The Efficacy of Topical Negative Pressure in the Management of Infected and Non-infected Wounds

Alper Sari, MD; Atilla Fesli, MD; Tolga Yener, MD; Yavuz Basterzi, MD; Ferit Demirkan, MD

Abstract: This study evaluates the efficacy of a vacuum-assisted closure (V.A.C.® Therapy, KCI, San Antonio, Tex) device in the comparative management of clean and infected wounds. Vacuum-assisted closure was applied to 57 wounds of 51 patients. Methods. Our protocol consisted of debridement of all necrotic tissue followed by vacuum-assisted closure therapy along with appropriate antibiotic administration. In 5 cases with peripheral circulation impairment, vacuum-assisted closure therapy was terminated due to a poor tissue response. In the remaining 52 wounds, healthy granulation tissue generation was observed. Wound cultures obtained from these patients prior to the start of vacuum-assisted closure proved the presence of infection in 31 wounds, while the other 21 wounds were free of infection. Results. The average sizes of the infected and non-infected wounds were 55.77 cm2 and 57.94 cm2 prior to the start of vacuum-assisted closure, respectively, while they were reduced to 48.28 cm2 and 45.70 cm2 after the last session. At the conclusion of vacuum-assisted closure therapy, 42 wounds were skin grafted and 10 wounds were covered with skin/muscle flaps. Conclusion. Vacuum-assisted closure therapy is a reliable tool in the management of almost any type of wound, whether infected or not infected, unless major circulatory impairment interferes and acted as a contributory factor in wound formation.

Address correspondence to:
Alper Sari, MD
Mersin University
Çiftlikköy Merkez
Kampusu 33343
Phone: 903 243 37 4300

     Clinical trials regarding the beneficial effects of controlled negative pressure on wound healing were initially reported in Russian journals in the mid-1980s, then in German and French journals in 1996, and later in English journals in 1997.1–9 Since its first application as an alternative to conservative wound management modalities, the indication spectrum of topical negative pressure application has grown tremendously.10 At first, it was indicated for enhancing healing of nonhealing chronic wounds, but later it was proven effective in the management of a variety of different wound scenarios, such as acute wounds, exposed surgical hardware, skin graft take, and as an aid in the survival of muscle flaps.10,11

     We have been using topical negative pressure therapy in the management of both acute and chronic wounds since November 2004. The present study evaluates the effectiveness of this unique wound management modality in the management of clean and infected wounds.

Material and Methods

     Patients who presented to the authors’ clinic between November 2004 and December 2007 with either clean or infected wounds and were treated with topical negative pressure therapy with the vacuum-assisted closure device (V.A.C.® Therapy, KCI, San Antonio, Tex), were included in the study.

     Initially, the localization and etiology of the wounds were documented. The wound beds and the type of tissue covering the wound beds were evaluated. The presence or absence of granulation tissue and necrotic components were also recorded. The wound beds were further inspected for the presence of any exposed anatomic structures, such as tendon and/or bone, or the presence of an exposed alloplastic material. The presence of exudation and perilesional erythema were considered signs of inflammation or infection.

     The surface area of each wound was measured with a digital wound area measurement device (Visitrak®, Smith and Nephew, London, UK) before and after vacuum-assisted closure therapy was completed. The time from the start to the end of vacuum-assisted closure therapy and the number of applied vacuum-assisted closure sessions were also recorded.

     Any necrotic tissue, including sequestrated bone fragments and non-viable tendons, were debrided prior to the start of vacuum-assisted closure therapy. Tissue cultures and antibiograms were obtained. All the obtained tissue specimens were directly plated on blood agar and eosin methylene blue (EMB), and were incubated for 48 hours at 37˚C. All isolates were identified morphologically and biochemically by a standard laboratory procedure. An appropriate antibiotic was administered if either a positive wound culture was obtained or if clinical signs of infection were seen.

     Each vacuum-assisted closure session started with cleansing the wound surface with 10% povidone iodine solution and rinsing with saline. The sterile polyurethane foam was cut and shaped into an appropriate size for the wound and was applied to the wound surface. An occlusive drape was placed over the foam to prevent air leakage. The pressure on the device was set to -125 mmHg and was applied continuously for the first 48 hours. After the first session, vacuum-assisted closure was either applied continuously or intermittently due to patient compliance. The start and end of the vacuum cycles caused discomfort in some patients, thus they were switched to continuous therapy. The incidence of air leak was also higher when intermittent therapy was applied to certain areas where airtight drape application was made difficult.

     The presence of proliferating, brightly colored granulation tissue and fresh bleeding from the wound bed were considered signs that the wound was improving. Vacuum-assisted closure therapy was terminated and the wound was covered with either a skin graft or a flap if high quality granulation tissue was observed in wound bed.

     In cases where a wound did not respond positively to vacuum-assisted closure therapy after the first two sessions (ie, no granulation tissue had formed, or the wound bed appeared dry and ischemic), vacuum-assisted closure was discontinued and other treatment options were considered. The data of these patients were excluded from the study.

     Statistical analysis was performed using SPSS 11.5 for Windows. The comparison of the responses of infected and noninfected wounds to vacuum-assisted closure therapy in terms of the wound surface area decrease rates was performed via the Mann-Whitney U test; P ≤ 0.05 was considered significant.


     Vacuum-assisted closure therapy was applied to 57 wounds of 51 patients. In 5 cases, the treatment was terminated due to poor or no tissue response after the first two vacuum-assisted closure sessions, (2 of these cases presented with infected, and 2 with noninfected, lower extremity wounds caused by peripheral arterial disease; 1 patient presented with an infected lateral thorax wound due to radiation necrosis).

     In 52 wounds of 46 patients a good tissue response (manifested by the formation of bright red granulation tissue) after the first two sessions of vacuum-assisted closure therapy was recorded, and the treatment protocol was continued as planned.

     Of the 46 patients, 32 were male and 14 were female. The patient ages ranged from 9 to 80 years with a mean of 42.6 years. These patients presented with a total of 52 wounds, 35 of which were lower extremity, 9 gluteal, 5 upper extremity, and 3 on other body parts. In 16 patients, acute traumatic wounds were caused by motor vehicle accidents. Diabetic foot ulcers and pressure ulcers were the second and third most common causes and were documented in 10 and 9 patients, respectively. Other less common etiologic factors, such as thermal injuries, frost bite injuries, venous foot ulcers, and foot ulcers caused by systemic diseases (eg, sickle cell disease) were documented in other patients.

     The examination of the wounds that responded well to vacuum-assisted closure before and at the end of the treatment protocol is summarized in Table 1.

     Wound cultures obtained before the start of vacuum-assisted closure therapy showed presence of infection in 31 wounds, while the other 21 wounds were not infected. The most common pathogen isolated in wound cultures was Pseudomonas aeruginosa followed by Staphylococcus aureus (Table 2). An average of 4 (8 days) and 5 (10 days) vacuum-assisted closure sessions were applied to infected and noninfected wounds, respectively. The average size of the infected and non-infected wounds was 55.77 cm2 and 57.94 cm2 before the start of vacuum-assisted closure therapy, respectively. Wound size was reduced to 48.28 cm2 (infected) and 45.70 cm2 (noninfected) after the last session of vacuum-assisted closure therapy (Figure 1). Overall, a 19.27% decrease in surface area was recorded for infected wounds and a 20.84% decrease was recorded for noninfected wounds. In terms of wound surface area measurements, no statistically significant difference between the responses of infected and noninfected wounds to vacuum-assisted closure therapy was recorded (P > 0.05).

     All 52 wounds, both infected and noninfected, that responded well to vacuum-assisted closure therapy after the first two sessions demonstrated healthy granulation tissue, which eventually covered the wound bed at the end of the treatment period. Wound cultures were taken again after the completion of vacuum-assisted closure therapy, and in all but 4 wounds, negative cultures were obtained. Re-examination of the wounds with positive tissue cultures revealed the presence of residual, non-vital structures such as tendon, fascia, or sequestrated bony pieces within the wound. These were debrided once more and appropriate antibiotics were continued until eradication of infection was demonstrated by tissue cultures.

     Out of 52 wounds managed by vacuum-assisted closure therapy, 42 were skin grafted and 10 were covered with skin or muscle flaps.

Case Reports

     Case 1. A 53-year-old man presented to our clinic with a nonhealing leg ulcer located on the distal one-third of his right leg. The wound occupied the area over the lateral malleolus with a surface area of 61.6 cm2. The wound bed was covered with fibrinous debris and necrotic tissue along the wound margins (Figure 2A). The wound bed and margins were cleared of all necrotic tissue via surgical debridement (Figure 2B). Tissue culture obtained from the wound bed revealed the presence of Pseudomonas infection. Appropriate antibiotics and vacuum-assisted closure were started. After the fourth vacuum-assisted closure session the wound bed was covered with healthy bright red granulation tissue without any purulent discharge, and the wound surface area was decreased to 57.9 cm2 (Figure 2C). After refreshing the wound edges by limited excision of the margins, definitive closure was achieved by the application of a split-thickness skin graft. Twelve months postoperative follow-up revealed an uneventful healing (Figure 2D).

     Case 2. A 48-year-old woman presented to our clinic with a pressure ulcer located on her left ischial region of 2 months’ duration. Her medical history revealed that she suffered from transverse myelitis for 3 years, and that this was the first pressure ulcer she experienced. At her initial examination the wound bed was covered with necrotic tissue. Purulent discharge from the wound cavity was evident (Figure 3A). After meticulous debridement of all necrotic tissue along with the walls of the ulcer cavity, a skin defect with a 26.2 cm2 surface area was created (Figure 3B). The tissue culture revealed the presence of Pseudomonas infection. Appropriate antibiotics along with vacuum-assisted closure therapy were started. After 4 vacuum-assisted closure sessions, healthy granulation tissue formation within the wound bed was observed and the wound surface area decreased to 17.8 cm2 (Figure 3C). Definitive closure was achieved with the rotation of a fasciacutaneous flap to the defect (Figure 3D). No wound complications were noted in the postoperative period.


     The use of negative pressure wound therapy grew tremendously because physicians now had the opportunity to apply standardized, evenly distributed negative pressure to the entire wound surface. Initially, the vacuum-assisted closure system was developed to manage nonhealing chronic wounds, but it soon proved to be effective in the management of acute wounds— particularly in wounds that resulted from high-energy trauma with severely compromised tissue.10 Studies have shown that vacuum-assisted closure therapy can be reliably used in a variety of wound scenarios in a reproducible manner, and can enhance local blood flow and granulation tissue formation.11–13

     After combining the results of their research experience and that of other studies found in the literature, Morykwas et al11 stated that when subatmospheric topical negative pressure was applied in a controlled way and with an optimal intensity (they advocated the use of -125 mmHg pressure), it enhanced local blood flow and supported the formation of granulation tissue. The effects of vacuum-assisted closure were attributed to the mechanical forces deforming the tissues, which eventually increased the rate of mitosis, and decompressed vessels by removing excess interstitial fluid and relieving edema.8,11,14

     In 5 patients in the present study, vacuum-assisted closure therapy was ineffective because regional circulation was not adequate. In 4 of these patients, the wounds were located in the lower extremity and were complicated by peripheral arterial occlusion. Whenever blood flow (especially to the lower extremity) is impaired due to occlusion of major arteries, a wound management modality alone will not be effective unless circulation is re-established surgically. After experiencing these difficulties in patients with inadequate circulation, it became our practice to postpone the application of vacuum-assisted closure therapy until definitive surgery returned blood flow to the wound environment. However, a management protocol of definitive surgical debridement of all nonviable tissue followed by vacuum-assisted closure therapy along with appropriate antibiotic administration evoked the generation of a healthy, bright red granulation tissue within the wound bed and reduction of the wound margins in all 46 patients who did not have any underlying vascular pathology. The effectiveness of this approach did not differ among infected and noninfected wounds, as the granulation tissue formation rates and the wound surface area reduction rates were not different between these groups.

     The management of infected, nonhealing wounds has always been a controversial issue. Not one definitive modality—either medical or surgical—exists that can facilitate wound closure in every clinical case. Most physicians would agree that the sine qua non of a successful wound management protocol is the debridement of all nonviable tissue along with the foci of infection. During debridement, one must decide whether to remove or retain the exposed bones, tendons, and alloplastic materials.

     In the patients presented here, lower extremity wounds caused by a motor vehicle accident were the most commonly encountered injury. In such wounds, due to the forcefulness of the trauma, progressive tissue necrosis caused by vessel damage and thrombosis is to be expected. In our patients, every effort was made to save an exposed vital structure, such as a bone or tendon, and only nonvital and severely infected tissue components were debrided.

     Bihariesingh et al12 reported their success in the management of complex soft tissue defects occurring after orthopedic surgical procedures. In their series, even when the wounds were complicated with exposed bone or alloplastic materials, a treatment protocol of not removing the implant but aggressive debridement and vacuum-assisted closure application, followed by flap closure, led to an eventless closure of the wounds without a need to remove the exposed structures or materials. It has been stated that in such scenarios, vacuum-assisted closure helped to relieve chronic edema, increase local blood flow within the wound environment, and induce granulation tissue formation.11,12,14 In most cases of high-energy trauma to the lower extremity, the situation is complicated by the presence of open tibial fractures. We have found vacuum-assisted closure therapy to be promising in such situations as it can act as a reliable temporary cover for the bone until adequate granulation tissue has formed and the patients’ accompanying medical problems can be resolved. However, Bhattacharyya et al15 recently reported that the management of open tibial fractures with vacuum-assisted closure, until definitive surgical intervention, did not warrant the elongation of an infection-free preoperative waiting period. The authors concluded that the classical 7-day time limit is still pertinent in order to close the wound with a minimal risk of infection, even if the wound is managed with vacuum-assisted closure.

     The literature reports conflicting results on the direct influence of vacuum-assisted closure on the microbiological profile of an infected wound. Classically, it is accepted that surgical closure of an infected wound is contraindicated unless the foci of infection is debrided, as bacteria may easily populate within the wound cavity and spread to neighboring tissues. In the same way, closed wound management modalities may carry a risk of spreading infection when applied over infected wounds. This is generally correct for stationary conventional wound dressings as they cannot efficiently remove wound exudate that contains bacteria, cellular debris, and in some instances, high amounts of proteases. Contrary to these conventional wound dressings, the vacuum-assisted closure system removes exudate from the wound surface via the application of a continuous or intermittent negative pressure to the entire wound surface. As the exudate is removed, cellular debris, bacteria, and proteases, are also expected to be taken away from the wound.8,11 In an experimental porcine model study, Morykwas et al8 reported a significant reduction in the bacterial levels in wound tissue treated with topical negative pressure therapy. Nevertheless, in a prospective, randomized, clinical study, Mouës et al14 compared quantitative bacterial loads after the wounds were managed by either vacuum-assisted closure or conventional moist gauze dressings and found that the results did not differ significantly between the two groups. Still, the authors observed that vacuum-assisted closure therapy was more effective in the management of infected wounds as it caused a faster reduction of wound surface area and faster formation of red granulation tissue within the wound than the conventional dressings. Weed et al16 reported similar conclusions and stated that bacterial bioburden within a wound was not lessened with the application of vacuum-assisted closure in their series, but that this did not interfere with successful wound healing outcomes. They reported that even in the presence of unchanged or increased bacterial numbers during its application, vacuum-assisted closure was effective enough to support wound healing.


     This case series presents vacuum-assisted closure therapy as a reliable tool in the management of almost any wound type, whether infected or not infected, unless major circulatory impairment interferes and was a contributory factor in wound formation.

     Nevertheless, it must be remembered that the application of topical negative pressure to infected or noninfected wounds must be regarded as an important component of a wound management protocol that should also include debridement of devitalized tissues, tissue culture evaluation, and proper systemic antibiotics administration (if necessary).

     Any physician who treats wounds inevitably struggles with a multitude of obstacles in the unforgiving field of wound healing. In the current practice of wound care, any relevant tool that facilitates wound closure is appreciated, especially if it is capable of generating consistent granulation tissue and can assist in wound reduction, even in the presence of infection. These precedents can be obtained with proper application of topical negative pressure on wounds, and the authors believe that the indication spectrum of this unique wound management modality will continue to increase exponentially.

From Mersin University, Department of Plastic and Reconstructive Surgery, Mersin, Turkey

Back to top