The Impact of Negative Pressure Wound Therapy on Orthopaedic Infection

By hastening the resolution of edema and improving local microcirculation, topical negative pressure wound therapy (TNP) aids the establishment of early wound coverage. Its use in the setting of type III open fractures is reviewed. The author’s initial use of TNP for closed surgical incisions and how it morphed its way into being applied to closed surgical wounds with heightened likelihood for infection is presented. Several case studies are presented to illustrate the role and the technique for management of acute or subacute infections involving bone and implant.

Key points

  • Topical negative pressure wound therapy (TNP) is used as a dynamic dressing for high-energy wounds following surgical debridement. The author provides support for its utility in this setting as a means of minimizing tissue demarcation (tissue death), the “fuel” of infection.

  • TNP is helpful in effecting early wound coverage by eliminating edema and by acting as a dynamic bolster for an applied split skin graft or “artificial skin.”

  • TNP has been shown to enhance the survival of random pattern flaps—not infrequently a consequence of the surgical extensions placed on a limb’s transversely directed traumatic wound.

  • The use of TNP in the setting of closed surgical wounds has emerged as a safeguard to surgical site infection in certain problematic wound types such as those occurring in the obese patient or those occurring in periarticular fractures of the ankle and tibial plateau.


The author’s personal experience dates from the early 1990s when negative pressure was first introduced by 2 of his colleagues at Wake Forest Medical Center, Lou Argenta, MD and Michael Morykwas, PhD, as a strategy for potentially shortening the hospital stay of patients with decubitus ulcers. The author was one of a small group of clinicians at Wake Forest who gained an early experience with “the VAC” (vacuum-assisted closure, or formerly, the “DecubiVAC”). Given his subspecialty in orthopedic trauma, his proximity to Drs Argenta and Morykwas, and the ready availability of the product, his use of topical negative pressure in wound management quickly became a commonality. He has been a strong proponent of its use in different clinical settings. Those settings, which are relevant to infection, are elaborated upon in this article.


The author’s personal experience dates from the early 1990s when negative pressure was first introduced by 2 of his colleagues at Wake Forest Medical Center, Lou Argenta, MD and Michael Morykwas, PhD, as a strategy for potentially shortening the hospital stay of patients with decubitus ulcers. The author was one of a small group of clinicians at Wake Forest who gained an early experience with “the VAC” (vacuum-assisted closure, or formerly, the “DecubiVAC”). Given his subspecialty in orthopedic trauma, his proximity to Drs Argenta and Morykwas, and the ready availability of the product, his use of topical negative pressure in wound management quickly became a commonality. He has been a strong proponent of its use in different clinical settings. Those settings, which are relevant to infection, are elaborated upon in this article.


Bernard a raison. La peste est rien. C’est la terrain qui est tout. (Bernard is right. The germ is nothing. It is the environment which is everything.)

Ramon Gustilo, MD, teacher, author, and mentor to virtually all orthopedic traumatologists, cited the above in his book entitled, Orthopaedic Infection and wrote, “ It behooves every surgeon to understand fully the implications of Pasteur’s statement. ” Wounded nonviable tissue is the fuel of infection and needs to be excised by surgical debridement. For high-energy wounds, the policy of following the initial debridement with a return to the operating room for a “second look” at 36 to 72 hours has become a protocol at many centers, knowing that the tissues in or adjacent to the zone of injury commonly “demarcate” during that span of time. Tissue “demarcation” is an ill-defined but widely acknowledged concept.

An interesting feature noted in the author’s early experience with topical negative pressure wound therapy (TNP) was the fact that when it was used for high-energy wounds following an initial debridement (the author’s preference was to set the transmitted negative pressure at −50 mm Hg or −75 mm Hg continuous), there was no or minimal secondary necrosis seen at 36 to 72 hours. In other words, there appeared to be something about TNP that prevented the cascade of cellular events that would otherwise proceed to tissue necrosis (or “demarcation”) in those patients.

Relevant to these observations was the study by Morykwas and colleagues on the histology of porcine skin burn wounds, half (on one side of the midline) were managed with saline dressing changes and half (on the other side of the midline) were managed with TNP. Punch biopsies of the burn wounds were obtained on postburn days 1, 3, 5, and 9 ( Fig. 1 ). The basis for the demise of the tissues managed with saline dressing changes is the lack of dynamic microcirculatory flow through the capillary bed probably due to an engorged interstitium in the tissues. Additional support for this statement is provided by the study of Langfitt and colleagues. It is now understood that the reason high-energy wounds “demarcate” by the time of the “second look” is the same as was exhibited in the above described porcine burn study of Morykwas. Given this, one might logically predict that the same phenomenon should apply not only to the tissues at risk in a high-energy extremity wound, a skin burn in a porcine model, but also to any tissue whose initial insult is followed by a second wave of necrosis, such as any infarct with a “reperfusion injury.” This line of thought foreshadowed the results of recent studies by Lindstedt and colleagues, Argenta and colleagues, and Jordan and colleagues on preservation of blood flow effected by TNP in an ischemic porcine heart model as well as Argenta and colleagues in preservation of tissue in a traumatic rat brain model evidenced by water content, histologic sectioning, MRI spectroscopy, injury cavity area, and cortical volume. Zheng and colleagues later obtained similar findings for ameliorating spinal cord injury and traumatic brain injury. A subatmospheric pressure of 100 mm Hg was found to be more efficacious in this porcine model of traumatic brain injury.

Fig. 1

Punch biopsies of the burn wounds were obtained on postburn days 1, 3, 5, and 9. There were no histologic changes seen on either the saline dressing side ( A ) or the TNP side ( B ) at day 1. On day 3, the notable finding on the saline dressing side was a significant zone of coagulation necrosis ( C ). There were no changes on the TNP side ( D ). On day 5, there is a clear zone of cleavage between the necrotic zone and the underlying living tissue on the saline dressing side ( E ). There were no changes on the TNP side ( F ). On day 9, there was significant sloughing of necrotic tissue with loss of specialized skin elements ( G ). There were no significant changes on the TNP side ( H ).

From Morykwas MJ, David LR, Schneider AM, et al. Use of subatmospheric pressure to prevent progression of partial-thickness burns in a swine model. J Burn Care Rehabil 1999;20(1):15–21; with permission.

Morykwas, in his original study of the effects on the porcine animal model, showed that the setting of −125 mm Hg on intermittent mode maximized the granulation tissue response. Thus, there may be several ideal pressures: one for maximizing blood flow in the setting of porcine myoischemia (−50 mm Hg, continuous); one for minimizing the immediate and shortly thereafter effects following a porcine traumatic brain injury (−100 mm Hg) ; and one for maximizing porcine growth of healthy granulation tissue (−125 mm Hg, intermittent).

Given the success at avoidance or minimization of secondary necrosis in high-energy wounds, the author has continued to use and recommend the use of continuous TNP at the −50 mm Hg level applied at the time of completion of the initial debridement. An intermediary layer of xeroform or “artificial skin” (eg, Integra; Integra Lifesciences, Plainsboro, NJ, USA) for desiccation-prone tissue is commonly used. At the time of the “second look” at 48 to 72 hours, the wound is excisionally debrided if any dead (“demarcated”) tissue is present and closed with incisional TNP application. The ideal pressure for sparing tissue demarcation following high-velocity trauma has yet to be completely derived. The concept of a “dose response curve” may apply depending on the nature of the tissue and the nature of the inciting event, which resulted in the demarcation. Supportive of a −50 mm Hg negative pressure level is the author’s experience as well as the work of Lindstedt and colleagues demonstrating a dose response curve where topical negative pressure levels were correlated with maximum myocardial microvascular blood flow as well as the work of Langfitt and colleagues on skeletal muscle.

In the clinical arena, one wound that generates significant soft tissue trauma is the war wound inflicted by shrapnel. Peterson and colleagues start their recent paper on the topic with a 1914 quote from William Osler while attending to British casualties of WWI: “ This is an artillery war in which shrapnel do the damage, tearing flesh, breaking bones and always causing jagged irregular wounds. And here comes in the great tragedy—sepsis everywhere, unavoidable sepsis!… The surgeons are back in the pre-Listerian days and have wards filled with septic wounds. The wound of shrapnel and shell is a terrible affair, and infection is well nigh inevitable .” Dr Peterson’s commentary is: “Ninety years later, his (Osler’s) quote remains pertinent. War wounds are distinct from peacetime traumatic injuries because these higher velocity projectiles and/or blast devices cause a more severe injury and accompanying wounds are frequently contaminated by clothing, soil, and environmental debris.” Peterson and colleagues were stationed on the USNS Comfort , an echelon 3 facility during the Iraq war. They reported that “surgical management of wounds was similar for all patients—WIA (wounded in action), as well as Iraqi civilians. Aggressive debridement of all necrotic tissue was performed in the operating room upon arrival to the USNS Comfort . Further wound care included daily wet-to-dry dressing changes and wound VAC therapy (TNP) depending on the availability of suction on board USNS Comfort . Additional wound debridements were performed as necessary and dressing changes on large wounds were performed in the operating room to assist in patient comfort.” Therefore, in essence, these investigators reported on 211 patients, 56 of whom had shrapnel wounds, managed with initial debridement in the operating room and with subsequent daily wet-to-dry dressing changes in the “follow-on” (postoperative) period and sporadically applied TNP “depending on availability of suction on board Comfort. ” The reported infection rate among the shrapnel-wounded patients in their series was 32.7%.

In November 2006, Leininger and colleagues reported their series of 88 high-energy shrapnel wounds in 77 patients also managed in Iraq at an echelon 3 US Air Force field hospital facility in Balad, with a similar patient demographic consisting of American WIA, prisoners of war, and Iraqi civilians. All were high-energy shrapnel wounds treated with aggressive debridement and irrigation. However, unlike the Peterson report, negative pressure (TNP) dressings were placed at the time of the initial procedure.

To quote the investigators, We believe the most significant benefit was the protection of the wound from the ward environment.

The VAC system isolated the tissue injury, yet still kept it clean and free of exudate. Dressing changes could be accomplished every 2 or 3 days, rather than 3 times per day, allowing us to do them in the cleaner environment of the operating room. The physiologic benefits of such a dressing include the clearance of wound exudate, enhanced granulation from local vasodilatation, and mechanical wound contraction because of pressure differential. . . .” Their wound infection rate was 0%. The overall wound complication rate was 0%. The investigators conclude with the statement, “experience with these patients suggests that conventional wound management doctrine may be improved with the wound VAC (TNP), resulting in earlier more reliable primary closure of wartime injuries.”

Of interest is that these 2 studies were contemporaneous, and the patient demographic for each study was similar, a combination of high-velocity shrapnel-wounded soldiers and civilians. Both facilities were echelon 3 centers, which is a rough measure of injury severity according to the Department of Defense triage system. Acknowledged is the fact that both studies were retrospective and without control groups (level 4 evidence). In any case, the question arises, was the difference in infection rate attributable to differences in the timing and fidelity of the application of TNP, and if so, was there more secondary necrosis in the Peterson and colleagues series to account for their higher infection rate?


Topical negative pressure can be used for staging the coverage of an open fracture wound following debridement and stabilization of the fracture. The use of TNP in this setting helps to accelerate the resolution of edema as well as coverage of small areas of exposed bone, implant, or tendon. The same can be said about fasciotomy wounds after debridement, where topical negative pressure can act as a bridge between the initial (often edema-laden) fasciotomy wound and later closure or skin grafting. If skin grafting is elected, then TNP can act as a dynamic bolster for the applied graft. The author typically uses a 3:1 split graft in this setting to enable continued evacuation of edema and promotion of epithelialization from the graft.

One of the properties demonstrated in the initial series of basic studies done by Morykwas was that survival of random pattern pedicle flaps was higher (by 20%) with TNP. When surgical extensions are put on transversely directed traumatic wounds or when an extension of the original open wound is needed to better expose a fracture for appropriate debridement or osteosynthesis, the application of a topical negative pressure dressing may for that reason help to thwart the development of a partial wound necrosis (and secondary infection) in this setting ( Fig. 2 ).

Fig. 2

Radiograph of an open, right talus fracture dislocation in a 66-year-old male bicyclist struck by a car ( A ). Initially managed by referring surgeon with debridement, reduction of the subtalar joint, pin fixation, and bridging external fixation ( B, C ). A postoperative computed tomographic scan revealed a persisting significant intra-articular displacement ( D, E ). The original wound had a surgical extension (done at the time of initial debride-ment) from the anterior-most point of the traumatic wound directed medially (roughly from edge of sponge) ( F ). To expose the talus adequately, a surgical extension was needed from roughly the mid portion of the existing wound taken 2 or 3 inches toward the base of the fourth metatarsal ( G ). The reflected flaps warrant careful scrutiny in the postoperative period for viability and healing. The use of TNP in this setting is supported by improvement in the survival of at-risk flaps by 20% in the porcine model reported by Morykwas and colleagues. The postoperative x-rays and wound are seen in figures ( H–J ). There was a healthy granulation at their apex ( J ), which went on to heal uneventfully ( K ).

Artificial skin

John Burke, MD and Ionnas Yannas, PhD and their team of researchers at Harvard/MIT developed an artificial skin in the late 1970s primarily as a means of providing coverage for burn patients following an escharotomy. It consisted of 2 layers: the superficial one was a thin layer of silicone serving as a bacterial moisture barrier and the second was a composite of bovine collagen within a glycosaminoglycan matrix. The latter was to serve as a matrix for ingrowing host cells derived from the underlying tissue bed over which it was applied. Once this layer “took,” a very thin skin graft was applied. This overall process took about a month, and although wound-healing time in the setting of the burn patient was reported to be better than autograft, allograft, or xenograft, wound infection and “percentage of graft take” were problematic.

These latter issues were “surmounted” by Drs Anthony DeFranzo and Joseph Molnar, who were both early users of “the VAC” (TNP) at Wake Forest. Each were fully aware of the properties of enhancing the “take” of skin grafts both in their own practice and as reported earlier by their colleagues, Schneider and colleagues. Molnar and colleagues reported the results of Integra grafting with a VAC for their first 8 consecutive cases. The treated wounds included exposed bone in 62.5%, tendon in 37.5%, joint in 50%, and bowel in 25%. The mean time for clinically assessed vascularization of the Integra was 7.25 days (range 4–11 days), with an average incorporation of 96%. The split-thickness skin graft adhered to the bed by 4 days, with a success rate of 93%.

Currently, the incorporation of a dermal substitute (Integra or ACell [ACell Inc, Columbia, MD, USA] and similar) is greatly enhanced by the use of TNP as a dynamic bolster. The reliability of the technique has enabled it to assume the last “rung” on the “reconstructive ladder” for treating soft tissue defects before resorting to a soft tissue flap. Thus, the use of TNP either alone or as a bolster for a skin graft or an “artificial skin” graft is an option for providing early, stable coverage over desiccation-prone tissue such as bone devoid of periosteum, tendon without paratenon, joint capsule, and fascia. One should keep in mind the provision of early stable wound coverage is one of the essential tenets of open fracture management and is a key to the avoidance of infection. High-energy wounds are not infrequently devoid of coverage of bone, tendon, and/or implants.

Topical negative pressure wound therapy: role in management of the IIIB fracture

The author recently reviewed his own experience with the use of combined TNP and Integra for type IIIB open fractures. This work was submitted and accepted for presentation at the 2016 Georgia Orthopedic Society Annual Meeting. This study was a retrospective study of 18 consecutive type IIIB open fractures in 17 patients needing soft tissue coverage. Two patients died during their hospital stay, leaving 16 fractures in 15 patients for analysis. There were 7 tibial shaft fractures, 6 fractures of the ankle, 2 fractures of the foot, and 1 fracture of the humerus. Wound dimensions averaged 52 cm 2 . All had exposed bone at the base of the wounds.

Fixation constructs were intramedullary nail (6), plate and screws (5), bridging external fixation with supplemental screws/pins/plate (4), or screws alone (1). Follow-up averaged 14.8 months (6–37). All wounds healed, although 2 of the 16 required a reapplication of the Integra and one patient required 2 (a small fraction of the total) reapplications. Total time to bone healing (inclusive of a patient with bone loss who successfully underwent the Masquelet technique and consolidated their fracture and bone graft) was 5.5 months (range: 3–11) for the tibia, 3.4 months (range: 3–6) for the ankle, 6.2 months (range: 3.3–9) for the foot, and 3.5 months for the humerus.

The author’s experience with this coverage technique includes the fact that the wounds can take on a worrisome appearance, and it is understandable why many surgeons “throw in the towel” on the technique ( Fig. 3 ). The author is reminded of the observation of Ioannis Yannas, the MIT materials scientist, who, along with Harvard surgeon Dr John Burke, developed Integra in the 1970s. The last molecule category he tested was collagen, and the experiments on animals “didn’t look so good.” Rather than accept defeat in this decade-long research effort, Dr Yannas had the courage to look carefully at their “failures.” The “ah-ha” moment was when he realized that the prolongation of the time for incorporation of the graft was due to the fact that the collagen was actually impairing formation of scar tissue, and by so doing, was allowing precursor cells from the underlying muscle tissue to be drawn unencumbered to the surface and grow as primordial skin (without hair follicles and sweat glands).

Oct 6, 2017 | Posted by in ORTHOPEDIC | Comments Off on The Impact of Negative Pressure Wound Therapy on Orthopaedic Infection

Full access? Get Clinical Tree

Get Clinical Tree app for offline access