➢ Negative-pressure wound therapy (NPWT) has been widely adopted as one of the primary modalities for treating traumatic wounds and aiding in soft-tissue reconstruction. Despite the abundance of peer-reviewed publications, very few high-level studies exist in the orthopaedic literature.
➢ NPWT must be considered as an alternative to traditional skin-graft bolsters, particularly in large, complex wounds with irregular contours or exudation.
➢ NPWT should be used as a temporary wound dressing (if primary closure is not possible) between debridements and as coverage for soft-tissue injuries associated with open fractures.
➢ Incisional NPWT should be considered as an alternative to conventional dressings in high-risk wounds or patients.
➢ Currently, there is inadequate evidence for or against the use of NPWT in combination with antibiotic beads for open fracture management or in cases involving periprosthetic infection or osteomyelitis.
Management of soft-tissue injuries that either are or are not accompanied by open fractures has proven to be challenging. The first clinical reports of negative-pressure wound therapy (NPWT) were published in 19931. NPWT has been widely adopted and indications have expanded to include high-risk closed incisions, skin grafts, and flap coverage. Several review articles2-5 exist on NPWT in the orthopaedic literature.
This article includes historically relevant clinical and basic-science studies in addition to more current literature obtained from a PubMed search (with use of the terms “negative pressure wound therapy” and “vacuum-assisted closure”) of papers published from 2009 through 2013. More than 1300 studies (including a number of Level-I to Level-IV-evidence clinical studies related to acute orthopaedic traumatic and surgical wounds) were identified as being published during that time period.
The primary components of NPWT systems are reticulated open-pore foam or gauze, a semi-occlusive drape, and a suction device with a fluid reservoir. Several manufacturers supply commercially available systems suitable for inpatient or outpatient use; these systems include a variety of foam sizes, pressure settings, modes of therapy, battery life, and reservoir sizes.
The most common foam types available are polyurethane ether sponges (pore size, 400 to 600 µm), polyvinyl alcohol sponges (pore size, 60 to 270 µm), and silver-impregnated sponges. Early studies6 indicated that polyurethane ether sponges have the best potential for granulation. Unfortunately, polyurethane ether sponges are also associated with increased tissue ingrowth, which may be undesirable near blood vessels, nerves, tendon, and ligaments. Therefore, polyvinyl alcohol sponges may be more desirable in such locations to limit tissue ingrowth7. The semi-occlusive dressing is provided in most commercially available systems; however, many surgeons use iodine-impregnated dressings, which are available in large sizes and may be beneficial in minimizing skin and/or wound colonization8,9. The canister size ranges from 45 mL to 1000 mL; one disposable system now has no canister and relies on evaporation.
NPWT systems can be used at many pressure settings and in either continuous or intermittent modes. Morykwas et al.10 indicated that pressure set at −125 mm Hg increased granulation volume (63% more as compared with that in controls) and intermittent cycles (five minutes on and two minutes off) at −125 mm Hg induced the greatest increase in granulation volume (103% as compared with that in controls). A recent systematic review found very weak evidence from which to formulate treatment recommendations with any one NPWT variable, including foam or gauze type and pressure setting11.
NPWT costs can vary depending on the contract between manufacturer and provider. Hospital charge structure can range from a one-time fee to daily rental fees plus disposable charges. Two Current Procedural Terminology (CPT) codes exist for NPWT depending on wound size relative to 50 cm2. CPT codes 97605 and 97606 equate to 0.77 and 0.86 total facility relative value units, respectively. The use of NPWT may reduce long-term costs associated with the use of conventional dressings by improving patient outcomes and requiring fewer dressing changes and less nurse utilization12,13; however, no outcome studies exist in the orthopaedic literature that show a formal cost analysis of NPWT versus controls.
Mechanisms of Action
Wounds left open for extended periods often contract and may not reapproximate with primary closure. NPWT foam shrinks in three dimensions and thus may reduce wound size. At −125 mm Hg, polyurethane ether foam decreases volume by approximately 80%16. Deformation is dependent on the magnitude of suction17, foam volume, pore volume fraction, and the tissue-specific deformability18.
Tissue changes in response to physiologic stresses. NPWT applies a mechanical load to wound beds, resulting in extracellular matrix modulation, cellular proliferation, and neovascularization. Cell shape and gene expression is altered, resulting in the differentiation of myofibroblasts and upregulation of paracrine signaling pathways, including fibroblast growth factor (FGF), transforming growth factor beta (TGF-β), and vascular endothelial growth factor (VEGF)19-23. Alterations in local blood flow produce superficial tissue hypoperfusion and deep tissue hyperperfusion24-26. Hypoxia is present in areas of hypoperfusion, resulting in a VEGF gradient, promoting sprouting angiogenesis27. Areas of hyperperfusion facilitate improved oxygen and nutrient delivery.
Stabilization of Wound Environment
NPWT requires a semi-occlusive drape that limits fluid evaporation and does not allow passage of proteins from the wound except in a controlled manner to the NPWT reservoir. NPWT evacuates such fluid with its associated proteins, theoretically stabilizing the osmotic and oncotic gradients at the wound surface. In addition, the occlusive dressing near perforations in the NPWT drape prevents the desiccation and scab formation that is often seen with use of wet-to-dry dressings28-30. Maintaining a moist wound environment has been shown to improve the rate of wound repair and epithelialization in animal models and humans31,32.
Edema Control and Exudate Clearance
Edema impedes wound-healing. After surgery in dependent areas, elevation and judicious compression limit edema and local interstitial pressures to facilitate cellular proliferation and transfer of oxygen and nutrients. Similarly, NPWT drains extracellular fluid and decreases edema. Compared with conventional dressings, NPWT allows earlier closure of swollen lower-extremity fasciotomy wounds33 and decreases extremity circumference after trauma34.
Secondary Effects of NPWT
NPWT has been reported to increase wound perfusion; however, this has not been fully established. Early studies used noninvasive laser Doppler techniques to indirectly measure blood flow and identified increased blood flow in the wound35. Wackenfors et al.25 indicated that a hypoperfusion zone exists within 1.5 cm of the wound edge in a porcine model. Borgquist et al.36 identified a superficial hypoperfusion zone within 5 mm of the wound edge and a deeper hyperperfusion zone that varies in size inversely with suction pressure. Recent studies utilizing alternative perfusion measuring techniques (thermodiffusion or invasive laser Doppler) question the notion of increased local perfusion37,38.
Soft-Tissue Wounds Associated with Open Fractures
Very little high-level evidence exists with regard to the use of NPWT for the management of open fractures (Table I). To our knowledge, only one randomized, controlled trial exists.
Stannard et al.39 conducted a randomized controlled trial comparing NPWT and wet-to-dry dressings in the management of severe open fractures. Fifty-nine patients with a combined total of sixty-two open fractures were enrolled. The mean follow-up was twenty-eight months. Seven of twenty-five fractures in the control group developed deep infections as compared with two of thirty-seven fractures in the NPWT group (p = 0.012). There were two acute infections in the control group versus zero in the NPWT group. The relative risk ratio was 0.199 (95% confidence interval [CI], 0.045 to 0.874), indicating that patients who had been treated with NPWT were one-fifth as likely as control patients to develop an infection. Study groups were very similar with respect to wound size, number of debridements, and time to closure. This study was not blinded. As the only randomized controlled trial on NPWT for acute wounds with open fractures, it may be inherently biased because of a potential conflict of interest, as it was sponsored by Kinetic Concepts, Inc. (KCI, San Antonio, Texas).
Similarly, in a retrospective investigation by Blum et al.40, the use of NPWT was compared with the use of conventional dressings for soft-tissue injuries associated with open tibial fractures. All patients required delayed soft-tissue coverage. Blum et al. identified 229 open tibial fractures; 166 fractures were managed with NPWT, and sixty-three fractures were managed with conventional dressings. There were no significant differences between groups except for the fact that the more severely injured patients (i.e., those with an Injury Severity Score of >15) were treated with NPWT. The NPWT group had appreciably more Gustilo-Anderson type-III (especially type-IIIB) injuries. Eighty-seven percent of patients had a minimum one-year follow-up. The use of NPWT significantly increased over time (χ2, p < 0.001), and deep infections trended down (χ2, p = 0.08). Despite more severe injuries and a greater need for free flaps, fewer deep infections occurred with use of NPWT versus use of conventional dressings (fourteen [8.4%] of 166 versus thirteen [20.6%] of sixty-three, p = 0.011). Multivariate analysis indicated that NPWT use (odds ratio, 0.22; 95% CI, 0.09 to 0.55; p = 0.0013) was a predictor of deep infection. In fact, deep infection risk reduction was 78%, consistent with the findings of Stannard et al.39.
Several retrospective noncomparative case series have evaluated the use of NPWT in the management of open fractures. Dedmond et al.41 identified fifty open tibial fractures with an average follow-up of 19.6 to 20.5 months; eighteen patients had less than one year of follow-up. The overall infection rate requiring surgery in patients with a type-III fracture was 20%, and five patients (10%) required late amputation. Infection and nonunion outcomes with NPWT were similar to those in previous studies prior to NPWT use. The authors suggested that NPWT may reduce the need for free flaps; however, given the poor intraobserver and interobserver reliability of the Gustilo-Anderson classification, this factor is difficult to substantiate without use of a control group. In addition, Dedmond et al. used NPWT for extensive periods of time prior to wound coverage (average duration of use, 12.7 days), whereas Stannard et al.39 only used NPWT for an average of three to four days. Attempting to use NPWT for extensive periods to decrease the need for coverage may explain the lack of improvement in infection rates.
Parrett et al.42 reviewed the records of 290 consecutive patients who had open tibial fractures. Patients were divided into three four-year time periods, from 1992 to 2003. Approximately 40% of patients in each time period had Gustilo-Anderson type-III open fractures. NPWT utilization in all open fractures increased from 0% to 47% from period 1 to 3 and was used in 74% of all type-III fractures in period 3. Subsequently for all open fractures, free tissue transfer decreased from 20% to 5%, whereas closures by split-thickness skin grafts, delayed primary intention, and secondary intention increased from 28% to 53% from period 1 to 3. Time to definitive coverage (seven days), rates of deep infection requiring surgery (approximately 15% in patients with type-III fractures) and amputations (4% to 6% in patients with type-III fractures) remained consistent among periods. When used for extended periods of time (longer than one week), NPWT may move patients down the “reconstructive ladder” for coverage, thus decreasing the need for free tissue transfer.
Several small retrospective studies have investigated how the timing of wound coverage in the era of NPWT has affected infection rates in open fracture management. Bhattacharyya et al.43 identified thirty-eight patients with Gustilo-Anderson type-IIIB open tibial fractures who had at least one year of follow-up. Eleven patients (29%) developed an infection that required surgical intervention. Time to wound coverage was the only parameter associated with infection. Patients who developed an infection had wound coverage at a mean of 8.9 days as compared with 4.8 days in patients who did not develop an infection (p = 0.029). Age, sex, tobacco use, diabetes, Injury Severity Score, fixation type, number of debridements, and coverage technique were not statistically different between those infected and those not infected. The authors concluded that early wound coverage results in lower infection rates and that coverage in less than seven days resulted in an infection rate of 12.5% as compared with a 57% infection rate in wounds in which coverage was achieved in seven or more days (p < 0.008). Similarly, Hou et al.44 identified thirty-two patients with Gustilo-Anderson type-IIIB open tibial fractures. All patients with nonsalvageable injuries upon initial presentation were excluded. The mean follow-up was 2.4 years. Time to coverage averaged 10.9 days after an average of 3.3 debridements. Nine (28%) below-the-knee amputations were performed; in seven of these, wound coverage was achieved after seven days. Seven (63%) of the eleven patients in whom an infection developed required an amputation. Of the patients who had coverage by seven days, there was only a single one who developed an infection (p = 0.011). These data are consistent with the findings of Stannard et al.39. NPWT should not delay wound coverage following severe trauma.
Two retrospective studies exist that report low infection rates with temporizing NPWT and delayed wound coverage. Fleischmann et al.1 identified fifteen patients with severe open fractures in which NPWT was used prior to definitive coverage at a mean of 7.25 days. Only one (6.7%) infection occurred. Steiert et al.45 identified forty-two patients who underwent delayed soft-tissue reconstruction (i.e., a delay of more than three days). After debridement and fixation, the average time to plastic-surgery referral was 18.5 days after the injury. Thirty-three lower and ten upper-extremity injuries were included, and 67% of patients had sustained polytrauma. Although coverage occurred after a mean of twenty-eight days, only one infection (a rate of 2%) was reported. The authors concluded that, even though early flap coverage is preferred, this strategy is often not possible in the patients with polytrauma.
One clinical study compared NPWT to antibacterial modalities. Warner et al.46 retrospectively identified twenty-four patients with blast injuries resulting from improvised explosive devices who were treated at a combat support hospital and whose wounds were temporized with either an antibiotic bead pouch or with NPWT. Most wounds were traumatic amputations. The antibiotic bead pouch group required a combined total of twenty-six operations and no reoperations after delayed primary closure, whereas the NPWT group required a combined total of forty-eight operations and four reoperations after closure with NPWT (all wound specimens tested positive on culture). Closure averaged eight days with the antibiotic bead pouch versus twelve days with NPWT. A cost analysis was performed for each group, with the analyses indicating that NPWT treatment cost approximately $12,000 more ($17,966.34 for antibiotic bead pouch treatment versus $29,531.52 for NPWT).
Two recent studies evaluated a combination NPWT-antibiotic bead pouch technique in open fracture animal models. In twenty swine with simulated open femoral fractures, Large et al.47 evaluated NPWT (with vancomycin and tobramycin-impregnated polymethylmethacrylate beads) in comparison with antibiotic bead pouch treatment in the contralateral, control limb. Antibiotic elution was similar between antibiotic bead pouch treatment and NPWT only if fascial closure was possible. While periosteal biopsies trended toward lower antibiotic levels with NPWT, antibiotic concentrations in the wound bed were maintained for seventy-two hours (the standard NPWT dressing-change interval) above the necessary minimum inhibitory concentrations for most common bacteria, with the exception of Enterococcus faecalis, for all groups. Stinner et al.48 created a contaminated open tibial fracture model in twenty-one goats, the wounds in which were inoculated with Staphylococcus aureus. After debridement and irrigation, specimens were randomized to vancomycin antibiotic bead pouch or a combination NPWT-vancomycin bead therapy. The antibiotic bead pouch group had significantly fewer bacteria as compared with those in the augmented NPWT group (11% ± 2% versus 67% ± 11% of baseline, respectively; p = 0.01). When analyzing the spatial distribution of bacteria, the antibiotic bead pouch group had fewer bacteria, primarily located at the wound edge, and the augmented NPWT group had bacteria throughout the wound and adjacent to the beads. The authors concluded that NPWT diminished the effectiveness of the antibiotic bead pouch. In another basic science study, Stinner et al.49 identified a reduction in bacterial counts in a goat open fracture model with the use of silver foam as compared with the use of standard foam. Silver was especially effective against S. aureus but also against Pseudomonas aeruginosa to a lesser degree. More clinical studies are necessary to investigate the effectiveness of NPWT in combination with local antimicrobial systems to maximize the beneficial effects of both techniques.
Closed Incisional Management
Some studies have extrapolated NPWT to closed incisions (Fig. 1 and Table II), and commercially available incisional NPWT kits are on the market. NPWT has been found to promote the fibroblast proliferation that is responsible for scar formation21. Timmers et al.35 reported increased skin perfusion under an NPWT dressing; however, more recent literature37,38 questions this mechanism. Tissue edema and the presence of a seroma or hematoma can have adverse effects on wound-healing with primary closure. Investigators in one study50 created a high-risk wound in a porcine model to evaluate NPWT for difficult primary closures. After seventy-two hours, incisional NPWT resulted in improved wound tensile strength and increased energy to wound failure as compared with controls. Cross-sectional analysis indicated that hematoma formation was 15% smaller with NPWT. Another study51 used finite element analyses, which indicated that incisional NPWT counteracts tensile lateral stresses around the incision by approximately 50% and returns the stress distribution to that of intact tissue. Kilpadi and Cunningham52 created a high-risk primary closure model in eight swine. NPWT was compared with conventional dressings. Stable isotope-labeled nanospheres were injected into the wound dead space. After four days of therapy, the NPWT group had 63% less seroma and/or hematoma volume without any fluid collection in the NPWT reservoir. NPWT resulted in approximately 50% more nanospheres in regional lymph nodes and more nanospheres in the lungs, liver, and spleen, but not the kidney. The authors concluded that incisional NPWT significantly decreased hematoma and/or seroma volumes by increasing lymphatic clearance and not drainage into the reservoir.
In a large, multicenter study, Stannard et al.53 investigated incisional NPWT for high-risk lower-extremity fractures (calcaneal, tibial plateau, and pilon fractures). Two hundred forty-nine patients with a combined total of 263 fractures were randomized, resulting in 122 fractures in the standard postoperative dressing (control) group and 141 fractures in the NPWT group. There were no significant differences between groups. There was no difference in hospital length of stay after surgery or duration of drainage. NPWT resulted in fewer overall infections (fourteen [10%] in the NPWT group versus twenty-three [19%] in the control group, p = 0.049). NPWT also resulted in fewer wound dehiscences (twelve [8.5%] in the NPWT group versus twenty [16.5%] in the control group, p = 0.044). This study was funded by KCI and thus may be inherently biased.
Two smaller randomized controlled trials indicated equivalent outcomes with NPWT as compared with conventional dressings. Masden et al.54 enrolled ninety-three patients with multiple comorbidities and high-risk closures of lower-extremity amputations and abdominal wounds. Forty-three patients were in the control group, and fifty patients were in the NPWT group. There were no significant differences between groups. There was no significant difference in infections (13.5% versus 6.8%, p = 0.46) and wound dehiscence (29.7% versus 36.4%, p = 0.54) between control and NPWT groups, respectively. This study used silver-impregnated foam with a nonstick silicone interface, which is different from most published literature. Given the infection numbers provided and assuming 80% power, a sample size of 400 subjects would be needed in each group to determine if a difference truly existed. Howell et al.55 investigated NPWT versus conventional dressings in obese patients who underwent a combined total of sixty total knee arthroplasties. Twenty-four knees were in the NPWT group, and thirty-six were in the control group. There were no significant differences in days to a dry wound (4.3 days in the NPWT group versus 4.1 days in the gauze group) or infections (one infection in the NPWT group versus one infection in the gauze group). This study was aborted prematurely when fifteen knees (63%) with NPWT developed skin blistering. This study also was underpowered, as the initial power analysis required 140 total patients for 80% power to identify a 50% reduction in patients with wound drainage.
Several retrospective case series exist with regard to incisional NPWT56. Most recently, Reddix et al.57 identified 235 patients who underwent acetabular surgery during a ten-year period in which incisional NPWT was utilized. Three (1.27%) deep infections and one (0.43%) wound dehiscence occurred. When compared with their institution’s historical controls in sixty-six similar patients from the five immediately preceding years, both infection (four, 6.06%) and wound dehiscence (two, 3.03%) were significantly reduced with NPWT (p = 0.0414).
Stannard et al. hypothesized that incisional NPWT may be beneficial with established hematomas or seromas58. Pachowsky et al. conducted a randomized controlled trial in which conventional dressings were compared with NPWT after total hip arthroplasty in nineteen consecutive patients59. There was no significant difference in surgical drain output. Five of ten patients with conventional dressings had wound secretions after five days, compared with only one of nine NPWT patients. Ultrasound on postoperative day ten indicated significantly smaller (p = 0.021) mean seroma volumes and standard deviation with NPWT (1.97 ± 3.21 mL) compared with controls (5.08 ± 5.11 mL). There were no secondary surgeries for drainage or infection. Although this is significant, with no follow-up data it is difficult to assess whether this small change in seroma size is clinically important. Hansen et al.60 retrospectively reviewed the records of 109 patients who had persistent drainage after treatment with incisional NPWT after hip arthroplasty. These patients were followed until failure of NPWT or for a minimum of one year. There was no comparative group. In eighty-three patients (76%), drainage resolved without surgery. Twenty-six patients (24%) required surgery for superficial or deep debridement or implant removal.
Management of Infected Wounds
There is a paucity of literature published on the role of NPWT in treating active osteomyelitis and/or implant infections.
Tan et al.61 retrospectively identified sixty-eight patients who had been treated for Cierny-Mader types II to IV osteomyelitis. A treatment algorithm regarding implant preservation versus conversion to alternative fixation was utilized on the basis of stability of implants, fracture union, and severity of infection. After debridements, thirty-five patients with open wounds were treated with NPWT and a similar group of thirty-three patients were treated with conventional dressings. Only one infection recurrence occurred in the NPWT group compared with seven in the conventional dressing group (p < 0.05). With use of NPWT and standard debridement, polyethylene exchange, and antibiotics, Kelm et al.62 successfully managed twenty-six of twenty-eight patients with early infection after hip arthroplasty (mean follow-up, thirty-six-weeks).
NPWT manufacturers have recently marketed a new therapy mode: negative pressure wound therapy with instillation (NPWTi). This method utilizes intermittent suction and infusion cycles during which solutions such as saline solution, antiseptics, or antibiotics can be introduced into the wound. Timmers et al.7 retrospectively identified thirty patients who received NPWTi with polyhexanide antiseptic for the treatment of osteomyelitis and compared them with ninety-four matched patients with osteomyelitis who were managed with local antibiotic depots. There were significantly (p < 0.0001) fewer infection recurrences in the instillation therapy group (three [10%]) than there were in the historical control group (fifty-five [58.5%]). Also, the instillation therapy group had an overall shorter length of stay and fewer surgical procedures (p < 0.0001). Lehner et al.63 conducted a prospective, single-arm, observational study to investigate the effectiveness of NPWTi in thirty-two patients with periprosthetic joint infection. The mean duration of therapy was 16.3 days, with a mean of 3.5 dressing changes. Twenty-seven patients (84%) retained the implant, and the infection was eradicated in twenty-four patients (75%). While this study reported the successful treatment of periprosthetic joint infection with NPWTi, it had several limitations, including no control group, very minimal follow-up (mean, twenty-five weeks), and no clinical data on fracture union.
Management of Skin Grafts and Flaps
An abundance of literature consistently indicates improved outcomes with NPWT for skin grafts and/or skin substitutes. Traditionally, bolster dressings are applied to secure skin grafts to wound beds; however, irregularly contoured wounds, wound exudates and hematomas, poor vascularity, and shear stresses often contribute to bolster failure. Several randomized controlled trials exist on this topic64-67.
Azzopardi et al.68, in their evidence-based literature review from the past decade, made three recommendations. First, NPWT application to split-thickness skin grafts may promote blood flow at the graft bed and surrounding wound while stimulating angiogenesis and basement membrane integrity (quality of evidence, grade C). Second, NPWT may reduce graft loss due to shear or lift off and reduce exudate and hematoma formation as compared with the results obtained with traditional bolster dressings (quality of evidence, grade B). Lastly, NPWT may result in improved survival of split-thickness skin grafts as compared with the survival rate achieved with use of traditional bolsters (no grade of recommendation assigned; only two studies had Level-II to Level-V evidence). The authors recommended NPWT as the standard of care for split-thickness skin grafts applied to complex, large, exuding wounds with irregular contours.
A handful of basic science, retrospective, and small prospective studies exist on the use of NPWT in local and free-flap management. Most studies69-71 indicated favorable results with NPWT, presumably through improved perfusion and reduced edema and/or clearance of venous congestion. However, to our knowledge, no high-level evidence exists to support the use of NPWT with local or free flaps.
The U.S. Food and Drug Administration (FDA) has published an update regarding the number of severe injuries and deaths that have been attributed to NPWT72. Since 2007, the FDA reported twelve deaths and 174 injuries. In most cases, adverse events occurred away from hospitals, in homes or in long-term care facilities. Most serious adverse events were attributed to bleeding as a result of NPWT being placed in proximity to large blood vessels or grafts, near the sternum or groin, or used in patients receiving anticoagulation. Other common adverse events were related to ongoing infection or retention of foam in wounds that required additional surgeries. The FDA-recommended contraindications to NPWT, and risk factors associated with the use of NPWT are summarized in Tables III and IV, respectively.
Levels of Evidence and Recommendations
The literature surrounding NPWT is vast, and NPWT has been widely adopted as one of the primary modalities for treating traumatic wounds and aiding in soft-tissue reconstruction. Despite the abundance of peer-reviewed publications, very few high-level-evidence studies exist. Most high-level studies had some affiliation with a NPWT vendor, thus introducing the possibility of bias. In efforts to provide clinicians with high-quality recommendations, the International Expert Panel on Negative Pressure Wound Therapy (NPWT-EP) performed an evidence-based review of the literature73. The authors provided levels of recommendations in which NPWT must (quality of evidence, grade A), should (quality of evidence, grade B), or may (quality of evidence, grade C) be utilized.
To our knowledge, only three retrospective clinical studies40,44,74 have been published since the NPWT-EP review on open fracture management in 2011; thus, we believe that these recommendations are still reasonable. NPWT should be used temporarily after debridement until definitive coverage when primary closure is not possible (quality of evidence, grade B). NPWT may be used to decrease the complexity of technique required for definitive closure or coverage (quality of evidence, grade C). With regard to all soft-tissue scenarios without fractures, only grade-C recommendations were given. While we do not recommend the use of delayed wound coverage for open fractures with NPWT (quality of evidence, grade C)—to our knowledge, only low-level evidence, which is sometimes conflicting, exists.
In agreement with the recommendations of Azzopardi et al.68, the highest recommendations of the NPWT-EP were related to split-thickness skin-graft procedures and indicated that NPWT must be considered to improve split-thickness skin-graft success (quality of evidence, grade A), should be considered in high-risk patients or wounds (quality of evidence, grade B), and should be left undisturbed for three to seven days on continuous pressure (quality of evidence, grade B).
The NPWT-EP did not review the role of NPWT with regard to closed incisions. A Cochrane review article4 indicated that there is no significant evidence for the use of NPWT on surgical wounds that can heal by primary intention. The Cochrane review article also included studies in which split-thickness skin grafts were used, but infection and wound dehiscence was the primary end point in only one of the three prospective randomized controlled trials that were cited. We believe that limited Level-I evidence53 in combination with Level-III and Level-IV evidence56,57 indicates that NPWT should be used over closed, high-risk incisions in trauma patients (quality of evidence, grade B) and may be used over other closed incisions in efforts to prevent infection and wound dehiscence (quality of evidence, grade C).
Because the number of Level-III and Level-IV studies is limited, there are insufficient data on which to base NPWT recommendations on the following scenarios: in combination with an antibiotic bead pouch for the treatment of open fractures, for the treatment of wounds that are associated with osteomyelitis, or for the treatment of wounds resulting from periprosthetic infection.
Source of Funding: No external funding sources were used in the execution of this research.
Investigation performed at the University of Missouri, Columbia, Missouri
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2014 by The Journal of Bone and Joint Surgery, Incorporated