality ranges from 0 to 18 % within the first 24 h and 7 to 38 % for a later start.1–11 After esophageal resection, anastomotic leakage might occur between 0 and 18 % with a mortality ranging from 0 to 50 %.12–15 There are no strict guidelines
238
on how to proceed in these life-threatening situations. Therefore, therapy differs depending on institution and primary attending discipline. Endoscopic vacuum therapy (EVT) is increasingly being recognized as a new promising method for repairing upper GI defects of different etiology. Compared to a stent insertion, endoscopic vacuum therapy enables regularly viewing of wound conditions and adjustment of therapy. EVT can be used as a definite therapy or a supportive measure combined with surgical revision. In our department, EVT has become the treatment of choice for perforation and anastomotic leakage of the upper gastrointestinal tract. Therefore, we would like to present the outcome of our first 21 patients treated with EVT. In addition, we propose our treatment algorithm including combination of EVT and surgery. Compared to other series, we have combined endoscopic and surgical therapies in a relevant proportion of patients.
Patients and Methods All patients treated with EVT in the upper GI tract between March 2011 and March 2015 were prospectively documented and included into this report. There was an indication for the EVT either for anastomotic leakage or perforation. Endoscopic evaluation was mandatory in all cases. Additional CT scans were usually done in all cases with a manifest mediastinitis either before or after sponge placement. Contraindication for sponge placement was considered a circular necrosis of the gastric tube after esophagectomy. All sponges were placed with sufficient anesthesia or sedation in the operating room, ICU, or the endoscopy unit. In case of expected or manifest mediastinitis, the procedure was done with an anesthesiologist taking responsibility for the patient’s vital status.
Fig. 1 EVT with intracavitary (a) and intraluminal (b) sponge placement
J Gastrointest Surg (2016) 20:237–243
Until fall of 2014, polyurethane sponges of the BEndosponge^ system (B. Braun, Melsungen, Germany) originally designed for the rectum had to be modified for use in the upper GI tract. This was done by shortening and tailoring the sponge in most cases. After removing the original tube, the sponge was fixed to the distal end of a nasogastric tube by at least two sutures at either end of the sponge. On the distal end, an additional loop was attached for easier pulling the sponge. The sponge was grasped with endoscopic forceps and either pushed or pulled in the right place intracavitary or intraluminal (Fig. 1a, b). Whenever possible, the sponge was placed extraluminally within the extramural cavity. Depending on the extent of defects and progress of healing, sponge placement could be changed from initial intracavitary to intraluminal localisation, according to our treatment algorithm (Fig. 2). After sponge placement, the suction needs to be initiated under endoscopic control assuring thereby that the sponge is fixed in the proper position. In contrast to the pelvis or abdominal wall, EVT works at the esophagus intra- and extraluminally in a closed system with intrapleural negative pressure that partially adds to the vacuum. After insertion of the sponge and putting on the vacuum, the sponge-bearing nasogastric tube was channeled through the nose and connected again with an electronic vacuum pump system (V.A.C. by KCI®, Wiesbaden) to maintain a continuous negative pressure of 125 mmHg. Optionally—if the patient did not have a jejunal feeding tube already—an additional nasogastric tube was inserted under endoscopic control and channeled through the opposite nostril during the same session. A scheduled change of sponges took place every third day. A CT scan was only performed if there was evidence of sepsis suggesting a mediastinal or intrapleural focus, or if there was an unfavorable course of the patient during the first 24 h after sponge placement. During a median follow-up of 17 (range, 2–45) months, all patients were surveyed for food intake and general condition.
J Gastrointest Surg (2016) 20:237–243 Fig. 2 Treatment algorithm for surgical EVT of upper GI defects, University of Rostock
239 Anastomotic leakage
Esophageal perforation
Endoscopical examination
Necrosis or severe disseminated infection
Vital mucosa and localized infection
+/- Computed tomography
Sponge placement
EVT
Intraluminal – e.g. early stage after smaller iatrogenic lesion
EVT + Surgery
Intracavitary – e.g. anastomotic leakage with mediastinitis
in case of clinical deterioration: Computed tomography +/- endoscopy
Discontinuation of successful EVT depending on: size of cavity, granulation, amount of secretion
The technique was reviewed and registered as official treatment option at the ethics board of the University of Rostock. All patients or authorized persons, respectively, agreed to the treatment with written informed consent. Since November 2014, the new BEsosponge^ system (B. Braun, Melsungen, Germany) was used. In comparison to the modified rectal system, the sponge is smaller, and a long overtube is supplied for placement of the sponge.
Results Fifteen male and six female patients with a median age of 72 (range, 49–80) years were treated with EVT. In total, 126 polyurethane sponges were inserted in these 21 patients with different upper gastrointestinal defects, comprising anastomotic leaks, spontaneous (Boerhaave syndrome) as well as iatrogenic perforation. Mean distance of lesions (perforations and anastomotic leaks) treated with EVT was 32 cm (range, 18–44) from dental arch. Mean size of perforations was 5.4 cm (range, 1–17). Type and localization of perforations are listed in Table 1. Six out of ten esophageal perforations were detected within the first 24 h. Anastomotic leaks were detected at a mean of eight postoperative days (range, 1–18). Successful healing of the various defects was achieved in 19 of 21 patients (90.5 %). Median duration of EVT was 15 (range, 3–46) days with an average number of five (range, 1–14) sponge insertions per patient. The median changing interval of sponges was three (range, 2–4) days (Table 2).
Initial sponges were placed intraluminally in 11 and intracavitary in ten patients. In case of perforation, successful EVT was achieved in ten of ten patients. Duration of therapy was significantly shorter in patients with an iatrogenic perforation. The different success rates and parameters of EVT for perforation and anastomotic leakage are shown in Table 3. Nine of 21 (43 %) patients received EVT in combination with surgical revision. Seven of nine combined therapies were conducted in patients after anastomotic leakage. Indications for operative revision were pleural empyema in three patients. In three patients, a partial anastomotic necrosis was excised, and oversewn and intraluminal EVT was initiated. In two
Table 1
Type and localization (size) of esophageal perforation
Patient (n=10)
Type of perforation
Localization of perforation (distance from dental arch [cm])
1 2 3 4 5 6 7 8 9 10
Iatrogenic Iatrogenic Iatrogenic Iatrogenic Iatrogenic Boerhaave Iatrogenic Boerhaave Iatrogenic Traumatic
18–20 29–30 23–40 35–42 18–22 37–40 20–25 36–40 35–41 18–23
240 Table 2
J Gastrointest Surg (2016) 20:237–243 Results of EVT in the upper GI tract (n=21)
Table 4 EVT
Incidence of mediastinitis, empyema, and complications after
Gender, n (%)
6 (30 %) female 15 (70 %) male
Mediastinitis
9/21 (43 %)
Age in years (range) Duration of therapy in days (range)
72 (49–80) 15 (3–46)
Pleural empyema Discontinuation of therapy
3/21 (14 %) 2/21 (9.5%)
Follow-up in months (range) Total number of sponges
17 (2–45) 126
Sponge insertions per Patient (range)
5 (1–14)
Changing interval of sponges in days (range)
3 (2–4)
Multi-organ failure
Total Anastomotic leakage Perforation Iatrogenic Boerhaave
1
Stenosis
patients with total anastomotic breakdown, the anastomosis was redone, and an intraluminal sponge was placed additionally. Another operative revision took place after esophageal perforation following cervical tumor operation. Successful therapy of the perforation was achieved via intraluminal EVT in this very proximal position (18–20 cm from dental arch). EVT failed in two patients; in one patient, discontinuity resection due to anastomotic necrosis and severe mediastinitis after esophagectomy was inevitable. EVT was initiated 5 days before, in retrospect a wrong indication. Another patient died due to fulminant sepsis following anastomotic leakage after esophagectomy for carcinoma of the esophago-gastric junction. At postoperative day 7, operative revision had to take place due to a partially necrotic anastomosis that was excised and oversewn before intraluminal EVT was started. Because of further deterioration of the patient, a second redo-operation with pleural lavage and sponge replacement was conducted the next day. Three more changes of the intraluminal vacuum sponge were performed with local but not systemic control of septic progress. The patient died of multi-organ failure 17 days after the initial operation and 10 days after beginning of salvage therapy. During a median follow-up of 17 (range, 4–45) months, stenosis with dysphagia after EVT was detected in one of 21 (5 %) patients (Table 3). A 64-year-old patient who was treated with EVT for anastomotic leakage after esophageal resection developed a relative anastomotic stenosis. The stenosis was successful treated by endoscopic balloon dilatation (three times). Another patient developed a stenosis after colliquative necrosis and esophageal perforation. However, manifestation of stenosis in this case was most probably arising from the Table 3 Parameters and success rates of EVT for perforation and anastomotic leaks
1
Discontinuity resection Complications (long-term)
1/21 (5 %)
extensive colliquative necrosis and could not be ascribed to the EVT. The 55-year-old man mistook a bottle of convectionoven cleaner for a bottle of white wine and accidentally ingested a large swig of the alkaline solution. After 1 week of conservative treatment, the patient was experiencing increasing dysphagia. Endoscopic dilation was performed but led to a large esophageal perforation 23–40 cm from the dental arch. We immediately commenced EVT with extraluminal and intraluminal sponge placement. Following this, the patient underwent ongoing therapy with both esophageal bougienage and the use of intraluminal and extraluminal sponge placement. EVT was continued for 18 days with changes being made at intervals of 3 days. Further bougienage was required because of persisting esophageal stenosis. At the time of discharge, the patient was managing oral food intake and reported a good quality of life.
Discussion Esophageal perforation or anastomotic leakage after esophageal or gastric resection is accompanied with a mortality rate up to 38 %.1–15 Bartels and Siewert defined treatment principles as follows: first, goal is to drain the leakage and prevent further contamination. The second goal is to feed the patient enterally and to preserve continuity of the upper GI tract.16 Physically, there is respiration-dependent suction of fluid into the mediastinum after manifestation of an esophageal leak, representing a perfect environment for bacterial growth inevitably leading to severe mediastinitis. Placing a stent in order to close a leakage leaves medium and bacteria in the mediastinum. The infectious focus itself
Number of patients
Duration of therapy (day)
Number of sponges
Success rate
21 11 10 8 2
15 (3–46) 12 (6–27) 15 (3–46) 14 (3–19) 40 (34–46)
5 (1–14) 4 (2–7) 5 (1–14) 5 (1–7) 14 (13–14)
19/21 (90.5%) 9/11 (82 %) 10/10 (100 %)
J Gastrointest Surg (2016) 20:237–243 Table 5 Published data of EVT for upper GI defects (n=107)
241
Literature
Number of patients
Duration of therapy (day)
Success rate (%)
Kuehn et al., J Gastrointest Surg17 Bludau et al., Surg Endosc 201318 Brangewitz et al., Endoscopy19 Schniewind et al., Surg Endosc15 Schorsch et al., Chirurg 20
9 14 32 17 35
18 12 23 – 11
89 86 84 88 91
will not be eliminated. Therefore, antibiotic treatment and—in many cases—additional drainage via chest tubes might be mandatory. An operative approach may generalize a primarily well-confined infectious focus. In most cases, redo anastomosis is hardly possible, and the patient may finally end up with a cervical fistula after esophagectomy. Endoscopic therapy allows cleaning of the extramural cavity by lavage and placement of a polyurethane sponge into the cavity via the esophageal wall defect (Fig. 1a). Continuous suction and drainage via this open-pored sponge decrease bacterial contamination, secretion, and local edema. At the same time, perfusion of the esophageal wall and granulation is promoted. With EVT, a new system was introduced to manage mediastinitis and treat the patient endoscopically with very promising result. In total, there is published data of 107 patients with upper GI defects treated with EVT showing success rates from 84–91 % (Tables 4 and 5). However, there is no evidence-based recommendation on how to exactly treat a perforation or leakage in the upper GI tract with stent placement being the most common therapy. Schniedewind and colleagues15 performed an intention to treat analysis comparing results of stenting, operative revision, and EVT for anastomotic leakage after esophagus resection. In systemically ill patients matched for APACHE II scores, endoscopic endoluminal vacuum therapy patients had lower mortality (12 %) compared to surgically treated (50 %, p= 0.01) cases and patients managed by stent placement (83 %, p=0.0014). Brangewitz et al. compared the outcome of 32 patients after EVT and 39 patients after stent insertion in a retrospective analysis.19 The overall closure rate of leakages was significantly higher in the EVT group than in the stent group (84.4 % vs. 53.8 %). Additionally, they found significantly more strictures in the stent group (28.2 vs. 9.4 %). Our own treatment algorithm17 evolved during the first year of sponge application and currently resembles the description of Schorsch et al.20 We only perform a CT scan, if there is clinical evidence of mediastinitis or pleural empyema, or if the patient takes an unfavorable course in the first 24 h after sponge placement. The success of EVT is reflected by the size of the extramural cavity, accumulation of secretion, and
formation of granulation tissue (Figure 2). An endoscopic reevaluation is performed 2–3 days after cessation of sponge therapy. Especially in complex individual cases with esophageal wall destruction, EVT can be an alternative to surgery avoiding esophageal resection. As described above, a 17-cm long perforation by dilatation following corrosive esophageal injury could be managed successfully by EVT ultimately avoiding resection and the establishment of a cervical fistula.21 In contrast to other published patient series, we combined EVT with surgery in a relevant proportion of our patients. Nine of 21 (43 %) patients received EVT to control esophageal defects that could not be closed by suture due to extent or late presentation. In these cases, the esophageal wall defect was controlled by an intraluminal sponge whereas the concomitant mediastinitis was dealt with by established surgical concepts. In these cases, the sponge therapy resulted in salvage of the esophagus in seven of nine (78 %) patients. From our experience, EVT and surgery should be seen as complementary procedures. Mediastinitis and pleural empyema were the main indications for operative therapy as inevitable adjunct procedure to sponge therapy in these individual cases. Next to endoscopic experience and skills, surgical expertise is required to assess the extent of perforation or anastomotic conditions in order to adapt therapy appropriately.
Conclusion Our single-center experience confirms the initially promising results of EVT for control of various upper GI defects that have been reported so far. EVT completes the armamentarium in the treatment at even advanced stages with extended esophageal wall defect, severe local infection, and complex individual situations in which a stent insertion would be impossible. EVT can serve as an isolated therapy or as part of a more complex concept of esophageal salvage combined with conventional surgery for the control of mediastinal or pleural sepsis. Still, a multi-center study is missing to proof the definite benefit of EVT for patients with esophageal defects.
242
J Gastrointest Surg (2016) 20:237–243
Compliance with ethical standards The study was reviewed and registered as official treatment option at the ethics board of the University of Rostock. All patients or authorized persons, respectively, agreed to the treatment with written informed consent.
17.
18.
19.
References 1.
2.
3.
4. 5. 6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Sung SW, Park JJ, Kim YT, Kim JH. Surgery in thoracic esophageal perforation: primary repair is feasible. Dis Esophagus. 2002; 15: 204–9. Port JL, Kent MS, Korst RJ, Bacchetta M, Altorki NK. Thoracic esophageal perforations: a decade of experience. Ann Thorac Surg. 2003; 75: 1071–4. Brinster CJ, Singhal S, Lee L, Marshall MB, Kaiser LR, Kucharczuk JC. Evolving options in the management of esophageal perforation. Ann Thorac Surg. 2004; 77: 1475–83. Gupta NM, Kaman L. Personal management of 57 consecutive patients with esophageal perforation. Am J Surg. 2004; 187: 58–63. Braghetto I, Rodríguez A, Csendes A, Korn O. An update on esophageal perforation. Rev Med Chil. 2005; 133: 1233–41. Erdogan A, Gurses G, Keskin H, Demircan A. he sealing effect of a fibrin tissue patch on the esophageal perforation area in primary repair. World J Surg. 2007; 31: 2199–203. Eroglu A, Turkyilmaz A, Aydin Y, Yekeler E, Karaoglanoglu N. Current management of esophageal perforation: 20 years experience. Dis Esophagus. 2009; 22: 374–80. Griffiths EA, Yap N, Poulter J, Hendrickse MT, Khurshid M. Thirty-four cases of esophageal perforation: the experience of a district general hospital in the UK. Dis Esophagus. 2009; 22: 616–25. Chak A, Singh R, Linden PA. Covered stents for the treatment of life-threatening cervical esophageal anastomotic leaks. J Thorac Cardiovasc Surg. 2011; 141: 843–4. Abbas G, Schuchert MJ, Pettiford BL, Pennathur A, Landreneau J, Landreneau J, Luketich JD, Landreneau RJ. Contemporaneous management of esophageal perforation. Surgery. 2009; 146: 749–55. Vallböhmer D, Hölscher AH, Hölscher M, Bludau M, Gutschow C, Stippel D, Bollschweiler E, Schröder W. Options in the management of esophageal perforation: analysis over a 12-year period. Dis Esophagus. 2010; 23: 185–90. Hölscher AH, Vallböhmer D, Brabender J. The prevention and management of perioperative complications. Best Pract Res Clin Gastroenterol. 2006; 20: 907–23. Liu JF, Wang QZ, Ping YM, ZhangYD. Complications After Esophagectomy for Cancer: 53-Year Experience with 20,796 Patients. World J Surg. 2008; 32: 395–400 Schweigert M, Beattie R, Solymosi N, Booth K, Dubecz A, Muir A, Moskorz K, Stadlhuber RJ, Ofner D, McGuigan J, Stein HJ. Endoscopic stent insertion versus primary operative management for spontaneous rupture of the esophagus (Boerhaave syndrome): an international study comparing the outcome. Am Surg. 2013; 79: 634–40. Schniewind B, Schafmayer C, Voehrs G, Egberts J, von Schoenfels W, Rose T, Kurdow R, Arlt A, Ellrichmann M, Jürgensen C, Schreiber S, Becker T, Hampe J. Endoscopic endoluminal vacuum therapy is superior to other regimens in managing anastomotic leakage after esophagectomy: a comparative retrospective study. Surg Endosc. 2013; 27: 3883–3890 Bartels H, Siewert JR. Therapie der Mediastinitis am Beispiel des Ösophaguskarzinoms. Chirurg. 2008; 79: 30–7.
20. 21.
Kuehn F, Schiffmann L, Rau BM, Klar E. Surgical endoscopic vacuum therapy for anastomotic leakage and perforation of the upper gastrointestinal tract. J Gastrointest Surg. 2012; 16: 2145–50. Bludau M, Hölscher AH, Herbold T, Leers JM, Gutschow C, Fuchs H, Schröder W. Management of upper intestinal leaks using an endoscopic vacuum-assisted closure system (E-VAC). Surg Endosc. 2014; 28: 896–901. Brangewitz M, Voigtländer T, Helfritz FA, Lankisch TO, Winkler M, Klempnauer J, Manns MP, Schneider AS, Wedemeyer J. Endoscopic closure of esophageal intrathoracic leaks: stent versus endoscopic vacuum-assisted closure, a retrospective analysis. Endoscopy. 2013; 45: 433–438 Schorsch T, Müller C, Loske G. Endoscopic vacuum therapy and anastomotic insufficiency of the esophagus.. Chirurg. 2014; 85: 1081–93. Kuehn F, Klar E, Schwandner F, Alsfasser G, Gock M, Schiffmann L. Endoscopic continuity-preserving therapy for esophageal stenosis and perforation following colliquative necrosis. Endoscopy. 2014; 46 Suppl 1 UCTN: E361-2.
Primary Discussant Vic Velanovich, MD (Tampa, FL) Perforations and anastomotic leaks of the foregut are vexing problems. Up until recently, our only options were either external drainage and wait for the perforation to heal or surgery for an attempt at primary repair. Both have not always had satisfactory results. The conservative approach, if it worked, would take a long time, and if it did not, sepsis would continue. Surgery, if it worked, still required an operation with a high complication rate, and if it did not, the repair would fall apart. Recently, endoluminal approaches have increased our armamentarium in treating leaks and perforations. These include stents, clips, glues, and endoscopic suturing devices. One of these new approaches is endoscopic vacuum therapy, what I like to call endo-vacuum-assisted closure (VAC) therapy. The authors present a series of 21 patients treated over 4 years with excellent results. I have personally used this therapy on a few of my own patients and have been very impressed. I think that this therapy can be a game-changer. Combining endo-VAC therapies with and without external percutaneous drainage may become the standard of care for esophageal perforations and leaks. I have three questions: 1. Are there any perforations or disruptions which are too big for endo-VAC therapy? 2. How do you decided when endo-VAC therapy has failed and another approach is needed? 3. There are reports of combining endo-VAC therapy with stenting. Do you have any experience or opinion with this approach? Congratulations on an excellent study. I look forward to future work from your group. Closing Discussant Dr. Kuehn Dear Dr. Velanovich, Thank you very much for the purposeful summary of our work and the important questions that you have addressed. Our opinions follow point by point: 1. Are there any perforations or disruptions which are too big for endo-VAC therapy? From our experience, we cannot say that there are any limitations for EVT concerning size of defect. For example, a 55-year-old developed a coagulative necrosis after ingestion of an alkaline solution. After 1 week of conservative treatment, the patient was experiencing increasing dysphagia. Endoscopic dilation was performed but led to a large esophageal perforation 23–40 cm from the dental arch. We immediately commenced EVT with extraluminal and intraluminal sponge placement. Up to four sponges were placed at the same time. Following this, the patient underwent ongoing therapy with
J Gastrointest Surg (2016) 20:237–243 both esophageal bougienage and the use of intraluminal and extraluminal sponge placement. EVT was continued for 18 days with changes being made at intervals of 3 days. The 17-cm long perforation in a highly infectious environment was successfully healed with EVT. 2. How do you decided when endo-VAC therapy has failed and another approach is needed? We think that EVT cannot really fail if the indication is right. However, in some cases, therapy has to be extended. In contrast to other published series, we combined EVT with surgery in a relevant proportion of our patients. Nine of 21 (43 %) patients received EVT to control esophageal defects that could not be closed by suture due to extent or late presentation. In these cases, the esophageal wall defect was controlled by an intraluminal sponge whereas the concomitant mediastinitis was dealt with by established surgical concepts. In these cases, the sponge therapy resulted in salvage of the esophagus in seven of nine (78 %) patients. From our experience, EVT and surgery should be seen as complementary procedures. Mediastinitis and pleural empyema were the main indications for operative therapy as inevitable adjunct procedure
243 to sponge therapy in these individual cases. EVT can serve as an isolated therapy or as part of a more complex concept of esophageal salvage combined with conventional surgery for the control of mediastinal or pleural sepsis. 3. There are reports of combining endo-VAC therapy with stenting. Do you have any experience or opinion with this approach? Yes, there is one report in which two patients were treated successfully with EVT after stent placementx.20 Here, EVT was used as a rescue strategy for leakages that were refractory to stent therapy. They concluded that BIf stent therapy fails or the perianastomotic abscess cavity is large and complex to drain from outside, the endoscopic two-modality approach can be considered.^ However, since we do not stent these defects in the first place anymore, we do not have any experience with this approach. References: 1. Gubler C, Schneider PM, Bauerfeind P. Complex anastomotic leaks following esophageal resections: the new stent over sponge (SOS) approach. Dis Esophagus. 2013; 26: 598–602.