Promoting Wound Healing Using Bilayer Negative Pressure Wound Therapy to Minimize Subcutaneous Dead Space: A Porcine Model Investigation and Retrospective Clinical Review

Negative pressure wound therapy (NPWT), developed by Morykwas and Argenta in 1997^1,2 to manage persistent wounds, has been shown to be a beneficial advancement in wound care. Its use has expanded dramatically, encompassing many different disciplines^3,4 and wound types such as complex diabetic wounds, pressure ulcers, vascular ulcers,^5-7 open fractures, lacerations, degloving injuries, and burns.^8-11 Recent clinical studies^12,13 support the use of NPWT to bolster skin grafts in order to increase graft adherence and to stabilize wounds,^12,13 as well as in certain high-risk incisions (incisional NPWT) such as sternal incision,^14,15 laparotomy incision,¹⁶ and perineal incision,¹⁷ to reduce complication rates.

The mechanisms of action of NPWT involve wound edges contraction, deformation of the wound surface by mechanical forces, protection and maintenance of a moist wound environment, removal of exudate, physically secluding bacteria from the wound, and increasing blood flow to the wound.^4,18-20 Bilayer NPWT, a variation of incisional NPWT, uses a combination of regional NPWT over skin flaps connected to the same suction source with a drain underneath. This type of negative pressure arrangement is capable of decreasing dead space and promoting wound healing; it has been reported²¹ to promote healing of refractory seroma after mastectomy and aid in reducing the complication rate of the saphenous vein donor site after bypass surgery. In 2017, research indicated this therapy could significantly reduce postoperative complications, such as lymphedema, incision infection, wound dehiscence, and skin flap necrosis, in patients undergoing coronary artery bypass graft involving great saphenous vein harvesting.²²

The authors clinically observed the bilayer NPWT seemingly accelerating wound fluid removal compared with standard NPWT; however, the specific mechanism of this therapy remains unclear. The present study aimed to evaluate and describe the efficacy (including the rate and amount of drainage) of bilayer vs traditional NPWT for fluid removal in a porcine (pork loaf) model. In addition, bilayer NPWT application in patients undergoing inguinal lymph node dissection (ILDN) was explored.

Porcine model

Fresh, square-shaped pork loaves were purchased from the supermarket. A cutaneous flap model (10 cm × 10 cm × 1 cm) was made with a sharp scalpel. The flap was elevated to the subcutaneous level and closed with a running suture after a drain and a 12-gauge catheter were placed in the cavity (Figure 1A). For bilayer NPWT, the flap was covered with a gauze pad and sealed in the standard fashion with an airtight, transparent adhesive film made of polyurethane and acrylic acid that was hypoallergenic and less likely to cause skin irritation. (Figure 1B). The deep drain and suction tube were connected to the negative pressure source (ActiV.A.C. Therapy System; 3M) with a Y-shaped tube (Figure 1C, black arrow). Gauze pad thickness was measured with a Vernier caliper. In the control group, the flap was covered with a gauze pad, and the drain from the cavity was connected to a negative pressure source. Only 1 wound (a bilayer or control) was created in one pork loaf; in both groups (n = 6), the pork loaves were discarded after each test for quality control.

Diluted diatrizoic acid (370 mg/L, diluted with saline by a ratio of 1:5) was used as a contrast medium to reproduce a lymphocele. This fluid was injected into the wound cavity through the 12-gauge NPWT catheter by a dynamic injector (Ulrich) (Figure 1C, white arrow) at a rate of 0.2 mL/second. To explore the pattern of fluid removal, both the bilayer NPWT group and control group underwent computed tomography (CT) (Siemens) 10 minutes after the initiation of the diluted diatrizoic acid injection (Figure 1C). After CT, 3-dimensional (3D) images were obtained by volume reconstruction with an experienced technician.

Herein the focus of this investigation was to visualize the fluid distribution pattern within the wound pockets in each group (n = 6/group), as the existence of fluid may result in the detachment of flaps from wound bases, thereby prolonging the wound healing. To test the efficiency of fluid removal, 20 mL of diluted diatrizoic acid was injected into the cavity in both groups before suction was initiated; subsequent fluid was injected at 0.2 mL/s. Computed tomography was performed at 0, 5, and 20 minutes each after initiation of constant fluid injection in both groups. Subtraction software (Siemens Subtraction) was used to reconstruct layers of 0.5-mm thickness and the 0.4-mm space images that displayed the fluid distribution in the cavity by CT. Then the initial contrast agent area was measured; the remaining contrast agent area was measured at 5 minutes and 20 minutes, each after injection in both groups.

In addition, the authors suspected the interfacial pressure and pressure in the cavity under the flap were related to the size of the suction and thickness of the gauze in the bilayer NPWT group. Pressure generated by bilayer NPWT was measured with a pressure sensor kit (Honeywell; Figure 1D, 1E); a pressure sensor was placed over the flap (underneath the gauze pad), and a second pressure sensor was inserted into the subcutaneous cavity. The authors measured the pressure at the flap interface and in the cavity under the flap under different gauze thickness and suction in the bilayer NPWT group (n = 6). Each porcine model was repeated 3 times and the data were averaged.

Porcine model: data collection and analysis

The authors calculated the ratio of residual to initial contrast medium area at 5 minutes and 20 minutes after injection for both groups of porcine models (N = 12). The data were statistically tested by repeated measure analysis of variance (ANOVA) with SAS, version 9.3 (SAS Institute). The surface pressure and intracavitary pressure of the porcine model (n = 6) under different gauze thicknesses and different pressures were collected, and the data were statistically tested by repeated measure ANOVA with the aforementioned statistical software. A value of P < .05 was considered statistically significant.

Retrospective patient data review

Although it was verified that the fluid free condition generated by bilayer NPWT relied on extra pressure over the flap in the porcine model, the model proposed in the current study was different from the clinical operation; it could not observe the effect of the bilayer NWPT on flap healing, so the authors applied the bilayer NWPT in inguinal lymph node dissection (ILND).

An institutional review board approved a retrospective review of consecutive patients (electronic data from Changhai Hospital, Shanghei, China) undergoing ILND over a 4-year period (February 2010–March 2014) in the authors’ institution. Informed consent was given by all included patients. Inclusion criteria consisted of patients who received bilayer NPWT after ILND from February 2010 to March 2014. Exclusion criteria included: patients who had not received ILND for the first time; if ILND was not performed as a standard procedure or by a designated physician; or if bilayer NPWT was terminated during treatment.

Demographic and clinical data including age, sex, body mass index (BMI), comorbidities (diabetes, hypertension, cardiovascular disease, cancer), smoking status, location of ILND, and involvement of sentinel lymph node were extracted from medical records. Occurrence of complications including hematoma, infection, lymphocele, wound dehiscence, partial flap loss, pulmonary embolism (PE), and deep vein thrombosis (DVT) was noted.

All clinical procedures were performed following informed consent by the patients and in accordance with clinical guidelines. Inguinal lymph node dissection in patients with suspected malignancy was performed in the standard manner as described by Basset²³ by 2 experienced surgeons (authors HB, JL) at Changhai Hospital. In general, an S-shaped incision was created, and the flap was elevated to the subcutaneous level (Figure 2A). The sartorius muscle was dissected from the iliac end to cover the femoral artery, vein, and nerve after removal of the soft tissue with lymph node(s) (Figure 2B). A drain was placed inside the cavity before the wound was closed using interrupted sutures (Figure 2C, 2D). After placing a gauze pad (2 cm in thickness) over the flap, another suction tube was placed in the gauze, and the entire system was sealed with adhesive film in a standard fashion (Figure 2E). The deep drain then was connected to the suction tube with a Y-shaped connector, and the NPWT suction source was provided at -125 mm Hg continuously for 9 days to 12 days (Figure 2F). Sentinel lymph node biopsy was performed. Drain volumes were recorded for all patients.

Postoperatively, all patients received standard care including nutrition, first-grade nursing (bed rest, observe and record the amount of drainage, body cleaning), analgesic treatment, and preventive anti-infection treatment; patients were allowed protected lower extremity motion in the hospital bed. Patients were kept in the hospital postoperatively until deemed surgically and medically fit for discharge.

Retrospective review: data collection and analysis

General patient demographics (age, body mass index, sex, underlying disease, type of cancer, ILND side, and sentinel lymph node biopsy) and postoperative complications (hematoma, infection, wound dehiscence, lymphocele, partial flap loss, PE, DVT) were retrieved from the medical records and recorded in Excel. Continuous data in normal distribution were expressed as mean ± standard deviation.

Porcine model

Distribution of the diluted diatrizoic acid fluid was clearly visible using 3D CT reconstruction. The injected fluid flowed through a small tunnel in the cavity, entered into the deep drain immediately, and was discharged from the drain (Figure 3A, 3E). In the control group, the injected fluid occupied half of the entire cavity before being collected by the drain (Figure 3B). As seen in Figure 3A and 3B, the majority of the flap was firmly attached to the wound base with the aid of bilayer NPWT, while a vast area of the flap was detached from the wound base by dispersed fluid in the control group.

Efficient fluid removal from the cavity was observed with bilayer NPWT (Figure 3D, 3E); 5 minutes after initiation of suction, nearly 70% of the fluid was removed (Figure 3D), and 20 minutes after initiation of suction, most of the area within the cavity was fluid-free (Figure 3E). In the control group, dynamic suction removed some fluid and compressed the cavity (Figure 3H); however, a large area of the wound base contained fluid at 20 minutes and the mock seroma (Figure 3I) remained, a significant difference in fluid between the bilayer NPWT and control group (n = 6, P < .05) (Figure 3F).

With regard to pressure, increased suction power and increased gauze thickness led to increased pressure over the flap. However, this correlation between pressure or gauze thickness and suction was not linear; suction increased less when the gauze was 15 mm or thicker or when pressure was -125 mm Hg or higher (Figure 4A, 4B). This pressure within the cavity, although weaker than that over the flap surface, showed the same pattern as the flap surface pressure when the gauze thickness and suction power were changed (Figure 4A, 4B). In the control group, slight interface pressure existed when suction was initiated; however, this pressure did not change significantly when the suction power increased (Figure 4B, black bar).

Retrospective study

A total of 46 consecutive patients (26 males, 20 females; average age, 57.6 years [range, 39–82 years]; average BMI, 24 kg/m² (range, 19–29 kg/m²]) underwent ILND in the designated time frame (Figure 5). Many had diabetes (n = 22), hypertension (n = 27), and smoked (n = 28). Almost all lymph nodes tested positive for cancer (n = 44). Indications for ILND included lower limb melanoma (n = 32), squamous cell carcinoma (n = 12), and syringocarcinoma (n = 2). Mean post discharge follow-up duration was 3 months. All other patient characteristics are reported in Table 1.

A detailed consent form was signed before ILND; benefits and risks of this procedure, including that of the bilayered NPWT, were thoroughly explained to the patients. The average dissected tissue volume was 22.3 cm × 18.6 cm × 1.8 cm (from 26 cm × 22 cm × 2 cm to 19 cm × 16 cm × 0.8 cm). Postoperatively, patients were hospitalized for a mean of 12.5 days (range, 9–14 days). When drainage was not greater than 10 mL over a 24-hour period, the drains were removed regardless of the output. The average duration of bilayer NPWT was 12.5 days (range, 9–15 days). No hematomas, seromas, surgical site infections, or DVT/PE occurred in the series. Three patients experienced partial flap loss. Most of the flap loss occurred in the middle of the incision, with an average size of 3.2 cm × 1.6 cm (from 4.0 cm × 2.0 cm to 1.2 × 0.7 cm). These 3 patients received wound care and secondary healing was seen. No patients endured complete flap loss that required repeat operation. Additionally, postoperative pain, although not well-documented, was normally well controlled by patient-controlled analgesia, and patient-reported pain was usually none to mild.

The benefits of NPWT in wound care quickly led to expansion of its use beyond its original purpose. The proposed mechanisms of NPWT on wound closure include stimulated angiogenesis and increased blood flow to promote the formation of granulation tissue, reduce exudation and local edema, and decrease rates of wound infection.^4,18-20 Furthermore, the NPWT system exerts micromechanical forces on individual cells in the wound bed, resulting in stretching of the cellular cytoskeleton. This leads to the release of second messengers that promote cell mitosis and new tissue formation.^18,24

In addition to wound management, NPWT has been applied directly over the skin to promote healing. Incisional NPWT is the application of NPWT over a closed incision, reducing lateral tension across the wound and controlling the incidence of infections.²⁵ Clinical studies have shown that incisional NPWT can be beneficial to reduce complication rates in contaminated surgeries (sternal incisions^14,15 and laparotomy incisions^16,17), surgeries involving artificial implants beneath the incision (hernia repair,²⁶ vascular surgery,²⁷ and joint surgery²⁸), and arthroplasty procedures. In addition to use over incisions, NPWT has been applied over pedicled random style cutaneous flaps,²⁹ pedicled muscle flaps,³⁰ and free muscle flaps,³¹ although this approach is controversial.

Clinical studieshave shown that bilayer NPWT may accelerate the healing process of refractory seroma after mastectomy²¹ and reduce the risk of total complications at the saphenous vein donor site after bypass surgery.²² Although the bilayer NPWT mechanism was hypothetical, this therapy had been safely applied over a series of free flaps, and rapid fluid removal capability was observed.^32,33 In other research, this therapy helped facilitate safe closure of infectious laparotomy wounds.³⁴

Inguinal lymph node dissection

Although the previous study showed that bilayer NPWT can be safely applied in refractory postmastectomy seroma²¹ and after great saphenous veins harvesting for coronary artery bypass graft,²² it is unclear whether it is safe to apply this therapy to flaps in ILND. The inguinal lymph node dissection, also known as the groin dissection, was devised by Basset in 1912²³ and variations have been developed by surgeons from different countries, including expansion of the scope of surgery,³⁵ modification of the surgical incision,^36,37 and robotic surgery.³⁸

Inguinal lymph node dissection is used for the staging and treatment of malignant skin tumors in the lower limbs and perineum. Indications for this operation include malignant melanoma, squamous cell cancer, and penile cancer. Not generally considered a preventive operation, ILND has been shown to be a safe first approach to complete lymph node dissection in patients with a positive sentinel lymph node biopsy.³⁹ Despite numerous studies and subsequent progress using ILND, the incidence of surgical complications from this procedure remains high. Södermanet al⁴⁰ performed a meta-analysis of complications following inguinal and ilioinguinal lymphadenectomies that showed an overall complication rate of 52%, including lymphocele, wound dehiscence, flap necrosis, hematoma, seroma formation, wound infection, and chronic scrotal or lower extremity edema. These short-term complications serve to interrupt wound healing and may result in severe consequences that require multiple procedures for reconstruction.^41-43

In this study, bilayer NPWT was applied after ILND to evaluate the safety and efficacy of this therapy with promising results. Bilayer NPWT may safely be applied in the surgical region after ILND and can provide potential benefits in terms of wound healing. Compared to historical studies, the total complication rate in this study was dramatically reduced from 52%⁴⁰ to 6.5%. Similar to a typical incisional NPWT, bacteria were physically secluded from the wound, and the total post-ILND infection rate significantly decreased. Literature reported a post-ILND infection rate of 6%,⁴⁴ but no infection occurred in this series. Wound dehiscence and lymphocele were successfully controlled because the flaps were secured to the wound base with pressure. Interestingly, the flap necrosis rate was not high in this case series (6.5%) and was lower than reported in the meta-analysis (10%), supporting the hypothesis that pressure generated by NPWT may not disturb the blood supply, even in compromised flaps.

This study is limited by the use of a porcine model. Although the flap model in the pork loaf may reflex mechanical feature partially, it does not represent live tissue. This study is additionally limited by the small number of included patients and lack of a control group; rigid conclusions cannot be achieved in the current analysis. The pressure used in most NPWT studies, as in the current research, was set at -125 mm Hg. The authors could not determine whether this pressure was the most appropriate for bilayer NPWT, underscoring the need for further studies to investigate the safest and most effective settings for bilayer NPWT. Additionally, the authors did not routinely document postoperative pain, which may be considered a limitation.

Obesity significantly increases the risk of wound complications, including infection and wound breakdown after inguinal lymph node dissection.⁴⁵ Studies conducted in Japan likely comprise patients with a lower BMI score than noted in patients in the United States.⁴⁶ In the present study, the authors did not consider the effect of obesity on ILND. In a Japanese study⁴⁷ in which patient BMI was 22.5 ± 3.5 kg/m², bilayer NPWT had a positive effect on ILND. The present authors plan to further investigate the therapeutic effect of bilayer NPWT in patients who are obese.

This 2-part study involving a porcine model and retrospective data from patients undergoing ILND followed by bilayer NPWT showed bilayer NPWT may aid wound healing by minimizing the subcutaneous dead space. The porcine model showed that bilayer NPWT was advantageous to traditional power suction in exudation removal. The authors encourage future randomized controlled trials to confirm the benefits for wound healing

Authors: Wenjin Wang, MD; Jialing Li, MB; Junhui Li, PhD; and Hongda Bi, PhD

Affiliation: Changhai Hospital, Shanghai, China

Correspondence: Hongda Bi, PhD, Department of Plastic Surgery, Changhai Hospital, 168 Changhai Road, Shanghai 200433, China; bihongda0411@163.com

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