Objectives: Wound dehiscence is a common surgical complication, especially among pediatric liver transplant recipients in our center. In 2013, we introduced negative pressure wound therapy as a preventive treatment. We herein report the clinical outcomes of this intervention.
Materials and Methods: We conducted a retrospective review of the 26 pediatric liver transplant recipients in our center since 2011. We excluded 1 girl whose wound could not be closed due to bowel edema. The first 13 of the 25 remaining patients were treated with conventional wound management (conventional group). The latter 12 were treated with prophylactic negative pressure wound therapy (prophylactic group). Incidences of surgical complications and patient characteristics were compared between groups.
Results: Wound dehiscence occurred in 7 of the 13 patients in the conventional group and 3 of the 12 patients in the prophylactic group. When restricted to dehiscence that required surgical debridement, there were 6 cases in the conventional group and no cases in the prophylactic group. Although background data showed that liver insufficiency in the prophylactic group was more severe, this group had a lower incidence of wound dehiscence (P = .015).
Conclusions: Prophylactic negative pressure wound therapy is thought to be effective for preventing wound dehiscence among pediatric liver transplant recipients.
Key words : Biliary atresia, Surgical complication, Wound dehiscence
Wound dehiscence, which is one of the most fundamental complications, cannot be easily avoided. Negative pressure wound therapy (NPWT), which can induce secondary healing, is currently one of the most promising wound treatments for wound dehiscence.1 We have experienced a number of severe cases of wound dehiscence in the clinical setting; many were in patients who had undergone pediatric liver transplant. After curative NPWT, patient wounds were shown to be eventually cured. During this process, frequent dressing changes and irrigation were needed. In some cases, wound debridement was also required; these are stresses that could be avoided if severe wound dehiscence could be prevented.
In the present decade, the application of NPWT to closed wounds from the end of operation has been introduced for the prevention of wound complications in high-risk patients. To date, several studies have investigated the prophylactic use of NPWT. However, only a few of these studies investigated the use of NPWT in solid-organ transplant recipients. In March 2013, we used NPWT prophylactically for the first time in the treatment of a 1-year-old girl who had undergone living-donor liver transplant. Since then, we have performed prophylactic NPWT in the treatment of 13 pediatric liver transplant recipients in our center. In the present study, we compared the clinical backgrounds and outcomes of patients who were treated with and without prophylactic NPWT and assessed the efficacy of this therapy.
Materials and Methods
We conducted a retrospective review of 26 pediatric liver transplant recipients (< 18 years old) who were treated at our center from January 2011 to January 2016. We excluded 1 girl in whom wound closure could not be achieved due to bowel edema. The first 13 of the 25 patients, were treated with conventional wound dressing (conventional group). The latter 12 patients received prophylactic NPWT immediately after transplant (prophylactic group). In the prophylactic group, continued curative NPWT was applied to patients who developed dehiscence.
For detailed collection of data, we stratified wound dehiscence into 2 categories: “minor dehiscence,” which could be cured without surgical debridement, and “major dehiscence,” which required surgical debridement. Patient demographic data, Pediatric End-Stage Liver Disease (PELD) scores,2 incidence of wound complications, and clinical course were compared between the 2 groups.
In the conventional group, the abdominal walls were closed with either running or knotted absorbable sutures, the fascia and subcutaneous layers were closed with knotted absorbable sutures, and the skin was closed with either knotted or running buried sutures, simply knotted sutures, or with skin staplers. After closure of the skin, hydrocolloid wound dressings were applied and changed according to the amount of discharge until the wound healed.
In the prophylactic group, the abdominal wall and skin were closed in the same manner as for the conventional group. After wound closure, the wound fields were cleaned with saline-soaked gauze. Liquid transparent silicon foam sheets (Mepilex Transfer; Mölnlycke Health Care, Gothenburg, Sweden) were applied to cover the closed wounds. A suction tube was then placed on the silicon foam sheet with Adapt Skin Barrier Paste (Hollister Incorporated, Libertyville, IL, USA) to prevent air leakage from the tube site. Next, a thin dressing film (Airwall; Kyowa Limited, Osaka, Japan) was applied to cover all of the Mepilex Transfer sheets. A suction tube was then connected to a continuous aspirator (Mera Sacume MS-008; Izumi Kokako-gyo Co. Ltd., Tokyo, Japan) to provide continuous negative pressure (-50 hPa). Figure 1 shows an example of our prophylactic NPWT. These dressings were changed every 1 to 3 days, depending on the amount of discharge, which clogged the transparency of the silicon sheets. The dressing was changed to a normal dressing when the discharge stopped and the wound healed.
All patients received the same immunosuppression protocol (a combination of tacrolimus and low-dose steroids).3 Tacrolimus was initiated 1 day before transplant at an oral dose of 0.075 mg/kg/day. The target trough level was 10 to 15 ng/mL within 3 weeks after transplant, then 10 ng/mL until 3 months, and 5 to 10 ng/mL thereafter. Steroids were administered just before and after graft reperfusion, each at a dose of 10 mg/kg. Intravenous methylprednisolone was administered continuously for the first 6 days after transplant (2.0 mg/kg/day on days 1-3 and1.0 mg/kg/day on days 4-6), followed by oral prednisolone (0.5 mg/kg/day) for 2 weeks from day 7. From 3 weeks to 3 months posttransplant, a dose of 0.3 mg/kg/day was administered. From 3 months after transplant, the dose was gradually reduced to 0.1 mg/kg/day. Methylprednisolone was withdrawn at 6 months posttransplant. At acute rejection, a bolus injection of intravenous methylprednisolone (10 or 20 mg/kg/day) was administered, after which the dosage was tapered (steroid pulse therapy).
Statistical analyses were performed using the JMP 11 software program (SAS Institute Inc., Cary, NC, USA). We used chi-square test and the Fisher exact test to investigate the significance of the categorical variables and t test for continuous variables. Variables that demonstrated statistically significant differences in univariate analysis were included in a multivariate stepwise regression analysis to determine the significance of their contribution to postoperative wound complications. All tests were 2-tailed. P values < .05 indicated statistical significance.
Our 25 study patients included 11 boys and 14 girls. The median age was 8 months (range, 28 days to 15 years). The PELD score has an age limit (< 13 y); therefore, 1 boy, who was over 13 years of age, was excluded from the PELD score comparisons. The primary diseases of the patients were biliary atresia (n = 23), hemochromatosis (n = 1), and primary sclerosing cholangitis with ulcerative colitis (n = 1). Twenty-four patients underwent living-donor liver transplant, and 1 patient had deceased-donor liver transplant. During their hospitalizations, 9 patients (3 in the conventional group vs 6 in the prophylactic group; P = .15) showed acute rejection and were treated with steroid pulse therapy. Three patients underwent a reoperation. These included 1 patient in the conventional group who had intestinal perforation and 2 patients in the prophylactic group (1 with intra-abdominal bleeding and 1 with portal vein thrombosis). Among these cases of rejection and reoperation, 2 patients with rejection and 1 patient with thrombosis died during hospitalization (1 in the conventional group and 2 in the prophylactic group). At the time of death, wounds were recorded as being healed, and prophylactic NPWT had been completed.
The demographics of the 2 groups are compared in Table 1. A univariate analysis showed that patient sex, age, height for age, body weight for age, primary disease, graft size for body weight, bleeding during operation, use of skin stapler, and acute rejection were not significantly different. The only significant difference was the PELD score, which was significantly higher in the prophylactic group (6.8 vs 15.7; P = .03).
In total, 10 patients experienced wound dehiscence during hospitalization. In 4 patients, second healing was achieved with continuous (curative) NPWT alone, without any surgical procedures (defined as “minor dehiscence” in this study). The other 6 patients required reopening of the wound surface and debridement to achieve second healing (“major dehiscence”). In all cases, dehiscence was limited to the subcutaneous layers, and muscle layers were maintained. This meant that all cases of dehiscence in the present study were categorized as superficial dehiscence. No surgical site infections, as defined by guidelines from the US Centers for Disease Control and Prevention,4 occurred among our study patients. The distribution of wound dehiscence in the 2 groups is shown in Table 2. Although wound dehiscence occurred in both groups, major dehiscence, which requires surgical debridement, only occurred in the conventional group (P = .015).
The predictors of wound dehiscence are reviewed in Table 3. Our univariate analysis showed that lower PELD score, lower total bilirubin, smaller graft size, and larger patient body weight were significant predictors of total wound dehiscence. When restricted to major dehiscence, the predictors included female sex, lower PELD score, lower total bilirubin, smaller graft size, taller patient height, and conventional wound dressing without prophylactic NPWT. Our multivariate analysis showed that lower PELD score and conventional wound dressing were both independent predictors of major wound dehiscence (not shown in tables; P = .0016 and P = .0145, respectively).
The typical clinical course of the wound in the prophylactic group is shown in Figure 2. Prophylactic NPWT was initiated just after the operation (Figure 2, A and B). On postoperative day 2, wound discharge clogged the silicon foam (Figure 2C). The wound was in good condition, without redness (Figure 2D). The wound was healed at discharge on postoperative day 20 (Figure 2E).
In the present study, we evaluated the effects of prophylactic NPWT among our pediatric liver transplant recipients. The incidence of major wound dehiscence, which required surgical debridement, was significantly lower in the prophylactic group than in the conventional group. Our multivariate analysis showed that both a lower PELD score and conventional wound therapy were significant predictors of major wound dehiscence. The inclusion of a lower PELD score (score of liver insufficiency) as a risk factor may be controversial. This result may indicate that the effectiveness of prophylactic NPWT was better than expected or that preoperative liver insufficiency may not be as much of a risk factor for postoperative wound dehiscence.
Many factors can be considered to be risk factors for wound complications in liver transplant. Most of our pediatric liver transplant patients had biliary atresia. Among patients with biliary atresia, in addition to the use the of steroids,5 immunosuppressive medications,6 longer operation time, and larger amount of blood loss during operation,7,8 other risk factors may include liver insufficiency, malnutrition, and increased collateral blood flow in the body wall resulting from liver cirrhosis. Once acute rejection occurs, not only the rejection itself but also the steroid pulse therapy will affect the prognosis of wound healing. The incidence of wound dehiscence among liver transplant recipients is reported to be up to 27%.7 Although advancements in preoperative and postoperative management of liver transplant patients have occurred, wound dehiscence remains a frequent and fundamental surgical complication among liver transplant recipients. Once wound dehiscence occurs, either frequent dressing changes or the initiation of curative NPWT is inevitable. In most cases, wound debridement is also required, which would be highly stressful, especially for pediatric patients. We initiated prophylactic NPWT to prevent these avoidable stresses.
In 1997, NPWT was first introduced.9 The primary target of the therapy was chronic, complicated wounds. The described mechanisms of action for NPWT include the reduction of edema,10 increased perfusion,9 and enhanced development of granulation tissues.11 On the other hand, the prophylactic use of NPWT was first reported by Stannard and associates in 2006 for incisions after high-energy traumas.12 The primary target of prophylactic NPWT has been high-risk wounds, such as wounds in patients with diabetes mellitus, cardiac operation incisions, and incisions on the extremities. Today, NPWT is used for various types of wounds. There is already 1 case report about the use of NPWT for treatment of a renal transplant recipient.13 In addition, several studies have investigated the use of NPWT in the treatment of wound dehiscence in liver transplant recipients.14,15 To the best of our knowledge, this is the first report of the prophylactic use of NPWT in the treatment of liver transplant recipients.
Although several types of prophylactic NPWT devices and systems are already commercially available, in our study, we implemented a “handmade” version of NPWT using commonly available medical materials and devices. One reason for this is that, as of 2016, no NPWT devices are covered for prophylactic use by the Japanese national health insurance system. The other reason is the benefit of a handmade system, which would allow the shape and materials to be chosen on an individual basis. In pediatric liver transplant recipients, the application of NPWT dressings can easily be interfered with and limited by the placement of abdominal drainage tubes. After the operation, a window area near the operative incision is also needed to perform postoperative color Doppler ultrasonography. We had no difficulties in performing ultrasonography examinations in our cases. The transparency of the Airwall, a film dressing that we used in the present study, did not interfere with the ultrasonographic examinations. Furthermore, like other NPWT systems, our handmade NPWT kept the wound discharge inside the dressing, which kept the wound and the clothes of the patient clean. Thus, our handmade prophylactic NPWT system was positively received by our staff members and patients.
With regard to NPWT complications, there have been reports of infection, pain, and bleeding at the wound site. Although wound site infections are frequently reported in patients who receive NPWT, no cases of infection were identified in our series. The use of NPWT in cases of wound infection was originally discouraged. For the use of NPWT in cases of wound infection, we usually choose NPWT combined with continuous irrigation,16 which is thought to be effective in such cases. Another complication of NPWT is pain17 at the wound site. There were no pain episodes among our 25 study patients. This may be because the negative pressure that we applied (-50 hPa or -37 mm Hg) was weaker than that with commercially available NPWT devices (typically -125 mm Hg). The other complication of NPWT is bleeding. Although arterial bleeding has been reported in patients undergoing NPWT,18 there were no cases of bleeding among our patients. Thus, NPWT is thought to be a safe and effective method of wound treatment in pediatric liver transplant recipients.
In conclusion, we report that the incidence of wound dehiscence was decreased in the prophylactic NPWT group. Prophylactic NPWT is thought to be a useful treatment for pediatric liver transplant recipients.
DOI : 10.6002/ect.2018.0076
From the 1Department of Pediatric Surgery, Graduate School of Medical Sciences,
Kyushu University, Fukuoka, Japan; and the 2Department of Nursing, Kyushu
University Hospital, Fukuoka, Japan
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare.
Corresponding author: Genshiro Esumi, Department of Pediatric Surgery, Graduate School of Medical Sciences, Kyushu University, 3-1-1, Maidashi, Higashi-ku, Fukuoka 812-8582, Japan
Phone: +81 92 642 5573
Figure 1. Example of Our Prophylactic Negative Pressure Wound Therapy System
Figure 2. Typical Clinical Course of Wound in the Prophylactic Group
Table 1. Patient Demographics
Table 2. Distribution of Wound Dehiscence in Each Group
Table 3. Predictors of Wound Dehiscence in Pediatric Liver Transplant Recipients