Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) Review
The debridement of chronic wounds: a systematic review
M Bradley N Cullum T Sheldon
Health Technology Assessment NHS R&D HTA Programme HTA HTA
How to obtain copies of this and other HTA Programme reports. An electronic version of this publication, in Adobe Acrobat format, is available for downloading free of charge for personal use from the HTA website (http://www.hta.ac.uk). A fully searchable CD-ROM is also available (see below). Printed copies of HTA monographs cost £20 each (post and packing free in the UK) to both public and private sector purchasers from our Despatch Agents. Non-UK purchasers will have to pay a small fee for post and packing. For European countries the cost is £2 per monograph and for the rest of the world £3 per monograph. You can order HTA monographs from our Despatch Agents: – fax (with credit card or official purchase order) – post (with credit card or official purchase order or cheque) – phone during office hours (credit card only). Additionally the HTA website allows you either to pay securely by credit card or to print out your order and then post or fax it.
Contact details are as follows: HTA Despatch Email: [email protected] c/o Direct Mail Works Ltd Tel: 02392 492 000 4 Oakwood Business Centre Fax: 02392 478 555 Downley, HAVANT PO9 2NP, UK Fax from outside the UK: +44 2392 478 555 NHS libraries can subscribe free of charge. Public libraries can subscribe at a very reduced cost of £100 for each volume (normally comprising 30–40 titles). The commercial subscription rate is £300 per volume. Please see our website for details. Subscriptions can only be purchased for the current or forthcoming volume.
Payment methods Paying by cheque If you pay by cheque, the cheque must be in pounds sterling, made payable to Direct Mail Works Ltd and drawn on a bank with a UK address. Paying by credit card The following cards are accepted by phone, fax, post or via the website ordering pages: Delta, Eurocard, Mastercard, Solo, Switch and Visa. We advise against sending credit card details in a plain email. Paying by official purchase order You can post or fax these, but they must be from public bodies (i.e. NHS or universities) within the UK. We cannot at present accept purchase orders from commercial companies or from outside the UK.
How do I get a copy of HTA on CD? Please use the form on the HTA website (www.hta.ac.uk/htacd.htm). Or contact Direct Mail Works (see contact details above) by email, post, fax or phone. HTA on CD is currently free of charge worldwide.
The website also provides information about the HTA Programme and lists the membership of the various committees. The debridement of chronic wounds: a systematic review
M Bradley1* N Cullum2 T Sheldon3
1 NHS Centre for Reviews and Dissemination, University of York,UK 2 Centre for Evidenced Based Nursing, University of York,UK 3 York Health Policy Group, University of York,UK
* Corresponding author
Competing interests: none declared
Published September 1999
This report should be referenced as follows:
Bradley M, Cullum N, Sheldon T. The debridement of chronic wounds: a systematic review. Health Technol Assess 1999;3(17 Pt 1).
Health Technology Assessment is indexed in Index Medicus/MEDLINE and Excerpta Medica/ EMBASE. Copies of the Executive Summaries are available from the NCCHTA web site (see overleaf). NHS R&D HTA Programme
he overall aim of the NHS R&D Health Technology Assessment (HTA) programme is to T ensure that high-quality research information on the costs, effectiveness and broader impact of health technologies is produced in the most efficient way for those who use, manage and work in the NHS. Research is undertaken in those areas where the evidence will lead to the greatest benefits to patients, either through improved patient outcomes or the most efficient use of NHS resources. The Standing Group on Health Technology advises on national priorities for health technology assessment. Six advisory panels assist the Standing Group in identifying and prioritising projects. These priorities are then considered by the HTA Commissioning Board supported by the National Coordinating Centre for HTA (NCCHTA). This report is one of a series covering acute care, diagnostics and imaging, methodology, pharmaceuticals, population screening, and primary and community care. It was identified as a priority by the Pharmaceutical Panel and funded as project number 93/29/01. The views expressed in this publication are those of the authors and not necessarily those of the Standing Group, the Commissioning Board, the Panel members or the Department of Health. The editors wish to emphasise that funding and publication of this research by the NHS should not be taken as implicit support for the recommendations for policy contained herein. In particular, policy options in the area of screening will be considered by the National Screening Committee. This Committee, chaired by the Chief Medical Officer, will take into account the views expressed here, further available evidence and other relevant considerations. Reviews in Health Technology Assessment are termed ‘systematic’ when the account of the search, appraisal and synthesis methods (to minimise biases and random errors) would, in theory, permit the replication of the review by others.
Series Editors: Andrew Stevens, Ruairidh Milne and Ken Stein Editorial Assistant: Melanie Corris
The editors have tried to ensure the accuracy of this report but cannot accept responsibility for any errors or omissions. They would like to thank the referees for their constructive comments on the draft document.
ISSN 1366-5278
© Crown copyright 1999
Enquiries relating to copyright should be addressed to the NCCHTA (see address given below).
Published by Core Research, Alton, on behalf of the NCCHTA. Printed on acid-free paper in the UK by The Basingstoke Press, Basingstoke.
Copies of this report can be obtained from:
The National Coordinating Centre for Health Technology Assessment, Mailpoint 728, Boldrewood, University of Southampton, Southampton, SO16 7PX, UK. Fax: +44 (0) 1703 595 639 Email: [email protected] http://www.hta.nhsweb.nhs.uk Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Contents
List of abbreviations ...... i Acknowledgements ...... 25
Executive summary ...... iii References ...... 27
1 Introduction ...... 1 Appendix 1 MEDLINE search strategy ...... 31 Mechanical debridement ...... 1 Non-mechanical debridement ...... 2 Appendix 2 CINAHL and EMBASE 2 Methods ...... 3 search strategy ...... 33 Literature search ...... 3 Study selection and data extraction ...... 3 Appendix 3 Additional databases Data synthesis ...... 3 searched ...... 35 Assessment of outcome measures ...... 3 Appendix 4 Expert advisory panel ...... 37 3 Results ...... 7 Debriding agents versus traditional or Appendix 5 Quality assessment ...... 39 control therapy ...... 7 Comparisons between debriding agents ...... 9 Appendix 6 Summary of studies ...... 41 Cost-effectiveness ...... 9
4 Discussion ...... 19 Health Technology Assessment reports published to date ...... 71 5 Conclusions ...... 23 Implications for policy ...... 23 Health Technology Assessment Recommendations for research ...... 23 panel membership ...... 75
Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
List of abbreviations
C comparator*
I intervention*
CI confidence interval
CRD Centre for Reviews and Dissemination
ES effect size
F female*
HTA Health Technology Assessment
ITT intention-to-treat
M male*
NS not significant*
OR odds ratio
RCT randomised controlled trial
SD standard deviation*
SEM standard error of the mean*
* Used only in tables and figures
i
Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Executive summary
Background Interventions The following interventions were identified as A wide variety of debridement methods and agents that would be used primarily for wound products are available, all of which have diverse debridement. properties, costs and levels of acceptability. There is currently wide variation in their use and a lack of • Dextranomer polysaccharide beads or paste consensus on how to treat specific wound types. • Cadexomer iodine polysaccharide beads or paste • Hydrogels Objectives • Enzymatic agents • Adhesive zinc oxide tape • To summarise the evidence for the relative • Surgery or sharp debridement effectiveness and cost-effectiveness of different • Larval (maggot) therapy. debriding agents on wound healing. • To identify areas for future research. Other interventions that are believed to have a debriding function, such as hydrocolloid dressings and antibiotics/antiseptics, were not included in Methods this review as debridement was not the primary reason for their application. Data sources A range of electronic databases and several wound No RCTs were found that evaluated the effective- care journals were searched; organisations, manu- ness of surgical debridement, larval therapy, or that facturers, researchers and healthcare professionals compared debridement with no debridement. concerned with wound care were contacted for additional trials. The reference sections of obtained Dextranomer polysaccharide beads or paste studies were also searched for further trials. versus traditional or control treatment Nine trials met the inclusion criteria, five of which Inclusion criteria found a statistically significant difference between Studies were considered for inclusion if they were treatments: three favoured dextranomer poly- randomised controlled trials (RCTs), published or saccharide, and two favoured traditional treatment. unpublished, that assessed the effectiveness of a recognised debriding agent as identified by an Cadexomer iodine polysaccharide versus expert panel, and assessed patients with chronic traditional or control treatment non-healing wounds (pressure sores, leg ulcers, Nine trials met the inclusion criteria, of which sinuses and surgical wounds healing by secondary three had a statistically significant result that intention). Studies were only included if they used a favoured cadexomer iodine polysaccharide. quantifiable and objective measure of healing rate. Hydrogels versus traditional or control treatment Data synthesis Only one trial out of four that compared a For each trial an odds ratio and/or effect size was cal- hydrogel with a traditional or control treatment culated for all objective outcomes. Where possible the found a statistically significant difference between analysis was performed on an intention-to-treat basis, treatments, which suggested a small benefit and 95% confidence intervals were included when suf- from treatment with a hydrogel dressing compared ficient detail to allow their calculation was provided. with a hydrocolloid dressing.
Enzymatic agents versus traditional or Results control treatment None of the five trials in this category showed a Forty-seven reports describing 35 RCTs were statistically significant outcome in favour of either identified that met the inclusion criteria. treatment for wound closure. In fact, one trial iii Executive summary
showed an increase in wound size with both the unclear whether wound debridement is a beneficial enzyme collagenase and the control treatment; process that expedites healing. however, the increase was significantly less in the enzyme-treated group. There is insufficient evidence to promote the use of one debriding agent over another. There Adhesive zinc oxide tape versus traditional was only a single comparison between two treatment debriding agents that produced a significant A single trial meeting the inclusion criteria result (hydrogel significantly reduced necrotic showed that adhesive zinc oxide tape was more wound area compared with dextranomer effective in eradicating or reducing by more than polysaccharide paste). 50% the necrotic area of diabetic foot ulcers than a hydrocolloid dressing. Implications for policy There is little evidence to identify which agents Cadexomer iodine polysaccharide versus other are the most effective. Pending the availability of debriding agents improved data on relative effectiveness, other Two trials were comparisons with dextranomer considerations, such as cost-minimisation, may polysaccharide and one trial was a comparison reasonably guide decisions on the use of with a hydrogel. None of the trials had statistically debriding agents. significant results. Recommendations for research Dextranomer versus other debriding agents Much of the research is of poor quality, and Four RCTs comparing dextranomer polysaccharide direct comparisons are few. In the trials reviewed, with another debriding agent included two trials sample sizes were rarely sufficient to detect with enzyme formulations, namely collagenase and clinically important effects, and poor baseline streptokinase/streptodornase, and two trials using comparability frequently confounded outcome a hydrogel as the comparator. Only one of the two measures. Several important messages can be comparisons with a hydrogel showed a statistically identified for future studies. significant benefit associated with the hydrogel. • Recruitment numbers should be based on a Hydrogel versus hydrogel sample size calculation. A single trial was found. There was no statistically • The proportion of wounds healed should be significant difference between the two treatments. used as an objective outcome measure. Where healing rates are based on wound area, both the Enzymatic agent versus enzymatic agent percentage and absolute change in area should One trial that compared the enzyme preparation be given. streptokinase/streptodornase with the enzyme • Experimental groups should be comparable trypsin was included in the review. No statistically at baseline. significant difference between the two treatments • Baseline data and intervention details should was found. always include a thorough description of how the patients were nursed and report the use Cost-effectiveness of concurrent treatments, including Cost-effectiveness has not been thoroughly assessed secondary dressings. in studies of debriding agents. The unit cost for • Comparisons between debriding agents each treatment is stated in some studies, and a few are required and should use agents that contain further details on other important vari- are recommended for wounds of a ables, such as nursing time or number of dressing similar nature. changes. However, no study provides sufficient • Assessment should be blind to treatment. detail from which a reliable cost-effectiveness • Survival rate analysis should be adopted for analysis can be constructed. all studies that assess wound healing. • All RCTs should be published. • Detailed cost-effectiveness analyses should be Conclusions seen as a priority for future trials. • The frequent use of surgical debridement No studies were found that compared debridement and the increasing interest in larval therapy with no debridement and without these studies it is indicate that RCTs in these areas are needed. iv Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Chapter 1 Introduction
ound healing is an efficient and natural TABLE 1 Classification of debriding agents2 W process that normally requires no special treatment. However, chronic non-healing wounds Mechanical Non-mechanical can occur when there is some underlying factor debridement debridement preventing healing, and in such cases intervention Surgery (scalpel) Polysaccharide beads or is considered necessary. For example, pressure paste (dextranomer, sores are initially acute wounds caused by ischaemic Wet-to-dry dressings cadexomer iodine) death of tissue due to excessive pressure and will (saline gauze) usually heal readily when pressure is relieved and Enzymatic agents the blood supply restored. However, in certain Bio-surgery (maggot larvae) (streptokinase/ patients, particularly the elderly and persons with streptodornase, low mobility, it may not be easy to resolve these trypsin, collagenase) causative factors and chronic wounds can develop. Hydrogels
A characteristic of most chronic wounds is an accumulation of devitalised tissue and cellular Mechanical debridement exudate at the outer surface. These products result from a restriction of nutrients to the damaged Surgical or sharp debridement has been practised epithelium and form either a dry, hard eschar or, for several centuries and is still commonplace as in the case of deep moist wounds, a slough that today. The technique is simple, requiring only frequently hardens on the outside with exposure the use of sterile scissors or a scalpel, but it does to the air. The accumulation of these products require some degree of skill to avoid aggravating in the wound bed is generally regarded, though the wound. It is considered that sharp debridement not experimentally proven, to prevent or delay is best suited to wounds where there is a clear granulation and epithelialisation. The removal distinction between healthy and devitalised tissue of this tissue by a process termed debridement is and where a patient is unlikely to experience therefore thought to facilitate healing.1 significant pain.2
Although the clinical evidence in support of Bio-surgery, which exploits the feeding behaviour debridement is lacking, treatment is regarded of certain insect larvae, has an equally long medical as a necessity for patient acceptability and for history and is presently enjoying a resurgence of the prevention of infection:2 moist slough is interest.3 Sterile maggots (fly larvae) are considered malodorous and visually repugnant, and provides ideal debriders, having a ferocious appetite for an ideal culture medium for pathological organ- necrotic material while actively avoiding newly isms. The treatment of infection with topical formed healthy tissue.4 Maggots are placed directly prophylactic antibiotics is often unsuccessful onto the affected area and held in place by a close- in sloughy wounds as the active agent is unable net-dressing.5 These animals may also perform an to penetrate and diffuse through the necrotic antimicrobial role that is beneficial for wound debris. Debridement will remove this barrier healing, as the antibacterial agent allotoxin is and may thereby assist in the treatment actively excreted from the maggot’s body.2 Evi- of infection. dence for clinical effectiveness is restricted to case reports and expert opinion; it remains untested Although many products used during the treat- by randomised controlled trials (RCTs). ment of a wound may have a debriding function, only a few are commonly used specifically for Wet-to-dry debridement is used infrequently in the this purpose. Despite having diverse properties, UK but is common in many European countries costs and levels of acceptability, these specific and the USA. The wound is soaked in saline to debriding agents can broadly be classified as moisten hard material before the application of mechanical or non-mechanical interventions a moist gauze pad over the affected area. As the (Table 1 ). devitalised tissue drys it re-hardens and becomes 1 Introduction
attached to the gauze, when the dressing is hydrophilic in nature. The beads are made from a changed the adhered material is pulled free.6 modified starch infused with iodine at a concentra- This method is less discriminating than those tion of 0.9%. Absorption of fluid from the wound previously discussed, removing both healthy and results in a slow controlled displacement of iodine granulating tissue from the wound bed. Patients from the matrix, which acts as a bactericidal agent.10 are known to experience pain, and newly formed The slow and consistent release of iodine overcomes epithelial tissue may be significantly damaged.7 the problem of iodine inactivation by protein ab- sorption in the wound. The antibacterial property, biodegradability and high rate of fluid absorption Non-mechanical debridement distinguish cadexomer iodine from dextranomer.
Non-mechanical techniques have become Hydrogels are a group of agents that were pri- increasingly popular for wound cleansing. These marily developed as debriding agents. These gels treatments are usually easy to apply and may have are biologically inert and have a significant water additional properties that are beneficial for wound content. They complement the body’s natural healing. Such interventions include enzymes, debriding process by providing an advantageous hydrogels and specific chemical formulations, environment for autolysis, while still acting to such as cadexomer iodine and dextranomer. preserve living healthy tissue.2 The hydrogel is usually applied directly into the wound bed and Several different enzyme preparations are held in place by a non-adherent dressing. Once available that digest slough and necrotic tissue. the gel is fully hydrated it is unable to absorb the In the UK, only the formulation containing copious quantities of exudate that are released by streptokinase and streptodornase (Varidase some wounds. For this reason hydrogels are often Topical®, Wyeth Laboratories) is licensed for used in conjunction with a highly absorbent dress- use.8 This enzyme aggressively digests the proteins ing. In addition to the amorphous gel, hydrogels fibrin, collagen and elastin which are commonly are also available in a sheet form. Several types found in the necrotic exudate of a wound.9 of hydrogel are available manufactured under Other enzymatic debriding agents are available different trade names (Intrasite® Gel, Smith & and used internationally; these include trypsin Nephew Healthcare Ltd; Sterigel®, Seton Health- and collagenase. care Group plc; Granugel®, CovaTec UK Ltd).
Dextranomer polysaccharide is supplied as In clinical practice there is wide variation in the anhydrous, porous beads with a diameter of use of debriding agents and no consensus on which 0.1–0.3 mm or as a paste. The beads are highly agent is most appropriate for each type of wound.11 hydrophilic and rapidly absorb exudate from a As a result, many clinicians combine debriding necrotic sloughy mass. Prostaglandins, hormones agents in a dual or triple therapy in an attempt and other relatively small molecules enter the to maximise healing rates, but such measures are matrix of the beads, while larger particles such as untested and are not endorsed by manufacturers. bacteria and wound debris become concentrated This review, commissioned by the NHS Health at the surface of the dextranomer layer. When Technology Assessment (HTA) Programme, the beads are changed by washing with saline attempts to summarise the evidence to date for the absorbed and trapped necrotic material the relative effectiveness and cost-effectiveness of is removed.10 these different interventions and to identify the gaps in our research knowledge. In this way it is Cadexomer iodine is similar to dextranomer, hoped that clinical practice may better reflect consisting of small spherical beads that are the results of scientific research.
2 Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Chapter 2 Methods
Literature search the following interventions as debriding agents: dextranomer polysaccharide, cadexomer iodine A systematic review of primary research was polysaccharide, hydrogels, various enzymatic undertaken using the NHS Centre for Reviews agents, adhesive zinc oxide tape, surgery and and Dissemination (CRD) structured guidelines.12 larval therapy. Other interventions such as hydro- Nineteen electronic databases of medical literature colloid dressings and antibiotics/antiseptics were were searched up to October 1997 using a sensitive not considered by the panel to be debriding agents search strategy designed in collaboration with an as debridement may occur only as a secondary information specialist in CRD (appendices 1–3). function following their application rather than This was supplemented by a handsearch of five being the primary reason for their use. specialist wound care journals, twelve conference proceedings and systematic reviews held on the Data from included trials were extracted by a NHS CRD Database of Abstracts of Reviews of single reviewer into data extraction tables and then Effectiveness (DARE). The bibliographies of all checked independently by a second reviewer. retrieved and relevant publications were searched for further studies. Companies with an interest in wound care products were approached for un- Data synthesis reported trials. An advisory panel of experts in wound management, established to comment For each trial an odds ratio (OR) and/or effect on the review as it progressed, were also asked to size (ES) was calculated for all objective outcomes. identify any additional trials (appendix 4). Rele- Where possible the analysis was performed on an vant economic studies were identified by adding intention-to-treat (ITT) basis and 95% confidence economic-related search terms to those used in the intervals (CIs) were included when sufficient detail search for clinical trials. Authors of trials published to allow their calculation was provided. The results after 1985 were contacted and asked to provide from each study were plotted onto graphs, and details of any trial-based economic evaluations. individual study details are presented in structured tables containing information relevant to study validity. Heterogeneity between studies indicated Study selection and that pooling of results would be inappropriate. data extraction Studies varied in duration of follow-up, nature of the comparator treatment, wound types, setting Retrieved studies were assessed by a single reviewer and baseline comparability of patient groups. for relevance to this review, and decisions on final inclusion were checked by a second reviewer; disagreements were resolved through discussion. Assessment of outcome measures Trials, irrespective of date, language and publi- cation status, were included if they were human- A single standard outcome measure for wound based RCTs, which evaluated the efficacy of a healing does not exist. Both objective and subjec- recognised debriding agent in relation to wound tive measures are widely used by researchers but healing and had an outcome measure that was little effort has been made to determine the considered an objective measure of healing. validity of these measurements. Where study details were lacking the authors were invited to provide further information. Most subjective outcomes, such as estimates of oedema, erythema, granulation, pus and debris, Many treatments commonly used in wound are unlikely to be measured consistently between management may have a debriding function; wounds. Unless assessment is blinded to treatment only a few, however, are specifically used for this allocation, subjective outcomes are likely to result purpose. An expert panel was consulted to identify in significant biases. For example, in one trial those treatments that they believed were primarily cadexomer iodine was judged by visual assessment used for debridement. The panel identified to be more effective than dextranomer, but a later 3 Methods
assessment of quantitative planimetric results TABLE 2 Percentage and absolute measures of wound healing indicated that there was no difference between the can give different results for relative effectiveness two treatments.13 This appears to be a characteristic feature of research in this area; subjective measure- Group A Group B ments almost invariably overestimate the relative Baseline mean area (cm2)5060 effectiveness of the experimental treatment com- pared with objective measures in the same trial. Follow-up mean area (cm2)3543
Objective measures of debridement are usually Mean of % reduction in area 30 28 based on wound area. Planimetry, often aided by Mean absolute reduction in computer assessment, is the most frequently used area (cm2)1517 tool for calculating wound area, though other methods, such as the measurement of wound Using % change, wounds in group A appear to have healed diameter or the weight of a celluloid tracing drawn more rapidly than those in group B. The converse is true when around the area of the wound, are also used. the outcome is expressed as absolute change in wound area
Measurements of wound volume are infrequently reported in the literature. These methods are often TABLE 3 Groups with similar means may have different distributions cumbersome and their accuracy has not been suffi- ciently demonstrated.14 Computerised image ana- Group C Group D lysis may in the future, as the equipment becomes more affordable and portable, prove to be a useful Wound size at baseline (cm2) 10, 10, 30, 30 4, 4, 4, 70 technique for the assessment of wound volume.15 Mean area (cm2) 20 20.5
However, even though objective measures reduce Standard deviation (SD) 11.5 33 or eliminate subjective biases and reduce random measurement errors, they have certain inherent Groups C and D have approximately the same mean biases if the patients being compared have wounds area. If both groups of wounds heal at the same rate (the with different baseline size. treatments are equally effective) it could be expected that the three small wounds in group D will heal before those in group A change in wound area is often expressed as the C.Therefore measuring outcome based on the number of wounds healed within a certain time period will be biased. percentage change, which unlike the absolute Similarly, the percentage change in area will appear greater change in area, takes into account the initial size in group D, while the absolute change in area will appear of the wound. For two wounds healing at the same greater in group C linear rate (as measured by diameter reduction), percentage area calculations will show a larger change for a small wound than a big wound. The outcome is based on the change in area, the result converse is true when the absolute change in area can only be considered valid if it is obtained either is measured, as for any unit reduction in wound against the anticipated direction of the bias for radius a bigger area reduction will occur for a large wound size, or where percentage area change and wound. This has important consequences for the absolute area change are in the same direction. If validity of trial results, where there is poor com- baseline data are not given then it is not possible to parability in initial wound size at baseline between determine the direction of bias and the validity of the treatment groups. This is illustrated in Table 2. the result cannot be assessed.
In large trials, random allocation should ensure The change in area of a wound also assumes a that the mean wound size and variance in each linear rate of healing or at least a predictable rate. group are similar. In a small trial, random allo- However, this may not always be the case, with some cation is unlikely to result in an even distribution wounds enlarging prior to healing while others of wound sizes. This problem will persist in small initially decrease rapidly in size before experi- trials, even when the average wound size appears encing a slower rate of healing. For this reason to be comparable between groups, because the complete healing is seen as the most valuable distribution of wound sizes about the mean is likely outcome in studies of wound healing. Unfortun- to differ. This is illustrated in Table 3. ately, to follow all wounds to complete healing can require an extended period of follow-up that is not In a trial where there is poor comparability ideal for most studies, where financial resources 4 between groups for wound size at baseline, and the and time are limited. For this reason many studies Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
in the literature report changes in wound size size at baseline or by the variability of healing rather than complete healing. rate between wounds, they remain the most reliable assessment of wound healing as they Despite the potential for the objective outcomes reduce the bias of the assessor, which cannot discussed to be biased by differences in wound be estimated.
5
Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Chapter 3 Results
total of 47 RCTs met the review inclusion Debriding agents versus A criteria. Twelve studies (26%) had a shared traditional or control therapy data set with another publication, and of these: two contained interim results later published Twenty-eight trials compared a recognised in full,16,17 three were reproduced as papers debriding agent with a traditional or control presented at international conferences,18–20 one therapy. The definition and nature of traditional was presented at an international conference therapy varied widely between trials. In many trials, and later published as a drug company bulletin,21 a single traditional treatment was not used in the six studies were repeat publications of results comparative group, in extreme cases a different previously published in other journals,22–27 two treatment regimen was used for each patient of which22,23 were identical to a previously in the control group. The absence of a single published paper,28 except for a change in the standard comparator makes interpretation first author. Where duplication occurred and of the results across studies difficult. sufficient detail could not be extracted from a single publication, the studies were pooled and Dextranomer polysaccharide versus incorporated to represent a single trial entry. traditional or control treatment In total, 35 individual trials from the 47 Nine RCTs met the inclusion criteria (Figure 1; relevant publications were identified appendix 6, Table 5). Four trials showed a for inclusion. statistically significant result, three of which favoured treatment with dextranomer poly- The majority of trials had methodological saccharide,29–31 and one which favoured weaknesses (appendix 5, Table 4 ). Fewer than treatment with calcium alginate dressings.32 10% of studies reported an a priori estimate In an additional trial, the author’s own statis- of the number of participants required to have tical analysis indicated a significant improve- sufficient power to detect a clinically important ment for those wounds treated with a collagen effect as statistically significant. It was rare sponge dressing when compared with for a trial to have more than 50 patients in dextranomer.33 However, insufficient data each arm, and typically fewer than 30 were were available to test the analysis further. included. Twenty-six per cent reported that The remaining four trials had statistically the outcome had been assessed by someone insignificant results.34–37 blind to treatment allocation. Appropriate patient characteristics were recorded by Trials of dextranomer polysaccharide versus treatment group in 90% of studies, but ulcer traditional or control treatment can be grouped size at baseline was reported in only 62%. into categories which are defined by the nature Withdrawals occurred in most trials and were of the comparative therapy. recorded by group and cause in 72% of trials where it was appropriate, but less than one Dextranomer polysaccharide versus Eusol fifth performed an ITT analysis. Seventy-four Two small studies compared dextranomer poly- per cent of trials clearly stated the inclusion saccharide with Eusol; both were unable to detect criteria for patients to the trial, but information a statistical difference between the treatments.35,36 that indicated whether or not that participants The study by Nasar and Morley36 was confounded had been truly randomised to alternative by the switching of treatment for three wounds treatments was given in only 37%. that were considered to be healing slowly from the Eusol group to the dextranomer group. Figures 1–7 show the OR and/or ES calculated These patients were excluded from the analysis from the results for each of the included studies. because of the bias that was hence introduced. Where possible, ITT analysis has been used; In this same study, the cost of materials for where such analysis was inappropriate the average treatment time was calculated, and sample size used in the calculation is given showed that Eusol was 1.6 times more costly in parentheses. than dextranomer. 7 Results
Dextranomer polysaccharide versus saline soaks iodine polysaccharide when compared with Saline soaks were evaluated against dextranomer hydrogen peroxide and zinc paste,38 hydrogen polysaccharide in two trials, which were both peroxide or potassium permanganate,22,23,28 conducted over a 2-week period.31,34 and mechanical debridement by wet-to-dry dressings.39 However, high withdrawal rates22,23,28 The trial by Eriksson and co-workers34 found no and inappropriate movement of some patients statistically significant difference between the between groups during the trial39 may have treatments for the percentage of ulcers healed by introduced confounding factors, which 50% or 25%. The trial by Sawyer and co-workers,31 would have implications for the validity however, found a statistically significant effect in of these results. favour of dextranomer beads for the healing of venous leg ulcers. Despite the absence of baseline In two trials the authors own statistical analysis data in this latter trial it is unlikely that such a indicated that cadexomer iodine resulted in large difference in healing rates between the two a significant reduction in wound size when treatments could be explained merely in terms of compared to traditional dressing regimes.40,41 poor group matching at baseline. However, in both studies the papers provided insufficient information for CIs to be calcu- Dextranomer polysaccharide versus other lated which prevented further analysis in traditional or control therapies this review. One study found that dextranomer poly- saccharide beads were significantly more The remaining trials comparing a traditional effective than povidone-iodine for the treat- treatment with cadexomer iodine polysaccharide ment of venous leg ulcers.29 However, high were unable to detect a statistically significant dropout rates were reported, which could effect in favour of either treatment.42–45 This have introduced a bias. Dextranomer poly- may have been a consequence of small sample saccharide beads have also been found to be size or a reflection of the diversity of traditional more effective when compared with antiseptic/ therapies evaluated in a single trial. In one trial, antibiotic treatments.30 This is in contrast to 12 of the 13 patients in the traditional group a small study comparing dextranomer poly- were treated with a variant on the basic saccharide beads with an air dried combination standard therapy.41 of sugar and egg white, which did not find a statistically significant difference between the Hydrogels versus traditional or two treatments.37 control treatment Four trials were included that compared a Dextranomer polysaccharide paste was out- hydrogel with a traditional or control treatment performed by calcium alginate dressings for (Figure 3; appendix 6, Table 7 ).46–49 Only one of the healing of pressure sores over an 8-week these trials showed a statistically significant differ- period.32 These results were statistically signifi- ence between treatments. In this comparison a cant despite a bias that favoured the smaller hydrogel dressing showed a small improvement sores in the dextranomer polysaccharide in the number of wounds healed when compared group. with a hydrocolloid dressing.46
Insufficient data were provided to determine Enzyme treatments versus traditional or the CIs required to test statistical significance control treatment in a study that compared dextranomer beads Only one of the five trials included in the review with a collagen sponge dressing.33 However, the showed a statistically significant difference be- author’s own analysis suggested a statistically tween an enzymatic debriding agent (collagenase) significant reduction in healing time occurred and a control treatment (Figure 4; appendix 6, with the collagen sponge dressings. Table 8 ).50 However, the wounds in this trial actually increased in size and the predicted Cadexomer iodine polysaccharide versus bias was in the direction of enzymatic treatment. traditional or control treatment Despite the majority of trials showing a trend in Nine RCTs compared cadexomer iodine favour of enzyme treatment, statistical significance polysaccharide with a traditional or control could not be demonstrated. All the trials in this therapy (Figure 2; appendix 6, Table 6 ). Three category suffered from small sample size;9,37,51,52 trials (five papers) had a statistically significant only one of the included studies had more 8 result favouring treatment with cadexomer than 15 patients in each arm.52 Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Adhesive zinc oxide tape versus the treatments used in both studies appear to traditional treatment have been similar, it is likely that the formulation In a single trial an adhesive zinc oxide tape was of Intrasite used by Colin was a modification of compared with a hydrocolloid dressing for the that employed by Thomas and Fear. Interestingly, treatment of necrotic diabetic foot ulcers (appen- in the latter study, treatment was continued beyond dix 6, Table 9 ).53 The adhesive zinc oxide tape was the 14-day follow-up period in a selected group of more effective in eradicating or reducing necrotic patients. After 28 days, a further four wounds were area than the hydrocolloid dressing (OR = 4.44; fully cleansed in the dextranomer polysaccharide 95% CI: 1.34, 14.70). Treatment was discontinued group, compared with no additional successes in in nine patients (four from the adhesive zinc oxide the hydrogel group. This may suggest that over a tape group and five from the hydrocolloid dressing longer period the results in this trial would have group) due to a significant increase in necrotic reflected those reported by Colin and co-workers. area (> 100%). Common adverse effects seen in both groups were maceration of skin edges, A non-significant benefit was found in favour pain and oedema. of dextranomer polysaccharide when compared with the enzyme collagenase, this benefit was recorded against the direction of wound size Comparisons between bias.37 In the only other trial to compare dextra- debriding agents nomer polysaccharide with an enzyme preparation (streptokinase/streptodornase), no difference Nine trials were identified that directly compared between the treatments was found.58 two debriding agents. Seven trials compared either dextranomer polysaccharide or cadexomer iodine Hydrogel versus hydrogel polysaccharide with another debriding agent, one Only one study made a direct comparison be- trial compared two enzyme treatments, and a tween two hydrogels (Intrasite Gel and Granugel) further trial was a comparison between hydrogels. (Figure 7; appendix 6, Table 12 ).59 This trial found no statistically significant difference between the Cadexomer iodine polysaccharide groups after 21 days treatment. The total cost for versus other debriding agents treatment per person from baseline to final assess- Three RCTs were included that compared ment was calculated to be £43.25 for Granugel cadexomer iodine polysaccharide with another and £40.25 for Intrasite, while the average cost debriding agent (Figure 5; appendix 6, Table 10 ). per 10 cm2 reduction in area was £8.59 and £13.38, Dextranomer polysaccharide13,54 was the com- respectively. The data appear to indicate that treat- parator in two trials and a hydrogel was used ment with Granugel may be more cost-effective in the third trial.55 None of the trials showed a than with Intrasite. However, this is misleading as statistically significant effect, and in two the point the analysis is based on absolute changes in ulcer estimate of relative effectiveness was close to size. The Granugel group had larger ulcers at the line of no effect.13,55 baseline, suggesting that this group was more likely to experience larger changes in absolute ulcer area. Dextranomer polysaccharide versus other debriding agents Enzyme versus enzyme Four RCTs were included that compared dextra- Only one RCT directly compared two enzymatic nomer polysaccharide with another debriding agents (Figure 8; appendix 6, Table 13 ).60 No agent (Figure 6; appendix 6, Table 11 ). Two trials statistically significant difference was found were comparisons with enzyme formulations between wounds treated with streptokinase/ (collagenase or streptokinase/streptodornase), streptodornase and those treated with trypsin. and two were comparisons with a hydrogel.
Both hydrogel trials attempted to record Cost-effectiveness debridement directly by employing an outcome measure based on wound cleansing (the reduction Cost-effectiveness has not been thoroughly assessed in area of necrotic tissue covering the wound in studies of debriding agents. The unit cost for each bed).56,57 The study by Colin and co-workers56 treatment is stated in some studies, and a few con- failed to find a statistically significant effect for tain further details on other important variables, 100% cleansing. However, the smaller study by such as nursing time or number of dressing changes. Thomas and Fear57 found a large statistically sig- However, no study provides sufficient detail for a nificant effect in favour of the hydrogel. Although reliable cost-effectiveness analysis to be constructed. 9 Results
Goode et al, 197935 Comparison: Traditional vs. Dextranomer Mean baseline area (cm2): Not stated Not stated Sample size (N): 10 10 No. of wounds healed without secondary closure (n/N): 1/10 (10%) 1/10 (10%) 1.00 (0.05, 18.32) Traditional therapy: Eusol Wound type: Surgical wounds (bowel or appendicectomy) Follow-up time: Until healed or ready for secondary suture
Nasar and Morley, 198236 Comparison: Traditional vs. Dextranomer Mean baseline area (cm2): Not stated Not stated Sample size (N): 9 9 No. of wounds reduced by > 25%, clean and granulating (n/N): 5/9 (56%) 6/9 (67%) 1.6 (0.24, 10.81) Traditional therapy: Eusol Wound type: Pressure sores Follow-up time: Until wound was clean and granulating
Sawyer et al, 197931 Comparison: Traditional vs. Dextranomer Mean baseline area (cm2): Not stated Not stated Sample size (N): 31 35 No. of wounds healed (n/N): 3/19 (16%) 15/18 (83%) 26.70 (4.639, 153.29) Traditional therapy: Saline soaks, or mechanical debridement where indicated Wound type: Venous leg ulcers Follow-up time: 3 weeks
Eriksson et al, 198434 Comparison: Traditional vs. Dextranomer Mean baseline area (cm2): Not stated Not stated Sample size (N): 27 26 No. of wounds reduced by > 50% (n/N): 6/27 (19%) 2/26 (8%) 0.29 (0.05, 1.60) No. of wounds reduced by 25% or more (n/N): 21/27 (78%) 24/26 (92%) 3.43 (0.62, 18.85) Traditional therapy: Saline soaks Wound type: Venous leg ulcers Follow-up time: 2 weeks
Groenewald, 198029 Comparison: Traditional vs. Dextranomer Baseline area (cm2): < 6 cm2: 7 ulcers 7 ulcers 6–12 cm2: 21 ulcers 20 ulcers > 12 cm2: 22 ulcers 23 ulcers Sample size (N): 50 50 No. of wounds healed (n/N): 2/50 (4%) 11/50 (22%) 6.77 (1.42, 32.38) No. of wounds reduced by 50–100% (n/N): 7/50 (14%) 24/50 (48%) 5.67 (2.14, 15.00) Traditional therapy: Povidone-iodine, zinc oxide, and gauze Wound type: Venous leg ulcers Follow-up time: 3 weeks
OR 0.01 0.1 1 10 100 (95% CI) ES –40 –20 0 20 40 (95% CI) Favours alternative Favours Dextranomer
FIGURE 1 Dextranomer polysaccharide compared with traditional or control treatments. Study results are presented as OR and/or ES enclosed by their 95% CI. An arrow by a trial indicates that the results were biased by poor comparability between groups for wound size at baseline; the direction of the arrow suggests which intervention was favoured by this bias. Convergent arrows suggest that the groups were reasonably comparable for wound size, while divergent arrows indicate that the bias could not be determined from the 10 data presented Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Sayag et al, 199632 Comparison: Traditional vs. Dextranomer Mean baseline area (cm2): 20.1 (SD 12.9) 16.1 (SD 12.5) Sample size (N): 47 45 No. of wounds reduced by 40–100% (n/N): 35/47 (74%) 19/45 (42%) 0.25 (0.10, 0.61) No. of wounds reduced by > 75% (n/N): 15/47 (32%) 6/45 (13%) 0.33 (0.11, 0.94) Traditional therapy: Calcium alginate dressing Wound type: Pressure sores Follow-up time: Until the wound reached 40% of its original size or on completion of 8 weeks’ treatment
Garcia et al, 198430 Comparison: Traditional vs. Dextranomer Mean baseline area (cm2): Not stated Not stated Sample size (N): 22 22 Mean % reduction in wound area: 85.2 93.1 7.9 (2.27, 15.53) (SD 10.8; n = 22) (SD 7.4; n = 22) Mean % reduction in wound circumference: 70.1 80.0 9.9 (–2.70,22.50) (SD 20.8; n = 22) (SD 20.6; n = 22) Traditional therapy: Potassium permanganate and hypertonic saline or antibiotic ointments Wound type: Secondary perimalleolar ulcers Follow-up time: 3 weeks
Parish and Collins, 197937 Comparison: Control vs. Dextranomer Mean baseline area (cm2): 5.76 20.25 Sample size (N): 9 14 No. of wounds healed (n/N): 0/9 (0%) 6/14 (43%) 14.53 (0.71, 298.23) Traditional therapy: Sugar and egg white Wound type: Pressure sores Follow-up time: 4 weeks
Palmieri, 199233 Comparison: Traditional vs. Dextranomer Mean baseline area (cm2): Not stated Not stated Sample size (N): 24 24 Mean no. of days to healing for leg ulcers: 36 (n = 6) 60 (n = 6) –24 Mean no. of days to healing for pressure sores: 20 (n = 6) 47 (n = 6) –27 Traditional therapy: Collagen sponge Wound type: Leg ulcers and pressure sores Follow-up time: Until all wounds had healed
OR 0.01 0.1 1 10 100 (95% CI) ES –40 –20 0 20 40 (95% CI) Favours alternative Favours Dextranomer
FIGURE 1 contd Dextranomer polysaccharide compared with traditional or control treatments. Study results are presented as OR and/or ES enclosed by their 95% CI. An arrow by a trial indicates that the results were biased by poor comparability between groups for wound size at baseline; the direction of the arrow suggests which intervention was favoured by this bias. Convergent arrows suggest that the groups were reasonably comparable for wound size, while divergent arrows indicate that the bias could not be determined from the data presented
11 Results
Apelqvist and Ragnarson-Tennvall, 199642 Comparison: Traditional vs. Cadexomer iodine Mean baseline area (cm2): Not stated Not stated Sample size (N): 19 22 No. of wounds healed (n/N): 2/19 (11%) 5/22 (23%) 2.5 (0.42, 14.72) Traditional therapy: Gentamicin or streptokinase/streptodornase; dry saline Wound type: Diabetic foot ulcers Follow-up time: 12 weeks
Laudanska and Gustavson, 198838 Comparison: Traditional vs. Cadexomer iodine Mean baseline area (cm2): 35.2 (SE 8.1) 27.5 (SE 7.0) Sample size (N): 30 30 No. of wounds healed (n/N): 7/30 (23%) 16/30 (53%) 3.76 (1.24, 11.39) Mean % reduction in wound area: 54 (n = 30) 71 (n = 30) 17 Traditional therapy: Hydrogen peroxide and zinc paste; or saline dressings, potassium permanganate and gentian violet Wound type: Chronic venous leg ulcers Follow-up time: 6 weeks
Ormiston et al, 198544 Comparison: Traditional vs. Cadexomer iodine Mean baseline area (cm2): 10.2 (SD 8.7) 12.1 (SD 13.9) Sample size (N): 30 31 No. of wounds healed (n/N) 7/30 (23%) 12/31 (39%) 2.08 (0.68, 6.32) Mean % reduction in wound area: 61 (n = 30) 83 (n = 30) 22 Mean absolute reduction in wound area per week: 0.46 0.89 0.43 (SEM 0.1; n = 30) (SEM 0.1; n = 30) Traditional therapy: Gentian violet, Polyfax® ointment and a Melolin® pad Wound type: Chronic venous leg ulcers Follow-up time: 12 weeks
Steele et al, 198645 Comparison: Traditional vs. Cadexomer iodine Mean baseline area (cm2): 17.59 (SE 3.97) 12.64 (SE 2.91) Sample size (N): 30 30 No. of wounds healed (n/N): 1/30 (3%) 3/30 (10%) 3.22 (0.32, 32.91) Mean % reduction in wound area: 18 (n = 29) 22 (n = 28) 4 Traditional therapy: Antibiotics, antiseptics, hydrophilic agents, bland agents, steroids, dry dressings Wound type: Venous leg ulcers Follow-up time: 6 weeks
Harcup and Saul, 198640 Comparison: Traditional vs. Cadexomer iodine Baseline area (cm2): 9.08 (SD 1.37) 7.74 (SD 1.04) Sample size (N): 31 41 Mean absolute reduction in wound area (cm2): 0.9 (n = 31) 2.81 (n = 41) 1.91 Mean % reduction in wound area: 9.91 (n = 31) 36.30 (n = 41) 26.39 Traditional therapy: Support bandage, dry dressing and an antiseptic cleanser (i.e. Eusol) Wound type: Chronic leg ulcers Follow-up time: 4 weeks
Lindsay et al, 198641 Comparison: Traditional vs. Cadexomer iodine Mean baseline area (cm2): Not stated Not stated Sample size (N): 28 patients entered the trial; data available for only 25 Mean % reduction in wound area: 4.2 (n = 13) 33.6 (n = 12) 29.4 Traditional therapy: Non-adherent dressings and a support bandage or one of several alternative treatments Wound type: Chronic venous leg ulcers OR Follow-up time: 4 weeks 0.01 0.1 1 10 100 (95% CI) ES –60 –30 0 30 60 (95% CI) ES Polyfax®, Dominion; –1.0 –0.5 0 0.5 1.0 (95% CI) Melonin®, Smith & Nephew Healthcare Ltd Favours alternative Favours cadexomer iodine
12 FIGURE 2 Cadexomer iodine polysaccharide compared with traditional or control treatments Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Moberg et al, 198343 Comparison: Traditional vs. Cadexomer iodine Mean baseline area (cm2): 12.4 (SEM 4.3) 9.6 (SEM 1.8) Sample size (N): 19 19 Mean absolute reduction in wound area (cm2): 2.5 2.9 0.4 (–3.05,3.85) (SEM 1.1; n = 18) (SEM 1.3; n = 16) Mean % reduction in wound area: 9.6 30.9 21 (–5.83, 48.43) (SD 31; n = 18) (SD 46; n = 16) Traditional therapy: Saline dressings, enzymatic agents, non-adhesive dressings Wound type: Pressure sores Follow-up time: 3 weeks
Skog et al, 1983;23 Hillström, 1988;22 Troëng et al, 198328 Comparison: Control vs. Cadexomer iodine Mean baseline area (cm2): 34.0 (SEM 5.7) 20.1 (SEM 4.4) Sample size (N): 45 50 Mean % change in wound area: +5 –34 39 (8.16, 69.84) (SEM 15; n = 36) (SEM 5; n = 38) Traditional therapy: Dilute hydrogen peroxide or potassium permanganate and a non-adherent dressing Wound type: Chronic venous leg ulcers Follow-up time: 6 weeks
Holloway et al, 198939 Comparison: Traditional vs. Cadexomer iodine Mean baseline area (cm2): 11.2 20.1 Median baseline area (cm2): 9.75 (range, 3–37) 10.7 (range, 0.6–136.0) Sample size (N): 37 38 Mean absolute reduction in wound area per week (cm2): 0.41 0.95 0.54 (0.19,0.90) (SEM 0.13; n = 27) (SEM 0.12; n = 27) Mean reduction in wound area per week vs. baseline circumference (cm2/wk/cm2): 0.03 0.04 0.01 (–0.02,0.04) (SEM 0.01; n = 27) (SEM 0.01; n = 27) Traditional therapy: Wet-to-dry dressings and saline-soaked gauze Wound type: Venous leg ulcers Follow-up time: 24 weeks OR 0.01 0.1 1 10 100 (95% CI) ES –60 –30 0 30 60 (95% CI) ES –1.0 –0.5 0 0.5 1.0 (95% CI) Favours alternative Favours cadexomer iodine
FIGURE 2 contd Cadexomer iodine polysaccharide compared with traditional or control treatments
13 Results
Darkovich et al, 199046 Comparison: Traditional vs. Hydrogel (Biofilm®) Mean baseline area (cm2): 9.2 (range, 0.4–64) 11.0 (range, 0.2–100) Sample size: 63 60 No. of wounds healed (n/N): 15/63 (24%) 26/60 (43%) 2.45 (1.13, 5.30) Mean % reduction in wound area: 40 68 28 Mean absolute reduction in wound area (cm2): 3.7 (n = 63) 7.5 (n = 60) 3.8 Traditional therapy: DuoDerm® Wound type: Pressure sores Follow-up time: 60 days
Brown-Etris et al, 199647 Comparison: Traditional vs. Hydrogel (Transorbent®) Mean baseline area (cm2): Not stated Not stated Sample size: 63 77 No. of wounds healed (n/N)37/63 (59%) 39/77 (51%) 0.72 (0.37, 1.41) Mean absolute reduction in wound area: stage II sores (2–30 cm2): 2.3 (n = 12) 3.6 (n = 12) 1.3 stage III sores (2–30 cm2): 5.2 (n = 36) 6.3 (n = 42) 1.1 stage III sores (31–80 cm2): 4.3 (n = 2) 24.5 (n = 3) 20.2 Traditional therapy: DuoDerm Wound type: Pressure sores Follow-up time: 10 weeks
Lum et al, 199648 Comparison: Traditional vs. Hydrogel (Intrasite) Mean baseline area (cm2): 18.6 36.3 (range, 1.0–72.5) (range, 1.5–160) Sample size (patients): 3 5 Mean absolute change reduction in wound area (cm2): +1.9 (n = 3) –20.4 (n = 5) 22.3 Mean % reduction in wound area: +10 (n = 3) –56 (n = 5) 66 Traditional therapy: Saline alone or with eusol where the wound was infected Wound type: Pressure sores Follow-up time: 8 weeks
Mulder et al, 199349 Comparison: Control vs. Hydrogel (Clearsite®) Mean baseline area (cm2): Not stated Not stated Sample size (N): 21 20 Mean % reduction in wound area per week (cm2): 5.1 8 2.9 (–6.45,12.25) (SD 14.8; n = 21) (SD 14.8; n = 20) Median % reduction in wound area per week (cm2): 7.0 (n = 21) 5.6 (n = 20) –1.4 Control treatment: Saline soaks Wound type: Pressure sores Follow-up time: 8 weeks
Comparison: Traditional vs. Hydrogel (Clearsite) Mean baseline area (cm2): Not stated Not stated Sample size (N): 20 20 Mean % reduction in wound area per week (cm2) 3.3 8 4.7 (–11.55,20.95) (SD 32.7; n = 20) (SD 14.8; n = 20) Median % reduction in wound area per week (cm2) 7.4 (n = 20) 5.6 (n = 20) –1.8 Traditional therapy: DuoDerm Wound type: Pressure sores Follow-up time: 8 weeks OR 0.01 0.1 1 10 100 (95% CI) ES –100 –50 0 50 100 (95% CI) Biofilm®, Braun Biotrol; Transorbent®, Braun Medical; ES Duoderm®, UK equivalent Granuflex, ConvaTec Ltd; 10 5 0 5 10 (95% CI) Clearsite®, New Dimensions in Medicine Favours alternative Favours hydrogel
14 FIGURE 3 Hydrogels compared with traditional or control treatments Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Westerhof et al, 198752 Comparison: Control vs. Enzyme (Elase®) Mean baseline area (cm2): 20.2 (range, 2–48) 25.7 (range, 2–273) Sample size (N): 17 16 No. of wounds sucessfully grafted or healed (n/N): 6/17 (35%) 11/16 (69%) 4.03 (0.94, 17.23) Mean baseline area (cm2): 22.3 (range, 3–48) 14.9 (range, 2–48) Sample size (N): 8 6 No. of complex wounds successfully grafted or healed (n/N): 3/8 (38%) 3/6 (50%) 1.67 (0.20, 14.27) Mean baseline area (cm2): 18.1 (range, 2–40) 36.4 (range, 3–273) Sample size (N): 9 10 No. of venous ulcers successfully grafted or healed (n/N): 3/9 (33%) 8/10 (80%) 8.00 (1.00, 64.00) Control: Placebo (freeze-dried powder, no active ingredients) Wound type: Leg ulcers Follow-up time: Until ulcers ready for grafting
Lee and Ambrus, 197550 Comparison: Control vs. Enzyme (Santyl®) Mean baseline volume (cm3): 1.25 (SD 1.62) 15.44 (SD 19.92) Sample size (N): 11 17 Mean % increase in wound volume: 13.14 78.79 65.65 (1.72, 129.58) (SD 49.8; n = 11) (SD 94.6; n = 17) Control therapy: Placebo (heat-inactivated enzyme) Wound type: Pressure sores Follow-up time: 4 weeks
Gordon, 197551 Comparison: Control vs. Enzyme (Chymacort®) Mean baseline area (cm2): 14.5 (SD 22.13) 15.82 (SD 19.67) Sample size (N): 9 10 No. of wounds healed (n/N): 0/9 (0%) 2/10 (20%) 5.59 (0.23, 133.71) Mean absolute reduction in ulcer area (cm2): 6.1 10.5 4.40 (–9.92,18.72) (SD 13.2; n = 8) (SD 15.0; n = 10) Mean % change in wound size +47 –71.8 118.80 (–50.20, 287.50) (SD 252.8; n = 8) (SD 22.5; n = 10) Control therapy: Hydrocortisone and neomycin palitate Wound type: Leg ulcers Follow-up time: 6 weeks
Ågren and Strömberg, 19859 Comparison: Traditional vs. Enzyme (Varidase) Mean baseline area (cm2): 5.8 4.2 (range, 1.2–26.0) (range, 1.2–18.2) Sample size (N): 14 14 Median % change in wound area: –2.4 (n = 14) +18.7 (n = 11) –21.1 Traditional therapy: Zinc oxide dressing Wound type: Pressure sores Follow-up time: 8 weeks
Parish and Collins, 197937 Comparison: Traditional vs. Enzyme (Santyl) Mean baseline area (cm2): 5.76 10.24 Sample size (n/N): 9 11 No. of wounds healed (n/N): 0/9 (0%) 1/11 (9%) 2.71 (0.098, 75.05) Traditional therapy: Sugar and egg white Wound type: Pressure sores Follow-up time: 4 weeks
OR 0.01 0.1 1 10 100 (95% CI) ES Elase®, manufacturer unknown; –200 –100 0 100 200 (95% CI) Santyl®, Knoll Pharmaceutical Co; Chymacort®, Armour Pharmaceutical Company Ltd Favours alternative Favours enzyme
FIGURE 4 Enzymatic agents compared with traditional or control treatments 15 Results
Tarvainen, 198854 Comparison: Dextranomer vs. Cadexomer iodine Mean baseline area (cm2): Not stated Not stated Sample size (N): 13 14 No. of wounds healed (n/N): 2/13 (15%) 7/14 (50%) 5.50 (0.88, 34.48) Wound type: Leg ulcers Follow-up: 8 weeks
Moss et al, 198713 Comparison: Dextranomer vs. Cadexomer iodine Mean baseline area (cm2): 25.5 (SD 29.5) 19.7 (SD 19.8) Sample size (N): 21 21 Mean % reduction in wound area: 3 (CI: 4, –9; n = 21) 4 (CI: 5, –14; n = 21) 1 (–10.28, 12.28) Wound type: Venous leg ulcers Follow-up: 6 weeks
Stewart and Leaper, 198755 Comparison: Hydrogel (Scherisorb®) vs. Cadexomer iodine Median baseline area (cm2): 2.6 2.8 (range, 0.4–74.4) (range, 0.2–50.5) for range Sample size (N): 49 46 No. of wounds healed (n/N): 14/49 (29%) 14/46 (30%) 1.09 (0.45, 2.64) Wound type: Leg ulcers of mixed aetiology Follow-up: 10 weeks for median
OR 0.1 0.2 1 5 10 (95% CI) ES –20 –10 0 10 20 (95% CI)
Scherisorb®, presently Intrasite Gel Favours alternative Favours cadexomer iodine
FIGURE 5 Cadexomer iodine polysaccharide compared with other debriding agents
16 Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Colin et al, 199656 Comparison: Hydrogel (Intrasite) vs. Dextranomer Baseline area: 4 cm2: 15 ulcers 18 ulcers 4–13 cm2: 25 ulcers 25 ulcers > 13 cm2: 27 ulcers 25 ulcers Sample size (N): 67 68 No. of wounds 100% cleansed (n/N): 13/67 (19%) 14/68 (21%) 1.08 (0.46, 2.51) No. of wounds 75–100% cleansed (n/N): 33/67 (49%) 29/68 (43%) 0.77 (0.39, 1.51) Median % reduction in wound area: 35 7 –28 (range,–185–+91; (range, –340–+98; n = 53) n = 43) Median % reduction in non-viable tissue: 74 (range, –103–+100; 62 (range, –340–+100; –12 n = 53) n = 43) Wound type: Pressure sores Follow-up time: 3 weeks
Thomas and Fear, 199357 Comparison: Hydrogel (Intrasite) vs. Dextranomer Mean baseline (cm2): 22.2 (SD 23.4) 15.6 (SD 16.2) Sample size (N): 21 20 No. of wounds fully cleansed (n/N): 8/20 (40%) 1/20 (5%) 0.08 (0.01, 0.71) Wound type: Pressure sores Follow-up time: 2 weeks
Parish and Collins, 197937 Comparison: Enzyme (Santyl) vs. Dextranomer Mean baseline (cm2): 10.24 20.25 Sample size (N): 11 14 No. of wounds healed (n/N): 1/11 (9%) 6/14 (43%) 7.50 (0.74, 75.72) Wound type: Pressure sores Follow-up time: 4 weeks
Hulkko et al, 198158 Comparison: Enzyme (Varidase) vs. Dextranomer Mean baseline (mm): Not stated Not stated Sample size (N): 13 18 Mean % reduction in wound diameter: 11.9 (n = 12) 14.5 (n = 15) 2.6 Mean absolute reduction in wound diameter (mm): 9.8 (n = 12) 9.3 (n = 15) 0.5 Wound type: Venous ulcers Follow-up time: 2 weeks
OR 0.01 0.1 1 10 100 (95% CI) ES –30 –15 0 15 30 (95% CI)
Favours alternative Favours dextranomer
FIGURE 6 Dextranomer polysaccharide compared with other debriding agents
17 Results
Gibson, 199559 Comparison: Hydrogel (Intrasite) vs. Hydrogel (Granugel) Mean baseline area (cm2): 18.7 24.9 Sample size (N): 32 30 Mean % reduction in wound area after 21 days: 10 (n = 32) 14 (n = 30) 4 Mean % reduction in slough area after 21 days: 32 (n = 32) 37 (n = 30) 5 Median absolute reduction in wound area (cm2) after 7 days: 0.54 (n = 32) 0.74 (n = 30) 0.2 Median absolute reduction in slough area after 7 days (cm2): 1.12 (n = 32) 1.51 (n = 30) 0.4 Wound type: Leg ulcers Follow-up time: 3 weeks
OR 0.01 0.1 1 10 100 (95% CI) ES –10 –5 0 5 10 (95% CI)
Favours Intrasite Favours Granugel
FIGURE 7 Comparison between hydrogels
Hellgren, 198360 Comparison: Enzyme (Trypure®) vs. Enzyme (Varidase) Mean baseline area (cm2): Not stated Not stated Sample size (N): 19 21 No. of wounds fully cleansed (n/N): 5/19 (26%) 13/21 (62%) 3.58 (0.90, 14.16) Mean % granulated area at 14 days: 56 (n = 16) 60 (n = 21) 4 Wound type: Leg ulcers Follow-up time: 3 weeks
OR 0.01 0.1 1 10 100 (95% CI) ES –10 –5 0 5 10 (95% CI)
Trypure®, Novo Laboratories Favours Trypure Favours Varidase
FIGURE 8 Comparison between enzyme treatments
18 Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Chapter 4 Discussion
here are no RCTs comparing debridement The underlying bias is frequently in the direction T with no debridement. A limited number of of the treatment effect, suggesting that the study trials have employed an inert or placebo com- results should be viewed cautiously. A statistically parison, but the extent to which even secondary significant study result favouring one treatment dressings contribute to wound healing is unclear. over another can only be considered valid if it is Given the multi-faceted way that topical debriding obtained either: against the anticipated direction agents are thought to work, the obvious study to of the bias for wound size, or where percentage answer the question of whether debridement is area change and absolute area change are in the a necessity for wound healing would be a com- same direction. When baseline data are not given parative evaluation of surgical, or sharp for wound size it is not possible to estimate the debridement with no debridement. degree of bias, and in the absence of other indicators, validity of the trial result should The majority of trials investigated debridement be considered doubtful. of either pressure sores or venous ulcers and for these wounds the debriding agent generally showed Analysis of studies where patients received a a benefit over traditional or control treatments. traditional or control therapy in the comparison However, without conclusive evidence that debride- arm indicates that the validity of the outcome ment is an important procedure for wound healing, measure, according to baseline comparability, it is not possible to determine whether the benefit was doubtful in about 70% of the trials. In the over traditional treatment is a true assessment of remaining 30%, where wound size bias was effectiveness, or a consequence of harm induced unlikely to have affected the outcome measure, by the traditional therapies. the OR or ES was in the direction of the debriding agent in over 88% of cases, though only two studies Although evidence in the form of RCTs is lacking, obtained statistical significance. In other words many clinicians believe that debridement facilitates those studies with the most reliable results indicate wound closure by removing necrotic tissue that acts that treatment with a known debriding agent is as a barrier to new tissue growth. This suggests that favourable to traditional or control therapy. if debridement really does aid wound closure then the effectiveness of a debriding agent should be Analysis by intervention type suggests that both measured by an outcome based on wound healing. dextranomer polysaccharide and cadexomer iodine Even though the debriding agent is not necessarily polysaccharide may improve wound healing when used throughout the entire healing process an compared with a traditional or control therapy. outcome measure based on healing remains valid However, evidence for the use of hydrogels or as long as both comparison groups follow a similar enzymatic agents in preference to traditional or schedule of nursing care after the debridement control treatment is less convincing as, with the period. In this way, any difference in healing rates exception of one trial,46 these debriding agents between groups can be attributed to the debriding did not show a statistically significant improvement agent used. Some researchers however, have in wound closure. The lack of evidence for the attempted to estimate the effectiveness of these effectiveness of enzymatic agents suggests that agents by measuring the degree of debridement careful consideration should be given to the expressed as the percentage area of wound covered continued use of streptokinase/streptodornase, in necrotic material. This measurement may not given that its topical application is known to induce be a reliable indication of treatment effect as the the production of anti-streptokinase antibodies, extent of debridement does not appear to have which could theoretically put patients at risk if been scientifically validated as a surrogate or given streptokinase immediately following a proxy measure of wound healing. myocardial infarction.61
The results reported in many of the included In drawing these conclusions on effectiveness studies are biased by a lack of comparability it is necessary to bear in mind that many of the between the groups for wound size at baseline. traditional treatments employed as comparators 19 Discussion
are no longer used in clinical practice, and some with a single debriding agent. Although it could may even have a detrimental effect on healing. This be argued that this reflects typical clinical practice implies that relative effectiveness, as recorded in it makes interpretation of the results very difficult. many of the trials, may be an over-estimate of what It is likely that within such a varied group of we might expect to obtain in a similar trial today. traditional treatments there will be considerable variation in effectiveness. It is possible also that Recently, there have been several non-systematic one of these treatments may be equally, if not narrative reviews in nursing-orientated publi- more effective, than the debriding agent under cations, which have suggested that hydrogels evaluation, but that its beneficial effect is masked are the non-mechanical debridement agent of by the other less effective treatments within the choice.2,62 In terms of scientific evidence, such group. It is also possible that treatment care may a statement can only be supported by trials that vary depending on the care givers familiarity with directly compare two or more debriding agents; each of the traditional treatments employed. such trials, however, are relatively rare in the This is less likely to be an issue in the application literature and are generally of poor quality. Using of a homogeneous debriding agent. the criteria previously described, around 40% of comparisons between debriding agents had a valid Different debriding agents may be more appro- result, but only one trial found a statistically signifi- priate for different types of wound because the cant difference between treatments.57 The small mode of action employed by each agent to facili- number of reliable trials and the diversity of com- tate wound healing differs. For example, two parisons implies that there is insufficient evidence studies compared the same hydrogel with dextra- to promote the use of one debriding agent over nomer polysaccharide for the treatment of pres- another on the basis of effectiveness alone. sure sores.56,57 Dextranomer polysaccharide is thought to actively absorb water and necrotic In wound care research it is not uncommon material from the wound bed, whereas the for trials to be associated with funding from the hydrogel is believed to liquefy the wound and pharmaceutical industry. This may in part explain maintain a moist environment that facilitates the lack of direct comparisons between different, auto-debridement.57 Clearly these two agents commercially important debriding agents. A work in very different ways and their effectiveness pharmaceutical company is unlikely to be willing is likely to depend on the type of wound to which to compare its product with that of a competitor they are applied. Dextranomer polysaccharide is if the product has only shown marginal benefits then recommended for the treatment of moist, over traditional or control treatments. yellow sloughy wounds, while a hydrogel, although a generalist agent, is recommended particularly A pooled estimate of effectiveness for each debrid- for dry necrotic wounds.10 Comparisons between ing agent has not been calculated; to do so would debriding agents should then take into account be inappropriate considering the heterogeneous the local nature of the wound under examination nature of the studies and the potential biases due when assessing effectiveness. to differences in wound size at baseline. Pooling would compound these biases and result in an over- The categories used in this review to group estimate of effectiveness, particularly considering debridement interventions could be misleading the likely influence of publication bias favouring because they may mask pharmaceutical develop- positive outcomes. If similar trials reported both ments. For example, the formulation of Intrasite absolute and percentage reductions in ulcer size Gel has changed over time, and that used in a trial then a pooled estimate could be calculated for each in the late 1980s is unlikely to be the same as that of the outcomes, which would provide a reasonable used in the early 1990s. The vehicle used for a estimate of effectiveness. Unfortunately, this is rarely treatment may also vary. For example, dextranomer the case and where sufficient details are provided polysaccharide can be applied as either beads or as there are other differences between studies, such a paste with polyethylene glycol, and the nature of as wound type, follow-up time, the presence of this vehicle may have consequences for effective- other underlying conditions such as ischaemia, ness. Considering all dextranomer polysaccharide and the variability of traditional treatments within trials together does not take into account these a single comparison group which make combining potential differences. However, only by grouping results inappropriate. these interventions under broad categories can a general overview of effectiveness be gained, and A multiplicity of traditional therapies are frequently as the data are not statistically pooled erroneous 20 employed in the same trial to act as a comparator interpretations are avoided. In addition, both Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
clinicians and patients alike are primarily not been validated and were likely to result in concerned with whether a treatment works misleading information. rather than its mode of action. If the subtle differences identified between products in the Larval (maggot) therapy has had a long history, laboratory do not translate into improved wound and is enjoying a resurgence of interest in the healing rates in clinical practice, then there is UK.4 However, evidence for effectiveness remains little benefit in discussing differences in restricted to case reports and anecdotal articles; product formulation. there is presently no research evidence to support claims of effectiveness. Considering the high Quality assessment indicates that most studies profile of this intervention, clinical trials, partic- included in the review have methodological flaws ularly RCTs, are urgently required. that affect their validity. In general, the RCTs are too small to ensure that wounds of different sizes Publication bias appears to be present in debride- are equally allocated to both experimental groups. ment studies with the majority of trials favouring This suggests that most trials will be subject to a the debriding agent under evaluation. This may be bias at baseline. Most trials had a short follow-up, a reflection of commercial interests in many of the which mitigates against the outcome of the number included trials. Even if a debriding agent was effec- of wounds healed. Studies that employ such an tive, one would expect by chance to get occasional outcome measure should analyse the data using results from studies that found in favour of the ‘survival analysis’, which takes into account both alternative treatment, particularly because of the whether and when a wound healed; in this way it is small sample size of the RCTs, giving more a more efficient method for estimating the rate of opportunity for chance variation. healing. Many of the studies also excluded sick, debilitated or confused patients and those with Future trials that assess wound healing need to diseases that affect healing, such as diabetes and be larger in terms of patient numbers than the rheumatoid arthritis. These are the very patients majority of those reviewed here. The design of whose wounds tend to be chronic and are there- smaller trials could be improved by the use of fore the most likely to be prescribed debriding matched or stratified randomisation to ensure agents in practice. In general then, the patients comparability between treatment groups at enrolled into a trial are a highly selected group baseline: patients should be stratified based on and are unlikely to be representative of the those wound type and size at baseline and these group- encountered in clinical practice. ings should then be randomised to treatment. If patients have been matched for wound size and Subjective outcome measures such as degrees type then matched pairs analysis should be used to of oedema, erythema, granulation, pus and debris analyse the data. But even with this type of design are frequently recorded by authors, but are not the trial still needs to be large enough to ensure considered in this review. These outcomes are comparability for other features and confounding more susceptible to bias as the assessor may have factors, such as age or duration of the wound. an unconscious preference for one of the treat- ments. Analysis of subjective outcomes nearly Planned and future studies would also benefit always produced a statistically significant result from the inclusion of a detailed cost-effectiveness in favour of the experimental treatment. Similarly, analysis. Several reports included in this review other variables that are likely to influence treat- detailed the costs incurred during the trial period, ment choice such as pain, comfort and quality but unfortunately none of the studies provided of life issues, are not addressed in this review sufficient detail for a reliable cost-effectiveness because the outcome measures employed had analysis to be constructed.
21
Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Chapter 5 Conclusions
here is some, if limited, evidence to suggest • Recruitment numbers should be based on an T that the use of a specific debriding agent is a priori sample size calculation, though this may beneficial for wound healing when compared be complicated by the lack of trials on which to with certain traditional or control treatments. base such a calculation. In contrast, there is little or no evidence to suggest • The proportion of wounds healed should be that one debriding agent is more effective than used as an objective outcome measure. Where any other. healing rates are based on wound area both the percentage and absolute change in area should be given. Implications for policy • Experimental groups should be comparable at baseline. In small RCTs, randomisation alone As the benefits shown over traditional or control will not achieve comparability; in such situations treatments are at best marginal, the choice of patients should be paired by baseline character- which debriding process to adopt should perhaps istics and then the individuals of each pair ran- be based on the relative cost of the treatments, domised to treatment. However, with particularly unless clinical experience suggests otherwise. small groups, pairing of patients is still unlikely There is little reliable research evidence to to ensure that all risk factors (known and un- suggest that the more expensive agents should known) are distributed equally between the be used in preference to some of the less costly groups, and hence, large sample size should ones. However, this decision should not be based still be the ultimate goal. on the net cost of each treatment as this does not • Baseline data and intervention details should take into account important variables, such as the always include a thorough description of number of dressing changes required, coverage how the patients were nursed and the use of the wound area, or nursing time. The financial of concurrent treatments including cost of a treatment can only be determined from secondary dressings. studies designed specifically to address this issue • Comparisons between debriding agents are and this should be seen as a priority for required and should use agents that are future research. recommended for wounds of a similar nature. • Assessment should be blind to treatment. • Survival rate analysis should be adopted for Recommendations for research all studies that assess wound healing. • All RCTs should be published. Much of the research into wound debridement • Detailed cost-effectiveness analyses should be is of poor quality and direct comparisons are few. seen as a priority for future trials. In those trials reviewed, sample sizes were rarely • RCTs of both surgical and larval debridement sufficient to detect clinically important effects, have not been performed. The frequent use and poor baseline comparability frequently of surgical debridement and the increasing confounded outcome measures. Several important interest in the larval therapy indicate that messages can be identified for future studies. these trials are needed.
23
Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Acknowledgements
his review was commissioned by the NHS to our expert panel (appendix 4) for their T R&D HTA programme. The authors are helpful comments on the review protocol indebted to Julie Glanville and Alison Fletcher and draft report. The report also benefited for assistance with the search, location and greatly from comments received from the collection of literature. We are also grateful HTA referees.
25
Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
References
1. US Department of Health and Human Services. 16. Fear M, Thomas S. Wound cleansing properties Treatment of pressure ulcers. Clinical Practice of Debrisan and Scherisorb gel: interim results Guidelines 1994;15:47. of a clinical trial. Proceedings of the 1st European Conference on Advances in Wound Management. 2. Bale S. A guide to wound debridement. J Wound London: Macmillan Ltd, 1992;1:162–4. Care 1997;6(4):179–82. 17. Sawyer PN, Dowbak G, Sophie Z, Feller J, Cohen L. 3. Thomas S, Jones M, Shutler S, Jones S. Using larvae A preliminary report of the efficacy of Debrisan in modern wound management. J Wound Care (dextranomer) in the debridement of cutaneous 1996;5(2 Suppl):60–9. ulcers. Surgery 1979;85(2):201–4. 4. Sherman RA, Wyle F, Vulpe M. Maggot therapy 18. Apelqvist J, Ragnarson-Tennvall G, Persson U. A for treating pressure ulcers in spinal cord injury comparative trial between Iodosorb and standard patients. J Spinal Cord Med 1995;18(2):71–4. treatment of deep diabetic ulcers including an evaluation of cost-effectiveness. Proceedings of the 5. Thomas S, Jones M, Shutler S, Andrews A. All 3rd European Conference on Advances in Wound you need to know about maggots – larval therapy. Management. 1993;3:126. Nurs Times 1996;92(46):63–4. 19. Colin D, Kurring PA, Quinlan D, Yvon C. The clinical 6. Thomas S. Wound management and dressings. investigation of an amorphous hydrogel compared London: The Pharmaceutical Press, 1990. with a dextranomer paste dressing in the manage- ment of sloughy pressure sores. Proceedings of the 7. Longe RL. Current concepts in clinical 5th European Conference on Advances in Wound therapeutics: pressure sores. Clin Pharmacol Management. London: Macmillan Ltd, 1996;5:151–5. 1986;5(8):669–81. 20. Mulder GD. A prospective controlled randomized 8. British National Formulary. British Medical study of the efficacy of Clearsite, DuoDERM, and wet- Association and the Royal Pharmaceutical Society to-dry gauze dressing in the management of pressure of Great Britain, 1998:35. ulcers. Proceedings of the 2nd European Conference 9. Ågren MS, Strömberg HE. Topical treatment of on Advances in Wound Management. 1992;2:45. pressure ulcers. A randomized comparative trial of 21. ConvaTec. Management of leg ulcers and pressure Varidase and zinc oxide. Scand J Plast Reconstr Surg sores. Company Bulletin. Uxbridge: Convatec 1985;19(1):97–100. 1996;1–4. 10. Thomas S. A handbook of surgical dressings. 22. Hillström L. Iodosorb compared to standard Bridgend: The Surgical Materials Testing treatment in chronic venous leg ulcers – a multi- Laboratory, 1992. center study. Acta Chir Scand 1988;544 (Suppl):53–6. 11. Thomas S. Nurses’ knowledge of dressings. 23. Skog E, Arnesjö B, Troëng T, Gjöres JE, Bergljung L, Community Outlook 1993;3(10):21–2. Gundersen J, et al. A randomized trial comparing cadexomer iodine and standard treatment in the 12. NHS Centre for Reviews and Dissemination. out-patient management of chronic venous ulcers. Undertaking systematic reviews of research on Br J Dermatol 1983;109(1):77–83. effectiveness. University of York, 1996;Report 4. 24. Eriksson G, Eklund AE, Kallings LO. The clinical 13. Moss C, Taylor AE, Shuster S. Comparison of significance of bacterial growth in venous leg ulcers. cadexomer iodine and dextranomer for chronic Scand J Infect Dis 1984;16(2):175–80. venous ulcers. Clin Exp Dermat 1987;12(6):413–18. 25. Groenewald JH. The treatment of varicose stasis 14. Brown-Etris MB, Pribble J, LaBrecque J. Evaluation ulcers: a controlled trail. Schweiz Rundsch Med Prax of two wound measurement methods in a multi- 1981;70(28):1273–8. center, controlled study. Ostomy Wound Manage 1994;40(7):44–8. 26. Leaper DJ, Stewart AJ. A comparative trial of Scherisorb gel and cadexomer iodine: problems of 15. Frantz RA, Johnson D. Stereophotography and community-based trials involving chronic leg ulcers. computerised image analysis: a three dimensional In: Turner TD, Schmidt RJ, Harding KG, editors. method of measuring wound healing. Wounds Advances in wound management. Chichester: 1992;4:58–64. Wiley, 1986:121–6. 27 References
27. Sawyer PN, Sophie Z, Dowback G, Cohen L, Feller J. 41. Lindsay G, Latta D, Lyons KGB, Livingstone ED, New approaches in the therapy of the peripheral Thomson W. A study in general practice of the vascular ulcer. Angiology 1978;29(9):666–74. efficacy of cadexomer iodine in venous leg ulcers treated on alternate days. Acta Therapeut 28. Troëng T, Skog E, Arnesjö B, Gjöres JE, Bergljung L, 1986;12:141–7. Gundersen J, et al. A randomised multicentre trial to compare the efficacy of cadexomer iodine and 42. Apelqvist J, Ragnarson-Tennvall G. Cavity foot standard treatment in the management of chronic ulcers in diabetic patients: a comparative study venous ulcers in out-patients. In: Fox JA, Fisher H, of cadexomer iodine ointment and standard editors. Cadexomer iodine. Stuttgart: FK Schattaüer, treatment. Acta Derm Venereol 1996;76:231–5. 1983:43–50. 43. Moberg S, Hoffman L, Grennert ML, Holst A. A randomized trial of cadexomer iodine in decubitus 29. Groenewald JH. An evaluation of dextranomer ulcers. J Am Geriatr Soc 1983;31(8):462–5. as a cleansing agent in the treatment of the post- phlebitic stasis ulcer. S Afr Med J 1980;57(20): 44. Ormiston MC, Seymour MT, Venn GE, Cohen RI, 809–15. Fox JA. Controlled trial of Iodosorb in chronic venous ulcers. Br Med J 1985;291(6491):308–10. 30. Garcia HG, Cobian RR, Martin JR, Konaizeh S, Rahola JG. The effect on the reparation process 45. Steele K, Irwin G, Dowds N. Cadexomer iodine in of dextranomer (Debrisan) in perimalleolar ulcers the management of venous leg ulcers in general due to chronic venous insufficiency. Clin Trials J practice. Practitioner 1986;230(1411):63–8. 1984;21(3):121–34. 46. Darkovich SL, Brown-Etris M, Spencer M. Biofilm 31. Sawyer PN, Hague S, Reddy K, Sophie Z, Feller J. hydrogel dressing: a clinical evaluation in the Wound healing effects of Debrisan on varicose, treatment of pressure sores. Ostomy Wound postoperative, decubitus, and sickle-cell ulcers in Management 1990;29:47–60. man. Vasc Surg 1979;13:251–6. 47. Brown-Etris M, Fowler E, Papen J, Stanfield J, Harris A, Tintle T, et al. Comparison and evaluation of the 32. Sayag J, Meaume S, Bohbot S. Healing properties performance characteristics, usability and effective- of calcium alginate dressings. J Wound Care ness on wound healing of Transorbent versus 1996;5(8):357–62. DuoDerm CGF. Proceedings of the 5th European 33. Palmieri B. Heterologous collagen in wound Conference on Advances in Wound Management. healing: a clinical study. Int J Tissue React London: Macmillan Ltd, 1996;5:151–5. 1992;14(Suppl):21–5. 48. Lum CC, Cheung W, Chow PM, Woo J, Hui E, Or KH. Comparative study of pressure sore manage- 34. Eriksson G, Eklund AE, Lidén S, Zetterquist S. ment. 1996 unpublished report provided by the Comparison of different treatments of venous leg author. (Conference abstract also available: Lum ulcers: a controlled study using stereophoto- CC, Cheung W, Chow PM, Woo J, Hui E, Or KH. grammetry. Curr Ther Res 1984;35:678–84. Use of a hydrogel in pressure sore management. A 35. Goode AW, Glazer G, Ellis BW. The cost- randomised controlled clinical trial. Proceedings effectiveness of Dextranomer and Eusol in the of the 5th European Conference on Advances in treatment of infected surgical wounds. Br J Clin Wound Management. London: Macmillan Ltd, Pract 1979;33:325–8. 1996;5:283.) 49. Mulder GD, Altman M, Seeley JE, Tintle T. 36. Nasar MA, Morley R. Cost-effectiveness in treating Prospective randomized study of the efficacy of deep pressure sores and ulcers. Practitioner hydrogel, hydrocolloid, and saline solution- 1982;226:307–10. moistened dressings on the management of 37. Parish LC, Collins E. Decubitus ulcers: a pressure ulcers. Wound Repair Regeneration comparative study. Cutis 1979;23(1):106–10. 1993;1(4):213–18. 50. Lee LK, Ambrus JL. Collagenous therapy for 38. Laudanska H, Gustavson B. In-patient treatment of decubitus ulcers. Geriatrics 1975;30(5):91–8. chronic varicose venous ulcers. A randomized trial of cadexomer iodine versus standard dressings. 51. Gordon B. The use of topical proteolytic enzymes J Int Med Res 1988;16(6):428–35. in the treatment of post-thrombotic leg ulcers. Br J Clin Pract 1975;29(6):143–6. 39. Holloway GA, Jr, Johansen KH, Barnes RW, Pierce GE. Multicenter trial of cadexomer iodine to treat 52. Westerhof W, Jansen FC, de Wit FS, Cormane RH. venous stasis ulcer. West J Med 1989;151(1):35–8. Controlled double-blind trial of fibrinolysin- desoxyribonuclease (Elase®) solution in patients 40. Harcup JW, Saul PA. A study of the effect of with chronic leg ulcers who are treated before cadexomer iodine in the treatment of venous autologous skin grafting. J Am Acad Dermatol 28 leg ulcers. Br J Clin Pract 1986;40(9):360–4. 1987;17(1):32–9. Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
53. Apelqvist J, Larsson J, Stenström A. Topical 58. Hulkko A, Holopainen YV, Orava S, Kangas J, treatment of necrotic foot ulcers in diabetic Kuusisto P, Hyvarinen E, et al. Comparison of dextra- patients: a comparative trial of DuoDerm and nomer and streptokinase-streptodornase in the MeZinc. Br J Dermatol 1990;123:787–92. treatment of venous leg ulcers and other infected wounds. Ann Chir Gynaecol 1981; 70(2):65–70. 54. Tarvainen K. Cadexomer iodine (Iodosorb) compared with dextranomer (Debrisan) in the 59. Gibson B. A cost effectiveness comparison of treatment of chronic leg ulcers. Acta Chir Scand two gels in the treatment of sloughy leg ulcers. Suppl 1988;544:57–9. Advanced Wound Care Symposium. April 1995; San Diego, USA. 55. Stewart A J, Leaper DJ. Treatment of chronic 60. Hellgren L. Cleansing properties of stabilized leg ulcers in the community: a comparative trypsin and streptokinase-streptodornase in necrotic trial of Scherisorb and Iodosorb. Phlebologie leg ulcers. Eur J Clin Pharmacol 1983;24(5):623–8. 1987;2:115–21. 61. Bux M, Baig MK, Rodrigues E, Armstrong D, 56. Colin D, Kurring PA, Quinlan D, Yvon C. Managing Brown A. Antibody response to topical sloughy pressure sores. J Wound Care 1996;5(10): streptokinase. J Wound Care 1997;6(2):70–3. 444–6. 62. Collier M. Hydrogels and their practical use in 57. Thomas S, Fear M. Comparing two dressings for wound debridement. Community Nursing Notes wound debridement. J Wound Care 1993;2(5):272–4. 1997;13:3–5.
29
Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Appendix 1 MEDLINE search strategy
EDLINE has been searched for RCTs from zinc/ or zinc oxide/ or ointments/ or M 1966 to October 1997 using a mixture of free anti-infective agents/ text terms and MEDLINE search headings: dermatologic agents/ or colloids/ or cushions/ or wheelchairs/ Wound infection beds/ or wound dressings/ Pilonidal cyst (debridement or dressing$ or compress$ Wounds and injuries or cream$ or (growth adj factor$)).tw. Wound healing (pressure-relie$ or (recombinant adj protein$) Leg ulcer or bandag$ or stocking$).tw. Varicose ulcer (antibiotic$ or (electric adj therapy) or laser$ or Skin ulcer nutrition$ or surg$).tw. Decubitus (homeopath$ or acupunture or massage or reflexology or ultrasound).tw. The MEDLINE search strategy used is as follows: (iloprost or alginate$ or zinc or paste$ or ointment$ or hydrocolloid$).tw. decubitus ulcer/ or foot ulcer/ ((compression adj therapy) or (compression adj leg ulcer/ or varicose ulcer/ bandag$) or wrap$).tw. pilonidal cyst/ (bed$ or mattress$ or wheelchair$ or (wheel adj skin ulcer/ chair) or cushion$).tw. diabetic foot/ ((wound adj dressing$) or vitamin$ or bind$ or ((plantar or diabetic or heel or venous or stasis or gauze$ or heals or healing).tw. arterial) adj ulcer$).tw. (diet or lotion$ or infect$ or reduc$ or (wound adj ((decubitus or foot or diabetic or ischaemic or healing)).tw. pressure) adj ulcer$).tw. (treat$ or prevent$ or epidemiol$ or aetiol$ or ((pressure or bed) adj sore$).tw. etiol$ or therap$ or prevalence or incidence).tw. ((pilonidal adj cyst) or (pilonidal adj sinus) or bedsore$).tw. or/15–35 ((diabetic adj foot) or (cavity adj wound)).tw. 14 and 36 ((varicose or leg or skin) adj ulcer$).tw. (decubitus or (chronic adj wound$)).tw. random allocation/ or randomized controlled ((sinus adj wound$) or (cavity adj wound$)).tw. trials/ controlled clinical trials/ or clinical trials phase I/ or/1-13 or clinical trials phase II/ clinical trials phase III/ or clinical trials phase IV/ or debridement/ or biological dressings/ or clinical trials overviews/ bandages/ single-blind method/ or double-blind method/ occlusive dressings/ or clothing/ or wound publication bias/ or review/ or review, academic/ healing/ review tutorial/ or meta-analysis/ or systematic antibiotics/ or growth substances/ or platelet- review/ derived growth factor/ ((random$ adj controlled adj trial$) or fibroblast growth factor/ or electrical stimulation (prospective adj random$)).tw. therapy.ti,ab,sh. ((random adj allocation) or random$ or (clinical lasers/ or nutrition/ or surgery/ or surgery, adj trial$) or control$).tw. plastic/ ((standard adj treatment) or compar$ or single- surgical flaps/ or skin transplantations/ or blind$ or double-blind$).tw. homeopathy/ or homeopathic/ (blind$ or placebo$ or systematic$ or (systematic acupuncture therapy/ or acupuncture/ or adj review)).tw. alternative medicine/ (randomized controlled trial or clinical trial).pt. or alternative medicine/ or massage/ or iloprost/ comparative or alginates/ study.sh. 31 Appendix 1
or/38-48 ((peptic adj ulcer) or (duodenal adj ulcer) or 37 and 49 traum$).tw. limit 50 to human ((aortocaval adj fistula) or (arteriovenous adj fistula)).tw. burns/ or wounds, gunshot/ or corneal ulcer/ or (bite adj wound$).tw. exp dentistry/ peptic ulcer/ or duodenal ulcer/ or stomach or/52–56 ulcer/ 51 not 57
32 Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Appendix 2 CINAHL and EMBASE search strategy
ther databases that have been searched for (debridement or dressing$ or compress$ or O RCTs are: cream$).tw. ((growth adj factor$) or pressure relie$ or EMBASE [to October 1997] (recombinant adj protein$) or bandag$).tw. CINAHL (Cumulative Index of Nursing and (stocking$ or antibiotic$ or (electric adj therapy) Allied Health Literature) [to April 1997] or laser$ or nutrition$ or surg$).tw. (iloprost or alginate$ or zinc or paste$ or The CINAHL search strategy used is as follows: ointment$ or hydrocolloid$).tw. ((compression adj therapy) or (compression adj pressure ulcer/ or foot ulcer/ or leg ulcer/ bandag$) or wrap$).tw. or skin ulcer/ (bed$ or mattress$ or wheelchair$ or (wheel adj diabetic foot/ or diabetic neuropathies/ chair) or cushion$).tw. diabetic angiopathies/ or diabetes mellitus/co ((wound adj dressing$) or vitamin$ or bind$ or pilonidal cyst/ or surgical wound infection/ gauze$ or heals or healing).tw. ((plantar or diabetic or heel or venous or stasis (diet or lotion$ or infect$ or reduc$ or etiol$ or or arterial) adj ulcer$).tw. (wound adj healing)).tw. ((decubitus or foot or diabetic or ischaemic or (treat$ or prevent$ or epidemiol$ or aetiol$ or pressure) adj ulcer$).tw. therap$ or prevalence or incidence).tw. ((pressure or bed) adj sore$).tw. “Bandages and dressings”/ or skin transplantation/ ((pilonidal adj cyst) or (pilonidal adj sinus) or or homeopathy/ or ointments/ or “beds and bedsore).tw. mattresses”/ ((diabetic adj foot) or (cavity adj wound)).tw. ((varicose or leg or skin) adj ulcer$).tw. or/14–34 (decubitus or (chronic adj wound$)).tw. 13 and 35 ((sinus adj wound$) or (cavity adj wound$)).tw. clinical trials/ or single-blind studies/ or double- or/1–12 blind studies/ control group/ or placebos/ or meta analysis/ debridement/ or biological dressings/ or occlusive ((random$ adj clinical adj trial$) or (prospective dressings/ adj random$)).tw. (bandages.ti,sh,ab,it. and “Bandages and ((random adj allocation) or random$ or controlled Dressings”/) or clinical trial$ or control).tw. compression garments/ or antibiotics/ (comparison group$ or (standard adj treatment) or electric stimulation/ or Laser Surgery/ or lasers/th compar$).tw. lasers/ or Nutrition Care (Saba HHCC)/ or diet (single-blind$ or (single adj blind) or double-blind therapy/ or Nutrition or (double adj blind)).tw. Therapy (Iowa NIC)/ (blind$ or placebo$ or systematic or (systematic adj surgery, reconstructive/ or surgery, plastic/ or review)).tw. surgical flaps/ (meta analysis or meta-analysis).tw. or (trials or trial surgical stapling/ or skin transplantation/ or or prospective).tw. alternative therapies/ (clinical trials).sh. or (comparative studies).sh . acupuncture/ or massage/ or zinc/ or ointments/ antiinfective agents, local/ or antibiotics/ or or/37–45 dermatologic agents/ 36 and 46 dermatology nursing/ or colloids/ or beds nad mattresses/ burns/ or wounds, gunshot/ or corneal ulcer/ or flotation beds/ or wheelchairs/ or exp dentistry/ positioning:wheelchair/ or positioning:therapy/ peptic ulcer/ or duodenal ulcer/ patient positioning/ or positioning/ or wound ((peptic adj ulcer) or (duodenal adj ulcer) or care/ or wound healing/ trauma).tw. 33 Appendix 2
(burn$ or (gunshot adj wound$) or (corneal adj ulcer) or dentist$ or (bite adj wound)).tw. or/48–51
47 not 52
34 Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Appendix 3 Additional databases searched
ISI Science Citation Index (on BIDS) DHSS Data (on Knight-Ridder Datastar) BIOSIS (on Silver Platter) EconLit British Diabetic Association Database EMBASE (on Knight-Ridder Datastar) CINAHL (on OVID CD-ROM) Index to Scientific and Technical Proceedings CISCOM, the database of the Reseach Council (searched on BIDS) for Complementary Medicine MEDLINE (on OVID CD-ROM) Cochrane Database of Systematic Reviews (CDSR) National Research Register (to locate ongoing Cochrane Wounds Group Specialised research in NHS) Trials Register Current Research in Britain (CRIB) NHS Economic Evaluation Database (NHS CRD) Database of Abstract of Reviews of Effectiveness Royal College of Nursing Database (CD-ROM) (DARE) System for Information on Grey Literature in Dissertation Abstracts Europe (SIGLE – on Blaise Line)
35
Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Appendix 4 Expert advisory panel
Dr Mary Bliss, Consultant Geriatrician, Department Dr Keith Harding, Director, Wound Healing of Medicine for the Elderly, Homerton Hospital, Research Unit, University of Wales College of Homerton Row, London E9 6SR, UK. Medicine, University Department of Surgery, Heath Park, Cardiff CF4 4XN, UK. Professor Nick Bosanquet, Imperial College School of Medicine, Department of General Practice, Deborah Hofman, Dermatology Department, Norfolk Place, London W2 1PG, UK. Churchill Hospital, Headington, Oxford OX3 7LJ, UK. Professor Andrew Boulton, Department of Medicine, Manchester Royal Infirmary, Oxford Vanessa Jones, Educational Facilitator, Wound Road, Manchester M13 9WL, UK. Healing Research Unit, University Department of Surgery, Heath Park, Cardiff CF4 4XN, UK. Dr Richard Bull, Department of Dermatology, Homerton Hospital, Homerton Row, London E9 6SR, UK. Dr Christina Lindholm, Nattarövägen 42, S-132 34 SaltsjöBoo, Sweden. Mr Michael Callam, Department of Vascular Surgery, Bedford Hospital, South Wing, Kempston Dr Raj Mani, Southampton University Hospitals Road, Bedford MK42 9DJ, UK. Trust, Medical Physics Department, Level D, Centre Block, Southampton SO9 4XY, UK. Mrs Carol Dealey, Research Fellow, University Hospital Birmingham NHS Trust, Department of Andrea Nelson, Department of Health Studies, Nursing, Queen Elizabeth Medical Centre, University of York, Heslington, York Edgbaston, Birmingham B15 2TH, UK. YO1 5DD, UK. Professor Peter Friedmann, Dermatology Unit, University Medicine, Level F, Centre Block Dr Steve Thomas, Director, Surgical Materials (Mail Point 825), Southampton General Hospital, Testing Laboratory, Bridgend General Hospital, Tremona Road, Southampton SO16 6YD, UK. Bridgend, Mid Glamorgan, UK.
Mr Brian Gilchrist, Department of Nursing Studies, Dr Ewan Wilkinson, Liverpool Health Authority, King’s College London, Cornwall House Annexe, Hamilton House, 24 Pall Mall, Liverpool Waterloo Road, London SE1 8TX, UK. L3 6AL, UK.
37
Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Appendix 5 Quality assessment
TABLE 4 Quality assessment of RCTs of debridement
Study Inclusion Total no. A priori Random- Appropriate Blinded Appropriate With- ITT and exclu- of wounds sample size isation baseline outcome outcome drawals† analysis sion criteria [arms] calculation procedure characteristics assessment measures stated stated reported*
Ågren and Strömberg, ✗ 28 patients [2] ✗✔✔c ✗✔ ✔a ✗ 19859
Apelqvist et al, 199053 ✔ 44 patients [2] ✗✗✔c ✔✔ ✔a ✗
Apelqvist and Tennvall, ✔ 41 [2] ✗✔✗ ✔✔✔b ✗ 199642
Brown-Etris et al, ✔ 140 [2] ✗✗✔ ✔✔✔b ✗ 199647
Colin et al, 199656 ✔ 135 [2] ✔✗ ✔c ✗✔ ✔a ✔
Darkovich et al, 199046 ✔ 123[2] ✗✗✔c ✗✔ ✔a ✗
Eriksson et al, 198434 ✔ 53 [2] ✗✗✔ ✗✔ N/A N/A
Garcia et al, 198430 ✔ 44 [2] ✗✗✔ ✗✔ N/A N/A
Gibson, 199559 ✗ 62 [2] ✗✗✔c ✗✔ ✗✗
Goode et al, 197935 ✗ 20 [2] ✗✔✔ ✗✔ N/A N/A
Gordon, 197551 ✗ 19 [2] ✔✗ ✔c ✔✔ ✔a ✗
Groenewald, 198029 ✗ 100 [2] ✗✗✔c ✗✔ ✔a ✗
Harcup and Saul, 198640 ✔ 72 [2] ✗✗✔c ✗✔ ✔a ✔
Hellgren, 198360 ✔ 40 [2] ✗✗✔ ✔✔✔a ✗
Holloway et al, 198939 ✔ 75 [2] ✗✗✔c ✗✔ ✔a ✗
Hulkko et al, 198158 ✗ 31 [2] ✗✗✔ ✗✔ ✔b ✗
Laudanska and Gustavson, 198838 ✔ 60 [2] ✗✗✔c ✗✔ ✔b ✗
Lee and Ambrus, 197550 ✗ 28 [2] ✗✗✔c ✗✔ ✔a ✗
Lindsay et al, 198641 ✔ 28 [2] ✗✗✔ ✗✔ ✔a ✗
Lum et al, 199648 ✔ 20 [2] ✗✔✔c ✗✔✔a ✗
Moberg et al, 198343 ✔ 38 patients [2] ✗✗✔c ✔✔ ✔a ✗
Moss et al, 198713 ✔ 42 [2] ✗✗✔c ✗✔ ✔b ✗
Mulder et al, 199349 ✔ 64 patients [3] ✗✔✔ ✗✔ ✔b ✗
Nasar and Morley, 198236 ✔ 18 [2] ✗✗✗ ✔✔✔a ✗
✔ = Yes; ✗ = No; N/A = not appropriate (no withdrawals) * Baseline characteristics: ✔ = one or more appropriate characteristics stated (but not initial wound size); ✔c = initial wound size stated † Withdrawals: ✔a = reported by group and with reason; ✔b = withdrawals but not reported by group or reason not given; ✗ = withdrawals not reported
continued 39 Appendix 5
TABLE 4 contd Quality assessment of RCTs of debridement
Study Inclusion Total no. A priori Random- Appropriate Blinded Appropriate With- ITT and exclu- of wounds sample size isation baseline outcome outcome drawals† analysis sion criteria [arms] calculation procedure characteristics assessment measures stated stated reported*
Ormiston et al, 198544 ✔ 61 [2] ✗✔✔c ✔✔ ✔a ✔
Palmieri, 199233 ✔ 48 [2] ✗✗✔ ✗✔ N/A N/A
Parish and Collins, ✗ 34 [3] ✗✗✔c ✗✔ N/A N/A 197937
Sawyer et al, 197931 ✗ 37 [2] ✗✔✗ ✗✔ N/A N/A
Sayag et al, 199632 ✔ 92 [2] ✔✔✔c ✗✔ ✔a ✔
Skog et al, 1983;23 ✔ 95 [2] ✗✗✔c ✗✔ ✔a ✗ Hillström, 1988;22 Troëng et al, 198328
Steele et al, 198645 ✔ 60 [2] ✗✔✔c ✗✔ ✔a ✗
Stewart and Leaper, ✔ 95 [2] ✗✔✔c ✗✔ ✔a ✗ 198755
Tarvainen, 198854 ✔ 27 [2] ✗✔✔ ✗✔ ✔a ✗
Thomas and Fear, 199357 ✔ 40 patients [2] ✗✔✔c ✗✔ ✔a ✔
Westerhof et al, 198752 ✔ 33 [2] ✗✔✔c ✔✔ ✔b ✗
✔ = Yes; ✗ = No; N/A = not appropriate (no withdrawals) * Baseline characteristics: ✔ = one or more appropriate characteristics stated (but not initial wound size); ✔c = initial wound size stated † Withdrawals: ✔a = reported by group and with reason; ✔b = withdrawals but not reported by group or reason not given; ✗ = withdrawals not reported
40 Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Appendix 6 Summary of studies
41 Appendix 6 continued < 0.05). p < 0.001; for p < 0.05). p patient showed a reduction a reduction patient showed in the I (dextranomer) group ( were Changes in area correlated to be highly found variables: with background glucose levels elevated healing in slower resulted as did a history of previous thrombosis. Bacterial flora remained the unchanged throughout of study irrespective treatment. I (dextranomer) does not induce contact dermatitis, eczema or allergic reactions. Statistical significance is after 1 and 2 weeks greater the initiation of treatment area (for circumference No withdrawals. the Pain as judged by No withdrawals reported. -test) t = 22) = 22) -test) = 22) t n n = 22) n n > 50 50 25 > 50 50 25 > 50 50 25 > 50 50 25 < 0.01; Student’s > 0.05, NS; Student’s p p % reduction in mean area of wound in mean area % reduction after 3 weeks: I: 93.1% (7.4 SD; C: 85.2% (10.8 SD; Number of wounds decreased in decreased Number of wounds % area: I:C: in increased Number of wounds % area: 2 6I:C: 8 5 in decreased Number of wounds 14 10 % volume: 0 0I:C: 1 3 in increased Number of wounds 1 3 % volume: 4 9I:C: 6 1 between No statistical difference 11 6 was found. treatments the two 0 0 2 5 3 6 ( % reduction in mean circumference % reduction after 3 weeks: I: 80.0% (20.6 SD; C: 70.1% (20.8 SD; ( All groups ): 2 All groups High History of glucose levels thrombosis Mean wound area: Mean wound Not stated. Other characteristics: Mean age (years):M:F ratio:Mean duration (months): 56.5 hospitalisation 33 patients required nine could be while the remaining as out-patients. treated 2.2:1 164 Mean wound area (cm area Mean wound Not stated. Other characteristics: Mean age (years):M:F ratio:Mean duration 70.1 (months): all comparable for Both groups characteristics except background 1:3.1 and a blood glucose levels elevated No details thrombosis history of previous negatively (both these factors are with ulcer healing): correlated I:C: 8% 30% 16% 38% = 27. n = 22. = 26. n n ) mixed with ) mixed ® = 22. n Treatment: I: Dextranomer polysaccharide beads (Debrisan), C: permanganate 1/5000 Potassium saline or antibiotic and hypertonic ointments, Prior to the initial treatment and Prior to the initial treatment at the startof of the second week treatment, bathed all ulcers were 15 min in water containing for 1 ml/l of 3% (w/v) potassium permanganate. Crust and debris removed. were C: in Sterilised gauze soaked 0.9% (w/v) sodium chloride in gauze was moistened water.The the day, through regularly sterilised glycerol and covered sterilised glycerol with gauze. was The dressing the first for day changed every for second day and every week the second week, Treatment: I: Dextranomer polysaccharide beads (Debrisan Inclusion criteria: Outpatients with venous leg ulcers. Exclusion criteria: Diabetes mellitus; manifest arterial insufficiency; clinical of erysipelaspicture or cellulitis. commencement Before the general health of all participants and the state of and arterialvenous peripheral evaluated. were circulation index ankle/arm of A pressure less than 0.75 was considered pathological and as such an exclusion criteria. Inclusion criteria: Patients with secondary perimalleolar ulcers and chronic insufficiency venous Exclusion criteria: Patients not submitted for surgical treatment. 34 30 1984 1984 Dextranomer polysaccharideDextranomer treatments versusor control traditional , Pharmacia Upjohn ® et al, et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Debrisan Garcia Garcia Spain type: Wound Secondary perimalleolar ulcers. Method of randomisation: Not stated. outcome: Objective and circumference Area measured of the wound planimetry. by Setting and length of treatment: Hospital and trial community-based period. a 3-week over Eriksson Sweden type: Wound leg ulcers. Venous Method of randomisation: Not stated. outcome: Objective measured and volume Area stereophotogrammetric by examination. Setting and length of treatment: a over Treated period. 2-week
42 5 TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) continued Three patients in the C group Three to have (Eusol) continued up to serious discharges for closure. after wound 5 days This did not occur in the I (dextranomer) group. healing was the Wound principal factor preventing hospital;discharge from those with I had a shortertreated a median of by hospital stay 2.2 days. The author states that the high cost of I is compensated shorter from the saving by hospital stay. There were no were There withdrawals. = 10) = 10) n n < 0.05; U-Test) Mann-Whitney p Number of wounds healed without Number of wounds secondary closure: I: 1/10 (10%) C: 1/10 (10%) to Mean time and range (days) secondary closure: I: 8.1 (range, 5–28; C: 1.6 (range, 6–22; ( IC ): 2 Mean wound area (cm area Mean wound Not stated. Other characteristics: Mean age (years):M:F ratio: 52.9Wounds: 50.9 AppendicectomyParamedian 1:0.4 primary Delayed 1:0.7 6suture abscessWound 7 4 either heavily were All wounds contaminated at operation and left 6 3 primary suture delayed open for (ten patients), or those which 4 4 and subse- closed primarily were an abscess that developed quently 6 of sutures the removal required and drainage (ten patients). = 10. n = 10. n All other nursing procedures were procedures All other nursing both groups. identical for Treatment: I: Dextranomer polysaccharide applied beads (Debrisan) were to a depth into the wound directly of 0.5 cm, a light pack. by followed twice Dextranomer was applied daily, C: Eusol and paraffin soaked changed ribbon gauze dressings twice daily, Inclusion criteria: wound Patients who developed after an appendicec- infections surgery. or bowel tomy Exclusion criteria: Not stated. Dextranomer polysaccharideDextranomer treatments versusor control traditional 35 1979 et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Goode UK type: Wound Surgical wounds. Method of randomisation: Sealed envelopes. outcome: Objective Time to secondary wound closure. was Each wound at the start,photographed during and at the end of treatment. Setting and length of treatment: Hospitalised surgical was patients.Treatment until it was continued to appropriate considered close the wound.
TABLE 5 contd TABLE 43 Appendix 6 continued Of 30 I (dextranomer) patients complaining of pain, improvement 20 showed within 24 hr, one after 3 days and one after 10 days. In the eight patients the remaining before worsened pain initially improving. Of the 35 C patients complaining of pain, seven after 24 hr,improved three after 3 days, and six after 7 days. In 13 the pain improving, before worsened patients the pain level in two unchanged,remained and in the pain worsened four later. without improving 30 patients, 15 in were each group not included in the final analysis.These patients were surgically treated after the wound was deemed to be clean (e.g. skin by grafting). additional Five patients dropped out of the trial and to replaced were the group keep at 50. number these (Reasons for not are withdrawals stated.) IC = 35) = 35) n n < 0.05) p Reduction in wound size at 21 days: Reduction in wound 0%:10%:25%:50%:75%:100%: 8Mean time to healing: 2I: 1 ( 4.44 weeks 26 6 1 7 11 1 2 3 2 C: ( 5.32 weeks ( IC IC Wound size: Wound < 6 cm:6–12 cm:> 12 cm:Other characteristics: 7 20Mean age (years): 23 Not stated M:F ratio: 21 7 Mean duration 22 (months): 1:3.5Depth of ulcer: 1:4.6 Shallow Not stated DeepRacial distribution (%): White 20ColouredBlack 30 29 right and of affected The number 21 almost equal in each left legs were 26 23 groups. of the two 25 22 to be shown The samples were 1 clinic of the overall representative population. 3 = 50. n Pseudomonas = 50. Treatment: I: Dextranomer polysaccharide poured (Debrisan) beads were onto the ulcer surface to directly at least 2–3 mm thick. a layer form gauze pad was used A multilayer in the ulcer and kept to cover gauze bandage, a standard place by n C: ointment was Povidone-iodine onto the ulcer and a swabbed bandage applied.Thepressure leg were and lower whole foot bound with a zinc oxide-impreg- nated gauze bandage, The ulcers and surrounding were in both groups leg area washed with a soft brush in initially solution,a povidone-iodine before with covered being temporarily in Eusol soaked gauze swabs solution. of In the presence 0.25% acetic acid solution infection was applied.The skin surrounding of was painted with tincture methiolate. of In the presence fungal occurring commonly skin of the surrounding infections area, antifungal an appropriate agent was applied. the sur- Where skin was eczematous, rounding flucinoline ointment was applied. Inclusion criteria: Outpatients with post-phlebitic stasis leg ulcers. Exclusion criteria: Not stated. Dextranomer polysaccharideDextranomer treatments versusor control traditional 29 Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Groenewald, 1980 South Africa type: Wound leg ulcers. Venous Method of randomisation: Not stated. outcome: Objective of The circumference was measured the wound overlay. a millimetre by was taken A photograph at each visit. Setting and length of treatment: Hospital clinic. Patients and 21 days for treated two by assessed daily independent investigators.
44 5 contd TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) continued Wounds treated with C treated Wounds observed to be (Eusol) were associated with a rise in to 11 mmol/l. blood urea Cost of materials calculated for each treatment for time in treatment average that group. was C treatment than I. costly 1.6 times more I:Three wounds. due to patient Two death; one as a of patient result discomfort. C: wounds. Four One due to patient death; three to switched dextranomer (two and after 16 days one after 48 days). Number of wounds attaining the Number of wounds end point: I: 6/9 (67%) C: 5/9 (56%) end point: Mean time to reach I: 39.3 days C: 62 days IC Mean age (years):M:F ratio: 83.2Mean duration 77.4 (months):Anaemia, hypoalbuminaemia, Not stated hypovitaminosis, and high blood if present. corrected were urea Not stated of diabetic Scrupulous control patients was ensured. Patients with urinary incontinence were catheterised. on the mostly were sores Pressure or heel in both groups. foot Mean wound area: Mean wound Not stated. Other characteristics: = 9. n = 9. n Treatment: I: Dextranomer polysaccharide as a stiff paste (Debrisan) applied and the first 3 days for twice daily thereafter, daily C: Eusol and paraffin packs were and to the wound applied changed three were dressings and the first 3 days for times daily twice daily.thereafter Melolin used throughout were dressings tape. micropore held in place by A sachet was used each time savlon was changed, the dressing Prior to initiation of the trial all cut off and sloughs were hardened on a large nursed all patients were cell ripple mattress.The con- only was ultraviolet therapy current inches to 12 square light applied first degree of skin to produce erythema masked with the sore rays. the ultraviolet from Inclusion criteria: Patients with deep pressure similar of approximately sores size. Exclusion criteria: Urinary tract infection. 36 Dextranomer polysaccharideDextranomer treatments versusor control traditional Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Nasar and Morley, 1982 UK type: Wound sores. Pressure Method of randomisation: Not stated. outcome: Objective was measured area Wound and using celluloid squares wound the entire photographed. End point was reached was clean when the wound and appear- and granulating ed to be less than 25% its original size (= healed). Setting and length of treatment: Hospital-based trial. made Assessments were an inde- by 3 days every pendent observer and a once a taken photograph was week.Treatment until the wound continued the end point,reached or of 94 days. a maximum for
TABLE 5 contd TABLE 45 Appendix 6 continued patients with any of the patients with any treatments. No withdrawals. No withdrawals. reported by No side-effects -test) -test) t t = 6) = 6) = 6) = 6) n n n n < 0.001; Student’s < 0.005; Student’s < 0.05; test) Fisher’s p p p ( Pressure sores: Pressure I: 47 ( C: 20 ( ( Mean time to healing (days): Leg ulcers: I: 60 ( C: 36 ( Number of patients with healed at 4 weeks: wounds I: 4/7 (57%) C: 0/5 (0%) Number of wounds healed Number of wounds at 4 weeks: I: 6/14 (43%) C: 0/9 (0%) ( of surface IC √ All groups Wound area: Wound Not stated. Other characteristics: Age range (years): 58–75 M:F ratio:Mean duration (months): type: Wound 1: 0.6 Leg ulcers: Not stated Diabetic gangrene: sores:Pressure 12 traumatic:Post 12 12 12 Mean wound size = Mean wound (cm): area I: 4.5 C: 2.4 between No statistical difference ulcer size for the groups Other characteristics: Age range (years):M:F ratio:Mean duration (months): 29–57 Not stated 32–70 Not stated = 24. n = 24. = 9 wounds from = 9 wounds n n = 14 wounds from seven from = 14 wounds five patients. five I: Dextranomer polysaccharide to a beads (Debrisan) applied depth of at least 3 mm covered with a dry dressing. Changed 1–3 depending on exudate,times daily n patients. C: Sugar and egg white applied after a saline wash. Changed four times a day. to dry Allowed and not covered, I: Dextranomer polysaccharide applied beads (Debrisan) were bead and to the wound directly daily, replaced C: directly Collagen sponge applied after saline nebulis- to the wound ation.The was checked dressing and if the collagen day every or partially sponge was swollen sponge was more reabsorbed the without removing applied one.previous sponge and Greasy non-allergenic tape regular completed the dressing, sharp debrided were All wounds prior to randomisation. In addition to ensure treated were all wounds at bacterial cultures negative baseline. Inclusion criteria: sores, Patients with pressure home. in a nursing residing Exclusion criteria: Not stated. Inclusion criteria: leg ulcers,Venous pressure sores, diabetic gangrene, sores,pressure post traumatic wounds, burns and radioactive ulcers. Note: given Data are leg ulcers and for here only. sores pressure Exclusion criteria: with treatments Additional drugs (with the exception of digitalis). 37 Dextranomer polysaccharideDextranomer treatments versusor control traditional 33 Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Palmieri, 1992 Italy type: Wound Leg ulcers and pressure sores. Method of allocation: Not stated. outcome: Objective Time to healing. Setting and length of treatment: clinic.Treatment Wound until all was continued had healed. wounds USA type: Wound sores. Pressure Method of randomisation: Not stated. outcome: Objective Number of ulcers healed. Setting and length of treatment: (nursing Community trial. home) 4-week Parish and Collins, 1979
46 5 contd TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) continued Pain from the wound Pain from in 31 patients improved with I (dextranomer) treated patients treated and in five with C (saline). in two The pain worsened with I and in patients treated with C.14 patients treated I: No withdrawals C: No withdrawals. < 0.05; test) chi-squared p Number of wounds healed Number of wounds at 3 weeks: I: 15/18 (77%) C: 3/19 (16%) ( Mean wound area: Mean wound Not stated Other characteristics: Mean age (years): Not stated M:F ratio:Mean duration (months): Not stated Not stated ® = 18. n = 19. n Both treatments were covered were Both treatments with gauze and a cling bandage. the above elevated Legs were as possible.heart as frequently in pHiso Hex soaked All wounds for 30 min prior to treatment. for C: Soaks only. Repeated twice daily, Treatment: I: Dextranomer polysaccharide to a beads (Debrisan) applied depth of 2–3 mm. convex For mixed the beads were wounds a paste. to form with glycerol once dextranomer applied Fresh or twice daily, unless heavy frequent more exudate required changing, Inclusion criteria: Outpatients or hospitalised patients with cutaneous ulcers origin.of mixed Note: the only ulcers are venous for results here. presented Exclusion criteria: Not stated. Dextranomer polysaccharideDextranomer treatments versusor control traditional 31 1979 ; unknown manufacturer ® et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments pHiso Hex Sawyer Sawyer USA type: Wound leg ulcers. Venous Method of randomisation: Sealed envelopes. outcome: Objective in epithelised % increase area. Method of not stated. measurement Setting and length of treatment: Hospital and medical clinic.centre Assessments a minimum performed were for (daily of once a week hospitalised patients) for 3 weeks.
TABLE 5 contd TABLE 47 Appendix 6 On average the number of the number On average changes per week dressing was similar: 4.28 (1.49 SD) C and 4.52 (1.42 SD) for I. for 8% of the C patients and 33% of I patients experienced effects. adverse = 2); = 15). = 1); n = 6); = 5); n n n n = 1); = 2). n n I: 22 Withdrawals: death ( event adverse ( deterioration or stagnation of ulcer after ( 4 week C:Ten withdrawals: C:Ten death ( ( transfer deterioration or stagnation of ulcer after 4 week ( included in All were and few the analysis to considered were at improved have the last evaluation. End point data not available were one patient in C for due to admission unit. a special care deterioration of health ( ): 2 = 0.0002 -test) -test) t t p 40% (cm ≥ ): 2 = 0.002; exact test) Fisher’s = 0.0004; Student’s = 0.0001; Student’s 40% reduction in area: 40% reduction p p p Logrank test) between the groups, Logrank test) between in C and with a median of 4 weeks in I. 8 weeks Mean wound area reduction per reduction area Mean wound (cm week Number of wounds with > 75% Number of wounds in area: reduction 1: 6/45 (13%) C: 15/47 (32%) with sores Number of pressure ≥ I: 19/45 (42%) C: 35/47 (74%) ( curves of the number Cumulative 40% reduction of patients reaching ( significantly differed I: 2.15 (3.60 SD) C: 3.55 (2.18 SD) ( I: 0.27 (3.21 SD) C: 2.39 (3.54 SD) ( per reduction area Mean wound those only using the data from week patients reaching -test). t ): 2 IC I: 16.1 (12.5 SD) C: 20.1 (12.9 SD) Other characteristics: Mean (SD) age (years): 80.4 (9.1)M:F ratio: 81.9 (8.9) Mean (SD) 1:2.8duration (months): 1:2.9 3.0 (3.2) grade: Wound 3.5 (3.8) IIIIV between No significant difference (Student’s groups the two 30 patients had multiple Where 15 one was selected only wounds study. for 33 14 located on were sores Pressure the sacrum, ischium, trochanter and heels. Mean wound area (cm area Mean wound :1218–19. ) applied directly directly ) applied ® = 45. 126 n = 47. n 1990; Treatment: I: Dextranomer polysaccharide to a paste (Debrisan) applied the wound depth of 3 mm over surface, C: Calcium alginate dressings (Algosteril on to wound to cover the entire to cover on to wound area, a sterile gauze was In both groups as a secondaryapplied dressing. were No other local treatments saline solution the used except for use of which was not restricted. inspected and were Dressings or at least every changed daily depending on the degree 4 days of exudate. Yarkony’s Arch Dermatol Arch * . 2 60 hospitalised ≥ 8 weeks, with a pressure ≥ Inclusion criteria: Patients aged for graded III or IV ( sore classification) and surface 5–100 cm from area Exclusion criteria: than half the total ulcer More tissue; had granulating area necrotic by covered wound plaque; requiring infection active local or systemic antibiotic therapy; failure; renal severe heel ulcers when combined with end- of the lower stage arteriopathy limbs; or radiotherapy receiving therapy. cytotoxic Classification of pressure ulcers. of pressure Classification Dextranomer polysaccharideDextranomer treatments versusor control traditional 32 et al. 1996 , Beiersdorf by UK manufactured (now Les Laboratories Brothier Ltd) ® et al, Yarkony GM, Yarkony Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Algosteril * Sayag Sayag France type: Wound sores. Pressure Method of randomisation: Sealed envelopes. outcome: Objective measured of wound Area planimetry,by digitised calcu- twice and the area computer.The lated by values was mean of the two used to determine individ- area.ual wound A photo- of each was taken graph evaluation. at every wound Setting and length of treatment: trial based A multicentre (17 special- at 20 centres of elderly ising in the care in people and three dermatology). Assess- made on a ments were the same basis by weekly at each researcher was centre.Treatment terminated when the 40% of reached wound the initial area, or after a of 8 weeks. maximum
48 5 contd TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) continued Mean weekly cost was 903 Mean weekly C, I and 1421 SEK for SEK for of which the major part was staff and costs for transportation to related of dressing frequency changes. were reactions No adverse with either recorded treatment. and/or 2 ulcer type,Wagner patients III).Two excluded were because of hospital- isation (one for infarc- myocardial tion, one became and later febrile died), these hospitalisations not associated were with treatment. One patient was non- excluded for compliance with treatment. One other patient was excluded because of insufficient data. 35 patients for remained clinical evaluation. I: withdrawals Five C: One withdrawal patients were Two the excluded from due to analysis violation of inclu- sion criteria (size > 25 cm Number of wounds healed: Number of wounds I: 5/22 (23%) C: 2/19 (11%) ): 2 Mean wound area (cm area Mean wound Not stated. Other characteristics: Mean age (years):M:F ratio: Not stated Mean duration (months): Not stated Not stated ) was ® :64–122. 2 = 19. n ) was used. ® 1981; = 22. n Foot Ankle Foot Wounds were cleaned with sterile were Wounds in accord- saline prior to dressing ance with the manufacturers guidelines. applied were Dressings nurses. healthcare regular by when the both treatments For ulcers had stopped exudating, gauze (Jelonet Vaseline Treatment: I: Cadexomer iodine polysac- charide ointment (Iodosorb Footwear of patients was Footwear or special footwear corrected pressure. to relieve provided Oral in case of given antibiotics were signs of infection. changed once daily for the first for changed once daily or every and then daily week thereafter day second or third of to the degree according exudation, C: Gentamicin solution twice daily (80 mg/ml) was given cellulitis was present. where streptokinase Streptodornase/ was used for Topical) (Varidase lesions and changed moist necrotic twice daily. Dry saline gauze was and dressing used as an absorptive changed once or twice daily to exudation, according 2 , Ltd Healthcare Smith & Nephew ® ; patients with 2 ) with an ulcer area > 1 cm ) with an ulcer area * Inclusion criteria: Caucasian outpatients old with previously > 40 years diabetes mellitus.Allknown exudating and were wounds grade the ankle (Wagner below I–II a deep abscess; or osteomyelitis grade III–IV); (Wagner gangrene thyroid patients undergoing gland investigation; patients to the study to adhere unlikely protocol. wounds, In patients with several meeting the largest wound only the selection criteria was chosen. and a systolic toe pressure and a systolic toe pressure > 30 mmHg or a systolic ankle > 80 mmHg. pressure Exclusion criteria: > 25 cm Wounds Cadexomer iodine polysaccharide treatments versusor control traditional , Ltd; Healthcare Smith & Nephew Jelonet ® 42 Wagner FW.Wagner The dysvascular foot: diagnosis and treatment. a system for Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Iodosorb * Apelqvist and Tennvall, Apelqvist and 1996 Sweden type: Wound ulcers. Diabetic foot Method of randomisation: Computer allocation. Strati- on fication was performed size and type of wound grade I–II). (Wagner outcome: Objective determined of wound Area wounds photographing by against a graduated scale. length and The maximum mul- width were maximum tiplied together to calculate the area. the In addition healed of wounds number was recorded. Setting and length of treatment: trial with Open controlled blinded photo-evaluation. treated Patients were either at home or on visiting an outpatient clinic. made at Assessments were 1, 4, 8 and 12 weeks.
TABLE 6 TABLE 49 Appendix 6 continued Six patients treated with I Six patients treated experienced mild transient burning, pain, or itching. were effects No adverse reported in the C group. During the 4-week treatment During the 4-week period those patients C reported a 27% receiving in pain,decrease while those with I reported treated a 66% reduction. 15 patients did not complete the treat- ment: failed to four to treat- respond ment; died from two to causes unrelated their ulcers; nine out and dropped failed to return follow-up. for not Six patients were included in the ana- lysis: to felt were two been inappro- have admitted to priately the study because too their ulcers were small; lacked four adequate follow-up information. were Withdrawals in equivalent pro- both portions from groups. I:Three patients withdrawn were due after 2 weeks to various reasons, including diarrhoea, erythema, oedema, and ulcer irritation unhappiness with treatment. These patients included in were at the analysis 4 weeks. ): 2 ): 2 = 27) = 27) = 27) = 27) n n < 0.001; analysis n n p /wk/cm 2 = 41) = 31) = 31) = 41) n n n n = 0.0025; of covariance) analysis = 0.0720; of covariance) analysis < 0.01 I vs. C p p I: 0.95 (0.12 SEM; C: 0.41 (0.13 SEM; ( size in wound Mean reduction vs. after baseline circumference (cm 24 weeks I: 0.04 (0.01 SEM; C: 0.03 (0.01 SEM; ( was area in wound The reduction baseline from different significantly ( in both groups of covariance). Mean reduction in wound size per in wound Mean reduction (cm after 24 weeks week C: 9.91 ( p of covariance) (analysis C: 0.90 ( of wound in mean area % reduction at 4 weeks: I: 36.30 ( Mean wound area at 4 weeks: area Mean wound I: 2.81 ( ): 2 All groups ): 2 IC Mean wound area (cm area Mean wound I: 20.1 C: 11.2 Other characteristics: Mean age (years): 63.0M:F ratio: 61.5 Mean duration (months): 1:0.8 No statistical differences 1:1.5 groups the two between 29.5reported. Although the mean 11.4 duration and initial mean area in the I group. greater are Mean area of wound (cm of wound Mean area I: 7.74 (1.04 SD) C: 9.08 (1.37 SD) Other characteristics: Mean age (years):M:F ratio:Mean duration (months): 16.9 (%): wounds Additional 67.8 26 1:2.2 = 38. n = 41. n = 31. n = 37. ), absorbent dress- ® ), n ® ) and medicated dressings ® C:Wet-to-dry dressings with dressings C:Wet-to-dry sterile 4 x inch saline-soaked gauze pads, were Patients in both groups changing their for responsible once a day. dressings own Both with a covered were treatments toe-to-knee elastic compression bandage. Treatment: I: Cadexomer iodine (Iodosorb) powder polysaccharide after sprinkled onto the wound with saline,irrigation and covered with a dry gauze dressing, Treatment: I: Cadexomer iodine polysac- charide beads (Iodosorb) applied to a depth of 3 mm after cleansing or a with sterile saline swabs of water or saline. gentle stream with a dry sterile Covered bandaging by and secured dressing or stocking. Cadexomer iodine was changed daily, C: a support bandaging or Primarily stocking and a dry dressing. 21 of a variant the participants received which included on this treatment amongst others: Crêpe bandaging (Elastocrepe (Sofra-Tulle ings (Melolin), topical antibiotics (Polyfax Inclusion criteria: stasis Out-patients with venous a minimum for ulcers present of 3 months and otherwise in health. good Exclusion criteria: < 2 cm in wounds Initially diameter; or suspected proven cause of wounds; non-venous with the inability to comply treatment; major medical disorders; iodine allergy; significant arterialclinically disease. Inclusion criteria: Out-patients with exuding leg ulcers not chronic to existing responding treatment. Exclusion criteria: Concomitant, serious or life- disease;threatening a suspected malignant ulcer; insulin-requiring diabetes; pregnancy; iodine sensitivity; psychiatric disease; intelligence;low dementia; or to affect other conditions likely ability to comply the patient’s with the trial. 40 Cadexomer iodine polysaccharide treatments versusor control traditional 39 1989 et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Elastocrepe, Ltd; Healthcare Smith & Nephew Sofra-Tulle, Hoechst Marion Roussel USA type: Wound leg ulcers. Venous Method of randomisation: Not stated. outcome: Objective was of the wound Area planimetry by measured tracings. on transparent was A colour photograph at each evaluation. taken Setting and length of treatment: A multicentre 24 trial conducted for or until the wound weeks to be was considered healed. were Measurements intervals at 2-week taken and the first 8 weeks for intervals.then at monthly Holloway Holloway Harcup and Saul,Harcup 1986 UK type: Wound leg ulcers. Chronic Method of randomisation: Not stated. outcome: Objective was mea- of wound Area tracing the outline by sured on to a plastic sheet. Setting and length of treat- ment: treated Patients were at home under the guidance of a general practitioner. re- were Patients or carers changing dress- sponsible for ings. made Assessments were and 4 weeks. 2 weeks every
50 6 contd TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) continued Application of I resulted in Application of I resulted pain than with C.more Serum concentrations of iodine increase protein-bound with I treatment. is But there no disturbance of thyroid function. patients on I complained Five of stinging. Four withdrawals Four the first before assessment: three due to social reas- ons and one due to failure. cardiac patients Three withdrawn were after completing the trial. One in was each group excluded due to large ulcer area (ten times that of the mean ulcer area).The third patient had rheu- matoid arthritis which interfered with assessment. 30 patients in each completed group the trial. < 0.05; p -test) t = 30) = 30) n n < 0.01; Student’s p ( matched pairs) Wilcoxon Number of ulcers healed or shallow at 6 weeks: I: 16/30 (53%) C: 7 / 30 (7%) Mean % reduction in wound area at area in wound Mean % reduction 6 weeks: I: 71% ( C: 54% ( significantly were Both treatments baseline. from different ( ): 2 IC -test) t > 0.05; Student’s p Other characteristics: I: 27.5 (7.0 SE) C: 35.2 (8.1 SE) ( Mean age (years): 64.8M:F ratio: 63.5 Mean duration (months): 1:0.8 between No statistical difference 1:1.5 groups. the two 19.1 15.0 for available only Details are 60 of the 67 participants initially randomised to treatment. Mean wound area (cm area Mean wound = 30. = 30. n n Treatment: I: Cadexomer iodine powder in a daily (Iodosorb) applied with a light 3–4 mm layer, covered elastic bandage, C: and peroxide Dilute hydrogen saline zinc paste (in addition dressings, dilute potassium per- manganate and Gentian violet with a light covered applied) were elastic bandage, complete given All patients were 6 weeks. for bed rest Inclusion criteria: Hospitalised patients with ulcers that had venous chronic for outpatient treatment resisted than 3 months. more Exclusion criteria: < 2 cm;Wounds iodine sensitivity; peripheral severe arterial disease. Cadexomer iodine polysaccharide treatments versusor control traditional 38 Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Laudanska and Gustavson, 1988 Poland type: Wound leg ulcers. Venous Method of randomisation: Not stated. outcomes: Objective perimeter was The wound onto a transparency drawn calculated by and the area of planimetry.The number healed was wounds recorded. Setting and length of treatment: trial.6-week Measure- at 0,ments taken 1, 2, 4 and 6 weeks.
TABLE 6 contd TABLE 51 Appendix 6 continued Mean wound duration was Mean wound of 20 months so a reduction on I is 4 weeks 33% in only impressive. However, no data of other measures given are been employed have that may size. wound to reduce I caused three patients to I caused three experience smarting after application, one patient had and minor skin irritation another had an exacerbation of psoriasis. of the pain as a result Overall less in was significantly wound with I. patients treated and I was easy to apply remove. I: One patient was after withdrawn due to an 2 weeks allergic reaction. C: One patient was after withdrawn on the 4 weeks of peri- discovery pheral vascular disease. The results based on 25 are patients suggesting was a further there withdrawal, prob- in the I group. ably I:Three withdrawals.Two they patients felt getting worse were and one had skin and irritation a oedema around sacral ulcer and chose not to continue. C: One withdrawal the wound where and the had grown patient was moved to another hospital. -test or t = 18) = 18) = 16) n = 16) n n n = 12) = 13) n n < 0.005 I vs. C; of analysis < 0.05 for both treatments when both treatments < 0.05 for < 0.02; correlated p p p -test or Fisher’s exact test). -test or Fisher’s Mean % reduction in wound area at area in wound Mean % reduction 4 weeks: I: 33.6 ( C: 4.2 ( ( variance). Mean decrease in area of wound of wound in area Mean decrease at 3 weeks: I: 2.9 (1.3 SEM; C: 2.5 (1.1 SEM; ( with baseline;compared correlated t of in mean area % reduction at 3 weeks: wound I: 30.9% (46 SD; C: 19.6% (31 SD; exact test) Fisher’s ( All groups ): 2 IC Mean wound area (cm area Mean wound I: 9.6 (1.8 SEM) C: 12.4 (4.3 SEM) Other characteristics: Mean age (years): 72.6M:F ratio: 80.1 Mean duration (months): 1:4.3 the for available only are Values 1:2.6 the from patients not withdrawn 6.2study. 6.2 Mean wound area: Mean wound Not stated. Other characteristics: Mean age (years):M:F ratio:Mean duration (months): 20.1 No. 66.7 of patients with ulcers:additional All female 11 = 19. n = 19. n C: was vari- treatment Standard able. It included: saline dressings, enzyme-based debriding agents, dressings, and non-adhesive subject to: All patients were attention to nutrition; improve- ment of hygiene; of removal using decubitus by pressure mattresses, turning the patient 2 or 3 hrs,every and optimal mobilisation. Treatment: I: Cadexomer iodine polysac- (Iodosorb) charide powder to a depth of 3 mm. daily applied running water by Removed swab, saline or wet Treatment: the trial,28 patients entered but the groups allocation between was not stated. I: Cadexomer iodine powder to a depth of (Iodosorb) applied not less than 3 mm after cleansing with sterile saline swabs, or a of water or saline. gentle stream with a dry sterile dress- Covered bandaging or by ing and secured stocking.The was dressing changed on alternate days. C: a non-adherent Primarily plus support dressing bandaging or stocking.The was dressing changed on alternate days.Twelve a of the participants received which variant on this treatment included amongst others: Crêpe bandaging (Elastocrepe), absorbent (Melolin),dressings Povidone iodine and medicated dressings (Sofra-Tulle). Inclusion criteria: Hospitalised patients with sores. pressure Exclusion criteria: Confirmed or suspected malignancies, moribund, iodine sensitivity, psychiatric illness, psoriasis,severe other any a criteria that might make a patient unsuitable for clinical trial or unable to consent. informed give Inclusion criteria: Out-patients with chronic leg ulcers and not venous to existing responding treatment. Exclusion criteria: Concomitant serious or life- disease,threatening suspected malignant change in the ulcer; insulin-dependent diabetes; pregnancy; iodine sensitivity; psychiatric disease; very low intelligence; dementia; other any the condition that might affect with patients ability to comply the conditions of trial. Cadexomer iodine polysaccharide treatments versusor control traditional 43 41 1983 1986 et al, et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Lindsay Lindsay UK type: Wound leg ulcers. Venous Method of randomisation: Not stated. outcome: Objective was of the wound Area traced onto a plastic sheet and measured. Setting and length of treatment: Patients treated at home under the guidance of a general practitioner. were Patients or carers changing for responsible dressings. Assessments 2 weeks made every were 4 weeks. for Moberg Sweden type: Wound sores. Pressure Method of randomisation: Not stated. outcome: Objective of the wound Perimeter was traced and the area planimetrycalculated by of the or measurement longest diameter. Setting and length of treatment: trial. 3-week
52 6 contd TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) continued Patients were able to Patients were bandage their adequately wounds. own patients had difficulty Two patients I and three removing complained of stinging and itching. Pain was assessed on a scale the individual of 0–100 by patient. Pain was perceived to be similar in both groups. on patients on C and five Two eczema,I developed puritis or a rash. I: One patient was admitted to surgery, hospital for and had inappropriately ulcers (this dressed patient was the excluded from baseline data further set).Two patients were withdrawn: one died of a perforated ulcer and the other had difficulty in the I from removing two ulcer.These patients were included in the analysis. C: No withdrawals. -test). t -test). ): t = 30) 2 = 30) n n = 30) = 30) n n < 0.02; Student’s > 0.05; analysis) chi-squared < 0.0001; Student’s p p p ( % reduction in mean area of wound in mean area % reduction at 12 weeks: I: 83% ( C: 61% ( per week Reduction in mean area (cm 12 weeks over Both treatments are statistically are Both treatments baseline (p < 0.001; from different of covariance). analysis Number of ulcers healed after 12 weeks: I: 12/31 (39%) C: 7/30 (23%) ( I: 0.89 (0.1 SEM; C: 0.46 (0.1 SEM; ( ): 2 IC I: 12.1 (13.9 SD) C: 10.2 (8.7 SD) Other characteristics: Mean age (years): 67.3M:F ratio: 70.3 Mean duration (months): 1:1.3 for given Baseline data was only 1:2.6 30 of the 31 patients treated 45.9with I (see withdrawals). 15.9 Mean wound area (cm area Mean wound = 31. n = 30. C: Gentian violet, by followed applied topical antibiotics (Polyfax) layer, and covered in a generous pad (Melolin), with a non-adherent n with a was covered Each dressing crêpe bandage and a cotton bandage. I: Cadexomer iodine (Iodosorb), powder polysaccharide with a 3–5 mm deep and covered gauze pad, Inclusion criteria: Out-patients with chronic leg ulcers persisting for venous at least 3 months. Exclusion criteria: aetiology, Ulcers of non-venous peripheral vascular disease (ABPI < 0.7), poor compliance expected due to concomitant or mental disability, physical problems. travelling Almost all patients selected and bandage able to dress were ulcers. their own 44 Cadexomer iodine polysaccharide treatments versusor control traditional 1985 et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments ABPI, index ankle brachial pressure Ormiston UK type: Wound leg ulcers. Venous Method of randomisation: Sealed envelopes. outcomes: Objective was of the wound Perimeter onto a transparency drawn calculated by and the area was wound planimetry.The also photographed.The healed of wounds number was recorded. Setting and length of treatment: trial.12-week Patients assessed at regular were intervals.After 12 weeks to allowed patients were cross-over.
TABLE 6 contd TABLE 53 Appendix 6 continued One patient on C and four on I complained of pain of the dressing. application This subsided 30–60 min later. a One patient developed rash in the I group, but tests that this was due to showed and not coal tar derivatives the treatment. Another acquired patient in this group the ulcer an itch around which subsided after 2 weeks. Thyroxine, tri-iodothyronine, binding serum thyroxine index and thyroid protein did not change significantly the study period. over I: patients did Four not meet the selec- tion criteria; one a rash; developed took holidays; two had beta- two Strepto- haemolytic coccus infection; had missing two information; and one experienced of recurrence ulcer pain. C:Three patients did not meet the selection criteria; had beta- four Strepto- haemolytic coccus infection; one developed squamous cell ;carcinoma and one experienced a dramatic increase in ulcer size. = 36) = 38) n n ): 2 < 0.02; matched pairs Wilcoxon p ( signed-ranks test, two-sided). I: –34% (5 SEM; C: +5% (15 SEM; Mean % change in wound area after area Mean % change in wound (cm 6 weeks ): IC 2 (1.99) (3.54) (18.3) (14.3) I: 20.1 (4.4 SEM) C: 34.0 (5.7 SEM) Other characteristics: Mean (SEM) age (years):M:F ratio:Mean (SEM) duration (months): 68.1 72.1 26.5 1:2.8 22.2 type: Wound 1:3.5 VenousOtherDepth of ulcer: DeepSuperficial 37 superficialVery 1 30 between No significant difference 1 groups. the two 26 6 11 available only Baseline details are 1 23 the 74 patients completing for 12 the original the trial from 95 randomised. Mean wound area (cm area Mean wound , ® = 45. n -values given are for patients for are -values given = 50. n Treatment: I: Cadexomer iodine polysac- (Iodosorb) applied charide powder to a depth of approx. 3 mm after washing in running water, and with a dry dressing, covered n C: cleaned ulcers were Each day or peroxide with dilute hydrogen dilute potassium permanganate were dressings and non-adherent applied. Paraffin-impregnated most commonly were dressings used, and but saline dressings occasionally bland ointments were used. In case of difficulties the was able to modify physician the treatment. Other therapies used included: Salvstrumpa merbromine and systemic merbromine antibiotics, during the first In both groups when exudate was heavy days few could be changed the dressings twice daily, the remainder but for of the trial this was seldom necessary were and dressings changed daily. with treated All patients were bandages applied compression either at the clinic or a nurse by home. trained Some patients were bandages to change their own ( after withdrawals). remaining ; history of iodine 2 sensitivity; peripheral arterial disease. Inclusion criteria: Out-patients with chronic leg ulcers that failed to infected treatments. to current respond Exclusion criteria: Ulcers with diameters smaller less than than 2 cm and areas 3 cm Cadexomer iodine polysaccharide treatments versusor control traditional 28 22 23 , unknown manufacturer 1983 ® 1983; et al, et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Salvstrumpa Skog Skog Hillström, 1988; Troëng Sweden type: Wound Leg ulcers. Method of randomisation: Not stated. outcome: Objective was of the wound Area planimetry, by measured or measuring the greatest by diameters. All wounds against photographed were a ruler scale at each assessment. Setting and length of treatment: trial in ten A multicentre centres.Assessments were made after 1, 2, 4 and 6 at which point the weeks trial was terminated.
54 6 contd TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) Pain directly after application Pain directly I. common for was more pain was similar in Day-to-day after 6 weeks’ both groups treatment. I: 2 C: 1 due to Withdrawals hospital admission and lack of cooperation. = 29) = 28) n n = 0.31; analysis) chi-squared p % reduction in mean area of wounds in mean area % reduction at 6 weeks: I: 22% ( C: 18% ( ( healed at Number of wounds 6 weeks: I: 3/30 (10%) C: 1/30 (3%) ): 2 IC = 0.32; statistical method p Mean wound area (mm area Mean wound I: 1264 (291 SEM) C: 1759 (397 SEM) ( not stated). Other characteristics: Mean age (years): 69.5M:F ratio: 73.4 Mean duration (months): 1:2.5 between No statistical differences 1:2.6 groups. the two 16.6 16.3 57 of for available only Data are the 60 participants. = 30. n = 30. n Treatment: I: Cadexomer iodine poly- (Iodosorb), saccharide powder with gauze, covered C:A used variety of agents were including: antibiotics, antiseptics, agents,hydrophilic bland agents, and drysteroids dressings. All with covered were treatments gauze, times a changed three Dressings week. crêpe All patients wore bandages. . 2 Exclusion criteria: Arterial disease, diabetes, rheumatoid arthritis, neuro- logical disease, tissue connective disease, hospital and ongoing treatment. Inclusion criteria: leg Out-patients with venous than more for ulcers present 3 months, and larger than 2 cm Cadexomer iodine polysaccharide treatments versusor control traditional 45 1986 et al, 0, 2, 4 and 6 weeks. Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Steele UK type: Wound leg ulcers. Venous Method of randomisation: table.Random number outcomes: Objective perimeter was The wound onto a transparency drawn calculated by and the area computer planimetry. healed Number of wounds was also recorded. Setting and length of treatment: from Patients recruited primary practices. care trial. Six-week at taken Measurements
TABLE 6 contd TABLE 55 Appendix 6 continued No significant differences in No significant differences infection, clinical wound odour, or dressing changes/week. 19 randomised not patients were included in the did as they analysis not complete the of the first 3 weeks study, or missed or more two sequential weekly visits. ): 2 ): ): 2 2 ) ) ) 2 2 2 = 3) = 12) (NS) = 36) (NS) = 2) (NS) = 12) = 42) n n n n ( ( ( ( n n ( ( 2 2 2 2 2 2 C: 4.3 cm I: 24.5 cm Stage II sores (2–30 cm Stage II sores (31–80 cm Stage III sores for available Insufficient data were of the subgroupings: analysis Stage II (31–80 cm Number of wounds healed Number of wounds at 10 weeks: I: 39/77 (51%) C: 37/63 (59%) of wounds Reduction in mean area at 10 weeks: (2–30 cm Stage II sores Stage IV 31–80 cm I: 3.6 cm I: 6.3 cm Stage IV (2–30 cm C: 2.3 cm C: 5.2 cm IC All groups Mean wound area: Mean wound Not stated Other characteristics: Mean age (years):M:F ratio 70 %2%Duration (months): 1:1 < 11–3 38% 49% 4–6 10% 14% 7–12 13 > 12 Location: 23%Sacrum 31% TrochanterHeel 16% 16% 18% Ischium 16% 4% 13% Malleolus 10% 17% 33%Spine 6% 26% 4% 37% Knee 0% 2% 2% = 63. n = 77. I: (Transorbent), dressing Hydrogel n C: dressing Hydrocolloid (Duoderm CGF), Evaluation took place weekly, every changes occurred dressing frequently. or more 7 days or 2 75% granu- and < 1 cm ≥ 2 , or > 1 cm deep; 2 Inclusion criteria: with one or Patients > 18 years sores. pressure more Stage II, III or IV only. size between Wound 2 cm and 80 > 80 cm deep; non-infected; clinically eschar-free, with wound lation base with fixed margins; adequate nutritional mouth tube or by intake hyperalimentation. Exclusion criteria: or Stage IV Stage I sores with exposed tendon or sores bone; size < 2 cm wound wounds covered with necrotic covered wounds base wound eschar or necrotic containing > 25% slough; diagnosis or suspicion of at study wound osteomyelitis site; carcinomatosis; or signs clinical symptoms of wound infection; inadequate nutritional intake; tract, sinus tunnelling margin or > 0.5 cm of wound undermining. 47 1996 et al, Hydrogel dressings versus traditional or control treatments versus or control dressings traditional Hydrogel Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Brown-Etris Brown-Etris USA type: Wound sores. Pressure Method of randomisation: Not stated; stratification to according occurred and stage. surface area outcome: Objective assessed reduction Area planimetry gravimetric by tracing onto with wound plastic film and photo- graphy. Independent analysis biostatistical analysis by firm. of Change in level margin under- wound mining assessed. Setting and length of treatment: trial. A multicentre Participation was until 10 weeks, or when treat- ment change was indicated healed, or the wound came first. whichever
56 7 TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) continued Patients appeared to prefer I to prefer Patients appeared because of the lack (hydrogel) of odour, cushioning and lightness. in C (hydro- The gel layer to degrade colloid) was found which necessitated easily mechanical cleansing of the wound, which damaged the healing tissue layers. I: One patient was excluded because enlarged the wound > 10% per day. by One patient was excluded because the wound by decreased > 25% per day. C:Three patients excluded were because their enlarged wounds > 10% per day. by One patient was excluded because the wound by decreased > 25% per day. ): 2 ): 2 I: 3.5 C: 5.5 area in wound Mean reduction (cm at 60 days Mean wound area at 60 days (cm at 60 days area Mean wound I: 7.5 (68% reduction) C: 3.7 (40% reduction) healed Number of wounds at 60 days: I: 26/60 (43%) C: 15/63 (24%) days: Mean treatment I: 12 C: 11.3 ): 2 IC All groups (range, 30–98) Mean area of pressure sore (cm sore of pressure Mean area I: 11.0 (0.2–100 range) C: 9.2 (0.4– 64 range) Other characteristics: Mean age (years):M:F ratio: 75 Ratio of grade I:II ulcers: 1:1.6 Serum albumin (gm/dl):No. of stage I wounds: 27No. of stage II wounds: 1:1.3 35 1:1.6 31 36 was a significant difference There the age of patients in between 2.8 setting (69 years) the acute care 2.7 facilities and the extended care (83 years). , 2 = 63 wounds. n = 60 wounds. instead utilising 4 x inch dressings. changed usually were Dressings and washed in 3–4 days every reapplication.saline before All on the same type of patients lay mattresses. pressure-reducing All wounds were initially cleansed initially were All wounds and saline. peroxide with hydrogen skin were Patients with an oily a 1.25 inch for to allow degreased the wound. adhesion belt around Although this was not maintained was > 20 cm the wound where I: (Biofilm), dressing Hydrogel n C: dressings Hydrocolloid (Duoderm), ). 2 Inclusion criteria: facilities Patients in acute care homes with stage I and nursing ulcers sores or II pressure (size > 2 cm Exclusion criteria: Receiving radiation therapy; infection, tracts or fistulas in the sinus wound; a blood sugar level > 180 mg/dl; no improved status. nutritional 46 Hydrogel dressings versus traditional or control treatments versus or control dressings traditional Hydrogel 1990 et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Darkovich Darkovich USA type: Wound sores. Pressure Method of randomisation: Not stated. outcomes: Objective of the wound Perimeter was traced and in some to cases photographed determine size.The number healed was of wounds also recorded. Setting and length of treatment: trial unless 60-day Maximum healed,wound patient discharged or withdrawn clinician.by Measurements at each dressing taken change or at least weekly intervals.
TABLE 7 contd TABLE 57 Appendix 6 continued = 0.046). p = 0.15; NS). p Estimated manpower was general care for required 5.8 and 2.72 hr/day/patient, with less time used in the arm ( treatment duration of stay The average in the I (Hydro- was 4.8 days in the and 6.4 days gel) group group treatment) C (standard ( I: No withdrawals. C: patients Two died from pneumonia. IC ): 2 = 0.87) = 0.02) p p = 0.94;Wilcoxon sign test or p Mean size of pressure sore at final sore Mean size of pressure assessment (cm ( rank sum test were Mann-Whitney statistical analysis) used for at area Mean % change in wound final assessment: I: –56% C: +10% (HK$): Cost of local treatment Nursing time:Pharmaceutical items: 785 1658 less with I was statistically Treatment with C. than treatment expensive 1884 280 C: 20.5 ( I: 15.9 ( IC Mean wound area (cm2): area Mean wound I: 36.3 (range, 1.5–160) C: 18.6 (range, 1.0–72.5) Other characteristics: Mean age (years):M:F ratio: 80Mean duration 78.4 (months): significant No statistically 1:4 the two between difference 1:1.5 was reported. groups Not stated = 8 wounds n Eusol for at Eusol for 2 / 1 = 12 wounds = 12 wounds n ), ® from five patients. five from C:The was cleansed sore dependent with a frequency on the amount of discharge (for cleansing was sores uninfected while saline only by performed cleansing was sores infected for with performed from five patients. five from the same All patients received which included: general treatment bathing at least on alternate days; 2 hrs (7 am–11 pm) turning every 3 hrs (11 pm–7 am); and every of fluid at least 1.5 litres given unless contraindications per day existed; support nutritional and monitoring; to be vitamins only of deficiency; if evidence given anti- to be given agents only microbial of systemic sepsis. with evidence most three times a day followed by followed times a day most three rinsing with saline), Treatment: I: (Intrasite dressing Hydrogel with a hydrophilic Gel) covered dressing foam polyurethane (Allevyn ; grade I 2 1 cm ≤ 65 years with pressure 65 years ≥ Inclusion criteria: Hospitalised post acute patients aged sores. Exclusion criteria: sores Pressure sores; for patients considered only;supportive care patients agents for antimicrobial requiring management of systemic sores; pressure from infection did not give patients (or carers) the study. consent for Hydrogel dressings versus traditional or control treatments versus or control dressings traditional Hydrogel 48 1996 , Ltd Healthcare Smith & Nephew ® et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Allevyn Lum (unpublished) Hong Kong type: Wound sores. Pressure Method of randomisation: Patients with an odd were admission number assigned to the control group, those with an even assigned to were number group. the treatment outcome: Objective size was measured. Wound the time spent In addition staff and cost nursing by of pharmaceuticals was monitored. Setting and length of treatment: hospital. Post-acute progression Treatment weekly was monitored and a final assessment was made after 8 weeks, complete healing, discharge or death.
58 7 contd TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) One patient in the C1 (hydro- group treatment colloid) had mild irritation, and another had minor sensitivity to the C1 dressing. One case of inflammation in the I (hydrogel) occurred group, and another patient had excoriation which was to I. related possibly no adverse were There to C2 (saline). reactions Three patients Three not evaluable were and their data are in not presented the baseline characteristics. I:Three patients omitted from were the final analysis. were No reasons these for given withdrawals. C1: No withdrawals. C2: No withdrawals. > 0.05; non-parametric test) > 0.05; non-parametric test) p p Mean % reduction in wound area area in wound Mean % reduction per week: I: 8 (14.8 SD) C1: 3.3 (32.7 SD) C2: 5.1 (14.8 SD) ( area in wound Median % reduction per week: I: 5.6 C1: 7.4 C2: 7.0 ( ): 2 IC1C2 Mean wound area (cm area Mean wound Not stated. Other characteristics: Mean age (years): 56.7 63.1M:F ratio: 57.2 stage: Wound II: 1:3.6 1:5.6 1:9.5 III:Race (patients): Black:White:Hispanic: 8 14 significant No statistically 9 13 the three between differences 4 18 17 5 was reported. groups 1 3 16 1 14 6 0 = 23. n = 20. n = 21. C2: Saline solution and moistened gauze, times a day, changed three n changed either by were Dressings giver,the patient or care after appropriate had received they instructions. in to enrolled 67 patients were the trial; for analysed data were 64. only Treatment: I: (Clearsite), dressing Hydrogel changed twice a week, C1: dressing Hydrocolloid (DuoDerm), changed twice a week, 18 years and 18 years ≥ 1.5 cm x 0.5 and ≥ 10 cm x 10 cm were included. 10 cm x were Inclusion criteria: In-patients and out-patients with sores. stage II and III pressure Sores ≥ Patients were expectancy of at had a life least 2 months. Exclusion criteria: or those with Stage IV wounds tendon, bone, capsule, or fascia exposure; pregnancy; chemo- therapy; infection; prior wound undermining of the extensive ulcer (> 1 cm);AIDS; patients > 10 mg of receiving corticosteroids. Hydrogel dressings versus traditional or control treatments versus or control dressings traditional Hydrogel 49 1993 et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Mulder USA type: Wound sores. Pressure Method of randomisation: Computer allocation. outcomes: Objective perimeter was The wound traced on to a transparency determined and the area computer analysis.by In the largest length, addition width and depth of each and was measured wound at taken a photograph each assessment. Setting and length of treatment: trial at three A multicentre independent sites.Assess- size were ments of wound 8 weeks for made weekly was or until the wound healed. possible, Where each patient was evaluated the same investigator by the trial.throughout Efforts made to standardise were observations between investigators.
TABLE 7 contd TABLE 59 Appendix 6 continued I (enzyme) was associated in wound with an increase size.This be due to may debride- wound excessive ment, or inhibition of tissue the enzyme. by growth Pseudo- I:Three patients withdrawn were because of unsuccessful treatment. In one of these patients a skin reaction on the occurred heel after 3 weeks of treatment. In another patient developed necrosis to 8x its original size. In the third patient monas aeruginosa developed infection after 6 weeks. C: No withdrawals All withdrawals included in were the analysis. % median change in wound area at area % median change in wound final assessment: I: +18.7% C: –2.4% ): 2 IC ): 4 5 n Median age (years):M:F ratio: 86Diabetes mellitus ( 81 1:3.7 1:1.8 I: 4.2 (range, 1.2–18.2) C: 5.8 (range, 1.2–26.0) Other characteristics: Median wound area (cm area Median wound ) 2 = 14. = 14. n n C: (400 mg ZnO/cm Zinc oxide Treatment: I: Streptokinase/streptodornase (Varidase enzyme preparation to a sterile gauze applied Topical) compress. changed Dressings twice daily, applied to a sterile gauze applied compress. changed Dressings once daily, with secured were All dressings acrylic-basedporous tapes.Where were existed they wounds multiple uniformly,all treated the but only largest was monitored. loosely Prior to treatment material was attached necrotic removed, but no surgical debride- thereafter. ment was performed antibiotics. No patients received its usual followed Nursing care procedure. Inclusion criteria: in-patients and out- Elderly patients with one or more sores. pressure necrotic Exclusion criteria: Not stated. Enzyme preparations versus traditional or control treatments versusor control traditional Enzyme preparations 9 Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Ågren and Strömberg, Ågren 1985 Sweden type: Wound sores. Pressure Method of randomisation: Patients consecutively what matched in pairs (for not stated). Each member of the pair was randomly allocated to one of the treatments. two outcome: Objective was traced area Wound by and the size measured planimetry. A photograph at each was taken assessment. Setting and length of treatment: Single-blind trial for 8 weeks. One of the authors was responsible measuring all the for intervals. at weekly wounds An independent surgeon another hospital from assessed the photographs.
60 8 TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) continued Pain caused by treatment was treatment Pain caused by insufficient to cause in either group. withdrawal One wound treated with I treated One wound (enzyme) experienced mild bleeding and a burning sensation. C: One patient defaulted after of the first week and treatment was excluded from the analysis. I: No withdrawals. I: Patients were from removed 30 6 to day day (termination of the trial). C: Patients were from withdrawn 10. 3 to day day No patient in this continued group after to be treated 10 days. < 0.05; I vs. p < 0.01; statistical test p Mean % change in wound volume volume Mean % change in wound at completion of the trial: I: +13.14 (59.8 SD) C: +78.79 (94.6 SD) in both groups sores (Pressure in size.) increased method not stated) healed after Number of wounds 6 weeks: C: 0/9 (0%) I: 2/10 (20%) Approximate mean wound area area mean wound Approximate at 6 weeks: C: 8.40 (9.79 SD) I: 5.37 (7.82 SD) (C vs. baseline, baseline, ): 3 ): 2 All groups All groups < 0.01). ): 2 p Approximate mean wound mean wound Approximate (cm area C: 14.50 (22.13 SD) I: 15.82 (19.67 SD) Other characteristics: Mean age (years):M:F ratio: 61 was no statistical difference There with groups the two between baseline to the given respect 1:9.5 characteristics. Not stated. (cm volume Mean wound 11 patients with 28 advanced included in were sores pressure the study. disease All had chronic con- in poor physical and were dition. had neoplastic disease; Four heart had atherosclerotic four disease or had a cerebrovascular accident, or both; had Parkin- two disease;son’s and one had a neck fracture. femoral I: 15.44 (19.92 SD) C: 1.25 (1.62 SD) Other characteristics: Mean age (years): 67.6 (13.7 SD) M:F ratio: is volume Baseline wound between different significantly ( groups 1:2.7 Mean wound area (cm area Mean wound = 17. n = 9. = 10. n n = 11. n Treatment: C: A Formulation – acetate and hydrocortisone palitate, neomycin I: B (Chymacort). Formulation was the same as This treatment A but with the formulation enzymes of the pancreatic addition and trypsinchymotrypsin in the of 1:6, proportion with a covered were All wounds dressing;non-adherent a tube gauze; an absorbent gauze dressing area; the wound pad over and a 4-inch crêpe bandage finally the toes to knee from applied changed at the were Dressings clinic. instructed not Patients were and to with dressings to interfere ambulant. fully remain Before application of either application Before was washed the wound treatment with sterile saline (pH 7.5). Each once was applied application unless more to each wound daily cleansing was required frequent because of contamination from incontinence of urine, faeces or both. with a covered were All wounds sterile gauze pad. Treatment: I: Collagenase enzyme preparation at 250 units per (Santyl) applied gram of white petrolatum, C: Placebo (heat-inactivated Santyl) as in the same proportions applied I, for Inclusion criteria: Patients with leg ulcers to be post- considered thrombotic. Exclusion criteria: Not stated. Inclusion criteria: Patients with advanced sores. pressure Exclusion criteria: Not stated. Enzyme preparations versus traditional or control treatments versusor control traditional Enzyme preparations 50 51 Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Lee & Ambrus,Lee & 1975 USA type:Wound sores. Pressure Method of randomisation: Not stated. outcome:Objective Two diameters of the wound and a colour measured taken.photograph In addition mould was made a volume No.1, or Kerr with Jeltrate determined and the volume water displacement. by Setting and length of treat- ment: Setting not stated. for treated Patients were or until complica- 4 weeks or the tions developed patient died. Measurements and on weekly taken where completion of the study. Gordon, 1975 UK type: Wound Leg ulcers. Method of randomisation: Not stated. outcome: Objective was estimated area Wound measuring the two by largest diameters and together.multiplying Each was photographed. wound Setting and length of treatment: Patients attend- leg ulcer ing a clinic for treatment. Assessments made at weekly were intervals 6 weeks. for
TABLE 8 contd TABLE 61 Appendix 6 continued patients with any of the patients with any treatments. The authors state that debridement with I (enzyme) the wound helped to make to skin grafting. receptive of complex leg The presence extended the have ulcers may needed length of treatment prior to grafting. No withdrawals. reported by No side-effects Four patients Four excluded were the analysis from reasons for to the unrelated therapy. These were withdrawals not included in the baseline data. % of wounds successfully grafted or successfully % of wounds needing no grafting: I: 69 (95% CI: 44, 86) C: 35 (95% CI: 17, 59) % of complex ulcers successfully grafted or needing no grafting: I: 50 (95% CI: 19, 81) C: 38 (95% CI: 14, 69) stasis ulcers successfully % of venous grafted or needing no grafting: I: 80 (95% CI: 49, 94) C: 33 (95% CI: 12, 65) Number of wounds healed Number of wounds at 4 weeks: I: 1/11 (9%) C: 0/9 (0%) Number of patients with healed ulcers at 4 weeks: I: 1/5 (20%) C: 0/5 (0%) ): IC IC 2 of surface √ I: 25.7 (2, 273 range) C: 20.2 (2, 48 range) Other characteristics: Mean age (years):M:F ratio: 75.9Mean duration 73.5 (months): stasis ulcers:Venous 1:2.2 10Complex ulcers: 1:1.3 19.8 9 only Baseline characteristics are 6 13.8 30 of the 34 patients. for available 8 Mean wound area (cm area Mean wound Mean wound size = Mean wound (cm): area I: 3.2 C: 2.4 between No statistical difference ulcer size. for the groups Other characteristics: Age range (years): 28–59M:F ratio: 32–70 Mean duration (months): Not stated Not stated = 16. n = 9 wounds from = 9 wounds n = 11 wounds from five from = 11 wounds n = 17. n five patients. five I: Collagenase enzyme preparation after a saline daily (Santyl) applied with a dry wash and covered dressing, patients. C: Sugar and egg white applied after a saline wash. Changed four times a day. to dry and Allowed not covered, Treatment: I: (Elase; Enzyme preparation a containing powder freeze-dried and 15,000 U 25 U fibrinolysin desoxyribonuclease), C: (freeze-dried Placebo powder ingredi- without active powder ents), first cleaned with were Wounds saline.The and placebo treatment in 30 ml dissolved were powders of saline and used to soak gauze pads.The pad was applied soaked with a to the ulcer and covered and bandaged. paraffin dressing changed three were Dressings familiar nurses by times a day with the procedures. the for enrolled 34 patients were trial with 37 leg ulcers. Patient available only allocation details are those patients completing for the study. Inclusion criteria: leg ulcers Patients with chronic skin grafting. for referred Exclusion criteria: Pregnancy; to hypersensitivity Elase; by in ability to be treated method. the usual pregrafting Inclusion criteria: sores Patients with pressure home. in a nursing residing Exclusion criteria: Not stated. 37 52 Enzyme preparations versus traditional or control treatments versusor control traditional Enzyme preparations 1987 et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Parish and Collins, 1979 USA type: Wound sores. Pressure Method of randomisation: Not stated. outcome: Objective Number of ulcers healed. Setting and length of treat- ment: (nursing Community trial. home) 4-week Westerhof Westerhof The Netherlands type: Wound leg ulcers. Chronic Method of randomisation: Patients allotted a code number, one per leg.Treat- also random- ments were ised and codes concealed the investigators. from outcome: Objective healed Number of wounds skin where and the number grafting was a success.A was colour photograph standardised by taken at each photography assessment period. Setting and length of treat- ment: Assessments were made twice weekly. Both the patient and assessor blind to the treatment were given. Final assessment was independent two made by was investigators.Treatment until the assessor continued was confident to initiate grafting.
62 8 contd TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) Common adverse effects Common adverse seen in both groups were macer- usually which were ation of the skin edges. The compliance of patients as excellent. was reported I: patients Four because of > 50% of in area increase tissue, necrotic associated pain and oedema or signs of cellulitis. C: patients Five because of > 50% of in area increase tissue, necrotic associated pain and oedema or signs of cellulitis. = 18) = 17) n n C: ( 5% increase Number of wounds reduced by by reduced Number of wounds at 5 weeks: 50% or more I: 14/21 (67%) C: 6/21 (29%) area Mean % change in wound at 5 weeks: I: ( 60% reduction ): IC 2 ): 2 I: 1.5 (0.5, 10.5 SD) C: 1.6 (0.9, 19.2 SD) Other characteristics: Mean age (years):M:F ratio: 63Mean duration of diabetes:(years) 62 22Insulin: 1:1.2Systolic toe 1:0.5 19 (mmHg):pressure Systolic ankle 66 (mmHg):pressure 17 104 68 114 18 I: 2.2 (1, 10.5 SD) C: 2.2 (0.9, 20.4 SD) of wound area Mean necrotic (cm Mean wound area (cm area Mean wound = 22. = 22. n n ), ® Treatment: I:Adhesive tape zinc oxide (MeZinc C: dressing Hydrocolloid (DuoDerm), the same offered All patients were treatment:additional the foot and external was corrected wear on the ulcer relieved. pressure cleaned with sterile Ulcers were to according saline and dressed guidelines. the manufacturers for changed daily were Dressings 3 days and every the first week afterwards. No surgical debridement was allowed. in area with in area 2 Inclusion criteria: diabetic Patients with necrotic ulcers (superficial full foot ankle thickness skin ulcer below and with systolic toe pressure 45 mmHg or absence of above cutaneous erythema; ulcers 1–25 cm between > 50% of area covered with covered > 50% of area tissue).Wheredry/wet necrotic than one ulcer was more there the largest was chosen. Exclusion criteria: patch test;Positive clinical signs of cellulitis; the ulcers where of these dressings application was inappropriate. 53 1990 Zinc oxide tape versustreatment Zinc oxide traditional et al, , Health Care Molnlycke ® Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results WithdrawalsMeZinc Comments Apelqvist Sweden type: Wound ulcers. Diabetic foot Method of randomisation: Not stated. outcomes: Objective tissue of necrotic Area and a colour measured taken.Aphotograph successful outcome was had area when the necrotic 50% or more. by reduced blinded. Evaluations were Setting and length of treatment: Outpatient clinic. made Assessments were 5 weeks. for weekly
TABLE 9 TABLE 63 Appendix 6 continued Both agents were similar Both agents were time and pain dressing for the patient. experienced by of one week Cost for per person: treatment I: £4.68 C: £3.33 The antibacterial properties of I (cadexomer polysac- of little charide) were advantage in the treatment leg chronic of resistant ulcers. One patient was from withdrawn 4 the trial at week because of poor compliance. Data available only are the 42 patients for the for remaining treatment 6-week period. I: 13 patients. C: 11 patients. patients died; Three withdrawn 18 were request;by had two allergies; one was lost to follow-up. -test) t -test). t > 0.05; sample two > 0.05; analysis). chi-squared p p Number of wounds healed Number of wounds at 10 weeks: I: 14/46 (30%) C: 14/49 (29%) ( % reduction in mean wound area area in mean wound % reduction at 6 weeks: I: 4 (95% CI: 5, –14) C: 3 (95% CI: 4, –9) ( No significant change in wound No significant change in wound 6 weeks baseline over size from (paired Number of wounds healed at Number of wounds 6 weeks: I: 0/21 (0%) C: 0/21 (0%) ): ): 2 2 IC IC < 0.01; Mann-Whitney p > 0.05; U test). Mann-Whitney p I: 19.7 (19.8 SD) C: 25.5 (29.5 SD) Other characteristics: Median age (years): 70M:F ratio:Median duration 68 (months):Ischaemia: 1:3.5 between No statistical differences 1:4.2 groups.the two only Data are 75 42 of the 43 patients. for available 6 61 5 Median wound area (cm area Median wound Median wound area (cm area Median wound I: 2.8 (range, 0.2–50.5) C: 2.6 (range, 0.4–74.4) ( Other characteristics: Mean age (years):M:F ratio: 70Mean duration (months): 77 between No statistical differences 1:1.4 with the exception the groups 1:2.8 of age ( 12.0U test). 24.0 = 46. n = 49. n = 21. n = 21. n I: Cadexomer iodine (Iodosorb) powder polysaccharide to a depth of 3 mm, applied C: dressing Hydrogel (Scherisorb/Intrasite) applied to a depth of 5 mm, covered were Both treatments with gauze, dress- a non-adherent ing and padding. A crêpe bandage, ulcer a those with a venous or for supplied if stocking were Tubigrip them. to wear the patient agreed I: Cadexomer iodine powder polysaccharide (Iodosorb), C: Dextranomer polysaccharide beads (Debrisan), with a covered Both agents were pad,non-adhesive cotton-wool wadding, stockinette and a firm elastic bandage. a occurred If bacterial infection course of oral antibiotics 2-week was allowed. = 43). n Inclusion criteria: Out-patients with leg ulcers of various aetiologies. Exclusion criteria: receiving steroids, cytotoxics, or antibiotics; poor nutritional status; abnormal thyroid function; unable to give consent. informed Inclusion criteria: Out-patients with unresponsive leg ulcers for venous > 3 months. Patients with insufficiency ischaemic venous included. were Exclusion criteria: Not stated recruited 56 patients were and observed on a variety of treatments.Thosestandard not were after 6 weeks improving admitted to the trial ( 55 13 Cadexomer iodine polysaccharide versus other debriding agents 1987 et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Moss UK type: Wound leg ulcers. Venous Method of randomisation: Not stated. outcomes: Objective perimeter The wound onto poly- was drawn and the thene squares calculated by area computer planimetry. of wounds The number healed was also recorded. Setting and length of treatment: trial.6-week Measure- at 0,ments taken 2, 4 and 6 weeks.After 6 weeks allowed patients were to cross-over. UK type: Wound Leg ulcers. Method of randomisation: Computer allocation. outcome: Objective healed. Number of wounds Setting and length of treatment: trial.10-week Measure- for weekly ments taken and then once 6 weeks at 8 and 10 weeks thereafter. Stewart and Leaper, 1987
64 10 TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) Three patients in the I group Three (cadexomer polysaccharide) and one in the C group (dextranomer polysaccharide) complained of pain during the period.Thistreatment pain of the in two was severe patients, one in each group,treatment and resulted the study. from in withdrawal I:Three patients had bacterial infection; one patient experi- enced pain; and in one patient the increased wound in size. patients had C:Two bacterial infection; and one patient experienced pain. Number of wounds healed Number of wounds at 8 weeks: I: 7/14 (50%) C: 2/13 (15%) IC Mean wound area: Mean wound Not stated. Other characteristics: Mean age (years):M:F ratio: 67.7Mean duration 68.8 (months): 1:2.5 1:3.3 54.8 12.2 = 13. n = 14. n C: Dextranomer polysaccharide (Debrisan), powder to a depth Both agents applied with a clean of 3 mm and covered compress. bandage A compression the leg. around was applied by changed once daily Dressings soaking. cleansed were Wounds a applying before mechanically dressing. new Treatment: I: Cadexomer iodine powder polysaccharide (Iodosorb), Inclusion criteria: with Out-patients > 18 years exuding leg ulcers chronic at one of three presenting clinics. Exclusion criteria: Insulin-dependent diabetes mellitus; rheumatoid arthritis tissue and other connective diseases; or known goitre allergy to iodine. Cadexomer iodine polysaccharide versus other debriding agents 54 Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Tarvainen, 1988 Finland type: Wound leg ulcers. Chronic Method of randomisation: Sealed envelopes. outcome: Objective Number of ulcers healed. Length of treatment: Outpatient clinic. 8-week trial. taken Measurements at 0, 2, 5 and 8 weeks.
TABLE 10 contd TABLE 65 Appendix 6 continued Five adverse events were events adverse Five reported, one in the C group in the I and four (hydrogel) (dextranomer polysac- group charide).The one con- only related to be dressing sidered one was pain reported by patient in the I group. to be easier C was found than the I. and remove apply to be C was also found associated with less pain. I: 19 lost to follow- up, died, two four reactions adverse to (one related pain on application of the agent, no details on the other three). C: 11 lost to follow-up, two died, and one reaction. adverse IC = 0.03; Rank Sum test). Wilcoxon = 0.20; Rank Sum test) Wilcoxon p p Median % reduction in wound area in wound Median % reduction at 21 days: I: 7 (–340, 98% range) C: 35 (–185, 91% range) ( of degree by Number of wounds cleansing at 21 days: 100%75–99%50–74%25–49%0–25%Deteriorated 14 15 in 13 median % reduction Overall 13 20 non-viable tissue at 21 days: 6 13 12 I: 62 (–340, 100 range) 7C: 74 (–103, 8 100 range) 9 ( 5 IC IC IC 1825 15 25 25 27 1827 15 23 24 28 2 2 2 2 2 2 Wound area: Wound < 4 cm 4–13 cm > 13 cm Non-viable tissue area: < 3 cm 3–9 cm > 9 cm Other characteristics: Median age (years)M:F ratio 81 duration: Sore 79 < 1 month 1–3 months 1:1> 3 months 1:1.4 22 grade: Sore 35110 24 1121016 28 34538 15 41213 well were Authors state groups matched. written All patients gave of capable consent and were participating in the trial. = 68. n = 67. n The two interventions were The two with the in accordance applied instructions.Anmanufacturers’ absorbent plastic film dressing (Melolin) was used as a standard- ised secondaryboth for dressing treatments. than a patient had more Where the largest was only one wound in the trial.evaluated Other with the treated were wounds if this same randomised dressing appropriate was considered the clinical investigator. by Treatment: I: Dextranomer polysaccharide paste (Debrisan), C: (Intrasite dressing Hydrogel Gel), 16 years with pressure with pressure 16 years Inclusion criteria: patients Male and female ≥ that area in any present sores needed cleansing. Exclusion criteria: Pregnancy, immunodeficiency, of the wound, clinical infection more black eschar covering hard than 20% of the wound, diabetes, the demands inability to follow any in for of the protocol reason, non-consenting patients. 56 Dextranomer polysaccharideDextranomer versus other debriding agents 1996 et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Colin France type: Wound sores. Pressure Method of randomisation: Not stated. outcome: Objective in reduction Percentage of non-viable tissue area x (% yellow area (wound + % black tissue) x 1/100). taken were Photographs at the initial and final assessment. Setting and length of treatment: Open, multicentre, multi- national, trial. parallel group were centres Six different with approximately involved of patients equal numbers in each trial.Assessments 7 days made every were was until the wound cleansed or on the completion of 21 days.
66 11 TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) continued No adverse effects were effects No adverse reported in patients treated with I (dextranomer patients polysaccharide).Two (enzyme) in the C group experienced pain, and in one it was severe. with I treated One wound in size.Thereincreased was reported in no size increase the C group. patients with any of the patients with any treatments. I:Three patients withdrawn,were no given. reasons was C:Treatment interrupted 7 the day between 14 in one and day patient because of pain. No withdrawals by reported No side-effects Number of wounds healed at Number of wounds 4 weeks: I: 6/14 (43%) C: 1/11 (9%) Number of patients with healed at 4 weeks: wounds I: 4/7 (57%) C: 1/5 (20%) Mean reduction in wound size in wound Mean reduction (mm): at 14 days I: 9.3 C: 9.8 size at in wound Mean % reduction 14 days: I: 14.5 C: 11.9 (NS) IC IC of surface √ Mean wound size: Mean wound Not stated. Other characteristics: Median age (years): 68.8M:F ratio 65.1 Ulcer duration: < 1 month1–12 months> 12 months Not stated exuding and were All wounds 1 8 also necrotic. of them were many 9 1 6 6 Mean wound size = Mean wound (cm): area I: 4.5 C: 3.2 between No statistical difference size. wound for the groups Other characteristics: Age range (years) 29–57M:F ratio: 28–59 Mean duration (months): Not stated Not stated = 18. n = 11 wounds n = 13. n = 14 wounds from seven from = 14 wounds Treatment: I: Dextranomer polysaccharide to a beads (Debrisan) applied depth of at least 3 mm covered with a dry dressing. Changed 1–3 depending on exudate,times daily n patients. C: Collagenase enzyme daily (Santyl) applied preparation after a saline wash and covered with a dry dressing, patients. five from C: Streptokinase/streptodornase (Varidase enzyme preparation Topical).One ampule was diluted saline. with 20 ml of physiological The solution was either placed or a onto the wound directly was soaked sterile compress in the solution prior to application. sheet was placed A waterproof on top of the compress.Appli- changed twice cations were daily, was excised from necrosis All hard before in both groups wounds commencement of the trial. Treatment: I: Dextranomer polysaccharide beads (Debrisan) in a glycerin to the paste (ratio 4:1) applied at least 3 mm in a layer wound thick.The was covered wound with a dry sterile compress. rinsing by removed were Dressings with water or saline irrigation changed twice a were Dressings and dextranomer was applied day moist wounds, to only Inclusion criteria: sores, Patients with pressure home. in a nursing residing Exclusion criteria: Not stated Inclusion criteria: leg In-patients with venous ulcers. Exclusion criteria: Not stated. 37 Dextranomer polysaccharideDextranomer versus other debriding agents 58 1981 et al, Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Parish and Collins, 1979 USA type: Wound sores. Pressure Method of randomisation: Not stated. outcome: Objective healed. Number of wounds Setting and length of treatment: (nursing Community trial. home) 4-week Hulkko Hulkko Finland type: Wound leg ulcers. Venous Method of randomisation: Not stated. outcome: Objective Longest and shortest diameters of each wound measured,were the pro- duct of which was used of wound as a measure also were size.Wounds at each photographed assessment. Setting and length of treatment: Hospital-based trial. made Assessments were at 7 and 14 days.
TABLE 11 contd TABLE 67 Appendix 6 C (hydrogel) had to be C (hydrogel) frequently changed more than I (dextranomer polysaccharide). with frequent However, even the cost of C per dressing the I. patient was less than for Mean cost per patient: I: £44.70 C: £22.60 (NB. successes costed Only not failures) Up to 14 days: I: because of three difficulty in applying the dressing. Classed as failures in the results. C: one patient because the case were report forms mislaid. Up to 28 days: Withdrawals the over occurred period follow-up patients four leaving and in the C group patients in the two I group. = 0.008; exact test) Fischer’s p Number of wounds cleansed Number of wounds at 14 days: I: 1/20 (5%) C: 8/20 (40%) ( all ulcers were After 14 days no showing reassessed.Wounds of debridement were evidence and withdrawn. classed as failures were wounds The remaining another 14 days. for followed cleansed after wounds Additional at 28 days: follow-up I1: – 5/20; 4/20 (overall 25%) C: – 8/20; 0/20 (overall 40%) ): IC 2 I: 15.6 (16.2 SD; range 1.5–68.9) C: 22.2 (23.4 SD; range, 2.6–91.4) in slough: covered area % wound I: 75.3 (22.4 SD; range, 20–100) C: 73.5 (29.7 SD; range, 20–100) Other characteristics: Mean age (years):M:F ratio: 81.0Ratio of grade 83.5 3:4 ulcers 39 of the for given only are Values 1:5.740 patients entering the trial. 1:3.8 5.7:1 3.8:1 Mean wound area (cm area Mean wound = 20. n = 20. Treatment: I: Dextranomer polysaccharide beads (Debrisan) made into a 600 glycol paste with polyethylene paste was applied and water.The a layer to a depth of 10 mm over net, of polyamide C: (Intrasite dressing Hydrogel to a depth of 5 mm. Gel) applied plastic with a perforated Covered held in film absorbent dressing or a bandage,place with tape n changed as were Dressings required, cleansed and the wound reapplication. with saline before Inclusion criteria: Hospitalised patients with sores. grade 3 or 4 pressure had to be covered The wounds with or partially covered slough. yellow/brown Exclusion criteria: Age < 16 years, insulin- dependent diabetes, immuno- suppression, pregnancy, cellulitis of the surrounding and redness of infection). tissue (indicative 57 Dextranomer polysaccharideDextranomer versus other debriding agents Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Thomas and Fear, 1993 UK type: Wound sores. Pressure Method of randomisation: Computer-generated randomisation code. outcome: Objective of wounds The number cleansed.fully Cleansing was determined by measuring the % of total covered area wound in slough. Setting and length of treatment: trial.After2-week 14 days showing those wounds were no improvement while the withdrawn were wounds remaining a further for followed 14 days.
68 11 contd TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1) person from baseline to final person from assessment are: I £43.25 for C.and £40.25 for reported were Both products and convenient to be safe to use. Not stated. per treatment costs for Total ): 2 ): 2 = 0.25; Rank Sum test) Wilcoxon = 0.358; Rank Sum test) Wilcoxon = 0.62; Rank Sum test) Wilcoxon p p p I: 1.511 C: 1.118 ( % reduction in mean area of wound in mean area % reduction after 21 days: I: 14 C: 10 ( % reduction in mean area of ulcer in mean area % reduction in slough after 21 days: covered I: 37 C: 32 I: 0.744 C: 0.540 ( Median reduction in slough over in slough over Median reduction period (cm a 7-day Median reduction in wound area area in wound Median reduction period (cm a 7-day over ): IC 2 Mean wound area (cm area Mean wound I: 24.9 C: 18.7 Other characteristics: Mean age (years):M:F ratio: 77 since first Years ulcer: 76 in treated 88% of patients were 1:2.3the community. 1:3.5 8 4 = 32. n = 30. Both hydrogels were covered with covered were Both hydrogels either an absorbent plastic film (Melolin) or a knitted dressing (Tricotex). viscose dressing changed daily. were Dressings Treatment: I: (Granugel), dressing Hydrogel n C: (Intrasite dressing Hydrogel Gel), Inclusion criteria: leg ulcers Patients with sloughy of wound (> 10% surface area in slough). covered Exclusion criteria: Not stated. Hydrogel dressings versus hydrogel dressings versus dressings hydrogel Hydrogel 59 Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Gibson, 1995 UK type: Wound Leg ulcers. Method of randomisation: Not stated. outcomes: Objective area,Wound and the area in covered of wound slough. Setting and length of treatment: parallel trial A multicentre or 21 days conducted for was until the wound to be cleansed. considered taken were Measurements at 0, or 10 and 21 days was clean. when the wound
TABLE 12 TABLE 69 Appendix 6 13/19 patients in the I group (Trypure), and 4/21 patients (Varidase) in the C group reported pain associated with the dressing. Pain was in four reported a severe of the I patients, while pain as mild in was recorded the C group. In nine patients treated treated with I and in three with C, pain chronic in intensity. increased I:Three patients because of pain associated with the dressing.Another patients were five due to withdrawn being the wound cleansed. C: 13 patients were due to withdrawn cleansing. wound -test) t < 0.01 for both treatments < 0.01 for p Mean wound area at 21 days: area Mean wound in ulcer area. was no decrease There of total wound area % granulated at 14 days: area I: 56% C: 60% ( with baseline. compared Student’s cleansed fully Number of wounds at 21 days: I: 5/19 (26%) C: 13/21 (62%) at area A comparison of granulated was not made due to the 21 days rates. high withdrawal IC Mean wound area: Mean wound Not stated. of total ulcer area % granulated area: I: 25% C: 22% Other characteristics: Mean age (years):M:F ratio: 68.9Mean duration 69.7 (months): leg ulcers:Venous 1:1.4Arterial/venous and 14 1:4.3 arterial leg ulcers:Other leg ulcers, 15 Not stated 3vasulitis, decubitus:Concomitant 2 3 diabetes mellitus (patients): 3 was left Concomitant treatment the trial unchanged throughout period. 7 6 = 21. n = 19. Treatment: enzyme preparation I:Trypsin (Trypure). One ampule (50 mg) with 15 ml of saline, was mixed n C: Streptokinase/streptodornase (Varidase enzyme preparation Topical).One ampule (streptokinase-streptodornase 100,000 U and 25,000 with was mixed respectively) 20 ml of saline, of the test application Before treatment, each ulcer was washed with distilled water. Thereafter, impregnated compresses wet applied with enzymes were twice daily. 2 , > 400 cm 2 Inclusion criteria: Patients with leg ulcers covered with pus and debris. Exclusion criteria: < 3 cm Wounds or more than 6 cm deep. or more 60 Enzyme preparations versus enzyme preparations Enzyme preparations Study and design criteria Inclusion/exclusion Intervention details Baseline characteristics Results Withdrawals Comments Hellgren, 1983 UK type: Wound Leg ulcers. Method of randomisation: Not stated. outcomes: Objective onto a drawn area Wound and weighed transparency measure- an indirect to give size.ment of wound At a photo- each evaluation was taken.graph Cleansing was also determined from covered area the % wound tissue.with granulating Setting and length of treatment: A randomised double-blind trial. Assessments were made at 7, 14 and 21 days. Both the patient and blind to the assessor were given.Treatment treatment terminated before was only if the ulcer was 21 days clean,totally or if serious occurred. side-effects
70 13 TABLE Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Health Technology Assessment panel membership This report was identified as a priority by the Pharmaceutical Panel. Acute Sector Panel Current members Chair: Dr Katherine Darton, M.I.N.D. Ms Grace Gibbs, Dr Duncan Keeley, Professor Francis H Creed, Mr John Dunning, West Middlesex University General Practitioner, Thame University of Manchester Papworth Hospital, Cambridge Hospital NHS Trust Dr Rajan Madhok, Mr Jonathan Earnshaw, Dr Neville Goodman, East Riding Health Authority Professor Clifford Bailey, Gloucester Royal Hospital Southmead Hospital Dr John Pounsford, University of Leeds Services Trust, Bristol Mr Leonard Fenwick, Frenchay Hospital, Bristol Ms Tracy Bury, Chartered Freeman Group of Hospitals, Professor Mark P Haggard, Dr Mark Sculpher, Society of Physiotherapy Newcastle-upon-Tyne MRC University of York Professor Collette Clifford, Professor David Field, Professor Robert Hawkins, Dr Iqbal Sram, NHS Executive, University of Birmingham Leicester Royal Infirmary University of Manchester North West Region Past members Professor John Farndon, Professor Cam Donaldson, Mrs Wilma MacPherson, Professor Michael Sheppard, University of Bristol* University of Aberdeen St Thomas’s & Guy’s Hospitals, Queen Elizabeth Hospital, Professor Richard Ellis, London Birmingham Professor Senga Bond, St James’s University Hospital, Dr Chris McCall, University of Newcastle- Professor Gordon Stirrat, Leeds General Practitioner, Dorset upon-Tyne St Michael’s Hospital, Mr Ian Hammond, Professor Alan McGregor, Bristol Bedford & Shires Health Professor Ian Cameron, St Thomas’s Hospital, London & Care NHS Trust Dr William Tarnow-Mordi, Southeast Thames Regional Professor Jon Nicholl, University of Dundee Health Authority Professor Adrian Harris, University of Sheffield Churchill Hospital, Oxford Professor Kenneth Taylor, Ms Lynne Clemence, Dr Gwyneth Lewis, Professor John Norman, Hammersmith Hospital, Mid-Kent Health Care Trust Department of Health University of Southampton London Diagnostics and Imaging Panel Current members Chair: Dr Barry Cookson, Mrs Maggie Fitchett, Professor Chris Price, Professor Mike Smith, Public Health Laboratory Association of Cytogeneticists, London Hospital Medical University of Leeds Service, Colindale Oxford School Dr Peter Howlett, Professor David C Cumberland, Dr William Rosenberg, Dr Philip J Ayres, Portsmouth Hospitals NHS Trust University of Sheffield University of Southampton Leeds Teaching Hospitals Professor Alistair McGuire, NHS Trust City University, London Dr Gillian Vivian, Professor Adrian Dixon, Royal Cornwall Hospitals Trust University of Cambridge Dr Andrew Moore, Dr Paul Collinson, Editor, Bandolier Dr Greg Warner, Mayday University Hospital, Mr Steve Ebdon-Jackson, Dr Peter Moore, General Practitioner, Thornton Heath Department of Health Science Writer, Ashtead Hampshire Past members Professor Michael Maisey, Professor MA Ferguson-Smith, Professor Donald Jeffries, Professor John Stuart, Guy’s & St Thomas’s Hospitals, University of Cambridge St Bartholomew’s Hospital, University of Birmingham London* London Professor Andrew Adam, Dr Mansel Hacney, Dr Ala Szczepura, Guy’s, King’s & St Thomas’s University of Manchester Dr Ian Reynolds, University of Warwick School of Medicine & Dentistry, Nottingham Health Authority London Professor Sean Hilton, Mr Stephen Thornton, St George’s Hospital Professor Colin Roberts, Cambridge & Huntingdon Dr Pat Cooke, RDRD, Medical School, London University of Wales College Health Commission Trent Regional Health Authority of Medicine Ms Julia Davison, Mr John Hutton, Dr Jo Walsworth-Bell, St Bartholomew’s Hospital, MEDTAP International Inc., Miss Annette Sergeant, South Staffordshire London London Chase Farm Hospital, Enfield Health Authority
* Previous Chair 75 continued Health Technology Assessment panel membership
continued Methodology Panel Current members
Chair: Professor Ann Bowling, Professor Jeremy Grimshaw, Dr Nick Payne, Professor Martin Buxton, University College London University of Aberdeen University of Sheffield Brunel University Medical School Dr Stephen Harrison, Professor Margaret Pearson, Dr Mike Clarke, University of Leeds NHS Executive North West University of Oxford Professor Doug Altman, Mr John Henderson, Professor David Sackett, Institute of Health Sciences, Department of Health Centre for Evidence Based Oxford Professor Michael Drummond, University of York Professor Richard Lilford, Medicine, Oxford Regional Director, R&D, Dr David Armstrong, West Midlands Dr PAG Sandercock, Dr Vikki Entwistle, University of Edinburgh Guy’s, King’s & St Thomas’s University of Aberdeen School of Medicine Professor Theresa Marteau, Dr David Spiegelhalter, & Dentistry, London Guy’s, King’s & St Thomas’s Professor Ewan Ferlie, School of Medicine & Institute of Public Health, Imperial College, London Dentistry, London Cambridge Professor Nick Black, London School of Hygiene Professor Ray Fitzpatrick, Dr Henry McQuay, Professor Joy Townsend, & Tropical Medicine University of Oxford University of Oxford University of Hertfordshire Past members
Professor Anthony Culyer, Professor George Davey-Smith, Mr Nick Mays, Professor Charles Warlow, University of York * University of Bristol King’s Fund, London Western General Hospital, Edinburgh Professor Ian Russell, Professor Michael Baum, Professor Stephen Frankel, University of York Royal Marsden Hospital University of Bristol Dr Maurice Slevin, Dr Rory Collins, Mr Philip Hewitson, St Bartholomew’s Hospital, University of Oxford Leeds FHSA London
Pharmaceutical Panel Current members
Chair: Professor Rod Griffiths, Dr Andrew Mortimore, Dr Frances Rotblat, Professor Tom Walley, NHS Executive Southampton & SW Hants Medicines Control Agency University of Liverpool West Midlands Health Authority Mr Nigel Offen, Essex Rivers Dr Eamonn Sheridan, Dr Felicity Gabbay, Mrs Jeanette Howe, Healthcare, Colchester St James’s University Hospital, Transcrip Ltd Department of Health Leeds Mrs Marianne Rigge, Mr Peter Golightly, Professor Trevor Jones, The College of Health, Mrs Katrina Simister, Leicester Royal Infirmary ABPI, London London Liverpool Health Authority Dr Alastair Gray, Mr Simon Robbins, Health Economics Research Ms Sally Knight, Camden & Islington Dr Ross Taylor, Unit, University of Oxford Lister Hospital, Stevenage Health Authority, London University of Aberdeen
Past members
Professor Michael Rawlins, Ms Christine Clark, Dr Tim Elliott, Dr John Posnett, University of Newcastle- Hope Hospital, Salford Department of Health University of York upon-Tyne* Mrs Julie Dent, Dr Desmond Fitzgerald, Dr Tim van Zwanenberg, Dr Colin Bradley, Ealing, Hammersmith & Mere, Bucklow Hill, Cheshire Northern Regional University of Birmingham Hounslow Health Authority, Health Authority London Professor Keith Gull, Professor Alasdair University of Manchester Breckenridge, RDRD, Mr Barrie Dowdeswell, Dr Kent Woods, Northwest Regional Royal Victoria Infirmary, Dr Keith Jones, RDRD, Trent RO, Health Authority Newcastle-upon-Tyne Medicines Control Agency Sheffield
76 * Previous Chair Health Technology Assessment 1999; Vol. 3: No. 17 (Pt 1)
Population Screening Panel Current members
Chair: Professor Howard Cuckle, Professor Dian Donnai, Professor Alexander Markham, Professor Sir John University of Leeds St Mary’s Hospital, Manchester St James’s University Hospital, Grimley Evans, Leeds Radcliffe Infirmary, Oxford Dr Tom Fahey, Dr Carol Dezateux, University of Bristol Dr Ann McPherson, Institute of Child Health, General Practitioner, Oxford Ms Stella Burnside, London Mrs Gillian Fletcher, Altnagelvin Hospitals Trust, National Childbirth Trust Dr Susan Moss, Londonderry Institute of Cancer Research Dr Anne Dixon Brown, Dr JA Muir Gray, Mr John Cairns, NHS Executive, Institute of Health Sciences, Dr Sarah Stewart-Brown, University of Aberdeen Anglia & Oxford Oxford University of Oxford
Past members
Dr Sheila Adam, Dr Anne Ludbrook, Professor Catherine Peckham, Professor Nick Wald, Department of Health* University of Aberdeen Institute of Child Health, University of London London Professor George Freeman, Professor Ciaran Woodman, Charing Cross & Westminster Dr Connie Smith, Professor Theresa Marteau, Centre for Cancer Medical School, London Parkside NHS Trust, Guy’s, King’s & St Thomas’s Epidemiology, Manchester London Dr Mike Gill, Brent & Harrow School of Medicine Health Authority & Dentistry, London Ms Polly Toynbee, Journalist
Primary and Community Care Panel Current members
Chair: Ms Judith Brodie, Mr Andrew Farmer, Dr Chris McCall, Dr John Tripp, Age Concern, London Institute of Health Sciences, General Practitioner, Dorset Royal Devon & Exeter Oxford Healthcare NHS Trust Mr Shaun Brogan, Professor Richard Hobbs, Dr Robert Peveler, Daventry & South Northants University of Birmingham University of Southampton Primary Care Alliance Mr Kevin Barton, Professor Allen Hutchinson, Professor Jennie Popay, East London & City Mr Joe Corkill, University of Sheffield University of Salford Health Authority National Association for Dr Phillip Leech, Patient Participation Department of Health Ms Hilary Scott, Professor John Bond, Tower Hamlets Healthcare University of Newcastle- Dr Nicky Cullum, Dr Aidan Macfarlane, NHS Trust, London upon-Tyne University of York Oxfordshire Health Authority Professor David Mant, Dr Ken Stein, Dr John Brazier, Professor Pam Enderby, Institute of Health Sciences, North & East Devon University of Sheffield University of Sheffield Oxford Health Authority
Past members
Professor Angela Coulter, Professor Andrew Haines, Mr Lionel Joyce, Professor Dianne Newham, King’s Fund, London* RDRD, North Thames Chief Executive, Newcastle City King’s College London Regional Health Authority Health NHS Trust Professor Martin Roland, Professor Gillian Parker, University of Manchester* Dr Nicholas Hicks, Professor Martin Knapp, University of Leicester Oxfordshire Health Authority London School of Economics Dr Simon Allison, & Political Science Dr Mary Renfrew, University of Nottingham Mr Edward Jones, University of Oxford Rochdale FHSA Professor Karen Luker, Professor Shah Ebrahim, University of Liverpool Royal Free Hospital, London Professor Roger Jones, Guy’s, King’s & St Thomas’s Dr Fiona Moss, Ms Cathy Gritzner, School of Medicine Thames Postgraduate Medical King’s Fund, London & Dentistry, London & Dental Education
* Previous Chair 77 continued Health Technology Assessment panel membership
National Coordinating Centre for Health Technology Assessment, Advisory Group
Current members Chair: Ms Lynn Kerridge, Professor James Raftery, Professor Andrew Professor John Gabbay, Wessex Institute for Health Health Economics Unit, Stevens, Wessex Institute Research & Development University of Birmingham Department of Public for Health Research Health & Epidemiology, & Development Dr Ruairidh Milne, Professor Ian Russell, University of Birmingham Department of Health Sciences Wessex Institute for Health & Clinical Evaluation, Research & Development University of York Professor Mike Drummond, Ms Kay Pattison, Dr Ken Stein, Centre for Health Economics, Research & Development North & East Devon University of York Directorate, NHS Executive Health Authority
Past member Dr Paul Roderick, Wessex Institute for Health Research & Development
78 HTA Commissioning Board
Current members Chair: Professor Martin Eccles, Professor Alison Kitson, Dr Sarah Stewart-Brown, Professor Charles Florey, Professor of Director, Royal College of Director, Institute of Department of Epidemiology & Clinical Effectiveness, Nursing Institute Health Sciences, Public Health, Ninewells University of Newcastle- Dr Donna Lamping, University of Oxford Hospital & Medical School, upon-Tyne Senior Lecturer, Department of University of Dundee Professor Ala Szczepura, Dr Mike Gill, Public Health, London School of Hygiene & Tropical Medicine Director, Centre for Director of Public Health & Health Services Studies, Professor Doug Altman, Health Policy, Brent & Harrow Professor Alan Maynard, University of Warwick Health Authority Director of ICRF/NHS Centre Professor of Economics, for Statistics in Medicine, Dr Alastair Gray, University of York Dr Gillian Vivian, Oxford Director, Health Economics Professor Jon Nicholl, Consultant, Royal Cornwall Research Centre, Director, Medical Care Hospitals Trust Professor John Bond, University of Oxford Research Unit, Professor of Health Services University of Sheffield Professor Graham Watt, Research, University of Professor Mark Haggard, Professor Gillian Parker, Department of General Practice, Newcastle-upon-Tyne MRC Institute of Nuffield Professor of Woodside Health Centre, Hearing Research Mr Peter Bower, Community Care, Glasgow Dr Jenny Hewison, University of Leicester Independent Health Advisor, Professor Kent Woods, Newcastle-upon-Tyne Senior Lecturer, Dr Tim Peters, Department of Psychology, Regional Director of R&D Reader in Medical Statistics, NHS Executive, Trent Ms Christine Clark, University of Leeds Department of Social Medicine, Honorary Research Pharmacist, University of Bristol Hope Hospital, Salford Professor Sir Miles Irving Dr Jeremy Wyatt, (Programme Director), Professor Martin Severs, Senior Fellow, Health & Professor Shah Ebrahim, Professor of Surgery, Professor in Elderly Public Policy, School of Public Professor of Epidemiology University of Manchester, Health Care, Policy, University College, of Ageing, University of Bristol Hope Hospital, Salford Portsmouth University London
Past members Professor Ian Russell, Dr Michael Horlington, Professor Sally McIntyre, Dr Mark Williams, Department of Health Head of Corporate Licensing, MRC Medical Sociology Unit, Public Health Physician, Sciences & Clinical Evaluation, Smith & Nephew Group Glasgow Bristol University of York* Research Centre Professor David Sackett, Professor Martin Knapp, Centre for Evidence Based Director, Personal Social Medicine, Oxford Professor David Cohen, Services Research Unit, Professor of Health Economics, London School of Economics Dr David Spiegelhalter, University of Glamorgan & Political Science MRC Biostatistics Unit, Institute of Public Health, Professor Theresa Marteau, Cambridge Director, Psychology & Genetics Mr Barrie Dowdeswell, Research Group, Guy’s, King’s Professor David Williams, Chief Executive, & St Thomas’s School of Department of Royal Victoria Infirmary, Medicine & Dentistry, Clinical Engineering, Newcastle-upon-Tyne London University of Liverpool * Previous Chair Health TechnologyAssessment 99 o.3 o 7(t1 Debridement of chronic wounds 17 (Pt 1) No. 3: Vol. 1999;
Copies of this report can be obtained from:
The National Coordinating Centre for Health Technology Assessment, Mailpoint 728, Boldrewood, University of Southampton, Southampton, SO16 7PX, UK. Fax: +44 (0) 1703 595 639 Email: [email protected] http://www.hta.nhsweb.nhs.uk ISSN 1366-5278