A Four-step Technique for Effluent Diversion of Enteroatmospheric Fistulas

Background. Isolation of the enteroatmospheric fistula (EAF) opening and prevention of contamination of the rest of the wound by effluent are important factors in the management of EAF. Objective. The aim of this study is to describe an easily reproducible technique for effluent control in patients with EAF. Materials and Methods. A retrospective analysis was conducted on all patients who underwent the present technique between 2013 and 2015. The surgical technique included condom-EAF anastomosis, fistula ring creation, negative pressure wound therapy (NPWT), and adaptation of an ostomy bag. Results. A total of 7 patients with a Björck grade 4 abdomen were included. All fistulas were located in the small bowel with a median number of 2 EAFs (range, 2–3) in each patient, and the majority had moderate output volume. The mean number of NPWT changes was 10 (range, 5–18), the mean time of NPWT use was 75.7 days (range, 60–120 days), and the mean length of stay was 108.2 days (range, 103–160 days). The mean time of ostomy formation to restitution of bowel continuity was 14.3 months (range, 8–20 months). Open anterior component separation was employed in all cases for closure of the abdominal wall. No mortality, ventral herniation, or refistulization was registered in the study. The mean follow-up time was 8.5 months (range, 6–12 months). Conclusions. This is an easily reproducible and safe technique for effluent control in patients with Björk grade 4 abdomen with established EAF.

Enteroatmospheric fistula (EAF) is defined as the occurrence of an enteric fistula in the middle of an open abdomen (OA), consequently creating a communication between the gastrointestinal tract and external atmosphere.¹ Unique features that define EAF are the abscess of a fistula tract, lack of well-vascularized surrounding tissue, and location within an OA resulting in spillage of enteric content into the peritoneal cavity.¹  

Some of the etiologic factors associated with EAF are anastomotic leak, severe wound infection, burst abdomen, severe trauma, bowel ischemia, missed enterotomies, use of OA techniques (damage control surgery), and inflammatory bowel disease.^1,2 Open abdomen techniques and methods for temporary abdominal wall closure have been associated with an EAF development rate of 14% to 25%.^2,3 A higher incidence of EAF in septic OA compared with nonseptic OA (12.1% vs. 3.7%) has been reported.¹ Fistula formation rate related to negative pressure wound therapy (NPWT) can be as low as 5%, although Giudicelli et al² suggest the relationship between NPWT and EAF could be as high as 14.6%. The mortality rate due to EAF is reported to be between 36% and 64%.^1-5 In addition, the development of EAF increases the hospital length of stay (LOS) by 4-fold and hospital costs by 4- to 5-fold.⁶ 

There are several treatment options for EAF, reflecting the lack of uniform data, evidence-based results, and standard technique.^1-4 The isolation of the EAF opening and the prevention of the contamination of the rest of the wound by fistula effluent are important factors in the management of this entity.⁴ 

The aim of this study is to describe a novel, easily reproducible 4-step technique for effluent control in patients with EAF and the surgical outcomes of a series of patients treated with this method. 

The investigators conducted a retrospective analysis of all patients who underwent this 4-step technique for effluent control of EAF between January 2013 and December 2015 at the Hospital General Dr. Manuel Gea González (Mexico City, Mexico). Patients with incomplete data were excluded. This study was conducted in accordance with local audit, ethics, and governance protocols. 

Data of patients regarding age, sex, and indication of the first emergency surgery and OA were analyzed. All EAFs recorded in this study were new fistulas that occurred following an emergency laparotomy performed in the authors’ hospital. All EAFs included in this study had protruding visible mucosa. Enteroatmospheric fistula was defined as a pathological opening in the intestinal lumen directly into the atmosphere or an exposed fistula occurring in an OA with no overlying soft tissue.^1,2,5 

An initial stabilization phase including reanimation with fluids, sepsis control, electrolyte imbalance, and severe anemia correction was applied to all patients. A stepwise multidisciplinary team approach, including a nutritionist, infectious disease specialist, and ostomy specialist, were involved in patient care. 

Based on previously described nutritional support interventions,^2,5-8 oral or enteral nutritional support was the preferred route of feeding. In patients with proximal fistulas, the authors introduced a feeding tube through it to provide enteral nutrition. Parenteral nutrition was employed in all patients in order to cover their energetic needs. Nutritional support followed the recommendations of the American Society for Parenteral and Enteral Nutrition⁷ and the European Society for Parenteral and Enteral Nutrition.⁸

Effluent control was achieved with the surgical technique described in the next subsection. The purpose of this technique is to transform the EAF into a floating stoma and, after several dressing changes, transform this floating stoma into a “true” stoma (“ostomization of the fistula orifice”). In most instances, an elective operation for closure or resection and anastomosis was performed after correcting malnutrition status and maintaining well-balanced metabolic parameters as required. Nutritional status after nutritional support intervention (enteral and parenteral) was evaluated by clinical, biochemical (albumin, prealbumin, and transferrin), and anthropometric measurements (body mass index).^2,5-8 

The EAF variables included the number of fistulas, output volume, anatomic site of fistula (proximal or distal), and location inside the OA (either superficial or deep) and were described in each patient. Effluent characteristics were evaluated and classified in low (< 200 mL/day), moderate (200–500 mL/day), or high (> 500 mL/day) output volume.⁵ The standard OA classification as proposed by Björck et al⁹ was employed as well as the newest classification by Di Saverio et al.¹ 

Surgical characteristics consisted of duration of OA with NPWT, number of dressing changes, time to fistula resection, and type of abdominal wall closure. In addition, the time in intensive care unit (ICU; days), overall hospital LOS, and 6-month mortality were evaluated. Rates of refistulization after the last surgery were included. 

Data are presented as mean or median (minimum and maximum values, range) and presented. All data were analyzed using SPSS, version 22.0 (IBM Corp, Armonk, NY).

Surgical technique
The aim of this 4-step technique is to control fistula drainage and protect the wound and surrounding skin while creating a floating stoma. At the final stage, the floating stoma is converted into a true stoma (“ostomization of the fistula”). This stoma facilitates subsequent management of the effluent and restores bowel continuity.

For patients with more than 1 EAF, the technique was applied to the fistula with the higher output, more proximal location, and largest fistula orifice (with protruded mucosa). The more distal, smaller, and lower output fistulas without protruding mucosa were controlled with primary suture closure and NPWT after covering with a patch of hydrophilic polyvinyl alcohol foam (V.A.C. WHITEFOAM Dressing; KCI, an Acelity Company, San Antonio, TX). For patients with more than 1 high-output and protruded mucosa fistula, a separate condom-fistula ring appliance was constructed for each EAF with the rest of the abdominal wound dressed with black NPWT foam (V.A.C. GRANUFOAM Dressing; KCI, an Acelity Company), thus bridging and connecting the negative pressure.

After completely debriding the wound, the 4-step surgical technique includes the following: 

Step 1: Condom-EAF anastomosis (Figure 1). A condom (made of latex rubber) cut by the reservoir side according to the fistula size was utilized. The condom was anastomosed to the EAF opening with a running 4-0 nonabsorbable (polypropylene) suture, full-thickness bite. This is an important part of the procedure because the effluent will be diverted through the condom to the ostomy bag.

Step 2: Fistula ring creation (Figure 2). A fistula ring technique, as described by Verhaalen et al,¹⁰ was performed. The fistula isolation “ring” was constructed with the black NPWT foam cut into a circular shape. The foam is completely encompassed with a drape (V.A.C. Drape; KCI, an Acelity Company). The fistula ring was placed surrounding the EAF and the condom. 

Step 3: NPWT (Figure 3). After the condom was anastomosed to the fistula opening and isolated with the impermeable barrier provided by the fistula ring, the rest of the laparostoma was dressed applying black foam encompassed with the NPWT drape or dressed with the white foam (hydrophilic polyvinyl alcohol foam). Then, the occlusive drape was applied over the entire dressing area, with an orifice for the fistula ring and condom. Next, V.A.C. SensaT.R.A.C. (KCI, an Acelity Company) was placed onto the foam, away from the condom, with continuous negative pressure set at -125 mm Hg to -150 mm Hg. 

Step 4: Ostomy bag (Figure 4). Once isolation of the effluent was achieved, a drainage or ostomy collection bag was placed over the ring. The condom was pulled through to the ostomy bag. For high-output EAFs, the authors introduced the tip of a flexible suction tube to the ostomy bag (secured with a suture or an adhesive tape) to recollect the effluent, facilitating quantification and avoiding premature dysfunction of the NPWT device. 

The NPWT was changed on demand based on dysfunction of the system by obstruction or leakage (or at least once weekly). The condom was changed whenever the union with the fistula leaked effluent to the wound. In every change of the system, the authors closed the edges of the wound (at the top and bottom of the skin defect and, whenever possible, the fascial defect) to reduce the area of the wound with every change. Dressing changes were performed under regional anesthesia and sedation in the operating room. After several dressing changes, the EAF was converted to a true stoma; at the final stage, this stoma was surrounded by skin (Figure 5). Afterwards, patients followed up in the office. When nutritional and inflammatory statuses were corrected, patients were scheduled to a definitive surgery with restitution of bowel continuity and abdominal wall closure/reconstruction. 

A total of 7 patients were included in the study. Table 1 presents the baseline characteristics, EAF features, and postoperative outcomes. All patients were previously healthy, with a mean age of 38.5 years (range, 23–67 years); 4 (57.1%) of the patients were male. The indications for primary surgery were abdominal sepsis in 5 patients (71.4%) and abdominal trauma in 2 (28.6%). In all patients, OA treatment was necessary after a complication of the index surgery (Table 1). Complications included anastomotic leaks (n = 4, 57.1%) and missed enterotomies (n = 3, 42.9%); all patients developed severe intra-abdominal sepsis (severe peritonitis). The indication for OA was the need for relaparotomy for lavage and drainage of collections/abscesses. The EAF developed as a complication of the OA in all patients. 

All 7 patients were classified as Björck grade 4 (frozen OA with adherence/fixed bowel, impossible to close surgically, and with established fistula).⁹ A total of 16 EAFs were registered in this series; all fistulas were located in the small bowel (proximal as per Di Saverio et al¹ classification) with a median number of 2 EAFs per patient (range, 2–3) and moderate output volume as the most frequent presentation (as per Di Saverio et al¹ classification). All EAF were located superficially (ie, draining on top of a granulating abdominal wound).¹ The 4-step technique was performed successfully in all 7 patients. Patients 5 and 7 required 2 separated condom-fistula ring appliances as previously described (Table 1). The mean number of NPWT changes was 10 (range, 5–18), mean time of NPWT use was 75.7 days (range, 60–120 days), and mean hospital LOS was 108.2 days (range, 103–160 days). No complications or new fistulas related to NPWT were recorded. 

All patients included in this study underwent restitution of bowel continuity, with bowel resection (enterectomy) and primary anastomosis performed. Although the exact distance between each fistula was not recorded, the diseased bowel containing all active fistulas at the moment of restitution of bowel continuity was resected in all cases. The mean time of ostomy formation to restitution of bowel continuity was 14.3 months (range, 8–20 months). Open anterior component separation was employed in all cases for closure of the abdominal wall. No complications after restitution were found. No cases of short bowel were recorded. No mortality, ventral herniation, or refistulization was registered during the mean follow-up time of 8.5 months (range, 6–12 months).

The OA complicated with EAFs results in a complex wound with persistent inflammation of the surrounding granulation tissue or skin because of persistent soiling and chemical irritation by intestinal content, local infection, sepsis, electrolyte imbalance, and nutritional depletion.¹¹

With the advent of damage control surgery for the treatment of abdominal trauma and sepsis, OA has emerged as a common treatment modality. Another indication of OA is prevention or treatment of intra-abdominal hypertension/compartment syndrome. Although OA has its role in acute care surgery, it is associated with significant complications, including fluid and protein loss, loss of bowel function, loss of abdominal wall domain, prolonged hospital LOS, and, the most feared complication, fistula formation.⁶ 

The fistulogenic nature of the OA was recognized due to increasing rates of incidence; it is thought to be a consequence of the continuous exposition of the bowel to desiccation and frequent dressing changes.¹¹ Spontaneous closure of this type of fistula (EAF, high-output, and with protruding mucosa) is unlikely,³ thus controlling the fistula effluent and protecting the wound and skin are the biggest challenges in managing these patients.¹¹

Fistula management should aim to completely divert the fistula output.¹ Perhaps one of the most important advances in managing EAF is the introduction of NPWT as an adjunct for managing effluent and wound complications. Early case reports and small case series have been published since 2000 on the treatment of fistulas with NPWT.^12-16 These reports recognized the advantage of using NPWT in enterocutaneous fistula but not in the context of OA. D’Hondt et al¹¹ used NPWT to collect the effluent, protect the surrounding skin, and promote granulation tissue formation. Transferring this knowledge to the OA complicated with EAF (once considered a contraindication) has been slow but with increasing evidence that NPWT can be employed successfully in this scenario.

In addition, NPWT not only provided effluent control but also contributed to increased granulation, wound contraction, microenvironment control, and accelerated wound healing.⁴ Some of the previously described techniques for effluent diversion and their disadvantages that differ from the present technique include NPWT with handmade dressings (does not provide effective fistula effluent diversion)¹; floating stoma(very traumatic and technically difficult)¹⁷; fistula NPWT (risk of pooling of effluent under the sponge, area around the fistula exposed to bowel content)¹⁸; tube NPWT (risk of enlargement of fistula, not applicable to fistulas larger than the catheter)¹⁹; NPWT chimney (risk of pooling of effluent under the sponge, complex and time consuming)²⁰; baby bottle nipple NPWT (small fistula only)²¹; fistula patch (not tested in peritonitis, must remain until fistula takedown)²²; fistula suspension (technically difficult, only if adjacent to well-vascularized dermis)²³; biologic dressing (expensive, not applicable if sepsis)¹; and fistula plug (complex device, difficult care, risk of obstruction).¹ 

Some advantages of the present technique (4 steps with condom) that differ from the others are that it is a quick, easy technical application; has a low cost; has accessible materials; is applicable with ongoing sepsis or peritonitis; provides effective fistula effluent diversion and easy quantification; has a good seal; avoids cannulization of the orifice; is applicable to any size of EAF orifice (especially with large protruding mucosa); allows progressive abdominal wall closure; protection of spillage of enteric contents under the sponge to avoiding early dysfunction of NPWT; and is feasible in cases of multiple fistulas. The drawbacks of the 4-step technique include, in some cases, trauma due to laceration of the bowel with the sutures and risk of enlargement of the fistula. This technique is unique because it combines the benefits of abdominal NPWT plus the isolation of the fistula with the condom-fistula ring complex, providing excellent outcomes regarding effluent control and wound healing and transforming EAF into a more manageable ostomy. 

Results of different series treating OA complicated by EAF with NPWT are listed in Table 2.^{5,11,17,18,22,24} In comparison with previously published work, this study adds the rate of refistulization and rate of ventral herniation during follow-up to the literature, which the authors consider an important variable in these cases. 

In 2011, D’Hondt et al¹¹ proposed the treatment of small bowel fistulas in the OA with topical NPWT. They covered the abdomen with 3 layers of nonadherent paraffin gauze and the entire wound was covered with black NPWT foam and received -125 mm Hg continuous pressure. Small fistulas without protruding mucosa were covered with polyvinyl alcohol foam; a hole was cut in the black foam for larger fistulas with protruding mucosa. The dressings were changed every 4 days.¹¹  

One of the most recent large case series was published in 2017 by Bobkiewicz et al.⁵ This series included 10 patients with 12 EAFs with protruding mucosa that were treated with a Foley catheter placed directly into the intestine lumen; in cases with moderate- or high-output fistulas, they implemented the fistula NPWT technique.⁵ 

Also, in 2017, Yetisir et al²⁴ reported a new technique called “fistula isolation by suturing the penrose drain to mucosa of orifice of EAF”; after suturing the drain, they placed NPWT foam and administered NPWT at -75 mm Hg. They achieved isolation of the fistula with Adapt Skin Barrier Paste (Hollister, Libertyville, IL).²⁴

Regarding the optimal time to close or resect the fistula and repair the abdominal wall defect, Demetriades²⁵ reported that ideally it should be performed 4 to 6 months after treatment in patients with an EAF, but a later publication¹ reported that fistula takedown and abdominal wall reconstruction should be delayed optimally by at least 8 to 12 months. 

This study’s limitations are largely attributed to its small sample size and retrospective nature. Unfortunately, it is very difficult to perform either a case-matched retrospective study or randomized controlled trial due to the complexity of this condition. Also, important to note, this technique is an off-label use of the NPWT manufacturer’s indications. Despite this, the authors consider this technique a feasible, safe, and effective treatment for patients with EAFs with good functional results. 

This study provides a new, safe, easily reproducible, and effective technique to the armamentarium against the OA complicated with EAF. The present technique demonstrates the basis for excellent effluent control, avoids septic complications from persistent intestinal soiling, allows clean granulation tissue formation, protects the skin, increases the rate of wound healing, transforms a difficult-to-treat fistula orifice into a well-controlled fistula simulating an ostomy, and facilitates future restitution of bowel continuity and abdominal wall construction.

Authors: Adolfo Cuendis-Velázquez, MD; Mario Trejo-Avila, MD; Elisafat Arce-Liévano, MD; Eduardo Cárdenas-Lailson, MD; Carlos Sanjuan-Martínez, MD; and Mucio Moreno-Portillo, MD

Affiliation: Department of General and Endoscopic Surgery, Hospital General Dr. Manuel Gea González, Mexico City, Mexico

Correspondence: Mario Trejo-Avila, MD, Department of General and Endoscopic Surgery, Hospital General Dr. Manuel Gea González, Calzada de Tlalpan 4800, Mexico City 14090 Mexico; mario.trejo.avila@gmail.com 

Disclosure: The authors disclose no financial or other conflicts of interest.