Table 1. Negative Pressure Wound Therapy (NPWT) Peer Reviewer Comments

Disposition of Comments, Negative Pressure Wound Therapy Devices

Comments received from draft review on negative pressure wound therapy devices.

Table 1. Negative Pressure Wound Therapy Devices (NPWT) Peer Reviewer Comments

NameCommentsECRI Response
Peer Reviewer 1

This is a thoroughly researched, well written, and interesting report. I have two major and one minor concern.

  1. The report is based on answering four key questions, which in general are based on exploring differences between the NPWT devices that are currently available. Due to the proliferation of these devices, these seemingly are very important questions. It would seem to me that first it would be important to demonstrate an advantage of using this class of device over standard care. Why would we care if one device is better than another, unless the class of devices offers an advantage over standard therapy? Certainly the authors have access to all of the important studies and could easily address this key issue. This issue was recently addressed by three reviews Gregor S et al Negative Pressure Wound Therapy Arch Surgery 2008 143: 189-196; Vikatmaa et al Negative Pressure Wound Therapy: a Systemic Review on Effectiveness and Safety Er J Vasc Endovasc Surg 2008 36, 438-448; and Ubbink DT et al Topical negative pressure for treating chronic wounds, Cochrane Database of Systemic Reviews 2008 Issue 3. Art. No CD001898. In general these reviews acknowledge that NPWT might improve wound healing but overall there was insufficient evidence to clearly prove efficacy over standard therapies. This certainly does not mean that NPWT does not work but it does indicate that there is not enough evidence to reach a firm conclusion about efficacy or effectiveness for this class of therapy. Without beginning able to firmly reach this conclusion, it seems like it would be impossible or premature to demonstrate that one of many different forms of NPWT is superior to another.
  2. As noted above, the overall theme of the key questions is to explore or demonstrate clinical advantages between the various devices that are currently marketed to provide NPWT. This a very difficult task because the market has been dominated by one device for many years. In fact, in comparison the additional devices that are currently marketed to provide NPWT have been available for a very short period of time. It would be very difficult for adequate studies of these newer devices to have been conducted and published. At this time, as you have shown, it is very unlikely that many of these newer devices were even available long enough to conduct comparative efficacy or effectiveness studies. As a result, the logistics of answering your key questions were not in your favor.

Finally, a minor comment about the ultimate outcome of a wound care trial. The FDA, perhaps wrongly, has been very resistant to accepting any outcome other than a healed wound as a proper outcome for a chronic wound healing study (Guidance for industry: Chronic cutaneous ulcer and burn wounds-developing products for treatment. Food and Drug Administration; 2006 Jun). The majority of your discussion evaluates outcomes that were not from studies that used this final study outcome. It is very important that you place your results in this perspective so that health care providers interested in wound care realize that your key questions may not be based on studies conducted with the goal of a "healed" outcome. This perspective is important especially if comparisons are to be made by a reader between NPWT and other wound care modalities.

In support of the upcoming coding review of NPWT devices, the Center for Medicare & Medicaid Services (CMS) requested that ECRI Institute perform an independent technology assessment of the various NPWT devices. The focus of this report was the comparative effectiveness of NPWT devices and the significant therapeutic distinction they may offer. We were not asked to address the effectiveness of this modality over standard care.

Reviews by Gregor S et al Negative Pressure Wound Therapy Arch Surgery 2008 143: 189-196; Vikatmaa et al Negative Pressure Wound Therapy: a Systemic Review on Effectiveness and Safety Er J Vasc Endovasc Surg 2008 36, 438-448; and Ubbink DT et al. Topical negative pressure for treating chronic wounds, Cochrane Database of Systemic Reviews 2008 Issue 3. Art. No CD001898 are included in the review and may be referenced in the report by searching for the following reference numbers: (1-3).

Key questions were formulated for the report to test the hypothesis that one NPWT device or its components may provide significant therapeutic distinction compared to another NPWT device or its components. These questions were structured using the "PICO" framework: patients, intervention of interest, comparator, and outcomes.

For assistance in clarifying relevant outcomes, we looked for guidance from the U.S. Food and Drug Administration (FDA). According to the FDA, improved wound healing and improved wound care are important clinical outcomes with the use of a wound-treatment device. Specifically, the most important outcomes to consider under the category of improved wound healing are percent of patients with complete wound closure and time to complete healing (partial healing for facilitation of surgical wound closure). We therefore incorporated "time to complete wound closure" and "percent of wounds healed" as important patient-oriented outcomes. In addition, for cases that are not expected to result in complete wound closure but rather to advance the wound to a stage where healing is possible we incorporated "time to 50% reduction of wound initial volume", "percent change in wound volume", and "improved wound condition" as important outcomes of interest.

The analytic framework then drives the development of inclusion and exclusion criteria specific for each key question. Criteria may vary for each question, i.e., the inclusion of single arm studies for key question 3 only.

Next steps include an extensive search of the literature and in this case, a review of an additional 1,400 submissions. Items were evaluated for possible inclusion in the report based on their relevance to the key questions and the pre-determined inclusion exclusion criteria.

An obvious key component of this screening process is to affirm appropriate outcome measures are included in each study. This determination is addressed in the screening phase of our methods and does not drive this process.

Lastly, a very small number of studies were excluded in this report for "irrelevant outcomes". Of the 497 articles that remained at the article level only nine (2%) of the articles were excluded for this reason. Specifically these studies were focused on cellular and biochemistry measurements which were not suitable outcomes for this report.

Peer Reviewer 2
  1. Strength—One particular strength of this Evidence Report is the inclusion of studies with both Intermediate and Patient-Oriented Outcomes (Pain, Bleeding, Mortality, Infection). Much of the chronic wound evidence base includes studies with only intermediate outcomes, following subjects for inadequate periods of time. A second strength is the inclusion of new information on unstageable wounds, and pressure ulcer development from the inside out. NQF is proposing a new two-stage classification of pressure ulcers—partial thickness and full-thickness, in addition to the unstageable category. This might make it easier for those assessing wounds, and allow for more accurate comparisons and outcomes analysis.
  2. It is not clear what "wound volume" and "improved wound condition" refer to in the narrative. This is not typical wound care terminology. More familiar are the terms area or dimensions of the wound, or length, width, and depth. It is not clear what wound condition refers to—exudates, slough, necrotic tissue?
  3. Infection/bacterial load may not be a very helpful measure. All chronic wounds have some degree of bacterial colonization. It is considered a wound infection if the colony count exceeds 105 micro-organisms per gram of tissue or swab. Even if clinical infection is absent, however, wound healing may be impaired when the bacterial colonization counts exceed 105 organisms per gram of tissue.
  4. Are saline gauze and foam dressing sets the only two relevant dressings for the comparative analysis of components of NPWT systems? I have come across use of non-adherent porous wound dressings and antimicrobial dressings in conjunction with NPWT.
  5. Diabetic foot ulcers are a major cause of amputations in diabetic patients—this should be emphasized.
  6. Infection of chronic wounds occurs when host defenses are overwhelmed, and demonstrate erythema, edema, warmth, pain and purulent exudates. Fever, chills and leukocytosis suggest the development of bacteremia or septicemia. It is common for chronic wound infections to be polymicrobial, often with Staph aureus and anaerobes. These wounds should be cultured and micro-organism sensitivities determined. Systemic antibiotics are recommended for clinically infected ulcers. Patients may be treated with antimicrobial dressings or topical antibiotics to reduce wound bioburden, but their use remains controversial. Topical antiseptics are chemical agents that are broadly toxic to microbes. Antibiotics are narrow-spectrum antimicrobial agents. Microbial resistance may develop with the use of topical antibiotics, while antiseptics may cause damage to healthy tissue.
  7. In the discussion of venous ulcers (and elsewhere), mechanical debridement is presented as one option or treatment choice. In almost all cases, mechanical debridement (wet-to-dry, whirlpool) is contraindicated—it is a nonselective form of debridement and very painful to the patient. Instead, sharp (use of lasers, forceps, scissors), enzymatic (exogenous) or autolytic (endogenous enzymes) debridement is preferred.
  8. Dressings should be selected based on the principal of moisture management—absorptive if the wound is draining exudates, and hydrating if the wound is dry or has minimal drainage. The optimal dressing will change over time as wound goes through healing stages. Unfortunately, this is seldom acknowledged in RCTs and CPGs. Also, strong evidence supports the use of graduated compression during the healing process for venous leg ulcers, and continuous use of compression stocking to prevent recurrence once the ulcer has healed. Evidence also supports the use of multilayer compression bandages rather than single layer in promoting healing of chronic venous leg ulcers.
  9. Page 48. I am not sure if the "cause" of many wounds can be eliminated or minimized. Instead, recommend the use of the term "contributing factors" to wound occurrence—which would include reducing friction or shear forces, improving nutritional status, and so on.
  10. Later in evidence report, the discussion of debridement is again inadequate. Surgical debridement is performed less frequently for chronic wounds compared to other available approaches. It may be done initially to allow staging of the wound, and then be followed up with techniques requiring fewer resources and less expertise. Nurses can be trained to perform sharp debridement (using forceps or scissors), and enzymatic and autolytic debridement are more common in long-term care settings. Chemical debridement is a term not often used—enzymatic and autolytic debridement should be specified.
  11. The evidence base for selection of dressings is inadequate. Studies have not been powered to find significant differences between different versions of the same type of dressing (hydrocolloid versus hydrocolloid). Dressings remain the same over the course of the study in spite of changing wound characteristics. Moist gauze dressings are not the same as wet-to-dry dressings. The use of moist gauze dressings, as pointed out, is time intensive, and may lead to skin maceration. Gauze is not as effective as "modern" dressings, but studies have been underpowered to find significant differences, intermediate outcomes have been used, limited comparisons made, and sample sizes have been small and highly selected (small wounds). The most important point is that dressings need to be selected to match the characteristics of the wound at that point in time.
  12. A 3rd category of dressings is antimicrobial dressings (in addition to gauze and moisture management "modern"). Some antimicrobial dressings contain iodine or silver. These are different from use of topical antimicrobials (neomycin) and antiseptics (Dakin's solution).

Very few wound measurements were reported in length, width and depth. Study authors equally reported wound size as a two-dimensional measurement of area or a three-dimensional measurement of volume. We have updated the Analytic Framework and table headings to reflect that both measurements may be reported.

Improved wound condition as presented in the Analytic Framework is defined as a reduction in wound exudate and infectious materials; the promotion of granulation tissue formation and perfusion; an improvement in graft appearance; a reduction in odor; and a greater rate of epithelialization. The Analytic Framework has been updated to read as such.

Remaining comments all pertain to discussion in our Background Section. This section of the report has been updated as per suggested.

Peer Reviewer 3

The WNDP 1109 NPWT Draft Report Document was read in detail. In my current full time faculty position at UCSD Medical Center, I am and have been extensively involved with the use of NPWT for patients with all wound etiologies. My major concern has been the misuse and extensive overuse (prolonged periods of time) with NPWT without evidence to support the prescribed indication.

I believe that the above document is a thorough and comprehensive review of the current data and information on NPWT and fully agree that evidence is lacking to support NPWT use over other standards of care to attain complete closure. I have not been able to find any evidence supporting one NPWT over the other.

There are errors in the document, in the description of venous ulcer etiologies and treatments. These wounds are primarily treated by compression, rarely develop severe infections and respond well in the majority of cases to chronic wounds. Also, the information on NPWT and its effect on matrix formation are not correct from a cellular biology or physiological standpoint. NPWT does promote granulation in a wound but does not do so my denaturing or destroying existing matrix. Also, NPWT removes exudate and does not necessarily promote a wound environment.

Not all dressing categories are listed in the dressing section. Collagen/combination and antimicrobial dressings are not discussed. Allografts and xenografts have been successfully used to treat chronic wounds and are also excluded.

Overall, this is an outstanding review that provides extensive evidence on the lack of well designed randomized controlled trials for NPWT. The review also supports that lack of evidence for use of NPWT to attain complete wound closure. NPWT tends to be used beyond its potential at a significant cost. Data is needed to support length and value of use.

Comments are all pertaining to discussion in our Background Section. This section of the report has been updated as per suggested.
Peer Reviewer 4I think that this is a very through and thoughtful review of the published data on negative pressure wound therapy. My only comment is that at least one major wound journal, Wounds: A Compendium of Clinical Research and Practice is not indexed through pub med and medline. They have articles published on negative pressure wound therapy. You may have reviewed the journal and I missed it in my review. However, I wanted to be sure that this journal was not overlooked. Thank you very much for giving me the opportunity to review the document.The journal Wounds: A Compendium of Clinical Research and Practice is indexed in the Embase and CINAHL databases which were included in our searches for this report.

Note: Peer reviewer comments are not presented in alphabetical order.

Return to Document

Current as of September 2009
Internet Citation: Table 1. Negative Pressure Wound Therapy (NPWT) Peer Reviewer Comments: Disposition of Comments, Negative Pressure Wound Therapy Devices. September 2009. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/findings/ta/negative-pressure-wound-therapy/table1.html