Disclosure: This study was funded by Providence Hospital Research Department, Southfield, Michigan. Negative pressure wound therapy (NPWT) (Vacuum-Assisted Closure®, V.A.C.®, Kinetic Concepts Inc., San Antonio, Texas) has been shown to accelerate wound healing.[1] This may be due to removal of interstitial fluids, which could increase oxygen and nutrient delivery to tissues, leading to increased vascularity and granulation tissue, decreased bacterial colonization, and removal of inhibitory factors from chronic wound edema. The favorable effect of NPWT as a bolster for skin grafts was demonstrated in a clinical study.[2] The purpose of the current study is to elucidate the effect of NPWT on the histology of the healing skin graft. Study Under general anesthesia, two rectangular wounds were made on the back of a pig using a dermatome. Multiple stab incisions were made in the harvested split-thickness skin grafts before they were applied with staples over the wounds from where they originated. Each graft measured 15x5cm and was subdivided into five different zones, each one measuring 5cm2. At the end of the procedure, a bolster dressing, consisting of a layer of gauze (Xeroform® 3% bismuth tribromophonate, Kendall Healthcare, Boston, Massachusetts) covered with absorbent cotton balls soaked with normal saline was tied as a compressive dressing over the graft on the left side using nylon sutures. The right side received the NPWT system, which was applied over the gauze and preset to continuous negative pressure of 125mmHg (Figure 1). A specially designed jacket was placed on the animal’s back and torso to protect the grafts. Management of the animals, including pain control, was conducted under an IACUC-approved Providence Hospital protocol. Three days postoperatively, the dressings were removed, and both grafts took nicely without any signs of infection (Figure 2). Four 3mm punch biopsies were obtained from designated areas on both grafts. This procedure was performed under general anesthesia. At the end of the procedure the Xeroform gauze, sponge gauze (Versalon®, Kendall), and latex-free elastic bandages (Dynaflex®, Johnson & Johnson Wound Management, Somerville, New Jersey) were applied as dressings over both skin grafts. Subsequent procedures to collect histological samples were performed every other day on postoperative Days 5, 7, 9, and 11. The animal was euthanized on postoperative Day 11. The histological samples were embedded in paraffin and fixed with haematoxylin and eosin. These were anonymized to ensure their unbiased laboratory evaluation by the histopathologist. The following histological characteristics were studied (Figure 3 and Table 1): Wound edema (0=none, 1=mild, 2=moderate, 3=severe); plane of separation between the graft and the recipient wound bed (0=none, 1=very narrow, 2=narrow, 3=wide, 4=very wide). The statistical test was performed with repeated measures one-way analysis of variance (SPSS, 11.5; SPSS Inc., Chicago, Illinois). For wound edema, the probability of obtaining the respective means over the five time points (Days 3, 5, 7, 9, and 11) for the two groups, NPWT and bolster, by chance is less than 0.001. For the plane of separation, the p value is 0.012. Results See Figure 3 and Table 1. As was noticed in the authors’ study, NPWT application led to decreased wound edema throughout the follow-up period. This effect was even seen in the biopsies obtained on postoperative Day 11, even though NPWT was only applied the first three days after grafting. In contrast, the side covered with the tie-over bolster dressing showed persistent significant edema throughout the experiment. Furthermore, a decrease in the plane of separation between the graft and the wound bed was seen on the NPWT side on Days 3, 5, and 7. Complete disappearance of this plane with replacement by fibroblasts and collagen deposition was seen under the graft on Days 9 and 11. In contrast, this plane remained visible on the tie-over bolster side even on Days 9 and 11. Gross examination of the slides showed a similar pattern of progression on both sides, from acute inflammatory reaction with neutrophils present in the field to chronic inflammatory reaction with lymphocytes, histocytes, and a few giant cells, with a small difference in favor of the side covered with NPWT. The inflammatory reaction was centered at the superficial aspect of the plane of separation in the NPWT side as compared to the bolster side where cell distribution was more homogeneous on either side of the plane of separation (Figure 3, NPWT Day 3 and 5). Discussion Clinical studies on using NPWT as a bolster for skin graft confirmed that this technique is efficacious in increasing graft take due to total immobilization of the graft, thereby limiting shear forces, eliminating fluid collection, bridging the graft, and decreasing bacterial contamination.[3–5] In the authors’ study, two of the four biopsies obtained from the bolster side on postoperative Day 3 showed blood accumulation under the graft despite the multiple stab incisions made on the graft prior to application and the pressure applied by the tie-over dressing, when no fluid collection was seen on the NPWT side (Figure 3). In the authors’ clinical practice, the authors use NPWT as a bolster for the skin graft in certain situations, such as when an increased risk of bleeding is present, oozing from the wound is expected, when the wound bed is edematous at the time of grafting, or in cases where a regular dressing is difficult to apply and secure (e.g., an area with irregular contouring, such as extremities, axilla, perineum, or the inguinal area). NPWT avoids shearing forces and applies equal pressure on these grafted areas. In general, the authors remove NPWT on postoperative Day 3 to assess the take and rule out infection. This prompted the authors to only apply NPWT for three days in this experiment to study the histological effect and correlate the results with clinical observation. Although these results are limited to one individual animal, the positive findings suggest that additional studies of the process need to be performed to appreciate the variability of individual response to the effect of NPWT on the healing process of skin grafts. Conclusion The authors’ histological observations demonstrate the favorable effect of NPWT on skin graft take. This effect is translated by decreased edema with the removal of interstitial fluids and decrease of the plane of separation between the graft and the wound bed, which could increase oxygen and nutrient delivery to tissues leading to accelerated wound healing. This brief communication may prompt further clinical investigations to confirm our findings.