Efficacy of a povidone‐iodine foam dressing (Betafoam) on diabetic foot ulcer

Abstract

This study aimed to assess the efficacy of a new povidone‐iodine (PVP‐I) foam dressing (Betafoam) vs foam dressing (Medifoam) for the management of diabetic foot ulcers. This study was conducted between March 2016 and September 2017 at 10 sites in Korea. A total of 71 patients (aged ≥19 years) with type 1/2 diabetes and early‐phase diabetic foot ulcers (Wagener classification grade 1/2) were randomised to treatment with PVP‐I foam dressing or foam dressing for 8 weeks. Wound healing, wound infection, patient satisfaction, and adverse events (AEs) were assessed. The PVP‐I foam and foam dressing groups were comparable in the proportion of patients with complete wound healing within 8 weeks (44.4% vs 42.3%, P = .9191), mean (±SD) number of days to complete healing (31.00 ± 15.07 vs 33.27 ± 12.60 days; P = .6541), and infection rates (11.1% vs 11.4%; P = 1.0000). Median satisfaction score (scored from 0 to 10) at the final visit was also comparable between groups (10 vs 9, P = .2889). There was no significant difference in AE incidence (27.8% vs 17.1%, P = .2836), and none of the reported AEs had a causal relationship with the dressings. The results of this study suggest that PVP‐I foam dressing has wound‐healing efficacy comparable with foam dressing, with no notable safety concerns. This study was funded by Mundipharma Korea Ltd and registered at ClinicalTrials.gov (identifier NCT02732886).

1. INTRODUCTION

Diabetic foot ulcer is one of the most significant complications of diabetes. It is defined as ulceration associated with neuropathy and/or peripheral arterial disease of the lower limb in a patient with diabetes.1 The prevalence of diabetic foot ulcer in diabetic patients is 4% to 10%.2, 3 In Korea, from an epidemiological survey of almost 4 million diabetic patients, the prevalence of foot ulcer was 99.7 per 100 000.4 The lifetime risk of diabetic patients developing a diabetic foot ulcer is 15%.2, 3 While the majority (60%‐80%) of ulcers eventually heal,1 delayed healing carries increased risk of infection, leading to amputation. Between 40% and 70% of non‐traumatic amputations of lower limbs are reported to occur in patients with diabetes.5

Although prevention of diabetic foot ulcer is important, better outcomes can be achieved through early treatment and effective management, including prevention of secondary infection and stopping progression to tissue necrosis. Timely management may prevent 85% of foot amputations.6 Current management of diabetic foot ulcers takes into account the specialised nature of the wound, including poor blood circulation. Goals for effective diabetic ulcer management include wound protection and prevention of additional infections while maintaining a moist environment to promote development of granulation tissue and absorption of wound exudate. Polyurethane (PU) foam dressings may address some of these needs as they are designed to maintain a moist environment at the wound surface and absorb wound exudate. In a recent retrospective study of diabetic foot ulcer treatment, complete healing rates in a 3‐month period were higher with PU foam dressing than with saline gauze,7 suggesting that these properties may support better healing of this type of chronic wound.

As conventional wound dressings provide only a basic protective barrier against pathogens, it is of interest to understand whether, and under what circumstances, dressings with added antimicrobial agents can positively influence overall wound healing or clinical outcomes. Povidone‐iodine (PVP‐I) is an antimicrobial agent that has been used in a wide range of wound‐healing applications for over 60 years. It has broad‐spectrum antimicrobial activities (against bacteria, viruses, and fungi), anti‐inflammatory properties, and a well‐established safety profile.8, 9 Some reports have suggested that PVP‐I may interfere with wound healing; however, in vivo studies have demonstrated that PVP‐I does not disturb wound healing, especially when lower doses are used (<5% PVP‐I).10

PVP‐I foam dressing (Betafoam) is a PU foam dressing impregnated with 3% PVP‐I, designed to address additional specificities of wound care, such as antimicrobial protection, in addition to exudate management and maintaining a moist wound environment. In in vitro studies, PVP‐I foam dressing was shown to have more rapid antimicrobial activity compared with two silver foam dressings (99% reduction of S aureus and P aeruginosa in 1 minute [PVP‐I foam dressing] vs 10 minutes [Allevyn Ag] and vs 20 minutes [PolyMem Ag]; P < .05), as well as having the fastest fluid absorption and greatest fluid retention capacity among the tested dressings.11 Clinical studies of PVP‐I foam dressing have also shown good efficacy and safety for the management of acute partial‐thickness burn wounds12 and split‐thickness skin graft donor sites.12 In particular, for these types of wound, no delays in wound healing were observed with PVP‐I foam dressing when compared with other conventional wound dressings.12, 13 These observations suggest that PVP‐I foam dressing may be suitable for the management of other types of wounds, including diabetic foot ulcers.

The objective of this multicentre study was to prospectively evaluate the efficacy and safety of this PVP‐I foam dressing for the management of diabetic foot ulcers, which often do not heal well.14 We compared the PVP‐I foam dressing with a control foam dressing (Medifoam) that had similar composition and properties, obtained from the same manufacturer. As this antimicrobial‐containing PVP‐I foam dressing has not previously been studied in humans for this specific wound type, we evaluated its safety, performance, and handling characteristics, compared with the control foam dressing, in terms of both clinical parameters and patient‐reported outcomes.

2. METHODS

This multicentre, open‐label, randomised study was conducted between March 2016 and September 2017 at 10 sites in the Republic of Korea. The local independent ethics committee at each site reviewed and approved the study protocol and documents used for informed consent prior to study initiation. The study was conducted in accordance with Korea Good Clinical Practice, the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use guidelines, the Declaration of Helsinki, and applicable regulations in Korea. All patients provided written informed consent before any study‐related procedures were performed. This study was registered at ClinicalTrials.gov (identifier NCT02732886).

2.1. Patients

Key inclusion criteria included: aged ≥19 years, Type 1 or 2 diabetes with glycosylated haemoglobin (HbA1c) ≤10%, early‐phase (Wagner classification grade 1 or 2) diabetic foot ulcer, post‐debridement ulcer size of ≥1 by 1 cm², no clinical signs of infection or necrosis of the ulcer, and no serious blood circulation issue in the foot. Key exclusion criteria included: pregnancy; known hypersensitivity to PVP‐I foam dressing, foam dressing, or PVP‐I; hyperthyroidism, other thyroid dysfunctions, or treatment with radioiodine; other comorbidities that may affect wound healing (eg, acute/chronic renal failure, autoimmune disease, immunocompromised patient, etc.); requirement for continuous antibiotic or growth factor‐containing treatment; and requirement for a skin tissue transplant. All patients provided written informed consent before study enrolment.

2.2. Study design

Eligible patients were randomised to treatment with PVP‐I foam dressing or foam dressing in a 1:1 ratio by block randomisation, with allocation codes prepared in advance in the order of study participation. The pre‐generated randomisation list was sent to investigators prior to study initiation. Study treatment was not blinded.

The duration of the study was up to 8 weeks, with randomisation at Day 0 (visit 1) followed by an 8‐week treatment period (visits 2 to 17). After randomisation, the dressings were applied to the target ulcer after washing with normal saline and debridement if necessary. In cases where there are two or more ulcer lesions, the largest ulcer was selected as the target ulcer. If ulcer lesions were of the same size, the ulcer with a higher Wagner grade was selected as the target ulcer. The dressings were cut to an appropriate size to cover the target ulcer and were changed at subsequent visits (scheduled every 3‐4 days). If the patient changed the dressing between site visits, or the dressing was changed at an unscheduled site visit, this was recorded in a patient diary and in the case report form. Wound assessments were conducted at each visit. If complete wound healing was confirmed, the end‐of‐study visit procedures were conducted, and the patient was considered to have completed the study.

2.3. Study endpoints

The primary efficacy endpoint was the proportion of patients with complete wound healing of the target ulcer at Week 8.

Secondary efficacy endpoints included the proportion of patients with complete wound healing of the target ulcer at Week 4, proportion of patients with ≥50% wound healing at Week 8, number of days to complete healing of the target ulcer, rate of change in target ulcer size, incidence of infection at the target ulcer and patient satisfaction, and the total and mean number of the dressing changes per day.

Safety endpoints included the incidence of adverse events (AEs) and changes in clinical laboratory variables (haematology, blood chemistry, and urinalysis) and vital signs.

2.4. Study assessments

2.4.1. Wound healing

At each visit, the wound was assessed for healing by visual inspection by the investigator. Complete wound healing was defined as a completely epithelialised state with no discharge.

2.4.2. Wound size

Wound size was determined by placing a transparent film on the wound site and tracing the outline of the wound. The trace area was then measured for length, width, and area using Visitrak digital planimetry (Smith & Nephew Wound Management, Inc, Largo, Florida).

2.4.3. Infection

At each visit, the wound was checked for infection using a clinical signs and symptoms checklist. Signs and symptoms on the checklist included pain in the wound area, erythema, oedema, heat, exudates, delayed healing, granulation tissue status, and foul odour. Infection was defined as purulent exudates or at least two of the following symptoms: erythema, increasing pain around the wound, oedema, or heat.

2.4.4. Satisfaction

Patient satisfaction was assessed at every other visit from Visit 3 onwards on a 10‐point numeric rating scale from 0 (not satisfied at all) to 10 (very satisfied).

2.4.5. Safety

AE data were collected via spontaneous patient reporting, patient interview, and examination by the investigator. A three‐grade scale was used to assess the severity of AEs: mild—mild discomfort but does not interfere with daily activities; moderate—significant discomfort that interferes with daily activities; and severe—prevents normal daily activities. Serious AEs (SAEs) were any AEs that resulted in death, were life‐threatening, required inpatient hospitalisation or prolongation of existing hospitalisation, resulted in persistent or significant disability or incapacity, or resulted in a congenital anomaly or birth defect or other medically important events.

2.5. Statistical analyses

This study planned to recruit 70 patients in total, 35 per group. As this was the first exploratory study designed to estimate healing rates and other clinical parameters of this particular PVP‐I foam dressing for this chronic wound type, no formal hypotheses were tested, and formal statistical sample size estimation was not performed. The Safety Set consisted of all patients to whom the PVP‐I foam dressing or foam dressing was applied at least once. The Full Analysis Set (FAS) included patients in the Safety Set in whom primary efficacy endpoint data (complete wound healing at Week 8) was obtained. The Per Protocol Set (PPS) included patients in the FAS who completed the study without any major protocol violations. Demographic baseline data, clinical medical history, and safety data were analysed in the Safety Set, and efficacy data were analysed in both the FAS and PPS.

Demographic variables, diabetes and ulcer history, efficacy, and safety outcomes were summarised by treatment group using descriptive statistics. Differences between the treatment groups were analysed using the χ ² test or Fisher's exact test. P‐values <.05 were considered statistically significant. All statistical analyses were performed using SAS statistical analysis software (version 9.4; SAS Institute, Cary, North Carolina).

3. RESULTS

3.1. Patient disposition and baseline characteristics

A total of 87 patients were screened, of which 71 were deemed eligible and randomised to PVP‐I foam dressing (36 patients) or foam dressing (35 patients) (Figure 1). All 71 patients met the criteria for both the Safety Set and the FAS. Among them, 22 patients withdrew prior to study completion; reasons for premature discontinuation included withdrawal of consent (PVP‐I foam dressing three; foam dressing two), protocol violations (PVP‐I foam dressing nine; foam dressing nine), physician's decision (PVP‐I foam dressing two; foam dressing one), AEs (PVP‐I foam dressing two; foam dressing one), and lack of efficacy (PVP‐I foam dressing zero; foam dressing two). The PPS included 40 patients who completed the study without major protocol deviations (PVP‐I foam dressing 20; foam dressing 20).

Figure 1.

Patient flow through the trial

The mean age of the patients was 62 years (range 30‐88 years), and 79% were male. Demographic variables, duration of diabetes, medical history, and baseline ulcer characteristics (duration, aetiology, location, Wagner classification), except for the leg the ulcer was located on, were comparable between treatment groups (Table 1).

Table 1.

Demographic variables and baseline ulcer characteristics (Safety Set)

Characteristic	PVP‐I foam dressing n = 36	Foam dressing n = 35	P‐value
Gender, n (%)			.7247a
Male	29 (80.6)	27 (77.1)
Female	7 (19.4)	8 (22.9)
Age (years), mean ± SD	64.5 ± 11.1	60.7 ± 12.8	.1834b
Duration of diabetes (months), mean ± SD	91.6 ± 102.9	61.9 ± 42.9	.5677b
Duration of target ulcer (months), mean ± SD	1.68 ± 1.86	1.69 ± 1.42	.5603c
Leg with target ulcer, n (%)
Left	23 (63.9)	11 (31.4)	.0062a
Right	13 (36.1)	24 (68.6)
Aetiology of target ulcer, n (%)
Neurological‐ischaemic	21 (58.3)	22 (62.9)	.6966a
Neurological	15 (41.7)	13 (37.1)
Location of target foot ulcer, n (%)
Basal	15 (41.7)	21 (60.0)	.1224a
Dorsal	21 (58.3)	14 (40.0)
Wagner classification, n (%)
Grade 1	25 (69.4)	27 (77.1)	.4638a
Grade 2	11 (30.6)	8 (22.9)

Abbreviation: PVP‐I, povidone‐iodine; SD, standard deviation.

^a χ ² test.

^b2‐sample t test.

^cWilcoxon rank sum test.

3.2. Wound healing and dressing changes

The proportion of patients with complete wound healing at Week 8 was 44.4% (16 patients) in the PVP‐I foam dressing group and 44.1% (15 patients) in the foam dressing group (P = .9781, Table 2). At Week 4, the proportion of patients with complete wound healing was 22.2% in the PVP‐I foam dressing group and 18% in the foam dressing group (P = .6324, Table 2). Both groups were comparable in the proportion of patients with ≥50% wound healing at Week 8 (PVP‐I foam dressing vs foam dressing: 69.4% vs 80.0%, P = .4030), mean number of days (±SD) to complete healing (31.00 ± 15.07 vs 33.27 ± 12.60 days, P = .6541), and rate of change in ulcer size (length, width, area) (Figure 2).

Table 2.

Wound healing and infection rates of the target ulcer (Full Analysis Set)

	PVP‐I foam dressing	Foam dressing	P‐value
	(N = 36)	(N = 35)
Complete wound healing, n (%)
Within 8 weeks	16 (44.4)	15 (44.1)	.9781a
Within 4 weeks	8 (22.2)	6 (17.6)	.6324a
≥50% wound healing within 8 weeks, n (%)	25 (69.4)	28 (80.0)	.4030a
Time to complete healing of the target ulcer (days), mean ± SD	31.0 ± 14.1	33.3 ± 12.6	.6541b
Local infection, n (%)
Yes	4 (11.1)	4 (11.4)	1.0000c
No	32 (88.9)	31 (88.6)

Abbreviation: PVP‐I, povidone‐iodine; SD, standard deviation.

^a χ ² test.

^b2‐sample t test.

^cFisher's exact test.

Figure 2.

Percentage change rate of ulcer length (A), width (B), and area (C) from Visit 1 at each visit (FAS). Values are mean ± SD

The total number of dressing changes was assessed in patients with complete wound healing. In the PVP‐I foam dressing group (n = 16), the mean number of dressing changes (±SD) was 29.9 ± 15.1; in the foam dressing group (n = 15), it was 32.2 ± 12.5 (P = .6549). The mean number of dressing changes per day (±SD) was comparable between the groups (PVP‐I foam dressing 0.95 ± 0.03; foam dressing 0.93 ± 0.02).

The analyses on the PPS were comparable with those on the FAS.

3.3. Infection

Overall, four patients in each group had an infection at the target ulcer (11.1% PVP‐I foam dressing, 11.4% foam dressing, P = 1.0000) (Table 2).

3.4. Satisfaction

On a numeric rating scale from 0 (not satisfied at all) to 10 (very much satisfied), mean satisfaction scores (visits 3, 5, 7, 9, 11, 13, 15, and 17) ranged between 8.23 and 9.23 in the PVP‐I foam dressing group and 8.09 and 8.96 in the foam dressing group, with no statistically significant differences between the groups at any visit (Figure 3).

Figure 3.

Patient treatment satisfaction per visit. Patients rated their treatment satisfaction at every other visit on a numeric rating scale ranging from 0 (not satisfied at all) to 10 (very satisfied). Values are mean ± SD

3.5. Safety endpoints

The incidence of AEs was 27.8% (n = 10) in the PVP‐I foam dressing group and 17.1% (n = 6) in the foam dressing group (P = .2836, Table 3). All AEs were assigned a causality assessment of “definitely not related” by the investigator. The majority of AEs were mild in both groups (Table 3). Five patients reported six SAEs in the PVP‐I foam dressing group (localised infection, cellulitis, ankle fracture, diabetes mellitus inadequate control, extremity necrosis, peripheral vascular disorder). One patient reported one SAE in the foam dressing group (peripheral swelling). There were no significant differences between groups in clinical laboratory findings, vital signs, and physical examinations.

Table 3.

Adverse events (Safety Set)

	PVP‐I foam dressing		Foam dressing		P‐value
	(N = 36)		(N = 35)
	Incidence, n (%)	Events	Incidence, n (%)	Events
Adverse events	10 (27.8)	14	6 (17.1)	7	.2836a
Mild	9 (25.0)	12	4 (11.4)	5
Moderate	1 (2.8)	1	2 (5.7)	2
Severe	1 (2.8)	1	0 (0.00)	0
Unexpected adverse Events	9 (25.0)	13	6 (17.1)	7	.4175a
Serious adverse events	5 (13.9)	6	1 (2.9)	1	.1987b
Adverse events leading to drop out	2 (5.6)	2	1 (2.9)	1	1.0000b
Adverse device effects	0 (0.0)	0	0 (0.0)	0	‐
Skin‐related adverse events at the application site	1 (2.8)	1	0 (0.0)	0	1.0000b

Abbreviation: PVP‐I, povidone‐iodine.

^a χ ² test.

^bFisher's exact test.

4. DISCUSSION

As the first prospective clinical study of this PVP‐I foam dressing for diabetic wound management in humans, this was designed as an exploratory study to document clinical performance parameters and safety profile across multiple study sites. Existing clinical data for this PVP‐I dressing relate to its use of dressing clean wounds (skin graft donor sites)13 and partial‐thickness burn wounds,12 whose characteristics and healing profile may be quite different. In terms of wound‐healing parameters, the present findings are consistent with an earlier study of foam dressing for diabetic foot ulcers7 and generally support the use of foam dressings for non‐infected low‐grade diabetic ulcer wounds.

Results from this prospective 8‐week study indicate that both PVP‐I foam dressing and control foam dressing promoted significant healing of diabetic foot ulcers over 8 weeks (44% in both groups achieved complete healing within 8 weeks), with comparable efficacy and safety between the two dressings. In terms of healing rates, the findings appear consistent with those observed in an earlier retrospective study of patients with diabetic foot ulcers. In that study, patients were treated using a standardised wound management protocol with either foam dressing or saline gauze, and the results suggested foam dressing to have greater wound‐healing properties in diabetic foot ulcer than normal saline gauze as assessed by the proportion of patients reaching complete wound healing within 12 weeks (63.5% vs 39.4%, P < .05).7

In our study, other measures, including ulcer wound healing, changes in ulcer size, and infection rates, were not significantly different between groups. The number of dressing changes was comparable, and patients in both groups were highly satisfied with their treatment. The safety profile of the two dressings also appeared to be similar with no statistically significant difference in the incidences of AEs observed.

Besides documenting standard wound‐healing parameters, it was envisaged that comparison with a similar but non‐antimicrobial foam dressing would provide some information on effects related to the PVP‐I component of the dressing. Prevention of infection is a major goal of chronic wound management, and PVP‐I foam dressing, which contains the antimicrobial agent PVP‐I, is posited to have an added advantage over standard foam dressing in this respect. In the present study, although target ulcer wounds were assessed at each visit for signs of infection, the study was not specifically designed to detect differences in infection rates between dressing groups. Future studies designed to evaluate infection control or prevention in patients with chronic or non‐healing ulcer wounds may be necessary to clarify the potential advantages of PVP‐I foam dressing for these applications. In addition, comparisons with dressings that contain other types of antimicrobial agents (eg, silver) may also be of interest.

Some antimicrobial‐containing dressings are known to delay wound healing15; however, in this study, 3% PVP‐I in PVP‐I foam dressing did not appear to prolong wound‐healing time. The mean number of days to complete wound healing and rate of change in ulcer size at each visit were comparable between the PVP‐I foam dressing and foam dressing groups in this study. These results are consistent with those from recent clinical studies showing similar wound‐healing time (no delay of healing) between the PVP‐I‐containing foam dressing and other conventional wound dressings.12, 13 Interestingly, one of these studies demonstrated significantly shorter wound‐healing time of partial‐thickness burn wounds with PVP‐I foam dressing than with two other dressings that did not contain antimicrobial agents (Allevyn and petrolatum gauze).16

As limited information was available to estimate the difference in healing rates with PVP‐I containing foam dressing vs the control foam dressing, the target study sample size was based on practical considerations. Whilst several secondary endpoints (proportion of patients with complete wound healing at 4 weeks, patient report of treatment satisfaction, and total number of dressing changes) showed numerical trends in favour of PVP‐I foam dressing, there were no statistically significant differences in efficacy endpoints. In addition, because of the requirement for clean and uninfected target ulcer wounds at study entry, only the incidence of secondary infections could be compared, and the number of events was low in both groups. Nonetheless, this multicentre study of PVP‐I foam dressing for diabetic foot ulcers paves the way for future studies to evaluate the use of PVP‐I foam dressing for different aspects of chronic wound care.

In conclusion, the findings of this study suggest that PVP‐I foam dressing has wound healing efficacy and safety comparable with foam dressing in diabetic foot ulcer management.

CONFLICT OF INTEREST

K.K. is an employee of Mundipharma Korea Ltd. All other authors have no relevant relationships to disclose.

TRADEMARK STATEMENT

®: Betafoam and Medifoam are registered trademarks of Genewel, Seongnam, Korea.

Gwak HC, Han SH, Lee J, et al. Efficacy of a povidone‐iodine foam dressing (Betafoam) on diabetic foot ulcer. Int Wound J. 2020;17:91–99. 10.1111/iwj.13236