You are on page 1of 9

CLINICAL ISSUES

Manuka honey vs. hydrogel a prospective, open label, multicentre, randomised controlled trial to compare desloughing efcacy and healing outcomes in venous ulcers
Georgina Gethin and Seamus Cowman

Objective. Comparison of desloughing efcacy after four weeks and healing outcomes after 12 weeks in sloughy venous leg ulcers treated with Manuka honey (Woundcare 18+) vs. standard hydrogel therapy (IntraSite Gel). Background. Expert opinion suggests that Manuka honey is effective as a desloughing agent but clinical evidence in the form of a randomised controlled trial is not available. There is a paucity of research which uses Manuka honey in venous ulcers. Design. Prospective, multicentre, open label randomised controlled trial. Method. Randomisation was via remote telephone. One hundred and eight patients with venous leg ulcers having 50% wound area covered in slough, not taking antibiotics or immunosuppressant therapy were recruited from vascular centres, acute and community care hospitals and leg ulcer clinics. The efcacy of WoundCare 18+ to deslough the wounds after four weeks and its impact on healing after 12 weeks when compared with IntraSite Gel control was determined. Treatment was applied weekly for four weeks and follow-up was made at week 12. Results. At week 4, mean % reduction in slough was 67% WoundCare 18+ vs. 529% IntraSite Gel (p = 0054). Mean wound area covered in slough reduced to 29% and 43%, respectively (p = 0065). Median reduction in wound size was 34% vs. 13% (p = 0001). At 12 weeks, 44% vs. 33% healed (p = 0037). Wounds having >50% reduction in slough had greater probability of healing at week 12 (95% condence interval 112, 97; risk ratio 33; p = 0029). Infection developed in 6 of the WoundCare 18+ group vs. 12 in the IntraSite Gel group. Conclusion. The WoundCare 18+ group had increased incidence of healing, effective desloughing and a lower incidence of infection than the control. Manuka honey has therapeutic value and further research is required to examine its use in other wound aetiologies. Relevance to clinical practice. This study conrms that Manuka honey may be considered by clinicians for use in sloughy venous ulcers. Additionally, effective desloughing signicantly improves healing outcomes. Key words: nursing, randomised control trials, research, tissue viability, venous leg ulcer, wound care
Accepted for publication: 11 June 2008

Introduction
The premise that wound healing is impaired in the presence of devitalised tissue is well supported within the literature (Sieggreen & Malkebust 1997, Baharestani 1999, NICE
Authors: Georgina Gethin, PhD, RGN, Dip HE wound care, Dip Anatomy, Dip Applied Physiology, Faculty of Nursing and Midwifery, Royal College of Surgeons in Ireland, Dublin, Ireland; Seamus Cowman, MSc, PhD, RNT, RGN, RPN, FFNMRCSI, PG Cert (Adults), Dip N (London), Faculty of Nursing and Midwifery, Royal College of Surgeons in Ireland, Dublin, Ireland

2001). Its removal, when consistent with treatment goals, is considered the rst step in wound bed preparation of the chronic wound (Falanga 2002). Debridement methods range from surgical to natural autolytic debridement and the method is dependent on the condition of the wound bed,
Correspondence: Georgina Gethin, Lecturer, Research Centre, Faculty of Nursing and Midwifery, Royal College of Surgeons in Ireland, 123 St Stephens Green, Dublin 2, Ireland. Telephone: 00 353 1 4022202. E-mail: ggethin@rcsi.ie

466

2008 The Authors. Journal compilation 2008 Blackwell Publishing Ltd, Journal of Clinical Nursing, 18, 466474 doi: 10.1111/j.1365-2702.2008.02558.x

Clinical issues

An open label, multicentre, prospective study

resources, knowledge and patient and clinician treatment goals. Venous leg ulcers (VLU) continue to affect up to 2% of the population (Nelzen et al. 1996, OBrien et al. 2000, Moffatt et al. 2004). While compression remains the primary treatment option for uncomplicated VLU the plateau in healing rates of 3250% after 12 weeks of compression and recurrence of 50% within three months of healing (Monk & Sarkany 1982, Thomson et al. 1996, Moffatt et al. 2004) warrants investigation into which ulcers are healing and the characteristics of those that do not. Systematic reviews of dressings used in the management of VLU have failed to demonstrate improved healing outcomes above that expected with compression alone (Bradley et al. 1999, Palfreyman et al. 2006). Arguably, failure to stratify healing outcomes based on the condition of the wound bed at baseline may account for this. Essentially one would not expect a sloughy, large, heavily colonised VLU to have the same rate of healing as a clean granulating ulcer. While the use of honey in wound management dates back 2000 years (Forrest 1982, Molan 2006), evidence of efcacy in modern literature is predominantly in the acute wound domain. A specic type of honey, namely Manuka honey (MH) from the Leptospermum scoparium bush found throughout New Zealand and Australia, is gaining recognition as a wound management agent (Cooper 2001, Gethin & Cowman 2005, Molan 2006). However, claims of MH as a desloughing agent are based on in vitro research and case studies and have not been tested heretofore in a rigorous clinical study. The desloughing capacity of MH may permit it to work synergistically with compression therapy to treat VLU and improve outcomes.

2000) determined that 156 subjects randomly allocated to two equal groups of 78 were required. The power of the study was based on an 80% power to detect a minimum difference of 20% at the 5% two-sided significance level.

Selection and description of participants


Adults presenting with VLU having 50% wound area covered in slough and able to provide written informed consent were considered for inclusion. Inclusion and exclusion criteria are listed in Table 1 and were based on Ankle Brachial Pressure Index (ABPI) of 08 where no other causative aetiology was present and the ulcer appeared to be venous on clinical examination. Baseline data collection included patient gender, age, wound size and duration, ulcer location and history of recurrence. Medical information included history of deep venous thrombosis (DVT), hypertension, trauma or surgery to the affected limb, diabetes, immunosuppression and current medications. The uppermost ulcer or the largest ulcer was selected if more than one was present, providing this ulcer met the inclusion criteria. Each participant could only contribute one ulcer to the study.

Randomisation
Following screening and when consent was provided, patients were randomised via remote phone allocation to either treatment group. The allocation sequence was generated

Table 1 Inclusion and exclusion criteria

Study outcomes
The primary outcomes were determination of the ability of MH to deslough VLU over four weeks and measurement of its effects on healing rates after 12 weeks when compared with a standard agent.

Methods
Study design
This prospective, open label, 12-week, multicentre, randomised controlled clinical trial (RCT) was conducted between February 2003January 2006. Ethical approval from each area was obtained. The formula for comparison of proportions in two independent groups (Daly & Bourke

Inclusion criteria Having a venous ulcer Being over 18 years of age Able to provide written informed consent Having 50% of wound bed covered in slough Ulcer size < 100 cm2 Exclusion criteria Less than 18 years Unable to provide written informed consent Having an ulcer > 100 cm2 Ulcer diagnosed as being malignant Having a cavity wound Clinical diagnosis of wound infection Currently taking antibiotics for any reason Currently taking oral immunosuppressant Having poorly controlled diabetes Having previously enrolled into the study Pregnant women or lactating mothers

2008 The Authors. Journal compilation 2008 Blackwell Publishing Ltd, Journal of Clinical Nursing, 18, 466474

467

G Gethin and S Cowman

using serially numbered, sealed, opaque envelopes, prior to the study by two persons independent of the study.

Statistical analysis
All statistical analyses were performed using SPSS version 14 and Stata release 92 by a statistician blinded to the treatment allocations on the intention-to-treat population. All patients were included in the nal analysis. All tests were conducted at the two-sided a = 005 level of signicance. Regression was used to model the effects of treatment on wound size. Ordered logistic regression was used to model the effects of treatment on slough as this variable departed markedly from the normal distribution and exhibited clustering of values. Binomial regression was used to model binary outcomes which allowed the treatment effect to be expressed as a risk ratio (RR). Slough was calculated as the percentage of the wound bed that it covered at the time of assessment. The difference from baseline to end of week 4 assessment was calculated as: [original percentage of wound area covered in slough nal percentage of wound area covered in slough]. Wound area was re-expressed as wound diameter and changes were calculated as: [original wound diameter nal wound diameter]. The chi-square test was used to compare categorical demographic and baseline variable response proles. Independent t-test analysed baseline demographic and wound variables. Wilcoxon-signed rank test analysed median differences in wound size. In addition to treatment effects, subset analyses were performed comparing the Margolis scores (Margolis et al. 2000) of each group at baseline and determination of outcomes at week 12 of all wounds based on this score.

Treatments
All wounds were cleansed with warm tap water at each visit prior to assessment and treatment. MH dressing (Woundcare 18+; Comvita, Te Puke, New Zealand), a viscous orange/ brown topical agent, at a dosage of 5 g/20 cm2 was applied weekly. The control group had hydrogel therapy (HT; IntraSite Gel; Smith & Nephew, Hull, UK) at a dosage of 3 g/20 cm2 applied weekly. The viscosity of the agents accounted for the dose differential. HT was selected as this represented the routine practice for sloughy wounds. Secondary dressings were standardised as Allevyn hydrocellular foam (Smith & Nephew). Compression therapy was continued in all cases, the most common form being four-layer bandages. The treatment period was four weeks. After this time, all patients received follow-on treatment based on clinical assessment of the wound by the local investigator, which varied from patient to patient. Follow-up was conducted at week 12.

Wound assessment
All wound assessments were completed at baseline and at weekly intervals for four weeks by the local clinical investigator who had received training in leg ulcer management. A determination of healed or not healed was made at week 12. Weekly wound measurements were conducted using Visitrak digital planimetry (Smith & Nephew); the percentage of the wound bed covered in slough was assessed quantitatively. A pretrial inter-rater reliability (IRR) study determined the level of agreement among three raters when assessing slough as a percentage within the wound bed. The inter-class correlation for slough was 095 (95% condence interval, CI 088098, p < 0001). The presence or absence of other tissues, including granulation, epithelial and necrotic, was recorded at each assessment. Blinded outcome assessment was not possible because of obvious differences in the colour and presentation of the products, specically orange staining of the peri-wound skin when MH was used.

Results
Demographic and baseline values
The ow of participants through the trial is presented in Fig. 1. One hundred and eight patients were enrolled from 10 sites, including acute and community hospitals, vascular and leg ulcer clinics over 34 months. There were no statistical baseline differences between groups (Table 2). Complete screening data was only available from one large centre. The reasons for non-inclusion from this centre are presented in Table 3.

Specic withdrawal criteria


Patients who commenced antibiotic or immunosuppressant therapy for any reason during the treatment period were withdrawn. The rationale being, that any improvement in the wound in the four weeks could not be condentially attributed to the topical agent and may have been because of antibiotics or steroid treatment.
468

Reduction in slough
After four weeks, 80% (n = 86) of all wounds had a reduction of >50% slough. There was no statistically signicant difference at week 4 between treatments (95% CI 0370, 100; p = 0367; Table 4). It is noteworthy that a slough reduction of at least 50% by week 4 was associated

2008 The Authors. Journal compilation 2008 Blackwell Publishing Ltd, Journal of Clinical Nursing, 18, 466474

Clinical issues

An open label, multicentre, prospective study

Assessed for eligibility n = 256 *

Excluded n = 148 Reason for exclusion: Did not meet criteria n = 142 Refused to participate n = 6 See Table 2 Randomised n = 108

Manuka-honey group (a) n = 54

Hydrogel group (b) n = 54

Received allocated treatment n = 54 Did not receive n = 0

Received allocated treatment n = 54 Did not receive n = 0

Lost to follow up n = 0 Withdrawal n=9 Reason for withdrawal: Infection in study wound n=6 Infection elsewhere n=1 Patient request n=1 Non-compliance with treatment n = 1

Lost to follow up n = 0 Withdrawal n = 17 Reason for withdrawal: Infection in study wound n = 12 Infection elsewhere n=1 Patient request n=4 Non-compliance with treatment n = 0

Analysed

n = 54

Analysed

n = 54

Figure 1 Flow of participants through the trial.

*complete screening data from one centre only. Numbers randomised are from all centres

with a higher probability of healing at 12 weeks across all groups (95% CI 11297; RR 33, p = 0029).

Healing outcomes
Epithelisation was visible at an earlier stage in the MH group and this difference was statistically signicant (v2 = 9906; p = 0042). Healing outcomes are presented in Table 4. The median wound size at baseline for all wounds was 47 cm2 and this reduced by 23% to 365 cm2 at week 4. The MH group had a 34% reduction in median size vs. 13% in the HT group; this difference was statistically signicant (z = 4609; p < 0001). The mean reduction in wound size in all wounds after four weeks was 659% (SD 233) in those that healed at 12 weeks vs. 19% (SD 232; p < 0001) in those that did not heal.

The Margolis index (Margolis et al. 2000), while intended as a potential predictor to healing at 24 weeks, was included here for outcomes at 12 weeks to aid interpretation of the results. Patients scored 1 for wound duration 6 months and scored 1 if the size was 5 cm2. Thus, a higher score indicates a poorer prognosis for healing at 24 weeks. Sixty nine per cent of those with a score of zero at baseline healed vs. 21% of those with a score of 1 vs. 95% of those with a score of 2. When adjusted for Margolis score, the healing rate at 12 weeks was signicantly higher in the MH group (OR 31, 95% CI 115835, p = 0025).

Adverse events
There were no adverse events directly attributable to either wound agent. Exactly 24% (n = 26) withdrew (n = 9 MH,
469

2008 The Authors. Journal compilation 2008 Blackwell Publishing Ltd, Journal of Clinical Nursing, 18, 466474

G Gethin and S Cowman Table 2 Baseline characteristics of participants by treatment group Honey treatment (n = 54) 6850 (1350) 3946 (4050) 1052 (1230)* 54 855 (217) 14 10 5 Hydrogel therapy (n = 54) 6830 2993 987 42 7815 19 8 4 (1510) (3520) (1290)* (233)

Characteristics Age (years)* Wound duration (weeks)* Wound size (cm)* % wound bed covered in slough Hypertension Current smoker History of deep vein thrombosis in affected leg Ulcer recurrence Margolis score = 0 Margolis score = 1 Margolis Score = 2

27 18 16 20

31 25 17 12

Values are mean* (SD) or median where specied.

Table 3 Reason for non-randomisation based on screening data from one centre Reason for non-randomisation Clean wound bed Ankle Brachial Pressure Index < 08 On antibiotics Unable to attend for duration of study Ulcer > 10 10 cm2 Patient request Unable to consent Infection in study wound On steroid therapy Not a venous ulcer Other Total Number (%) 15 47 3 12 6 6 3 8 4 38 6 148 (10) (31) (2) (8) (4) (4) (2) (5) (2) (25) (4)

n = 17 HT), reasons for which are presented in Table 4. The low numbers of withdrawals in each group precluded meaningful statistical analysis. However, analysis of time to withdrawal because of infection using survival analysis shows that there is a difference but it is just short of statistical signicance using the log-rank test (p < 007).

Discussion
There is little knowledge of healing outcomes specic to the condition of the wound bed in sloughy VLU. This is important if we are to gain deeper understanding of which ulcers heal at 12 weeks and advance the knowledge related to the appropriate use of topical agents in wound management.
470

This RCT represents the rst and largest multicentre, clinical trial designed to assess the effectiveness of MH for the treatment of sloughy VLU. A pragmatic study design rather than an efcacy trial was conducted so that ndings may be applicable to the general patient population and emulate routine practice. For the primary outcome, a statistically signicant differential efcacy in desloughing between the two treatment groups could not be shown. However, they are clinically signicant in that a greater percentage of slough was removed by MH than by the HT. Placing the primary outcome results in the perspective with other studies is challenging because of study differences in inclusion criteria and duration of treatment. One can conclude that in the desloughing of VLU, MH is slower than larval therapy or curettage (Wayman et al. 2000, Williams et al. 2005), but superior to some hydrogels, enzymatic agents, hydrocolloids, parafn gauze or cadexomer iodine (Hansson 1998, Wayman et al. 2000, Konig et al. 2005). These ndings advance the knowledge gained from other research on desloughing agents, as other studies failed to follow patients for 12 weeks (Flanagan 1995, Fear & Thomas 1996, Wayman et al. 2000, Konig et al. 2005). The second observation period of 12 weeks demonstrated the benet to healing outcomes when wounds were desloughed. After 12 weeks, all wounds with a reduction in slough of 50% had higher healing rates than those where 50% slough was removed (RR 33; p = 0029) supporting the premise that removal of devitalised tissue promotes wound healing. In comparison, one study which followed patients for 20 weeks did not demonstrate superiority of healing rates at 12 weeks compared with those in this RCT (Williams et al. 2005). Williams et al. (2005) reported a healing rate of 146% (n = 5) at 12 weeks when wounds were debrided using curettage. A possible explanation for the poorer healing rates may lie in the fact that the latter debrided wounds as a one-time event, whereas debridement was continued over four weeks in the present study. The benet of on-going or maintenance debridement on healing rates has been demonstrated previously in the management of diabetic foot ulcers (Steed et al. 1996) but not in VLU. It is therefore proposed that a four-week observation period can guide clinicians on whether to continue with, or change a treatment based on the reduction in wound size. But, to determine efcacy of an agent on healing, outcomes should be evaluated over at least 12 weeks. MH may deslough wounds through a combination of pathways. First, the honey exerts an osmotic effect within the wound bed as it draws uid from the deeper wound tissue to

2008 The Authors. Journal compilation 2008 Blackwell Publishing Ltd, Journal of Clinical Nursing, 18, 466474

Clinical issues Table 4 Study outcomes at weeks 4 and 12 based on treatment group Honey therapy, week 4
2

An open label, multicentre, prospective study

Characteristic Wound size (cm )* Wound size (cm2) Healed, n (%) % wound bed covered in Slough % reduction in slough Withdrawal Infection in reference wound Infection elsewhere Patient request Non-compliance Did not attend follow-up

Honey therapy, week 12

Hydrogel therapy, week 4 824 (1211) 366

Hydrogel therapy, week 12

p-value (95% CI) 016 (01, 055) 0001 003 (102, 188) RR 138 006 (292, 089) 005 (028, 3108) 007

825 (1157) 356 24 (44) 29 (35) 67 (3641) 9 6 1 0 1 1 (166) (111) (185) (185) (185)

18(33) 43 (44) 526 (451) 17 (3148) 12 (222) 1 (185) 3 (55) 0 1 (185)

Values are mean* (SD) or median where specied. CI, condence interval; RR, risk ratio.

the surface (Chirife et al. 1982). This process aids cleansing and removal of devitalised tissue (Chirife et al. 1982). Second, honey maintains a moist wound environment (Condon 1993, Cooper 2001, Molan 2001), which facilitates autolysis (Sieggreen & Malkebust 1997, Baharestani 1999, Ayello & Cuddigan 2004). In addition to autolysis, MH may also deslough by lowering wound pH (Gethin et al. 2008). Research has shown a statistically signicant reduction (p < 0001) in pH of chronic wounds after two weeks of MH use (Gethin et al. 2008). In the latter study, when slough remained in the wound bed, the pH was elevated at 77 and the wound size increased by 6%. Conversely, when slough was removed, the pH was 76. A clinically and statistically signicant nding from this study was that a reduction in 01 pH units correlated with an 81% reduction in wound size (p = 0012; Gethin et al. 2008). In determining the efcacy of MH to deslough VLU, it was important from a clinical perspective to ascertain if any benet was observed on wound size and healing rates. The four-week treatment period was guided by clinical experience and research indicates that a 30% reduction in wound size is a good predictor of healing (Tallman et al. 1997, Kantor & Margolis 2000, Margolis et al. 2000, Meyer et al. 2002, Flanagan 2003, Steed et al. 2006). However, stopping the trial at this stage would fail to demonstrate effects on healing outcomes, which, it could be argued, are the only outcomes of interest to clinicians and patients. After four weeks, the mean 19% reduction in wound size could be interpreted as a poor outcome when compared with the recommended 30% (Flanagan 2003). However, a probable reason for this when compared with other research (Tallman et al. 1997, Kantor & Margolis 2000, Margolis

et al. 2000, Gethin & Cowman 2005), may be related to the compromised state of the wound bed at the start of the study wherein slough was the predominant tissue type. It may be that the recommended 30% reduction is most appropriately applied to clean granulating wounds and not sloughy VLU. At the four-week time point, MH outperformed the HT with a median reduction in wound size superior in the MH-treated wounds. Wound healing trajectories have demonstrated signicant differences in healing curves of healing vs. nonhealing VLU (Steed et al. 2006). This study is in agreement with Steed et al. (1996) as ulcers that healed had signicantly greater reduction in wound size at four weeks that those that did not heal. This study has used the Margolis score on a specic subset of VLU, i.e. 50% slough and reported outcomes after 12 weeks. The Margolis score represents a step towards a targeted therapeutic approach to wound management, as it considers both size and duration. Sixty-nine per cent healing occurred in wounds with a zero score. This is distinctly different from the 95% healed in those with a score of 2. Wounds treated with MH had statistically superior healing rates after 12 weeks when size and duration were considered. One could thus conclude that the expected healing rates of clean granulating VLU, <5 cm2 and <6 months duration would be much greater than the 39% healed in this study. This gure may set a benchmark against which other topical agents can be compared. Based on the observations made in this study, it is recommended that future research on the efcacy of topical agents in the treatment of VLU be stratied according to the condition of the wound bed together with the size and duration of the wound. Thus, agents that have not been able to demonstrate superiority over standard care to date may be able to report more clinically meaningful and
471

2008 The Authors. Journal compilation 2008 Blackwell Publishing Ltd, Journal of Clinical Nursing, 18, 466474

G Gethin and S Cowman

statistically signicant outcomes. The 39% healing rate after 12 weeks compares favourably with the 4050% reported in other VLU studies (Duby et al. 1993, Thomson et al. 1996, Wilkinson et al. 1997, Bale & Harding 2003) but less than one study in which 55% healing was achieved when compression and MH dressings were used (Jull et al. 2008). The current study differs from that of Jull et al. (2008) as the median wound size in their study was 27 cm2, exactly half of that in our study. Additionally, the majority of patients had a zero Margolis score in the Manuka group in the latter study while the majority had a score of 2 in this study. This underscores the differences in expected healing outcomes based on the condition of the wound at baseline. Furthermore, the mean ulcer duration was 346 weeks in the present study, arguing potentially longer healing times than predicted by other research (Kantor & Margolis 2000). Further comparison with other studies designed to evaluate the use of dressings in VLU management is limited as many studies fail to state the condition of the wound bed at the start of treatment (Blair et al. 1988, Thomas et al. 1997, Franks et al. 2007). One RCT was identied which used honey (not Manuka) in the management of chronic wounds, but this study excluded VLU (Oluwatosin et al. 2000). Direct comparison is not applicable to a further study which compared honey with IntraSite Gel (Ingle et al. 2006). The latter study excluded wounds of the legs and used a local, unsterile, unprocessed honey. Infection was the most common reason for withdrawal at 166% (n = 18). Although prospective patients with clinical signs of infection were excluded from the trial, research has shown that chronic ulcers are persistently colonised subclinically (Trengove et al. 1996, Bowler & Davies 1999). Infection rates are concerning given that in both groups, 18 patients failed to complete the four-week treatment period because of interrupted treatment owing to wound infection. Although, no statistical differences were noted in the frequency of infection between the groups, the two study treatments differed in terms of their antimicrobial properties and this may explain the differences in infection rates in sloughy wounds.

enrol 156 participants would have necessitated a further 18 months of recruitment which represented an unrealistic timeframe for this type of study. This study has several strengths most notably that it is the largest study of desloughing in VLU and the only study to have determined the efcacy of MH to deslough. Furthermore, it establishes the healing outcomes of VLU based on the condition of the wound bed. The direct comparison of honey with another agent provides more clinically meaningful interpretation of the ndings than studies in which standard care or usual care, which is not standardised, is used.

Implications for practice


There has been an exponential rise in demand for evidence to support clinical practice. While the clinician is faced with an ever-increasing choice of topical agents to manage wounds, Cochrane reviews have identied that the evidence for many of these agents is based on methodologically inferior research studies (Bradley et al. 1999, OMeara et al. 2001, Palfreyman et al. 2006). This study has contributed to evidence-based practice through the conduct of a well-designed RCT which answered a clinically relevant question. Whether clinicians use MH in their armoury of products will remain a choice based on their local protocols for purchase and selection of products. Nonetheless, this study will allow them to base their decision on an unbiased assessment of objective evidence rather than on a subjective compilation of opinion.

Conclusion
When tested in the rigours of an RCT containing the largest sample size in a study of debridement of VLU, MH-treated wounds exhibited an increased incidence of healing, more efcacious desloughing and lower rates of infection as compared with the controls. MH has a benecial therapeutic effect in modern wound management.

Contributions
Study design: GG; data collection: GG; data analysis: RC*, GG, SC and manuscript preparation: GG, SC. *RC is Prof. Ronan Conroy, DSc, Associated Professor of Biostatistics in RCSI conducted the primary data analysis.

Strengths and limitations


The prestudy pilot phase identied that when MH was used orange staining of the peri-wound skin was noted, introducing the possibility of ascertainment bias. However, this was minimised by the use of percentage tissue type, the IRR of which had been established, wound measurement and the multicentre study. A second limitation was failure to enrol the predetermined number of participants. The duration of accrual extended to 34 months. Continuation of the trial to
472

Acknowledgement
The authors are indebted to Prof. Ronan Conroy for statistical analysis and the tissue viability nurses in Sligo General Hospital.

2008 The Authors. Journal compilation 2008 Blackwell Publishing Ltd, Journal of Clinical Nursing, 18, 466474

Clinical issues

An open label, multicentre, prospective study Flanagan M (2003) Wound measurement: can it help us to monitor progression to healing? Journal of Wound Care 12, 189194. Forrest R (1982) Early history of wound treatment. Journal of Royal Society of Medicine 75, 198205. Franks P, Moody M, Moffat C, Hiskett G, Gatto P, Davies C, Furlong W, Barrow E & Thomas H (2007) Randomised trial of two foam dressings in the management of chronic venous ulceration. Wound Repair and Regeneration 15, 197202. Gethin G & Cowman S (2005) Case series of use of Manuka honey in leg ulceration. International Wound Journal 2, 1015. Gethin G, Cowman S & Conroy R (2008) Change in pH of chronic wounds when a honey dressing is used. International Wound Journal 5, 185195. Hansson C (1998) The effects of cadexomer iodine paste in the treatment of venous leg ulcers compared with hydrocolloid dressing and paraffin gauze dressing. Cadexomer Iodine Study Group. International Journal of Dermatology 37, 390396. Ingle R, Levin J & Polinder K (2006) Wound healing with honey a randomised controlled trial. South African Medical Journal 96, 831835. Jull A, Walker N, Parag V, Molan P & Rodgers A (2008) Randomized clinical trial of honey-impregnated dressings for venous leg ulcers. British Journal of Surgery 95, 175182. Kantor J & Margolis D (2000) A multicentre study of percentage change in venous leg ulcer area as a prognostic index of healing at 24 weeks. British Journal of Dermatology 142, 960964. Konig M, Vanscheidt W, Augustin M & Kapp H (2005) Enzymatic versus autolytic debridement of chronic leg ulcers: a prospective randomized trial. Journal of Wound Care 14, 320323. Margolis D, Berlin J & Strom B (2000) Which venous leg ulcers will heal with limb compression bandages? American Journal of Medicine 109, 1519. Meyer FJ, Burnand KG, Lagattolla NRF & Eastham D (2002) Randomized clinical trial comparing the efficacy of two bandaging regimes in the treatment of venous leg ulcers. British Journal of Surgery 89, 4044. Moffatt CJ, Franks PJ, Doherty DC, Martin R, Blewett R & Ross F (2004) Prevalence of leg ulceration in a London population. Quarterly Journal of Medicine 97, 431437. Molan P (2001) Potential of honey in the treatment of wounds and burns. American Journal of Clinical Dermatology 2, 1319. Molan P (2006) The evidence supporting the use of honey as a wound dressing. Lower Extremity Wounds 5, 4054. Monk BE & Sarkany I (1982) Outcome of treatment of venous stasis ulcers. Clinical and Experimental Dermatology 7, 397400. Nelzen O, Bergqvist D, Fransson I & Lindhagen A (1996) Prevalence and aetiology of leg ulcers in a defined population of industrial workers. Phlebology 11, 5054. NICE (2001) Guidance on the use of debriding agents and specialist wound care clinics for difcult to heal surgical wounds. National Institute for Clinical Excellence, Technology Appraisal Guidance No. 24. OBrien JF, Grace PA, Perry IJ & Burke PE (2000) Prevalence and aetiology of leg ulcers in Ireland. Irish Journal of Medical science 169, 110112. OMeara SM, Cullum N, Majid M & Sheldon TA (2001) Systematic review of antimicrobial agents used for chronic wounds. British Journal of Surgery 88, 421.

Funding
This study was funded by research grant-aid from the Research and Education Foundation in Sligo General Hospital, European Wound Management Association and the Health Research Board of Ireland. These funders had no role in design and conduct of the study, collection, analysis and interpretation of data or in the preparation of the manuscript, review or approval of manuscript, in protocol development, patient recruitment, data collection, data analysis or manuscript preparation. Grant-aid was obtained through an application, interview and international peer-review process.

References
Ayello E & Cuddigan J (2004) Debridment: controlling the necrotic/ cellular burden. Advances in Skin and Wound Care 17, 6675. Baharestani M (1999) The Clinical Relevance of Debridement. Springer-Verlag, Heidelberg. Bale S & Harding K (2003) Managing patients unable to tolerate therapeutic compression. British Journal of Nursing (Tissue Viability Supplement) 12, S7S10. Blair S, Blackhouse D, Wright D, Riddle E & McCollum C (1988) Do dressings influence the healing of chronic venous ulcers? Phlebology 3, 129134. Bowler P & Davies B (1999) The microbiology of infected and noninfected leg ulcers. International Journal of Dermatology 38, 573578. Bradley M, Cullum N, Nelson E, Petticrew M, Sheldon T & Torgerson D (1999) Systematic reviews of wound care management: (2) dressings and topical agents used in the healing of chronic wounds. Health Technology and Assessment 3, 3126. Chirife J, Scarmato G & Herszage L (1982) Scientific basis for use of granulated sugar in treatment of infected wounds. Lancet i, 560 561. Condon RE (1993) Curious interaction of bugs and bees. Surgery 113, 234235. Cooper RA (2001) How does honey heal wounds? In Honey and Healing (Munn P & Jones R eds). International Bee Research Association, United Kingdom. Daly LE & Bourke G (2000) Interpretation and Uses of Medical Statistics. Blackwell Science, Oxford. Duby T, Cherry G, Hoffman D, Cameron J, Dobloff-Brown D & Ryan T (1993) A randomised trial in the treatment of venous leg ulcers comparing short-stretch bandages, four layer bandage system and a long stretch-paste bandage system. Wounds A Compendium of Clinical Research and Practice 5, 276279. Falanga V (2002) The clinical relevance of wound bed preparation. In The Clinical Relevance of Wound Bed Preparation (Falanga V & Harding K eds). Springer Publication, Germany. Fear M & Thomas S (1996) IntraSite Gel compared with dextranomer paste in the management of pressure sores. In IntraSite Gel Clinical Reference Guide. Smith & Nephew, UK. Flanagan M (1995) The efficacy of a hydrogel in the treatment of wounds with non-viable tissue. Journal of Wound Care 4, 264 267.

2008 The Authors. Journal compilation 2008 Blackwell Publishing Ltd, Journal of Clinical Nursing, 18, 466474

473

G Gethin and S Cowman Oluwatosin OM, Olabanji JK, Oluwatosin OA, Tijani LA & Onyechi HU (2000) A comparison of topical honey and phenytoin in the treatment of chronic leg ulcers. African Journal of Medical Science 29, 3134. Palfreyman SJ, Nelson EA, Lochiel R & Michaels JA (2006) Dressings for healing venous leg ulcers [systematic review]. Cochrane Database of Systematic Reviews 3. Sieggreen M & Malkebust J (1997) Debridement: choices and challenges. Advances in Wound Care 10, 3237. Steed DL, Donohoe D, Webster MW & Lindsley L (1996) Effect of extensive debridement and treatment on the healing of diabetic foot ulcers. Diabetic Ulcer Study Group. Journal of The American College of Surgeons 183, 61. Steed DL, Hill DP, Woodske ME, Payne WG & Robson MC (2006) Wound-healing trajectories as outcome measures of venous stasis ulcer treatment. International Wound Journal 3, 4047. Tallman P, Muscare E, Carson P, Eaglstein H & Falanga V (1997) Initial rate of healing predicts complete healing of venous ulcers. Archives of Dermatology 133, 12311234. Thomas S, Banks V, Bale S, Fear-Price, Hagelstein S, Harding K, Orpin J & Thomas N (1997) A comparison of two dressings in the management of chronic wounds. Journal of Wound Care 6, 383 386. Thomson B, Hooper P, Powell R & Warin AP (1996) Four-layer bandaging and healing rates of venous leg ulcers. Journal of Wound Care 5, 213216. Trengove N, Stacy M, McGechie D & Mata S (1996) Qualitative bacteriology and leg ulcer healing. Journal of Wound Care 5, 277 280. Wayman J, Nirojogi V, Walker A, Sowinski A & Walker M (2000) The cost effectiveness of larval therapy in venous ulcers. Journal of Tissue Viability 10, 9194. Wilkinson E, Buttfield S, Cooper S & Young E (1997) Trial of two bandaging systems for chronic venous leg ulcers. Journal of Wound Care 6, 339340. Williams D, Enoch S, Miller D, Harris K, Price P & Harding K (2005) Effect of sharp debridement using curette on recalcitrant non-healing venous leg ulcers: a concurrently controlled, prospective cohort study. Wound Repair and Regeneration 13, 131 137.

474

2008 The Authors. Journal compilation 2008 Blackwell Publishing Ltd, Journal of Clinical Nursing, 18, 466474