Management of Open Abdominal Wounds With Intestinal Fistula Formation Using a Combination of NPWTi-d and New Generation Fistula Devices: A Case Report

Surgical site infections (SSI) are common complications of abdominal surgery.¹ They are usually managed by debridement followed by open treatment of the wounds. Negative pressure wound therapy (NPWT) has become widely accepted to promote tissue granulation and wound conditioning in the open treatment phase.^2,3 Since its introduction a few years ago, NPWT with instillation and dwell time (NPWTi-d; V.A.C. VERAFLO; 3M + KCI) has been shown to be an efficient treatment concept for infected wounds.⁴ The possibility of modifying the wound environment by instilling and allowing to dwell at regular intervals with solutions such as saline or hypochlorous acid, and reducing biofilms,⁵ while simultaneously promoting the formation of granulation tissue, make the technique valuable for the treatment of infected wounds.⁶

However, treating open abdominal wounds with intestinal fistulas, particularly in the small bowel, has been more challenging. In the authors’ practice, these wounds were managed using NPWT with custom-made devices such as baby bottle nipples, urinary catheters, and NPWT,⁷ sealed with stoma paste. The dressings applied in this way started to leak after only a few hours and did not prove to be stable enough to mobilize patients, as has been demonstrated in other research.⁸ The surrounding wound was regularly contaminated with stool, sometimes for hours, which led to painful maceration of the surrounding skin. Consequently, the care of intestinal fistulas was resource- and personnel-intensive and often led to patient dissatisfaction with their health care providers.

Since the introduction of special fistula devices (SFD; FISTULA SOLUTION Devices; 3M + KCI) for combined use with NPWT, patient care has improved.^9,10 These fistula devices are available in 3 different designs: one for simple superficial fistulas, another for deep-seated fistulas, and the last for marginal fistulas. They are made of an especially soft plastic material and adapt themselves to the wound bed. In addition, they are exactly the same height as a normal NPWT dressing and deform together with the surrounding foam when negative pressure is applied. This leads to a high stability of the dressings.

The authors’ experience with this technique has led to significantly greater stability and optimized tightness of the dressing-body seal. By reducing the number of dressing changes and the associated savings in material and working time, this approach appears more economical.¹¹ In this case report, the authors demonstrate the technique of combining NPWTi-d and SFDs use established in their department for the treatment of large fistulizing abdominal wounds.

An obese (body mass index, 55 kg/m²) 57-year-old female with Crohn disease had undergone a bowel resection due to sigmoid perforation in 1997, after which a transverse end colostomy was placed. She recently presented to the authors’ facility with an incarcerated parastomal hernia (Figure 1) with peritonitis (C-reactive protein [CRP]: 395 mg/L [normal: < 3.0 mg/L]; procalcitonin [PCT]: 0.96 ng/mL [normal: < 0.1 ng/mL]). During the emergency laparotomy, the incarceration with peritonitis was confirmed, and the hernia was repositioned into the intraabdominal cavity. No bowel resection was necessary. In the acute situation, an absorbable polyglycolic acid mesh (Safil, B Braun) was inserted in intraperitoneal on lay position as temporary hernia closure, traditional continuous NPWT with -125 mm Hg was applied, and the patient was stabilized with antibiotics and intensive care for 3 days, until the hernia was repaired using a 25 cm x 40 cm bovine dermal matrix mesh (Surgimend 3, 0 mm, Integra) in the intraperitoneal onlay and bridging position. The ostomy had to be placed in the midline through the mesh due to the retracted Crohn mesenterium (this also explained why stoma care had been difficult). The wound became increasingly dehiscent due to the constant stool contamination, and a progressive wound infection occurred. With regular debridement and NPWT, the wound was decontaminated over several weeks and increasingly brought to granulation; 8 weeks after the initial intervention, the wound was closed by cutaneous flap plasty.

Due to an occult intraoperative injury to the small bowel, the patient developed 2 intestinal fistulas. The subsequent wound infection with necrosis of the mobilized tissue flaps caused a ventral soft tissue defect measuring 60 cm x 50 cm (Figure 2). In addition, the patient developed life-threatening sepsis, necessitating systemic antibiotic and antifungal therapy which consisted of ciprofloxacin, linezolid, meropenem, and caspofungin according to the microbial findings until the systemic infection subsided. In the initial debridement, all of the necrotic and infected tissue was resected (Figure 3). Trying to close one of the fistulas using sutures failed during the further treatment period of several days. After initial debridement, negative pressure was applied as follows; the fistulas were isolated using SFDs. Initially, the dressings were changed in the OR; a stable change interval of 3 to 4 days was achieved from the beginning. After 6 dressing changes, the wound appeared macroscopic cleaner and the patient's clinical condition improved progressively. The systemic inflammatory parameters such as CRP, leukocytes, and PCT showed a significant decrease. The wound showed a continuously thickening layer of granulation tissue. Therefore, the NPWTi-d therapy was able to be changed to a conventional NPWT therapy combined with SFDs (Figure 4A). After 8 additional dressing changes on demand (between 2–4 days) over 3 additional weeks, the OR was no longer required for dressing changes and they could be performed bedside. However, as wound area decreased and the wound became more superficial, obtaining a dressing seal became more difficult, increasing dressing change frequency. In addition, it was not possible to achieve the desired split-skin coverage due to the lack of donor sites, so secondary healing continued. Eight months after the initial operation, the patient was discharged to a nursing home with a small granulated wound surrounding the fistulas (Figure 4B) but free of systemic infection.

Dressing application technique

When first implemented, the dressing application technique had 3 objectives: (1) wound decontamination, (2) granulation tissue generation, including on the bovine mesh, and (3) acid succus isolation. For this purpose, the SFDs (WOUND CROWN; 3M + KCI) were sealed with cohesive skin protection plates (Eakin) on the wound side and fixed with a hydrophobic polyurethane ether foam (V.A.C. GRANUFOAM Dressings; 3M + KCI) in the center of the fistula-bearing wound section. The upper fistula was located distal to its counterpart and produced almost no effluent and was covered with a skin protection plate to avoid contact with the NPWT foams in the early treatment phase. The peripheral wound sections and wound pockets were filled with a hydrophilic polyurethane ether foam dressing (V.A.C. VERAFLO Dressing; 3M + KCI). An instillation pad was attached to the right and left lateral to the hydrophobic polyurethane ether foam dressing. The negative pressure connectors were placed far into the flanks (Figure 5).

This arrangement of the dressings, the instillation, and negative pressure connections made it possible to generate a directed fluid flow from medial to lateral during the instillation phases; the fistula adapters were largely excluded from the instillation field by the more hydrophobic foam without losing their tightness.

For this patient, NPWTi-d settings were as follows:

• Instillation volume: This was determined using the fill assist after every dressing change (fill assist stopped when fluid could be detected in the flanks; instillation volume varied from 800 mL in the beginning (2 instillation devices placed serial) to 250 mL at the end of instillation therapy).

• Solution and soak time: An isotonic (0.9 %) saline solution was set to dwell for 10 minutes.

• Negative pressure wound therapy target setting: NPWT device was set at -125 mm Hg and left in place for 2 to 4 hours, depending on the macroscopic aspect (ie, clean = 4 hours, contaminated = 2 hours). Intensity was set to low.

Regular instillation with saline helped achieve rapid macroscopic cleansing and wound granulation even at the first dressing change 2 days after initial debridement. Although microbiological examinations revealed constant wound contamination, no significant systemic infections occurred while providing NPWTi-d and subsequent NPWT over almost 120 days.

Intestinal fistulas are a stressful situation for both the affected patients and the attending HCPs. Therefore, a new generation of fistula devices has been developed in recent years. Their design and adaptation to common NPWT foams have achieved a substantially higher degree of tightness and led to better patient care than previous with custom-made therapeutic devices such as baby bottle nipples or stoma paste rings.¹² However, as a cautionary note, they can only be used to a limited extent in infected and contaminated sites because airtight sealing of an infected wound is not recommended. As such, a concept to combine the technique of NPWTi-d, which is used on infected wounds, with SFDs to achieve the best possible patient care was developed. The combination of hydrophobic and hydrophilic foams provided maximum stability with the highest level of wound bed cleaning. Additionally, the authors utilized NPWTi-d on the highest point of the wound beside the fistula area and placed the negative pressure to the deepest points in the flanks. So, they reached a fluid stream heading away from the fistulas. The use of the hydrophobic polyurethane ether foam in the area around the fistula also led to stable positioning of the dressings because the instillation solution did not moisten and subsequently did not loosen the foam and fistula devices. The authors believe the combination of these different techniques led to the extended dressing change intervals of 2 to 4 days described above. By applying the hydrophilic polyurethane ether foam, the instillation solution could be distributed over the large and deep wound areas in the flanks. In addition, the previously described effect of increased granulation by NPWTi-d (up to 40%)⁶ was also noticed and utilized. The combination of these techniques facilitated a rapid reduction of wound depth and wound surface area through epithelialization from the wound edges as well as through rapid decontamination like others also described.^11,13,14

Combining NPWTi-d and SFDs helped keep the bacterial colonization of the wound from increasing; no further infection of the wound occurred determined by clinical aspect and systemic infection parameters. The initial septic disease was resolved using NPWTi-d in combination with transient systemic antibiotic therapy. In retrospect, systemic therapy may not be necessary because of the rapid local infection control. Although systemic therapy is gold standard in septic disease, treatment with NPWTi-d without systemic antimicrobial therapy should be the focus of further studies.

In this case report, NPWTi-d and SFDs were used together for the first time in the management of a complex fistulizing abdominal wound infection. The study demonstrated that their use in a continuous contaminated and temporarily superinfected laparostoma scenario can contribute to wound healing and fistula isolation, a technique that since this case report has been successfully repeated. Avoiding systemic antibiotic therapy by using an instillation component combined with NPWT in this patient cohort should be the topic of further studies.

Authors: Frank Werner Brennfleck, MD; Henrik Horst Gerhard Junger, MD; Slowik Przemyszlaw, MD; Christian Mauerer, MD; Stefan Martin Brunner, MD; Hans-Jürgen Schlitt, MD; and Matthias Hornung, MD

Affiliation: Department of Surgery, University Hospital Regensburg, Regensburg, Germany

Correspondence: Frank Werner Brennfleck, MD, Senior Physician, Universitätsklinikum Regensburg, Department for Surgery, Franz-Josef-Strauss-Allee 11, Regensburg, Bavaria 93053 Germany; frank.brennfleck@klinik.uni-regensburg.de

Disclosure: Dr Brennfleck is a paid speaker for 3M + KCI.