A randomized, controlled trial of negative pressure wound therapy of pressure ulcers via a novel polyurethane foam Marcus James Dermot Wagstaff, PhD, FRACS1; Sara Driver, RN2; Patrick Coghlan, MBBS3; John Edward Greenwood, AM, MD, DHlthSc3 1. Department of Plastic and Reconstructive Surgery, 3. Adult Burn Centre, Royal Adelaide Hospital, and 2. Royal District Nursing Service, Adelaide, South Australia, Australia
Reprint requests: A/Prof. J. E. Greenwood, Adult Burn Centre, Royal Adelaide Hospital, North Terrace, Adelaide SA 5000, Australia. Tel: +08 8222 2233, Mobile 0422 000809; Fax: +08 8222 5676; Email: [email protected] Manuscript received: May 15, 2013 Accepted in final form: December 22, 2013 DOI:10.1111/wrr.12146
ABSTRACT The objectives of the study were (1) to look for any local, clinically apparent response, within and around a debrided wound, to a novel biocompatible polyurethane foam during repeated, short-term implantation, and (2) to assess the material’s efficacy as a negative pressure wound therapy (NPWT) interface compared with a widely used, commercially available foam. Twenty pressure ulcers in 18 patients underwent surgical debridement, then randomization to receive novel treatment or control foam as the wound interface for NPWT. Dressing changes every 2–3 days allowed qualitative wound assessment and quantitative measurement to compare outcomes. No adverse reaction was observed in any patient receiving the new foam. The new “novel foam” performed as a NPWT interface as effectively as the control “standard foam.” In deep wounds, the new foam was easier to remove, fragmented less, and showed less retention than the control foam. No marginal in-growth occurred, making removal less traumatic and reducing bleeding from cavity wall granulations. The results support previous large animal studies, and independent ISO10993 testing, that the new foam is safe and biocompatible. Its efficacy as an NPWT interface, nontraumatic removal with low fragmentation and retention rate, favors the new material, especially in deep cavity wounds.
This study forms part of a project to create products based on a family of novel polyurethanes (NovoSorb, PolyNovo Biomaterials Pty Ltd, Port Melbourne, Victoria, Australia), which are biocompatible and bioabsorbable and can be formulated and processed to have a range of physical and tissue compatibility characteristics. One product is the biodegradable temporizing matrix (BTM), designed for dermal replacement in major burn injury. As this study represents the “first-in-human” experience using this polymer, a shortterm implantation (2–3 days) model was sought, allowing observation for signs or symptoms of allergic, local, or systemic reaction to its presence. Louis Argenta and Michael Morykwas introduced negative pressure wound therapy (NPWT) to clinical practice in two linked papers in the Annals of Plastic Surgery in 1997.1,2 Using a medical-grade, open-cell, polyurethane ether foam with pore sizes between 400 and 600 μm as an interface for an intermittent negative pressure of −125 mmHg, they demonstrated, in pigs, an increased wound blood flow, increased rate of granulation tissue formation, and increased bacterial clearance from the wound.1 In human chronic wounds, they additionally reported a reduction of peri-wound edema and corresponding fluid removal through the vacuum-assisted closure (VAC) system (that decreased with time), as well as the positive outcomes noted in the animal trials.2 Even in this preliminary work, the complication of bleeding secondary to traumatic removal of the foam, and disruption of the vascular granulations that had marginally invaded it, was reported.2 This Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
occurred especially where the foam had been left in situ in excess of 48 hours. There are a large number of people being treated in the community within South Australia suffering with decubitus ulceration of varied etiology and site, receiving daily silver alginate dressings. NPWT availability in the community is limited due to cost restriction. Additionally, NPWT had come under scrutiny by the US Food and Drug Administration because of an increasing number of concerns raised by practitioners and patients alike, with 12 deaths and 174 injuries attributable to interface issues between 2007 and 2011.3,4 The market leading interface for NPWT, Granufoam, is a standard, reticulated polyurethane foam and is the primary wound interface supplied by Kinetic Concepts Incorporated (KCI, San Antonio, TX). The walls between the foam “cells” have been “skeletonised” by short exposure to intense heat. As a result, it appears to be very effective in its transmission of the negative pressure to all parts of the wound cavity, stimulating exuberant granulation tissue production by the cavity walls. These granulations can, however, invade the pore cells at the margin of the foam. Removal of Granufoam under these conditions is difficult, disrupting the fragile granulations and causing bleeding,2 or causing foam fragments to be retained in the cavity wall. The subsequent “sharp” removal of these fragments (with scalpel or scissors) causes more trauma, pain, and bleeding, where failing to notice fragment retention can lead to infection. KCI produces a second NPWT foam interface. Initially called Versafoam, its trademark application 205
NPWT in pressure sores with new polyurethane foam compared to control foam
was abandoned in 2005. It is now marketed as White Foam Dressing™. This material (like NovoPore) is not reticulated and white in color. Unlike NovoPore, it is not bioabsorbable and has been the subject of serious adverse event reporting for unrecognized foam retention in a healed wound which subsequently resulted in suppurative abscess formation and cellulitis in a diabetic patient. It was not studied as the control material for several reasons. Firstly, its chemistry is fundamentally different from both NovoPore and Granufoam. Both the studied materials are polyurethanes, whereas KCI’s Versafoam is a polyvinyl alcohol polymer. As a result, it is highly hydrophilic and is supplied moist to avoid issues with swelling on exposure to water. It comes only in a 1 cm thickness and because of its moisture-retaining properties and inhibition of granulation invasion, it has become the most popular in superficial wounds where there are exposed structures that need to be kept moist (such as tendons on the dorsum of a degloved hand) and ingrowth needs to be discouraged. Its dimensions make it unsuitable for use in large, deep cavity wounds (such as ischial wounds). Because the integrating component of our BTM is an open-cell polyurethane foam, like Granufoam, with cell sizes comparable with Granufoam, but smaller pore dimensions (mean pore size 600–800 μm), the use of a block of this material as an interface for NPWT would meet the design requirements. A 30 mm foam dressing was specifically created for this NPWT study from the NovoSorb chemistry, and has been trademarked as NovoPore. The aims of this study were twofold. The primary purpose was to assess the safety and biocompatibility of a novel polyurethane foam (NovoPore) in humans. This safety has been shown in multiple small and large animal studies5–10 and independent ISO10993 testing. The secondary purpose was to ensure the efficacy of the new material in NPWT compared with an established control (Granufoam). As a safety study, not primarily designed to compare interfaces, the null hypothesis was applied (that there is no difference in efficacy between treatment material and control material as a negative pressure wound therapy interface [N = G]). From this point onwards, NovoPore will be referred to as “novel foam” and Granufoam as “standard foam.”
MATERIALS AND METHODS Approvals and registration
This pilot human study was approved by the Royal Adelaide Hospital Human Research Ethics Committee (Approval No.110612) and, subsequently, by the Australian Therapeutic Goods Administration under the Clinical Trial Notification scheme. The trial was then registered under the Australian and New Zealand Clinical Trials Registry (Trial No. ACTRN12611000815965: http://www.anzctr.org.au).
Patient recruitment and screening visit
The large majority of patients with decubitus ulceration in South Australia are managed in the community by their general practitioner, for medical input, and the Royal District Nursing Service (RDNS) for their dressings. RDNS partnership in this study was therefore essential, to identify and 206
Wagstaff et al.
recruit patients in community care and to provide a specialist wound care nurse to conduct all of the study dressings in the community following surgical debridement, avoiding issues of operator variability. Inclusion criterion was (Grade III/IV) decubitus ulceration (ischial, sacral, or heel) in patients between 18 and 75 years of age. Patients would be excluded if they demonstrated hollow viscera fistulation, were pregnant or lactating, were non-English speakers (thus unable to reliably give informed consent), or had a known history of allergy/reaction to polyurethane dressing materials or complications with NPWT. The identified subjects were visited at home by the primary investigator (PI; JEG) and the study nurse (SD). They were educated (along with their spouse or carer) relating to their wound(s), the potential benefits of surgical debridement, and the role of NPWT before their general health and their specific wound condition were assessed for suitability. The logistics of the study (surgery, timing of dressings, etc.) were discussed. An information sheet was left with the patient who was recontacted 48 hours later for their decision regarding the trial. Hospital treatment
Preoperatively Patients were admitted to the Burns Unit at the Royal Adelaide Hospital the day before proposed surgery to familiarize them with the unit, its staff and the operating surgeon, and to perform any preoperative investigations. In addition, they were assessed by our dietician, for their nutritional status, having blood tested for trace elements (zinc, selenium) and vitamins (A, C, and E). The opportunity for the patient to ask questions was provided before securing informed consent. Surgery Surgery was performed in the Burns Unit theater. The majority of patients had at least partial sensation or sensibility in their wounds and required general anesthesia for aggressive wound debridement. A few required no anesthesia and some required sedation for anxiety, or to reduce unpredictable spastic movement during debridement. The surgery was performed with the patient in a position to maximally expose the ulcer (prone with hips partly flexed in the case of ischial and sacral ulcers, lateral for heel ulcers). Once positioned, the wounds were swabbed for microbiology. In the case of ischial ulcers, the cavity was packed tight with surgical gauze swabs into which Bonney’s Blue (crystal violet) dye was injected before the opening was closed with a purse-string suture. The whole ulcer was excised including the capsule with reference to the stained packs. Hemostasis was secured within the cavity with bipolar diathermy. For sacral and heel ulcers, all diseased tissue was excised to viable deep tissue. Fragments of the excised tissue from all ulcers were sent for microbiological analysis. In all wounds where bone was exposed in the base of the ulcer, or where bony spurs or other excrescences merited removal, fragments of bone were obtained and sent to the pathology laboratory to exclude underlying osteomyelitis. The finding of osteomyelitis necessitated a prolonged course of the appropriate intravenous antibiotic. Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
Wagstaff et al.
NPWT in pressure sores with new polyurethane foam compared to control foam
Randomization Only when the debridement was completed and the internal boundaries of the cavity measured using Visitrak (Smith & Nephew, Hull, United Kingdom) was the wound randomized to treatment using sealed envelopes. Where the patient had two ulcers, each was randomized after its debridement. Obviously, because the novel foam (white) and standard foam (black) differed in both color and texture, blinding of the surgeon, nurse, and patient to the treatment option being received was impossible. It was important that the NPWT Trak-Pad (the flange through which delivered suction is transmitted to the foam interface) was not positioned over the wound foam itself, as this would represent a potential pressure insult. A foam “bridge” arrangement was established so that the Trak-Pad could be sited over a nonweight-bearing area. The wound edges and a strip of skin where the foam bridge would sit were protected using Duoderm thin hydrocolloid dressings (Convatec Inc., Skillman, NJ). The randomized material (“novel foam” treatment or “standard foam” control) was cut to size and inserted into the debrided wound cavity. The whole was sealed with a KCI polyurethane film. The KCI Trak-Pad was applied to the end of the bridge and connected to a KCI ActiVac pump set at −125 mmHg continuous vacuum pressure. The arrangement was checked for any leaks before the patients left theater. They were observed for the next 24 hours before being allowed home. The RDNS collaborative nurse checked on them the evening of discharge.
tissue, whereas reduction in peri-wound induration was shown by pressure and feel, and the time to healing was recorded (if occurring within the 8-week study period). Digital photographic assessment (with in-frame ruler) occurred at every time point. Specific comparison of the above outcomes between the treatment (“novel foam”) and control (“standard foam”) arms was performed. Data collection
All data in the community were collected by the study nurse in the study case review folder. Any concern or adverse event was similarly recorded and the PI informed. The PI visited the patient in the community on these occasions. Statistical analysis
Independent statistical analysis was performed by senior statisticians within the Data Management and Analysis Centre of the University of Adelaide using analysis of variance, multiple logistic regression analysis, and independent t-tests corrected for the numbers of comparisons (as appropriate for the dataset). The two-sided probability level of 0.05 was used as the criterion for significance. These analyses evaluated the percent change in wound size at 15, 35, and 50 days. Comparisons were made by “Group.” All analyses were conducted using SAS version 9.3 (SAS Institute Inc., Cary, NC).
Community treatment
Trial completion
Home visits and dressing changes took place on Monday, Wednesday, and Friday, with 72 hours “rest” over the weekend. If the patient had a problem with the NPWT (dressing displacement, seal loss and air leakage, excessive pump noise or alarming, etc), the study nurse would make an unscheduled visit (or the RDNS emergency line was called out-of-hours). All dressing changes were performed under aseptic conditions. Data were collected up to complete healing or to 8 weeks after surgery (whichever came first). Once the data collection period was completed, those patients treated with “novel foam,” and still requiring NPWT, were transferred onto “standard foam” dressings for ongoing treatment.
At the end of the 8-week study period, patients still requiring NPWT continued, but all were dressed with “standard foam.” Frequent poststudy review of the cohort separated those patients who had not healed within the study period from those requiring surgical flap repair, those requiring ongoing NPWT, and those able to return to simple, non-NPWT dressings.
Outcome measures
The surgeon’s perspective on ease of handling the treatment and control material for each patient’s first NPWT dressing following surgical debridement
Outcomes were chosen which would most appropriately support the study purposes. In regard to patient response to “novel foam,” outcomes included any symptoms of pain, other discomfort or itching, and signs of peri-wound redness, rash, blistering, and/or edema. Wound cavity “area” was measured by placing Visitrak “Depth” measuring probes into the extremes of the cavity and marking that spot onto the overlying skin with a marker pen. With regard to efficacy as a NPWT interface, reduction in this wound cavity area (which correlates well with wound volume) was observed by Visitrak acetate tracing and measurement, whereas reduction in bacterial population was identified clinically by odor reduction and wound appearance. Clinical assessment was supported by frequent wound surface sampling for microbial culture. The increase in wound bed vascularity was assessed by the presence of granulation Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
RESULTS The results consist of both quantitative and qualitative data and are most easily interpreted in a temporal sequence.
The operating surgeon (MJDW) had extensive experience with “standard foam” and the KCI VAC system preceding this trial. The reticulated nature of “standard foam,” and its immediate elasticity, makes cutting to size and shape with scissors easier than the denser “novel foam” (which is compressed between the blades of the scissors and takes a short, but tangible, time to reexpand). For this reason, the surgeon found it easier to shape using a surgical blade. The reticular nature of “standard foam,” however, generated multiple “shards” of foam material on cutting, which seemed to adhere to instruments, patient, and drapes, possibly by a “static” charge. This was not the case with “novel foam.” Both materials appeared to “compress” evenly and equally after the negative pressure was applied. 207
NPWT in pressure sores with new polyurethane foam compared to control foam
The nurse’s perspective—ease of use, ease of removal, material fragmentation, material retention, and workload
The study nurse (SD) compiled her independent thoughts on the materials she had been trialling. She wrote: After the first few patients, I began using the “novel foam” with increasing confidence, with experience that the new material was not being retained in the wound. This was important with larger pressure injuries with deep cavities and/or undermining, where it is impossible to visualise the entire wound depth. Clinician confidence is important in the community setting, where the patient has access to “periodical” nursing care rather than hospital-based “continual” care. “Novel foam” in its current form is dense and more difficult to shape than the control (“standard foam”). It was shaped using conventional general sterile scissors. I found it easier to compress the “novel foam,” whilst cutting it to the desired size. After obtaining the correct size, it had to be left for several seconds to reexpand so that its true profile could be established, at which point the “novel foam” often needed further modelling. With increasing familiarity, this became no more difficult than using “standard foam”. “Novel foam” did not display the “fine filament” shedding on cutting in contrast to “standard foam.” In the larger cavity wounds, “novel foam” was compressed before being implanted, potentially increasing the risk of “over-packing,” which may stent and retard shrinkage of the wound. Even in the most vascular of wounds, “novel foam” did not cause bleeding or trauma on its removal. Often, the control foam (particularly in highly vascular wounds) required saline soaking or forceful physical detachment in order to remove it. If removal required saline soaks for several minutes prior to foam removal and/or the foam had to be physically dislodged from the cavity walls by the use of implements, or gloved fingers, such incidents were recorded as “difficult.” Bleeding accompanied such traumatic removal. Furthermore, if the main piece of foam came out with ease, but additional cleaning or irrigation was required to remove residual fine filaments (specifically with “standard foam”), this also constituted a “difficult” removal. “Difficult” foam removals were only ever reported where “standard foam” was the interface in ischial wounds (Table 1). Of 50
Table 1. Ease of foam removal, foam fragmentation, and foam retention needing sharp debridement Type of wound/Material Ease of foam removal Ischial Sacral Heel Foam fragmentation All wounds Ischial ulcers Foam retention needing sharp debridement Ischial ulcers
208
“Standard foam”
Wagstaff et al.
dressing changes in ischial wounds treated with “standard foam”, “difficult” foam removals were recorded in 40 (80%). All “novel foam” dressing changes, in all wound types, were reported to be “easy” (72/72–100%). Foam fragmentation was recorded during dressing changes with “standard foam” in 21/69 (30.4%). However, all “standard foam” fragmentation occurred in deep cavity, ischial wounds 21/50 (42%). With “novel foam,” fragmentation was recorded in 14/72 (19.4%); with 13/42 (31%) ischial and 1/8 (12.5%) sacral (Table 1). Fragmentation is important because it increases the likelihood of foam retention within the wound. Retention of material was recorded in 5 patients, all with ischial sores—four with “standard foam,” one with “novel foam.” In the case of “novel foam,” the retained foam was successfully washed out of the cavity with saline with no subsequent infection. The four patients with “standard foam” retention represented 5 episodes (one patient had two episodes of retention in the trial period). With “standard foam” retention—sharp debridement and removal with scissors was required on 4 occasions, leading to infection, in the following few days, in 3 patients (Table 1). The retained material could be washed out with saline on one occasion, where infection did not follow. “Novel foam” is less suited to being used in “bridging.” This technique was required to ensure that the Trak-Pad™ sat on uninjured skin over the non weight-bearing hip, rather than over the foam-filled wound. By the third trial patient randomized to “novel foam,” I had ceased using the material as the bridge.
The patient’s perspective—the efficacy of each material as an interface for transmitting topical negative pressure to the wound
Excluding the patient who was withdrawn from the trial for failure of NPWT in the community, every wound improved markedly from surgical debridement to trial conclusion. Home dressings had to be scheduled to fit in with patient’s medical appointments, carer visits, showering, etc., and as such while subjects were assessed at multiple time points, these were not standard across subjects. Thus, because not all subjects were assessed exactly at 15, 35, or 50 days, the closest day to the cut point assessed (15, 35, or 50 days) was selected. The baseline wound areas for the two sacral ulcers were similar to that for the ischial ulcers and these were assigned to the same group in all analyses. Preliminary analysis indicated that the results from combining the groups did not differ from those obtained by removing the two sacral wound cases.
“Novel foam”
Day 15 key findings 40/50 (80%) difficult 8/8 easy 11/11 easy
8/8 easy 22/22 easy
21/69 (30.4%) 21/50 (42%)
14/72 (19.4%) 13/42 (31%)
4, with 3 sharp debrided
42/42 easy
1, with 0 sharp debrided
(1) There were no significant differences in percent change in wound cavity area (at 15 days) between groups (“standard foam” vs. “novel foam’) (Table 2a). (2) Group differences were not significant following adjustment for baseline wound cavity area (Table 2b). Day 35 key findings (1) There were no significant differences in percent change in wound cavity area (at 35 days) between groups (‘standard foam” vs. “novel foam’) (Table 3a). (2) Group differences were not significant following adjustment for baseline wound cavity area (Table 3b). Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
Wagstaff et al.
NPWT in pressure sores with new polyurethane foam compared to control foam
Table 2. (a) Unadjusted group mean wound areas (in cm2) at 15-day follow-up, and (b) group means adjusted for wound cavity area at baseline (15-day follow-up) Group
Mean
Lower CI
Upper CI
a
b
a
b
a
b
“Standard foam” “Novel foam”
59.59 64.26
58.93 64.99
46.25 50.20
46.21 51.58
72.93 78.32
71.65 78.40
Comparison
Mean difference
“Standard foam” vs. “novel foam”
Lower CI: mean diff.
Upper CI: mean diff.
p-Value
a
b
a
b
a
b
a
b
−4.67
−6.06
−24.05
−24.60
14.71
−12.48
0.64
0.52
Day 50 key findings (1) There were no significant differences in percent change in wound cavity area (at 50 days) between groups (‘standard foam” vs. “novel foam”) despite a large effect (median Group “standard foam” vs. “novel foam”: 52.26 v 28.10) (Table 4a). (2) Group differences were not significant following adjustment for baseline wound cavity area (Table 4b).
The Health Service/RDNS perspective—overall outcome for this patient group
Of 20 wounds, in 18 patients recruited into the trial, the outcomes were as follows: (1) Healing within the study period occurred in six wounds (five patients). (2) Healing outside the study subsequently in two wounds (two patients) (3) Surgical closure by posterior, hamstring V-Y advancement flap occurred in four wounds (four patients) and a gluteal rotation flap in one wound (one patient).
These results can be more easily appreciated graphically (Figure 1).
Table 3. (a) Unadjusted group mean wound areas (in cm2) at 35-day follow-up, and (b) group means adjusted for wound cavity area at baseline (35-day follow-up) Group
Mean
Lower CI
Upper CI
a
b
a
b
a
b
“Standard foam” “Novel foam”
41.57 44.58
41.43 44.74
26.94 29.06
27.02 29.45
56.21 60.11
55.85 60.03
Comparison
Mean difference
“Standard foam” vs. “novel foam”
Lower CI: mean diff.
Upper CI: mean diff.
p-Value
a
b
a
b
a
b
a
b
−3.01
–3.31
−24.35
−24.33
18.32
17.72
0.78
0.76
Table 4. (a) Unadjusted group mean wound areas (in cm2) at 50 day follow-up, and (b) group means adjusted for wound cavity area at baseline (50-day follow-up) Group
Mean
Lower CI
Upper CI
a
b
a
b
a
b
“Standard foam” “Novel foam”
40.50 29.77
40.13 30.13
24.63 13.90
24.81 14.79
56.37 45.64
55.48 45.46
Comparison
Mean difference
“Standard foam” vs. “novel foam”
Lower CI: Mean diff.
Upper CI: Mean diff.
p-Value
a
b
a
b
a
b
a
b
10.73
10.02
−11.71
−11.68
33.17
31.72
0.35
0.37
Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
209
NPWT in pressure sores with new polyurethane foam compared to control foam
Mean group wound cavity area as a percentage of original wound cavity area
100 90 80 70
p=0.52
60 50
p=0.76
40 p=0.37
30 20 10 0
0
5
10
15
20
25 30 Day
35
40
45
50
Figure 1. Efficacy of each material as an NPWT interface. At days 15, 35, and 50 postsurgery, the group means of the percentage wound cavity area reduction (from the mean original wound cavity area) are compared. No significant differences were achieved. NPWT, negative pressure wound therapy.
(4) Ongoing NPWT in a cleaner, smaller wound in two wounds (two patients) (5) Ongoing management of cleaner, smaller wounds in conventional dressings in two wounds (two patients) (6) Withdrawal from the study for NPWT failure in the community (one wound— one patient) (7) Death due to unrelated causes (pneumonia) in one patient (with one wound). (8) Patient lost to follow-up by his own volition after study completion in one patient (with one wound)
DISCUSSION Safety and biocompatibility
In the 20 ulcers in 18 patients observed in this study, the qualitative results show that neither the “novel foam,” nor the “standard foam,” was associated with any observable local or systemic reaction, or signs of allergy/sensitivity. Comparative findings
The US FDA concerns surrounding patient mortality and morbidity as complications of NPWT were focused in their released warnings of 2009 and 2011.3,4 The issues raised were bleeding, foam retention, and infection—phenomena often insignificant and frequently seen when using NPWT. However, certain population groups are at enhanced risk of serious morbidity, or death, as a result. Bleeding
Patients on anticoagulant therapy (heparin/fractionated heparin, warfarin, etc.), blood “thinners” (aspirin, dipyridam210
Wagstaff et al.
ole, etc.), or those with a bleeding diathesis (hemophilia, etc.) are considered “at risk” of bleeding problems with NPWT. Granufoam (“standard foam”) generated the majority of such incidences.4 The structure of “standard foam” facilitates the generation of vascular granulation tissue from the cavity walls and supports marginal integration of the foam structure, and this phenomenon seems to be much more prevalent in deep cavity wounds. Foam removal at dressing changes in such wounds was frequently difficult (80% in our series), with the foam having to be traumatically peeled away from the friable granulations, resulting in bleeding. Such blood loss is usually small and self-limiting. In this small cohort, “novel foam” interface wounds did not show any bleeding after foam removal. The ease of removal suggests that marginal invasion of the foam did not occur and was thus “atraumatic” on separation. The nature of the “novel foam” allows greater compression of the marginal foam cells under the negative pressure and may discourage ingrowth (at least over the maximal 72 hours of foam implantation in this study). Infection
The microbiological analysis of the predebrided wounds, debrided tissue, and wound cavities during NPWT revealed a wide array of organisms, many pathological. Their presence on swabbing, however, even in great numbers, was not frequently translated into clinical infection (either local or systemic), or a delay in wound healing/cavity size reduction. Infection did, however, frequently follow “standard foam” retention. This is defined as foam fragments becoming embedded in the wound cavity wall where they appear to act as a nidus for bacterial infection and is reported as the major mechanism of wound infection using NPWT.4 Apparent retention of “novel foam” was seen in only one patient (foam fragments visualized against the wall of the cavity after dressing removal); however, this “retention” was washed easily out of the wound with saline irrigation, without trauma or the need for sharp debridement. Infection did not follow this episode. The retention issue is intimately related to the bleeding issue, namely a problem of marginally integrated foam removal. Preventing marginal integration of the foam by cavity wall granulations appears to solve both issues. Handling and ease of (duration of) dressing changes
In the experience of both surgeon and nurse, “novel foam” was more difficult to shape. This reduced with familiarity. In the community, however, nursing time is precious and this would seem to be a negative attribute. However, 12 of the 20 wounds in this study were ischial and of the six dressed with “standard foam,” foam removal was difficult (and thus time consuming) in 80% of dressing changes. Foam retention and the need for sharp debridement augmented the time taken for dressing changes, and infection required further clinical (medical) intervention and prescription. The results suggest that the “novel foam” interface might be better suited to wounds where the full extent of the cavity cannot be easily visualized during dressing changes. This occurs where there are very large cavities (such as ischial ulcers), significant undermining of wound edges, or the presence of narrow “tunnelled” wounds or wound components. Additionally (although not addressed directly), it may provide Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
Wagstaff et al.
NPWT in pressure sores with new polyurethane foam compared to control foam
a safer interface where patients requiring NPWT are concurrently being treated with anticoagulants or blood thinners, or are immunosuppressed or immunocompromised and thus at risk of life-threatening infection originating in the wound as a result of foam retention. This is a small-scale trial primarily investigating the introduction of a novel polyurethane foam in a patient group with impaired sensation. Our experience is that early NPWT dressing changes in acute wounds with “standard foam” are painful, because of marginal ingrowth and traumatic foam removal. Given that the removal of “novel foam” was less traumatic, it may be less painful than “standard foam” in acute and highly sensate wounds. Differences in pain on dressing changes in acute wounds treated with NPWT would be an avenue for further study in a larger scale study.
ACKNOWLEDGMENTS The authors would like to thank Sue Templeton and Prof. Debbie Kralik of the Adelaide Royal District Nursing Service, who were pivotal in establishing the collaboration between hospital and community that allowed the study to proceed. Thanks also go to Valerie Cohn, Meredith Sullivan, and Brenton Sutton of KCI Australia for provision of pumps, control foams, and other accessories. Thomas Sullivan, Senior Statistician within the Data Management & Analysis Centre, Discipline of Public Health of the University of Adelaide, performed all the independent statistical analysis. Finally, thanks to the staff of the Adult Burns Unit, Royal Adelaide Hospital, for looking after these patients during their debridement hospital admission. Source of Funding: Funding for this study was provided by PolyNovo Biomaterials Pty Ltd, Port Melbourne, Victoria, Australia, who provided the treatment materials (NovoPore) and paid the RDNS to backfill the position of the study nurse. By virtue of the gratis supply of pumps, dressings and control foam, Kinetic Concepts Incorporated, San Antonio, Texas, additionally supported the trial. Conflict of Interest: A/Prof. John Edward Greenwood, AM, owns a 20% share of NovoWound Pty. Ltd. through his company, Skin Pty. Ltd. NovoWound Pty. Ltd. is a joint venture company established to explore commercialization opportunities of the NovoSorb polymers (including NovoPore) in the field of wound management. He receives no salary or other payment from this collaboration. Dr. Marcus James Dermot Wagstaff has a minor shareholding in Calzada
Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
Pty Ltd, the Australian Stock Market (ASX)-listed parent company of PolyNovo Biomaterials Pty Ltd. Neither remaining authors have any conflict of interest.
REFERENCES 1. Morykwas MJ, Argenta LC, Shelton-Brown EI, McGuirt W. Vacuum-assisted closure: a new method of wound control and treatment: animal studies and basic foundation. Ann Plast Surg 1997; 38: 553–62. 2. Argenta LC, Morykwas MJ. Vacuum-assisted closure: a new method for wound control and treatment: clinical experience. Ann Plast Surg 1997; 38: 563–77. 3. FDA Preliminary Public Health Notification on Negative Pressure Wound Therapy Systems. November 13, 2009. Available at http://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ PublicHealthNotifications/ucm190658.htm (accessed May 1, 2013). 4. FDA Safety Communication. UPDATE on serious complications associated with negative pressure wound therapy systems. February 24, 2011. Available at http://www.fda.gov/MedicalDevices/ Safety/AlertsandNotices/ucm244211.htm (accessed May 1, 2013). 5. Li A, Dearman BL, Crompton KE, Moore TG, Greenwood JE. Evaluation of a novel biodegradable polymer for the generation of a dermal matrix. J Burn Care Res 2009; 30: 717–28. 6. Greenwood JE, Li A, Dearman B, Moore TG. Evaluation of NovoSorb™ novel biodegradable polymer for the generation of a dermal matrix. Part 1: in-vitro studies. Wound Pract Res 2010; 18: 14–22. 7. Greenwood JE, Li A, Dearman B, Moore TG. Evaluation of NovoSorb™ novel biodegradable polymer for the generation of a dermal matrix. Part 2: in-vivo studies. Wound Pract Res 2010; 18: 24–34. 8. Greenwood JE, Dearman BL. Split-skin graft application over an integrating, biodegradable temporising polymer matrix: immediate and delayed. J Burn Care Res 2012; 33: 7–19. 9. Greenwood JE, Dearman BL. Comparison of a sealed, polymer foam biodegradable temporising matrix against Integra™ dermal regeneration template in a porcine wound model. J Burn Care Res 2012; 33: 163–73. 10. Dearman BL, Stefani K, Li A, Greenwood JE. Take of a polymer-based autologous cultured composite “skin” on an integrated temporising dermal matrix: proof of concept. J Burn Care Res 2013; 34: 151–60.
211
Reprint requests: A/Prof. J. E. Greenwood, Adult Burn Centre, Royal Adelaide Hospital, North Terrace, Adelaide SA 5000, Australia. Tel: +08 8222 2233, Mobile 0422 000809; Fax: +08 8222 5676; Email: [email protected] Manuscript received: May 15, 2013 Accepted in final form: December 22, 2013 DOI:10.1111/wrr.12146
ABSTRACT The objectives of the study were (1) to look for any local, clinically apparent response, within and around a debrided wound, to a novel biocompatible polyurethane foam during repeated, short-term implantation, and (2) to assess the material’s efficacy as a negative pressure wound therapy (NPWT) interface compared with a widely used, commercially available foam. Twenty pressure ulcers in 18 patients underwent surgical debridement, then randomization to receive novel treatment or control foam as the wound interface for NPWT. Dressing changes every 2–3 days allowed qualitative wound assessment and quantitative measurement to compare outcomes. No adverse reaction was observed in any patient receiving the new foam. The new “novel foam” performed as a NPWT interface as effectively as the control “standard foam.” In deep wounds, the new foam was easier to remove, fragmented less, and showed less retention than the control foam. No marginal in-growth occurred, making removal less traumatic and reducing bleeding from cavity wall granulations. The results support previous large animal studies, and independent ISO10993 testing, that the new foam is safe and biocompatible. Its efficacy as an NPWT interface, nontraumatic removal with low fragmentation and retention rate, favors the new material, especially in deep cavity wounds.
This study forms part of a project to create products based on a family of novel polyurethanes (NovoSorb, PolyNovo Biomaterials Pty Ltd, Port Melbourne, Victoria, Australia), which are biocompatible and bioabsorbable and can be formulated and processed to have a range of physical and tissue compatibility characteristics. One product is the biodegradable temporizing matrix (BTM), designed for dermal replacement in major burn injury. As this study represents the “first-in-human” experience using this polymer, a shortterm implantation (2–3 days) model was sought, allowing observation for signs or symptoms of allergic, local, or systemic reaction to its presence. Louis Argenta and Michael Morykwas introduced negative pressure wound therapy (NPWT) to clinical practice in two linked papers in the Annals of Plastic Surgery in 1997.1,2 Using a medical-grade, open-cell, polyurethane ether foam with pore sizes between 400 and 600 μm as an interface for an intermittent negative pressure of −125 mmHg, they demonstrated, in pigs, an increased wound blood flow, increased rate of granulation tissue formation, and increased bacterial clearance from the wound.1 In human chronic wounds, they additionally reported a reduction of peri-wound edema and corresponding fluid removal through the vacuum-assisted closure (VAC) system (that decreased with time), as well as the positive outcomes noted in the animal trials.2 Even in this preliminary work, the complication of bleeding secondary to traumatic removal of the foam, and disruption of the vascular granulations that had marginally invaded it, was reported.2 This Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
occurred especially where the foam had been left in situ in excess of 48 hours. There are a large number of people being treated in the community within South Australia suffering with decubitus ulceration of varied etiology and site, receiving daily silver alginate dressings. NPWT availability in the community is limited due to cost restriction. Additionally, NPWT had come under scrutiny by the US Food and Drug Administration because of an increasing number of concerns raised by practitioners and patients alike, with 12 deaths and 174 injuries attributable to interface issues between 2007 and 2011.3,4 The market leading interface for NPWT, Granufoam, is a standard, reticulated polyurethane foam and is the primary wound interface supplied by Kinetic Concepts Incorporated (KCI, San Antonio, TX). The walls between the foam “cells” have been “skeletonised” by short exposure to intense heat. As a result, it appears to be very effective in its transmission of the negative pressure to all parts of the wound cavity, stimulating exuberant granulation tissue production by the cavity walls. These granulations can, however, invade the pore cells at the margin of the foam. Removal of Granufoam under these conditions is difficult, disrupting the fragile granulations and causing bleeding,2 or causing foam fragments to be retained in the cavity wall. The subsequent “sharp” removal of these fragments (with scalpel or scissors) causes more trauma, pain, and bleeding, where failing to notice fragment retention can lead to infection. KCI produces a second NPWT foam interface. Initially called Versafoam, its trademark application 205
NPWT in pressure sores with new polyurethane foam compared to control foam
was abandoned in 2005. It is now marketed as White Foam Dressing™. This material (like NovoPore) is not reticulated and white in color. Unlike NovoPore, it is not bioabsorbable and has been the subject of serious adverse event reporting for unrecognized foam retention in a healed wound which subsequently resulted in suppurative abscess formation and cellulitis in a diabetic patient. It was not studied as the control material for several reasons. Firstly, its chemistry is fundamentally different from both NovoPore and Granufoam. Both the studied materials are polyurethanes, whereas KCI’s Versafoam is a polyvinyl alcohol polymer. As a result, it is highly hydrophilic and is supplied moist to avoid issues with swelling on exposure to water. It comes only in a 1 cm thickness and because of its moisture-retaining properties and inhibition of granulation invasion, it has become the most popular in superficial wounds where there are exposed structures that need to be kept moist (such as tendons on the dorsum of a degloved hand) and ingrowth needs to be discouraged. Its dimensions make it unsuitable for use in large, deep cavity wounds (such as ischial wounds). Because the integrating component of our BTM is an open-cell polyurethane foam, like Granufoam, with cell sizes comparable with Granufoam, but smaller pore dimensions (mean pore size 600–800 μm), the use of a block of this material as an interface for NPWT would meet the design requirements. A 30 mm foam dressing was specifically created for this NPWT study from the NovoSorb chemistry, and has been trademarked as NovoPore. The aims of this study were twofold. The primary purpose was to assess the safety and biocompatibility of a novel polyurethane foam (NovoPore) in humans. This safety has been shown in multiple small and large animal studies5–10 and independent ISO10993 testing. The secondary purpose was to ensure the efficacy of the new material in NPWT compared with an established control (Granufoam). As a safety study, not primarily designed to compare interfaces, the null hypothesis was applied (that there is no difference in efficacy between treatment material and control material as a negative pressure wound therapy interface [N = G]). From this point onwards, NovoPore will be referred to as “novel foam” and Granufoam as “standard foam.”
MATERIALS AND METHODS Approvals and registration
This pilot human study was approved by the Royal Adelaide Hospital Human Research Ethics Committee (Approval No.110612) and, subsequently, by the Australian Therapeutic Goods Administration under the Clinical Trial Notification scheme. The trial was then registered under the Australian and New Zealand Clinical Trials Registry (Trial No. ACTRN12611000815965: http://www.anzctr.org.au).
Patient recruitment and screening visit
The large majority of patients with decubitus ulceration in South Australia are managed in the community by their general practitioner, for medical input, and the Royal District Nursing Service (RDNS) for their dressings. RDNS partnership in this study was therefore essential, to identify and 206
Wagstaff et al.
recruit patients in community care and to provide a specialist wound care nurse to conduct all of the study dressings in the community following surgical debridement, avoiding issues of operator variability. Inclusion criterion was (Grade III/IV) decubitus ulceration (ischial, sacral, or heel) in patients between 18 and 75 years of age. Patients would be excluded if they demonstrated hollow viscera fistulation, were pregnant or lactating, were non-English speakers (thus unable to reliably give informed consent), or had a known history of allergy/reaction to polyurethane dressing materials or complications with NPWT. The identified subjects were visited at home by the primary investigator (PI; JEG) and the study nurse (SD). They were educated (along with their spouse or carer) relating to their wound(s), the potential benefits of surgical debridement, and the role of NPWT before their general health and their specific wound condition were assessed for suitability. The logistics of the study (surgery, timing of dressings, etc.) were discussed. An information sheet was left with the patient who was recontacted 48 hours later for their decision regarding the trial. Hospital treatment
Preoperatively Patients were admitted to the Burns Unit at the Royal Adelaide Hospital the day before proposed surgery to familiarize them with the unit, its staff and the operating surgeon, and to perform any preoperative investigations. In addition, they were assessed by our dietician, for their nutritional status, having blood tested for trace elements (zinc, selenium) and vitamins (A, C, and E). The opportunity for the patient to ask questions was provided before securing informed consent. Surgery Surgery was performed in the Burns Unit theater. The majority of patients had at least partial sensation or sensibility in their wounds and required general anesthesia for aggressive wound debridement. A few required no anesthesia and some required sedation for anxiety, or to reduce unpredictable spastic movement during debridement. The surgery was performed with the patient in a position to maximally expose the ulcer (prone with hips partly flexed in the case of ischial and sacral ulcers, lateral for heel ulcers). Once positioned, the wounds were swabbed for microbiology. In the case of ischial ulcers, the cavity was packed tight with surgical gauze swabs into which Bonney’s Blue (crystal violet) dye was injected before the opening was closed with a purse-string suture. The whole ulcer was excised including the capsule with reference to the stained packs. Hemostasis was secured within the cavity with bipolar diathermy. For sacral and heel ulcers, all diseased tissue was excised to viable deep tissue. Fragments of the excised tissue from all ulcers were sent for microbiological analysis. In all wounds where bone was exposed in the base of the ulcer, or where bony spurs or other excrescences merited removal, fragments of bone were obtained and sent to the pathology laboratory to exclude underlying osteomyelitis. The finding of osteomyelitis necessitated a prolonged course of the appropriate intravenous antibiotic. Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
Wagstaff et al.
NPWT in pressure sores with new polyurethane foam compared to control foam
Randomization Only when the debridement was completed and the internal boundaries of the cavity measured using Visitrak (Smith & Nephew, Hull, United Kingdom) was the wound randomized to treatment using sealed envelopes. Where the patient had two ulcers, each was randomized after its debridement. Obviously, because the novel foam (white) and standard foam (black) differed in both color and texture, blinding of the surgeon, nurse, and patient to the treatment option being received was impossible. It was important that the NPWT Trak-Pad (the flange through which delivered suction is transmitted to the foam interface) was not positioned over the wound foam itself, as this would represent a potential pressure insult. A foam “bridge” arrangement was established so that the Trak-Pad could be sited over a nonweight-bearing area. The wound edges and a strip of skin where the foam bridge would sit were protected using Duoderm thin hydrocolloid dressings (Convatec Inc., Skillman, NJ). The randomized material (“novel foam” treatment or “standard foam” control) was cut to size and inserted into the debrided wound cavity. The whole was sealed with a KCI polyurethane film. The KCI Trak-Pad was applied to the end of the bridge and connected to a KCI ActiVac pump set at −125 mmHg continuous vacuum pressure. The arrangement was checked for any leaks before the patients left theater. They were observed for the next 24 hours before being allowed home. The RDNS collaborative nurse checked on them the evening of discharge.
tissue, whereas reduction in peri-wound induration was shown by pressure and feel, and the time to healing was recorded (if occurring within the 8-week study period). Digital photographic assessment (with in-frame ruler) occurred at every time point. Specific comparison of the above outcomes between the treatment (“novel foam”) and control (“standard foam”) arms was performed. Data collection
All data in the community were collected by the study nurse in the study case review folder. Any concern or adverse event was similarly recorded and the PI informed. The PI visited the patient in the community on these occasions. Statistical analysis
Independent statistical analysis was performed by senior statisticians within the Data Management and Analysis Centre of the University of Adelaide using analysis of variance, multiple logistic regression analysis, and independent t-tests corrected for the numbers of comparisons (as appropriate for the dataset). The two-sided probability level of 0.05 was used as the criterion for significance. These analyses evaluated the percent change in wound size at 15, 35, and 50 days. Comparisons were made by “Group.” All analyses were conducted using SAS version 9.3 (SAS Institute Inc., Cary, NC).
Community treatment
Trial completion
Home visits and dressing changes took place on Monday, Wednesday, and Friday, with 72 hours “rest” over the weekend. If the patient had a problem with the NPWT (dressing displacement, seal loss and air leakage, excessive pump noise or alarming, etc), the study nurse would make an unscheduled visit (or the RDNS emergency line was called out-of-hours). All dressing changes were performed under aseptic conditions. Data were collected up to complete healing or to 8 weeks after surgery (whichever came first). Once the data collection period was completed, those patients treated with “novel foam,” and still requiring NPWT, were transferred onto “standard foam” dressings for ongoing treatment.
At the end of the 8-week study period, patients still requiring NPWT continued, but all were dressed with “standard foam.” Frequent poststudy review of the cohort separated those patients who had not healed within the study period from those requiring surgical flap repair, those requiring ongoing NPWT, and those able to return to simple, non-NPWT dressings.
Outcome measures
The surgeon’s perspective on ease of handling the treatment and control material for each patient’s first NPWT dressing following surgical debridement
Outcomes were chosen which would most appropriately support the study purposes. In regard to patient response to “novel foam,” outcomes included any symptoms of pain, other discomfort or itching, and signs of peri-wound redness, rash, blistering, and/or edema. Wound cavity “area” was measured by placing Visitrak “Depth” measuring probes into the extremes of the cavity and marking that spot onto the overlying skin with a marker pen. With regard to efficacy as a NPWT interface, reduction in this wound cavity area (which correlates well with wound volume) was observed by Visitrak acetate tracing and measurement, whereas reduction in bacterial population was identified clinically by odor reduction and wound appearance. Clinical assessment was supported by frequent wound surface sampling for microbial culture. The increase in wound bed vascularity was assessed by the presence of granulation Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
RESULTS The results consist of both quantitative and qualitative data and are most easily interpreted in a temporal sequence.
The operating surgeon (MJDW) had extensive experience with “standard foam” and the KCI VAC system preceding this trial. The reticulated nature of “standard foam,” and its immediate elasticity, makes cutting to size and shape with scissors easier than the denser “novel foam” (which is compressed between the blades of the scissors and takes a short, but tangible, time to reexpand). For this reason, the surgeon found it easier to shape using a surgical blade. The reticular nature of “standard foam,” however, generated multiple “shards” of foam material on cutting, which seemed to adhere to instruments, patient, and drapes, possibly by a “static” charge. This was not the case with “novel foam.” Both materials appeared to “compress” evenly and equally after the negative pressure was applied. 207
NPWT in pressure sores with new polyurethane foam compared to control foam
The nurse’s perspective—ease of use, ease of removal, material fragmentation, material retention, and workload
The study nurse (SD) compiled her independent thoughts on the materials she had been trialling. She wrote: After the first few patients, I began using the “novel foam” with increasing confidence, with experience that the new material was not being retained in the wound. This was important with larger pressure injuries with deep cavities and/or undermining, where it is impossible to visualise the entire wound depth. Clinician confidence is important in the community setting, where the patient has access to “periodical” nursing care rather than hospital-based “continual” care. “Novel foam” in its current form is dense and more difficult to shape than the control (“standard foam”). It was shaped using conventional general sterile scissors. I found it easier to compress the “novel foam,” whilst cutting it to the desired size. After obtaining the correct size, it had to be left for several seconds to reexpand so that its true profile could be established, at which point the “novel foam” often needed further modelling. With increasing familiarity, this became no more difficult than using “standard foam”. “Novel foam” did not display the “fine filament” shedding on cutting in contrast to “standard foam.” In the larger cavity wounds, “novel foam” was compressed before being implanted, potentially increasing the risk of “over-packing,” which may stent and retard shrinkage of the wound. Even in the most vascular of wounds, “novel foam” did not cause bleeding or trauma on its removal. Often, the control foam (particularly in highly vascular wounds) required saline soaking or forceful physical detachment in order to remove it. If removal required saline soaks for several minutes prior to foam removal and/or the foam had to be physically dislodged from the cavity walls by the use of implements, or gloved fingers, such incidents were recorded as “difficult.” Bleeding accompanied such traumatic removal. Furthermore, if the main piece of foam came out with ease, but additional cleaning or irrigation was required to remove residual fine filaments (specifically with “standard foam”), this also constituted a “difficult” removal. “Difficult” foam removals were only ever reported where “standard foam” was the interface in ischial wounds (Table 1). Of 50
Table 1. Ease of foam removal, foam fragmentation, and foam retention needing sharp debridement Type of wound/Material Ease of foam removal Ischial Sacral Heel Foam fragmentation All wounds Ischial ulcers Foam retention needing sharp debridement Ischial ulcers
208
“Standard foam”
Wagstaff et al.
dressing changes in ischial wounds treated with “standard foam”, “difficult” foam removals were recorded in 40 (80%). All “novel foam” dressing changes, in all wound types, were reported to be “easy” (72/72–100%). Foam fragmentation was recorded during dressing changes with “standard foam” in 21/69 (30.4%). However, all “standard foam” fragmentation occurred in deep cavity, ischial wounds 21/50 (42%). With “novel foam,” fragmentation was recorded in 14/72 (19.4%); with 13/42 (31%) ischial and 1/8 (12.5%) sacral (Table 1). Fragmentation is important because it increases the likelihood of foam retention within the wound. Retention of material was recorded in 5 patients, all with ischial sores—four with “standard foam,” one with “novel foam.” In the case of “novel foam,” the retained foam was successfully washed out of the cavity with saline with no subsequent infection. The four patients with “standard foam” retention represented 5 episodes (one patient had two episodes of retention in the trial period). With “standard foam” retention—sharp debridement and removal with scissors was required on 4 occasions, leading to infection, in the following few days, in 3 patients (Table 1). The retained material could be washed out with saline on one occasion, where infection did not follow. “Novel foam” is less suited to being used in “bridging.” This technique was required to ensure that the Trak-Pad™ sat on uninjured skin over the non weight-bearing hip, rather than over the foam-filled wound. By the third trial patient randomized to “novel foam,” I had ceased using the material as the bridge.
The patient’s perspective—the efficacy of each material as an interface for transmitting topical negative pressure to the wound
Excluding the patient who was withdrawn from the trial for failure of NPWT in the community, every wound improved markedly from surgical debridement to trial conclusion. Home dressings had to be scheduled to fit in with patient’s medical appointments, carer visits, showering, etc., and as such while subjects were assessed at multiple time points, these were not standard across subjects. Thus, because not all subjects were assessed exactly at 15, 35, or 50 days, the closest day to the cut point assessed (15, 35, or 50 days) was selected. The baseline wound areas for the two sacral ulcers were similar to that for the ischial ulcers and these were assigned to the same group in all analyses. Preliminary analysis indicated that the results from combining the groups did not differ from those obtained by removing the two sacral wound cases.
“Novel foam”
Day 15 key findings 40/50 (80%) difficult 8/8 easy 11/11 easy
8/8 easy 22/22 easy
21/69 (30.4%) 21/50 (42%)
14/72 (19.4%) 13/42 (31%)
4, with 3 sharp debrided
42/42 easy
1, with 0 sharp debrided
(1) There were no significant differences in percent change in wound cavity area (at 15 days) between groups (“standard foam” vs. “novel foam’) (Table 2a). (2) Group differences were not significant following adjustment for baseline wound cavity area (Table 2b). Day 35 key findings (1) There were no significant differences in percent change in wound cavity area (at 35 days) between groups (‘standard foam” vs. “novel foam’) (Table 3a). (2) Group differences were not significant following adjustment for baseline wound cavity area (Table 3b). Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
Wagstaff et al.
NPWT in pressure sores with new polyurethane foam compared to control foam
Table 2. (a) Unadjusted group mean wound areas (in cm2) at 15-day follow-up, and (b) group means adjusted for wound cavity area at baseline (15-day follow-up) Group
Mean
Lower CI
Upper CI
a
b
a
b
a
b
“Standard foam” “Novel foam”
59.59 64.26
58.93 64.99
46.25 50.20
46.21 51.58
72.93 78.32
71.65 78.40
Comparison
Mean difference
“Standard foam” vs. “novel foam”
Lower CI: mean diff.
Upper CI: mean diff.
p-Value
a
b
a
b
a
b
a
b
−4.67
−6.06
−24.05
−24.60
14.71
−12.48
0.64
0.52
Day 50 key findings (1) There were no significant differences in percent change in wound cavity area (at 50 days) between groups (‘standard foam” vs. “novel foam”) despite a large effect (median Group “standard foam” vs. “novel foam”: 52.26 v 28.10) (Table 4a). (2) Group differences were not significant following adjustment for baseline wound cavity area (Table 4b).
The Health Service/RDNS perspective—overall outcome for this patient group
Of 20 wounds, in 18 patients recruited into the trial, the outcomes were as follows: (1) Healing within the study period occurred in six wounds (five patients). (2) Healing outside the study subsequently in two wounds (two patients) (3) Surgical closure by posterior, hamstring V-Y advancement flap occurred in four wounds (four patients) and a gluteal rotation flap in one wound (one patient).
These results can be more easily appreciated graphically (Figure 1).
Table 3. (a) Unadjusted group mean wound areas (in cm2) at 35-day follow-up, and (b) group means adjusted for wound cavity area at baseline (35-day follow-up) Group
Mean
Lower CI
Upper CI
a
b
a
b
a
b
“Standard foam” “Novel foam”
41.57 44.58
41.43 44.74
26.94 29.06
27.02 29.45
56.21 60.11
55.85 60.03
Comparison
Mean difference
“Standard foam” vs. “novel foam”
Lower CI: mean diff.
Upper CI: mean diff.
p-Value
a
b
a
b
a
b
a
b
−3.01
–3.31
−24.35
−24.33
18.32
17.72
0.78
0.76
Table 4. (a) Unadjusted group mean wound areas (in cm2) at 50 day follow-up, and (b) group means adjusted for wound cavity area at baseline (50-day follow-up) Group
Mean
Lower CI
Upper CI
a
b
a
b
a
b
“Standard foam” “Novel foam”
40.50 29.77
40.13 30.13
24.63 13.90
24.81 14.79
56.37 45.64
55.48 45.46
Comparison
Mean difference
“Standard foam” vs. “novel foam”
Lower CI: Mean diff.
Upper CI: Mean diff.
p-Value
a
b
a
b
a
b
a
b
10.73
10.02
−11.71
−11.68
33.17
31.72
0.35
0.37
Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
209
NPWT in pressure sores with new polyurethane foam compared to control foam
Mean group wound cavity area as a percentage of original wound cavity area
100 90 80 70
p=0.52
60 50
p=0.76
40 p=0.37
30 20 10 0
0
5
10
15
20
25 30 Day
35
40
45
50
Figure 1. Efficacy of each material as an NPWT interface. At days 15, 35, and 50 postsurgery, the group means of the percentage wound cavity area reduction (from the mean original wound cavity area) are compared. No significant differences were achieved. NPWT, negative pressure wound therapy.
(4) Ongoing NPWT in a cleaner, smaller wound in two wounds (two patients) (5) Ongoing management of cleaner, smaller wounds in conventional dressings in two wounds (two patients) (6) Withdrawal from the study for NPWT failure in the community (one wound— one patient) (7) Death due to unrelated causes (pneumonia) in one patient (with one wound). (8) Patient lost to follow-up by his own volition after study completion in one patient (with one wound)
DISCUSSION Safety and biocompatibility
In the 20 ulcers in 18 patients observed in this study, the qualitative results show that neither the “novel foam,” nor the “standard foam,” was associated with any observable local or systemic reaction, or signs of allergy/sensitivity. Comparative findings
The US FDA concerns surrounding patient mortality and morbidity as complications of NPWT were focused in their released warnings of 2009 and 2011.3,4 The issues raised were bleeding, foam retention, and infection—phenomena often insignificant and frequently seen when using NPWT. However, certain population groups are at enhanced risk of serious morbidity, or death, as a result. Bleeding
Patients on anticoagulant therapy (heparin/fractionated heparin, warfarin, etc.), blood “thinners” (aspirin, dipyridam210
Wagstaff et al.
ole, etc.), or those with a bleeding diathesis (hemophilia, etc.) are considered “at risk” of bleeding problems with NPWT. Granufoam (“standard foam”) generated the majority of such incidences.4 The structure of “standard foam” facilitates the generation of vascular granulation tissue from the cavity walls and supports marginal integration of the foam structure, and this phenomenon seems to be much more prevalent in deep cavity wounds. Foam removal at dressing changes in such wounds was frequently difficult (80% in our series), with the foam having to be traumatically peeled away from the friable granulations, resulting in bleeding. Such blood loss is usually small and self-limiting. In this small cohort, “novel foam” interface wounds did not show any bleeding after foam removal. The ease of removal suggests that marginal invasion of the foam did not occur and was thus “atraumatic” on separation. The nature of the “novel foam” allows greater compression of the marginal foam cells under the negative pressure and may discourage ingrowth (at least over the maximal 72 hours of foam implantation in this study). Infection
The microbiological analysis of the predebrided wounds, debrided tissue, and wound cavities during NPWT revealed a wide array of organisms, many pathological. Their presence on swabbing, however, even in great numbers, was not frequently translated into clinical infection (either local or systemic), or a delay in wound healing/cavity size reduction. Infection did, however, frequently follow “standard foam” retention. This is defined as foam fragments becoming embedded in the wound cavity wall where they appear to act as a nidus for bacterial infection and is reported as the major mechanism of wound infection using NPWT.4 Apparent retention of “novel foam” was seen in only one patient (foam fragments visualized against the wall of the cavity after dressing removal); however, this “retention” was washed easily out of the wound with saline irrigation, without trauma or the need for sharp debridement. Infection did not follow this episode. The retention issue is intimately related to the bleeding issue, namely a problem of marginally integrated foam removal. Preventing marginal integration of the foam by cavity wall granulations appears to solve both issues. Handling and ease of (duration of) dressing changes
In the experience of both surgeon and nurse, “novel foam” was more difficult to shape. This reduced with familiarity. In the community, however, nursing time is precious and this would seem to be a negative attribute. However, 12 of the 20 wounds in this study were ischial and of the six dressed with “standard foam,” foam removal was difficult (and thus time consuming) in 80% of dressing changes. Foam retention and the need for sharp debridement augmented the time taken for dressing changes, and infection required further clinical (medical) intervention and prescription. The results suggest that the “novel foam” interface might be better suited to wounds where the full extent of the cavity cannot be easily visualized during dressing changes. This occurs where there are very large cavities (such as ischial ulcers), significant undermining of wound edges, or the presence of narrow “tunnelled” wounds or wound components. Additionally (although not addressed directly), it may provide Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
Wagstaff et al.
NPWT in pressure sores with new polyurethane foam compared to control foam
a safer interface where patients requiring NPWT are concurrently being treated with anticoagulants or blood thinners, or are immunosuppressed or immunocompromised and thus at risk of life-threatening infection originating in the wound as a result of foam retention. This is a small-scale trial primarily investigating the introduction of a novel polyurethane foam in a patient group with impaired sensation. Our experience is that early NPWT dressing changes in acute wounds with “standard foam” are painful, because of marginal ingrowth and traumatic foam removal. Given that the removal of “novel foam” was less traumatic, it may be less painful than “standard foam” in acute and highly sensate wounds. Differences in pain on dressing changes in acute wounds treated with NPWT would be an avenue for further study in a larger scale study.
ACKNOWLEDGMENTS The authors would like to thank Sue Templeton and Prof. Debbie Kralik of the Adelaide Royal District Nursing Service, who were pivotal in establishing the collaboration between hospital and community that allowed the study to proceed. Thanks also go to Valerie Cohn, Meredith Sullivan, and Brenton Sutton of KCI Australia for provision of pumps, control foams, and other accessories. Thomas Sullivan, Senior Statistician within the Data Management & Analysis Centre, Discipline of Public Health of the University of Adelaide, performed all the independent statistical analysis. Finally, thanks to the staff of the Adult Burns Unit, Royal Adelaide Hospital, for looking after these patients during their debridement hospital admission. Source of Funding: Funding for this study was provided by PolyNovo Biomaterials Pty Ltd, Port Melbourne, Victoria, Australia, who provided the treatment materials (NovoPore) and paid the RDNS to backfill the position of the study nurse. By virtue of the gratis supply of pumps, dressings and control foam, Kinetic Concepts Incorporated, San Antonio, Texas, additionally supported the trial. Conflict of Interest: A/Prof. John Edward Greenwood, AM, owns a 20% share of NovoWound Pty. Ltd. through his company, Skin Pty. Ltd. NovoWound Pty. Ltd. is a joint venture company established to explore commercialization opportunities of the NovoSorb polymers (including NovoPore) in the field of wound management. He receives no salary or other payment from this collaboration. Dr. Marcus James Dermot Wagstaff has a minor shareholding in Calzada
Wound Rep Reg (2014) 22 205–211 © 2014 by the Wound Healing Society
Pty Ltd, the Australian Stock Market (ASX)-listed parent company of PolyNovo Biomaterials Pty Ltd. Neither remaining authors have any conflict of interest.
REFERENCES 1. Morykwas MJ, Argenta LC, Shelton-Brown EI, McGuirt W. Vacuum-assisted closure: a new method of wound control and treatment: animal studies and basic foundation. Ann Plast Surg 1997; 38: 553–62. 2. Argenta LC, Morykwas MJ. Vacuum-assisted closure: a new method for wound control and treatment: clinical experience. Ann Plast Surg 1997; 38: 563–77. 3. FDA Preliminary Public Health Notification on Negative Pressure Wound Therapy Systems. November 13, 2009. Available at http://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ PublicHealthNotifications/ucm190658.htm (accessed May 1, 2013). 4. FDA Safety Communication. UPDATE on serious complications associated with negative pressure wound therapy systems. February 24, 2011. Available at http://www.fda.gov/MedicalDevices/ Safety/AlertsandNotices/ucm244211.htm (accessed May 1, 2013). 5. Li A, Dearman BL, Crompton KE, Moore TG, Greenwood JE. Evaluation of a novel biodegradable polymer for the generation of a dermal matrix. J Burn Care Res 2009; 30: 717–28. 6. Greenwood JE, Li A, Dearman B, Moore TG. Evaluation of NovoSorb™ novel biodegradable polymer for the generation of a dermal matrix. Part 1: in-vitro studies. Wound Pract Res 2010; 18: 14–22. 7. Greenwood JE, Li A, Dearman B, Moore TG. Evaluation of NovoSorb™ novel biodegradable polymer for the generation of a dermal matrix. Part 2: in-vivo studies. Wound Pract Res 2010; 18: 24–34. 8. Greenwood JE, Dearman BL. Split-skin graft application over an integrating, biodegradable temporising polymer matrix: immediate and delayed. J Burn Care Res 2012; 33: 7–19. 9. Greenwood JE, Dearman BL. Comparison of a sealed, polymer foam biodegradable temporising matrix against Integra™ dermal regeneration template in a porcine wound model. J Burn Care Res 2012; 33: 163–73. 10. Dearman BL, Stefani K, Li A, Greenwood JE. Take of a polymer-based autologous cultured composite “skin” on an integrated temporising dermal matrix: proof of concept. J Burn Care Res 2013; 34: 151–60.
211
