Combined Endoscopic Negative Pressure and Surgical Treatment of Anastomotic Insu�ciencies following Oncological Gastrectomy – Recent Results

Background and study aims: Management of esophago-jejunal anastomotic leakages (EJAL) following gastrectomy is challenging. Endoscopic negative pressure therapy (ENPT) is an emerging effective tool for treatment of gastrointestinal and anastomotic leaks. We have been using ENPT as �rst line therapy for EJAL after oncological gastric resections since 2018. The aim of the study was to present our results with this strategy in a case series. Methods: Nine consecutive patients were treated with ENPT for EJAL after oncological gastrectomy between 01.2018 and 12.2019. A retrospective analysis of patients’ and treatment-related data was performed. Results: Time to leakage detection was 6.00 ± 2.49 days after surgery. After 14.78 ± 9.66 days of ENPT, 6.25 ± 3.65 endoscopies and 38.11 ± 16.46 days of hospitalization, endoscopic treatment with ENPT combined with surgical debridement and drainage for sepsis control was effective in eight of nine patients. In one patient with a complete anastomotic dehiscence, treatment was changed to a stent-based therapy combined with surgery. Conclusions: ENPT is a new and promising option in the complication management of patients with anastomotic insu�ciencies following oncological gastrectomy. It can be recommended in combination with limited surgery to preserve the anastomosis and provide sepsis control. The time interval to diagnosis and the size of the insu�ciency are important for the success of ENPT in patients with EJAL.


Background
Gastric cancer is one of the leading causes of cancer-related death in the world, and multimodal treatment combining perioperative chemotherapy with radical resection and D2 lymphadenectomy is the established curative treatment.For subtotal and complete gastrectomy, the digestive reconstruction is performed with a Roux-en-Y or jejunal interposition, both of which include anastomosis between the esophagus/stomach remnant and jejunum.Esophagojejunal anastomotic leakage (EJAL) is still one of the most serious complications with an EJAL incidence between 0.5% and 11.0% associated with a high mortality rate [1][2][3].
Vital granulation tissue is formed after wound cleaning.
For endoluminal ENPT, the open-pore drainage system is placed within the lumen of the digestive tract at the entrance to the extraluminal cavity.This method can be used when the defects and cavities are small [11,12] [11,12].ENPT success for esophageal leaks has been reported since 2010 [5][6][7][8][13][14][15].Following the promising results and the increased use of this method in Europe, encouraging reports on the practice of endoluminal ENPT have been published worldwide.For endoluminal ENPT we used a drain covered by an open-pore polyurethane sponge (OPD: open-pore polyurethane sponge drain) or a handmade openpore lm drain (OFD).Characteristics of both open-pore devices are listed in Table 1.
After implementation of an SOP at our center, ENPT with OPD or OFD in endoluminal position has been the primary endoscopic therapeutic option for EJAL since January 2018.Before endoluminal ENPT became the standard for complication management for EJAL, stenting with self-expandable metal stents (SEMS) was our tool for management of this complication.Stent dislocations and stent-related complications like pressure ulcer and bleeding were the reasons for changing our EJAL complication management.The aim of this study is to present our results with ENPT in patients with EJAL.

Study design:
The local ethics committee of Tübingen University Hospital, Germany, approved this study (AZ: 752/2019BO2).The study is registered with ClinicalTrials.govunder reference number NCT04362605.
The study is adheres to the PROCESS guidelines from 2016.
All EJAL patients treated primarily with ENPT between January 2018 and December 2019 were considered for the study.Ethics Approval and Consent to Participate whether informed consent was obtained from all participants.Patients' records as well as the database were analyzed for EJAL therapyspeci c items.EJAL is de ned according to the de nition of anastomotic insu ciencies following esophageal or cardial resection of 2018 [16].

Endoscopic negative pressure therapy (ENPT):
OPD: The commercially available Eso-SPONGE ® System (B.Braun Melsungen AG, Melsungen, Germany) was used for endoluminal vacuum therapy.For positioning the OPD in loop-technique a loop (Mersilene TM , Polyester, 4 Ph.Eur., Ethicon®, Norderstedt, Germany) is xed at the distal end of the drain, gripped with an endoscopic grasper and placed under endoscopic view.Endoscopic placement was performed via oral intubation of the esophagus and nally with oro-nasal redirection and xation with plasters.The OPD with placed suture loop is illustrated in Figure 1.
OFD: The OFD for endoluminal therapy is handmade, as previously described by G. Loske et al. [11], by wrapping a thin open-pore double-layered drainage lm (Suprasorb® CNP, Drainage Film; Lohmann & Rauscher International GmbH & Co.KG, Rengsdorf, Germany) around the distal end of a small caliber redon drain, a gastric tube or the gastric segment of a naso-jejunal feeding tube (Freka® Trelumina, Fresenius Kabi Deutschland GmbH, Bad Homburg, Germany).Sutures (Mersilene TM , Polyester, 4 Ph.Eur., Ethicon®, Norderstedt, Germany) were used for xation of the drainage lm around the tube.Drain insertion took place via nasal positioning and endoscopic guiding with a grasper.Venting tubes in tri-lumen enteral feeding tubes had to be closed for ENPT.The handmade OFD on an intestinal feeding tube is shown in Figure 2.
Controlled Negative Pressure: A continuous vacuum of -125 mmHg is generated with electronic vacuum devices (KCI V.A.C. Ulta or V.A.C. Freedom; KCI USA Inc., San Antonio, TX, USA).
Procedural information: In patients with suspected EJAL rst an index endoscopy was performed under general anesthesia with endotracheal intubation.Standard gastroscopes with an outer diameter of 9.8mm were used with carbon dioxide insu ation.
De nition of an anastomotic leak is based on the endoscopic nding at the esophago-jejunostomy according to the CAES classi cation [16].Extraluminal uids and gases are radiologically assessed.Decision criteria for placement of OPD or OFD were the defect size, the extension of extraluminal collection and the patient's clinical condition.A CT scan in patients with suspected EJAL was done in all cases before or after endoscopy.Extraluminal collections were drained via limited surgical procedure as part of the presented concept.
Re-endoscopy and ENPT device change were performed after three to seven days until resolution of the complication.In patients with OPD change interval was three to four days, while in patients treated with OFD the interval was longer, namely up to seven days.Success was de ned as complete closure of the perforation.
In patients with a compensated clinical condition and endoluminal OFD the swallowing of liquids is allowed.

Results
Nine consecutive patients (4 females and 5 males with a mean age of 60.67 ± 12.35 years) were treated with ENPT for EJAL.Patients' characteristics are presented in Table 2.
Mean time of EJAL diagnosis was day 6.00 ± 2.49 after surgery.Three patients were treated at the ICU at the time of diagnosis.Symptoms that led to a more in-depth diagnosis were respiratory insu ciency, conspicuous secretion via drains, fever, and elevated in ammatory markers.After diagnosis of EJAL and start of ENPT all patients were treated and observed on ICU.Index endoscopy ndings varied strongly.Circumscribed insu ciencies, large leakages with secretion of putrid uids or brin-coated anastomosis with exposed clamps were seen.Endoscopic ndings are classi ed according to the CAES classi cation [16].Figures 3 and 4 show examples of endoscopic ndings.
First treatment mode in all patients with EJAL was endoluminal ENPT.In ve patients ENPT was performed with OPD and in four patients with OFD.In one obese patient with a complete anastomotic rupture ENPT was changed to SEMS therapy.Table 3 shows an overview of treatment-related characteristics per EJAL patient according to CAES classi cation.Treatment characteristics are summarized in Table 4.
To address extraluminal collection and sepsis control surgical debridement and drainage were performed in eight of nine patients.Enteral feeding was established in all patients via nutrition tubes combined with ENPT through OFD in ve patients and via needle catheter jejunostomy in three patients.
In eight patients the combined treatment with ENPT and surgery was successful.
Later endoscopic intervention for post-treatment of a stenosis of the anastomosis was not necessary in the analyzed patients.

Discussion
Because of multimodal treatment of gastric cancer the majority of patients with EJAL after gastrectomy are critically ill.The analyzed patients had a prolonged postoperative course.All were treated on the ICU/IMC and required invasive ventilation or closer monitoring.
For management of a suspected EJAL an SOP has been established with a generous indication for endoscopy in any case of worsening of the clinical condition or suspicion of an anastomosis problem.In most cases endoscopy is done before the CT scan and allows immediate treatment for either ENPT placement or stent placement.
Several articles favor the primary use of self-expandable fully or partially covered metal stents or sealing the leak with clips in patients with EJAL [3,[17][18][19][20].Both techniques lead to a more or less closing of the defect without addressing the extraluminal infected focus or uid collection.In our experience the use of stents for this indication is critical and reserved for patients with failed ENPT.Only one of our patients needed a stent after failed ENPT.Advanced clip systems like the over the scope clip (OTSC) system are used to close fresh perforations of the gastrointestinal tract as well as anastomotic leakages, but good perfusion of the wound edges is a prerequisite.We have no experience with the OTSC in EJAL patients.
The value of this strategy is well documented with excellent results in patients who experienced insu ciencies of the staple line after sleeve gastrectomy [29,30].
In this article we focus on the primary endoscopic treatment in patients who suffered from EJAL exclusively after oncological gastrectomy.The advantages of ENPT are internal endoluminal drainage, stimulation of granulation of the surrounding tissue followed by size reduction of the wound and nally preservation of the anastomosis.Complications associated with the use of stents such as migration, damage to the digestive wall, postinterventional stenosis, low tolerability, and high rates of dysphagia can be prevented with ENPT.Hemorrhage due to negative pressure is a possible complication of ENPT; it is documented in some patients with intracavitary position of the sponge (OPD) [24].No bleeding complication occurred in patients treated with ENPT with open-pore devices in endoluminal position.
Possible disadvantages of EJAL are the limited possibility for enteral feeding and repeat endoscopies.
The former can be ensured with a combined feeding tube as well as the possibility to swallow liquids in cases with OFD.A re-endoscopy interval of three to ve days ensured good assessment of the anastomotic region and leak healing with the possibility to immediately change strategy in the case of treatment failure.
The decision for the open-pore device, the use of OPD or OFD, in endoluminal position is taken by the endoscopist together with the surgeon focusing on the endoscopic and radiological ndings and the patient's condition.Because of the very good cleaning properties and wound size reduction effect of OPD, this device is favored by us for large insu ciencies and dirty wound ground.
In further studies the questions for the best time and usage of the different open-pore devices have to be answered.
Additional surgery as re-laparotomy, re-laparoscopy or re-thoracoscopy depends on primary operation, endoscopic and radiological ndings.Surgery is needed to address the extraluminal infection focus.
We are aware of the limitations of this retrospective small case series, but to the best of our knowledge this is the rst case series of primary ENPT performed in EJAL patients exclusively after oncological gastrectomy.The excellent results with complete healing of the anastomosis in eight of nine treated patients have led us to change our treatment algorithm.ENPT is now the primary treatment option in patients with EJAL.

Conclusion
ENPT is a promising new tool for treating EJAL.Furthermore, comparative studies of ENPT and other endoscopic treatment options for EJAL are needed to determine the best management options and indications for combined surgery. Tables

Table 1 :
Characteristics of used open-pore devices for ENPT in EJAL patients

Table 3 :
Overview of treatment-related characteristics per EJAL patient according to CAES classi cation