Disclosure: This study was financially supported by ConvaTec, a Bristol-Myers Squibb Co., Princeton, New Jersey.

Introduction

The management of chronic wounds represents a significant burden, not only in terms of direct costs to healthcare services[1,2] but also in terms of pain, economic loss, and impaired quality of life experienced by patients. Any open wound is at risk of infection with potentially pathogenic organisms,[3–7] which may delay the healing process with resultant increases in trauma to the patient and burden of wound care. Prevention of infection has driven the development of occlusive wound dressings since the 1960s, and a variety of dressings, such as hydrocolloids, polyurethane films, and foams, are now available. The primary purpose of these dressings is to support wound healing by maintaining an optimal moisture balance while presenting a barrier to colonization by potential pathogens.[8–10]

While systemic antibiotics are essential components of the management of nonhealing infected wounds, debate ensues over the role of systemic and topical agents in the treatment of nonhealing, noninfected wounds.[11] Other nonmicrobiological solutions need to be considered as part of the multidisciplinary management of chronic wounds.

The dressing used in this safety evaluation is composed of sodium carboxymethylcellulose fibers (SCF) and has been available, without ionic silver, for clinical use since the mid-1990s (Aquacel®, ConvaTec, A Bristol-Myers Squibb Co., Princeton, New Jersey). The dressing fibers absorb wound exudate by vertical wicking and form a cohesive gel as they swell. The gel matrix molds itself over the wound surface, eliminating dead space and maintaining an optimal moist wound healing environment. The SCF dressing has also been shown to immobilize exudate containing bacteria and polymorphonuclear leukocytes within the dressing.[12] There is now considerable clinical experience with the SCF dressing,[13–18] and it is also associated with increased caregiver and patient benefits in terms of ease of application and removal and reduction in pain at dressing change.[17]

Silver has been used widely for many years in wound care to help manage local infection.[19–22] It has been incorporated into a variety of topical creams and solutions, as well as dressings, and exerts its antiseptic action by inhibiting bacterial enzymes associated with respiration. The principal dose-limiting effect for silver in humans is argyria, an irreversible discoloration of the skin and mucous membranes that is associated with systemic consumption and is considered cosmetically disfiguring rather than toxic per se.[20] However, very high systemic doses of silver are required to result in argyria, and it is very unlikely that exposure to topical silver in dressings used for chronic wounds carries any significant risk.

Currently available silver-containing dressings all utilize different technologies to make silver available in the dressing and at the wound dressing interphase. Not all silver dressings, however, provide the benefits of moist wound healing.[23,24] Ionic silver has been incorporated into the SCF dressing (Aquacel® Ag) without affecting its physical performance characteristics at a concentration effective against wound pathogens in vitro. The principle inherent in the ionic silver dressing is that silver ions released from the fibers are available to act on the bacteria trapped by the SCF dressing. This principle is intended to help maintain a low bioburden within the dressing and to assist in creating an antimicrobial barrier. Nonhealing can be a sign of increased bacterial burden.[25] Due to its physical performance characteristics, the ionic silver dressing also handles the increased levels of exudate so often associated with infection.[26] Innovative products providing such characteristics as bacterial balance, moisture balance, and sometimes autolytic debridement may have roles in the management of the bacterial balance predominantly in the surface wound compartments.[27] The ionic silver dressing is intended for the management of a wide range of acute and chronic wounds based on the clinical experience with the SCF dressing.

To provide confirmation that the addition of ionic silver to the SCF dressing does not pose any additional safety risks, a small, open-label, pilot study was undertaken in subjects with chronic leg ulcers. The study was also intended to yield supportive evidence of the efficacy of the ionic silver dressing in the management of such wounds.

Methods

This was a multicenter, noncomparative, non-randomized, pilot trial conducted in France and Germany between September and December 2001. It was designed to evaluate primarily the safety and also the initial performance of the ionic silver dressing in the management of chronic leg ulcers.

To participate in the study, 18 subjects (8 men and 10 women) with a mean age of 78.2 years (range 63.0–91.0) provided written informed consent. All subjects had chronic leg ulcers that had been present for a mean of 2.59 years (range 0.12–7.00) and were more than 2cm2 in size. Subjects with wounds too large for management with the test product (>12cm in any dimension) were excluded from the study, as were those with a known hypersensitivity to any dressing component.

Subjects underwent initial assessments, including evaluation of their leg ulcer histories, etiology (confirmed by ankle-brachial pressure index [ABPI]), and ulcer and periulcer skin condition. Acetate tracings and photographs of each ulcer were obtained at baseline, as were wound cultures to determine the level of bacterial colonization. The ulcers were dressed with the ionic silver dressing and covered with a moisture-retentive secondary dressing (DuoDERM® Extra Thin, DuoDERM® CGF, or CombiDERM®, ConvaTec). It should be noted that all patients were treated with compression therapy where clinically indicated both prior to and during the study. Dressings were changed as clinically indicated but were not left in place for more than seven days.

The primary objective was to monitor safety through the incidence and nature of adverse events. Efficacy was evaluated by assessment of levels of pain and comfort experienced by the subjects while the dressings were in place and upon removal (pain was assessed by the Johns Hopkins Pain Rating Instrument[28] where 0 = no pain and 10 = intolerable pain). The dressing was also evaluated for nonadhesion and conformability, exudate management, autolytic debridement, clinical improvement, and healing rate. Reduction in bioburden was assessed by semiquantitative analysis of wound cultures taken at baseline and on days 7, 14, 21, and 28. The investigator rated the overall change in wound condition at final evaluation on day 28.

Data from all subjects having at least one dressing (the intention-to-treat population) were included in the analyses. Demographics and baseline characteristics were summarized using descriptive statistics. Quantitative efficacy variables were analyzed using one-way analysis of variance and qualitative variables by Fisher’s exact test. Analysis of the efficacy-related variables for ordered categories was performed using the Cochran-Mantel-Haenzsel test. All analyses were conducted using validated SAS programs with a 0.05 level of significance.

Results

Patient disposition and characteristics. The demographic and baseline characteristics of the 18 subjects enrolled are summarized in Table 1. There were 15 evaluable subjects; 13 subjects completed the 28-day study period, and two subjects completed the study on Days 25 and 27, having undergone at least 80 percent of the study period. Two subjects were withdrawn: one due to an adverse event unrelated to the test dressing (possible nephritis) and one due to pain experienced on initial application of the dressing. One subject was lost to follow up.

Infection, based purely on generally accepted clinical criteria, was present at baseline in 11 of 15 (73%) evaluable subjects (Table 2). The organisms isolated were typical of the microflora found in chronic wounds (Table 3).[11] Seven subjects were using topical antibiotics for their wounds at baseline. One subject was receiving oral antibiotics at baseline for a urinary infection; this subject developed a wound infection during the study and continued on systemic antibiotic therapy. There were no significant differences in demographic characteristics between subjects with infected and noninfected ulcers.

The majority of the ulcers were of venous or mixed etiology, and the mean level of pain experienced at baseline according to the Pain Instrument was 4.33 (range 0–10) (Table 1). Half of the ulcer beds showed evidence of granulation at baseline and the remaining half showed evidence of slough; epithelial tissue was minimal (Table 2). In the skin surrounding the ulcers, there was evidence mainly of erythema but also of maceration and cellulitis (Table 2). The dressings used most frequently by the subjects at baseline were alginate or hydrocolloid or were not specified (Table 2).

Adverse events. There were no serious adverse events. Ten adverse events were reported by seven subjects. All events were isolated incidences and were not correlated with infection status, with the exception of redness and inflammation, which was reported in three subjects with infected ulcers (27%) but not in those with no evidence of infection. Nine of the adverse events were mild or moderate in severity; one was severe (wound enlarged/deteriorated). Only one event was considered to be related to the dressing management (burning sensation in the wound three hours after application); the subject withdrew from the study and the event resolved fully. One other adverse event was considered to be probably related (discoloration of the wound bed on Day 18, which resolved spontaneously on Day 20). Both events were mild in severity.

Dressing changes. There were 129 dressing changes among the 15 evaluable patients in the study, 84 of which were for the infected ulcers. The mean wear time was similar in both infection-status groups; the overall mean wear time was 3.59 days. The majority of the dressing changes (72%) were for routine reasons, although maximum wear time was also a reason for dressing change for the noninfected ulcers. Exudate levels at each dressing change were rated as none or minimal on 35/129 (27.1%) dressing changes, moderate on 68 (52.7%) changes, and heavy on 26 (20.2%) changes. No leakage was detected in over two-thirds (69.8%) of the cases. Significantly more subjects with infected ulcers had no leakage compared to those with noninfected ulcers (86% vs. 40%; p<0.0001).

Pain. Overall, there was a significant reduction in pain scores from baseline to final assessment, both while the dressing was in place and upon removal (Figure 1). No pain while the dressing was in place was reported at 42/129 (32.6%) dressing changes. In a further 59 (45.7%) changes, levels of pain while the dressing was in place were <=4 on the rating scale. Scores of 9 or 10 were applied on only two (1.6%) occasions. Subjects with noninfected ulcers gave significantly lower pain scores while their dressings were in place than those with infected ulcers (1.13 versus 2.43; p=0.0042). Similar findings were recorded for pain on dressing removal. No pain was recorded at 59 (45.7%) dressing changes, and low levels of pain (score <=2.5) were recorded on 43 (33.3%) occasions. On all 129 dressing changes, the subjects rated the dressing as either very comfortable (13.2%) or comfortable (86.8%) while in place.

Adhesion and conformability. All 129 dressings were rated as easy to remove. Saline was used to aid removal in 62 (48.1%) dressing changes overall, more frequently in the infected (55%) than in the noninfected (36%) ulcers (p=0.04). Five changes (3.9%) resulted in mild or moderate trauma to the surrounding skin. For all subjects entering the study who had at least one dressing change (n=18), the investigators’ overall assessment for ease of removal was excellent for seven (38.9%) subjects, good for eight (44.4%) subjects, and fair for three (16.7%) subjects (Figure 2). Likewise, all dressings were reported to be easy to apply, and the investigators’ overall rating of excellent, good, or fair was applied to 44.4 percent, 33.3 percent, and 22.2 percent of subjects, respectively.

Autolytic debridement. At the end of the study, the mean percentage of slough for all ulcers had decreased from 48.6 to 16.5 percent while the percentage of granulation tissue and epithelium had increased, respectively, from 49.3 to 60.5 percent and from 1.6 to 21.9 percent (Figure 3). All but one ulcer showed a reduction in the percentage of slough with a corresponding increase in granulation tissue and new epithelium.

Clinical improvement and healing. In the final analysis of all subjects receiving at least one dressing (n=18), seven (39%) were assessed by the investigators as having marked improvement and 10 (56%) were assessed as having mild improvement (Figures 2 and 4). One subject (5%) showed marked deterioration; this patient had severe ischemia, was terminally ill, and was receiving intravenous morphine. Although no ulcers healed during the study period, a mean reduction in ulcer area of 23.9 percent was measured for all ulcers (Figure 3). Interestingly, for ulcers infected at baseline, a greater mean reduction in area (32.5%) was seen.

Incidence of infection. In all 11 cases of infection at baseline, the use of systemic antibiotics was not considered by the investigators to be clinically indicated. At the end of the study period, infection had resolved in 2 of 11 infected ulcers without the use of systemic antibiotics. The wound microflora observed in the study ulcers typically reflects that found in chronic wounds11 (Table 3). The mean bacterial count at baseline in infected ulcers was significantly higher in infected wounds than in noninfected wounds (2.72 log10 versus 1.00 log10; p=0.0004). At the final visit, the bacterial burden in infected wounds had declined from baseline to 2.27 log10 (–0.43 log10), whereas in noninfected wounds there was a small increase in count to 1.67 log10 (+0.83 log10). This latter increase may be explained by the fact that new infections developed during the study period in those ulcers not infected at baseline.

Discussion

The objectives of this study were to evaluate the safety and initial clinical performance of the new SCF-based ionic silver dressing in the management of chronic leg ulcers. The subjects had been affected by their ulcers for a mean duration of approximately 2.5 years prior to the first application of the ionic silver dressing, and at baseline, 11 of 18 ulcers were infected with microbial flora typical of that associated with wound infection.

The study was designed to provide information on the safety of the ionic silver dressing to determine whether the addition of ionic silver to the SCF dressing posed any new safety risks. The study would also yield preliminary information on the performance of the dressing prior to larger studies. Only two adverse events, both mild in severity and spontaneously resolving, were considered to be related or probably related to the use of the dressing, indicating that addition of ionic silver introduced no new safety risks to the established safety profile of the SCF dressing. Within the limitations of the study size, and despite the presence of infection in 60 percent of the ulcers at baseline as evidenced by bacterial counts, over 90 percent of the ulcers showed either mild or marked improvement, and although no ulcers healed during the study, the changes observed in slough, granulation, and epithelial cover indicated that the ionic silver dressing provided an environment supportive to healing. The fact that progression to healing was not observed is not unexpected in view of the short duration of the study, which was designed primarily to assess safety. The reduction in ulcer area seen over the 28-day period can be considered indicative of a beneficial effect in light of clinical experience in the management of chronic ulcer wounds, particularly as no concomitant antibiotic therapy was required and favorable clinical outcomes were seen in all but one case. Interestingly, resolution of infection was seen in two of the ulcers infected at baseline and only one ulcer became infected during the study.

In this study, the ionic silver dressing had a mean wear time of approximately 3.5 days. This compares very favorably with the mean wear time of the SCF dressing.[18] The ionic silver dressing was judged to be easy to apply, and its use was associated with a significant reduction in the subjects’ perception of pain, both while the dressing was in place and upon its removal. The absence of leakage in a major proportion of dressing changes was remarkable in view of the numbers of ulcers that were infected at baseline and, therefore, expected to generate more exudate than noninfected ulcers.

The findings from this study also suggest that ionic silver has been incorporated into the SCF without undue alteration to the performance properties of the base dressing, and that the ionic silver dressing retains the wound management properties of the SCF dressing.[12,17,26] The small evaluable study population of 15 patients does not suffice to provide any valid scientific conclusions. However, the overall findings suggest that this new dressing might have potential clinical benefit in assisting the management of chronic ulcer wounds that are infected or at risk of infection.