Treatment of Wound Dehiscence Utilizing Negative Pressure Wound Therapy With Instillation and Dwell Time in Emergency Abdominal Surgery: A Step-by-step Closure Protocol

Surgical wound dehiscence occurs in 0.25% to 3% of patients post-laparotomy.¹ Multiple local, systemic, and mechanical factors can contribute to abdominal wound dehiscence: infection, hematoma or seroma, ostomy, ileus with distended loops, pulmonary disease, hemodynamic instability, age greater than 65 years, hypoalbuminemia, systemic infection, obesity, uremia, malignant tumors, ascites, steroid therapy, chemotherapy or radiotherapy, and hypertension.^2,3 Technical defects in wall closure and emergency laparotomies are associated with increased risk of dehiscence.⁴ 

Dehiscence, whether partial or complete, is associated with a high mortality rate (20%)⁵ if bowel protrudes from the wound. Wound site infection increases both costs and hospital stay for the patient, and it induces an increase in morbidity and in mortality.⁶

In order to promote healing, the dressing must adequately cover the dehiscent wound, especially if the viscera are exposed. Wound management varies based on the wound size, type of exposed structures and organs, and potential comorbidities. ⁷

Negative pressure wound therapy (NPWT) is widely used in the management of both acute and chronic wounds.^8-10 Negative pressure wound therapy with instillation and dwell time (NPWTi-d) is an evolution of negative pressure therapy; once instilled, a topical solution dwells for established periods of time between negative pressure cycles to facilitate wound cleansing, wound bed preparation, and the healing of complex wounds in various surgical and medical specialties.^11,12

This case report presents the authors’ experience in the treatment of wound dehiscence with NPWTi-d (V.A.C. VERAFLO Therapy; 3M + KCI) in 13 patients undergoing emergency abdominal surgery. Additionally, this report provides the potential advantages associated with the authors’ step-by-step closure protocol.

The intent of this work is to formally establish a course of treatment in the case of surgical wound dehiscence. By following the included steps in the step-by-step closure protocol, there is a possibility to reduce the length of time of both healing and hospitalization. As the protocol, which was presented to all surgeons in the same operating unit, is schematic and divided into phases (known as T) that correspond to each dressing change, it allows all surgeons to intervene at any stage knowing exactly how to proceed.

In the authors’ Department of General and Oncological Surgery, from July 2017 to December 2019, 21 patients who were undergoing emergency abdominal surgery had wound dehiscence; 8 were treated with drainage and gauze packing while 13 patients were treated with NPWTi-d. The criteria for inclusion in the NPWTi-d group included excessive wound length; high degree of wound contamination; presence of comorbidities such as obesity, diabetes, heart disease, and smoking; and a history of previous abdominal surgery. The case series presented herein concerns the 13 patients treated with NPWTi-d.

The majority of patients were female (69.23%), and the average age was 60 years (range, 23–84 years). A total of 5 patients (38.46%) had hollow bowel perforation; of these, 3 underwent ileal or colic resection. Additionally, 4 patients (30.77%) had intestinal occlusion caused in 3 cases by bridle and in one by a Petersen hernia; 2 of these patients underwent intestinal resection for necrosis. Of the 3 oncological patients (23.08%), 2 had stenosing colon cancer and 1 had perforated colon cancer. Finally, 1 patient (7.69%) underwent extensive ileal resection due to intestinal infarction.

The abdominal incisions were preceded by hair removal and cleaning with iodopovidone. No incise drapes were used. Of the 13 patients, 4 (30.77%) underwent supra/subumbilical incision, 3 (23.08%) had xipho-pubic incision, 3 (23.08%) had umbilical-pubic incision, 2 (15.38%) had xipho-umbilical incision, and 1 (7.69%) had umbilical-pubic and Pfannenstiel incisions (occlusion after recent hysterectomy). 

On average, wound dehiscence occurred on the ninth postoperative day (range, 7–18 days), requiring a readmission for 10 patients (76.92%), while the other 3 had remained in the hospital since the operation due to respiratory problems or heart failure. Contaminated surgical wounds were classified as grade IV (Figure 1) in 11 patients (84.61%) and grade III in 2 patients (15.38%). 

Wound culture swabs were collected at time of the initial dressing, and the results are summarized in Table 1. In 4 patients (30.77%), the swabs showed contamination of the wound by several bacteria. In 7 patients, contamination by a single bacterium was found. Only 2 (15.38%) patients presented with negative culture swabs. 

Protocol

The step-by-step closure protocol was applied by all 4 surgeons in the general and oncological surgery unit of the authors’ hospital, under the supervision of the senior surgeon. 

The step-by-step closure protocol featured a minimum of 3 to a maximum of 7 steps (Table 2). Each dressing change was designated with T and a consecutive number. Every 3 days, the dressing was changed. If necrotic material was visible during the dressing change, sharp debridement was performed at the bedside. Starting at T₀, application of NPWTi-d occurred, with instillation of saline solution and a dwell time of 10 minutes. Depending on the extent of the wound and the thickness of the subcutaneous tissue, the amount of liquid instilled varied; it ranged from 50 mL for the supra/subumbilical wounds to 120 mL for the xipho-pubic wounds. In 11 (84.61%) of the cases, 3-hour cycles of continuous negative pressure at -125 mm Hg were used; medium-sized NPWTi-d dressings were applied (Figure 2A). In 2 patients (15.38%) with muscle fascia defect and ileal loop exposure, -100 mm Hg continuous negative pressure was applied and for 15 days; polyvinyl alcohol dressings (V.A.C. WHITEFOAM Dressing; 3M + KCI) were used in the deep portion (in contact with muscle fascia and loops) and medium-sized NPWTi-d dressings, applied on top, were appropriately shaped to remain below the edges of the aspirated wound (Figure 2B).

The swab was performed at T₀ and the first dressing change (T₁) was performed after 3 days, so most swab results were ready at T₁. At T₁, modifications, if necessary, of systemic antibiotic therapy based on antibiogram results were made. At T₂, step-by-step closure of subcutaneous tissue (in detached stitches of resorbable material) of skin in detached stitches (Premilene; B. Braun Medical Ltd) after infiltration of local anesthetic (lidocaine) was started. The closure affected both the cranial and caudal margins for about 3 cm to 5 cm at each step (Figure 3). The closure in detached stitches was made only if the bottom of the wound, in the caudal and cranial portions of the wound, was well sprinkled, well cleansed with saline and granulation tissue was present. Of note, the 2 patients (15.38%) with fascial defect and exposed ileal loops started step-by-step closure at T₃. Depending on the extent of the surgical wound, step-by-step closure was repeated at each dressing change until complete closure (Figure 4). At step T₃, patients 1, 2, 3, and 4 (Table 2) with supra/subumbilical wounds (ie, least extensive) were completely closed, while the 2 most complex cases (cases 12 and 13; Table 2) underwent 5 steps of closure with detached stitches (T₃ and T₇ ). Overall there were 7 applications of NPWTi-d with increasingly smaller-sized dressings. In cases 12 and 13, the traction performed by NPWTi-d was amplified with silk U-shaped detached stitches, with 1 stitch between the edges of the middle third location of the wounds (Figure 5). Patients 5 and 6 had xipho-umbilical incision and were completely closed at step T₄. Patients 7, 8, and 9 had umbilical-pubic incision and were completely closed at step T₄. Cases 10 and 11 had xipho-pubic incision and were completely closed at step T₆. 

The use of NPWTi-d reduced the number of dressing changes compared with saline-soaked gauze dressing. On average, 4.5 dressing changes were performed in the NPWTi-d group. In the traditional dressing group, an average of 17 dressing changes were made. In this second group, the dressing was changed daily in 70% of cases, because the exudate was abundant; in the remaining 30%, the dressing was changed every other day. In all 13 cases, the dressing changes always occurred at the patient’s bedside. No patient required analgesia or pain therapy.

The average time for complete closure, with the initial dressing change or T₀ being the application of the NPWTi-d (V.A.C. ULTA; 3M + KCI), was 13 days (range, 7–21 days). Instillation resulted in faster and better wound bed cleansing compared with traditional dressings; exudate management was better, and the average complete wound closure time was faster than with traditional dressings (13 vs. 27 days). 

In areas of necrotic material NPWTi-d did not replace surgical debridement.¹² The closure of the damaged muscle fascia in patients 12 and 13 was only partially possible; at the patient’s bedside, after infiltration with local anesthetic, detached stitches of resorbable material were applied, partially closing the fascial defect. Where it was not possible to close the fascia, only the skin was closed, causing an induced incisional ventral hernia. The step-by-step closure protocol made it possible to anticipate the healing of the ends of the wound when the bottom was well cleaned and grainy in that portion; at the same time the NPWTi-d continued to treat the sanious or partially necrotic bottom of the middle third of the wound.

The step-by-step closure protocol made it easier to approach and subsequently synthesize the edges of the wound in the middle third and, in synergy with NPWT, significantly reduced tension at the edges of the middle third of the wound in patients with xipho-pubic incision (patients 10, 11, and 12) and in umbilical-pubic plus Pfannenstiel incision (patient 13).

The step-by-step closure protocol also accelerated complete healing time, especially in the xipho-pubic wounds, because during the final dressing change at the last step (T₆), the skin stiches of the wound extremities that were applied at T₂ of patients 10 and 11 could be removed; for these patients, at least 9 days had passed (considering that each T represents a dressing change every 3 days—hence, T₃ + T₄ + T₅ equals 9 days).

The tractive force exerted by the NPWT on the wound edges was increased and sustained by means of detached U-shaped stitches made of nonabsorbable material in the 3 patients with previous xipho-pubic incision (patients 10, 11, and 12) and in the patient with previous umbilical-pubic and Pfannenstiel incisions (patient 13). No patient required a skin graft for complete closure.

Only in 1 patient, with a previous xipho-pubic incision, the middle third of the wound healed by secondary intention in 8 weeks; he underwent ambulatory follow-up, and advanced dressings were used.

Although treatment with NPWTi-d was used off label in the 2 patients with fascial defect, the integrity of the intestinal loops and muscle fascia was not altered; it did not negatively affect subsequent abdominal wall plastic surgery following the development of incisional ventral hernias (Figure 6). One of the 2 patients (case 13) underwent abdominal wall reconstruction using the component separation technique 6 months after discharge (Figure 7A). The extent and tenacity of visceral-visceral and viscero-parietal adhesions were much lower than expected after treatment with NPWTi-d (Figure 7B, 7C). The separation of the muscle fasciae, despite the defect treated with NPWT, was without complications (Figure 7D). 

Patient 12 declined wall reconstruction surgery because he had previously undergone bariatric surgery, which resulted in significant weight loss; thus, the patient was not in a sufficient general condition to be subjected to the component separation technique.

No patient had partial or total reopening of the wound. Additionally, no patient developed enterocutaneous fistulae. The aesthetic result was defined as satisfactory by 10 patients (76.92%).

The benefits of NPWTi-d in the treatment of complex wounds have been demonstrated in several cohort studies, case series, and meta-analyses. These studies have shown a reduction in the number of dressing changes, patient-reported pain, and total number of days of hospitalization.^11,13-15 In recent years, there has been evidence of improved skin perfusion and reduced tension at the edges of the wound.¹³

Experimental models have also shown that instillation in combination with negative pressure therapy accelerates granulation tissue formation by 43% compared to negative pressure therapy alone.¹⁶

While there are no definitive recommendations for the type of liquid to be used for instillation at this time, Ludolph et al¹³ showed NPWTi-d reduces bacterial load in contaminated wounds by favoring their closure.

The authors’ case series exhibited all of the known benefits of NPWTi-d and, in particular, the microdebridement that the instillation cycles are able to perform on dehiscent wounds. Saline solution was chosen because many studies^11,17,18 have shown no significant difference in terms of bacterial load reduction and debridement compared with an antiseptic liquid,¹² such as polyhexamethylene biguanide (0.1%) plus betadine (0.1%). The amount of liquid instilled was determined during the first treatments T₀: increasing by 10 cc at each test cycle and verifying the absorption capacity of the sponge, taking care to instill only as much liquid as could be absorbed by the sponge and not leak from the edges of the wound.

The step-by-step closure allowed for the gradual reduction of the wound surface area to be treated with NPWTi-d and to stabilize the closure at the extremities. The stitches applied in the subcutaneous tissue and on the skin reduced the tension on the remaining part of the dehiscent wound; the application of U-shaped stitches was useful in gradually bringing the edges closer, using the elastic capacity of the skin.

The limited number of patients presented limitations. It would be interesting to enlarge the case with a multicenter study. The number of patients with fascial defect treated with NPWTi-d in association with our protocol could be expanded and analyzed for any complications, such as enterocutaneous fistulas; it could be demonstrated that the surgical treatment of induced incisional hernia is feasible and safe. The synergy between step-by-step closure protocol and NPWTi-d in reducing wound edge force tension would require appropriate measuring instruments. The benefit of the protocol is so far based only on personal experience.

The synergistic action of step-by-step closure and NPWTi-d allowed for the treatment of infected dehiscent surgical wounds to obtain excellent results, even in patients with muscle fascia defect and loop exposure, and accelerated healing times by reducing the tension at the edges of the wound, by resorting in 1 case to healing by secondary intention and in no case to skin grafts. 

Authors: Raffaele Porfidia, MD¹; Simona Grimaldi, MD²; Maria Giovanna Ciolli, MD¹; Pietro Picarella, MD¹; Angela Romano, MD¹; and Sergio Grimaldi, MD¹

Affiliations: ¹Ospedale Convenzionato Villa Dei Fiori, Acerra - Naples, Italy; and ²Università degli Studi di Milano Facoltà di Medicina e Chirurgia, Milan, Italy 

Correspondence: Raffaele Porfidia, MD, Surgeon, Ospedale Convenzionato Villa Dei Fiori, General and Oncological Surgery, Corso Italia 157, Acerra, Naples 80011 Italy; lelloporf@hotmail.com

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