Negative Pressure Wound Therapy Technologies For Chronic Wound Care in the Home Setting

Disposition of Comments

Comments received from draft review of technology assessment on negative pressure wound therapy technologies for chronic wound care.

Table 2: Public Review Comments

Reviewer Name1 Reviewer Affiliation2 Section3 Reviewer Comments Author Response4
Anonymous Anonymous General The reviewers have undertaken a large body of work to identify citations evaluating negative pressure wound therapy (NPWT). However, they have been limited in their scope by the paucity of evidence that they were able to find. The focus of the review was for use of the NPWT within the home setting. This was not clearly defined and was broadly non-acute settings. The lack of evidence for this setting could have meant that the scope of the review was widened to review the evidence for NPTW. As cost was not examined as an outcome within the review the efficacy of NPTW on chronic wounds was likely to be the same in whatever setting. We developed the scope of the review in discussion with AHRQ and the nominating organization. The goal was to focus on the home setting. The setting was poorly defined in many studies and we clarified in the results our interpretation of setting, and noted this as a limitation of our review in the discussion. Cost was not an outcome of interest.
Anonymous Anonymous General The authors sought to answer four key questions related to identifying the range of NPWT available, whether NPWT improved outcomes for patients (>65) with chronic wounds in the home setting and what were the characteristics that predict outcomes for chronic wounds with NPTW. Thank you for reviewing our report.
Anonymous Anonymous General The questions would appear reasonable but a more general question regarding whether NPWT was effective should have been included as the premise seemed to assume that as NPTW was an established therapy there would be some form of effectiveness. An additional question of whether the costs outweighed the benefits compared to other therapies would also have been useful knowledge. Thank you for your comment. Cost analysis was not part of the project, thus no comment was made on cost effectiveness versus other therapies.
Anonymous Anonymous Introduction The introduction was disappointing in its scope and thoroughness. The review of chronic wounds did not extend to that of the types of wounds for which NPWT was suitable and time was spent detailing wounds for which it is not indicated such as venous ulceration. Thank you for your comment. There were multiple types of wounds which were excluded due to the scope of the project.
Anonymous Anonymous Introduction Additional references are also required within the introduction as unsupported statements are made. For example on line 33 they state that NPWT facilitates healing without any supporting evidence. We revised the statement in the fourth paragraph on page 1 to say, "NPWT refers to the application of negative pressure across a wound."
Anonymous Anonymous Introduction Little time is spent detailing the potential benefits of NPWT beyond a statement on line 47 "with the potential benefits..." Thank you for your comment. We do not feel that additional text is needed in introduction about the potential benefits.
Anonymous Anonymous Introduction The introduction also seems to preempt the findings with the paragraph starting on line 38 on page 2 which details the need for appropriate use. We do not feel that this statement preempts the findings of the review. Details about safety and efficacy are needed to help determine appropriate use.
Anonymous Anonymous Introduction The final paragraph of the introduction on page 2 line 43 seems a little vague and could be more explicit. We feel that the final paragraph in the introduction is sufficient and immediately leads into the specific questions for the review outline in Scope and Key Questions.
Anonymous Anonymous Methods The inclusion and exclusion criteria are vague and could be more explicitly stated and justified beyond a table of PICO. It would seem that comparative studies evaluating NPWT were sought and it included both RCTs and non-randomised studies. The basis and weight given to these were not stated and in parts of the review the same weight seemed to be implicitly given to both types of studies. We have expanded the text describing our inclusion and exclusion criteria in the Study Selection section of the Methods chapter. Methodological strengths and weaknesses for all studies are described and we considered all study designs in grading the strength of evidence.
Anonymous Anonymous Methods The search strategy was comprehensive if not extensive with only the key databases searched and no detail given of whether the grey literature was searched. We stated in the Methods that we had searched clinicaltrials.gov and conducting handsearching.
Anonymous Anonymous Methods The included outcomes did not include cost which was a surprise. We developed the protocol in discussion with AHRQ and the nominating organization. Cost was not an outcome of interest.
Anonymous Anonymous Methods The data extraction and synthesis were well defined and outlined. The proposed meta-analysis was planned but no detail of the proposed method was outlined. We added further detail about the planned meta-analysis in the Data Synthesis section of the Methods chapter.
Anonymous Anonymous Results The search identified 5332 citations and the authors should be commend for sifting through these and outlining their findings so comprehensively. Thank you for reviewing our report.
Anonymous Anonymous Results The key messages could have been more explicit. However, the authors were hampered by only finding one RCT and 6 non-randomised studies. A clear summary paragraph detailing the these in relation to their key questions would have aided the reader. We provided a summary of the key findings in the Key Findings and the Strength of Evidence section of the Discussion chapter.
Anonymous Anonymous Results The descriptions of the 7 included studies was comprehensive but a clear statement of the limitations of the studies should have been made. The authors seemed to concentrate overly on describing the findings of the studies and spent only a few lines detailing to the reader the potential bias and flaws of the included studies. The limitations are described under the section entitled, "Limitations of the Evidence Base."
Anonymous Anonymous Discussion / Conclusions The implications of the findings are neatly summarised in the opening paragraph of the discussion. However, the implications of the lack of evidence were not reviewed in sufficient detail. Some reflection on why only 7 out of 5332 were suitable for inclusion would have strengthened the review. The role of industry in promoting and generating the evidence for NPTW would also have aided the reader. We added several paragraphs to the Limitations of the Review section of the Discussion chapter about the scope of our review and how it may have affected the results of the review.
Anonymous Anonymous Discussion / Conclusions The gaps in the research are outlined and would be easily translated into new research. Thank you for reviewing our report.
Anonymous Anonymous Discussion / Conclusions The overall structure of the report was fine. Although the main points are outlined they are not reinforced to the reader. The conclusion was rather brief. We feel that the key points are sufficiently outlined in the discussion section.
Steven Brotman AdvaMed Introduction The introduction states that, "The exact mechanism by which these devices may promote wound healing is not known." AdvaMed is concerned that the inclusion of this statement, as written, may leave the non-scientific reader with the impression that there is limited understanding of the use of NPWT in treating wounds and that the therapy has not been investigated adequately. This, however, is not the case, as is demonstrated by the number of articles referenced throughout the technology assessment document. Since 1993, there have been over 900 Negative Pressure Wound Therapy-related peer-reviewed articles published, including 41 articles on randomized clinical trials (RCTs). Furthermore, there has been over 900 NPWT abstracts, 650 NPWT-related articles and over 70 NPWT book references published. Additionally, there are numerous articles specifically discussing the mechanism of action of NPWT. , , , , , AdvaMed encourages AHRQ to clarify or to revisit inclusion of this paragraph. We have revised this paragraph at the bottom of page 1 to cover multiple mechanisms of action of NPWT. Thank you for the list of references, some of which we added. However, the Lanceretto article refers to a specific subset of NPWT, which we don't think is generalizable.
Steven Brotman AdvaMed Methods AdvaMed is concerned about the reference to the use of NPWT to treat acute surgical wounds in only acute care settings. NPWT is frequently used to treat surgical wounds, especially dehisced wounds, in the home.  A large number of surgical wounds will become dehisced wounds and will take on chronic-wound characteristics in a home setting — resulting in the need to address their treatment through the use of NPWT. For this reason, AdvaMed believes that the authors should include references and evidence that speak to the use of NPWT in the treatment of dehisced surgical wounds in the home setting. This population of wound patients accounts for a very large percentage of those receiving NPWT in the home. Failure to include this subset of wounds does not enable a full evaluation of the usefulness or impact of these technologies in the home and misrepresents the burden of this type of wound being treated in the home setting and utilizing NPWT. Unless a separate review is planned to evaluate the impact of NPWT on dehisced surgical wounds, AdvaMed recommends including treatment for these types of wounds in the technology assessment. We have excluded the setting from the statement in Introduction in discussing surgical wounds. Chronic wounds were the specific focus of this review, as outlined in our key questions.
Steven Brotman AdvaMed Methods The methods section of the Technology Assessment sets out the specific inclusion and exclusion criteria for the selection of studies that were reviewed. Regarding wound healing outcomes, studies were only included if they dealt with one of the following parameters:

  • Complete Wound Healing;
  • Time to Complete Wound Healing;
  • Time to Surgical Readiness of the Wound Bed;
  • Mortality; or
  • Wound Healing Rate for Healed Wounds

By excluding studies that do not specifically address the endpoint of complete wound healing, the authors of the Draft Assessment are missing a substantial volume of evidence regarding the usefulness of NPWT's role as an adjuvant therapy in wound care at various stages during the lifecycle of a wound.

During the development of the review protocol, selection of outcomes, including whether to include wound healing rate, was discussed extensively amongst our review team and with the organization that nominated the topic. We have added rationale for excluding wound healing rate as an outcome to the Study Selection section of the Methods chapter. We have also included in the Limitations of the Review Process section of the Discussion chapter a paragraph regarding excluding studies that evaluated only wound healing rate.
Steven Brotman AdvaMed Methods Using the incidence of complete wound healing or the time to complete wound healing as the main outcome measure of a study is not always a feasible or desirable goal due to patient characteristics, comorbidity and the type of ulcer.  In addition, some technologies for the treatment of chronic wounds are not always appropriately evaluated by measuring wound closure, since wound closure may not be their intended goal. Examples of alternate endpoints include control of exudate, reduction of odor, stimulation of tissue growth to decrease wound volume and/or surface area, alleviation of pain and protection of the periwound skin. Thank you for your suggestions. The outcomes were predefined in the review protocol after discussion with AHRQ and the nominating organization.
Steven Brotman AdvaMed Methods Importantly, when combination or sequential interventions are necessary to achieve wound closure, relevant intermediate endpoints short of healing can be appropriate measures for product efficacy and effectiveness. It is well accepted that wound healing is a multi-phase process, thus endpoints other than wound closure are valid in research design. AdvaMed believes that the authors of the Draft Technology Assessment should reevaluate their data and analysis after they have included studies that are not limited solely to those that consider complete wound healing. Thank you for your suggestions. The outcomes were predefined in the review protocol after discussion with AHRQ and the nominating organization.
Steven Brotman AdvaMed Methods In addition, the following types of studies were excluded from analysis in the Draft Assessment:

  • Studies of Patients with Surgical or Traumatic Wounds;
  • Studies that did not have a Comparison Group;
  • Studies that Reported Wound Healing Rates Without Also Reporting Complete Wound Healing;
  • Studies Without Original Data (e.g., reviews, editorials and commentaries); and
  • Studies Conducted in the Hospital, Inpatient or Long-Term Care Setting

The comments above regarding studies concerning surgical wounds and those related to complete wound healing apply here as well. In addition, it appears that many studies that contained significant data regarding NPWT treatment in the home setting were excluded.

We developed the scope of the review in discussion with AHRQ and the nominating organization. We added a few paragraphs to the Study Selection section of the Methods chapter to describe the criteria outlined in Table 1 and to explain our rationale for some of criteria. Studies without original data are not included in systematic reviews, so we do not elaborate on this further. We discuss the more significant exclusions in the Limitations of the Review Process in the Discussion Chapter.
Steven Brotman AdvaMed Methods Wounds are often treated in a number of different settings (hospital, nursing facility, home health service, outpatient clinic, doctor's office, etc.) during the course of healing. In fact, the initiation of wound therapy often begins in the inpatient setting with care continuing in the patient's home following discharge. The likelihood of having a wound that is solely treated in the home setting, from inception to final healing without occasional consultations by professionals in other settings, such as outpatient/ambulatory or others is highly unlikely. We developed the scope of the review in discussion with the AHRQ and the nominating organization. We agree it can be difficult to define the setting and have noted a few places in methods, results and discussions, how we interpreted studies that did not explicitly state that care was in the home setting.
Steven Brotman AdvaMed Methods Appendix C of the Draft Technology Assessment lists over 150 excluded studies; many based on not being conducted in home or outpatient settings. Several of these studies were conducted predominately (approximately 90%) in the home setting. For example, in the randomized study in 342 diabetic foot ulcer patients by Blume, et al which was conducted across 37 centers, the proportion of home care therapy days to total therapy days was 9,471 of 10,579 (89.5%) for NPWT and 12,210 of 12,810 (95.3%) for AMWT (the controls).  In a population consisting of diabetic foot ulcers or patients receiving amputations for foot ulcers, it is inconceivable that some inpatient care would not be experienced, given the need for urgency for responses to those patients who did experience issues with suspected infection, in either control or treatment arms. To exclude patients from access to acute care if necessary would, of course, be unethical. It seems perverse to exclude such studies as not being "sufficiently" homecare based. AdvaMed believes that the authors of the Draft Technology Assessment should modify their exclusion criteria to include NPWT studies that were not conducted solely in the home care setting and provide a reassessment of the evidence, after these studies are included. We developed the scope of the review in discussion with the AHRQ and the nominating organization. We agree it can be difficult to define the setting and have noted a few places in methods, results and discussions, how we interpreted studies that did not explicitly state that care was in the home setting.
Steven Brotman AdvaMed Discussion / Conclusions The discussion section of the draft technology assessment refers to the "paucity" of well-designed and well-conducted studies of NPWT technologies. As noted in our comments to the methodology, AdvaMed is concerned that the authors may have excluded from consideration a number of studies that examined randomized comparative studies related to the use of NPWT in the home setting. It is typical for patients with certain conditions treated with NPWT in the home setting (i.e. diabetic foot ulcers) to also receive some inpatient care for their condition especially since preventing them from having access to acute care when needed would be unethical. Therefore, AdvaMed questions the author's definition of "sufficiently" homecare based and their subsequent decision to exclude many studies on those grounds. AdvaMed urges the authors to re-evaluate and to consider inclusion of those studies as they work towards finalization of the draft document. The setting was poorly defined in most studies. We noted our interpretation of setting in methods, a few places in the results, and in the discussion. We developed the scope of the review in discussion with AHRQ and the nominating organization. In the Study Selection section of the Methods chapter, we added our rationale for including patients only treated in the home setting. We clarified in the Study Selection section of the Methods chapter that we wanted all patients followed up in the home setting.
Steven Brotman AdvaMed Discussion / Conclusions The authors conclude that there is a need to develop consensus on study methods and that efforts should be made to standardize the conduct and reporting of NPWT studies. AdvaMed is concerned that this finding does not take into account the complex nature of chronic and acute wounds and the resulting need to employ a range of methodologies to treat the patients who have them — including patients who participate in studies. The lack of similarities in terms of patient symptoms, co-morbidities, and stages of healing require that all wound patients receiving NPWT and other technologies be treated in a manner specific to their condition and their care needs in order to promote the best care outcomes. These same factors also impact the rate of healing. Therefore, it is important to consider varying endpoints when evaluating these patients and the efficacy of various technologies, in this instance NPWT, in healing their wounds. Thank you for your comment. We agree that wound care is complex, and does not lend itself to simple studies given the many variables which are in play. Industry does play a vital role in creating important data and medical innovations for our patients. From a literature standpoint, we felt that there could be some common consensus made in terms of definitions, which could greatly strengthen the literature base.
Steven Brotman AdvaMed Discussion / Conclusions Wound care also poses significant challenges to the selection of comparator groups, in that most wounds require a number of different therapies or therapeutic approaches to take a wound from an early phase of healing to complete closure. If, in good clinical practice, treatment modifications are made to ensure healing, the active control arm of a study should allow for such modifications to ensure acceptable standard of care. In addition to the selection of the primary wound therapy within the control arm, other factors (i.e. offloading of pressure-related wounds, debridement and treatment of wound-related complications) should be discussed clearly in the protocol, with an approach for handling of the subject and subject's data. Data should be collected to understand the impact of these factors on overall outcome. Thank you for your comment. We agree that there are significant challenges that are inherent to wound care, which creates difficulties in studying the various methods used.
Vickie Driver AAWC General In the introduction of the report, AHRQ had stated "The findings and conclusions in this document are those of the authors who are responsible for its contents. The findings and conclusions do not necessarily represent the views of AHRQ." We anticipate that AHRQ is not in agreement with this research and its findings. The authors have produced a report that is of substandard quality compared to the reviews traditionally produced and released by the AHRQ. The draft reflects badly on the caliber and accuracy of information AHRQ will be providing to the Centers for Medicare and Medicaid Services. The contracted reviewers have taken some liberties in their exclusion criteria and certain statements. As a result, they have incorrectly "determined that the existing evidence is insufficient to draw any conclusions". We developed the scope of our review in conjunction with AHRQ and CMS and followed methods outlined in the AHRQ EPC Methods Guide. Both AHRQ and CMS reviewed the protocol and the draft report prior to posting for public comment.
Vickie Driver AAWC General The overall TA objective as stated in the report was, "To systematically review the efficacy and safety of negative pressure wound therapy (NPWT) for treatment of chronic wounds in the home setting." This objective identifies a subset of the clinical uses of NPWT in the home setting. It raises our first concern; who and why would only chronic wounds treated in the home setting be evaluated for safety and efficacy? [CMS, AHRQ or reviewers]? The scope of the project was to examine NPWT use for chronic wounds in the home setting. We acknowledge that this is limited, given that NPWT is used across the healthcare spectrum, and that these patients often change the location of their treatment based on a variety of factors. The key questions, from the nominating organization, specified use of NPWT for chronic wounds in the home setting.
Vickie Driver AAWC General The evaluation of safety and efficacy for NPWT use in the home is limited by selecting only chronic wounds. To truly meet the objective to assess safety and efficacy of the NPWT technology in the home setting, why not include all common wounds treated in the home setting? The objective as stated is too narrow. The scope of our review, and specific key questions to be addressed, were based on the request from the nominator. The specific focus of interest was chronic wounds.
Vickie Driver AAWC Methods We have a fundamental problem with the scrutiny and methodology applied to this TA. This concern extends to any technical assessment regarding wound care. If a technical review, after removing duplicate articles, identifies 5,332 unique citations but when the review criteria is applied excludes 5,173 [97%] of those studies, major alarms go off. We were very broad in the search strategy because of the variety of terms used to describe chronic wounds. Thus, the search retrieved over 5000 citations. We developed the eligibility criteria in discussion with the nominating organization. We applied the pre-defined eligibility criteria uniformly across all citations and followed standard methods for conducting a systematic review.
Vickie Driver AAWC Methods The review team eliminated 1,083 studies simply because each had less than 20 patients per study. There was no criteria or rationale in the report to justify this decision.  The research team did not pool those studies and conduct a meta-analysis to understand the value of this large number of studies and the representative large patient populations to answer questions on safety, adverse events, or healing. In fact, there was no suggestion by the reviewers to conduct a meta-analysis to further support the technology assessment data. We question, "Why not?" We clarified the inclusion/exclusion criteria in the Methods chapter in the Study Selection section and Table 1. The sample size criteria was developed in discussion with the nominator. We have added a rationale for this criterion. Also, we re-reviewed all of the studies that were excluded during full-text screening because of the sample size. All of the studies could have been excluded for other reasons (e.g., no comparison group, mix of patient populations, mix of settings, no original data). We also added to the Limitations of the Review Process a paragraph about excluding studies based on sample size. Finally, the data are not appropriate for meta-analysis. Quantitatively pooling heterogeneous studies of poor quality would provide misleading results
Vickie Driver AAWC Methods The reviewers excluded another 200 studies because they claim that NPWT was not used in the home setting, when in fact most of those discarded studies had the technology applied in the hospitals, as is customary, before being discharged to the home setting. In our clinical experience, we estimate more than 75% of all NPWT used in the home is initiated in the hospital setting. As a result of this decision by the reviewers, all that critical and usable data was ignored. Thank you for your estimate. We developed the scope of the review in discussion with AHRQ and the nominating organization. In the Study Selection section of the Methods chapter, we added our rationale for including patients only treated in the home setting. We clarified in the Study Selection section of the Methods chapter that we wanted all patients followed up in the home setting.
Vickie Driver AAWC Methods They further eliminated data in 298 studies, because they did not have a direct comparison group. While this may be appropriate for evaluating efficacy, it is not appropriate for determining safety. Uncontrolled trials documenting safety parameters should be included in a separate safety analysis. Additionally, many of these studies reported patient specific comparison through retrospective chart reviews, extracting registry historical data, or reported prior treatments to compare the impact of the treatment on the patients' wound healing. The wounds in the identified citations almost always included wounds that had not responded to standard care. Information from these 298 studies provided comparisons to prior treatments. Again, the reviewers do not seem to understand this concept. We developed the inclusion and exclusion criteria in discussion with AHRQ and the nominating organization. The intention was to limit the review to only include studies with a comparison group. We have added a statement to Limitations of Review about the potential use of other study designs for identifying adverse events.
Vickie Driver AAWC Methods In total, after eliminating thousands of available and relevant clinical studies, the team settled on 159 papers. After reviewing the full text of those remaining papers, only seven studies met their narrow criteria. This narrow list however included data for over 1,458 patients. If we assume the average number of applications for each patient is 10, at a very conservative assumption, the reviewers had data on over 14,580 applications of NPWT and yet they "determined that the existing evidence is insufficient to draw any conclusions". The eligibility criteria was pre-defined and developed in discussion with the nominating organization. We applied a systematic approach for determining the eligibility of the studies. The number of studies excluded reflects the breadth of our search, demonstrating how hard we tried to find all available studies that met our eligibility criteria. We then applied a systematic approach to grading the strength of the evidence, as outlined in the AHRQ EPC Methods Guide. Because the population of patients is very heterogeneous, we evaluated the evidence separately for each patient population. The strength of evidence was graded based on study limitations, consistency, directness, precision, and reporting bias. The number of applications does not factor into the strength of evidence, nor should it. We did not make any changes to the report based on this comment.
Vickie Driver AAWC General This review team eliminated data on thousands and thousands of patients treated with NPWT, with the overwhelming majority receiving the treatments in their home. The team determined criteria that seems to overly value RCT studies, studies with a direct comparator arm, and studies designed to only report complete closure results. Yet, in key question #2 that lists the meaningful clinical outcomes, "rate-to-healing" and "time to surgical readiness of the wound bed" are included as clinical endpoints. As with any systematic assessment of a body of evidence, we sought to find the evidence on NPWT from the strongest study designs. We clarified in the Study Selection section of the Methods chapter the types of outcomes that were included.
Vickie Driver AAWC Methods The specific criteria applied by the reviewers in determining which data was relevant was not pre-specified in the TA protocol and appear to be an ad-hoc approach. The reviewers seem to have assumed if NPWT is applied in the hospital, which is the case for a majority of NPWT treatment applications, that the full study data was to be eliminated. Most of these rejected studies had 80-90% of all the NPWT treatments reported in the studies applied and delivered in the patients' homes. NPWT is commonly applied in the inpatient facility to treat the wound initially and to allow the patient, with a complex wound, to be discharged from a facility to his/her home sooner. Use of NPWT helps to contain and managed these wounds and supports the healing process. With that as a clinical reality, why would the assessment team not want to include the 80-90% of the application experience with NPWT in the home as part of their pool of information to determine safety and efficacy? We developed the scope of the review in discussion with AHRQ and the nominating organization. In the Study Selection section of the Methods chapter, we added our rationale for including patients only treated in the home setting. We clarified in the Study Selection section of the Methods chapter that we sought studies with all patients followed up in the home setting.
Vickie Driver AAWC Methods As mentioned earlier, over a thousand studies were rejected from the review based solely on a sample size of 20 patients or less. This has left a significant number of valuable clinical usage data out of the technology analysis. We clarified the inclusion/exclusion criteria in the Methods chapter in the Study Selection section and Table 1. The sample size criteria was developed in discussion with the nominator. We have added a rationale for this criterion. Also, we re-reviewed all of the studies that were excluded during full-text screening because of the sample size. All of the studies could have been excluded for other reasons (e.g., no comparison group, mix of patient populations, mix of settings, no original data). We also added to the Limitations of the Review Process a paragraph about excluding studies based on sample size.
Vickie Driver AAWC Methods The reviewers stated in their methodology that evidence was downgraded because "complete healing" was not defined according to the FDA's definition of healing. This makes no clinical sense. NPWT is not always utilized for complete healing. NPWT may be initiated to bring a wound to a certain level of granulation coverage to allow changing treatment back to a standard-of-care approach for closure completion. NPWT therapy is also often applied to elicit a specific response to enhance wound progression, such as: 1) removal of necrotic debris, 2) management of excessive exudate detrimental to healing, 3) to obtain bacterial control and reduction of bioburden to progress healing, and 4) achieve a certain percentage of granulation bed for other treatment modalities [e.g. tissue grafting, surgical flaps, etc.]. These studies evaluated and reported on complete healing, but they usually did not provide a definition for complete healing (Yao 2013; Armstrong 2012; Lerman 2010; and Ford 2002). Lavery 2007 defined wound closure, but used different definitions for wound closure for the NPWT and control groups. Neither definition was as stringent as the FDA definition.
Vickie Driver AAWC Methods In fact, many of the Medicare Local Coverage Decision [LCD] medical coverage policies for NPWT specify discontinuing use of NWPT when a certain level of granulation tissue is present or a percent of healing has occurred. This review team appears unaware of the clinical instructions that restrict use of NPWT for Medicare coverage, which impacts how the device is used in everyday practice and is reflected in many of the clinical citations identified by the reviewers. During the development of the review protocol, selection of outcomes, including whether to include wound healing rate, was discussed extensively amongst our review team and with the organization that nominated the topic. We have added rationale for excluding wound healing rate as an outcome to the Study Selection section of the Methods chapter. We have also included in the Limitations of the Review Process section of the Discussion chapter a paragraph regarding excluding studies that evaluated only wound healing rate.
Vickie Driver AAWC Methods In Key Question #2 in the report, the acceptable clinical outcomes identified for data retrieval are not all correctly stated for chronic wounds or applied in the review. We have identified those which are incorrect or are an issue below. Thank you for your feedback.
Vickie Driver AAWC Methods Key Question 2, Part 1: AAWC does not believe the statement "compared with treatment with other wound care methods" was intended by the CMS to suggest they only wanted data on studies with direct comparator arms or to ignore the "real world" data in comparison to other treatments available from the identified clinical citations. The CMS often asks researchers for data that translates to the Medicare population at large, instead of restrictive data from RCT studies that have many exclusion criteria, to truly understand the applicability of safety and effectiveness of data. We developed the inclusion and exclusion criteria in discussion with AHRQ and the nominating organization. The intention was to limit the review to only include studies with a comparison group.
Vickie Driver AAWC Methods Clinical outcomes: "a) Complete wound healing by secondary intention (i.e., healing without surgical repair)": Chronic wounds are not left to close by secondary intention. This term applies to open surgical or traumatic wounds, which due to extensive tissue damage or infection, are treated and deliberately left open to close by secondary intention. This is possible because there are healthy wound edges available for cellular closure. Chronic wound healing is not through the process of secondary intention, but it is through a complex, multi-cellular process of inflammation, formation of granulation tissue, re-epithelialization, matrix formation and remodeling. The protocol, including specifying the outcomes to be considered, was developed in discussions with AHRQ and the nominating organization.
Vickie Driver AAWC Methods Clinical outcomes: "b) Time to complete wound healing by secondary intention" Same inaccurate use of terms intended for traumatic and / or open surgical wounds. The protocol, including specifying the outcomes to be considered, was developed in discussions with AHRQ and the nominating organization.
Vickie Driver AAWC Methods Clinical outcomes: "c) Time to surgical readiness of the wound bed" Use of NPWT to prepare a wound site for "surgical readiness" for flaps, tissue grafting, or primary closure is one aspect of the use of NPWT technology. Although listed as an outcome to be evaluated, the reviewers excluded all studies that did not follow the FDA definition of "complete wound closure" as an endpoint, and thereby they have contradicted and violated their own protocol. We did not state that we excluded all studies that did not follow the FDA definition of complete wound closure.
Vickie Driver AAWC Methods More concerning in this report is the absence of other accepted clinical uses and intended outcomes for NPWT: removal of necrotic debris, removal of excess exudate, and management of infection/ bioburden/ bacterial levels. NPWT is often used for chronic wounds to only reduce the exudative phase of healing or to only remove excess necrotic debris or to only treat the bioburden and not necessarily continue use to total closure. The reviewers eliminated thousands of studies that reported some of these clinically relevant and appropriate outcomes for NPWT that would provide data on efficacy and safety of the technology. We developed the scope of the review, including outcomes, in discussion with AHRQ and the nominating organization. The intermediate or surrogate outcomes suggested by this reviewer were not considered in this review.
Vickie Driver AAWC Methods Many aspects of actual search seem inconsistent with the objective. Were they only searching for US studies? N> 20? English speaking? All methods should match the objective. We have corrected the Study Selection section and Table 1 of the Methods chapter. We now state that we excluded non-English language articles and studies with a sample size less than 20.
Vickie Driver AAWC Methods Discrepancies between stated inclusion/exclusion criteria in Methods (Table 1) and apparently arbitrary exclusion reasons used in Figure 2, page 12. Eighty-two (82) excluded studies may have qualified for inclusion and may change the conclusions. We have corrected Table 1 to be consistent with Figure 2. Criteria were pre-defined and noted in our protocol.
Vickie Driver AAWC Abstract Abstract does not describe significant risks and benefits found in the studies. Abstract: Should contain all clinically relevant statistically significant findings. We disagree. The abstract should contain the findings of the systematic review, not of the individual studies included in the systematic review.
Vickie Driver AAWC Methods Objective. Should be " To systematically review comparative efficacy and safety." Table 1 states that only comparative trials were included in their search strategy. This should match the Objective. However, it they are truly looking for efficacy and safety, it would seem logical to add the non-comparative studies. Some text to explain why this was not done may be appropriate. We revised the first sentence of the Methods chapter to state that we conducted a systematic review "on the effectiveness and safety of negative pressure wound therapy (NPWT) compared with other wound care treatments or other NWPT devices in the home setting."
Vickie Driver AAWC Methods Review methods. Include, if they did, a statement that they tested for publication bias. Because of the small number of studies included, we are unable to quantitatively assess publication bias. Planned testing was outlined in our protocol.
Vickie Driver AAWC Results Results. All significant benefits and risks in relevant studies should be in Abstract. This would meet goal to inform clinical decisions where evidence does exist. At a minimum, state a general overview of significant efficacy results:
i. Second sentence: Six of the studies compared NPWT devices to AA intervention on BB wounds (X studies on Y subjects) or CC intervention on DD wounds (XX studies on YY subjects) etc., No significant differences were found between NPWT and EE interventions on VU (# studies; # subjects), DU (# studies; # subjects), PU (# studies; # subjects), arterial ulcers (# studies; # subjects) or mixed etiology ulcers (# studies; # subjects). One study (# subjects with XXX etiology ulcers) compared two different NPWT devices.
ii. Significant results above would include Yao's finding of more venous ulcers closed with NPWT for efficacy and,
iii. Schwien pressure ulcer results addressing safety objective: significantly fewer ER visits (p< 0.01) and hospitalizations (p< 0.05) in a large home care sample of patients treated with NPWT compared to controls (Page 18 Adverse Events of patients with pressure ulcers)
Thank you for your comment. We do not agree that the abstract should detail the individual study findings. This could be potentially misleading, as we found significant flaws in study designs (which are elucidated in the report) and may render some of the findings unreliable.
Vickie Driver AAWC Discussion / Conclusions Add statement on risk of bias being generally high or unclear; however, retrospective evidence supports safety of using NPWT on pressure ulcers in home care and suggests the merit of further pressure ulcer prospective RCTs of efficacy. Future RCTs can reduce bias by blinding outcome evaluations. We strongly agree that the risk of bias is high. However, we cannot make the suggested conclusions because, as the reviewer acknowledges, the risk of bias is high. We do state in the Research Gaps section the need for future studies to be designed in a way to minimize bias.
Vickie Driver AAWC Introduction Page 2, Paragraph 9 to match strategy used, should this be: "especially as it applies to the disabled and/or elderly population over 65 years of age." We have changed that sentence to read, "especially as it applies to the Medicare population (i.e., those over 65 years of age or with disabilities)."
Vickie Driver AAWC Introduction Page 3. Item 1.e. Clarify what wound healing rate means. Do they mean % ulcer area reduction from baseline per week on study? (e.g. for healed wounds this would be 100% / # of weeks on study) We have clarified wound healing rate to mean "percent ulcer area reduction or other measurement."
Vickie Driver AAWC Introduction Page 3. Item 3.f. Do they mean dropouts for reasons related to the NPWT? This indicates any subject lost to follow up from any cause, and no further data could be collected.
Vickie Driver AAWC Introduction Page 4. Key Q3, 1) Clarify parameters: amplitude, frequency, duty cycle and duration of suction applied. We have added the following to clarify parameters, "including amplitude, frequency, duration of suction, and other parameters as specified by the device."
Vickie Driver AAWC Introduction Page 4. Key Q4, Add 1) f: dryness of the wound surface This is not an outcome of interest. Thank you for your comment.
Vickie Driver AAWC Methods Page 11. Last line. Were the authors of the omitted studies contacted to identify setting? Same query applies for the 11 studies excluded in Figure 2 because authors could not abstract relevant data. To get valid answers representing the full body of data requires this effort. The authors did not try to ascertain if any data related to the home care setting. The researchers could also have contacted the companies who may have been be able to provide PDFs of full-text articles and of conference posters or presentations. We detailed in the report the authors contacted for further details. We did not contact all authors of meeting abstracts and/or those articles from which we could not abstract data. Many had non-informative results such as "Comparison of so many different techniques and patients is difficult; nevertheless we observed good and quick results in all the treated cases." Many also didn't present results for NPWT vs. other therapies (even though not everyone received NPWT).
Vickie Driver AAWC Methods Page 12. Several reasons for exclusion in Figure 2 seem not to be discussed in the Methods section or Table 1. Arbitrary inclusion/exclusion of studies can lead to biased results. Excluding 82 studies for reasons I could not find in the Methods section undermines credibility and relevance of this AHRQ review. Please clarify rationale for exclusion and add to Methods section if appropriate. Including all studies that address Objective would strengthen the generality and relevance of this review. Examples of arbitrary exclusion:

  • Less than 20 subjects (Potential relevance of these 44 studies if they meet inclusion criteria, combined in appropriate meta-analyses)
  • NPWT device not available in US (7 studies - exclusion rationale?)
  • Meeting or conference abstract (12 studies)
  • Not in English. (19 studies) despite the statement in Methods: "We had no date or language restrictions"?
We have corrected Table 1 to be consistent with Figure 2. We have added text description of the criteria listed in Table 1, along with rationale for some criteria, such as sample size. All criteria were pre-defined, developed in discussion with the nominating organization, and outlined in our registered protocol.
Vickie Driver AAWC Results Page 18. Adverse Events Line 2. Was the adverse event in the ulcer treated with NPWT or a different one? Please state in text for all such events. We have revised the sentence regarding drop outs for the Ford trial on page 18. We now state, "Although the study reported reasons for drop out (two patients died, three were lost to followup, and one was deemed noncompliant), they do not report the number of drop outs in each treatment group."
Kenton Fong Spiracur Results The "highest-quality" RCT included was the Ford et al study from 20021 . This study which was included in the analysis says, "Twenty-eight patients with 41 wounds were recruited from the plastic surgery clinic and inpatient physician referral at Boston Medical Center." In a direct conversation with Dr. Ford on 7/8/14, he reported over the phone that approximately 10% of NPWT treatment time in his study was done on inpatients. This is very common with wounds, as wound care patients often have multiple co-morbidities or requirements for wound debridement that require brief hospitalizations. As part of how these chronic wound patients are treated, having brief hospitalizations is common. In addition, the Yao 2012 study2 includes outcomes for a patient population that include the full spectrum of care settings as well. This study which was included in the AHRQ analysis states, "In this study, 171 NPWT patients and 171 nonNPWT patients reflected real-world patients in a continuum of care settings. These patients had multiple co-morbidities, multiple clinic visits and ulcer-related inpatient service (hospitalization)." Thus, some inpatient treatment was performed in these included studies with NPWT at least at a similar rate to other excluded studies in the AHRQ report. The key questions from this review specified chronic care in the home setting. We clarified in the Study Selection section of the Methods chapter that we were focused on patients in the home setting because many adverse events occur in the home. We acknowledge that hospitalizations can occur in this population, and indeed, hospitalizations related to the wound was an outcome of interest. However, we interpreted the Armstrong and Ford articles differently than the Blume article. We excluded some studies because some patients were followed up in the hospital setting or because it was unclear if all of the patients were followed up in the home setting.
Kenton Fong Spiracur Results Importantly, the largest randomized controlled trials with the vast majority of NPWT treatments done in the home-setting were excluded from this analysis. Specifically, the Armstrong et al 2005 study published in the Lancet3 looked at treatment of diabetic foot ulcer patients with NPWT after fore-foot amputation. Some of these patients may have been admitted to the hospital for a couple of days after surgery, but most if not all NPWT treatments after this type of surgery were done in the home setting. These patients do not remain in the hospital after their amputation until wound closure and are treated primarily in home as chronic wounds. A reviewer familiar with wound care would recognize that the majority of care would be provided in the home setting. Correspondence on 7/8/14 with the first author, Dr. David Armstrong, confirmed that 89% of treatment days were done in the home/outpatient setting4. The benefit of this study design was that it showed diabetic wound healing in a relatively standardized wound, and mimicked the real life use of NPWT in the home/outpatient setting after debridement (which is a common use of NPWT in the home/outpatient setting). The other large diabetic foot ulcer comparative NPWT RCT study by Blume et al in 20085 actually states on page 633 of the publication that for the NPWT treated group, the total days of therapy in the home setting were 9,471 of 10,579 total days of treatment. This equates to 89.5% of all therapy days were in the home setting. Again, both these studies were excluded based on care setting (i.e. having some inpatient treatment time). If the objective of the AHRQ TA study is to evaluate efficacy in the home setting of NPWT the aforementioned studies should be included. The argument that outpatient follow up only was performed in the included studies does not make sense based on a careful review of the included studies, as both included and excluded studies had follow-up protocols that were similar and set in the outpatient setting. We excluded the Armstrong 2005 study because it was conducted on post surgical patients. The Blume 2008 study was excluded because some of the patients were followed up in the inpatient setting. We expanded the Study Selection section of the Methods chapter to include our rationale for our inclusion and exclusion criteria.
Kenton Fong Spiracur General Thus, more studies on chronic wounds should be included in the analysis. Specifically, the largest randomized controlled trials need to be included as these studies are indicative of effectiveness in the home/outpatient care setting. Importantly, the conclusion that the report was "unable to draw conclusions about the efficacy or safety of NPWT for the treatment of chronic wounds in the home setting due to insufficient evidence" is inappropriate because of the exclusion of almost all studies in the literature in an almost arbitrary manner, including the exclusion of the largest randomized controlled trials done primarily in the home/outpatient settings. Thank you for your comments. The key questions for the review were specific to care in the home setting.
Kenton Fong Spiracur Methods It is unclear why the authors chose to exclude studies with treatment populations that have greater than 89% of dressing changes done in the outpatient/home setting if the objective is to analyze effectiveness in the outpatient/home setting. We developed the scope of the review in conjunction with our sponsors. We clarified in the Study Selection section of the Methods chapter that we were focused on patients in the home setting because many adverse events occur in the home. We clarified in the Study Selection section of the Methods chapter that we wanted all patients followed up in the home setting.
Kenton Fong Spiracur Methods The vast majority of studies done on chronic wound and NPWT have therapy primarily in the home or outpatient settings (ex. Ford 2002, Armstrong 2007, Blume 2008, Yao 2012), as this is where these wounds are typically treated. In addition, it is a normal part of treatment of these types of wounds to have some short inpatient hospitalizations related to either the wound treatment or due to treatment for co-morbidities in this frail patient population. These short hospitalizations are a part of normal care for wounds in this patient population, and this is evident to anyone clinically treating these patients. The alternative study design to exclude hospital treatment would be difficult as it might limit appropriate care and would not be consistent to how these patients are normally treated. Thus, having a study without any inpatient use of NPWT is not practical or realistic for evaluating chronic wound outcomes. Direct conversations with the primary authors (Armstrong and Ford) of two papers confirmed that only approximately 10% of NPWT therapy time was in the inpatient setting, and this same level was reported in one excluded RCT study as well (Blume 2008). Thus, it makes sense that these studies at a minimum and perhaps other studies that examine chronic wound treatment with NPWT and some small amount of inpatient treatment should be included in this analysis, just as the Ford 2002 study was included in this analysis. We developed the scope of the review in discussion with AHRQ and the nominating organization. We clarified in the Study Selection section of the Methods chapter that we were focused on patients in the home setting because many adverse events occur in the home. We acknowledge that hospitalizations can occur in this population, and indeed, hospitalizations related to the wound was an outcome of interest. However, we interpreted the Armstrong and Ford articles differently than the Blume article. In the Armstrong and Ford articles, it seems as if all of the patients were followed up in the home, but some were hospitalized for part of the followup period. The Blume article seemed to followup a majority of the patients in the home setting and minority of patients in the inpatient setting.
Kenton Fong Spiracur Methods To raise the question of efficacy and safety of a therapy based on care setting, there should be significant differences between those care settings that warrant investigation. There are not significant differences in how the intervention under investigation (NPWT) is utilized between care settings, as the protocols for treatment of wounds in all care settings are the same when concerning chronic wounds such as diabetic foot ulcer, venous ulcers, and pressure sores. The adverse events mentioned in the FDA warning concerning monitoring in the home setting of NPWT use was for electrically powered pumps and primarily for wounds that have high complication rates (such as NPWT treatment over vascular grafts or vascular structures such as heart (where bleeding can occur), deep tunneling wounds (where foam can be left behind and infection rates are higher), or very large acute wounds. All of these are not the typical chronic wounds that tend to be smaller and more superficial that are treated with NPWT in the outpatient or home setting. Importantly, these FDA warnings were for electrically powered NPWT pump systems and not mechanically powered or non-powered NPWT systems like the SNaP Wound Care System. To date, there have been no reportable adverse events to the FDA for the mechanically powered SNaP Wound Care System. We believe that the risk for severe complications such as exsanguination are remote with mechanically powered NPWT devices like the SNaP Wound Care System due to its exudate capacity and lack of an electrical mechanisms that could fail. This raises the question of what are the significant differences in potential safety that warrant a separate analysis of these types of devices in the home/outpatient setting. For most chronic wounds, the treatment regimen and protocol is identical to that of inpatient treatment. Thus, data from all care settings or mixed care-setting should be appropriate for determining efficacy. We have added to Limitations of Review section that this review was focused on one care setting. We do not feel that excluding studies that did not address the key questions, which specified home care setting, was arbitrary.
Kenton Fong Spiracur Methods The questions asked by CMS for evaluation by AHRQ are more effectiveness questions than efficacy-related questions as they asked the AHRQ to evaluate care setting specific questions. This distinction is important, as the criteria for evaluation of studies and the evaluation techniques used are different between efficacy and effectiveness evaluatons. As we move from efficacy to effectiveness, we agree that data from studies that mimic the actual use settings of the intervention are important. These studies do not stand alone in interpretation, but are interpreted in the context of existing literature on efficacy from all care settings. Importantly, for evaluation of effectiveness, the studies included should mimic how the intervention is used in real life. In the case of NPWT, for patient with chronic wounds they receive NPWT over a spectrum of care settings, but primarily in the outpatient/home setting. However, many of the core NPWT studies that were performed in the population of chronic wounds in the outpatient/home setting were not included in this analysis for arbitrary reasons as discussed above.   We have added to Limitations of Review section that this review was focused on one care setting. We do not feel that excluding studies that did not address the key questions, which specified home care setting, was arbitrary.
Kenton Fong Spiracur Methods It is obvious from the errors and omissions made in the evaluation of the studies included in this report that the persons evaluating these studies are unfamiliar with chronic wound care. If they did, they would have recognized that the care settings for the key studies in the field are primarily outpatient or home-based studies based on the duration of intervention and follow-up as well as the type of wound treated and inclusion/exclusion criteria utilized. In addition, as noted above, careful reading of the studies would have revealed these findings as well. We followed a systematic approach for evaluating the eligibility of the included studies. We have added to Limitations of Review section that this review was focused on one care setting.
Kenton Fong Spiracur Results The Armstrong et al 2012 study6 comparing two NPWT treatment modalities had comments that were addressed in manuscript, but may have been missed by the evaluators. We think that the reviewers are referring to the outcomes. This study reported on a variety of daily activity questions. However, we only reported on a few that reflected on a return to prior function, per the outcomes we were seeking.
Kenton Fong Spiracur Results The authors repeatedly comment that the wound size baseline characteristics may have influenced results. Multivariate analysis confirmed outcome findings even when wound size was taken into consideration, and these data demonstrate that wound size did not affect outcomes based on multiple types of analysis that are reported in the manuscript. Specifically, this analysis demonstrated that baseline wound size did not affect the outcomes on wound size reduction and full healing rates. Also, included in this analysis was a subset analysis with exclusion of larger size wounds which normalized wound sizes in each group and still resulted in non-inferiority on statistical analysis. Thus, the results and conclusions of this study should not be discounted based on this baseline characteristic. Understanding baseline characteristics, such as treatments received and wound characteristics, are important to understanding the results of the studies. We thus reported these details or noted when studies did not provide these details.
Kenton Fong Spiracur Results The authors are critical because blinding was not performed. Blinding was performed on analysis of Visitrak data tracings, but it would not have been feasible to blind subjects or investigators to treatment condition based on differences in size, sound, and weight of the two treatment interventions. The FDA guidance document7 on wound research says, "In general, blinding of subjects and investigators to the assigned treatment reduces bias and should be employed when feasible. Early trials of topical wound-treatment products often include a treatment group that receives only standard care (see section IV.B., Comparator Arms) to establish whether the vehicle has an effect on healing. Often the standard care only arm cannot be blinded. In other cases, especially for trials of some medical devices, it is impractical or unethical to implement a control treatment that mimics the test product for the purposes of blinding. In these situations, blinded assessment by a third-party evaluator should be considered." In this study, the assessment of objective measurement of wound tracings was done by blinded third party observer since it was impossible to feasibly blind for treatment groups the patients or the clinicians performing dressing changes. Thank you for the clarification about masking of analysis of tracings. The outcomes assessed with tracings were not outcomes of interest.
Kenton Fong Spiracur Results The authors are critical of who placed the dressing during the study was not described. The dressings were placed by the staff at the wound care centers. Typically the nurses do dressing changes, though occasionally doctors will do some. This is obvious to anyone who has been to a wound care center. Under GCP, the study staff is not allowed to interfere with treatment. In completing systematic reviews, we abstract information reported about the studies and do not make assumptions based on what may be typical practice. Details about the personnel applying an intervention is a standard piece of information collected and assessed in evaluating studies.
Kenton Fong Spiracur Results The authors state that it was unclear how incomplete outcome data was handled. The manuscript reported fully how missing data was handled as reported on page 333 of manuscript in the statistical analysis section, "If a wound was completely healed or a subject had discontinued prior to completion of 16 weeks, the last observation was carried forward for the intent-to-treat analysis." All outcomes are reported based on this unless specified in results section otherwise. Thank you for your comment and clarification. We have removed this statement from the text.
Kenton Fong Spiracur General There appears to be some confusion between home care setting and outpatient settings. The home setting is in the patient's home. Outpatient setting is in a clinic or physician office. Treatment is very different between these two settings, and it is unclear if this distinction is drawn in the analysis or appreciated by the authors. In home setting, patients have a visiting nurse come to perform dressing changes with only occasional supervision by a doctor. In the office setting the patient visits the clinic and a doctor may supervise each dressing change. Although these settings are different, the analysis seems to combine or blur these together. In our opinion, the differences in treatment in terms of efficacy are not significant between any care setting, inpatient, outpatient, or home (as the same therapy and protocols are used in each care setting), but for consistency the authors should keep these distinctions clear. It is the home setting that is the focus of CMS's key questions posed not the outpatient setting. Yet, a large portion of the study research focuses on the outpatient setting use of NPWT. The setting was poorly defined in most studies. We noted our interpretation of setting in methods, a few places in the results, and in the discussion. Limiting our review to only studies that explicitly stated that care was in the home would have limited the review greatly (i.e., to only one study).
Sue Lapp None General I have been involved in the Advanced Wound Care industry for over 25 years and seen great progress made in NPWT, and advanced wound care dressings. These advances have made a difference in the lives of many beneficiaries. In advocating for patients, I have seen the differences made in their ability to get back to their activities of daily living. After reading the draft,assessment, I was dismayed that "you were unable to draw conclusions about the efficacy or safety of NPWT for the treatment of chronic wounds in the home setting due to insufficient evidence". To me that represents a very large set back for the many beneficiaries that would benefit from NPWT in the home care setting. We agree that it was unfortunate that we were unable to draw conclusions due to insufficient evidence.
Sue Lapp None General While I recognize that you were looking for patient quality outcomes, many of those studies were done before the ACA mandated those reporting standards for future studies. The PCORI who looks at comparative effectiveness will only derive in the future that we are headed to "good, better, best" medicine and delivery. We cannot afford to make NPWT, the treatment of last resort. All the wound care products and NPWT are all "aids" when it comes to advanced wound healing. I agree however, that more must be done to fully understand the co-morbitities of the beneficiaries so as to assess the potential for complications. Thank you for your comments.
Sue Lapp None General Local Coverage Determinations for NPWT have always said that the documentation of the wound, type of wound, size and dimensions were essential in reporting that information on claims. The CERT rate for NPWT remains high. NPWT companies however, have been extremely diligent in their patient education materials, user guides, toll-free 800 call-in numbers when there is an issues with the unit, and 24-hour hotlines for patient concerns. They have printed those materials in Spanish to help with bi-lingual needs for beneficiaries. Teaching materials have pictures, symbols and trouble shooting guidelines. In the need to drive patient compliance, companies provide in-service education to providers so they understand the NPWT devices and can communicate effectively with the licensed professionals about the nuances of the unit. This is done so that the patient can have reasonable expectations about what will happen after leaving the acute care setting. Patient's have an opportunity to reach out at any time during their therapy if there is a questions regarding the unit and dressing changes. Many materials
have diagrams that clearly illustrate what to do if something does not appear right. And, fire and safety instructions are all clearly indicated in the User Instructions Guides, which are in accordance with the
standards for safety that the FDA mandates.
Thank you for your comments.
Sue Lapp None General I understand what the information from the assessment says. To fund additional long term large scale studies in a technology that is seeing smaller units, and mechanical NPWT being done without electricity does not seem necessary. If you want to have a study looking at specific outcomes for the beneficiary that generates outcomes data to see how the patient is doing long term with their ADLs then that seems like a better path. Thank you for your comments.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Introduction In the Introduction the authors state that "The exact mechanism by which these devices may promote wound healing is not known. Hypotheses for the healing effect include the removal of excess fluid while improving circulation to the wound bed,11, reducing bacterial load on wound surface, or the presence of a mechanical effect that aids wound healing.12" The authors cite amongst other references the original Wake Forest publication (Morykwas, Argenta, Shelton-Brown, & McGuirt, 1997). This seems to imply to a non-specialist reader that the therapy is poorly understood and relatively uninvestigated, whereas on the contrary, there has been rather a lot of progress and understanding developed over the NPWT mechanisms of action: see Glass & Nanchahal for an independent recent review (J Plast Reconstr Aesthet Surg 2012;65(8):989-1001.). In particular the reduction of bacterial bioburden in contaminated clinical wounds under NPWT has been extensively investigated and now widely accepted as not to be a major mechanism of action; hence the increasing interest in the use of antimicrobials, of one sort or another, in combination with NPWT (Siegel HJ, et al., Clin Orthop Relat Res 2014;472(3):830-5; Kim PJ, et al., Plast Reconstr Surg 2013;132(6):1569-79. We have revised this sentence in the Introduction to state, "There are many mechanisms by which these devices may promote wound healing." and now list a few of these mechanisms.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Introduction It is likely that several mechanisms are responsible for the benefit of NPWT, including: fluid removal, drawing the wound together, micro-deformation, and moist wound healing (Orgill DP, Bayer LR. Plast Reconstr Surg 2011;127 Suppl 1:105S-115S); systemic mobilization of endothelial progenitor cells (EPCs) during NPWT (Seo SG, et al., Exp Mol Med 2013;45:e62); and support [of] (neo-) angiogenesis and transformation of chronic non-healing wounds in a physiological wound healing process when combined with surgical debridement (Malsiner CC. et al., Int Wound J 2013 Sep 13. [Epub ahead of print]). In summary, the authors could have done a much better job of describing the mechanism of action of NPWT by reviewing more recent references in the literature. We have revised the first sentence of the fifth paragraph in the Introduction to state, "There are many mechanisms by which these devices may promote wound healing." We also incorporated some of these references.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Introduction It should be noted that the goal of NPWT is to improve wound healing to the point where application of traditional dressings that maintain a moist wound environment can be applied to finish the process. NPWT is ideal for chronic wounds that are "stuck" and unable to progress. NPWT is cost effective by helping certain wounds progress through the healing process which in turn reduces the hospital readmission rates and overall healthcare expenditures. We believe that this point was missed by the authors. We have revised the introduction to include mechanisms by which NPWT may promote healing and to list the aims of NPWT.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers   Although NPWT is often initiated in the hospital and patients' transition to the home, it is not a straight path to home care. Patients can transition to the home, long term care, or long term acute care facilities. Their wounds can also be treated in a hospital affiliated or independent wound clinic, physician office, or visiting nurse association in the home settings. Each of these settings has different standards of care and delivery methods and thus makes it difficult to define control groups and monitor study protocols. We developed the scope of the review in discussion with the AHRQ and the nominating organization. We agree it can be difficult to define the setting and have noted a few places in methods, results and discussions, how we interpreted studies that did not explicitly state that care was in the home setting.
Marcia Nusgart Alliance of Wound Care Stakeholders   The authors discuss the limitation to chronic wounds in the present review and although they acknowledge the extensive use of NPWT on acute and surgical wounds, they seem to imply that use of NPWT in "surgical" wounds is exclusive to acute care settings and not in the home. This is misleading and misrepresents the true burden of post-surgical wound dehiscence that is treated in the homecare situation with NPWT. Although it is an imprecise term, such wounds are often referred to as "sub-acute" and in a wound after 4 weeks of non-healing the biology of these wounds will quickly approach that of chronic wound with microbial contamination, poor extracellular matrix, edema and exudate (Schultz GS, et al., Int Wound J 2004;1(1):19-32.). A very significant proportion of wounds managed by NPWT in outpatient clinics or in the home setting are dehisced surgical wounds with delayed healing For example an analysis of Outcomes and Information Set (OASIS) data on outcomes in NPWT-treated Stage III or Stage IV pressure ulcers (N = 98) or surgical wounds (N = 464) found almost five times the number of post-surgical wounds treated at home compared to pressure ulcers (Baharestani MM, et al., Ostomy Wound Manage 2008;54(11):48-53.). Significantly, outcomes from early initiation of NPWT in the dehisced surgical wounds were excellent. Similar data are found overseas. In an analysis of NPWT in homecare in the UK (known as Community-based wound care in the UK) 68% of all NPWT episodes were for post-surgical dehisced wounds (Dowsett C, et al., Int Wound J 2012;9(5):544-52.). Unless a similar AHRQ review is scheduled to be undertaken shortly on the use of NPWT on dehisced surgical wounds, it seems arbitrary to exclude these wound types when management of these wounds represent a major use of NPWT in the home care setting and moreover that such wounds are considered to respond well to NPWT. Thank you for this information. We have deleted the reference to setting in this sentence in the Introduction.
Marcia Nusgart Alliance of Wound Care Stakeholders Introduction Question 2 Part A: what does "similar to Medicare patients" mean? The stated definition (age 65 or older or disabled) does not even begin to really define what this hypothetical population is. In discussion with the nominator during protocol development, it was decided to include studies regardless of age or disability. Few studies have a study population that is exclusively Medicare patients. This difference in population is discussed under Applicability. We have also clarified this in the Study Selection section of the Methods chapter.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Methods Why aren't hospital/inpatient/LTC studies results applicable to the home setting? There is no delineation between hospital use and home use. Patients are not followed in the inpatient setting long enough to make determinations that hospital-based NPWT is insufficient. Patients are often started inpatient then transferred to LTC or home use. The decision just to look at studies that exclusively had home settings is flawed because the use of NPWT does not change between settings and such a strategy excluded many valuable studies. The scope of the project was to examine NPWT use for chronic wounds in the home setting. We acknowledge in Discussion this limitation, given that NPWT is used across the healthcare spectrum, and that these patients often change the location of their treatment.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Methods Patient compliance with wound therapies, nutrition, psychosocial concerns, management of comorbidities and environment are all critical components in wound management. Management of the above can be closely monitored and documented in an acute care setting. The home environment presents greater challenges in ensuring the above variables are appropriately managed. Therefore, the home care environment presents many barriers to obtaining quality study data and it may be unrealistic to expect outcomes of the type proposed in the TA. We believe that having a clinician on the team who is well versed in the practice of NPWT at home would have made this TA much more useful. We agree that these are particular challenges of conducting research in the home environment. However, studies may still adhere to rigorous methods.
Marcia Nusgart Alliance of Wound Care Stakeholders Methods Whereas the protocol registered at Prospero specified that for RCTs the Cochrane risk of bias tool would be used and the Newcastle-Ottawa tool used for observational trials, it appears that in the actual assessment a modified Cochrane risk tool was used for all studies; moreover the predefined critical outcomes were somewhat different to the usual Cochrane tool and NOT pre-specified. Use of the Cochrane tool for risk of bias assessment for observational trials is WHOLLY inappropriate. We decided to use the Cochrane Risk of Bias tool to assess study quality for all studies. The Cochrane Risk of Bias tool can be used for any study design. Added considerations, such as appropriate comparison group, are also discussed in the report.
Marcia Nusgart Alliance of Wound Care Stakeholders Methods No inter-rating scores (e.g., kappa values) were published regarding the level of agreement between reviewers (risk of bias) for risk of bias assessment. We cannot tell if the two reviewers agreed most of the time or disagreed on every assessment. This is an unknown source of bias. We did not capture original rates of agreement. As noted in report, any disagreements were resolved through discussion.
Marcia Nusgart Alliance of Wound Care Stakeholders Methods Some studies were rejected for review based on a sample size of 20 or less. This was not pre-specified and is an ad-hoc approach that might leave out some useful studies. There is not rationale for using a sample size of 20. Moreover, the authors might have been able to perform a meta-analysis of some outcomes had they left these studies in. We clarified the inclusion/exclusion criteria in the Methods chapter in the Study Selection section and Table 1. The sample size criteria was developed in discussion with the nominator. We have added a rationale for this criterion. Also, we re-reviewed all of the studies that were excluded during article review because of the sample size. All of the studies could have been excluded for other reasons (e.g., no comparison group, mix of patient populations, mix of settings, no original data). We also added to the Limitations of the Review Process a paragraph about excluding studies based on sample size.
Marcia Nusgart Alliance of Wound Care Stakeholders Methods Evidence was downgraded when "inappropriate control groups" and "surrogate outcomes/endpoints" were utilized in studies. Although in some studies there can be some questions regarding these issues, downgrading the evidence on this basis was overly conservative and not justified. Having an appropriate comparison group and valid outcome measures are basic principles of research and are often used in quality assessment. We did not make any changes to report based on this comment.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Methods Evidence was also downgraded because complete healing was not defined according to the FDA's definition of healing. For observational trials and non-pivotal trials this is often inappropriate as most of these studies would not use such a definition in practice or pragmatic trials. Moreover, NPWT is often used to partially close a wound to prepare a wound for further surgery (e.g., flap closure). The authors knew this because one of their planned outcomes was "Time to surgical readiness of the wound bed." We would submit that then to downgrade evidence just because outcomes did not meet the FDA's definition of healing is illogical and arbitrary. These studies evaluated and reported on complete healing, but they usually did not provide a definition for complete healing (Yao 2013; Armstrong 2012; Lerman 2010; and Ford 2002). Lavery 2007 defined wound closure, but used different definitions for wound closure for the NPWT and control groups. Neither definition was as stringent as the FDA definition.
Marcia Nusgart Alliance of Wound Care Stakeholders Methods The level of evidence was cited as insufficient. However, in other systematic reviews higher levels of evidence have been found in a broader context. For example in a recent systematic review conducted by the VA they stated: "For diabetic ulcers … [T]here was moderate-strength evidence for improved healing with a biological skin equivalent (relative risk [RR], 1.58 [95% CI, 1.20 to 2.08]) and negative pressure wound therapy (RR, 1.49 [CI, 1.11 to 2.01]) compared with standard care… Compared with standard care, some advanced wound care therapies may improve the proportion of ulcers healed and reduce time to healing, although evidence is limited" (Greer N, et al., Ann Intern Med 2013;159(8):532-42.).  We included some details of studies and reviews of other wound types and settings in the discussion but these were not within our scope (i.e., did not meet our inclusion criteria for addressing NPWT used in treating chronic wounds in home setting).
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Methods Studies were excluded if they did not have a comparison group. While we understand this is reasonable for determination of efficacy/effectiveness, less common safety issues are usually reported in large case series. Excluding these can miss valuable data. The type of studies to be considered in this review were outlined in our protocol, after discussion with AHRQ and the nominating organization. We have added a statement to Limitations of Review about potential benefit of other study designs in identifying adverse events.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Methods Sometimes NWPT devices are used in heavily exudating wounds in which the patient is not a surgical candidate due to co-morbidities and attempts at alternative wound care has been attempted and has been unsuccessful. NWPT can allow for fewer dressing changes and decreases nursing utilization. It also increases patient comfort and quality of life. This was not included as an outcome. Thank you for your suggestion. As described, the outcomes suggested may be considered surrogate outcomes for pain and quality of life. We included pain and health-related quality of life as outcomes.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Methods Table 3. Manufacturer/Company – Mendela is incorrect. It should instead be "Medela." Thank you. We have made this correction.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Discussion / Conclusions The level of data in the literature (over 5,000 citations) is enough for AHRQ to make some sufficient declarative statement on the efficacy and safety of NPWT. Even the study conducted by Fife et al shows this. For the authors to state that Lavery et al's 2007 study on DFUs is insufficient, tells us that the authors were too stringent on the guidelines for clinical data. We are unclear if the "5,000 citations" noted refer to our search results or not. If this does, it is misleading to suggest that there are 5,000 citations addressing NPWT as these citations were identified using a broad, comprehensive search. We applied standards for selecting studies based on pre-defined eligibility criteria, assessing the risk of bias and grading the strength of the evidence as outlined in our protocol, and in the AHRQ EPC Methods Guide. As we noted in the review, the Fife and Lavery articles had severe methodological flaws and a high risk of bias. We are not able to draw any conclusions based on these studies.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Discussion / Conclusions Based on data analysis, to dismiss the observational studies as being too inconclusive is unfair given that carriers have adopted less stringent guidelines for acceptance of data that is in real-time. If observational and retrospective studies are acceptable for cellular and/or tissue-based therapies then they should be acceptable for NPWT. We question whether the authors really understand some of the observational studies: "…larger studies were retrospective and based on administrative databases." For example, the study conducted by Fife et al (Int Wound J 2008;5 Suppl 2:17-22.) is based on electronic health records and is NOT an administrative database. We used standard methods for assessing the risk of bias and for grading the strength of the evidence. As outlined in the AHRQ EPC Methods Guide. We revised the second paragraph of the Limitations of the Evidence Base to clarify that some studies also used a patient record database. Both administrative and patient record databases, being designed for other purposes, typically are not complete for research purposes.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Discussion / Conclusions A further concern is the omission of a number of randomized controlled trials (RCTs) from study inclusion because they were not entirely conducted in the home setting (Armstrong DG & Lavery LA, Lancet 2005;366(9498):1704-10.; Blume PA, et al., Diabetes Care 2008;31(4):631-6.). The study by Armstrong and Lavery was subjected to an economic analysis published by Apelquist et al in 2008 (Am J Surg 2008;195(6):782-8), which details where care was given: A total of 24.4% of all dressing changes were performed in the hospital, 18.3% in the outpatient clinic, and 42.1% during home care. In terms of the 10,908 days of total therapy, 9,719 (89.1%) and 1,189 (10.9%) were in the home care and inpatient settings respectively. Similarly, in the randomized study in 342 diabetic foot ulcer patients conducted by Blume et al., which was carried out in 37 centers, the proportion of home care therapy days to total therapy days was 9,471 of 10,579 (89.5%) for NPWT and 12,210 of 12,810 (95.3%) for AMWT (the controls). Given that this is a population consisting of diabetic foot ulcers or patients receiving amputations for foot ulcers, some inpatient care is absolutely necessary (i.e., this is "real life."). Moreover, in regard to inpatient vs. outpatient NPWT, the data are quite clear that this has evolved to a "bridge to outpatient" technology. Thus, we question the decision to exclude such studies as not being "sufficiently" homecare based. Could the authors have not conducted an analysis of these kinds of trials in terms of subpopulations? We developed the scope of the review in discussion with the AHRQ and the nominating organization. We noted a few places in methods, results and discussions, how we interpreted studies that did not explicitly state that care was in the home setting. We also included in the discussion details about some key studies, such as Blume, that did not meet our review eligibility criteria.
Marcia Nusgart Alliance of Wound Care Stakeholders Discussion / Conclusions The AHRQ TA included the issues related to the 2011 FDA safety communication regarding serious complications associated with NPWT devices used in the home. The Agency expressed concern that, although these devices can be used safely and effectively in that setting, greater risk mitigation is required to prevent patients using these devices from experiencing serious health problems. As stakeholders in the wound care community, the Alliance of Wound Care Stakeholders agreed with the FDA. The Alliance also took this very seriously and partnered in 2011 with the FDA staff to help them with their MedSun Program which was dedicated to developing a relationship with the clinical community to learn about, understand, and solve problems with the use of medical devices as they are used in the health care environment AND to provide useful safety information back to health professionals to improve patient safety. We helped them with questions in their MedSun NPWT survey and gave them lists of Alliance members who were clinicians that they could send the survey to complete. We invited FDA staff who had responsibility for these issues to speak at our meetings and to the largest wound care conference about the MedSun program and NPWT survey. Thank you for your comment and commitment to patient safety.
Marcia Nusgart Alliance of Wound Care Stakeholders Discussion / Conclusions The Alliance went a step further-by developing and endorsing a set of proposed standards specifically for NPWT suppliers. Although physicians, nurses, physical therapists and other health professionals have primary responsibility for managing and overseeing care of NPWT patients, suppliers of these products play a vital role in ensuring the safe and effective use of these devices in the home. The training and 24/7 support that suppliers provide to professional and caregivers is critically important and requires a different set of competencies than those required for simple functional products. For that reason, the Alliance believed that the competency of all NPWT suppliers, including those participating in the DMEPOS competitive bidding program, should be validated through the external accreditation program already established by CMS, much like the product-specific accreditation required for suppliers of prosthetics, orthotics, rehabilitation products and respiratory products. Validating the competency of suppliers to provide support NPWT patients and caregivers is no less important. Thank you for your comment and clarification.
Marcia Nusgart Alliance of Wound Care Stakeholders Discussion / Conclusions In support of the FDA's recommendations and to assist CMS in the development of NPWT quality and accreditation standards, the Alliance developed and endorsed a set of proposed standards specifically for NPWT suppliers. On the Alliance's website (under comments) is a 2011 letter to CMS Administrator Tavenner stating the need for these standards and that the accreditation process would ensure that NPWT suppliers have the expertise required for support of the clinicians and patients using these devices. We met with the Administrator and her staff to emphasize that CMS adopt these standards. Thank you for your comment and clarification.
Marcia Nusgart Alliance of Wound Care Stakeholders Discussion / Conclusions Since then, CMS used the standards to create interpretative guidelines which were incorporated into standards now used by accrediting bodies. Suppliers who want to furnish NPWT into the home must be accredited. Thank you for your clarification.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Discussion / Conclusions The authors are rather dismissive of the retrospective nature of the study that (Fife et al. performed in 2007 (published in 2008) which evaluated adverse events for the VAC. A point overlooked in the analysis in the TA is that every prospective RCT of NPWT a priori excluded patients on anticoagulants so as to reduce the risk of bleeding. All prospective trials of NPWT have also excluded patients with most major comorbidities (Carter MJ, et al. Adv Skin Wound Care 2009; 22:316-24.) What was unique about the 2008 Fife study is that it specifically looked for patients treated with NPWT in the home setting who were on coumadin, heparin or other medications associated with an increased risk of bleeding. Since patients taking these medications had been excluded from the prospective trials, the only way to analyze the safety of NPWT in the home setting for patients on these medications was via retrospective analysis of the data which had been prospectively obtained. Thank you for your clarification.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Discussion / Conclusions The published study referenced in this AHRQ TA was a subset of a much larger analysis performed at the request of the FDA. The FDA requested that KCI provide data on the safety of the VAC in the home setting in relation to moist wound care. Nearly 10,000 wound care patients were analyzed, approximately 10% of whom underwent NPWT treatment with the VAC. We think that the FDA report is informative and that AHRQ would find it useful. In that analysis, 200 patients on anticoagulants were found among the roughly 900 patients undergoing treatment with the VAC, none of whom had bleeding. Only one patient had VAC treatment discontinued due to blood-tinged drainage (not frank hemorrhage). That patient was on clopidogrel bisulfate and his wound was in the 99th percentile for size within this dataset. Thank you for this information. We did not locate the FDA report mentioned (we searched the FDA Web site) and are uncertain if it is publically available. Additionally, the commenter did not provide the report, so we could not include it.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Discussion / Conclusions The average number of major co-morbidities among patients treated with NPWT in this large data set was 8. As the 2009 Carter et al study showed, prospective trials have excluded patients with any serious diseases. So, while the AHRQ report is critical of studies utilizing retrospective data, we submit that analysis of registry data is the ONLY way to assess the safety of NPWT. This is because prospective trials a priori exclude patients with significant co-morbid diseases and patients on medications which increase the risk of bleeding. We disagree with this comment. Safety can be assessed through well-designed prospective cohort studies. Unfortunately, the information in registries is often not entered with research in mind and may not be complete. In our review, we included any type of study design, including registry data, so long as it had a comparison group. We assessed the quality of each study. We graded the strength of the evidence based on the study quality.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Discussion / Conclusions In summary, it appears there is much bias against using registries as the best way to follow adverse events in the real world. For example, AHRQ would not even mention The U.S. Wound Registry in their upcoming edition of the AHRQ book on registries. It appears to us that AHRQ is unsupportive of efforts of the wound care industry to collect this type of data and make clinicians aware of its availability despite our efforts at education. This registry is now expanding its abilities to receive data as a result of the QCDR process. Thank you for your comment. We are not involved in the AHRQ book mentioned.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Discussion / Conclusions Another key point is that these technologies require better operational definitions regarding when to start and stop. NPWT appears to be most helpful in reducing depth/complexity of wounds. It should then, generally speaking, be stopped in favor of less per-device expensive devices or just split thickness skin grafts when a suitable end point has been reached (Isaac et al, In: The Diabetic Foot: New York: Elsevier, 2013, 899-909). Thank you for your comment. We agree that better definitions are required.
Marcia Nusgart Alliance of Wound Care Stakeholders & Coalition of Wound Care Manufacturers Discussion / Conclusions Given that U.S. government agencies are unwilling to fund NPWT clinical studies, most of the cost of these trials must fall on the manufacturers. Adopting an unnecessarily narrow view point of the comparative clinical evidence does not encourage greater industry investment in clinical studies supporting innovations such as antimicrobials, instillation, or disposable NPWT devices. As all stakeholders have at heart the objectives of improving patient outcomes with the most cost efficient treatment protocols, we believe the conclusions of this TA, as it stands, does not serve patients. Thank you for your comment.  We agree that there needs to be some way of funding robust research to continue medical innovation. 
Marcia Nusgart Alliance of Wound Care Stakeholders Discussion / Conclusions Finally, we welcome the "characteristics of an ideal study" as outlined in Table 11. Thank you for reviewing our report.
Marcia Nusgart Coalition of Wound Care Manufacturers Discussion / Conclusions The AHRQ TA included the issues related to the 2011 FDA safety communication regarding serious complications associated with NPWT devices used in the home. The Agency expressed concern that, although these devices can be used safely and effectively in that setting, greater risk mitigation is required to prevent patients using these devices from experiencing serious health problems. As stakeholders in the wound care community, the Coalition of Wound Care Manufacturers agreed with the FDA. The Coalition also took this very seriously performed the following activities in 2011:

  • Worked with FDA staff to educate them on how manufacturers inform the caregiver, patient and clinician on the correct use of NPWT in the home. This was accomplished by:
    • Marcia Nusgart spoke with Nada Hanafi of FDA on this issue
    • The Coalition gave input to FDA staff on the NPWT MedSun survey
    • Marcia Nusgart invited FDA staff Mary Brady and Diana Rivi (who has responsibility for MedSun survey) to speak on her panel at two SAWC meetings.
  • Invited OIG staff (Christine Moundas) responsible for its NPWT report to speak on Marcia Nusgart' s panel at SAWC Spring meeting and met with her on this issue.
Thank you for your comment and clarification.
Marcia Nusgart Coalition of Wound Care Manufacturers Discussion / Conclusions Finally, we welcome the "characteristics of an ideal study" as outlined in Table 11 but question practically--what is the right wound study? Wounds are different by each co-morbidity, cause of wound and location of wound. Therefore a perfect clinical study is hard to define. Thus there are a variety of clinical study types. Total wound closure was one of the criteria used by AHRQ in their review, but not always a frequent endpoint for studies, i.e. percent closure etc. Thank you for your comment. Table 11 presents key characteristics for studies of NPWT in wound care. Rigorous methods are applicable for studies of different wound types and location. We also listed characteristics for methodologically strong studies, not based on, for instance, frequency of use of measures in studies to date.
Paul Rockar APTA General
  • The term "venous stasis" has not been considered accurate by many experts for a number of years. It implies stagnant, poorly oxygenated blood in the venous system. The more accepted and accurate terminology is "venous insufficiency" and similarly "venous insufficiency ulcers." Unless the terms are used to reflect terminology in specific studies we recommend the use of "insufficiency."
We have edited the report to use the term, "venous insufficiency ulcers."
Paul Rockar APTA General
  • The use of the terms "complete wound healing" versus "closure" was also unclear. These are very different but identifiable points in time. This is discussed in more detail under Methods.
We added the following to details in Methods about outcomes of interest: "Generally, closure of the surface of the wound by the growth of epithelium over the defect is healing. If the closure has durability over time it is complete wound healing. Complete wound healing was considered as defined by the study authors and we included their definitions, when provided."
Paul Rockar APTA Methods
  • For analysis of this technology, relying totally on published, randomized controlled trials may prove to be flawed. Requiring a control group of "standard of care" protocols, especially if it is expected that the trial be continued until 2 visits past complete reepithelialization, would be ethically challenging for wound management specialists. Randomized studies likely don't exist because practitioner experience indicates that NPWT is effective and it is difficult for professionals bound by ethical standards to set up a controlled trial, especially in the home environment, to prove something they believe would be to the detriment of the control group. Methods may need to be revised to obtain data needed to quantify effectiveness of this technology.
We conducted our research because there was a legitimate doubt about the effectiveness and safety of NPWT use in the home. Of particular concern is the number of injuries and deaths associated with NPWT, particular in the home environment, as noted by the Food and Drug Administration (see http://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm244211.htm). As our research demonstrates, there is insufficient evidence make conclusions about NPWT in treating chronic wounds in the home environment. To ignore this, and to conclude based on "practitioner experience" that NPWT is effective, would be against the basic principles of evidence-based medicine. "Practitioner experience" is considered very low quality evidence.
Paul Rockar APTA Methods
  • It is not clear why surgical wounds were excluded from the assessment considering the large number of patients discharged from the acute care environment with negative pressure devices and who are followed in the home setting with follow-up appointments in outpatient facilities. If the care is occurring in the home environment the logic for exclusion is not clear.
The scope of our review, and specific key questions to be addressed, were based on the request from the nominator. The specific focus of interest was chronic wounds.
Paul Rockar APTA Methods
  • There appears to be inconsistency in use of the terms closed and closure versus healed. As exclusion criterion for Outcomes was "studies that reported wound healing rates without also reporting complete wound healing." Wound management takes up to two years to "heal" a wound. Otherwise the wound is described as "closed." Clarity is needed around these terms. And as noted just above, a more appropriate measure would be wound healing rate. It is unclear what definition was used for "complete wound healing." This is significant as articles that had "wound healing rate" were excluded if they did not also have "complete wound healing." If the terms "closure" and "healed" were not accurately applied to the process a number of articles may have been inappropriately excluded.
We added the following to details in Methods about outcomes of interest: "Generally, closure of the surface of the wound by the growth of epithelium over the defect is healing. If the closure has durability over time it is complete wound healing. Complete wound healing was considered as defined by the study authors and we included their definitions, when provided."
Paul Rockar APTA Methods
  • Rate of wound healing is recommended as a more appropriate outcome measure than "complete wound healing." It is unclear why rate of healing articles were excluded. As to rate of healing, negative pressure devices are indicated to speed the wound contraction process of full thickness wounds and reduce bioburden. When granulation tissue is brought up to a level flush with the surrounding epidermis, NPWT is typically discontinued and other forms of treatment involving synthetic dressings and moist wound healing are instituted. Physical therapist experts note that NPWT must be removed to allow for epithelialization. Extended use is not indicated and would be very cost inefficient. Therefore using "complete wound healing" as an outcome appears to be an inappropriate measure.
During the development of the review protocol, selection of outcomes, including whether to include wound healing rate, was discussed extensively amongst our review team and with the organization that nominated the topic. We have added rationale for excluding wound healing rate as an outcome to the Study Selection section of the Methods chapter. We have also included in the Limitations of the Review Process section of the Discussion chapter a paragraph regarding excluding studies that evaluated only wound healing rate.
Paul Rockar APTA Methods
  • The appropriate comparison for study should be NPWT versus other dressings used. Debridement is part of wound management with or without NPWT if necrotic tissue is present. All other interventions, including pressure redistribution, glycemic control, etc., should be controlled.
We agree that an appropriate comparison would be other dressings used. However, as we note in the Results section, many studies did not provide details about the interventions the patients in the comparison group received. We clarify in the Study Selection section of the Methods chapter what we allowed as a comparison group.
Paul Rockar APTA Methods
  • One source of information that could be examined, if companies would share their data, are records that are kept on patient outcomes. Physical therapist experts report that with at least one company, Kinetic Concepts, Inc., the physical therapists were required by the company to provide progress reports every two weeks document healing rates, complications and a variety of other factors.
Thank you for this suggestion. We did not contact manufacturers of NPWT devices for this Technology Assessment.
Paul Rockar APTA Discussion / Conclusions As noted above in our comments under Methods we concur that there is a great need for research with more standardization. This is especially true relative to the skills and training of those providing the wound management, including but not exclusively the application of NPWT. It is management every time you change a NPWT dressing, and how it is applied that can make the difference. This is probably the area of greatest variability and cause for concern as poorly trained individuals are being called upon to apply NPWT devices in significantly compromised patients. We agree with your comment. We attempted to make this point by discussing the variability in the people who are applying the devices, as this can impact the results of a study, or of any study examining any wound therapy. Further guidance is necessary to ensure that our patients have adequately trained individuals performing their wound care.
Paul Rockar APTA Discussion / Conclusions The conclusion in the executive study calls for robust randomized study designs. With the current study design from 5,332 citations, 7 met the criteria established. We believe this may indicate the need for a closer look at the criteria used and accurate and appropriate definition of terms used for this process as noted above. We also suggest that alternative methods of study design that considers inclusion of high quality studies and study designs without control group interventions that would be considered unacceptable patient care to wound management specialists. Thank you for your comments. We agree that there is much variability in the process of wound care which make designing a study difficult.
Paul Rockar APTA Appendix B-3: Arm Details – VAC is not a proper term. It is the brand name of the product of Kinetic Concepts, Inc. (KCI) and the product (Vacuum Assisted Closure) and is V.A.C.®. The general term NPWT should be used. Thank you for your feedback. We have edited the data abstraction form to reflect this.
Paul Rockar APTA Appendix Wound etiology – Should use neuropathy not diabetes. Throughout document, the term "neuropathic" should be used instead of "diabetic" unless the study specifically uses the term "diabetic." Thank you for this suggestion. We will keep the term "diabetic foot ulcers" because some diabetic wounds reflect vascular occlusion.
Ron Silverman KCI Results We appreciate the work involved in compiling this TA and know that AHRQ would like to have this document as a future reference for those interested. In that light, we would like to make some comments regarding the accuracy of Table 3, which identifies the current NPWT devices commercially available in the US to treat patients with chronic wounds. Thank you for your feedback on Table 3.
Ron Silverman KCI Results
  • Barker's vacuum packing is a method rather than a device. The Barker vacuum packing technique involves constructing a dressing out of drains, towels, and adhesive film available in the operating room and connecting that dressing to wall suction for negative pressure.
We removed Barker from Table 3.
Ron Silverman KCI Results
  • There may also be other products in Table 3 that are no longer available on the market. If requested, we can provide a more up-to-date search of current NPWT technologies. For example, Kalypto Medical (St. Petersburg, FL) has been acquired by Smith and Nephew. (NOTE: Text in red below indicates suggested additions/revisions.)
We have noted that Kalypto Medical has been acquired by Smith & Nephew.
Ron Silverman KCI Results
  • Although Table 3 lists the different NPWT technologies, the majority of studies (approximately 94%) in the literature that discuss NPWT for the treatment for both acute and chronic wounds have used systems manufactured by KCI (San Antonio, TX). 
The purpose of Table 3 is to list the devices currently available. Statements on the use or evaluations of the devices are beyond the scope.
Ron Silverman KCI Results
  • Minor changes are requested to the KCI entry in Table 3 to correctly state the trademarks.
Thank you for the list of edits. We have made the suggested corrections to Table 3.
Ron Silverman KCI Results
  • Because the MiniV.A.C.® Therapy unit is no longer commercially available, we suggest removing it from the list.
We have removed MiniV.A.C. Therapy from Table 3.
Ron Silverman KCI Results
  • Additionally, V.A.C.Via™ Therapy, ActiV.A.C.® Therapy, V.A.C. ATS® Therapy, V.A.C. Freedom® Therapy, and Info V.A.C.® Therapy have all been cleared and indicated for use in the home care setting. We do not know if that is the case for the other products.
Thank you for the list of edits. We have made the suggested corrections to Table 3.
Ron Silverman KCI Results
  • It is probably important to note there are two types of NPWT systems commercially available: disposable systems (which are mechanical and electrical) and those that are considered durable medical equipment (DME). For example, the 2012 Armstrong study2 compared the disposable NPWT system (SNaP®, Spiracur, Inc., Sunnyvale, CA ) to the standard DME NPWT system (V.A.C.® Therapy) in a multicenter, comparative efficacy, randomized controlled trial (RCT).
We have already discussed this in the Introduction chapter and again in the Results chapter in the Characteristics of Negative Pressure Wound Therapy Compared with Other Characteristics. We feel that this is enough emphasis on the differences between devices.
Ron Silverman KCI Methods Key Questions 2-4 are questions that concern clinical outcomes of various patient conditions, disease states and other therapies along with NPWT. These questions are vast in nature and would take a significant amount of time to respond. Since there is limited time to respond to this TA, we wish to provide a few comments that speak to the general nature of the methodology of clinical study selection, efficacy and safety. Thank you for reviewing our report.
Ron Silverman KCI Methods We believe the methodology used for selecting the clinical studies may have been too stringent. Patients are unique in their presentation with varying co-morbidities, complications, and types of complex wounds. When evaluating clinical studies for inclusion, the full article needed to be thoroughly reviewed. Excluding studies based only on review of the abstract may have excluded useful studies that would have added to the evidence for NPWT use. Therefore, conclusions may not reflect true outcomes of NPWT use due to literature (eg, as Keith et al,3) that was inadvertently excluded due to limitations in the inclusion/ exclusion criteria. It is standard procedure in a systematic review to first review abstracts for eligibility and then the full text articles. The Keith 2007 study was a meeting presentation, and would have been excluded.
Ron Silverman KCI Methods
  • Studies that did not have a comparison group were not included, which may have excluded useful studies for supporting evidence. If the aim of the report is only to report on efficacy and safety, then prospective/retrospective studies without a control group should be included to provide these results.
The purpose of the Technology Assessment is to compare the effectiveness and safety of NPWT with other wound care methods or with other NPWT devices. We clarified this in the first sentence of the Methods. We did cite the FDA's warning for serious complications with negative pressure wound therapy, which was based on case reports of safety concerns. However, it is difficult to interpret the cause of many adverse events without a comparison group.
Ron Silverman KCI Methods
  • Studies that reported wound healing rates without reporting complete wound healing were excluded; however, it is important to note that complete wound healing is not always an end goal of NPWT. Other endpoints, such as time to wound closure (via grafts, flaps, biologics, etc), rate and degree of granulation tissue formation, and limb preservation are also important factors to consider, especially when dealing with chronic wounds.
Thank you for the suggestions about outcomes. The outcomes were defined in our protocol based on discussion amongst our review team and with AHRQ and the nominating organization.
Ron Silverman KCI Results
  • Evidence was downgraded because complete healing was not defined according to the FDA's definition of healing. For observational trials and non-pivotal trials, this is often inappropriate, as most of these studies would not use such a definition in practice or pragmatic trials.
These studies evaluated and reported on complete healing, but they usually did not provide a definition for complete healing (Yao 2013; Armstrong 2012; Lerman 2010; and Ford 2002). Lavery 2007 defined wound closure, but used different definitions for wound closure for the NPWT and control groups. Neither definition was as stringent as the FDA definition.
Ron Silverman KCI Results
  • Some evidence was also downgraded when "inappropriate control groups" and "surrogate outcomes/endpoints" were utilized in studies. Although in some studies there can be some questions regarding these issues, downgrading the evidence on this basis was overly conservative.
Having an appropriate comparison group and valid outcome measures are basic principles of research and are often used in quality assessment. We did not make any changes to report based on this comment.
Ron Silverman KCI Results
  • The Baharestani et al4 article, which examined the early versus late initiation of NPWT for homecare patients with pressure ulcers, was not included. The authors performed a retrospective analysis of Outcome and Assessment Information Set (OASIS) data for home care patients to determine whether early versus late initiation of NPWT affected the overall length of home care services. For pressure ulcer patients, the median length of stay for the early group (n=65) was 85 days compared to 166 days for the late group (n=33) (p<0.0001). The main finding from this study was that for each day NPWT initiation was delayed, almost 1 day was added to the total length of stay.
We included studies that compared NPWT with other wound care methods. Length of stay was not an outcome criteria of this review.
Ron Silverman KCI Results
  • The stringent inclusion/exclusion criteria may introduce a selection bias into the analysis – especially if the criteria are applied inconsistently (see comparison of Yao and Blume articles below).
    • The large Blume RCT was excluded from consideration because it had a "mixed population of inpatient and outpatient subjects; however, Blume states in his article that 89.5% of his patients were treated in the homecare setting.
    • However, the Yao article, which also reports both inpatient and outpatient treatment was included.
    • To be consistent with the selection of Yao, the Blume RCT should be included in the sample.
The inclusion/exclusion criteria were developed in discussion with AHRQ and the nominating organization. Participants in the Blume article were clearly treated in the inpatient setting whereas it is unclear if NPWT treatment occurred during a participant's inpatient stay in the Yao article. We make statements in the results and discussion about our interpretation of setting for this and other studies.
Ron Silverman KCI Results The safe use of NPWT products is extremely important to physicians and clinicians. A recent exploratory and descriptive study by Kaufman-Rivi et al6 (2013) surveyed wound care specialists and home healthcare providers to gather information on their use with NPWT. The authors concluded that "Overall, respondents thought that there was a definite benefit to NPWT, regardless of the care setting, and that it was a safe therapy when prescribed and administered appropriately."
In 2007 Dr. Caroline Fife provided a report on the safety of NPWT in an outpatient care setting using the Intellicure™ electronic medical record system.7 A total of 9,119 patient records were available, involving 20,371 wound and 161,913 outpatient visits. There were 931 patients who received V.A.C.® Therapy (approximately 10% of total patients). None of the data was from hospitalized patients. Below are the conclusions from the report (which was included in our 2007 response to AHRQ):7

  • There was either no significant difference in complication rates between NPWT and non-NPWT patients or NPWT appeared to provide a protective effect, particularly in regard to surrogates for wound infection in larger wounds.
  • There were no deaths during NPWT use.
  • The report concluded that NPWT was safe in outpatient use for all wound types.

To be thorough on safety, in 2011 the FDA issued an UPDATE safety communication regarding serious complications associated with NPWT systems.8 This report included the following statement, "Overall, respondents felt that there is a definite benefit to NPWT therapy, regardless of care setting and that it is a safe therapy when prescribed and administered appropriately."8 

Thank you for your comment. Fife's findings that are relevant to our outcomes are documented in the report.

1. Names are alphabetized by last name. Those who did not disclose name are labeled "Anonymous Reviewer 1," "Anonymous Reviewer 2," etc.
2. Affiliation is labeled "NA" for those who did not disclose affiliation.
3. If listed, page number, line number, or section refers to the draft report.
4. If listed, page number, line number, or section refers to the final report.

Return to Contents

Page last reviewed September 2014
Page originally created September 2014
Internet Citation: Negative Pressure Wound Therapy Technologies For Chronic Wound Care in the Home Setting. Content last reviewed September 2014. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/findings/ta/comments/npwt-comments2.html