VenUS II: a randomised controlled trial of larval therapy in the management of leg ulcers

Non-randomised controlled trial evidence

Case series have reported good outcomes when treating chronic wounds (several different aetiologies),52 foot ulcers53 and leg ulcers54,55 with loose larvae in terms of rapid debridement and improved healing. Studies have also compared larval therapy with conventional treatments in an ad hoc way. In an unblinded study, Sherman56 reported the results of 43 pressure ulcers treated with loose larvae compared with 49 treated with conventional therapy. The study concluded that loose larvae promoted rapid debridement, although the method of assessing debridement was not reported. Each of these case studies has serious methodological flaws and is at risk of selection and assessment bias. As a result, their conclusions, although providing a foundation for further work, cannot be used as evidence that larval therapy speeds debridement or healing.

Randomised controlled trial evidence

A search of MEDLINE and the Cochrane Wounds Group Specialised Trials Register (28 February 2008) identified one published randomised controlled trial (RCT) of loose larval therapy involving only 12 patients, all with venous leg ulcers. 57 This small trial found that venous leg ulcers treated with larval therapy debrided more quickly than those treated with hydrogel. The study did not follow participants until complete healing; therefore the extent to which larval therapy speeds venous ulcer healing remained unknown. The trial also reported that larval therapy was a cost-effective treatment; however, this analysis was limited.

A search of the Register of Current Controlled Trials and the National Research Register (throughout the trial and finally in May 2008) identified two further completed trials and one ongoing trial investigating the impact of larval therapy on debridement and/or healing. One further trial (Dompmartin; Table 2) was identified via a personal communication. No results could be obtained for the completed trial, see Table 2 for details.

Non-randomised controlled trial evidence

It has been proposed that whilst feeding, larvae also ingest and destroy bacteria and in doing so can alleviate symptoms associated with infection and also potentially reduce bacterial load. 26 Again, there has been very little recent research on this proposed antimicrobial action. Initial reports regarding the antimicrobial activity of larval secretions were in vitro findings published in the 1930s. 27 More recently Thomas et al. 39 reported that L. sericata larval secretions showed good antimicrobial activity against Streptococcus A and B and S. aureus in vitro, with some activity detected against Pseudomonas sp. and MRSA. A further in vitro study tested excretions/secretions from L. sericata larvae against MRSA and also reported antimicrobial activity. 36 A recent in vitro study also reported that whole body extractions of larvae were active against wound isolates of Gram-positive and Gram-negative bacteria including Pseudomonas aeruginosa, Klebsiella pneumoniae and MRSA. 37,38 However, there is relatively little in vivo work. Steenvoorde and Jukema81 treated 16 wounds infected with Gram-positive and/or Gram-negative bacteria with loose (n = 3) and bagged (n = 13) larvae and monitored the impact on wound flora. They reported a greater reduction in Gram-positive than Gram-negative bacteria after larval treatment.

The potential action of larvae on MRSA is of great interest because it is difficult to treat, has detrimental effects on patient health and is extremely costly to the NHS. The in vitro work above is supported by limited in vivo evidence. A prospective case series used loose larvae to treated diabetic foot ulcers that had been colonised with MRSA for more than 3 weeks. 82 In total, 13 participants were recruited and treated with larval therapy. The study reports that MRSA was eliminated in all but one ulcer. However, this evidence is severely limited by the lack of a control group so it is impossible to say whether MRSA would have disappeared without larval therapy, or there may have been some false-positive MRSA detection in the initial analysis.

Evidence from randomised controlled trials

As with the effectiveness of larval therapy on debridement and healing, although the existing research provides interesting data, more research is required to investigate the antimicrobial activity of larval therapy using RCTs. A search of MEDLINE, the Cochrane Wounds Group Specialised Trials Register and national and international research registers identified only one completed relevant RCT. This has not been published (as of 1 May 2008) and no results could be obtained for the completed trial (Table 2).

Primary objective

• To compare the effects of larval therapy and hydrogel on the time to complete healing of venous and mixed venous/arterial leg ulcers.

Secondary objectives

• To compare the cost-effectiveness of larval therapy with that of hydrogel.

• To compare the effects of larval therapy and of hydrogel on time to debridement of venous and mixed aetiology leg ulcers.

• To compare the effects of larval therapy and hydrogel on bacterial load and presence of MRSA.

• To assess staff and patient attitudes to and beliefs about larval therapy.

Exclusion criteria

People who:

• were currently in a trial evaluating other therapies for their leg ulcer

• had previously been entered into the trial

• were women of child-bearing potential

• were pregnant or lactating

• were allergic to hydrogel dressings or any of their components

• had grossly oedematous legs, which in the opinion of the recruiting health care professional were unsuitable for treatment with larval therapy and/or hydrogel

• were on anticoagulants (e.g. warfarin) and could not be admitted to a health care facility while receiving larval therapy (this exclusion became necessary during recruitment when the manufacturers of larval therapy added anticoagulation as a contraindication unless patients were closely monitored).

Ulcer area

Previously, ulcer area together with ulcer duration were identified as the two main prognostic factors for healing venous leg ulcers treated with compression. 85 Margolis et al. 86 then dichotomised these continuous variables, and found that an area of 5 cm2 and a duration 6 months were the cut-off points that maximised the differences between patients whose ulcers healed and those whose ulcers did not. In VenUS II baseline ulcer surface area was assessed using acetate tracings of ulcer perimeters (taken using a wound grid marked with 1 cm2 squares and marker pen). An assessment was then made by the recruiting nurse of whether the reference ulcer area was equal to or less than 5 cm² or greater than 5 cm2 for stratification purposes. The actual area was calculated at the York Trials Unit at a later date using the Mouseyes computer program. 87

Duration of reference ulcer

Longer ulcer duration is associated with longer time to healing and is an important prognostic variable;9,47 we therefore recorded duration of reference ulcer at baseline based on participant report.

Number of ulcer episodes on leg with reference ulcer

Data from VenUS I12 suggested that a greater number of ulcer episodes is prognostic of a longer healing time; we therefore recorded the number of ulcer episodes on the leg of the reference ulcer, based on participant report.

Sex

The sex of participants was recorded to allow comparison with the existing information on the population of people with ulcers in which women outnumber men. 88

Date of birth

Date of birth was recorded at recruitment so that age at recruitment could be calculated. Increased age has been associated with slower healing rates in one study. 89

Ulcer position and image

For monitoring purposes a record of the position of all ulcers on the trial leg including the reference ulcer was made. Digital photographs were also taken of all leg ulcers present at recruitment, including the reference ulcer.

Health-related quality of life

Participants were given a quality-of-life questionnaire booklet to complete immediately after recruitment. The questionnaire comprised the Short Form 12-item Health Survey (SF-12, Version 2, 4-week recall)90 and the European Quality Of Life-5 Dimensions (EQ-5D) instrument. 91

Ulcer microbiology

A swab was taken at baseline (described below) so that bacterial load in the ulcer could be assessed and the presence of MRSA ascertained.

Other

VenUS I reported that ulcer area, ulcer duration, number of ulcer episodes, ankle mobility (classified as full range of ankle motion, reduced range of ankle motion or fixed ankle), and body weight were prognostic for healing. These data were all collected at baseline, also recorded was an ulcer type variable (ABPI greater than 0.8 and treated with high compression; ABPI greater than 0.8 and not treated with high compression, and ABPI 0.6–0.8). We also assessed ulcer-related pain over the previous 24 hours using a visual analogue scale (VAS) (no pain to worst pain imaginable). The minimum value on the scale was 0 mm and the maximum was 150 mm.

Loose larvae group (Phase 1)

Participants received sterile larvae (L. sericata), (Zoobiotic, Bridgend Wales, UK). The number of pots of larvae to be used at a single application was determined by referring to a ‘calculator’ supplied by the manufacturer (Appendix 4.1). The calculator assesses ulcer size and percentage of ulcer covered with slough. The required number of larvae was then ordered from the manufacturer (Table 3). At the time of the trial it was not possible to order larvae for a Monday delivery.

Larvae were retained in the wound by securing a net mesh dressing with Sleek tape (Smith and Nephew, Hull, UK) onto strips of either hydrocolloid dressing or zinc paste bandage applied around the ulcer. Where wounds were dry and/or necrotic, the wound was kept moist by application of a saline-moistened gauze pad over the net mesh retention dressing. Larvae could be applied to an ulcer previously treated with Purilon hydrogel (because it does not contain propylene glycol which is toxic to larvae); however, the use of other dressings, including other hydrogels, between larval treatments was prohibited by the protocol.

Loose larvae were left on the ulcer for 3–4 days, although nurses could check the participant and larvae during this period (i.e. to rehydrate). The removal of larvae was a straightforward process. Once the mesh was removed the majority of larvae would fall out of the wound or move away from it and were caught in a suitable receptacle or retrieved with a forceps or a gloved hand. Any larvae remaining were gently removed manually or irrigated out of the wound with a jet of saline. Larvae were disposed of in accordance with the local guidelines for each site. Once the larvae were removed the treating nurse assessed the amount of slough/necrotic tissue remaining on the wound and decided whether a further application of larval therapy was required.

Where further application was required, Purilon hydrogel and the participant’s usual bandage were applied and more larvae were ordered for reapplication as soon as possible. The timescales of treatment were recorded. Once debridement was deemed complete by the treating nurse, or a decision was made to cease trial treatment before debridement, participants moved into Phase 2.

Bagged larvae (Phase 1)

Participants received sterile larvae (L. sericata), (Biomonde, Barsbüttel, Germany) within a sealed dressing. The number of bags was calculated according to the manufacturer’s specifications (Appendix 4.2) and ordered from the manufacturer (Table 4). At the time of the trial larvae could not be delivered for application on a Monday.

Bagged larvae were applied directly to the ulcer without the need for hydrocolloid/zinc paste retention strips. The bag was kept in place by a simple retention bandage, e.g. crepe. Bagged larvae remained in the ulcer for 3–4 days, although the nurses could check on the larvae during this period (i.e. to rehydrate). When ready for removal, the bag was simply lifted from the wound and disposed of intact. Larvae were disposed of in accordance with the local guidelines for each centre.

Where further application was required, Purilon hydrogel and the participant’s usual bandage were applied and more larvae were ordered for reapplication as soon as possible. The timescales of treatment were recorded. Once debridement was deemed complete by the treating nurse, or a decision was made to cease trial treatment before debridement, participants moved into Phase 2.

Control group (Phase 1)

Participants in the control group received Purilon hydrogel dressing (Coloplast A/S, Humlebæk, Denmark) covered with a knitted viscose dressing. Nurses discontinued the hydrogel treatment when they regarded the ulcer as debrided and the date was recorded. Once debridement was deemed complete by the treating nurse, or a decision was made to cease trial treatment before debridement, participants moved into Phase 2.

From baseline to withdrawal from trial (debridement) treatment (Phase 1)

Every nurse visit was recorded by the treating nurse along with location and reason for visit (Appendix 3.4–3.6, depending on trial arm). Nurses also recorded the number of applications of trial treatment. Visit information relating to trial debridement treatment was recorded until the reference ulcer debrided (as assessed by the treating nurse) or the participant was recorded as no longer receiving the trial treatment (both events resulting in the participant moving to Phase 2). During Phase 1, when the debridement treatment was removed participants were asked to complete a VAS (as completed at baseline) to assess how painful their ulcer had been over the last 24 hours (no pain to worse pain imaginable, where a higher score was worse).

We advised nursing staff that larval therapy and hydrogel should continue either until debridement or for a minimum of 6 weeks if debridement had not been achieved. Previous studies indicate that the median number of applications of hydrogel needed to achieve debridement is 15. 57

The reasons for deciding to stop delivering (withdrawing from) the trial treatment were recorded as:

• deterioration in the ulcer evidenced by clinically significant increase in reference ulcer size observed over 2 consecutive weeks

• no change in ulcer size but a significant increase in slough and/or necrotic tissue

• allergic reaction to larval therapy or hydrogel

• participant hospitalised and trial treatment not adhered to

If a participant stopped receiving the trial treatment they moved into Phase 2 of the study and follow-up continued as per the protocol.

From reference ulcer debridement/withdrawal from trial debridement treatment to trial exit (Phase 2)

Following ulcer debridement the trial treatment specified was a knitted viscose dressing and compression if appropriate (determined by ABPI and participant tolerance). Every nurse visit was recorded, along with location of visits and compression treatment received in the Dressing Log Booklet (Appendix 3.7, same for all three arms). If during Phase 2, a clinical decision was made to change the treatment for a participant who was still following the trial treatment protocol, a withdrawal-from-treatment form was completed with the reason for this change recorded and the new treatment noted.

Primary end point

The primary end point was defined as time to complete healing of the reference ulcer.

Ulcer healing was defined as complete epithelial cover in the absence of a scab (eschar). When nurses deemed the reference ulcer to be healed they were asked to seek independent verification from a nurse not involved in the care of the patient one week after the visit at which they regarded the ulcer as healed. If the two nurses did not agree that healing had occurred, the participant continued in the trial until they agreed that the reference ulcer had healed. Upon healing a ‘reference ulcer healed form’ was completed (Appendix 3.2).

Remote, blinded assessment of healing from photographs was also conducted. Nurses took a digital photograph of the reference ulcer every week until 6 months after recruitment and then monthly until healing or exit (Appendix 6). Before taking the photograph the nurses were requested to clean the skin surrounding the leg ulcer to reduce the risk of treatment detection in the photographs (e.g. remnants of zinc paste). The nurses were also asked to take an image 1 week after healing was regarded by them as having occurred.

Two blinded assessors independently assessed all photographs for each participant to determine a date for healing. This assessment process was piloted by two additional nurses before being undertaken for the main analysis. At the pilot stage it was noted that the trial nurses and clinical blinded outcome assessors viewed an ulcer as healed in the presence of a very small amount of eschar or scab. A revised definition for the blinded outcome assessment was agreed as:

Clinically, extensive epithelialisation is judged to have occurred (possibly under a very small amount of eschar/dry skin), no further primary wound contact dressing is judged to be required (maintenance treatment may be started).

The assessors discussed any discrepancies with referral to a third assessor for a final decision if required. The primary outcome was calculated using the date of healing as decided by the blind assessors. However, if no photographs were available for a participant then the date of healing decided by the treating nurse was taken as the healed date.

Collection of resource use data

During Phase 1 information recorded (Appendix 3.3–3.6) included reason for nurse visit and whether each visit was ulcer related; treatment received (used to calculate number of applications); the location of the visit and the compression system used.

During Phase 2 the Dressing Log Booklet (Appendix 3.7) recorded the reason for and the location of the visit and the compression system used.

Nurses were also asked to record details of participant hospitalisation, including the reason for the hospitalisation and length of stay (Appendix 3.8).

At baseline and at 3-monthly intervals thereafter participants were posted a questionnaire asking about health and social-care resource use during the preceding three months (Appendix 3.9). The questionnaire was designed for participant completion and was returned to the trial office using a reply-paid envelope. Participants indicated how many times in the previous month they had used health services (for example, seen a GP or nurse; or received hospital care) and any health service use that was ulcer related. The collection of self-reported resource-use data was continued until trial completion.

Time to debridement

Debridement was defined as a ‘cosmetically clean wound’. Nurses recorded whether a wound had debrided at the end of Phase 1; however, the main debridement outcome was assessed by remote, blinded outcome assessors from photographs of the reference ulcer using the definition: ‘a cosmetically clean wound requiring no further treatment for debridement’. If no photographs were available for a participant then the date of debridement healing decided by the treating nurse was taken.

Bacterial load

Data were collected using microbiological swabs taken at baseline and after each Phase 1 treatment up to debridement or for a maximum of 1 month (if the ulcer debrided within 1 month then weekly) and then monthly until healing or trial completion. A protocol for obtaining the swabs was provided for nurses. (Appendix 7) We used swabs, rather than biopsies, for sample collection. It was felt that the specialist skills required to take biopsies and the discomfort of this process for the participant (especially given the number of samples being taken) could not be justified based on existing data regarding the comparative quality of bacterial load data from each source.

Samples (dry swabs), identified by study number, participant’s date of birth and date taken, were sent by First Class post at ambient temperature to Micropathology Ltd, Coventry, UK. Upon receipt, samples were stored at – 70°C.

Bacterial load was assessed by measuring the amount of bacterial DNA present in wound samples. This molecular approach was able to provide a broad measure of bacterial load deemed suitable for this simple microbiological substudy. This approach was selected over other analytical methods that require specimen integrity, i.e. plating/culturing because of the logistical difficulties involved in processing a large numbers of viable samples from multiple centres across the UK. Because molecular analysis did not require sample preservation – swabs could be sent to a central location, stored and then analysed in batches. This batch analysis system and fact that bacterial load could be measured using a single assay, rather than requiring multiple cultures for subspecies growth, also meant that the molecular approach was much less resource intensive than qualitative analysis.

Bacterial DNA was released from swabs using a protocol adapted from Schabereiter-Gurtner et al. 92 Swabs were processed in batches that included negative and positive swabs to control for contamination and extraction efficacy (see ‘Quality control’). Swabs were broken off into labelled 2-ml sample tubes containing 500 µl XB buffer (150 mM sodium ethylenediamine tetraacetic acid, 225 mM NaCl, pH 8.5), 60 µl lysozyme (100 mg/ml) and 25 µl lysostaphin (1 mg/ml). The swabs were incubated at 37°C for 30 minutes and then subjected to three cycles of freezing (– 70°C) and thawing (72°C). Following this, 220 µl aliquots of each bacterial suspension were removed into a fresh tube and processed for nucleic acid extraction using the QIAamp DNA Blood Biorobot MDx kit combined with the Qiagen MDx Biorobot (Qiagen, Valencia, CA, USA) so that nucleic acids from 200 µl of sample were eluted into 200 µl TE buffer (10 mM Tris–HCl, 1 mM ethylenediamine tetraacetic acid, pH 8.0). The remainder of each sample containing the original swab was stored at – 70°C.

Nucleic acids from each swab were subjected to real-time quantitative polymerase chain reaction (PCR) using the following universal 16S ribosomal RNA gene primers: forward, 5′-AGA GTT TGA TCA TGG CTC AG-3′,93 reverse, 5′-ACC GCG GCT GCT GGC AC-3′. 94 The reaction mixture consisted of ReadyMix Taq PCR reaction mix (12 μl; Sigma-Aldrich, Poole, UK), combined primers (1 μl, 5 μM), MgCl₂ (1.5 μl, 25 mM), Evagreen fluorescent dye (1 μl; Biotium, Hayward, CA, USA) and template (10 μl, extracted nucleic acids). Real-time PCR was performed in a Rotorgene 3000 (Corbett Research UK Ltd, St Neots, UK) under the following parameters: 95°C for 40 seconds, followed by 30 cycles of 97°C for 20 seconds, 55°C for 20 seconds, and 72°C for 20 seconds (acquiring to the Sybr channel).

The bacterial load (10^x copies/ml) was calculated by comparison of extracted DNA levels with a standard curve, which was generated for each assay using 10-fold dilutions of DNA extracted from a culture of S. aureus that had been quantified by viable and particle-counting methods. Unknowns were calculated by interpolation and results were expressed as 16s ribosomal RNA gene copies per ml. The lowest limit of detection of the assay was 10³ copies/ml.

Positive and negative extraction controls were included in each batch of swabs processed. Positive controls were prepared by absorbing 500 µl of a 4-hour broth culture of S. aureus onto sterile swabs, which were stored at – 70°C until required. Negative controls were sterile swabs only. Extraction controls were processed using the same protocols as live samples and the extracted nucleic acids was analysed as described previously.

Positive extraction control swabs were expected to provide bacterial load results of at least 1 × 10⁴ S. aureus genome equivalents/ml and negative control swabs were expected to have a bacterial load of less than 1 × 10³ S. aureus genome equivalents for results from that batch of swabs to be accepted.

Each quantitative PCR assay also included a quality control sample that consisted of extracted nucleic acids of a known bacterial load [internal quality control (IQC)]. This IQC sample had to provide a result varying no more than 0.5 of a log from the predicted value for the results of that PCR assay to be accepted.

MRSA assay

All baseline swabs were tested for MRSA. Those participants with a positive baseline swab then had all further weekly/monthly swabs analysed for the presence of MRSA. Participants with negative swabs at baseline had one further swab analysed at 2 months to see if MRSA was detectable at the time when debridement treatment would have ceased in most cases.

MRSA was detected using a single-round PCR assay specific for the 3′-terminal region of the Staphylococcal Chromosomal Cassette containing the mecA gene responsible for methicillin resistance (SCCmec) when specifically inserted into the S. aureus genome at orfX.

Nucleic acids were obtained from swabs as described for the bacterial load swabs and subjected to single-round PCR using the following primers: forward rjmec 5′-TAT GAT ATG CTT CTC C-3′ and reverse orfX 5′-AAC GTT TAG GCC CAT ACA CCA-3′. 95 The reaction mixture consisted of PCR buffer (5 μl; Labmaster, Sevenoaks, UK), dNTPs (5 μl; Web Scientific, Crewe, UK), combined primers (1.5 µl, 5 μM), water (18.5 μl; Sigma-Aldrich, Poole, UK), MgCl₂ (1 μl, 25 mM; Sigma-Aldrich), and Immolase DNA polymerase (0.5 μl, Bioline, London, UK). The cycling parameters were 95°C for 20 seconds, followed by 45 cycles of 97°C for 20 seconds, 57°C for 20 seconds, and 72°C for 20 seconds. PCR products were detected by ethidium bromide and agarose gel electrophoresis. Products were referenced to a positive control (MRSA, kindly provided by Prof. C. Dowson, University of Warwick, UK) and results were scored as either ‘detected’ or ‘not detected’. The minimum detection threshold for MRSA was ascertained to be equivalent to 1 × 10³ CFUs/ml.

Health-related quality of life

Generic instruments were used to measure participants’ perceptions of health outcome in this trial; these have previously been shown to be sensitive to changes in venous ulcer healing status. 96 Generic instruments are also particularly useful for comparisons across groups of participants and have wide scope for use in economic evaluation. Their generic nature also makes them potentially responsive to side or unforeseen effects of treatment. The package of instruments was designed to be comprehensive yet brief and suitable and was administered at baseline and at 3, 6, 9 and 12 months’ follow-up.

Participants received the instruments by postal questionnaire to be completed either in the clinic waiting room or at home. The booklet included the EQ-5D and the SF-12. We used a layout of the SF-12 shown in previous work to yield improved response rates and quality. 97

A systematic review investigating ways of increasing questionnaire response rates reported that response to postal questionnaires was doubled [odds ratio = 2.02; 95% confidence interval (CI) 1.79 to 2.27] when a financial reward was included with the questionnaire, versus no incentive. 98 The response rate increased further when the incentive was not conditional on response, versus upon return of questionnaire (odds ratio = 1.71; 95% CI 1.29 to 2.26). Based on these data, we offered an incentive for participants to return their questionnaire. We notified participants in their 9-month questionnaires that once we received their final questionnaire they would receive £5 in recognition of their commitment to our study and the time they spent completing questionnaires.

Adverse events

An adverse event can be defined as ‘any undesirable clinical occurrence in a subject, whether it is considered to be device related or not’. 99 Although it is normal to monitor adverse events within an RCT, we had a mandatory responsibility to collect these data as part of VenUS II, given that it was a trial of an investigational medicinal product that is unlicensed in the UK. Both treatment-related and unrelated adverse effects were reported to the trial office on an adverse-event reporting form. The reporting clinician indicated whether, in their opinion, the event was related to trial treatment, or not. Events were also classed as serious or non-serious. Some events were always classified as serious (death, life-threatening risk, hospitalisation, persistent or significant disability/incapacity). For other events the treating nurse made a clinical decision about the seriousness. Nurses were asked to report serious adverse events, such as admission to hospital, directly to the Trial Coordinator. Nurses were asked to report all adverse events on the adverse event form and state whether they considered them to be related to the trial treatment (possibly/probably/definitely or not). We established a list of possible treatment-related adverse events a priori, based on reports in the literature and the VenUS I trial. These are described as follows.

Recruitment to the qualitative study

Baseline data

All baseline data were summarised by treatment group. Baseline data were summarised descriptively. No formal statistical comparisons were undertaken.

Primary analysis

Time to healing was derived as the number of days from randomisation until the date of healing as confirmed from the blinded photograph assessment (or the date recorded by the nurses if no photographs were available). Participants who did not heal were treated as censored and their date of trial exit, date of their last available assessment, or 365 days/trial cessation, as appropriate was used to calculate their duration in the trial.

An initial analysis of time to healing of the reference ulcer compared the three treatment groups using a log-rank test. Kaplan–Meier survival curves were constructed and the median time to healing and corresponding 95% CI for each group were calculated. A Cox proportional hazards model was used to adjust the analysis for the randomisation stratification factors (centre, baseline ulcer area) as well as duration, and ulcer type. Actual baseline area (as measured from the tracings) and duration of ulcer were used. A treatment contrast was included in the model to compare healing rates between loose and bagged larvae. It was decided a priori that if there was no evidence of any difference between the loose and bagged larval therapy then the hazard ratios and 95% CI would be presented for larval treatment overall (loose and bagged combined) compared with hydrogel. This strategy was not detailed explicitly in the original study protocol but was stipulated in the analysis plan, which was written before any data analysis. If there was a statistically significant difference (p ≤ 0.05) in healing rates between loose and bagged larvae then the hazard ratio and 95% CI would be presented separately for each type of larval treatment compared with hydrogel. The assumption of proportional hazards was assessed graphically using log–log plots of the estimated survivor function and Schoenfeld residual plots. We also included interaction terms between each variable and log (time) in the Cox model. The results from the adjusted analysis are presented as the primary results.

As there were a number of centres that only recruited a small number of participants (fewer than 10), sensitivity analyses were conducted to assess the impact of different methods of handling these small centres. This was done by: combining small centres, stratifying by centre, and by excluding centre from the model.

Secondary analyses

Health benefits

Resource use and unit costs

A cost for each trial participant was calculated as the product of resources used by their relevant unit cost. Resource-use data were collected in nurse questionnaires and self-reported participant questionnaires.

Analysis was carried out using 2006 costs in Pounds Sterling.

Three different elements were considered in the estimation of costs:

• trial debridement treatment

• health care provider time (visits to/from health care provider for leg-ulcer-related reasons)

• cost of compression therapy.

Other treatments, such as primary and secondary dressings or skin preparations, were assumed to be used equally across treatment arms: these resources were not included in the economic analyses.

Mean cost estimation

Resource use data collected are also subject to censoring and estimating the mean total cost based on complete case analysis will underestimate the true expected costs. An IPW regression (as described above) was used to account for censoring, possible baseline imbalances and randomisation stratification variables. Mean cost estimation was made by partitioning the study period into multiple time intervals (quarterly periods) and the IPW regression per period estimates were then summed to obtain total expected costs.

Cost-effectiveness

As in the clinical analysis, in the incremental cost-effectiveness analysis estimates of total cost and health benefit (ulcer-free days and QALYs) were reported for larval treatment (pooling data from the loose and bagged larvae groups) and hydrogel treatment. Nonetheless, for completeness descriptive measures on costs and health benefits were described for each of three trial arms.

To assess cost-effectiveness, we compared the mean difference in costs between trial arms with the mean difference in number of ulcer-free days and QALYs. The results of economic evaluation studies can be graphically represented in what is known as a cost-effectiveness plane where results are characterised in one of four potential scenarios (Figure 1).

FIGURE 1.

Cost-effectiveness plane showing cost and benefit differences between alternative treatments (NW, north-west; NE, north-east; SW, south-west; SE, south-east).

Alternative scenarios on the cost-effectiveness plane:

• If the results from the economic evaluation analyses of VenUS II fall in the north-west (NW) quadrant of the cost-effectiveness plane it implies that larval therapy is dominated by hydrogel, in other words, larval therapy is more expensive and is associated with fewer health benefits than hydrogel. In this case, the decision regarding adoption of larval therapy is straightforward and one should not adopt it.

• Similarly a straightforward decision can be made when the cost-effectiveness measure is positioned in the south-east (SE) quadrant. In this situation, larval therapy would be dominant, i.e. estimated to be less costly and more beneficial than hydrogel and should therefore be implemented.

• In the NE or SW quadrants, decisions are more difficult to make. Here one must evaluate whether the increased cost of the new intervention is worth the increased benefit (NE scenario), or if the reduced benefit conferred by the new treatment is justified by the reduced costs (SW).

To ascertain the cost-effectiveness of a health care intervention relative to another in the absence of dominance, one needs to conduct an incremental analysis of cost-effectiveness. The incremental cost-effectiveness ratio (ICER) is the most commonly used cost-effectiveness measure and combines costs and health in a single measure to which a decision rule for cost-effectiveness can be applied. It combines costs and benefits in a ratio of the mean difference in cost between the alternative treatments being compared with the mean difference in health benefits such that:

where: C1 is the mean cost associated with the technology under evaluation (larvae); C0 is the mean cost associated with the technology of comparison (hydrogel); B1 is the mean health benefit associated with the technology under evaluation (larvae); and B0 is the mean health benefit associated with the technology of comparison (hydrogel).

The decision rule for cost-effectiveness on the basis of the ICER indicates that a treatment strategy can be considered cost-effective only if the decision-maker’s willingness to pay for an additional unit of health benefit (QALYs, ulcer-free days) is greater (or equal) to the ICER. In this context, the decision-maker is responsible for establishing the willingness to pay.

Bacterial load

Table 28 shows summary statistics of the baseline bacterial count (logs to base 10) and the within-participant change over time (baseline to final) for three different groups of participants: (1) all participants (to last available swab); (2) only those participants who debrided (to last available swab during the debridement treatment phase); (3) participants who did not debride (change from baseline to last available swab). The average log bacterial count at baseline was 6.5 (approximately 3.1 × 10⁶ copies/ml) and the three treatment groups were similar. The summaries of the within-participant change from baseline show that on average, most participants, regardless of whether or not they debrided, had very little change in the bacterial load of their ulcer over time with median log differences in the range of – 0.5 to 0.2 copies/ml.

To investigate whether there were any differences between the larvae and hydrogel treatments in reducing the bacterial load of the ulcer, a repeated measures analysis (linear random coefficients model) was used. Two models were used; one analysed data from all available swabs for each participant; the other used only swab data up to the point of debridement for those participants who debrided and to the end of the trial for any participants who did not debride. The results for the model using all data and for the model using only data up to debridement are in Table 29. When using all swab data there was no evidence of a difference in bacterial load over time between larvae and hydrogel [p = 0.75, estimate of the mean log bacterial count (standard error) for larvae was 6.59 copies/ml (0.06) and for hydrogel it was 6.64 copies/ml (0.08)]. Time had a significant effect (p = 0.01) indicating that overall ulcer bacterial load decreased over time, but the interaction between treatment and time was not significant (p = 0.63) indicating that decreases in bacterial load over time did not differ between the larvae and hydrogel groups.

When analysing only swab data up to the point of debridement, there was also no evidence of a difference between the larvae and hydrogel groups [p = 0.86, estimate of the mean log bacterial count (standard error) for larvae was 6.72 copies/ml (0.07) and for hydrogel it was 6.73 copies/ml (0.09)]. Furthermore, there was no evidence of a time effect and bacterial load appeared to remain constant over time (p = 0.91), and again there was no evidence that the effect of time differed between the larvae and hydrogel groups (p = 0.65).

MRSA

Overall only 18 out of 267 participants (6.7%) had MRSA detected in their baseline ulcer swab, of whom seven were in the loose larvae group, five in the bagged larvae group and six in the hydrogel group (Table 30). MRSA was absent from swabs taken at the end of the debridement phase [to debridement if they debrided, or the end of the trial period if they did not debride (one loose and one hydrogel)] in 100% (5/5) of participants in the bagged larvae group, 57.1% (4/7) of participants in the loose larvae group and 50% (3/6) of the hydrogel participants. Three participants in the loose larvae group but none in the other groups had a recurrence of MRSA after debridement.

We compared the proportions of participants with MRSA detected at baseline who were free from MRSA by the end of the debridement treatment (Phase 1) – there was no evidence of a difference between the larvae and hydrogel groups (75% or 9/12) compared with the hydrogel group (50% or 3/6), Fisher’s exact test p = 0.34). This analysis was repeated for the proportion of participants testing negative for MRSA who were found to have MRSA at one or more follow-up assessments and again there was no evidence of a difference between the larvae and hydrogel groups (7.1% or 12/168) compared with 2.5% (2/81), Fisher’s exact test p = 0.16.

Trial debridement treatment

The number and duration of trial debridement treatments are described in Table 34. Participants receiving larval therapy obtained their first treatment application approximately 3 days later than participants allocated to hydrogel (because of the need for ordering and delivery). Those allocated to one of the larval therapy arms had, on average, 1.45 trial treatment applications before the debridement treatment was discontinued (i.e. participant was moved to Phase 2) or data were censored. Participants in the hydrogel arm received, on average, 9.2 applications of Phase 1 treatment before being moved onto Phase 2 or data were censored.

Nineteen participants never received the trial debridement therapy (see Figure 4). Data regarding nurse visits were missing for 14 participants, while five people received a treatment other than the trial treatment (one in the loose larvae arm and four in the hydrogel arm). The duration of Phase 1 (debridement) treatment was, on average, 30 days longer in the hydrogel arm (mean days of treatment 43 days) than the larval therapy arms (mean days of treatment 12 to 13 days).

The average estimated cost of the trial debridement treatment, per application was: loose larvae £71.70 (SD £13.40; minimum to maximum £51.50 to £132.50); bagged larvae £111.90 (SD 33.6; minimum to maximum £80.10 to £218.50) and hydrogel £1.50 (SD 0).

Visits to/from health care providers

Visits to and from health care providers were recorded by participants and used for the base-case analysis. These self-reported data suggested that the number of consultations with health care professionals was similar in each trial arm (Table 35) with the majority of nurse and hospital visits being related to ulcer treatment.

The proportion of visits that took place at home was similar between all arms (63% for the loose larvae arm and 64% for bagged larvae arm; 67% for hydrogel). A summary of the unadjusted ulcer-related costs of health care provider visits for each of the trial treatments is presented in Table 36.

Compression therapy

The use of high-compression bandaging through the trial was similar across all arms (Table 37).

Total costs

The cost of nurse visits was the major driver of total costs. In the base-case analysis, patient-reported data on ulcer-related visits to and from health care providers were combined with trial debridement treatment costs and compression therapy. Quarterly estimates are presented in Table 38.

To account for the censored nature of cost data, mean differences in ulcer-related costs between treatments were estimated using inverse probability weighted regression estimates of time to survival. The results of the base-case analysis show that larval therapy costs, on average, £96.70 more per participant per year (95% bias corrected CI –£491.90 to £685.80) (Table 39). This difference was not statistically significant.

Mean time to healing

The difference in estimated mean time to healing (over 12 months) favoured larval therapy. On average, participants treated with larval therapy healed 2.42 days before those in the hydrogel arm. However, this difference was not statistically significant (95% bias corrected CI of the difference was from – 40.95 days to 31.91 days) (see Table 40). Estimation through the restricted mean approach (see Chapter 2) returned 2.74 additional ulcer-free days for larval therapy users, confirming the appropriateness of the IPW regression.

Utility and QALYs

Quarterly utility scores per participant by trial arm are presented in Table 41 and unadjusted average QALYs per group are described in Table 42. The results show that, after adjustment for original imbalances in utility scores at baseline, stratification covariates and after accounting for the censored nature of data, individuals in the larval therapy arms had, on average, a better quality of life than individuals in the hydrogel arm [the annual difference in QALYs was 0.011 (95% CI – 0.067 to 0.071), see Table 43].

Scenario 1: nurse-reported data excluding amputation

Unadjusted costs estimated using nurse-reported visit data indicated that bagged larvae were more costly (Table 45). Conversely, patient-reported costs data indicated that loose larvae were more costly. The adjusted analysis of costs estimated smaller costs for larvae users (– £31.30); however, this was not statistically significant (Table 46). As only costs are subjected to sensitivity analysis, the cost-effectiveness/utility was based on the health benefits estimates of the base-case analysis (Tables 40 and 43). The decision to adopt larval therapy was associated with considerable uncertainty (Figures 13 and 14).

FIGURE 13.

Cost-effectiveness plane, scenario 1. QALY, quality-adjusted life-year.

FIGURE 14.

Cost-effectiveness acceptability curve, scenario 1. CE, cost-effectiveness; QALY, quality-adjusted life-year.

Scenario 2: nurse-reported data including amputation costs

The cost estimates for this scenario show that unadjusted total costs incurred by participants in the hydrogel group are higher than for the larval therapy groups (Table 47). Adjusted cost estimates (Table 46, scenario 2) confirm this result, suggesting that the use of larval therapy may reduce costs by £227 per patient per year. The CEAC and cost-effectiveness plane show that the probability of larval therapy being cost-effective varies between 55% and 75% for a willingness-to-pay range between 0 and £30,000 (Figures 15 and 16). The scenarios further explained the uncertainty surrounding the decision of adopting larval therapy. Although the number of participants who had an amputation is very low (three participants), the high costs incurred shifted the estimate of cost difference. As information on this event was not collected systematically but only through ‘reason for hospitalisation’ data, we might be underestimating the event rate and consequently biasing the cost estimates. Nevertheless, amputation is revealed to be an important cost driver for leg ulcer participants.

FIGURE 15.

Cost-effectiveness plane, scenario 2. QALY, quality-adjusted life-year.

FIGURE 16.

Cost-effectiveness acceptability curve (CEAC), scenario 2. CE, cost-effectiveness; QALY, quality-adjusted life-year.

One-way sensitivity analysis

As identified across other studies in wound care, both the setting and the duration of nurse visits are important in the evaluation of cost differences. A one-way sensitivity analysis was conducted to evaluate the influence of nurse visit costs on the cost difference. To this end, duration of nurse visits was increased/reduced by up to 13 minutes, i.e. the maximum reduction evaluated was 13 minutes and the maximum increase was 13 minutes from base-case values. As a consequence, a 5-minute reduction in clinic visit duration from the base-case analysis resulted in values of 31 minutes for a home visit and 17 minutes for a clinic visit (see Table 9 in Chapter 2).

The sensitivity analysis shows that homogeneously reducing visit duration increases the cost difference (Figure 17). A 5-minute reduction in each visit increases the cost difference to £132 [bias corrected 95% CI (£– 408.90 to £652.10)]. While the differences were not statistically significant, the uncertainty surrounding the cost difference reduces as the visit duration reduces.

FIGURE 17.

Influence of nurse visit duration on adjusted cost difference, one-way sensitivity analysis. Bias corrected 95% CI based on 2000 bootstrap replicates.

The ratio of clinic to nurse home visit duration was subjected to sensitivity analysis. In the base-case analysis this ratio was 40 : 22, and in the sensitivity analysis it varied between 30 : 22 and 60 : 22.

As the duration of home visits increases in relation to clinic visits, the cost difference reduces and the uncertainty increases, i.e. confidence intervals are wider (Figure 18). None of the scenarios considered in the sensitivity analysis revealed a statistically significant difference in costs at conventional levels of significance.

FIGURE 18.

Influence of home vs clinic nurse visit duration ratio on adjusted cost difference, one-way sensitivity analysis. Bias corrected 95% CI based on 2000 bootstrap replicates. The dashed line indicates the base-case analysis.

Pain

While some participants said they were able to ‘shrug off’ or tolerate pain from their ulcers, others described excruciating pain which kept them awake at night, or led them to depend on medication to enable them to sleep.

Varying levels of pain were attributed to the stage of healing of their ulcer, to external factors, such as hot weather, and to the application of compression bandaging.

they don’t hurt me, I’m one of those people who can suffer pain I think.

(P17)

you get quite depressed, the pain is obnoxious, absolutely obnoxious.

(P10)

you’re just going to sleep and then it ‘whoa, what the hell?, and it wakes you up. It’s only the ulcer telling you ‘I’m here’, isn’t it? The pain of it wakes you up…

(P7)

it’s a lot better, at the beginning I was taking a lot of pain killers, but I don’t take any now.

(P18)

I was having to keep the leg outside the covers because the heat was making it worse.

(P18)

the pain sometimes goes, like after a day or two on your tablets it will subside and it’ll be okay, but then you come to get it redressed and you know you are going to get pain after the redressing of the bandages.

(P11)

Mobility

The majority of the participants interviewed attributed lack of mobility to comorbidities such as osteoarthritis (resulting in stiff knees and hips) rather than to pain from their ulcer. However, the two youngest male respondents (P5 and P14) said that their leg ulcers restricted how far and for how long they could walk, which had a negative effect on their ability to work.

Although some participants were determined that they would not let their ulcer ‘slow them down’, others highlighted some of the physical and psychological consequences of immobility because of their ulcer pain. Reduced mobility was perceived as linked to social isolation and a dependence on others (P16). The importance of being able to drive to keep up social contact with his family and friends was stressed by P15.

I don’t let it slow me down, I mean I struggle with my walking anyway because I am waiting for an operation on my left hip…so my movement is relatively slow and limited.

(P2)

I was avoiding walking, you know, I need to do more walking to get back some level of physical fitness and get rid of that… [pats his stomach]

(P1)

I can’t walk, I can’t really walk more than 100 metres without feeling, well not severe, but average pain and I have to stop.

(P5)

I can move and walk for one hour, two hours, but more than that, no…it hurts me a lot and sometimes makes me fed up with life.

(P14)

it means that I can’t walk, I can’t go out, I can’t go into [name] or on the bus to [name] where I used to…so I do miss going out, yes I do…walking down to the post office to get my pension… but I haven’t been able to, so my daughter-in-law does a lot for me…

(P16)

I can drive, once I get into my car, not far, but I can drive, yes…

(P15)

Social embarrassment

Having to wear special clothes or shoes to accommodate compression bandaging was associated with social stigma by three (male) respondents. Asked about the effect of his leg ulcer on his life, P11 responded that it was:

Terrible…when I am in the bandages, the compression bandages…this is the only pair of shoes I can get in. I take 12s or 13s anyway, so when I’ve got the compression bandages, these are the only pair of 14s I’ve seen in my life, and I have had to take the laces out, so it means unless I go out in these shoes, I don’t go out for a meal or anything.

(P11)

P6 felt similarly socially curtailed because of difficulties in finding shoes to fit:

because you can’t get your shoes on when they put all these bandages on, you can’t get your shoes on and you can’t get dressed up or anything to go out…

(P6)

P3 described attracting unwanted glances due to his wearing shorts rather than full length trousers for comfort with his compression bandages.

if you can get over the embarrassment because you are walking around with a bandage on your leg, I could wear the pants and then my legs get very hot which is bad for them, so I tend to wear shorts, so people are always looking at me.

(P3)

The pervasive odour emanating from leg ulcers was regarded as a further potential source of social embarrassment:

it’s like rotten cabbage…it’s like a stinky cabbage…when it’s really on its bad side, it does not smell very nice.

(P6)

Nuisance factor

Having a leg ulcer was perceived as time-consuming (P3), activity restricting (P5, P6, P7, P11, P12, P14, P16) and a general nuisance (P10, P13, P15, P18) because of having to attend regular clinic appointments, spending time (and sometimes money) in consultations with doctors and nurses, and running into difficulties pertaining to care of the ulcer when travelling away from home.

it is time-consuming, it does restrict you to go anywhere, you’ve got to allow time for it.

(P3)

within a fortnight I saw a specialist, if you pay, you’ll get in that much earlier…

(P15)

when I went to see the Dermatologist at the hospital, she did suggest the pressure bandage…but that created its own problems, particularly as we go away in our motor home and each time we were going away, I was having to contact local surgeries in the area we were travelling to…

(P18)

Acceptability of larval therapy

The majority of the participants interviewed stated that they would be prepared to have larval therapy as a treatment for their leg ulcer. Five out of the 18 participants had tried larval therapy, and a further 10 participants said that they would be willing to try it.

Two respondents (P13 and P16; both female and aged 76 years) stated categorically that they would not find larval therapy acceptable under any circumstances. However, interestingly, one of these participants, (P16) revised her opinion during the course of the interview when she learnt that larvae were obtainable in ‘bagged’ form; this prompted her to say that she might be prepared to consider having larval therapy.

One respondent (P8) was equivocal about whether or not he would accept larval therapy. He said he would need to be convinced of its effectiveness before he could make a decision. Figure 19 summarises participants’ responses to the question posed during interview concerning acceptability of larval therapy.

FIGURE 19.

Patient willingness to try larval therapy.

Of the 15 respondents who said that they would be prepared to accept, or had accepted, larval therapy as treatment, 13 said that they would be prepared to accept either loose or bagged larvae. Of the two remaining participants, one said he was unsure about whether he would prefer to have loose or bagged larvae, stating that he would wish to know more about the effectiveness of the different types and to be shown them before he could express a preference, while P18 had been recruited to VenUS II and had been randomised to receive loose larvae. Given a choice, she said she ‘might have preferred’ to have had bagged larvae, but she emphasised that this was not a strong preference as she ‘didn’t mind having the loose’.

Prior experience of larvae generally

Male participants were more likely than female participants to have handled ‘maggots’. Of the 12 male participants, seven had handled maggots, in connection with fishing as a hobby as a child, two of whom (P5 and P11) mentioned that they had held or ‘rolled maggots’ in their mouths when fishing. Neither of the male participants who were born outside the UK (P12 and P14) had had any experience of handling maggots as a child.

None of the female participants mentioned that they had handled larvae. P9 said she had seen them as a child ‘in a tin’ belonging to her brother when he had been going fishing. P16 could not recollect if she had actually ever seen larvae, though she remembered that her son had used them when he went fishing when he was young.

Of the five participants who had received larval therapy (three male and two female), two of the males (P11 and P17) said that they had had experience of handling maggots, while the other three participants (one male and two females) (P15, P9, P18) had not.

Factors associated with a willingness to accept larval therapy

Factors associated with a reluctance or unwillingness to accept larval therapy

Factors associated with reluctance or unwillingness to accept larval therapy were evident in the accounts of three participants (P8, P13 and P16). These included: feelings of squeamishness or strong feelings of aversion to larval therapy; little or no previous contact with maggots; scepticism about the benefits of larval therapy; negative views of larval therapy shared by family members.

The accounts of the three participants, derived from their interviews, are presented here in the form of case studies or summary descriptions106 to illustrate how the interplay of these factors could lead the participants to say that they would be likely to reject larval therapy as a form of treatment for their leg ulcer (Boxes 1–3). During the course of her interview, P16 appeared to revise her attitudes towards larval therapy, as she learnt more about it, and as she reflected on a discussion she had had with her son just before the interview. P8 implied that he might be willing to accept larval therapy if there was strong enough evidence to support its effectiveness in healing ulcers, and if its application could be effected in a manner that did not distress his wife or cause any domestic upset. P13 seemed adamant that she would never consider larval therapy an acceptable form of treatment.

Setting where patients received larval therapy

Four of the five patients experienced larval therapy in their own homes, applied by nurses who were members of the community nursing team, generally well-known to participants, who referred to them by their first name and spoke highly of them and their dedicated approach to care.

P11 was the only participant in hospital on account of his leg ulcer when the larvae were applied, and his experience was in marked contrast to the others. He complained that he had been made to feel like a ‘peep show’ during his stay in hospital as nurses whom he did not know came to take a quick ‘peep’ at the larvae when they were being rehydrated, their use being something of a novelty on the ward.

A major potential drawback of receiving larval therapy in a community setting was highlighted by one the two participants who suffered intense pain soon after the larvae were applied. P18 described seeking advice, in vain, during a pain-filled and sleepless night, and how she had had to wait for the nurse to return the following day to have the larvae removed and the pain relieved.

I mean, I couldn’t take them off myself, I had to wait until the next day, if it had been happening during the day, I would have contacted the Sister and asked her to come round, but because it happened overnight I couldn’t do anything about it, and I didn’t get any sleep, I was just sitting watching the clock go round until the time that she was due to come, and just kept taking the painkillers…I rang [the out of hours service] at about half past three in the morning, and left a message and he [the emergency doctor] rang me back but you see, a lot of people don’t know anything about the larval therapy, so when I told him about it, it was something he had not come across before, and all he could suggest was keep taking the painkillers…

(P18)

Perceived competence of staff

Participants who received larval therapy in their home were pleased with the experience and praised the nurses who applied it. The manner in which the nurse broached the topic of using larval therapy was perceived as important by P17, who was pleased that the nurse had ‘asked’ his permission to use it, so that he felt drawn in to the decision-making process, though in fact he happily deferred to her expertise saying that he would try anything she recommended: ‘whatever you want, I’ll try’ (P17).

By contrast, P11 judged his experience of receiving larvae in hospital as a disappointment; his stay coincided with a Bank Holiday, the Tissue Viability Nurse was apparently unavailable, and he was critical of the nurses who had applied the larvae to his ulcer.

I was just a bit disappointed that they were actually put on by someone who really hadn’t done it before and wasn’t one of the Tissue Viability Nurses…it were people who had to read the instructions on the packet…and they were like feeling their way, and to me, something like that is a specialist thing and someone like a main person from, a district nurse type, I think should have been [there], and then as they got on, the seniority got less and less…they weren’t at the top of their game on that front…

(P11)

Expectations versus experience

Reflections on the experience of having larval therapy

Asked about their understanding of how larval therapy works, two of the male participants referred to larvae eating dead cells, and then eating each other. None of the participants seemed particularly well informed at a detailed level about how the larvae actually functioned; mostly, they referred to bits of information they had picked up from newspaper or magazine articles, or books they had read.

apparently, they devour the cells do they not, I think [name of nurse] said one eats the other one, and one eats the other one, and one eats the other one that’s eaten the bad skin…

(P15)

I knew alright about them, from reading books.

(P17)

Only participant (P18) referred to larval enzymes, in relation to the reaction she experienced at the outset of her treatment, the causal explanation offered by the nurse:

she said that there are a few people who get a reaction to the enzymes from the larvae, not the larvae itself, but from the enzymes…

(P18)

Participants believed that the views of people in general about larval therapy would be characterised by ‘fear of the unknown’ (P17) and distaste because of possible connotations of uncleanliness, leading to a degree of social stigma about having larval therapy:

for the majority of people, it is an unknown quantity…I am convinced that it’s people’s fear of them rather than anything else, you see the mere mention of maggots puts them off…

(P15)

P9 referred to people talking about maggots in a hushed tone, or preferring to refer to them euphemistically as ‘that new treatment’. P18 contrasted her attitudes towards larval therapy with those of her friends: ‘I know some people when I mentioned it to them seemed quite shocked at the idea, but it didn’t bother me…’ (P18)

With one exception (P11), participants seemed to retain their prior positive attitudes towards larval therapy, whether they had a ‘good’ or ‘negative’ experience of using them. Where the experience was deemed ‘good’, by the participant, their views of the beneficial effects were reinforced. Participants who seemed to have reported a ‘negative’ experience (P9 and P18), because of the pain they felt, still believed that they had benefitted from the treatment during the short time that they experienced it.

I would have maggots again tomorrow if necessary.

(P17)

I just think if they are the right thing for ulcers, then so be it…if they are going to help people…

(P15)

I don’t regret being in the trial, even though there was a reaction…

(P18)

I’ve had maggots which I thought were a wonderful, wonderful thing. I believe in them thoroughly…

(P9)

P11, who had had bagged larvae applied to his large necrotic ulcer in hospital, was the only participant who appeared disillusioned by his experience. Holding high expectations at the outset, he appeared disappointed in the small improvement he discerned in his ulcer. P11 implied that he thought the ulcer might have improved more markedly if more larvae had been applied: ‘I probably could have done with two bags bigger than what was there, or another bag…’ (P11). During his stay in hospital, P11 felt that the treatment had been mismanaged for a variety of reasons; in particular, he had been unhappy that the treatment was not applied by the Tissue Viability Nurse Specialist, as he had hoped.

Nurse participant characteristics

All nurse participants were female and their ages ranged from 30 years to 62 years; the median age was 45 years (Tables 51 and 52). Participating nurses had between 1 and 27 years’ experience of caring for people with leg ulcers (median was 9.8 years, mean was 11.3 years). Three participants (N8, N12 and N16) had no experience and three participants (N15, N18 and N20) had used larvae on only one or two occasions. Those nurses without any experience of using larval therapy were community nurses; all of the clinic nurses had experience of using larval therapy. Sixteen nurse participants had more extensive experience of using larval therapy. Of the 19 nurses who had used larval therapy, 17 (89%) had used loose larvae, 13 had used bagged larvae (68%) and 11 nurses had used both (58%).

Perceived utility, benefits and effectiveness of larval therapy

Nurse participants were unanimous in the belief that larval therapy was best suited to cleaning ‘dirty’ or ‘sloughy’ wounds to achieve a speedy debridement, and a dramatic, visible improvement in the condition of an ulcer, providing a morale booster for both patients and the staff caring for them:

I mean you can see the improvement…the area that they have cleaned up, it’s been dramatic really in most of them, it’s so good.

(N12)

it was quick, it was extremely quick, say if we put them on a Monday, by the time we got back, two to three days later, maybe 60–70% of the slough was gone, so fast, so quick fast….it did us all good, we thought, ahhh, right, we are getting somewhere after all these weeks and months something has finally happened…so from that point of view, the speed with which they can make a difference is significant I would say…

(N13, community nurse with 27 years’ experience of wound care)

you put them on and you can see instant results…all the slough is gone when you take them off, it is so good because it is so visual.

(N14)

The role of larval therapy in the ‘palliative’ care of wounds that were malodorous and offensive, and deeply unpleasant for patients and their families was valued:

if you’ve got an ulcer that full of slough, it smells, and the quickest way to get rid of that is to use maggots.

(N11)

The use of larval therapy following, or instead of, surgical debridement was described as a cost-effective and safe means of cleaning a ‘dirty’ wound quickly:

I’ll get them to come in and sharp debride the worse of it and then we’ll use maggots to finish off, and that’s probably my favourite way of using them, because otherwise I’ve got to use two or three applications and that’s expensive.

(N11, Tissue Viability Specialist Nurse)

sometimes the other option is the surgeon will say, we’ll debride it, and people don’t want to go and have surgery…and they [the patient] said, I don’t want it, and I say, well, let’s try maggots…it’s got to be cost-effective, and we don’t want MRSA, that’s another thing…

(N14)

The major perceived drawback of using larval therapy was that the results achieved in a short space of time were not maintained. Nurses described wounds as ‘going in a backwards direction’ (N3) soon after the removal of larvae; ‘after you take them off, the wound tends to go back to being sloughy’ (N1).

we’re not keeping it clean with an NA dressing, it is not enough…10 days down the line they’ve got a sloughy wound again.

(N2, clinic nurse)

I think they are right for the job that they do, but there is always a need once you take the larvae off, to have a different dressing that will keep the wound clean, because putting nothing on, just an NA Ultra, does not keep the wound clean.

(N1, Tissue Viability Nurse)

A further perceived limitation of larval therapy was that certain wounds, or certain types of patient, were deemed not suitable for treatment. Wounds with a thick eschar, or where the bottom of the wound bed could not be viewed clearly, and patients with vascular impairment, or those taking warfarin, were viewed as unsuitable for larval therapy:

I wouldn’t use it if I couldn’t see the bottom of the wound bed…if they’ve got a cavity…where that cavity was tracking…I would make sure that I would know where the base of the wound was, so you know what is going in and what is coming out…or is it a big thick eschar that you wouldn’t use maggots on because they can’t chew away at that…

(NP1)

if you put them on a leg that’s got vascular problems, they’re only going to reslough, so there is no point.

(N9)

I checked their drugs and they were on warfarin, and it’s a contraindication you see…

(N9)

Amongst the nurses, opinion was divided as to whether debridement with larval therapy promoted the healing of leg ulcers:

I don’t know whether they heal faster, I have no idea about that.

(N11)

they have, in my experience, enhanced the wound-care healing rates of wounds once they’ve been cleaned.

(N10)

they did a good job, but not completely, it was only partial, so I think expectation was a bit too high then, and the patient was pleased, but a little bit disappointed because they hoped that everything would be clean and ready for the healing process to take place.

(N13)

For a majority of nurses, larval therapy was considered a treatment to try when all else had failed, as a last resort:

traditionally, it has always been a last resort…if we’ve come up against a brick wall, then we try the larval therapy…

(N13)

A number of nurses appeared to work within a personal ‘hierarchy’ of tried and tested approaches to ulcer care, and larval therapy seemed to sit between first-line treatments, such as dressings and gels, and surgical debridement:

I would go for the simple choice first, and then move on to that [larval therapy], I wouldn’t use it as the first choice to debride, but I would definitely put it on a par instead of allowing the surgeon to come and play with a scalpel.

(N19)

Cost-effectiveness of larval therapy emerged as an important consideration in some nurses’ thinking about the value of the treatment. The current lack of evidence regarding its effectiveness meant that nurses were unsure whether or not it represented a cost-effective option:

the only reason I hesitate is the cost. If you’re using maggots because you believe that debriding a wound faster makes it heal faster, that’s one economic argument. If you’re using them as a palliative measure to reduce smell, that is a different argument. I find it much easier to justify the palliative one because I know they work, I don’t even need a trial, I know they work…

(N11)

sixty odd pounds per application, but then some dressings are very dear, desloughing can take so long…in the long run I think it is a matter of writing down how much you spent on using say a gel or some other method and gave it to a GP…it’s taken X weeks at that cost…and compare it, I am sure that in the long run it is cost-effective.

(N17)

Concerns were raised by two senior, experienced nurses, working in the community, that larval therapy was currently being overused, as the latest ‘fad’ in wound care:

we seem to have maggots fans [laugh]…there are some nurses will always leap on the latest idea…I have nurses who will want to use maggots on nearly everything, and I’ve got doubts about that frankly…if there is a bit of slough, and we must debride it, I’m saying, why don’t you put them into compression and see what happens, because in my experience that usually gives a satisfactory result…

(N11)

I have concerns that there’s people thinking ‘I’ll put maggots on’, that haven’t got enough knowledge or have had any training.

(N9)

Nurses’ perceptions of patients’ willingness to try larval therapy

In general, nurses believed that patients were open to the idea of trying larval therapy, as long as they could be reassured that the larvae could not escape from the dressing:

I am quite amazed at how accepting patients are of them.

(N17)

most of them are fine about maggots as long as they feel confident that they are not going to get out.

(N11)

Some patients were perceived as particularly enthusiastic about trying larval therapy as an innovative form of treatment, and likely to be disappointed if their request to try ‘maggots’ was not met:

I think they are keen to try things that are so called ‘new’ because they are cutting edge…they feel special when they are having them.

(NP2)

patients might approach you and say, ‘Can I have maggots?’ and then you have to say to them your ulcer does not meet the requirements for it, and they say, ‘what do you mean?’ and you say, ‘well if you put the maggots on they will die, there will be nothing for them to suck’, and they will be right disappointed, you know. [patients, not maggots]

(N4, clinic nurse)

Of the patients who appeared initially reluctant to try larvae, nurses believed that some might be open to persuasion, but not all:

some of them ‘Oh no, I couldn’t possibly have that,’ but you know, you talk to some of them and they could change their mind, and they’ll have them, but others are adamant…

(N3)

One group of patients considered highly likely to try larval therapy were those described as ‘willing to try anything’ because of their desperation to try any form of treatment that might offer hope of improvement for their ulcer, even one as ‘extreme’ as larval therapy. Nurses believed that these patients could, or would, overcome any feelings of fear or distaste in their desire to find a cure:

a lot of our patients have got to the point where they will try anything because they are absolutely fed up of having a chronic wound because it has such a great impact on their life.

(N6)

I think by the time they get to the stage of needing some fairly extreme therapy like that, they just don’t care, they just want to try anything and they will agree to it.

(N3)

ones that have had them have come to the end of the road…it’s sort of like desperation and so I think they would try anything, even if they had some sort of fear about them.

(N9)

Patients considered extremely unlikely to accept larval therapy were a small minority characterised as exhibiting strong feelings of fear or revulsion. Nurses attributed these feelings to: a lack of knowledge about larvae and how they function; associations of larvae with fishing maggots, dirt and death; worries about larvae emerging from the wound as full-blown flies; anxiety about experiencing an unpleasant ‘wiggling’ sensation; and the notion of ‘maggots’ being able to escape from the dressing to crawl over their leg, burrow into their flesh, or escape into their bedding.

you ask them why they are squeamish, and they say, can you imagine putting those things on to me, nibbling, and then you say, well, they don’t nibble, they suck…because I think they think it is something that is going to do more damage than good, because they don’t understand how they work.

(N4)

I think it is before they see them, because I think they think they are going to be big, like fishing maggots…

(N7)

occasionally, people say to me, they won’t turn into flies will they?

(N11)

what the patients worry about is, can they escape, can they get out?

(N18)

what they worry about is them escaping, or that they are going to eat into the skin and track through.

(N14)

others are just adamant, ‘Oh, the thought of it’…flies and maggots and things they think they are dirty although they are specially bred for it.’

(N12)

The fear identified by nurses as lying uppermost in patients’ minds, that larvae might escape, seemed to lead to an assumption that patients would prefer to have bagged rather than loose larvae:

they get it into their head that they are going to escape, so it constantly plays on their mind, whereas I think if you put bagged maggots on they would feel a lot safer…

(N5)

Referring to a patient recruited into VenUS II, N18 commented:

luckily, she got teabag maggots…I am sure we would have had a lot more discussion if she had got the loose ones, which she really didn’t want…it was the thought of them escaping, which they shouldn’t have done, because I would have taped them in well…

(N18)

Other factors which nurses believed might affect the likelihood of patients accepting to try larval therapy were the views of partners and other family members, and the approach nurses take when they initially raise the possibility of using larval therapy. Data relating to these factors are summarised in Boxes 4 and 5.

Acceptability of larval therapy

Overall, nurse participants expressed the general view that larval therapy was a beneficial form of treatment (in terms of debridement) for the treatment of leg ulcers, and one that was acceptable to the majority of nurses and patients.

Nurse participants classified themselves and nurses that they knew into three groups with regard to acceptability of larval therapy: ‘real converts’ (NP2) who thought ‘maggots are marvellous’ (N4); nurses who attempted to overcome their own feelings of squeamishness to benefit patients (see N17 below); and a minority of nurses who would not ‘entertain the idea’ of using larval therapy under any circumstances (N14).

At the point in time when they were interviewed, 19 of the 22 study participants had had experience of applying larval therapy, either in a clinic or community setting. Two participants (N8 and N16) had chosen not to apply larval therapy, because of their dislike of ‘creepy-crawlies’. A further nurse (N12) said that the relatively ‘junior’ position that she occupied on the community nursing team meant that she was not required to apply larval therapy as ‘the Sisters do them…’.

During interview, nurses discussed acceptability of larval therapy with reference to:

• feelings of squeamishness, distaste or disgust

• perceived potential side effects of larval therapy: pain and bleeding

• practical issues.

Ulcer healing

We did not find any evidence that a phase of treatment with either loose or bagged larvae reduced the time to leg ulcer healing compared with hydrogel. Loose and bagged larvae showed very similar results for healing and were considered as one group in the main analysis. The hazard ratio for healing for leg ulcers treated with larval therapy relative to hydrogel was 1.13 (95% CI 0.76 to 1.68) indicating that whilst larvae-treated participants were 13% more likely to heal, this difference was not statistically significant (p = 0.54). The CI around the hazard ratio means that the true value may range between a 68% increase in the chance of healing with larvae and a 24% reduction in the chance of healing with larvae.

The median healing times (236 days for the participants receiving larvae and 245 days for the hydrogel group) were longer than in our previous trial where the median healing time with 4LB was 92 days and 126 days with SSB. 12 The most likely explanation for the increased healing time in VenUS II is that inclusion was restricted to sloughy and necrotic leg ulcers and those associated with more arterial disease than in the previous study.

As previously,12,85,86 we found that baseline ulcer area and ulcer duration were statistically significant predictors of time to healing (p < 0.0001) with larger ulcers and those of a longer duration taking longer to heal.

Ulcer debridement

We found good evidence that larval therapy does debride leg ulcers more quickly than hydrogel. The hazard ratio for debridement of 2.56 (95% CI 1.76 to 3.71) indicates that the ulcers of participants receiving loose larvae were more than twice as likely to debride at any time during the trial than those of participants receiving hydrogel. The hazard ratio for bagged larvae relative to hydrogel was 2.05 (95% CI 1.39 to 3.03), again suggesting that this treatment was a more effective debriding agent than hydrogel. Whereas a previous RCT57 and a number of non-RCT studies53,54,56,134 have concluded that larval therapy is an effective debriding agent; these are the first data from a large, robust RCT to confirm these findings.

Although the median time to debridement was longer in the bagged larvae group than the loose larvae group (28 days versus 14 days) this difference was not statistically significant in the adjusted analysis. Hence, although there was a trend for loose larvae to debride more quickly than bagged larvae, this difference may have occurred by chance and our trial was not powered to detect a difference in debridement between the two larvae formulations, only a difference in time to healing between larvae and hydrogel.

In this trial there was a delay of approximately 5 days between randomisation and first application of larval therapy, because larvae had to be ordered from either Wales or Germany, and delivered for use (as is the case in normal clinical practice). After this initial delay, debridement in the larval therapy groups was rapid up to day 20. At day 20 approximately 10% of those treated with hydrogel were debrided compared with approximately 58% of people receiving larval therapy. After day 20 the rate of debridement seemed to slow in the larval therapy group and then plateau at approximately day 80, whereas in the hydrogel group debridement occurred at a steady rate between days 0 and 80. Although larval therapy was observed to be an effective debriding agent, data suggest that not all ulcers receiving this treatment will debride (20% of reference ulcers did not debride in the loose larvae group, 23% did not debride in the bagged larvae group). Reasons for this could include participant intolerance of larval therapy because of pain. Ulcer-related pain scores at the removal of the first debridement treatment were significantly higher for both loose and bagged larvae compared with hydrogel.

This trial did not investigate debridement as a longer term outcome so we do not know how many ulcers that did debride remained debrided. We do know that a number of participants were withdrawn from trial treatment in the larval therapy and hydrogel arms (trial treatment being debridement treatment in Phase 1 and knitted viscose dressings plus bandage in Phase 2) because of increased slough, suggesting that some ulcers resloughed after initial debridement. This is also supported by data from the qualitative interviews.

Bacterial load

Larval therapy has been proposed as an antimicrobial treatment for chronic wounds. Suggested mechanisms for its action are via secretions and digestive processes,37–40,135 and via debridement of bacteria-rich tissue which is thought to be important in controlling infection. 35,37,38

Data from VenUS II did not find evidence of an impact of larval therapy on the bacterial load of leg ulcers. Although bacterial load decreased over the duration of the trial – this decrease was not related to treatment group. Our aim was to investigate the antimicrobial action of larval therapy, so we have not yet conducted analyses to investigate the association between bacterial load and ulcer healing. These analyses will be reported separately.

Most previous research investigating the antimicrobial activity of larvae has been in vitro, using whole body extraction and isolated secretions. 37,38,44 Whereas this work is important, in VenUS II we assessed the impact of larvae on leg ulcers in vivo where the environment is very different to that of the Petri dish. Previously, Steenvoorde and Jukema81 investigated the antimicrobial action of bagged larvae in 16 patients, the majority with osteomyelitis, fasciitis necroticans or gangrene. A varying number of wound swabs were taken from patients one month before larval treatment, during the treatment and after the larval treatment. The swabs were cultured to assess the growth of Gram-positive or Gram-negative bacteria. Bacterial load was not quantified further. The study concluded that larvae may have an impact on Gram positive bacteria but have less effect on Gram-negative bacteria. However, as this study had no comparison control group, attributing microbiological changes to larvae is difficult. Additionally, because a number of swabs were taken from each individual during the study, special analysis is required to account for correlation within participants.

VenUS II is the first RCT we have identified to investigate and publish data on the antimicrobial action of larval therapy but we also recognise the limitations of the methodology employed. We only investigated an association between larval therapy and total bacterial load. Beyond identification of MRSA we did not have the resources to conduct qualitative investigation of bacterial wound flora so cannot draw any conclusions about the impact of larval therapy on other species.

We assessed bacterial load using quantitative surface swabs, where the sample was taken from viable tissue at the wound surface with light pressure to extract wound fluid. Nurses were instructed to use a standardised protocol for sample collection (Appendix 7). We did not collect wound biopsies, which are commonly cited as the gold standard for measurement of bacterial load because they remove deeper tissue for analyses. 68,136 However, the collection of wound biopsies requires specialist skills and is more painful for the patient: their use could only be justified if they offered significantly better bacterial load data than swabs and we found this difficult to justify based on existing data. Studies have compared swab techniques with wound biopsies in terms of quantifying bacterial load, however the focus has typically been on identifying wound infection (defined as at least 1 × 106 CFUs/g tissue). 62,63 Bill et al. 136 took surface swabs using Levine’s technique – swabbing a small area of the wound free from slough/necrotic tissue for five seconds with enough pressure applied to obtain fluid from the wound tissue – and wound biopsies from 38 patients with diabetic foot ulcers, venous leg ulcers and arterial leg ulcers. Wound biopsies identified 74% of wounds as having a bacterial load > 105 organisms/g tissue, compared with 58% of swabs. Sensitivity was reported as 79% and specificity 60%. Gardener et al. 137 compared the measurement of bacterial load in non-arterial chronic wounds using (1) swabbing wound exudate from the wound surface, (2) zigzagging a rotating swab over the whole wound area, and (3) Levine’s technique (the swabbing approach used in VenUS II was a hybrid of the Levine and the zigzagging approaches). The diagnostic value of each swabbing technique was assessed by comparing values with the result of a tissue biopsy (reference standard) where ulcers were classed as being either infected (≥ 1 × 106) or non-infected (1 × 106). All samples were cultured. The Levine technique performed best – at 90% sensitivity, specificity was 57% (critical threshold 3.7 × 104 organisms per swab) and the authors conclude that the Levine swabbing technique is an acceptable alternative to tissue biopsy in identifying clinical infection. Davies et al. 68 investigated the microbiology of venous leg ulcers again by culturing quantitative surface swabs and biopsies taken from the same 66 patients. The authors report an association between increased bacterial load and delayed healing whether bacterial load was measured by swab or biopsy, with no difference in prognostic value.

Existing data suggest therefore that swabs are a viable alternative to biopsies in clinical practice when diagnosing infection. Unfortunately, there are no published data reporting the agreement of bacterial load estimates from tissue biopsies and swabs, for example using the Bland–Altman method. 138 It is possible that swabbing could have underestimated the bacterial load assessment in VenUS II, although there should be no systematic differences between the trial arms. Arguably, if larval therapy did impact on bacterial load we would have expected to see this with the use of swabs.

A more serious limitation which offers an alternative explanation as to why no difference in bacterial load was observed between trial groups is that, unlike most studies investigating bacterial load and chronic wounds, the samples in VenUS II were analysed using molecular techniques. 139 These measured the presence of bacterial DNA in samples, rather than culturing samples or employing fluid microscopy or fluorescent live–dead staining (the cost of and practicalities of using these techniques were prohibitive in this study, which collected thousands of samples from mainly community-dwelling people across the UK). Molecular techniques have the advantage that bacterial load can be quantified from a single sample, though they do not differentiate between live and dead organisms. For this reason, the measure of bacterial load may not have been sensitive to treatment-related changes if bacterial DNA from dead organisms was detected. Despite this limitation, however, if larval therapy has a clinically important impact on bacterial load we would have expected to see an impact of this over time using the molecular techniques and we did not.

MRSA

To determine the approximate prevalence of MRSA in venous and mixed venous/arterial leg ulcers, a pilot study was conducted on the first 75 baseline swabs collected for VenUS II. MRSA was present in 16% of participants. We obtained funds to analyse all swabs at baseline and to follow-up all MRSA-positive participants. However, data for our pilot were obtained using molecular techniques that detected S. aureus and the gene for methicillin resistance in two separate tests (the duplex assay) so the methicillin-resistance gene could be detected in the absence of S. aureus. This method potentially overestimated the prevalence of MRSA. Following the pilot we employed a new method that detected methicillin resistance in S. aureus only (the single primer set or SPS assay; all previous sample were reanalysed). Using this new, more accurate approach to measure MRSA for all baseline samples, only 6.7% of participants had MRSA present in their leg ulcer at baseline. We are not aware of any other figures for community leg ulcer patients in the UK. This figure is lower than previously reported figures for a UK diabetic foot clinic (13%)79 (a study in which MRSA was assessed using MRSA-specific cultures). Other prevalence estimates of MRSA in ulcer patients are 31% for diabetic foot ulcers in patients admitted to a specialised diabetic foot unit in France80 and 30% in a retrospective analysis of the medical records of patients with chronic ulcers admitted to secondary care;140 in both cases it is unclear how MRSA was assessed.

Concern has been expressed over an apparent increase in the presence of MRSA in the community, and it has been suggested that larval therapy may remove localised MRSA from chronic wounds. We assessed the impact of larval therapy on MRSA after the debridement phase. Of the 18 participants with MRSA at baseline, 12 (66.7%) were MRSA-free following debridement; however, the small numbers involved mean that no conclusions can be made regarding difference between the treatment arms. One previous study conducted in vivo also investigated the impact of loose larvae on the presence of MRSA in 13 patients with diabetic foot ulcers where MRSA was detected using MRSA-specific culturing techniques. 82 The study reports that MRSA was eliminated in 12 of the 13 wounds (92%) treated with loose larvae. However, the conclusions drawn from this study are limited because there was no control group. We therefore do not know what would have happened to similar wounds over the same period had they received standard care. The VenUS II data highlight the importance of a comparison group since MRSA was eradicated in nine of 12 cases (75%) in the combined larvae group. However, MRSA was also eradicated in three of six (50%) of the hydrogel group.

Health-related quality of life

Changes in HRQoL from baseline were investigated using the SF-12. The poor physical health of people with leg ulcers is, as in other studies,12 emphasised by the low PCS. There was little change in the PCS score during the trial in both larvae and hydrogel groups. The MCS scores were higher than the PCS but again there was no evidence of a statistically significant change over time. We cannot conclude that larval therapy has any impact on participant HRQoL even though it increases the likelihood of debridement. The HRQoL measures in this study were generic rather than disease specific and so arguably may have missed changes in ulcer-specific dimensions. However, our previous work demonstrated that the SF-12 and EQ-5D are sensitive to, and thus able to measure, change in venous leg ulcer patients. 96

Adverse events

Overall there were no significant differences in the numbers and types of adverse events between trial treatments. Interestingly, however, more of the adverse events reported for patients receiving larval therapy were classified by the treating nurse as possibly, probably or definitely related to the treatment (though the actual numbers were small). It must be emphasised that this was an open trial and therefore nurses were probably more likely to attribute adverse events (such as ulcer deterioration) to a treatment with which they were relatively unfamiliar.

Analysis of ulcer-related pain scores suggests that participants had more pain when treated with larvae compared with hydrogel. This is the first report of pain associated with larval therapy in a large number of leg ulcer patients, with a control group for comparison. Previous reports regarding the association between pain and larval therapy have come from small qualitative studies suggesting that some patients have a reduction in pain when treated with larval therapy83 while some have increased pain. 54 As with the reporting of adverse events because participants were not blinded to treatment there is the possibility of some bias against the larvae. However, because this is a pragmatic trial and if larval therapy causes a perception of increased pain this remains an important finding. These data combined with data from the qualitative interviews (discussed below) suggest that there are some participants for whom larval therapy may lead to a substantial increase in pain which they attribute to the treatment. Reasons suggested in the literature as to why larval therapy may increase ulcer pain include a change in wound pH and participants with neuropathy having increased sensitivity to the movement of the larvae or to chemicals secreted by the larvae. 41,42

Ulcer deterioration was defined as either an increase in ulcer area, malodour, apparent allergy and ulcer bleeding. Some ulcer deterioration in the larval therapy arms was reported. It is not clear if this was related to ulcer bleeding, which has been discussed as a side effect of larval therapy. Deterioration in the hydrogel group may have been related to a lack of response to the treatment and subsequent withdrawal from trial treatment.

Acceptability

People with leg ulcers who had not received larval therapy were interviewed to assess their views about the acceptability of larval therapy with most stating that they would accept larval therapy as a treatment. Most of the participants had strong beliefs about the effectiveness of the treatment that stemmed from the media or anecdotal report. What was perhaps surprising is how generally acceptable larval therapy is to the relevant patient population; there were only two cases where interviewees would not consider larval therapy under any circumstance. In one case where a participant found the idea of larval therapy unacceptable the news that a bagged larvae formulation was available made larvae more acceptable; however, overall strong preferences were not expressed for bagged larvae. In one case a participant required effectiveness information to make an informed decision about whether to have loose of bagged larvae.

The findings from these patient interviews agree with previous work in which 35 UK patients with leg ulcers were asked about their preference for different types of larval therapy and their thoughts regarding the treatment. 142 In this study, larval therapy was acceptable to a majority (77%) of interviewees. As with data from our study, patients expressed a strong desire to heal their ulcers – expressed as a willingness to try different treatments. In this previous study 23% (8/35) of participants would not consider larvae to treat their leg ulcer – of these eight, seven were women. 142 In VenUS II, two participants who stated initially that they would not accept larvae were also women.

The nurse data mirrored the participant data. Most nurses were happy to use larval therapy, especially after seeing positive debridement results in practice. Some nurses had previously successfully overcome some squeamishness. However, there were a small number of nurses who would not consider the use of larval therapy at all because of the nature of the therapy. Although there have been previous studies on the acceptability of larval therapy to participants, this is the first that highlights how important it is that nurses are happy to deliver the treatment.

The role of the nurse seems important in reassuring the patient about the treatment, especially explaining how the larvae were secured into place so that they could not escape. The importance of the treating health professional in helping to inform patients to make a decision about having larval therapy have also been highlighted elsewhere. 83,142 The nurses interviewed for VenUS II recognised that there are a number of leg ulcer patients who would try anything in the treatment of their ulcer, whereas there are others who would never consider larval therapy. Unlike patient participants, nurses appeared to have a preference for bagged larvae because they were viewed as easier and quicker to apply. The nurses believed that the bagged larvae would be more acceptable to participants, but this was not borne out in the participant interviews. It remains possible that some nurse antipathy to larval therapy partly explains our poor recruitment to the trial.

It is interesting to note that several of the nurses interviewed anticipated the results of VenUS II in that their clinical experience was of larvae leading to rapid debridement, although they did note that this debridement was often only temporary. There was less certainty expressed about the impact of larval therapy on healing.

Patient experience

Of the five interviewees who had received larval therapy two reported associated pain, which was severe in one case. Nurses also recognised that larval therapy may be associated with pain, although the nature and severity of this pain may have been underestimated. When nurses had treated patients who had reported extreme pain resulting in early termination of larval therapy, the nurses outlined potential underlying reasons for the pain, including arterial disease.

Interestingly, participants who had received larval therapy and nurses who had used it seemed to recognise that the treatment cleaned wounds – there was little discussion about healing, apart from the uncertainty about the impact of larval therapy on healing by nurses. The anecdotal reports of both the patients and the nurses suggest that debrided wounds are liable to reslough after the removal of larval therapy.

Study size

Recruitment of sufficient numbers of eligible patients was a huge challenge for this trial and we did not reach our initial sample size despite an extension in time and funding. The reasons for this are probably complex. Anecdotally, nurses reported that there were fewer leg ulcer patients ‘on their books’ than 5 years previously, attributing this to increased use of compression bandaging. Second, fewer ulcers than we originally anticipated were ‘sloughy’. Indeed the main single reason for participant exclusion was ulcers not containing sufficient slough to be eligible. Because this is a prerequisite for using larval therapy, our experience in VenUS II suggests that performing a larger trial in the UK would be challenging. Information on these recruitment issues before the conduct of this full study would have been invaluable. This highlights the value of conducting pilot trials where feasibility data can be used to assess recruitment rates and issues and plan in advance for these when designing a full trial.

Blinded outcome assessment

Like most trials of wound treatment in a community setting, we were unable to conduct in person, blinded outcome assessment for reasons of logistics and resourcing. Although blinded outcome assessment is fairly easy to achieve in a clinic or hospital setting, most of our participants were seen in their own homes. To conduct blinded outcome assessment for people treated at home we would need blinded assessors to lurk outside patients’ homes while the treating nurse took away all evidence of current treatment, cleaned up the ulcer and then called the assessor in. Given the dispersed nature of leg ulcer patients through the community and the many nurses involved this was not logistically possible and would have doubled the cost of the trial. We are satisfied that our approach of remote, blinded assessment of serial photographs is robust and we would emphasise the importance of clinically active nurses making the assessments because ultimately whether a wound is regarded as clean, or healed, is a clinical judgement.

Although we did not find a difference in our primary outcome whether we used blinded or unblinded healing data, other wound trials have observed such as difference. 148 In a study of a dressing to treat exuding wounds the statistically significant effect in favour of the intervention seen using non-blinded data disappeared when blinded outcome assessment was used. 147 It is important to note that this study’s outcome was wound improvement rather than healing, which is arguably a more objective outcome.

The use of blinded outcome assessment means that we have more confidence in our primary outcome but it is important to note that undertaking such assessment is not easy. In some cases we did not receive photographs (e.g. where a camera was stolen or camera data cards were lost) and we had to fall back on unblinded data. Using real-time assessment of photographs would also help to ensure as complete a data set as possible was available for blinded assessment. In future trials we would investigate the e-mailing or texting of images to allow closer, contemporaneous monitoring of images. During our pilot phase of the blinded outcome assessment process we noted that the clinical definition of healing that nurses had used was slightly different from the definition that was in the protocol. Nurses were classifying ulcers as healed in the presence of small amounts of dry skin and eschar, and this did not agree with our protocol. Had we not undertaken the blinded outcome assessment we would not have detected this discrepancy. It is unclear if the interpretation of healing is an issue for other studies. If we had followed the protocol definition we may have underestimated healing and compromised the pragmatic nature of the trial. We consulted with senior tissue viability nurses who agreed with the trial nurses in that an ulcer could be classed as healed in the presence of small amounts of eschar as long as a dressing was not required. As a consequence we changed our definition for the assessment process. We did not change our protocol because the assessment occurred after all data had been collected. This change represents a limitation of the study. However, we do not anticipate that this change will have influenced our results.

Attrition

As with VenUS I, we observed a marked reduction in the patient questionnaire response rate over time. At 12 months from baseline almost 50% of HRQoL data were missing. This was in spite of our efforts to ensure the return of questionnaires (reminders and a financial incentive). Preliminary analysis suggests that those participants who healed were more likely to return data so we may have overestimated HRQoL scores. It is important to highlight that, overall, there was no evidence of differential attrition within trial group.

Statement of independence of researchers

The larvae manufacturers had no role in the design of VenUS II, nor in the collection, analysis, and interpretation of data.

Collaborations and contributions of the authors

The VenUS II collaborators (current and past) are: Una Adderley, Jacqui Ashton, Gill Bennett, JMB, Anne Marie Brown, Sue Collins, Ben Cross, NC, Val Douglas, CD, JCD, Andrea Ellis, Caroline Graham, Christine Hodgson, Gemma Hancock, Shervanthi Homer-Vanniasinkam, CI, June Jones, Nicky Kimpton, Dorothy McCaughan, Elizabeth McGinnis, Jeremy Miles, JLM, Veronica Morton, EAN, Sue O’Meara, Angie Oswald, Emily Petherick, Ann Potter, Pauline Raynor, Linda Russell, Jane Stevens, MS, Nikki Stubbs, DJT, Kath Vowden, Peter Vowden, Michael Walker, Shernaz Walton, Val Wadsworth, Margaret Wallace Judith Watson, Anne Witherow, and GW.

JD and PR were trial managers (PR between 2003 and 2005; JD 2005 and 2008). GW and MB designed and conducted the clinical analysis. MS and CI designed and conducted the economic analyses. NAC was the chief investigator, led the design of the trial, chaired the Trial Management Group, edited and approved the final draft of the report. CD advised on the collection and analysis of microbiological data. DMM conducted the qualitative work and analysis. JM performed the microbiological analysis. EAN and DJT contributed to the study design and co-ordination.

TSC members

Dr Su Mason: Principal Research Fellow (Independent Chair), Clinical Trials Research Unit, University of Leeds.

Professor Francine Cheater: Professor of Public Health Nursing, School of Healthcare, University of Leeds.

Professor Mike Campbell: Professor in Medical Statistics, School of Health and Related Research, University of Sheffield.

Consumer representatives

Mrs Joyce Golton

Mr Victor Beeston (deceased)

DMEC members

Professor Keith Abrams: Professor of Medical Statistics, Department of Health Sciences, University of Leicester (Chair).

Dr Michelle Briggs: Senior Research Fellow, School of Healthcare, University of Leeds.

Mr Alun Davies: Consultant General and Vascular Surgeon, Imperial College, London.

Publications

Dumville JC, Watson J, Raynor P, Torgerson, DJ. Research governance: a barrier to ethical research? QJM 2004;97:113–14. DOI: 10.1093/qjmed/hch026.

Raynor P, Dumville J, Cullum, N. A new clinical trial of the effect of larval therapy. J Tissue Viability 2004;14:104–5.

Petherick ES, O’Meara S, Nelson EA, Spilsbury K, Iglesias C, Torgerson, D. Patient acceptability of larval therapy – implications for the sample size calculation. BMC Med Res Methodol 2006;1:43.

Spilsbury K, Cullum N, Dumville J, O’Meara, S; Petherick, E; Thompson, C. Exploring patient perceptions of larval therapy as a potential treatment for venous leg ulceration. Health Expect 2008;11:148–59.

Dumville J, Worthy G, Bland JM, Cullum N, Dowson C, Iglesias C, et al. Larval therapy for leg ulcers (VenUS II): randomised controlled trial. BMJ 2009;338:b773.

Soares M, Iglesias C, Bland M, Cullum N, Dumville J, Nelson EA, et al. Cost effectiveness analysis of larval therapy for leg ulcers. BMJ 2009;338:b825.

Please read this document carefully.

We would like to invite you to take part in a research study. Before you decide whether to take part it is important for you to understand why the research is being done and what it will involve. Please take time to read the following information carefully. Feel free to discuss this with anyone else you wish to, for example, friend/nurse/doctor or relative. Ask us if there is anything that is not clear. We are happy to provide more information. Take as much time as you need to decide whether you want to take part.

Thank you for reading this.

What is the purpose of this study?

This is a study of larval therapy for the treatment of leg ulcers. Larval therapy involves the application of small, sterile maggots to a skin ulcer or wound. It is thought that larval therapy helps with healing but we need to find out if this really is the case.

Leg ulcers are common and can be very distressing for patients. Leg ulcers may contain dead skin cells and this dead tissue (called slough) is thought to delay healing. The removal of the dead tissue is undertaken in an attempt to speed up the healing process. Several different treatments can be used to clean a leg ulcer including different wound dressings and larval therapy. The use of hydrogel wound dressings is very common and simply involves applying a watery jelly to the ulcer. This treatment can work quite slowly, and it is thought that larval therapy may be a quicker way to clean up the ulcer and to help healing.

What is the treatment being studied?

We are looking at larval therapy (also called maggots). Larval therapy involves the use of larvae (sterile maggots). The larvae may be put onto the wound in a sealed bag or placed directly onto the wound and kept in place with a sealed dressing system. The main dressings containing the larvae are left in place for up to 5 days. The outer padding is changed as often as required. Most people only need one or two applications of larvae to clean up their ulcer (over 1 or 2 weeks), and can then have regular wound dressings until their ulcer heals.

Why have I been chosen?

Your nurse and/or doctor think that you might benefit from treatment to remove dead tissue from your leg ulcer. Across the UK about 300 people with leg ulcers will be asked to take part in this study.

Do I have to take part?

Participation in this study is entirely voluntary. It is up to you to decide whether or not to take part. If you do decide to take part you will be given this information sheet to keep and be asked to sign a consent form. If you do decide to take part you can still change your mind at any time. Your future care and treatment will not be influenced by your decision to take part or not. If you do agree to take part in this study and decide at a later time to withdraw then you are free to do so at any time without influencing your future care or treatment.

What will happen to me if I agree to take part?

We are interested in how quickly the ulcers heal, and also in your opinion about the treatment you receive. In order to compare the treatments we need to treat approximately 100 patients with the larval therapy in bags, 100 patients with loose larval therapy and another 100 with hydrogel wound dressings. If you agree to take part in this study you will be allocated to one of these three treatments. The decision regarding which treatment you will receive will be made after you agree to take part. The choice of treatment will be determined at random, that is we cannot predict which treatment you will receive. You will have an equal chance of receiving each treatment, in the same way that tossing a coin gives an equal chance of getting ‘heads’ or ‘tails’. Two people out of every three in the trial will receive larval therapy.

What do I have to do?

You will continue seeing your community nurse or clinic staff for the leg ulcer dressings to be carried out. The ulcer will be traced and photographed at the start of the study and then regularly to see if the ulcer is reducing in size. The study will last for 12 months. You will be asked to complete several questionnaires, at entry into the study, and at 3, 6 and 12 months after that. We will send you questionnaires about your leg and how it affects you, even if your ulcer has completely healed. We do not anticipate that you will have to see the nurse or attend your clinic more frequently than would normally be required. For this reason we will not be able to pay any travel expenses incurred.

There are no restrictions on your activity when you are in the study. You will continue with any other medical treatment, such as taking regular medication.

Why do the study?

Larval therapy may be more effective than using hydrogel wound dressings. It is therefore important to carry out this study so that patients with leg ulcers can be provided with the most appropriate and effective care. We are also interested in how you feel while you have the leg ulcer treatment so that nurses, doctors and patients can make future decisions about which treatments are comfortable. Without this information patients may receive inefficient care, and precious NHS money may be wasted.

Are there any alternatives to the treatments being studied?

There are a number of treatments for removing dead tissue from leg ulcers. Wound dressings to keep the ulcer moist can be used. The body can remove dead tissue beneath moist dressings. Less commonly, the nurse or doctor can sometimes remove dead tissue using forceps.

Are there any side effects from the larval therapy?

There may be an increase in the amount of fluid coming from your wound. This is perfectly normal. The colour of the liquid will also change and will be slightly pink. There may also be a change in the smell of the wound – this is caused by the maggots and is perfectly normal. The smell is usually masked by the layers of bandages on top and will go as soon as the outer dressing is changed. It is sometimes possible to feel a different sensation in the wound after the maggots have been applied. If you wish to have the maggot therapy discontinued for any reason then you can have it removed.

Are there possible disadvantages to taking part?

We do not anticipate that being in this trial will harm you. Should this occur, however, normal NHS negligence procedures will apply.

In an emergency you should contact your community nurse or clinic nurse. The name of a contact nurse and telephone number where they can be reached is provided below.

What are the possible advantages of taking part?

We hope that your ulcer will improve with either treatment (larvae or gel). We cannot guarantee that your ulcer will improve by your being in the trial. The information we get from this study may help us to treat people with leg ulcers better.

What if new information becomes available?

Sometimes during a research project, new information becomes available. If this happens, your nurse/doctor will tell you about it. They will discuss with you whether you want to continue in the study. If you decide to withdraw from the study your care will continue without the trial treatments. If you decide to continue, then you will be asked to sign an updated consent form.

If new information means that your nurse or doctor decides to take you out of the study, then she/he will discuss this with you. He/she will explain the reasons for this and arrange for your leg ulcer care to continue without the trial treatment.

What happens when the research study stops?

Larval therapy is only available in some clinics. Gel wound dressings are available to every nurse/doctor in the UK. After the research stops the gel treatment will continue to be available throughout the UK. Larval therapy may not be available in your area once the trial stops.

What if something goes wrong?

If taking part in this research project harms you, there are no special compensation arrangements. If you are harmed due to someone’s negligence, then you may have grounds for a legal action but you may have to pay for it. Regardless of this, if you wish to complain, or have any concerns about any aspect of the way you have been approached or treated during the course of this study, the normal NHS complaints mechanisms should be available to you.

Will my taking part in this study be kept confidential?

All information that is collected about you during the course of the research will be kept strictly confidential. Any information about you that leaves the hospital/surgery will have your name and address removed so that you cannot be recognised from it.

If you consent to take part in the research, the University of York (for purposes of analysing the results) may inspect any of your medical records. People from regulatory authorities may also look at them to check that the study is being carried out correctly. Your name, however, will not be disclosed outside the hospital/GP surgery.

Your GP will be notified of your participation in the trial.

What will happen to the results of the study?

The results of the study will be published in medical and nursing journals. This is likely to happen in 2008. You will be able to obtain a copy of the results from the University of York. You will not be identified in any publication arising from this study.

Who is organising and funding the research?

The study is being funded by part of the UK NHS (the Health Technology Assessment Programme). They have provided funds to pay for the larval therapy, for research nurses, and for the tests to monitor your wound (e.g. camera, wound tracing). The research nurses will work with your regular nurse/doctor to collect all the information needed. Your nurse or doctor is not receiving any money for including you in the trial.

The study is being organised by researchers from the University of York.

Who has reviewed the study?

The West Midlands Research Ethics Committee has reviewed the study. Your Local Research Ethics Committee has also been involved in reviewing the study.

What do I do now?

If you are interested in taking part please complete the enclosed questionnaire and sign the consent form, returning it to your study nurse in the envelope provided.

Where can I get more information about the study?

If you do not understand anything on this information sheet or would like further information please contact your local research nurse on the telephone number below.

Thank you for taking the time to read this information sheet

Appendix 3.1

(PDF download)

Appendix 3.2

(PDF download)

Appendix 3.3

(PDF download)

Appendix 3.4

(PDF download)

Appendix 3.5

(PDF download)

Appendix 3.6

(PDF download)

Appendix 3.7

(PDF download)

Appendix 3.8

(PDF download)

Appendix 3.9

(PDF download)

Appendix 4.1

Appendix 4.2

Approximate number of free-range sterile larvae required

Biobags are available in the following sizes:

Taking photographs – framing the picture

FIGURE 21.

Nikon Coolpix 4600.

Hold the camera steadily in both hands about 18 inches to 2 feet (45–60 cm) away from the patient. Zoom in using the optical zoom facility on the camera (PRESS T) ensuring that the whole of the wound area and the colour reference target card are included in the picture (see page 18 in your camera guide book).

When you press the ‘T’ button a white oblong box will appear in the top of the viewfinder and a bar will move towards the line (two-thirds along the box) until it stops. Stop pressing the ‘T’ button at this stage and take your photograph. Press the ‘W’ button to zoom out.

Specimen collection

Explain to the participant that you are about to take a wound swab from their leg ulcer.

Please DO NOT clean the wound before taking the swab to ensure the maximum numbers of bacteria are present.

Take the swab from an area of the wound containing viable tissue if possible (e.g. granulating tissue).

Avoid taking the sample from areas of the wound covered with slough or necrotic tissue.

Patient information sheet

Nurse information sheet

Patient interview schedule

This schedule includes generic questions for all participants. The participant will be asked specific questions dependent on their experience of larval therapy.

Introduction to the study

Explain about confidentiality and tape recording

1. 1. Background

   1. age

   2. ethnic origin

   3. household composition

2. 2. How long have you had a leg ulcer?

3. 3. What impact has the leg ulcer had on you?

4. 4. What treatments have you had on your leg ulcer?

The following questions will be posed to those participants who HAVE NOT had experience of maggots.

1. 5. What do you think about ‘natural treatments’ for leg ulcers (such as maggots)?

2. 6. How do you feel about maggots being used in the treatment of your leg ulcer?

3. 7. Why do you think you feel this way?

4. 8. Have you had any previous experience of maggots (e.g. fishing)?

5. 9. Have you talked to any members of your family about your leg ulcer/treatment of your leg ulcer?

6. 10. How do you think they would view the use of maggots in the treatment of leg ulcers?

The following questions will be posed to those participants who HAVE had experience of maggots.

1. 5. How did you feel when larval therapy was suggested for the treatment of your leg ulcer?

2. 6. Have you had any previous experience of maggots (e.g. fishing)?

3. 7. What is the maggot treatment like (e.g. do you have any pain)?

4. 8. How do you think the maggots are helping your wound?

5. 9. Have you talked to members of your family about the treatment of your ulcer?

6. 10. How do you think they view the use of maggots in the treatment of leg ulcers?

Is there anything else you would like to tell me about your leg ulcer and its treatment?

Reiterate about confidentiality

Thank you

Nurse interview schedule

This schedule includes generic questions for all participants. The participant will be asked specific questions dependent on their experience of larval therapy.

Introduction to the study

Explain about confidentiality and tape recording

1. 1. Background

   1. age

   2. ethnic origin

   3. employment details – position/grade and where based

2. 2. How long have you been involved in caring for patients with leg ulcers?

3. 3. What treatments have you used on leg ulcers?

4. 4. Have you been involved in using larval therapy before?

The following questions will be posed to those participants who have NOT HAD experience of maggots.

1. 5. How do you feel about maggots being used in the treatment of leg ulcers?

2. 6. Why do you think you feel this way?

3. 7. If a patient were prescribed larval therapy for the treatment of their leg ulcer, would this concern you?

4. 8. Have you had any previous experience of maggots (e.g. fishing)?

5. 9. Have you talked to colleagues about the use of maggots in the treatment of leg ulcers? If so, what were their views?

The following questions will be posed to those participants who HAVE had experience of maggots.

1. 5. Tell me about your experiences of using larval therapy. Were these loose or bagged?

2. 6. Did anything concern you about using maggots?

3. 7. Did you encounter any problems in the ordering or supply of the maggots?

4. 8. What did the patients say about the use of maggots on their wounds?

5. 9. Have you had any previous experience of maggots (e.g. fishing)?

6. 10. Have you talked to colleagues about the use of maggots in the treatment of leg ulcers? If so, what were their views?

Is there anything else you would like to tell me about the use of larval therapy?

Reiterate about confidentiality and anonymity

Thank you