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.

Project ID: WNDT0913 

The Agency for Healthcare Research and Quality's (AHRQ) Technology Assessment (TA) Program supports and is committed to the transparency of its review process. Therefore, invited peer review comments and public review comments are publicly posted on the TA Program Web site at http://www.ahrq.gov/research/findings/ta/index.html.  

This document presents the peer review comments and public review comments sent in response to the draft report, Negative Pressure Wound Therapy Technologies For Chronic Wound Care in the Home Setting.  

 Contents

Select for Table 1: Invited Peer Reviewer Comments.
Select for Table 2: Public Review Comments.

 Table 1: Invited Peer Reviewer Comments

Reviewer1 Section2 Reviewer Comments Author Response3
1 Overall Due to the heterogeneity of the study populations and the paucity of qualified studies the report has little value for the clinical decision maker or systems of care in determining the clinical effectiveness of the NPWT. While the RTC's may not be robust enough to provide high quality evidence, nonclinical decision makers may view the report as a go/no go evaluation. Although we were not able to draw conclusions about the comparative effectiveness and safety of NPWT to treat chronic wounds in the home environment, we were able to provide an evaluation of the studies that addressed this topic and provide guidance for future research.
1 Overall The target populations are well defined in each study however the relevance would depend on how the cohort that are being treated in any single system or practice match up to those studied. Thank you for your comment. We agree with your point.
1 Overall I think the key questions are appropriately stated. Thank you for reviewing our report.
1 Introduction The introduction describes the intent of the project. Thank you for reviewing our report.
1 Methods I think the exclusion of all but the use in the home setting may be too focused and may be missed by casual readers who may interpret this more widely to other use locations. That said the intent is for home use and as such the criteria fits. We revised the title as well as the Methods chapter to make it clear that the focus was on home use. We also provided our rationale for including only studies conducted in the home setting under the Study Selection section of the Methods chapter.
1 Results Based on my last comment if should be made clearer that this applies to home use only and opinion is not being rendered on use in other clinical settings. We revised the title as well as the Methods chapter to make it clear that the focus was on home use. We also provided our rationale for including only studies conducted in the home setting under the Study Selection section of the Methods chapter.
1 Discussion / Conclusions Yes with the caveat above Under the Limitations of the Review Process section of the Discussion chapter, we have added, "NPWT is used to treat a variety of wounds, including acute and chronic, and is used in a variety of settings. However, our review focused specifically on the comparative effectiveness and safety of NPWT to treat chronic wounds in the home setting."
1 Clarity and Usability The report is very well structured and organized but the main point is not as clearly presented as it could be. More needs to be included that clearly articulates that this does not pass judgement on clincal settings other than the home setting. Under the Limitations of the Review Process section of the Discussion chapter, we have added, "NPWT is used to treat a variety of wounds, including acute and chronic, and is used in a variety of settings. However, our review focused specifically on the comparative effectiveness and safety of NPWT to treat chronic wounds in the home setting."
2 Overall The report is on a useful topic. Thank you for reviewing our report.
2 Introduction Satisfactory introduction Thank you for reviewing our report.
2 Methods Page 13, lines 20-22--In the Outcomes section you say you excluded studies that included healing rates without complete closure rates. There are efforts to use percent of closure over a period of time as an endpoint rather than complete closure in an effort to limit the duration of some of these studies. The report by Kirsner, et al in Wound, Repair, and Regeneration addresses that issue well. In a recent FDA decision, they report they are willing to accept percent healing over a period of time as opposed to complete closure in certain studies. 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.
2 Results Page 16, lines 8-9--I would suggest that a sample of 7 studies out of 5332 hardly makes your continuing your review worthwhile. Why not take all the data available and see what the results are? You might not consider the data "reliable" to your standards but it would probably reflect the "real world" experiences of practitioners. That would be interesting data then you could tell us why it is not reliable! We disagree with the reviewer. First, the more than 5,000 citations were retrieved by a broad and comprehensive search. To suggest that those citations were studies of NPWT is misleading. Second, systematic reviews often find limited evidence to address the review questions. In such cases, it is useful to find or confirm the lack of evidence, and especially helpful to explicitly outline gaps and needs for future research. We feel that we have done so. Finally, the data are not appropriate for meta-analysis. Quantitatively pooling heterogeneous studies of poor quality would provide misleading results.
2 Results Page 20, lines 15-34--I do agree with you entirely that the comparator group should not include therapies such as tissue engineered skin and the others you mentioned. Those are active therapies. The comparator group should be non-active, standard of care therapies. The only exception would be if you are trying to show improvement in infection rates and then the comparator should be an anti-infective or antimicrobial. Thank you for your comment. We have reported, where available, details about the comparison.
2 Results Page 20, lines 46-47--I am not sure how reliable the measurements using the plaster impressions are. They have not been useful in our hands. The digital comeras which use lasers to measure volume are more reliable. I would suggest someone make this method the standard for RCTs when volume is at issue. We did not assess reliability of measurements. The use of plaster impressions allowed the study being described, on page noted, (Ford, 2012) to mask the personnel completing the outcome assessments. Thank you for the suggestion about laser measurements of digital photos.  We have kept our recommendations about outcome measurement in future research, noting generally the need for objective measures, ideally made by masked personnel.
2 Results Page 27, lines 5-37 and Page 37, lines 26-29--I am not sure why you insist on knowing the prior treatment for the wound. If there is a 1-2 week (or longer) run-in period for the study, the effect of whatever therapy is no longer an issue. Besides, each patient will, most likely, have had a different treatment. Evaluation of the many previous treatments would probably be "inconclusive." Besides, it really doesn't matter what the previous treatment has been since it obviously has not worked! The studies are to measure the results of this technology or treatment. Thank you for your comment. An exploration of the previous wound care modalities was part of the key questions put forward by the nominator. The prior treatment is also one of the characteristics that is useful in understanding heterogeneity in the study findings.
2 Results Page 27, lines 17-19--I agree with you that the role of debridement in the treatment of any wound in any trial should be described. The extent and frequency of debridement should be noted for each visit. Debridement can have a significant influence on the outcome of any treatment (ie. the Steed study when using Regranex) so its use should be recorded. Subgroup analysis of debrided groups would be interesting in most every wound treatment study. We agree. Unfortunately, these studies were not specific in reporting the occurrence and/or the frequency of debridement.
2 Results Page 28, lines 54-55--Blinding participants and study personnel in wound care studies is imposssible!! You finally suggest that in your final comments (page 35, lines 38-42). These studies are not like giving pills which you can color the same! We agree that this is difficult. Clearly as you suggest, digital photography is approaching a quantitative method. One can use photographs read by a third party who is unaware of the treatment allocation as a way to achieve masking of personnel assessing outcomes. The latter needs to be considered. We have added a statement to the discussion: "Several organizations, such as the FDA, and the European Wound Management Association have noted the need for objective outcomes with blinding of outcome assessors, where possible."
2 Results Page 30, lines 23-33--I have never been a part of a study that did not report adverse events. All adverse events that occurred are reported, but adverse events that might have occurred but did not are not reported. If you feel a list of potential adverse events that did not happen should be reported, the list should be made available to all investigators when the protocols are being developed. Thank you for your suggestion. Several clinical areas are working toward core outcomes to include in studies, including specific adverse events. Wound care therapy research may also benefit from core outcome sets.
2 Discussion /  Conclusions Page 35, lines 16-18--Does it really matter who is applying the dressings, bandages, or treatments? It does not matter whether it is a physician, nurse, physical therapist, or whomever. It is the skill of the applicant that counts. Just because physicians may be doing the treatment, does not mean the patient is getting the best treatment. Most nurses do a better job with treatments and dressings than do physicians! The center doing the trial has to be responsible for the skill of the "treaters." Even if it were listed in the manuscript who was doing the treatments, it does not mean anything concerning qualilty and effective treatment. Thank you for your comment.  Reporting the personnel applying the intervention is a methodological issue.  Consistency in how the treatment is applied for all study participants is also an issue. This is of more concern in the home environment, when the patient or family member may be involved in applying the therapy being evaluated or compared.
2 Discussion /  Conclusions Page 37,line 22--This is the first time you have mentioned patient compliance! This is probably the most important factor in the outcomes of a particular treatment than most other things you are considering. If the patient does not allow the treatment to be used appropriately, there will be a "less than optimal" outcome no matter what the treatment! I congratulate you on finally mentioning this; however, I have no idea how it can be measured or quantified. Thank you for your comments. We agree that compliance is important. Perhaps a standard compliance measure(s) can be designed and included in all wound studies. This once again confirms the importance of wound care developing standard clinical investigative parameters.
2 Clarity and Usability You state that no conclusions can be obtained from this review. I would suggest to the group who did the review that they develop a clinical protocol for wound care treatments and products that would meet all the requirements of what they consider a "good study" and make it available to all investigators as protocols for new studies are being developed. Maybe it would help improve the quality of the research if all knew exactly what is needed to meet the basic requirements of an acceptable study. I would suggest you involve well-known wound treatment practitioners who do wound care research in your development process. Thank you. We agree. Our Evidence-based Practice Center has previously prepared a future research needs document for AHRQ and published this in Wound Repair and Remodeling. In this review, we include details to be considered in designing a study in the discussion, including in Table 11.
3 General The report does seem to lack some focus. Some of this is down to the limitations of the evidence but the authors could tighten up some sections and provide a more balanced review of the evidence. We provide a synthesis and evaluation of the strength of evidence. We are uncertain in what way this was unbalanced and have not made changes in response to this comment.
3 General The target population of the use of NPWT in the homecare setting was difficult to define and the authors had to stretch their definition at times. They included studies that reported as taking place in an "outpatient setting" if it was reported or the authors interpreted that it was not in the acute hospital or assisted living. A less loose interpretation would have resulted in one fewer included study. 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).
3 General The clinical usefulness of the report is again limited by the evidence that was found. However, this could have been strengthened by a clearer definition and explanation of the alternative therapies to NPWT. Unfortunately, many of the studies did not describe the interventions received by the comparison group, the interventions received prior to NPWT, or any concomitant interventions.
3 General The key questions were explicitly stated but surprisingly the question of cost-effectiveness was not included. In light of the widespread use of NPTW some inclusion of the economic case for it would have been useful. We developed the scope of the review in discussion with AHRQ and with the nominator for this topic. Cost was not an outcome of interest.
3 Introduction The introduction could have been improved by a clear description of the role of NPWT in the care of wounds. Why is it used? What are the treatment aims? etc. At the end of the fourth paragraph of the Introduction, we added, "The aim of NPWT is to facilitate wound healing, promote granulation of the wound bed, and provide a bridge to surgical closure." 
3 Introduction The authors also seem to pre-empt their findings with an unsupported statement on Page 1 line 33 about NPWT facilitating healing. A clear description of why NPTW is used and the aims of treatment would help the reader. We revised the statement in the fourth paragraph on page 1 to say, "NPWT refers to the application of negative pressure across a wound."
3 Introduction An additional paragraph on cost would also be helpful. How much do the devices cost? What are the likely cost to health care providers/ Medicare etc? Several factors influence cost and cost was not an outcome we evaluated.
3 Introduction The line on Page 2 line 30/31 about previous reviews suggesting possible benefit needs clarifying. We have revised this to be: "Due to a lack of studies, results from previous systematic reviews of NPWT have primarily illustrated limitations in the evidence."
3 Methods The methods are described clearly but the authors do need to add some detail and clarification. Thank you for your feedback. Please see our responses below about how we added detail to the Methods chapter.
3 Methods The scope of the searching was comprehensive if not extensive. The authors limited their search to the major databases and did not describe if they had searched the grey literature and why other databases e.g. SciCitation Index etc were not included. The authors also do not detail if any handsearching or reference checking were undertaken. We stated in the Search Strategy section under the Methods chapter that we searched 4 different electronic databases, handsearched the reference lists of included articles and reviews, and searched ClinicalTrials.gov. We did not make any edits to the report based on this comment.
3 Methods The authors stated that they were going to do a metanalysis but did not specify the type or how they assessed heterogeneity. Under the Data Synthesis section of the Methods chapter, we added that details about planned analysis are available in our protocol. Given that we were unable to conduct such analysis we do not feel it necessary to include those details here.
3 Methods Their classification of the strength of evidence seems to be subjective and would be strengthened by including any basis for these categories. We used, and presented, the strength of evidence categories and descriptions from the AHRQ EPC Methods Guide.
3 Results The searching identified a large volume of citations (5332) but only 7 were included. The flowchart detailing the literature search and exclusions was detailed but id raise some questions. They excluded studies with less than 20 participants - why and was this a priori? Please define "Does not apply" They excluded non-English citations but did not specify this in the inclusion/ exclusion in the Methods section. All eligibility criteria were defined a priori and specified in our protocol. We added text to describe the inclusion/exclusion criteria outlined in Table 1. The sample size criteria was developed in discussions with the nominating organization. We have added rationale to the methods section. We also re-reviewed all of the studies that were excluded during article review because of the sample size. All of the studies would have been excluded for other reasons as well (e.g., no comparison group, mix of patient populations, mix of settings, no original data). "Does not apply" means that it did not address any of the study questions.
3 Results I am not sure of the usefulness without context of the list of NPWT available in the US. How do they compare in cost, effectiveness, evidence?  We included the list as it addressed Key Question 1. Key questions were provided by the nominator.
3 Results The authors breakdown the reporting of the 7 included studies linked to their key questions. A chapter highlighting the limitations and risk of bias of the included studies prior to this would be helpful. Thank you for this suggestion. We provide descriptive text and figures for risk of bias. This was found on page 27 of the draft report.
3 Results On page 14 the studies designs are detailed and the citations described. However, the usefulness of ann RCT of 35 comparing NPWT to three different interventions is not commented on. This is the same on page 15 line 24 which details that it was unclear what NPWT was being compared to in four out of the 7 studies. We added sentences to the Interventions section and the Study Quality section to address the heterogeneity of interventions received in the Ford trial.
3 Results In the study quality no comment is made on the reporting and analysis of the citations. No comment or highlighting is made of the funding of the trial by Industry where appropriate. Industry sponsorship is not necessarily a quality issue. However, industry sponsorship does raise concerns about reporting bias, and we discuss this on page 20 of the report under the Strength of Evidence section.
3 Results No detail is given or comment on the appropriateness of the interventions being compared to NPTW. Lavery e t al compared NPTW to wet to dry dressings and reported a benefit to NPTW. However, most other interventions compared to wet to dry dressings would also likely show a benefit. In the Interventions section, we removed Lavery, 2007 from the statement that says, "Details about the therapy provided in the comparison group were not provided in four studies." Later in that paragraph, we added, "Lavery et al. used a pooled analysis of wet-to-moist dressings as a comparison group."
3 Results The description of Schwien et al on page 18 does not detail that the historical controls were 2,288 compared to 60 who had NPTW. The potential for data fishing was not highlighted or how and whether the matching was conducted or appropriate. This was also true of describing the matching in Lerman et al on page 19 line 25. Thank you for your comment. In the Study Quality section, we have added these details.
3 Results The authors do report in the final paragraph that five out of seven studies were funded from industry and that one did not report the source of funding but this is not picked up on or highlighted further in the report. Under Limitations of the Evidence Base in the Discussion section, we discuss the issues with industry-sponsored trials.
3 Discussion /  Conclusions The discussion does seem to be rather short and not very discursive of the issues highlighted in the results. We added paragraphs to the Limitations of the Review Process and the Limitations of the Evidence Base sections of the Discussion chapter to discuss some of the issues that were mentioned in the Results chapter.
3 Discussion /  Conclusions One key finding that is not commented upon is the shear volume of the excluded studies in relation to those that were found to fulfil the inclusion criteria. This sort of inclusion rate is typical in systematic reviews with comprehensive searches and we do not feel is a key finding.
3 Discussion /  Conclusions A summary of the points in relation to the key questions is given in bullet form but not expanded upon or critically analysed. We added paragraphs to the Limitations of the Review Process and the Limitations of the Evidence Base sections of the Discussion chapter to discuss some of the issues that were mentioned in the Results chapter.
3 Discussion /  Conclusions A section on the use of NPWT in other settings is reported but tended to be descriptive rather than analytical. I am not clear as to what Table 9 adds to the clarity of the findings. We excluded many studies because the setting was unclear, or because the setting was a mix of inpatient and outpatient patients. We included Table 9 to discuss how the results of studies from other settings may or may not support our research findings.
3 Discussion /  Conclusions The section on future research is clear and should be translated into key research projects. Thank you for reviewing our report.
3 Clarity and Usability Overall the report was well structured if a trifle repetitive and descriptive. The conclusion was very short only consisting of 6 lines. We are unfortunately limited to qualitative synthesis due to the nature of the evidence. We do not feel that further text is needed in the conclusions.
4 General The conclusions seem at first blush disappointing because the authors describe an abysmal state of the art in one of the few supposed successes in the wound care field. But after discarding the vast majority of papers published (leaving only 7/5332) because most did not explicitly state they were dealing with exclusively outpatient NPWT management, the authors are setting an artificially elevated bar. No one hospitalizes a chronic wound patient until the full course of NPWT is concluded (except for Argenta's original description of the technology)- most patients are started on NPWT as an inpatient and are transitioned to the outpatient setting. Thus the review might well have discarded well-done studies for this methodological standard which is not an accepted standard. The key questions are phrased articulately, and the authors have done an admirable job of defining the terms, the available technology, and the analysis of the 7 articles is superb. Thank you for your feedback. The key questions were provided by the nominating organization. Note that our search was broad and comprehensive so to suggest that all of the search results were studies of NPWT is not correct. Also, we allowed for patients to have started NPWT in the hospital setting, so long as all patients were discharged and followed in the home setting.
4 Introduction No issues except the overly brief explanation of possible mechanisms of action. Omitted, for example, are the elevated levels of angiogenic and stimulatory cytokines seen in NPWT wound fluids. We have added some details about mechanisms of NPWT. However, a full analysis of the literature about the mechanisms was beyond the scope of our review.
4 Methods Page 8 Line 10, setting out the Clinical Outcomes - there is a reason why surrogates are now accepted for efficacy even by the FDA other than complete wound healing (complete re-epithelialization without evidence of wound drainage for two days on visits two weeks apart). In my estimation, succeeding in partial wound closure should either be a Clinical Outcome or a Patient-centered Outcome, since it should be straightforward to provide evidence that smaller wounds are easier to care for and have fewer sequelae/better QOL. 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.
4 Results The analysis is clear and forthright, and the figures are easy to understand. Again, my fear is that they excluded studies that should have been included because the abstract did not identify the study as entirely outpatient-based. The only way an article was excluded during the abstract review was if the setting was clearly defined in the abstract as not being an outpatient setting. If the setting was not clearly defined in the abstract, the reviewers included the article for further consideration at full-text screening.
4 Discussion /  Conclusions The PICOS framework in Figure 10 is clear, concise, and should be posted by every clinical research coordinator and potential PI wanting to do research on this subject! Thank you for reviewing our report.
4 Clarity and Usability The study is too limited to be of utility in informing policy or practice decisions.Most of us who use NPWT do so because of first-hand empirical feedback of efficacy, and unfortunately, this report neither confirms nor denies that gestalt. We agree that it is unfortunate that the evidence is insufficient to allow for conclusions about the efficacy and safety of NPWT for chronic wound care in the home setting.
5 General The report has significant clinical research questions. The four main parameters of this report: 1) the target population (elderly), 2) the type of health problem (chronic wounds), 3) the environment of care (home), and 4) the treatment technology (NPWT) are all growing rapidly, and they are all critical clinical issues to be included in the research studies. Thank you for reviewing our report.
5 General There is lack of sufficient evidence for "home setting". Selection of studies that were conducted in the "outpatient" setting means that the reviewers moved a bit away from the scope of the review. 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).
5 General Two problems with the target population:
1) The target population is referred to as patients of 65 years and over, similar to Medicare patients. Definition of Medicare patients also includes "disabled patients", which means patients younger than 65 years old may have been involved in the studies. One of the studies specifically mentions that they excluded patients under 18, indicating the possibility of inclusion of the patients between 18 and 65 years old.
2) Elderlies, being the target population is not mentioned anywhere in the purpose/objective of the review, while in other sections the target population seems to be an important criteria for selection of the studies
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.
5 KQ1 Key question 1
1) It refers to the existing NPWT "Technologies" commercially marketed in US. This gives the impression that the study is considering different types of technologies used in NPWT systems while Table 3 is just a list of brands and manufacturer’s names taken from FDA website. Recommendation: Maybe the word "Technologies" should be replaced by "Systems" or "Devices" to avoid any confusion.
Thank you for your comment. The key questions were provided by the nominator. The key questions were further refined in developing our protocol.  We appreciate the feedback is meant to clarify the questions but we are not comfortable changing the questions at this stage.
5 KQ1 Having pointed that out, the main concern for key question 1 is why this is considered a key question of the review, to begin with? The list can be presented as inclusion of NPWT devices that are approved by FDA in US. It only needs a brief description of how this list was made and why the other NPWT devices are excluded. Thank you for this suggestion. Key Question 1 was addressed as it was requested by the nominator.
5 KQ2a Key question 2 Part A states: "Does the home use of the NPWT significantly improve –Any- of the following outcomes as compared to…." The word "any" can be interpreted in different ways: Is the question looking for improving of one of the 3 categories and its subsets or one of subsets in each of 3 categories? I.e. the answer is qualified as valid if it is "yes" to all subsets of the "Clinical outcome" and none of the subsets of the "Patientcentered outcome" and "Adverse events", or at least one subset of each category should have positive response? In other words; is the question seeking a YES, for instance, to improving the "Time to surgical readiness of the wound" from the first category, a YES to improving of "pain" from the second category and a YES to less "blood transfusion" from the 3rd category? Thank you for your comment. The key questions were provided by the nominator. The key questions were further refined in developing our protocol.  We appreciate the feedback is meant to clarify the questions but we are not comfortable changing the questions at this stage.
5 KQ2a Additionally, this section needs editorial review; not all the phrases are appropriate to follow the word "improve", the questions need to be worded differently so each one has a clear meaning, standing alone. I.e. "lowers the incidence of bleeding and use of blood transfusion." And, "reduces the number of extremity amputations". Thank you for your comment. The key questions were provided by the nominator. The key questions were further refined in developing our protocol.  We appreciate the feedback is meant to clarify the questions but we are not comfortable changing the questions at this stage.
5 General General comment on the key questions 2-4:
Too many key questions; the high number of questions can be problematic in itself when it comes to finding the answers for all of them.
Thank you for this suggestion. The Key Questions were suggested by and refined with input from the nominator. Having multiple questions did not limit our ability to answer the key questions.
5 Introduction Introduction presents a clear description of only two of the main parameters of the review; "chronic wounds", and "NPWT system", but is missing further description of the "target population" and "home care setting" and why this review has questions about them. We have added the following sentence to the end of the Introduction, "Given the variability in care that may occur in the home environment, studies for this particular population, especially in regards to safety, are of interest."
5 Introduction The justification for the review is clear, understandable and important, but a brief explanation of the approach to be taken is missing in the introduction. Under the Scope and Key Questions section, we had already stated that this was a systematic review.
5 Methods Protocol development, and study selection are presented clearly. Search criteria are explicitly stated but do not seem to be logical for the purpose of this review. They are too broad and most likely that is the reason why too many unrelated citations were generated for the search. Because there is no standard terminology for chronic wounds, we purposefully kept the search strategy broad so we wouldn't miss any relevant studies.
5 Methods In Table 1: "Population and condition of interest" indicates inclusion of "studies with any population regardless of age or disability." This seems to be inconsistent with the report’s target population specified as 65 years and older. 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.
5 Methods The studies’ applicability was assessed in terms of "the degree to which the study population, interventions and outcomes are "typical" for the treatment of…." What is considered "typical"? , It needs to be more specific. Thank you for your comment. We have revised that sentence to: "We assessed the applicability of studies in terms of the degree to which the study population (age, duration of ulcer, comorbidities), interventions (treatment, cointerventions, duration of treatment), and outcomes may be applicable to the treatment of individuals with chronic wounds who are treated with NPWT in the home setting."
5 Methods The reason why the two studies by Armstrong DG, Marston were included in this review is missing. Comparison of two different NPWT devices seems to be unrelated to the questions of this review. The Armstrong study applies to Key Question 3, which evaluates differences in NPWT devices. We have added this comparison to the general objective statement.
5 Results Flow of the information is slightly confusing in the result section. The main topics and subtopics should be organized in a way that the reader can easily move from one main topic to another. In the Data Synthesis section of the Methods chapter, we added how we organized the results chapter. We now state, "Because the different chronic wound populations are not homogenous, we analyzed the results for each wound etiology separately when possible."
5 Results Tables, figures, and appendices are adequate and descriptive. The only concern is the length of the appendix D and its tables. The way tables appear on multiple pages is not easy to read and follow. We have revised Appendix D so the tables are easier to read. We reformatted the tables by changing the column widths and eliminating unnecessary columns.
5 Discussion /  Conclusions Table 10 "Characteristics of the studies" demonstrates too many limitations in each selected study, raising the question whether they were chosen appropriately. How we can expect to find answers to all the questions this review raises by examining these studies? Each of the studies listed in Table 10 met our inclusion criteria for the systematic review. We did not exclude studies based on quality. The finding from our review was that the evidence was insufficient to draw any conclusions. One benefit of reviews that find very limited evidence is to point out the weaknesses in the evidence, and to suggest possible areas for future research.
5 Discussion /  Conclusions The "Research gaps" section refers to the limitations clearly and provides useful guidance to narrow down these gaps for future studies. Thank you for reviewing our report.
5 Discussion /  Conclusions Discussion and conclusion sections are in depth and are described adequately. This section clearly identifies inefficiencies and limitations of the selected studies and the reasons why they do not provide answers to the questions of this report. Thank you for reviewing our report.
5 Discussion /  Conclusions The report provides useful guidance for future research projects. Thank you for reviewing our report.
5 Clarity and Usability The report’s structure and organization needs a lot of improvement. The report chapters follow standard TA format. We have clarified at beginning how results are presented.
5 Clarity and Usability Conclusions can be used in formation of policy and/or practice in research but not in the clinical settings. Conclusion is too short. We agree that there are no conclusions that can be used in clinical setting. The evidence is insufficient to draw conclusions about the use NPWT for chronic wound care in the home setting. We do not feel the need to add any further text to the conclusion.
5 Clarity and Usability Using the PICOS framework, any researcher should be able to easily set appropriate questions and parameters for their studies. However, since the majority of researchers don’t bother to start with a framework, Table 11 in page 33 is the most important and useful piece of information that is provided by this review. Thank you for reviewing our report.
6 General The introduction should be more rigorous, concrete, and up to date on the scope and causation of chronic wounds. While we were considering in our review several different types of chronic wounds, each with their own etiology, we did not conduct a review on the causes of chronic wounds. However, we did add in the second paragraph of the Introduction that cardiovascular disease, diabetes, and obesity, in addition to an aging population, are drivers for an increasing prevalence of chronic wounds.
6 Introduction There are general statements made that should be fortified with concrete data and figures. For example, in the 3rd sentence, first paragaph: "with billion of dollars spent annually on wound care." -- the magnitude of billions deserves specificity: how many billions, and when was this estimated, and cite or reference by whom. Fuzzy statement at the outset can call into question the rigor of the remainder of the manuscript. We edited this sentence to state that expenditures were estimated to be $25 billion dollar and cited Chandan 2009.
6 Introduction In the 3d paragraph, it is appropriate to describe chronic ulcers as common in the elderly. However, other major drivers are cardiovascular disease, diabetes, obesity, all of which are significant risk facts that are increasing in incidence. Some description of the burden of chronic wounds in these skyrocketing populations must be mentioned. Otherwise, the discussion seems focused soley on pressure ulcers. In fact, pressure ulcers are where there is the least amount of scientifically rigorous data on wound care interventions. More is available in venous and diabetic ulcers. This needs to be addressed before acceptance. We combined the second and third paragraphs so that there is less emphasis on pressure ulcers. In addition to the aging population, we also mention the growing prevalence of cardiovascular disease, diabetes, and obesity as drivers for the increasing prevalence of chronic wounds.
6 Introduction The 4th paragraph describes wound management, but jumps rapidly to technologies and products identified as adjunctive. Yes, one of the fundamental clinical issues leading to wound chronicity is lack of proper diagnosis, and lack of basic good standard wound care practice by poorly trained physicians. If these (compression, off loading, moist wound environment, wound cleansing and removal of necrotic and senescent tissue) are properly managed, many wounds would not become "chronic" --they may still heal slowly, but not become refractory to healing. The jump to technology is a common knee jerk response in the wound care field, and this manuscript for this journal should first discuss the need for, and lack of, adequate basic wound management as one of the causative factors in today's clinical reality of wound management. Although we agree that there are many other fundamental clinical issues which lead to wound chronicity, we do not feel the need for greater detail in providing the introduction for the review.  We discuss factors which are important in wound healing and in conducting wound research in the discussion.
6 Introduction Line 34: again the casual use of "multi-million dollar industry" should be avoided. Cite the exact economic or market size figure. In fact, NPWT market size exceeds 1 billion worldwide. The authors should state the actual market value if making such statements. We edited this sentence to state that expenditures were estimated to be $25 billion dollar and cited Chandan 2009.
6 Introduction Line 42: the paragraph starting here is too dismissive of the many scientific studies that have been done to characterize the multiple mechanisms of action of NPWT. These include: decreasing proteolytic enzyme activity that is destructive to newly generated tissue; activation of cell signaling that leads to growth factor expression; induction of angiogenesis to promote granulation; and stimulation of survival signaling pathways. Stating "the exact mechanism is not known" it is misleading because it implies there is a single 'exact' mechanism. There is not. It is a combination of factors that contributes to the promotion of wound healing. A search of the literature will easily yield reviews of these mechanisms. We have revised this paragraph at the bottom of page 1 to list multiple mechanisms of action of NPWT.
6 Introduction Paragraph beginning line 47: state what are the common causes of "death" as it is not an obvious risk for readers not familiar with this technology. We have edited this sentence to read, "death from infection or bleeding."
6 Methods The methodology is adequate, although it should be noted that despite sharing chronicity as a unifying feature, the wound types included -- diabetic, venous, pressure, arterial -- are not homogeneous populations in of themselves, and further that each wound type is distinct at the molecular and in the clinical sense from the other types. Hence, there is an intrinsic flaw in grouping analysis of these disparate wound types together. It would be like analyzing 4 different types of cancer together for the effect of a new cancer treatment. This is a common criticism and weakness in wound care research. We agree. We added a statement to the Data Synthesis section of the Methods chapter to make this more explicit. We now state, "Because the different chronic wound populations are not homogenous, we analyzed the results for each wound etiology separately when possible."
6 Methods The other implicit limitation of the study, is that wound clinical studies are notorious for their difficulty in appropriate enrollment, and many studies are known to be flawed by contamination with inappropriate or misdiagnosed patients, or lack of good standard wound care. Although we agree with this reviewer, we do not have enough information about the study populations to state if they were misdiagnosed or not.
6 Results The results support the obervation that there is much to be desired in clinical investigation in the wound care field. The design of properly controlled, uniform, and adequately powered studies, with a clear and consistent set of outcomes measures -- are woefully lacking. Thank you for reviewing our report.
6 Results The lack of any, or in some cases, consistent and well articulated analysis of patient centered outcomes is a problem for interpreating the broad clinical utility of NPWT. We agree that patient-centered outcomes are important in evaluating the utility of NPWT. Our analysis and interpretation of patient-centered outcomes is limited to what is reported in the literature. Unfortunately, these outcomes were often not reported. When they were reported, studies used either a surrogate marker (Fife 2008) or a non-validated exit interview (Armstrong 2011). We added these limitations to the Limitations of the Evidence Base section of the Discussion chapter.
6 Results Adverse events need to be clarified in the analysis for whether they are likely to be due to the NPWT intervention itself, or as a consequence of the level of acuity or level of patient (or clinician) compliance with good standard wound care practices. It sould be clarified is "some studies is not report adverse events" means this was looked at but no adverse events occurred; or that this was a missing part / deficiency of the study. For the outcomes of emergency room visits and unplanned hospitalizations/surgeries, we only reported these if the study stated that these were related to the NPWT or treated wound. We reported on all other relevant adverse events that were reported in the included studies. We clearly stated that the study did not report on adverse events.
6 Results The call out of a single study with venous leg ulcers, and 2 studies with mixed ulcer populations is notable. What about diabetic or pressure or arterial? What happened to those studies? This should be described why they were not specifically analyzed for the Results. These were mentioned in the introduction. Again, its critical for a manuscript in this journal to accurately represent the breadth of chronic wounds in which NPWT is used, and with adequate specificity. We report on all the studies that met our eligibility criteria. Table 4 lists the number of studies that compared NPWT with other wound treatments in each of the patient populations.
6 Discussion/ Conclusions The discussion describes the flaws and weaknesses in the evidence base with NPWT. However, it does not, and should, point out and discuss areas of evidence for clinical benefit of NPWT, even in individual studies with other weaknesses. Without discussion discussion, the conclusion of this report is a scathing review of the inadequacies of NPWT clinical evidence. This does not help the reader-stakeholder hoping to clean insights for policymaking. Its easy to criticize weakness, but what are the positive findings within the evidence reviewed? It is from those clues, that further research initiated. Unfortunately, this review does not take this opportunity, so overall, it is a "we tried to look at the evidence, but it wasn't there". Unfortunately, the available evidence is insufficient and does not allow us to draw any conclusions on the effectiveness or safety of NPWT to treat chronic wounds in the home setting.
6 Clarity and Usability The report does not yield true assessment of NPWT does a technology through systematic review. Rather, it revealed and highlights, but does not full discuss, the inadequacies of study design and clinical investigation in the wound care field as a whole, and involving NPWT in particular. Therefore, the report is not that useful for evaluating NPWT itself. This does not help inform policy making or practice concerning NPWT. We agree that it is unfortunate that the evidence is insufficient to allow for conclusions about the efficacy and safety of NPWT for chronic wound care in the home setting.
7 General I realize that CMS may have dictated the definition of "chronic wounds," but there are chronic wounds that are surgical or traumatic in origin that could be treated with NPWT. Thank you for your comment. The scope of the review, including definition of chronic wound, was developed in discussion with AHRQ and the nominating organization. Surgical and traumatic wounds were excluded based on those discussions.
7 General The population is defined, and the topic is clinically meaningful. Thank you for reviewing our report.
7 General Page 13, Line 31 – correct terminology would be that Smith & Nephew acquired Kalypto Medical (not that Kalypto became S&N) Thank you. We have made this correction to Table 3.
7 General Line 45: company name is Medela, not Mendela Thank you. We have made this correction to Table 3.
7 General Page 32, Line 21/22: Charlson Index [Mary Charlson developed it.] Thank you. We have made this correction to the Research Gaps section.
7 General Page 33, Line 7/8 – "Charlson Index" not "Charleson Thank you. We have made this correction.
7 Introduction Background information is appropriate. Thank you for reviewing our report.
7 Methods Perhaps this was defined by CMS, but I don‘t understand why the review was limited to chronic wounds (since treatment of wounds of shorter duration may be started in acute care and treatment continued in the home care setting), and why traumatic or surgical wounds (which can become chronic) would have been excluded. Thank you for your comment. The scope of the review, including definition of chronic wound, was developed in discussion with AHRQ and the nominating organization. Surgical and traumatic wounds were excluded based on those discussions.
7 Methods I also wondered why noncomparative studies were excluded from consideration for harms. Some harms (such as retention of foreign bodies from the NPWT dressings) are pretty specific to NPWT, and could be found in noncomparative studies (limited to home care setting). I had searched the MAUDE database for harms specific to home care and found a number that could be useful for your Discussion chapter. Thank you for your suggestion. The nominating organization requested a specific limit to studies with comparison groups. We have added a statement about this limit, for assessing adverse events, under Limitations of Review.
7 Results The level of detail in the text and appendices is appropriate. Thank you for reviewing our report.
7 Results I think this excluded study should have been included because treatment primarily occurred in the home care setting (89.5% of treatment days in the NPWT group and 95.3% of treatment days in the advanced moist wound care therapy group): Blume PA, Walters J, Payne W, et al. Comparison of negative pressure wound therapy using vacuum-assisted closure with advanced moist wound therapy in the treatment of diabetic foot ulcers: a multicenter randomized controlled trial. Diabetes Care. 2008 Apr;31(4):631-6. PMID: 18162494. Excludes as "Not conducted in home setting or outpatient setting." We included studies that followed all patients in the home setting. For the Blume article, we interpreted this to be a majority of patients were followed in the home setting and some patients followed in the acute care setting.  We include Blume in the Discussion where we note key articles that did not meet our review eligibility criteria.
7 Results Studies to consider that aren’t in the excluded studies list (perhaps they were excluded at the abstract level):
1. Hiskett, G. Clinical and economic consequences of discharge from hospital with on-going TNP therapy: A pilot study. J Tissue Viability 2010;19(1):16-21. Abstract: Despite practical advantages (such as exudate management and reduced dressing changes) the adoption of topical negative pressure therapy (TNP) in home care may be restricted through logistical issues and a perception that the intervention is more expensive than the more traditionally utilised wound management products. This pilot study followed the experiences of 20 subjects with a variety of acute and chronic wounds who received TNP either in hospital (n=10), at home (n=5) or in both care settings (n=5). All except one subject showed both reductions in wound surface area and improved appearance of the wound bed during the course of treatment ranging from 2 to 74 days.
2. Begum, SS, Papagiannopoulos, K. The Use of Vacuum-Assisted Wound Closure Therapy in Thoracic Operations. Ann Thorac Surg 2012. Abstract: . . . . The patients were discharged home with a portable VAC therapy system in situ. [if you expand scope of wound types]
3. Trueman P, Flack S, Loonstra A, Hauser T. The feasibility of using V.A.C. Therapy in home care patients with surgical and traumatic wounds in the Netherlands. Int Wound J 2008 Jun;5(2):225-31. PMID:18494628 [if you expand wound types]
4. Trueman P. Cost-effectiveness considerations for home health V.A.C. Therapy in the United States of America and its potential international application. Int Wound J 2008 Jun;5 Suppl 2:23-6. PMID:18577135
Thank you for these suggested articles. The Hiskett et al 2010 study was included in our search, but excluded at abstract review level because it did not have a comparison group. The Begum et al 2012 study was excluded at abstract review level because it did not evaluate patients with chronic wounds; it evaluated post-operative patients. The Trueman et al 2008 article was excluded at abstract review because it did not evaluate patients with chronic wounds; it evaluated patients with surgical and traumatic wounds. The Trueman 2008 article was excluded because it was a review article.
7 Discussion/ Conclusions I would suggest additional discussion of the potential for harms in the home care setting. We added a statement to the Limitations of Review section that other study designs may have provided further details about adverse events.
7 Clarity and Usability The major conclusions are not unlike those for most wound care studies, unfortunately. We agree. Thank you for your comment.
7 Clarity and Usability It may be helpful to CMS if you cite the work of others (who come from an industry perspective) in the Research Gaps section to bolster your statements. This is text we included in our CMS report on Skin Substitutes that you may find helpful:
The FDA guidance document on chronic wound suggests that "blinding of subjects and investigators to the assigned treatment reduces bias and should be employed when feasible."10 Other organizations interested in improving the quality of evidence in wound management have also suggested that blinding be a part of any study of chronic wounds. POWER considers blinding of "patients, clinical assessment, and analysts where possible" as part of a minimum set of criteria for RCTs.108 The European Wound Management Association (EWMA) believes that "outcome assessors should be blinded to interventions whenever possible."106 The Center for Medical Technology Policy (CMTP) also recommends blinding the individual assessing wound healing outcomes.113 Blinding of patients to wound care treatment is not always possible because of visible differences in the treatment devices, dressings, or wound care routine. …. FDA has suggested that "in these situations, blinded assessment by a third-party evaluator should be considered."10 Blinding of the individuals recording data on complete wound healing, wound size, and other outcomes should be possible and greatly minimizes the potential that results will be affected by the evaluator’s expectations.106  … As mentioned earlier, FDA considers complete wound closure to be "one of the most objective" wound healing endpoints.10  Objective as opposed to subjective endpoints provide less opportunity for bias and "in general, blinding becomes less important to reduce observer bias as the outcomes become less subjective."114
Thank you for this suggestion. We have added a statement to the discussion: "Several organizations, such as the FDA, and the European Wound Management Association have noted the need for objective outcomes with blinding of outcome assessors, where possible."
8 General The goal of this technology assessment was to provide a systematic review of available published literature data on the use of Negative Pressure Wound Therapy (NPWT) systems for chronic wound care in the home environment. The authors were guided by several key questions that were developed by the Centers for Medicaid and Medicare (CMS). These questions probed safety and effectiveness outcomes for patients who are similar to Medicare patients with chronic wounds under care in a home environment. Other questions were related to comparative effectiveness of NPWT versus other methods of wound care as well as predictors of successful wound healing as it relates to NPWT settings or patient characteristics. Thank you for reviewing our report.
8 General Although 8,263 publications were originally identified, the authors identified only 8 publications (7 studies) that met the pre-determined inclusion and exclusion criteria. Given that there were only 7 studies from which to draw information, the reporting of these data were limited to summary reporting. No further analyses of these data were conducted. Since the report essentially is providing a summary of the data reported from these 8 publications, there is no further comment to be made on the reported information. Thank you for reviewing our report.
8 General The authors were presented with a difficult task as available literature is not abundant in high quality clinical trials data that explore the key questions posed by CMS. This is reflected in the final conclusion made by the study authors, specifically that "existing evidence is insufficient to draw any conclusions." Thank you for reviewing our report.
8 General The authors were presented with a difficult task as available literature is not abundant in high quality clinical trials data that explore the key questions posed by CMS. This is reflected in the final conclusion made by the study authors, specifically that "existing evidence is insufficient to draw any conclusions." Thank you for reviewing our report.
8 General The authors may have been further limited by the approach to address all key questions together. For example, it may have been less limiting to develop unique sets of inclusion and exclusion criteria that apply to a specific key question. Literature review to answer key question 3 (characteristics of NPWT administration) independent of use environment (home or hospital) may have produced additional literature articles that may have reported relevant data that were otherwise excluded because the article did not specifically discuss home use. This is because some of the NPWT system characteristics that correlate to successful clinical outcome may be independent of use environment. We agree that the characteristics of the NPWT system that correlate with outcome(s) may be unrelated to the setting of care. However, the key questions from the nominating organization were specific to the home setting.
8 General Key question 2 part D was related to comparative assessment of NPWT and other standard wound care modalities. The inclusion and exclusion applied by the authors were focused solely on literature presenting data on NPWT usage. However, additional useful information may have been gained if the literature review was expanded to include literature reporting on non-NPWT wound therapies used on patient population and wound types that match literature on NPWT use. The focus of this review was NPWT and so we did not seek to extrapolate from the literature on non-NPWT therapies. Key question 2, D was seeking a subgroup analysis for key question 2 based on treatments received prior to and/or concurrently with NPWT. Thus evidence about non-NPWT would not be helpful in addressing this question.
8 General Although the focus of the review was presented as being on patients who are representative of the Medicare population, the report does not specifically discuss the relevance of the 7 studies to this population. Also, patients greater than 65 years old and disabled patients potentially represent two distinct populations with distinct co-morbidities. Likely study combining these two populations would be difficult due to the distinct patient population characteristics. 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.
8 General Overall, the report was able to identify 8 publications that provide data that address the key questions and conclude that there is lack of available information to provide definitive answers to the key questions posed by CMS. However, this study may have attempted to answer too many questions all at once. It would be extremely difficult to find publication of sufficiently high quality data that can answer questions relating to Medicare population (>65 years of age or disabled), use in home environment, use on chronic wounds, comparative analysis against other modes of wound therapy, correlative information on NPWT settings and clinical outcomes, and correlative information on patient characteristics and clinical outcomes. If the key questions that are independent of others were assessed individually, the authors may be able to assess data from publications that were otherwise excluded. Thank you for your comments about the key questions. These questions were posed by the nominating organization and further refined in our protocol. The elements of the research questions (population, intervention, comparison, outcomes, setting, study design) cannot be addressed as separate elements but are used to fully define the questions that were addressed in our review.

1. Peer reviewers are not listed in alphabetical order.
2. If listed, page number, line number, or section refers to the draft report.
3. If listed, page number, line number, or section refers to the final report.

Return to Contents
Proceed to Table 2: Public Review Comments

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