Effectiveness of Negative Pressure Closure in the Integration of Split Thickness Skin Grafts

Abstract

Objectives:

To determine the effectiveness of the negative pressure closure (NPC) technique in the integration of split-thickness skin grafts (STSG) to the recipient site.

Methods:

Randomized, double-masked, controlled trial. Setting: A tertiary burn unit. Patient characteristics: Between May 2003 and October 2004, 60 patients having wounds with skin loss which hindered primary closure, were incorporated to this study. We excluded patients with ≥20% of total body surface burns, polytraumatized, surgical contraindications, those who were enlisted in other clinical trials, and those who rejected the informed consent. Interventions: In all the patients, surgical cleaning of the recipient site and STSG were performed after which they were randomly assigned between 2 groups: a group that received a NPC dressing and were connected to the central aspiration system at −80 mm Hg versus a control group with similar dressing but without connection to negative pressure. Loss of STSG area at the fourth postoperative day, days of hospital stay.

Results:

Sixty patients were included. The median loss of the STSG in the NPC group was 0.0 cm² versus 4.5 cm² in the control group (P = 0.001). The median hospital stay was of 13.5 days in the NPC group versus 17 days in the control group (P < 0.001).

Conclusions:

The use of NPC significantly diminishes the loss of STSG area, as well as shortens the days of hospital stay. Therefore, it should be routinely used for these kinds of procedures.

This randomized controlled trial addresses the effectiveness of topic negative pressure versus placebo in the integration of partial thickness skin grafts. Sixty patients were evaluated showing a remarkable reduction in graft loss area, length of stay, and regrafting needing in the active group.

The management of complex wounds has been a challenge to physicians since historical times. During a great part of the history of medicine, the management of injuries has been based on the use of dressings and substances for external use (creams, ointments, etc.) that has resulted in discreet effectiveness and produced unsatisfactory results, especially in the handling of injuries with important loss of the dermo-epidermic layer where primary closure is impossible (such as in burn wounds, degloving, flap wounds, etc.).

One of the principal tools that the modern surgeon counts on is the split-thickness skin graft, a relatively simple versatile technique that can guarantee cutaneous covering for practically any defect. The integration of these grafts to the reception area consists of 3 phases: revascularization, lymphatic revascularization, and reinnervation.¹ The major causes of skin graft loss are the result of the formation of blisters or of hematoma under the graft that interferes directly with the serous imbibition and revascularization process, and the infection of the graft that frequently leads to partial or total graft loss. Thus, further interventions are required to close the skin defect and discharge the patient successfully.

The efficacy of negative pressure closure (NPC)² was initially described by Morykwas and Argenta in the United States, and modified in Chile by Medina.³ The technique is based on the use of a closed sealed system that places negative pressure over a surface producing traction of soft tissues, diminishing the surface and depth of the injury^3–5 and improving its irrigation. We have modified the original Vacuum Assisted Closure (VAC) technique by using a less dense polyurethane dressing and providing a vacuum by connecting to the central aspiration system of the hospital, instead of connecting to a designated vacuum pump. These modifications were considered because there was a connection available to the central aspiration system at each patient's bedside, which did not add any extra cost to the procedure. The designed dressing costs (components described in Methods) around U.S. $4 per unit while the original VAC pump costs around U.S. $6800 to purchase or U.S. $30 a day to rent, and the original pack of 10 dressings have a cost between U.S. $225 and U.S. $420. The modified technique is safe and does not interfere with the vacuum seal.

Given the positive impact of the vacuum closure on the characteristics of integration of split-thickness skin grafts (STSG), it is our hypothesis that negative pressure diminishes the area of skin graft loss by improving the contact zone of the graft layer, and thus directly helping plasmatic imbibition and the vascularization process^6–8 by preventing the accumulation of serum and improving the initial integration.⁹

To prove our hypothesis, a prospective randomized trial was designed to evaluate the effectiveness of negative pressure during the first phase of the integration of skin grafts.

The main objective is to determine if the NPC diminishes the area loss of the skin grafts. The secondary objectives are 1) to determine if the NPC shortens hospital stay, or if secondary wound coverage procedures are needed; and 2) to determine if there is a relation between the area loss of the skin graft and the total grafted surface.

METHODS

A randomized, double-masked, controlled trial. The general design of the study is illustrated in Figure 1.

FIGURE 1. Design of the trial.

The study was performed in the Burn Unit of the Hospital del Trabajador (CPQ-HTS), in Santiago, Chile, a national referral center for work-related injuries, which account for more than 90% of the admissions to the hospital.

Between May 2003 and October 2004, all patients who were admitted at the hospital with acute traumatic injuries and skin loss, which hindered primary closure, underwent a surgical cleaning of their wound, and had a quantitative biopsy culture performed on their wound. Those patients whose biopsy culture had a bacterial count lower than 100,000 colony forming units per gram of tissue¹⁰ and who had given informed consent to enter the study were recruited.

We excluded patients who: had burns that covered 20%, or more, of their total body surface; were polytraumatized; had surgical contraindications because of medical, anesthetic (serious associated pathology), or surgical cause (hypoalbuminemia or systemic infection); or were enlisted in other clinical trials.

Treatment assignment was performed using computer-generated random numbers in permuted blocks of n = 6. The treatment allocation of each patient was performed by the nurse of the operating room who knew the corresponding assignment. The corresponding treatment was notified to the surgeon only once the skin graft had been performed so that the surgeon did not modify his technique according to the assignation of treatment.

Surgical Cleaning

All the patients were operated on by the same surgical team. They all received antibiotic prophylaxis with a single dose of 1g of intravenous cephazoline (Laboratorios Chile S.A. Marathon 1315.26888 Reg ISP S- 8353/01; Reg Sanit Guatemala PF 28899). The site to be grafted was cleansed with chlorhexidine (Instituto farmaceútico Labomed Reg. ISP 33944; Avda Presidente Frei Km 21, Santiago, Chile) and a saline solution. Afterwards, the devitalized tissue was removed and the borders were evened. The viability of the tissue was assessed visually by the surgeon, and infection was assessed by analyzing clinical parameters (edema, erythema, pain, and purulent discharge) and quantitative biopsy cultures.

Taking of the Grafts

The donor site was cleansed with chlorhexidine and a saline solution. Then, grafts of 0.12 mm of thickness were taken with an electric dermatome (Electric Dermatome Padgett, Model B, year 2001 Padgett Instruments Inc.) or a pneumatic dermatome (2001, Zimmer Patient Care Division).

These grafts were fenestrated with a scalpel no. 11 or meshed with a dermacarrier in a ratio of 1:1.5 (Zimmer Meshgraff II Indiana 2195–01), and they were fixed at the recipient site with either nylon sutures 4.0 (Ethicon, Johnson & Johnson, Sao Paulo, Brasil) or metallic staples (Weck-Visistat, Weckn closure systems T. M.). The method of taking, meshing, and fixing the graft was left up to the surgeons and had no influence on the randomization that was performed after the procedure. Once the graft was fixed, the patient was randomly assigned to one of the following groups:

Active Group

The STSG was covered with a single layer of paraffin gauze dressing (Jelonet, Smith & Nephew, England) (Fig. 2C); then, 3 sheets of polyurethane (high-density foam, Nuris Luisa, Santiago, Chile) with a fenestrated silicone drainage tube between the layers (Fig. 2D–E) was placed over the gauze and covered with a transparent adhesive dressing (Op site, Smith & Nephew) providing the vacuum seal (Fig. 2F). We used a double layer under the tube to prevent pressure ulcers at the bed of the suction tube. Then, the tube was connected to the central aspiration system at a pressure of −80 mm Hg, and the integrity of the vacuum seal was tested and reinforced with cotton dressing and an elastic gauze bandage if necessary.

FIGURE 2. A, Schematic drawing of the NPC dressing. B–F, The procedure sequence.

Control Group

The STSG was covered with the same 3 sheets of polyurethane along with a silicone fenestrated tube, translucent adhesive dressing, and flexible gauze, and only differed from the active group in that it lacked a connection to the central aspiration system (ie, negative pressure vacuum).

Postoperative Care

After the intervention, patients were confined to bed rest, allowing limited movement only for personal hygiene. No extended wound care was performed until the fourth postoperative day when all the wounds were uncovered.

Outcomes

All the wounds of the patients were uncovered on the fourth postoperative day, which was when the principal variable was measured (ie, area of graft loss).

The main outcome was the area loss of the skin graft in cm². Digital photographs, without flash, were taken from a distance of 40 cm from the wound to evaluate the areas of loss at the moment of uncovering the graft. (Camera used: Cybershot DSC–505V, Sony Corporation, Japan; Cybershot DSC-T1 Sony Corporation Japan). Afterward, these images were analyzed with the Autocad (Autodesk, Inc, 2001, Fremont, CA) software, measuring the loss in cm² (Fig. 3).

FIGURE 3. For the measurement of area loss, first the total area (left panel) is marked and then the area loss (right panel) is delimitated.

Secondary outcomes were: the need for regrafting, ratio between graft size and loss area, and number of days of hospital stay between the date of intervention and patient discharge from the burn unit and regrafting requirement. We must mention that, due to the characteristics of our hospital (health insurance system for associated workers), we generally perform nonambulatory procedures, independent of the area grafted, and discharge patients only once the wounds have fully healed to sure the best care for the wound and to avoid readmission due to improper self-care. As a matter of policy, we aim at early work reinsertion with the lowest readmission rate possible, especially due to the fact that our hospital is a national derivation center, and many of our patients reside from distant locations, which makes it impossible for them to travel back and forth to the hospital in terms of costs.

Other variables such as the total grafted area, age, sex, and cause of the wound were also considered.

Masking

The person in charge of measuring the areas in the photographic register was masked to which intervention the patient had received. That way, at the moment of evaluating the photograph, this person did not know whether or not the patient had undergone NPC. In addition, the data analyst was masked to the groups of intervention at the moment of analyzing the results.

Statistics

By making an adjustment of 20% for possible losses, 30 patients in each group will have a power of 90% to detect a difference of 15 to 5 cm² (SD 15 and 5 cm², respectively) with an alpha level of 5%.

The t test and Wilcoxon rank sum test were used for the analysis of continuous variables for simple comparisons while the analysis of variance and Kruskal-Wallis test were used for multiple comparisons according to their distribution (Shapiro-Wilk test). Categorical variables were analyzed by using the χ² or Fisher exact test. Adjustment for confounding variables was performed by using stepwise multivariate linear and logistic regression models.

All the patients were analyzed in the group to which they were assigned, adhering to the intention-to-treat principle.

The continuous variables are described as median and range, and the categorical variables as percentage unless they are indicated otherwise. A two-tailed alpha level of 5% was considered statistically significant. All the confidence intervals are expressed at 95%. The STATA software 7.0 (Stata Corporation) was used for data analysis.

An informed consent of all the patients in compliance with the Helsinki Committee Standards was obtained. The authors have no conflicts of interest with any of the manufacturing companies of the products used in the present study.

The study was examined and approved by the Institutional Review Board of the Hospital del Trabajador de Santiago, Chile.

RESULTS

Patients

Between May 2003 and October 2004, 60 patients were selected to whom 60 procedures were performed. There were no statistically significant differences in the demographic variables between the groups. The socio-demographic variables and baseline characteristics are shown in Table 1. Two grafts were meshed in the active group versus none in the control group (P = 0.492), and 25 grafts were fixed with staples instead of sutures in the active group versus 23 in the control group (P = 0.748). No patient developed nosocomial pneumonia or other complications.

TABLE 1. Patients Baseline Characteristics by Treatment Allocation Group

Among the burned patients, there were no smoke inhalation injuries.

Ineligible Patients

In our study, there were no drop-outs, drop-ins, or noncomplying patients. Furthermore, no patients refused to take part in the study.

Main Outcome

Main and secondary outcomes are shown in Table 2. The median graft loss was 0.0 cm² (range, 0.0–11.8 cm²) in the NPC group and 4.5 cm² (range, 0–52.9 cm²) in the control group (P = 0.001; Fig. 4). The median percentage of graft loss in the NPC group was 0.0% (range, 0.0%–62.0%) versus 12.8% (0%–75.9%) in the control group (P < 0.001).

TABLE 2. Main and Secondary Outcomes by Treatment Allocation Group

FIGURE 4. Box-plot explanation: upper horizontal line of box, 75th centile; lower horizontal line, 25th centile; horizontal bar within box, median; upper horizontal bar outside box, 90th centile; lower bar outside box, 10th centile, extreme values as dots. Note that all the stadigraphs are smaller (lesser) in the NPC group.

After we adjusted for total grafted area, the number of days from injury to intervention, sex, age, and wound etiology, the difference was maintained in favor of NPC (coefficient, −6.9; confidence interval, −10.8 to −3.0, P = 0.001), showing that independent of age, sex, and wound etiology, the NPC favorably impacts on the graft taken. Furthermore, in the same model, we found a direct relation between the total grafted area and the extension of the graft area loss (coefficient, 0.2; CI, 0.1 to 0.3, P < 0.001).

We did not identify any pattern of graft loss in the grafted area. The losses were uniformly distributed over the entire grafted area (at the center as well as at the edges).

Secondary Outcomes

Twelve patients in the control group versus 5 in the NPC group required regrafting (P = 0.045), which was directly related to the lost graft area (0 [0 to 14.0] vs. 11.8 [2.1 to 52.9] cm² of lost area between no regrafting vs. regrafting respectively). The median time from intervention to discharge from the unit was 8 days (range, 7–13 days) in the NPC group versus 12 days (range, 7–23 days) in the control group. Total length of stay was 13.5 days (range, 11–22 days) in the NPC group versus 17 days (range, 10–31 days) in the control group (P = 0.010). This difference maintained its statistical significance (P < 0.001) after it was adjusted for total grafted area, sex, age, days from admission to grafting, and wound etiology.

When analyzing the influence of the area grafted and graft loss, we observed that in the control group the area loss is directly related to the size of the graft (slope, 0.41; CI, 0.22 to 0.57; R2, 0.50, P < 0.001), while in the NPC group this relation is not observed (slope, 0.01; CI, −0.02 to 0.05; R2, 0.03, P = 0.387), as shown in Figure 5.

FIGURE 5. In the left scatter plot, the correlation between graft size and graft loss in the control group is shown. Note that this correlation is truncated with the use of NPC dressing (right scatter plot).

DISCUSSION

In our trial, we found clear differences in favor of the use of NPC in the outcome of STSG. These differences were not influenced by confounding variables such as the cause of the wound, patient's age, or total grafted area. The analysis of data showed a direct relationship between the size of the grafted area and the loss of area. This relation is truncated with the use of NPC, which directly decreased the need for secondary procedures and as a result a reduction of the days of hospital stay with all the benefits that this implies.

Through our work and with a rigorous and clear design, we confirmed the findings of previously published studies of other authors:^9,11–13 the use of VAC has a relevant positive impact on clinical parameters such as loss of graft and length of hospital stay.

Based on the results obtained, we think that the NPC is a safe, useful, inexpensive, and simple technique for the treatment of wounds that are covered with dermo-epidermic grafts and should be used routinely as a coadjuvant treatment of these kinds of wounds. The findings of our study challenge us to develop new systems that could be used on patients with larger defects, such as major burns, or patients who require extensive cutaneous covering.

ACKNOWLEDGMENTS

The authors thank Hector Roco, Ricardo Roa, and Patricio Léniz for their invaluable contributions in the patient recruitment and support throughout the trial.