Advice
Evidence review
Evidence review
Clinical and technical evidence
Regulatory agencies
No reports of adverse events were identified from searches of the Medicines and Healthcare Products Regulatory Agency website, or from the US Food and Drug Administration database: Manufacturer and User Device Facility Experience (MAUDE).
Clinical evidence
There is 1 published randomised clinical trial (n=100; Moffatt et al. 2014) and 3 published case series (Davis et al. 2009, Wood et al. 2010, Lafferty et al. 2011) on the Oxyzyme and Iodozyme dressings. A publication on 5 patients taken from a multicentre study of 31 patients was excluded because the full results of the study have not been published (Ivins et al. 2007). A report on 3 patients taken from a larger case-study programme (up to 200 patients in progress at the time of publication) was excluded because of incomplete reporting (Hampton et al. 2008). Queen et al. (2007) reported on 4 patients who healed out of a series of 20 patients (22 wounds). This study was excluded because of incomplete reporting.
Some other case studies contained useful information on technical issues. In 1 case report (Kerr 2007a) it was noted at the 3-week assessment that the dressing had been incorrectly applied, with the primary dressing applied on top of the secondary dressing. In another case report (Kerr 2007b) it was noted that an occlusive secondary dressing had been used instead of an air-permeable dressing as specified in the manufacturer's instructions for use.
Randomised controlled trial
Moffatt et al. 2014: Oxyzyme and Iodozyme
This was a UK-based, single-centre trial comparing the Oxyzyme and Iodozyme dressings to standard care, in the treatment of 100 patients with venous or mixed venous/arterial leg ulcers. Standard care was defined as a continuation of the current treatment regimen based on formulary recommendations for the South Staffordshire PCT Tissue Viability Service. Objectives and outcomes for Moffatt et al. (2014) are described in tables 1 and 2.
The primary outcome was complete healing of the ulcer within 12 weeks of treatment. Analysis was performed using the Kaplan–Meier method. No statistically significant difference was seen in the proportion of patients whose ulcers healed within 12 weeks or 24 weeks, when the Oxyzyme or Iodozyme dressings were compared with standard care.
The authors report no difference in patient-reported quality of life or pain when the Oxyzyme or Iodozyme dressings were used, compared with standard care.
Cost-effectiveness was a secondary outcome and is reported in a later section.
Case series
Davis et al. 2009: Oxyzyme only
This was an observational study of 100 patients with chronic, hard-to-heal wounds of mixed aetiology recruited from 27 European complex wound clinics (22 in England). Objectives and outcomes for Davis et al. (2009) are described in tables 3 and 4.
Wood 2010: Iodozyme only
This case series reported results from 45 patients (51 wounds) in 30 different centres in England. Objectives and outcomes for Wood et al. (2010) are described in tables 5 and 6.
The primary outcome was a clinical assessment of overall wound condition. At the end of the study, clinicians assessed the overall outcome as healed, improved (which included subjective improvement in wound bed condition), static or deteriorated. Data from patients who were withdrawn from the trial were included in the results, and assessment conducted at the time of withdrawal. It was noted that patients who were withdrawn from the trial with problems relating to the wound dressing might still be reported as having an improved wound bed condition, including patients who were withdrawn due to increased wound size. No clear guidance on how assessments should be made was reported. This outcome was not reported here due to the high likelihood of bias.
Clinicians were also asked to assess the dressings compared with other dressings used for similar wounds. There were 51 responses reported (the total number of clinicians involved was not reported).
Other outcomes measured at entry and during weekly assessments included wound size and depth, condition of the wound margins, wound bed and peri‑wound skin. The mean wound area decreased 19.8% from 13.1 cm2 (median 6.0, range 0.23–98.0) to 10.5 cm2 (median 4.0, range 0.0–98.0). Depth, condition of wound bed, wound margins and peri-wound skin are not reported.
During weekly clinic visits, patients were asked to report wound pain using an non-validated numbered visual analogue scale, and overall satisfaction on a 5-point scale. Average pain scores were reported at the start (3.7) and end of the trial (2.6), but without intermediate results, ranges or standard deviations. Overall satisfaction was not fully reported.
In vitro evidence
Four papers were identified that reported on in vitro testing of the antimicrobial properties of Oxyzyme or Iodozyme dressings. Two included both Oxyzyme and Iodozyme dressings in the dressings tested, but did not report the results in full (Thorn 2005, Thorn 2006). Two papers (Greenman 2006, Thorn 2009) reported evidence of a reduction in monoculture bacteria with the Iodozyme dressing, compared with control and other dressings. The tests used 2 monocultures, Staphylococcus aureus and Pseudomonas aeruginosa. This does not equate to evidence of antimicrobial efficacy on a wound, or the effect that this might have on healing, and should be interpreted with caution.
Table 1 Summary of the Moffatt et al. (2014) study
|
Oxyzyme / Iodozyme |
Control |
Analysis |
|
|
Randomised |
n=47 |
n=53 |
– |
|
Efficacy |
n=47 |
n=53 |
– |
|
Primary outcome: Healed ulcers at 12 weeks |
44.7% (21/47) |
49.1% (26/53) |
ITT |
|
Selected secondary outcomes: |
|||
|
Healed ulcers, Kaplan Meier Analysis, 12 weeks |
48.4% |
50.2% |
95% CI unknown, HR 1.05 p=0.87 |
|
Healed ulcers, Kaplan Meier Analysis, 24 weeks |
unknown |
unknown |
95% CI 0.7 to 1.84, HR 1.14 p=0.6 |
|
HRQoLat 12 weeks: n, Mean (SD)a, b |
|||
|
Activities |
n=38, 28.8 (25.8) |
n=49, 31.3 (24.9) |
0=perfect health, 100=worst possible health |
|
Psychological |
n=39, 36.9 (26.7) |
n=49, 32.4 (27.2) |
– |
|
Symptom distress |
n=39, 36.1 (25.4) |
n=49, 32.6 (22.4) |
– |
|
HRQoLat 24 weeks: n, Mean (SD) |
|||
|
Activities |
n=39, 23.1 (23.3) |
n=45, 25.3 (27.9) |
– |
|
Psychological |
n=38, 28.3 (28.1) |
n=45, 27.0 (27.4) |
– |
|
Symptom distress |
n=37, 26.1 (27.7) |
n=46, 23.5 (26.2) |
– |
|
Safety |
n=47 |
n=53 |
– |
|
Patients reporting adverse events |
18 events in 11 patients |
8 events in 7 patients |
Details of adverse events are not given. Reported as mainly relating to pain. |
|
Reported as relating to dressing |
3 events |
0 events |
– |
|
Abbreviations: CI, confidence interval; HRQOL, health related quality of life; ITT, intention to treat; n, number of patients; a Patient reported outcome using VLU-QOL tool, where 0 is perfect health and 100 is worst possible health. b p values are reported, however it is unclear which values they refer to and therefore they have not been included in this summary. |
|||
Table 2 Summary of the RCT: Moffat et al. (2014)
|
Study component |
Description |
|
Objectives/hypotheses |
To determine the relative effectiveness of Oxyzyme/ Iodozyme compared with standard care (defined as a continuation of existing care). Effectiveness was defined as complete ulcer closure (100% re-epithelialization) at 12 weeks. |
|
Study design |
Randomised controlled trial |
|
Setting |
Single-centre trial at an NHS leg ulcer service. Patients were evaluated weekly for 12 weeks, or until their ulcer was healed if sooner, with a further follow up at 24 weeks. |
|
Inclusion/exclusion criteria |
The study included adults with a venous ulcer (ABPI>0.8) or a mixed ulcer (ABPI>0.6), whose ulcer was present for less than 1 year, and had healthy peri-wound skin. The study excluded patients with ABPI<0.6, or if unable to tolerate correct compression levels. Patients with recent acute deep vein thrombosis, surgery for venous insufficiency, or arterial reconstruction or angioplasty were also excluded. Other exclusion criteria included active cellulitis at the start of the trial, treatment for thyroid disorders,or peri-wound maceration. |
|
Primary outcomes |
Complete ulcer closure (100% re-epithelialization) at 12 weeks |
|
Statistical methods |
Kaplan–Meier analysis was used to examine the rate of healing |
|
Participants |
100 patients participated in the study, 53 were randomised to the control and 47 to the intervention. There was no power calculation. |
|
Results |
At 12 weeks, 44.7% (21/47) of ulcers treated with Oxyzyme and Iodozyme had completely healed. 49.1% (26/53) of ulcers treated using standard care had completely healed. Kaplan–Meier analysis resulted in a hazard ratio (active treatment:control) of 1.05 at 12 weeks (p=0.87) and 1.14 at 24 weeks (p=0.60). Neither result was statistically significant |
|
Conclusions |
There is no statistically significant difference in clinical outcomes between Oxyzme /Iodozyme and standard care for venous leg ulcers. |
|
Abbreviations: ABPI, ankle to brachial pressure index; CI, confidence interval; ITT, intention to treat; n, number of patients; RR, relative risk.
|
|
Table 3 Summary of the objectives of Davis et al. (2009) case series
|
Study component |
Description |
|
Objectives |
To assess the performance of the Oxyzyme dressing with a variety of hard-to-heal chronic wounds in clinical settings. To examine the effect of the Oxyzyme dressing on the wound bed. To obtain patient and carer feedback on the Oxyzyme dressing. |
|
Study design |
Case series (a report of multiple cases and small series from 27 centres) |
|
Setting |
27 European complex wound clinics (22 in England, 2 in Germany, and 1 each in Iceland, Sweden and Turkey). All patients had been receiving treatment in the clinics before recruitment. Each patient received 6 weeks' treatment with the Oxyzyme dressing. Wound area measurements were taken at baseline and at 6 weeks or at withdrawal. |
|
Inclusion/exclusion criteria |
Inclusion criteria:
Exclusion criteria:
Consecutive patients who met the inclusion criteria were recruited by investigating clinicians. There was an average of 3 patients from each clinic (range 1–12). Wound area measurements were taken from digital photographs when these were available and from clinician's area assessments otherwise. |
|
Primary outcome |
Not specified, but wound area is the only quantitative outcome reported. |
|
Statistical methods |
No statistical analysis was reported. Mean, median and range were reported, but not standard deviation. |
|
Participants |
100 patients were included in the study. There were 6 subgroups identified: arterial leg ulcer (14), diabetic foot ulcer (13), mixed aetiology leg ulcer (13) other chronic wounds (8), pressure ulcer (13) and venous leg ulcer (39). At baseline 10 ulcers measured less than 1 cm2, 49 ulcers measured 1–10 cm2, 39 measured 10–100 cm2 and 2 measured more than 100 cm2. |
|
Results |
38 patients were withdrawn before the end of the 6-week study period for the following reasons:
Mean wound area changed from 15.7 cm2 (median 5.0, range 0.2–250.0) to 10.2 cm2 (median 2.5, range 0.0–96.0) from baseline to study end (6 weeks or at withdrawal). |
|
Conclusions |
Without a control group it is impossible to draw any conclusions about how the Oxyzyme dressing compares with other dressings. |
|
Abbreviations: n, number of patients |
|
Table 4 Summary of the outcomes of Davis et al. (2009) case series
|
Oxyzyme dressing |
||||||||
|
Recruited |
n=100 patients |
|||||||
|
Completed |
n=62 patients |
|||||||
|
Primary outcome: mean wound area after 6 weeks or at withdrawal |
Wound type |
Mean baseline wound area (cm2) (median, range) |
Mean end point wound area (cm2) (median, range) |
Mean area reduction (%) over 6 weeks |
||||
|
Arterial leg ulcer n=14 |
42.8 (17.2, 1.0–250) |
23.0 (15.7, 1.7–96) |
46.3 |
|||||
|
Diabetic foot ulcer n=13 |
10.1 (7.2, 0.2–31) |
4.7 (2.1, 0.2–20) |
53.7 |
|||||
|
Mixed aetiology leg ulcer n=13 |
13.0 (9.0, 2.0–38) |
9.9 (8.0, 0–36) |
23.7 |
|||||
|
Other chronic wounds n=8 |
15.1 (6.1, 1.0–70) |
13.8 (4.4, 0–70) |
8.4 |
|||||
|
Pressure ulcer n=13 |
11.9 (9.1, 0.4–35) |
10.3 (6.5, 0.1–35) |
13.1 |
|||||
|
Venous leg ulcer n=39 |
10.2 (5.0, 0.3–50) |
6.8 (2.5, 0–34) |
33.2 |
|||||
|
Total |
15.7 (5.0, 0.2–250) |
10.2 (2.5, 0–96) |
35.0 |
|||||
|
Selected secondary outcomes: |
||||||||
|
Pain/comfort scores at final clinic visit n=100 |
Very comfortable/comfortable |
Discomfort |
Pain |
Not applicable/not rated |
||||
|
55% (55/100) |
9% (9/100) |
21% (21/100) |
15% (15/100) |
|||||
|
Safety |
n=100 patients |
No adverse events reported, not a clinical trial. |
||||||
|
Total patients withdrawn |
38% (38/100) |
Patients withdrawn from study, not all due to dressing-related issues. |
||||||
|
Infection |
13% (13/100) |
|||||||
|
Deterioration |
7% (7/100) |
Maceration or an increase in wound area |
||||||
|
Pain/ discomfort |
6% (6/100) |
|||||||
|
Other dressing-related reasons |
5% (5/100) |
Bleeding from a large non-healing surgical wound (n=1), wound static after 3–4 weeks (n=3), suspected pyoderma gangrenosum (n=1)). |
||||||
|
Non-dressing related reasons |
7% (7/100) |
e.g. hospitalisation |
||||||
|
Abbreviations: n, number of patients |
||||||||
Table 5 Summary of the objectives of Wood et al. (2010) case series
|
Study component |
Description |
|
Objectives/hypotheses |
To assess the performance of the Iodozyme dressing within normal clinical practice |
|
Study design |
Case series (a report of multiple cases and small series from 30 centres). |
|
Setting |
Recruitment over 12 months, in 30 centres in England, ranging from district nurse bases to complex wound care clinics. Patients continued to be treated in the same care setting. Assessment of the wound was made at entry to the study and at weekly clinic visits during the 6-week duration. |
|
Inclusion/exclusion criteria |
Inclusion criteria:
Exclusion criteria:
The method section stated consecutive recruitment, however the abstract reports that clinicians chose patients to be included. Many centres would have recruited very small numbers of patients, so consecutive recruitment is not meaningful. The mean number of patients per centre was 1.5, the range was not reported. Various wound aetiologies were included. |
|
Primary outcomes |
Primary outcome is stated as clinician's overall assessment of the product, but wound area is the only quantitative measurement quantitative outcome fully reported in the paper. |
|
Statistical methods |
No statistical analysis was reported. Mean, median and range were reported, but not standard deviation. |
|
Participants |
45 patients (21 men, 24 women), with 51 wounds of various aetiologies, all considered difficult to heal, and that had been static or deteriorating for 4 weeks. One patient had 2 arterial leg ulcers, 1 patient had 6 diabetic foot ulcers, the remainder had single wounds. The median wound duration prior to the study was 13 months (range 1–312). Median age was 75 years (range 27–93).The median wound area at the start of the study was 6 cm2 (range 0.23–98.0). Inclusion of very small wounds may lead to measurement uncertainties, and moderate improvements appearing as large percentage changes. |
|
Results |
The mean wound area decreased from 13.1 (median 6, range 0.23–98.0) to 10.5 (median 4, range 0–98.9) cm2. |
|
Conclusions |
A reduction in mean wound size is seen, however the number of withdrawals, potential for bias and lack of comprehensive reporting make it difficult to draw firm conclusions. |
|
Abbreviations: n, number of patients When patients were withdrawn, the outcome is included at point of withdrawal. Two patients withdrew due to increase in wound size, 1 due to reduction. |
|
Table 6 Summary of the outcomes of Wood et al. (2010) case series
|
Iodozyme dressing |
Analysis |
||||||||
|
Started treatment |
n=45 patients (51 wounds) |
43 patients, with 1 wound each. 1 patient with 2 arterial leg ulcers. 1 patient with 6 diabetic foot ulcers. |
|||||||
|
Completed treatment |
n=33 patients |
12 patients were withdrawn from the study, not all related to dressing. |
|||||||
|
Primary outcome: wound area after 6 weeks or at withdrawal a |
Wound type |
Mean baseline area (cm2) (median, range) |
Mean endpoint area (cm2) (median, range) |
Mean area reduction (%) over 6 weeks |
|||||
|
Arterial leg ulcer n=10, wounds=11 |
14.8 (11.2, 3.3–36.6) |
11.6 (9.2, 1.0–30.0) |
21.7% |
||||||
|
Diabetic foot ulcer n=6, wounds=11 |
6.1 (5.0, 0.4–28.0) |
4.7 (2.3, 0.5–20.0) |
23.3% |
||||||
|
Miscellaneous n=7, wounds=7 |
8.9 (7.5, 0.2–27.3) |
5.9 (3.9, 0.0–21.1) |
33.5% |
||||||
|
Pressure ulcer n=4, wounds=4 |
5.0 (4.5, 3.0–8.0) |
4.4 (3.5, 2.4–8.0) |
12.6% |
||||||
|
Surgical wound n=8, wounds=8 |
13.0 (6.0, 2.3–50.0) |
7.1 (3.5, 0.0–29.8) |
45.6% |
||||||
|
Venous leg ulcer n=10,wounds=10 |
25.0 (10.4, 2.1–98.0) |
24.0 (7.6, 0.0–98.0) |
4.1% |
||||||
|
Total n=45, wounds =51 |
13.1 (6.0, 0.23–98.0) |
10.5 (4.0, 0–98.9) |
19.8% |
||||||
|
Selected secondary outcomes: |
Clinician assessment of wound condition is not reported due to the likelihood of bias in assessment. Other secondary outcomes were not reported in sufficient detail to tabulate. |
||||||||
|
Clinician assessment of productb n=51 clinicians |
Much better |
Better |
Similar |
Worse |
Much worse |
||||
|
8% (4/51) |
69%(35/51) |
18% (9/51) |
6% (3/51) |
0% (0/51) |
|||||
|
Safety |
n=45 patients |
No adverse events reported, not a clinical trial |
|||||||
|
Total patients withdrawn |
27% (12/45) |
Patients withdrawn from study, not all due to dressing-related issues |
|||||||
|
Bleeding |
2.2% (1/45) |
Withdrawn after 1 week |
|||||||
|
Hospitalisation |
4.4% (2/45) |
Withdrawn after 1 week and 3 weeks, not related to dressing |
|||||||
|
Slippage |
2.2% (1/45) |
Patient with sacral pressure ulcer withdrawn after 1 week due to dressing slippage and faecal contamination. |
|||||||
|
Pain |
4.4% (2/45) |
Withdrawn after 1 week and 4 weeks |
|||||||
|
Clinical decision |
2.2% (1/45) |
Withdrawn after 5 weeks, change to the Oxyzyme dressing following reduction in wound size |
|||||||
|
Clinical decision |
2.2% (1/45) |
Withdrawn after 5 weeks when wound became static (despite initial improvement) |
|||||||
|
Maceration |
2.2% (1/45) |
Withdrawn after 1 week |
|||||||
|
Increased wound size |
4.4% (2/45) |
Both withdrawn after 3 weeks due to increase in wound area |
|||||||
|
Inflammation |
2.2% (1/45) |
Withdrawn after 1 week due to inflammation around wound |
|||||||
|
Abbreviations: n, number of patients a Primary outcome is stated as clinician's overall assessment of the product, however wound area was the only quantitative measurement technique reported. b Clinicians were asked for their overall rating of Iodozyme, compared with other treatments used for similar wound types. When patients were withdrawn, the outcome is included at point of withdrawal. Two patients withdrew due to increase in wound size, 1 due to reduction. |
|||||||||
Table 7 Summary of the objectives of Lafferty et al. (2011) case series
|
Study component |
Description |
|
Objectives |
To assess whether a modern wound dressing would be able to reduce the burden of chronic wounds and reduce overall costs. |
|
Study design |
A single-centre non-comparative case series |
|
Setting |
Single UK leg ulcer clinic. All patients who attended the clinic were assessed for suitability. |
|
Inclusion/exclusion criteria |
Inclusion criteria:
Exclusion criteria:
Wounds exhibiting signs of local infection were treated with the Iodozyme dressing. |
|
Primary outcomes |
Not specified, but wound area is the only quantitative outcome reported. Each patient had measurements of wound area using the LUTM telemedicine system weekly for the first 6 weeks and then at weeks 12, 16 and 20. |
|
Statistical methods |
A healing rate calculated from the 10 wounds remaining in the study at week 20 was compared with a historical healing rate (2 healed/26 treated) in the clinic over 6 months before the study using a chi-squared statistic. These calculations of healing rate are invalid and not comparable. |
|
Participants |
17 wounds in 13 patients were included, of which 3 wounds were treated with the Iodozyme dressing and the remaining 14 with the Oxyzyme dressing. Wound duration at recruitment ranged from 2–420 months. |
|
Results |
Results are presented for 11 wounds in 9 patients. No data are presented for 6 wounds in 4 patients withdrawn before week 12. One further patient with 1 wound withdrew at week 13. For the 11 wounds reported there was a reduction in wound area from baseline to week 20 from 75.3 cm2 to 8.8 cm2 (88.4%). |
|
Conclusions |
Without a control group it is impossible to draw any conclusions about how the Oxyzyme dressing compares with other dressings. |
|
Abbreviations: LUTM, leg ulcer telemedicine; n, number of patients |
|
Table 8 Summary of the outcomes of Lafferty et al. (2011) case series
|
Oxyzyme and Iodozyme dressings |
Analysis |
|||
|
Recruited |
n=13 patients (17 wounds) |
|||
|
Completed |
n=8 patients (10 wounds) |
|||
|
Primary outcome: reduction in wound area over 20 weeks |
– |
Week 0 wound area (cm 2 ) |
Week 20
wound area |
Results are only reported for 9 patients (11 wounds) out of 13 patients (17 wounds) recruited. *Patient withdrawn at week 13. Wound measurement is carried forward from week 12. |
|
Wound 1 |
6.1 |
2.3 |
||
|
Wound 2 |
2.5 |
0 |
||
|
Wound 3 |
1 |
0.8* |
||
|
Wound 4 |
30 |
0 |
||
|
Wound 5 |
2.5 |
1.3 |
||
|
Wound 6 |
2.8 |
1.3 |
||
|
Wound 7 |
5 |
0 |
||
|
Wound 8 |
16.1 |
0 |
||
|
Wound 9 |
5.3 |
3.1 |
||
|
Wound 10 |
1.5 |
0 |
||
|
Wound 11 |
2.5 |
0 |
||
|
Total area (%) |
75.3 (100.0%) |
8.8 (11.6%) |
||
|
Area reduction |
0% |
88.4% |
||
|
Selected secondary outcomes: |
||||
|
Safety |
n=13 patients (17 wounds) |
– |
||
|
Patients reporting serious adverse events |
5/13 (38.5%) patients withdrawn |
– |
||
|
Death |
1 (1/13, 7.7%) patient with 2 wounds died. |
– |
||
|
Hospitalisation |
2/13 (15.4%) patients (3 wounds) were hospitalised for non-wound related reasons. |
– |
||
|
Non-adherence |
2/13 (15.4%) patients were withdrawn for patient non-adherence issues. |
– |
||
|
Abbreviations: n, number of patients |
||||
Costs and resource consequences
The use of the Oxyzyme or Iodozyme dressings would not need a change in existing treatment facilities or care pathways. No additional equipment would be needed for its use, other than an air-permeable covering dressing, which is already commonly used with hydrogel wound dressings.
It is anticipated that the prescription of the Oxyzyme or Iodozyme dressings would need approval from a specialist such as a tissue viability nurse, or a wound clinic. This is usually needed for more challenging wounds and higher cost dressings and does not mean a change in NHS practice.
The Oxyzyme and Iodozyme dressings have a higher acquisition cost than many other dressings used for hard-to-heal wounds. Any anticipated cost or resource saving would be based on a reduced duration of wound treatment, or reduced frequency of dressing changes.
Published cost studies
A randomised controlled trial (Moffatt et al. 2014) reported cost savings through the use of the Oxyzyme and Iodozyme dressings compared to standard care. The trial also compared clinical outcomes and was described in the clinical evidence section and tables 1 and 2.
The cost of wound treatment was estimated from the number and type of dressings and nurse time during the 12 weeks of the study. The number of dressings used during the weekly assessments, and any additional nurse visits, was recorded. Nurse time was estimated from previous studies combined with information gathered during the trial, but the detail for this calculation is not reported, and there appears to be an assumption that the time is equal for both arms of the study. Cost of nurse time was based on Personal Social Services Research Unit costs for 2011 (Curtis et al. 2011) for a district nurse.
The mean cost per patient treated with Oxyzyme or Iodozyme dressings was £436.33, compared with £525.54 per patient for standard care. The mean cost per ulcer healed at 12 weeks or earlier was £976.54 using Oxyzyme or Iodozyme dressings, compared with £1071.29 per patient for standard care. The cost saving is based on a reduction in the mean number of nurse visits needed per patient, which was 14.8 visits for people who had standard care compared with 10.04 visits for those who had Oxyzyme or Iodozyme dressings.
Moffatt et al. (2014) also included a Markov analysis to model the costs over the 12 week period of patients being treated. The health states used in the model were: treatment according to study protocol, withdrawn from treatment for 2 weeks, treated off-study, or healed. Very little additional information is given about the model inputs and assumptions. The Markov model found a greater number of total ulcer free weeks in the control group (187 weeks), compared with the Oxyzyme or Iodozyme dressing group (175 weeks). The model still showed the total cost of the Oxyzyme or Iodozyme dressing group (£23,801) to be lower than that of the comparator (£28,831). The model results in a cost saving per healed week of £419.25 for people treated with Oxyzyme or Iodozyme dressings.
The resources, costs and results from this study are shown in table 9.
A case series (Lafferty et al. 2011) reported cost savings through the use of the Oxyzyme and Iodozyme dressings. The series also looked at clinical outcomes and was described in the clinical evidence section and tables 5 and 6. The cost of treating wounds using standard care was estimated from the number and type of dressings and clinician time during the 4 months before enrolment in the study. The cost of treating wounds with the Oxyzyme and Iodozyme dressings was estimated from the number and type of dressings and the clinician time over the first 6 weeks of the study. Cost of clinician time was based on Personal Social Services Research Unit costs for 2006 (Curtis et al. 2007). The average weekly cost of treatment with the dressings was £85.40 per patient, compared with £76.80 per patient for standard care. Faster healing with the Oxyzyme and Iodozyme dressings was assumed based on the 6 wounds that healed out of the 10 wounds receiving 20 weeks' treatment. This was compared with a historical healing rate of 2 patients healed out of the entire clinic cohort of 26 patients in the 6 months before the study. It is unclear whether some patients in the historical cohort were also patients included in the study. The authors projected a potential saving of £23,036 over 52 weeks from the use of the Oxyzyme and Iodozyme dressings compared with standard treatment.
The cost estimation (Lafferty et al. 2011) is summarised in table 10.
Table 9 Summary of the cost estimate from Moffatt et al. (2014)
|
Oxyzyme and Iodozyme dressings |
Standard care |
|
|
Resources |
Oxyzyme and Iodozyme dressings Secondary dressings Dressing packs Bandages Nurse time |
Dressings – based on types used prior to study Secondary dressings Dressing packs Bandages Nurse time |
|
Quantity of resource |
Dressings, packs and bandages are based on the recorded number and type of dressings used during 12‑week trial. The number of visits was recorded during the 12‑week trial. Nurse time estimated as 27 min contact per dressing change. This is based on previous studies and information collected during this study, but no details are given. The assumption is that times are equal for both trial arms. |
Dressings, packs and bandages are based on the recorded number and type of dressings used during 12‑week trial. The number of visits was recorded during the 12‑week trial. Nurse time estimated as 27 min contact per dressing change. This is based on previous studies and information collected during this study, but no details are given. The assumption is that times are equal for both trial arms. |
|
Value of resource |
Published drug tariff prices for dressings and bandages (British National Formulary) Mean number of visits was 10.04, over 12 weeks Nurse time: £60 per hour for District Nurse home visit time (Curtis et al. 2011) |
Published drug tariff prices for dressings and bandages (British National Formulary) Mean number of visits was 14.8, over 12 weeks Nurse time: £60 per hour for District Nurse home visit time (Curtis et al. 2011) |
|
Cost of dressings, bandages, packs |
£7763.37 |
£6739.49 |
|
Staff cost |
£12,744.00 |
£21,114.00 |
|
Mean cost per patient treated |
£436.33 |
£525.54 |
|
Mean cost per healed ulcer |
£976.54 |
£1,071.29 |
|
Total ulcer-free weeks (Markov) |
175 |
187 |
|
Total cost of care for 12 weeks (Markov) |
£23,801 |
£28,831 |
Table 10 Summary of the cost estimate from Lafferty et al. (2011)
|
Oxyzyme and Iodozyme dressings |
Standard care |
|
|
Resources |
Oxyzyme and Iodozyme dressings Bandages Clinician time |
Dressings – based on types used prior to study Bandages Clinician time |
|
Quantity of resource |
Estimated weekly resource use per wound over the first 6 weeks of treatment. Faster healing rate (60%) assumed based on 6 wounds healing of the 10 that completed the 20 week study. Using 6/26 wounds originally assessed would give a healing rate of 23.1%. |
Based on recorded number and type of dressings and clinician time spent over the 4-month period before the study commenced. Assumed 8% healing rate based on 2 wounds healed out of 26 treated in the previous 6 months. A statement that 8 of the 26 wounds originally assessed (30.7%) were improving and continued on the same care regimen suggests considerable uncertainty in the standard care healing rate. |
|
Value of resource |
Published drug tariff prices for dressings and bandages Clinician time: £36 per hour (Curtis et al. 2007) |
Published drug tariff prices for dressings and bandages Clinician time: £36 per hour (Curtis et al. 2007) |
|
Projected 52-week cost per wound |
£16,055 |
£39,091 |
Strengths and limitations of the evidence
One randomised controlled trial and 3 case studies were found for the Oxyzyme or Iodozyme dressings, with only the RCT providing evidence comparing these dressings with standard treatment for chronic wounds. Many of the limitations of the evidence are common in wound care research, as discussed in NICE's key therapeutic topic on wound care products.
The Moffatt et al. (2014) study was a randomised controlled trial comparing Iodozyme and Oxyzyme with standard care in an NHS setting for 100 patients. There was no formal power calculation so the results should be interpreted with caution as in underpowered studies, statistical non-significance could be caused by an insufficient sample size. The methods of this study are well reported and the outcomes are relevant to this briefing. The trial was funded by Archimed, now a subsidiary of the manufacturer Crawford Healthcare, and 1 of the authors was formerly an Archimed employee. No trial protocol was publically available before the trial commenced. There were no eligibility restrictions based on criteria for ulcer size or duration. Not all results were fully and clearly reported.
The statistical analysis of quality of life results is unclear, and data on patient reported pain is not presented.
For the cost comparison, the nurse time was estimated from previous studies and information gathered during the trial. No more detail is given for this estimation, and there appears to be an assumption that the time is equal for both arms of the study. Nurse time is the largest component of the treatment cost and more detailed information would have been useful in assessing the appropriateness of the assumptions made. For example, if the time taken to change dressings in the standard care group was shorter, this would impact on the results of the study.
Very little additional information was given about the Markov model inputs and assumptions. The model shows an additional cost per healed week for the group of patients treated with standard care, compared to the people treated with the Oxyzyme or Iodozyme dressings. This additional cost is likely to be based on the greater number of dressing changes in the control group, combined with an assumption of equal nurse time. Additional information would allow a better assessment of the model.
The proportion of patients withdrawn from the 3 case studies was very high, resulting in a strong likelihood of bias.
Two case series had very similar protocols. Davis et al. (2009) is a case series using the Oxyzyme dressing for the treatment of chronic wounds (n=100 patients) of various aetiologies conducted in 27 European centres, with a mean of 3 patients per centre (range 1–12). Wood et al. (2010) is a case series using the Iodozyme dressing for the treatment of chronic wounds of various aetiologies with 45 patients in 30 centres in England.
Each centre only treated a small number of patients. There was no meaningful consecutive recruitment and therefore a likelihood of recruitment bias.
The large number of centres included means that the variety in practice in wound care was realistically reflected. However, with small numbers in each centre and a variety of wound types, the probable differences in treatment and outcome measurement may have masked a common effect.
Wound area was based on non-blinded clinician assessment if digital photographs were not available. This meant there was a high likelihood of bias for such wounds (number not reported). Patients whose wounds increased in size were withdrawn, so the study could only show a reduction in mean wound area.
A key reported outcome was the overall condition of the wound at the end of the study. This was determined by non-blinded clinical opinion as healed, improved, static or deteriorating. 'Improved' included a reduction in size or an improvement in the wound bed, so that wounds could be classified as improved even when the wound size had increased, or the patient had to be withdrawn due to problems with the dressing. Assessment of wound bed condition was very subjective unless carefully defined and reported, which was not apparent in this paper. These factors make the probability of bias sufficiently high that the outcome was not reported here.
All authors were current or past employees of Archimed Wound Care Division, the manufacturer of the Oxyzyme and Iodozyme dressings when the study was published.
In addition, Davies et al. (2009) included a large number of very small wounds (10 ulcers <1 cm2 and 49 ulcers 1–10 cm2). For very small wounds a small change in wound measurement resulted in a large percentage change.
Lafferty et al. (2011) was a study of the Oxyzyme and Iodozyme dressings in patients with chronic wounds (n=13 patients, 17 wounds) referred to a single UK leg ulcer clinic. The author did not report the results of all the patients recruited into the study (13 patients with 17 wounds recruited, 9 patients with 11 wounds reported). Statistical analysis was applied to the 11 wounds in 9 patients reported, resulting in a strong likelihood of bias. The historical comparator group differed from the patients in the study because:
-
it included all patients attending the clinic in a 6‑month period, with no inclusion or exclusion criteria
-
individual patients were not followed for a specified time period; some may have attended the clinic for considerably less than the 20‑week study period of the intervention group
-
there were no withdrawals.
For these reasons the historical comparator group was not comparable to the reported patients in the study.
Of the initial cohort of 17 wounds in 13 patients, 3 wounds were treated with the Iodozyme dressing and 14 with the Oxyzyme dressing. However, after withdrawals it was unclear how many were being treated with the Iodozyme dressing. The study did not compare the Oxyzyme and Iodozyme dressings with any other treatment so it was not possible to form any conclusions about their effectiveness. Two authors were current or past employees of Archimed, the manufacturer of the dressings at the time of study publication.
The cost analysis undertaken in Lafferty et al. (2011) was flawed. The healing rate for the Oxyzyme and Iodozyme dressings was calculated as a proportion of wounds completing the 20‑week study (60% [6/10] healed) rather than the total number of wounds in patients recruited (35% [6/17] healed), or the total number of wounds in patients assessed for the study (23% [6/26] healed) if it is assumed that no excluded patients healed. The comparator group included all patients attending the clinic in the previous 6 months, so the patient populations were not comparable. The cost savings claimed were based on the assumed differences in healing rates. The costs of treating patients for whom the study dressings were not suitable were ignored, as were the costs of treating the patients withdrawn.