The National Pressure Ulcer Advisory Panel (NPUAP) would like to commend AHRQ for the enormous undertaking of synthesizing available information regarding the treatment of pressure ulcers (PU). In particular, the sources for information were broader than are usually undertaken including soliciting unpublished information from stakeholders. The population of patients with pressure ulcers is very diverse; unlike renal failure or congestive heart failure, there are no reliable lab tests for skin failure and the outcome of a treatment for any individual is also dependent on multiple co-morbidities. The difficulty of this technology assessment is compounded by the fact that healing is a continuum and no one modality is likely to be appropriate from initiation of treatment to wound closure.
This review provides an excellent platform to call for the implementation of certain basic approaches [eg methodology for wound measurement, clinical endpoints, use of consistent staging terminology and perhaps partitioning healing research to partial thickness (Stages I & II) and full thickness (Stages III & IV)] which would allow more accurate comparisons and meta-analyses in the future.
NPUAP recognizes that this document is still in draft format-some of the comments are likely editing-related. There is a concern, however, that the results in the Structured Abstract/Executive Summary will be highlighted without the significant caveats that are presented regarding strength of evidence. For example, if there is only low strength evidence that wound healing is similar with collagen compared to standard care dressings, AHRQ’s own definition suggests that there is low confidence that this evidence reflects the true effect and that further research is likely to change the confidence in the estimate of the effect. A conclusion of further research is warranted’ for the areas with low confidence might convey the message more accurately.
Other selected areas of concern are:
- The 2007 NPUAP staging system is a six stage system which includes Unstageable and Suspected Deep Tissue Injury (Executive Summary and Introduction). Research on PU healing is typically limited to Stages I-IV but the NPUAP-EPUAP Clinical Practice Guidelines, 2009 reaffirmed the six stage system.
- Use of the term vacuum-assisted closure (ES-2) instead of Negative Pressure Wound Therapy (NPWT) is confusing.
- ES-5 Population and Interventions sections seem to imply that direct research was undertaken rather than that these were selection criteria for the documents that were reviewed.
- ES-13 Table A Key Outcome is listed as Alternating pressure beds but the conclusion relates to alternating pressure chair cushions.
- ES-19 and the Key Points (p. 78) related to surgery for pressure ulcers cite independent variables-for example ‘Sacral ulcers have lower recurrence rates after surgery than ischial pressure ulcers.’ The location of the ulcer, whether or not the patient has a spinal cord injury, and whether or not bone debridement is done during surgery are defining characteristics of the patient’s wound.
- ES-22 cites the review by Reddy (2008) as being the most current, comprehensive evidence about the effectiveness of pressure ulcer treatments. This analysis and recommendations had a concerning degree of selection bias created by limiting reviewed studies to randomized, controlled trials. Not all treatments or interventions are appropriate for this type of study.
- Page 18 (Results) discusses a study of Air Fluidized beds which apparently backstaged the ulcers. Stage III and IV ulcers do not heal to Stage II. This ‘study’ also failed to provide data for the control group.
- The Adjunctive Therapy Tables (13-15) starting on page 87 lack the ‘quality’ evaluation in the first column. In addition, the methodology for selecting documents led to a rating of ‘insufficient evidence’ of harm (p 108) for NPWT despite warnings from the FDA in 2009, 2011 and 2012. This should not be construed to imply that NPWT should not be used but rather that it warrants attention and monitoring unlike most topical dressings.
Evaluating the effectiveness of support surfaces for treating pressure ulcers by combining results from distinct studies is hindered by a number of factors including variability among products with similar features (e.g. Low air-loss). Mechanistically, support surfaces affect pressure ulcer treatment by redistributing pressure away from the injured site and creating a microclimate that does not adversely affect the healing process. Low air-loss is a feature of a support surface “that provides a flow of air to assist in managing the heat and humidity (microclimate) of the skin.” (NPUAP-EPUAP Clinical Practice Guidelines, 2009). Notice that the term “low air-loss” refers to a feature and not a category of product. Clearly, if a bed only provided flow of air to the skin without redistributing pressure it would not be an effective support surface. Therefore, the extent to which a surface with low air-loss affects healing is almost certainly also related to how the surface redistributes pressure. A study to isolate the effect of the low airloss feature would need to be conducted on the same mattress with and without the air escaping through the cover. There is no such study in the literature.
The statement in the structured abstract on page v that reads, “there is no overall benefit to low air-loss beds compared to standard foam mattresses” is potentially misleading because the small number of studies conducted did not control for the additional factor of pressure redistribution. Therefore, there is insufficient evidence in the literature to compare surfaces with low air-loss features to foam mattresses without low air-loss features.
The statement in the structured abstract on page v that reads, “different mattress brands are comparable in performance “ is unclear. Does “different mattress brands” refer to only support surfaces with air-fluidized features or all support surfaces? Or, does it follow from the second key point on page 17 referring to alternating pressure beds? If it is the latter, this is a potentially misleading broad generalization.
A comment on page 17 reads, “Currently there is no universally accepted classification of support surfaces.” While this technically correct, there has been broad consensus reached between the US, European, Japanese, and others that the general (nonmutually exclusive) categories of support surfaces are reactive support surface, active support surface, integrated bed system, non-powered, powered, overlay, and mattress. (NPUAP-EPUAP Clinical Practice Guidelines, 2009)
The information contained in the Structured Abstract , Executive Summary and body of the paper is very similar-perhaps it could be condensed to create a less formidable, more useable document. Even the studies rated as ‘good’ frequently have low numbers of patients and many sites; many of the studies resemble pilot studies-further complicating the task. NPUAP would be honored to collaborate with AHRQ in addressing some of the above cited concerns in greater detail than is permitted by the limited commentary section.
This document can serve as a scaffolding on which to build improved information regarding comparative effectiveness of PU treatments through continuous literature review as better quality studies become available.
In summary, we do not find any factual errors or significant oversights. Our comments and suggestions are meant to provide consistency and cohesion to the document. The above correction will produce a valuable document for clinician and academicians. Thank you for the opportunity to participate in the comment period for ARHQ’s Pressure Ulcer Treatment Strategies: A Comparative Effectiveness Review.