Conversion of an Enterocutaneous Fistula Associated With an Open Abdominal Wound Into a Drain-Controlled Enterocutaneous Fistula

   Open abdominal wounds with enterocutaneous fistulae present health care providers and patients with complex wound management issues. Frequently, large wounds with exposed bowel are present and require the utilization of multiple wound care modalities to provide a method of control and isolation of fistula effluent to allow for maximum wound healing. This case study presents a unique approach to management of low-output enterocutaneous fistula with an open abdominal wound. Through the use of negative pressure wound therapy, a standard surgical drain, and optimized nutrition, fistula drainage was redirected and the abdominal wound healed, leaving a drain-controlled enterocutaneous fistula.

  Open abdominal wounds with associated enterocutaneous fistulae provide some of the most complicated wound management challenges for patients, physicians, and nurses.¹ In contrast to an enterocutaneous fistula not associated with a large open wound, an enterocutaneous fistula with an open abdominal wound provides unique challenges because of the additional concerns of open fascia and exposed bowel. These fistulae are frequently surrounded by adherent bowel in various stages of healing without an appropriate surface for placement of an appliance to control the fistula drainage. Control of fistula drainage and protection of surrounding tissue and skin is a principle of early fistula management.²   Options for wound care and fistula management have expanded as new techniques have been reported and health care teams with advanced experience and interest have developed. One of the greatest advances in recent years has been the innovative use of negative pressure wound therapy (NPWT) devices to aid in control and healing of complex abdominal wounds.^3,4 This case report describes a unique method to convert an enterocutaneous fistula into a manageable drain-controlled enterocutaneous fistula, allowing for separate abdominal wound healing using an NPWT device. 

  Patient “C” is a 25-year-old female who was admitted to Froedtert Hospital (Milwaukee, WI) with worsening headaches and vision loss from a known intraventricular tumor. She underwent a resection of an atypical central neurocytoma with concomitant laparoscopic replacement of a ventriculoperitoneal shunt. During this same hospitalization, she developed a strangulated bowel obstruction requiring multiple operations, externalization of the shunt, and 2 small bowel anastomoses prior to ultimate closure of the fascia of the abdomen. During this entire time, total parenteral nutrition (TPN) was administered and intermittent enteral feeding was attempted. However, the patient failed to tolerate enteral feeding as evidenced by increased abdominal distention and gastrointestinal ileus on an abdominal series. Therefore, a CT scan was performed, demonstrating a large pelvic hematoma. Drainage of this hematoma was attempted through interventional radiology, but failed. During this procedure, fluid was sampled and was found to be positive for bacteria. Ultimately, 4 weeks following her last abdominal operation, the patient was taken back to the operating room for open drainage of this infected hematoma. On postoperative day 1, it was noted there was enteric fluid draining from her midline incision. Upon immediate re-exploration, a small bowel fistula deep within the abdominal cavity was identified. The abdomen was washed out and fibrin glue was applied to the evident hole in the bowel. A Jackson-Pratt drain was placed near the intra-abdominal fistula site, and exited through the patient’s lateral abdominal wall. The abdominal fascia was closed, but the skin was left open. The next day, 200 cc total of thin enteric fluid was noted between the wound and the drain. At this point, the authors chose to utilize NPWT to redirect the enteric drainage away from the midline wound in an effort to optimize wound healing and control the fistula output.   Modifying the natural wound geography to separate fistula effluent form the wound bed required laying a second, flat 10 mm closed-suction drain below the dehisced fascia with exit through the lower aspect of the wound. The black GranuFoam sponge (KCI, San Antonio, TX) was then placed on top of the fascia, up to the skin level of the abdominal wall. The drape was placed on top of the sponge per standard application with adherence around the tubing of the second drain. Continuous suction was placed at 125 mm Hg (Figure 1). All enteric drainage was successfully rerouted into the 2 drains with no enteric effluent noted in the NPWT system or sponge. One vacuum-assisted closure dressing change was performed on postoperative day 6 using the same technique. Once a good bed of granulation was evident on postoperative day 9 (Figure 2), the wound was approximated with the drape; linear fenestrations were made into the drape along the midline of the approximated wound; a long thin strip of the sponge was placed over the fenestrated area, and a second piece of drape over the entirety. The T.R.A.C. Pad (KCI, San Antonio, TX) was placed per routine application in the middle of the sponge and suction was initiated at 125 mm Hg. Without any dressing change between postoperative days 9 and 14, the wound was almost completely healed on postoperative day 14, while the surgical drain continued to redirect enteric effluent with persistent volumes between 100 mL and 200 mL per day. The patient remained nil per os with full TPN support throughout this period. She was discharged to a long-term care facility 3 days after redirection of fistula effluent. On admission to the facility, her abdominal drains were inadvertently pulled without requiring replacement. She was evaluated in the Medical College of Wisconsin General Surgery Clinic on postoperative day 29, and found free of abdominal wound drainage. The fistula appeared healed (Figure 4). 

  Isolation of enteric fluid from an enteric fistula deep within the abdomen or a deep subcutaneous space can be a challenging clinical problem. The goal for these wounds is to keep the wound clean and separate from the enteric drainage that inhibits healing through increased inflammation and bacterial load.¹ Attempting to heal the subcutaneous tissue over a drainage device, such as a Jackson-Pratt surgical drain, was the authors’ method to heal the midline abdominal incision while providing a separate portal for draining the thin enteric effluent. This unique technique provided a successful strategy that improved wound management. Some advantages in this case that the authors feel optimized success were: low-output fistula; thin enteric output; optimized nutrition with high-protein TPN; visible fascial sutures that prevented the sponge from making direct contact on the bowel; a deep subcutaneous layer that provided a large barrier between the effluent and skin; and an intact pliable outer abdominal wall skin needed for tension-free approximation and closure.   The low output and thin consistency of the enteric effluent was easily managed through a Jackson-Pratt drain. Higher fistula output volumes or thicker fluids may not have been able to pass through this type of drain and might have prevented this system from working. Also, nutrition is essential for wound healing. This patient had started on TPN 1 month prior to the development of her enterocutaneous fistula. Despite initially low serum albumin and prealbumin (1.6 g/dL and 10 mg/dL, respectively), this patient showed signs of quick healing. She had maximum protein in her TPN of 2.5 g/kg at first sight of the enteric fistula. Finally, her thick subcutaneous tissue provided an area that was able to granulate and provide a firm base for healing once the wound was approximated. A thinner subcutaneous layer may not have provided the depth necessary to approximate the wound to close without tension.   Attempting different wound management measures using a NPWT has allowed for innovative and creative wound care techniques for difficult-to-control abdominal fistulae. The negative pressure hastens wound granulation⁵ which allows for the development of a surface that could be approximated along the subcutaneous tissue. Approximation of the wound was completed with the use of a drape to provide a tension-free closure. Also, the negative pressure increases the blood flow at the level of the wound once it was approximated in order to allow for improved healing and closure.⁶

  Use of NPWT with enterocutaneous fistulae has been found useful for control and diversion of the effluent while promoting healing of the surrounding wound.⁷ However, in this case, the enterocutaneous fistula was actually converted to an easier-to-control enterocutaneous fistula that, ultimately, was able to heal in a relatively short time period. As part of the standard recommendations for evaluation of a new fistula, sources for potential sepsis, such as undrained intra-abdominal fluid collections and identification of potential causes of an enteric fistula, are to be explored.¹ In this case, after control of the fistula through the drain, there was no further clinical evidence of intra-abdominal sepsis. Also, because of the patient’s transfer to a long-term care facility and lack of scheduled follow-up, no further delineation of the fistula anatomy was able to be completed as recommended by standard of care.¹ By the time the patient returned to the clinic, the fistula and wound were healed and she was clinically well.

The authors are from the Medical College of Wisconsin, Milwaukee, WI.

Address correspondence to: Colleen M. Trevino, RN, NP, PhD Medical College of Wisconsin 9200 W. Wisconsin Avenue Milwaukee, WI 53226 ctrevino@mcw.edu

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