Reepithelialization of partial thickness porcine burns treated with silver sulfadiazine, triple antibiotic or petrolatum

SUMMARY

Topical antibiotic application and occlusive dressings stand as the current standard of care for partial thickness burn (PTB) treatment. Silver sulfadiazine (SSD) is the most widely used topical antimicrobial agent for acute burn management worldwide. Despite its antimicrobial benefits, there is emerging evidence that SSD might delay wound reepithelialization. We hypothesized that when compared with a topical antibiotic ointment, wound reepithelialization with SSD would be delayed. We created 21 partial thickness burns on the backs of 3 anesthetized female domestic pigs (30kg) using an established vertical progression model. The burns were randomly assigned to topical treatment with SSD, triple antibiotic ointment, or petrolatum ointment applied daily for two weeks, followed by twice weekly for another 2 weeks with intermittent dressing changes. The primary outcome was percentage wound reepithelialization at day 21. Our study had 80% power to detect a 25-percentage difference in wound reepithelialization at day 21. All wounds were completely reepithelialized within 28 days of burn creation. At day 21 the mean (SD) percentages for wound reepithelialization were 73.6 (38.2), 93.2 (16.3), and 65.0 (48.9) in wounds treated with SSD, triple antibiotic, and petrolatum respectively, P=0.049. There was no significant difference among treatment groups in scar depth at day 28. Pseudoeschar formation and erythematous hypersensitivity reactions were common findings on gross analysis with SSD and topical antibiotics respectively. We conclude that SSD delayed reepithelialization of partial thickness porcine burns when compared with triple antibiotic ointment. There were no significant differences in scar depth among the groups.

Introduction

It is estimated that nearly 500,000 patients in the United States seek medical attention for burn injuries annually, with about 40,000 of those patients requiring acute inpatient hospitalization.¹ While a majority of burn injuries presenting to emergency departments are small and superficial, burn-related wounds account for about 3,300 deaths annually and pose a significant healthcare burden.^1,2 Burn injuries result in the loss of the skin barrier and require a robust treatment protocol involving frequent wound cleansing, topical antibiotics and dressings.³ The current standard of care for partial thickness burn (PTB) is the application of topical antibiotic agents or occlusive dressings. This treatment protocol facilitates a moist environment that promotes wound healing in burn injuries.⁴

Topical agents are a growing topic of interest as they provide pain control, promote healing through reepithelialization, and prevent wound infection and dehydration.⁵ Silver sulfadiazine (SSD) is a wide-spectrum antimicrobial topical agent developed over half a century ago at a time when burn-related wound sepsis was prevalent. SSD offers broad spectrum antimicrobial coverage against gram-positive and gram-negative bacteria as well as fungi.⁶ The introduction of SSD led to a significant reduction in burn-related sepsis incidence and mortality, and currently stands as the most commonly used topical agent worldwide.⁷ Recent systematic reviews of topical agents, however, suggest that the high concentration of silver in SSD leads to inhibition of keratinocytes and delays wound healing in non-infected burns.⁸

Much of the current literature on novel burn wound therapies has used SSD as the control comparator, while viable alternative topical agents have not been thoroughly compared to SSD.⁸ One such alternative is triple antibiotic ointment, which is also widely used for burn management. Triple antibiotic ointment contains antibacterial and antifungal agents embedded in a petrolatum-based ointment that, in theory, retains the antimicrobial properties of SSD without the negative impact on keratinocyte migration and proliferation associated with SSD.⁹ Considering the negative effects of SSD on wound healing, we believe that SSD should no longer be used as the control comparator. We propose that other topical antibiotics, such as triple antibiotic ointments, should be used instead. However, at present, triple antibiotic ointments have not been studied sufficiently to make a comprehensive claim regarding their performance in relation to SSD.⁸ Additionally, it is unclear whether the wound healing promoting properties of triple antibiotics are due to the antimicrobial agents or the petrolatum ointment.

The objective of this study was to compare the healing of partial thickness burns when treated with SSD versus triple antibiotic ointment using a validated porcine model. We also compared the healing of partial thickness burns treated with triple antibiotic ointment or petrolatum ointment alone. We hypothesized that partial thickness porcine burns treated with triple antibiotic ointment would reepithelialize faster compared to those treated with SSD, and that wound reepithelialization with triple antibiotics would be similar to petrolatum ointment alone.

Methods

Study design and setting

We conducted a prospective, randomized experiment in a porcine model to compare healing in partial thickness burns treated with SSD, triple antibiotic ointment, and petrolatum ointment. The porcine model is a widely validated burn model due to the high degree of similarity between human and porcine skin,¹⁰ making it an optimal model for burn wound analysis.¹¹ This study was approved by the Institutional Animal Care and Use Committee (IACUC) and was conducted in the Division of Laboratory Animal Research (DLAR) of an academic university hospital. Results reported from this study follow the recommendations of Animal Research: Reporting of in vivo Experiments (ARRIVE).¹²

Animal handling, sedation and anesthesia

We used three (3) female Yorkshire pigs weighing approximately 30-35 kg. The animals were acclimated for one week prior to the initiation of the experiment. The acclimation period included standard pig chow, continuous access to water, and exposure to recurring cycles of light (12 hours) and darkness (12 hours). The pigs were pre-medicated with an intramuscular injection of 0.01 mg/kg dexmedetomidine and 10 mg/kg ketamine post-overnight fast. Once prone positioning was achieved, general anesthesia was maintained with 2-4% isoflurane in room air administered via an endotracheal tube (for painful or prolonged procedures) or nose cone (for dressing changes). For painful procedures, the animals were administered 0.02 mg/kg buprenorphine intramuscular injection, and a 50 μg/hr transdermal fentanyl patch was applied to the animals’ dorsum. The experimental region (flank and back) was prepped using hair clippers and the skin was cleaned with soap and water followed by 70% isopropyl alcohol. At procedure completion, dexmedetomidine was reversed using an intramuscular injection of 0.1 mg/kg atipamezole. Animal handling and care was in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals national guidelines.¹³

Creation of burns

In each of the three pigs, we created twenty-one (21) deep partial-thickness burns for a total of sixty-three (63) burns. Partial thickness burns were created using a previously validated vertical progression model.¹⁴ This model utilizes a 150 gram aluminum bar preheated to 80°C applied to the pigs’ dorsum for 20 seconds (Fig. 1, left). The necrotic epidermis was removed using the blunt end of tissue forceps to simulate the formation and removal of blisters (Fig. 1, right), a feature of burn injuries absent in the porcine model.¹⁵ The result was 2.5 cm by 2.5 cm deep partial thickness burns, involving about half the dermal layer.

Fig. 1.

Vertical progression porcine burn model. Left: creation of deep partial thickness burn on porcine dorsum using a preheated aluminum bar. Right: removal of the necrotic epidermis to simulate blister rupture.

Treatments

In each of the three pigs, the burns were divided into three (3) groups of seven (7) partial thickness burns and each group was randomly assigned to one of three treatments in a 1:1:1 ratio (Fig. 2, left). Each pig had three groups of seven burns treated with petrolatum ointment, triple antibiotic ointment (Bacitracin Zinc, Neomycin Sulfate, Polymyxin B Sulfate; Taro Pharmaceuticals, Hawthorne, NY), or SSD (Ascend Laboratories, LLC, Parsippany, NJ). Burns were covered with their assigned respective topical agent daily for two weeks followed by 2-3 applications per week for another two weeks. On each treatment application, the burns were dressed using a non-adherent gauze (Telfa, Cardinal Health, Dublin, OH) that was stapled to the skin to prevent cross contamination between treatment groups (Fig. 2, right). The dressings were then wrapped with outer gauze wraps (Sof-Form, Medline Industries, Inc., Mandekein, IL) and adhesive dressings (Tenoplast, BSN Medical S.A.S, Vibraye, Frane) to avoid premature dislodgement of the dressings and treatments. The wounds were cleaned with soap and water and photographed at each treatment application. Full thickness 4-8 mm punch biopsies were obtained on dressing change days 10, 14, 17, 21 and 28, as described below.

Fig. 2.

Burn treatment and dressings. Left: deep partial thickness burns treated with petrolatum ointment, triple antibiotic, and silver sulfadiazine. Right: dressing application with isolated treatment groups.

Wound assessments

Burns were visually inspected for signs of wound reepithelialization and digital photographs were obtained on the following days: 1-4, 7-11, 14, 17, 21, 24 and 28. Wounds were considered reepithelialized when they appeared covered with an opalescent layer of epidermis on visual inspection and were not wet to touch with tissue paper. Signs of erythema, edema and purulent discharge were noted if present.

Histopathology

An 8-millimeter (mm) punch biopsy was taken from the unburned flank of each pig as a control for uninjured skin. Full-thickness 4-mm punch biopsies (Integra, York, PA) were obtained from all wounds at days 10 (upper left), 14 (lower right), 17 (lower left), and 21 (upper right) for determination of percentage wound reepithelialization. The biopsy sampling site was standardized to approximately 10 mm away from the burn edges in each of the four corners to prevent reepithelialization from the uninjured margins. On day 28, an additional full thickness 8mm punch biopsy was taken from the center of each individual burn for scar depth analysis.

Biopsy samples underwent histomorphometric analysis by a board-certified dermatopathologist blinded to treatment assignments. Samples were bisected and then fixed in 2% formaldehyde. Tissue sections were stained with Hematoxylin and Eosin using the internal camera of a EVOS microscope (ThermoFisher Scientific, Waltham, MA).

Outcomes

The primary outcome was percentage wound reepithelialization 21 days after burn creation. Percentage wound reepithelialization was calculated as the length of the neo-epidermis divided by the total epidermis length of the tissue specimens multiplied by 100. This has been shown to have excellent interobserver agreement.¹⁵ A secondary outcome was scar depth 28 days after injury. Scar depth was calculated as an average of three measurements obtained from the center and periphery of the wounds. Scars were defined histologically as thin, horizontal-oriented collagen fibers, as confirmed with visualization under polarized light.¹⁶

Data analysis

Continuous data are presented as means and standard deviations. Categorical data are presented as numbers and percentages frequency of occurrence. Percentage wound reepithelialization was compared among groups with analysis of variance (ANOVA). Post hoc analysis was performed using the Tukey test. The significance level was at 0.05. A sample of 21 wounds in each group had a power of 80% to detect a difference of 25 percentage points in wound reepithelialization between the groups.

Results

We examined three Yorkshire pigs with twenty-one partial thickness burns for a total of sixty-three burns (n=63). After burn creation and removal of the necrotic epidermis, burns were randomly assigned to treatment with petrolatum ointment (n=21), triple antibiotic ointment (n=21) and SSD (n=21). Study outcomes are summarized in Table I. On macroscopic analysis, appreciable pseudoeschar formation was noted in wounds treated with SSD. On the other hand, many of the wounds treated with the triple antibiotic ointment demonstrated erythematous allergic reactions that resolved by the end of the study.

Table I.

Study outcomes

Outcome	SSD	Triple antibiotic	Petrolatum	P Value
Mean (SD) % Reepithelialization
Day 10	10.4 (20.3)	3.3 (12.1)	7.9 (13.9)	0.34
Day 14	45.1 (49.1)	48.1 (42.3)	37.4 (44.1)	0.74
Day 17	81.9 (38.1)	89.7 (27.0)	86.3 (20.0)	0.70
Day 21	73.6 (38.2)	93.2 (16.3)	65.0 (48.9)	0.049
Mean (SD) mm Scar depth
Day 28	3.9 (1.1)	3.4 (0.9)	4.2 (0.9)	0.07

Histological reepithelialization

Percentage reepithelialization was measured using periodic punch biopsy samples obtained throughout the experimental course. The mean (SD) percentage of wound reepithelialization at days 10, 14, 17 and 21 for wounds treated with SSD, triple antibiotic ointment, and petrolatum is summarized in Table I. We found a significant difference in wound reepithelialization on day 21 (P=0.049) (Fig. 3). The mean day 21 percentage reepithelialization (SD) for partial thickness burns treated with SSD, triple antibiotic ointment and petrolatum were 73.6% (38.2), 93.2% (16.3) and 65% (48.9) respectively. All wounds completely reepithelialized by day 28 post-injury.

Fig. 3.

Representative micrographs of mean percentage wound reepithelialization 21 days after injury

Scar depth

Scar depth was measured histologically using a punch biopsy sample obtained on day 28 post-injury. The mean (SD) scar depths of partial thickness burns treated with SSD, triple antibiotic ointment and petrolatum were 3.9 mm (1.1), 3.4 mm (0.9), and 4.2 mm (0.9), respectively (Fig. 4).

Fig. 4.

Representative micrographs of day 28 scar depth

Discussion

Infection prevention is a major challenge to the treatment of acute thermal injuries.¹⁷ Early intervention with a topical antimicrobial agent and wound coverage is therefore critical to promoting wound healing. Recent literature suggests that topical silver, especially in high concentrations, delays wound healing and should not be the standard of care for uninfected wounds.¹⁸ Our study demonstrates that partial thickness burns treated with triple antibiotic ointment reepithelialize faster than those treated with SSD in a porcine model, confirming our initial study hypothesis. Since SSD delays reepithelialization, we do not believe that SSD should continue to be used as the control treatment in future preclinical and clinical studies. We also found that wound reepithelialization was faster with the triple antibiotic ointment compared with the petrolatum ointment alone. This suggests that use of topical petrolatum, without the addition of the topical antimicrobial agents, is inferior and should not be used.

We also found a high rate of pseudoeschar formation with SSD, often seen in clinical use. A major disadvantage of pseudoeschar formation is that it makes wound assessment for closure and infection difficult. As a result, some practitioners may overestimate the rate of wound infection, potentially leading to overuse of systemic antibiotics with the risk of allergic reactions and selection for resistant organisms. We also noted local erythematous reactions surrounding burns treated with the triple antibiotic. This has also been described clinically and suggests that alternative topical antibiotic preparations, such as bacitracin, may be preferable.

While reepithelialization was slower in wounds treated with SSD, all wounds were completely closed by 28 days and there were no significant differences among the groups in scar depth. It is possible that the study was underpowered to detect differences in scar depth. It is also possible that longer-term follow up may have detected differences in scarring among the groups.

The importance of early wound closure is suggested by several prior studies demonstrating higher rates of hypertrophic scarring in burns that fail to close by 21 days after injury.¹⁹ Indeed, early excision and grafting of burns predicted to require more than 2-3 weeks to close, is the current standard of care. Timely wound closure is also essential for the prevention of wound infection in addition to reducing scarring and functional disability.²⁰ Delayed wound healing is associated with a significantly higher risk of mortality.²¹ Thus, topical agents that promote fast wound healing are critical for quick patient recovery and infection prevention.

Our study has several notable limitations. Of note, this study used pigs and not humans. While the porcine model is a validated model for studying burns, there are appreciable anatomical and physiological differences between humans and pigs. There is less vasculature in the pig dermis and its hair follicles, and the endothelium of cutaneous blood vessels does not produce alkaline phosphatase.^22,23 There is also a significant difference in the presence of sweat glands. Pigs have predominantly apocrine sweat glands with confined eccrine sweat glands.²³ Humans have widely distributed eccrine sweat glands that serve as reservoirs of keratinocytes during wound epithelialization.²⁴

Despite these differences, we believe the many similarities between pig and human wound healing mechanisms including reepithelialization make the porcine model the optimal preclinical model to study topical agent treatments for burns. In comparison to other animal models, pigs are physiologically and metabolically more similar to humans. Pigs are relatively hairless, have a relatively thick epidermis, distinct rete pegs, dermal papillae, and dense elastic fibers in the dermis.^22,25Additionally, pigs and humans share the following similarities: epidermal turnover time, keratinous proteins and lipid composition of the stratum corneum, biochemical properties of collagen, and skin immune cells, including dendritic cells.^26,27,28 However, we recognize that our results may not translate to humans, and future studies comparing triple antibiotic ointment (or other formulations of antimicrobial agents) to SSD are needed to explore the comparison between topical agents in human burns.

Our study is also subject to sampling bias since the rate of reepithelialization may not always be uniform across the entire wound. We used 4 mm punch biopsies from standardized wound sites to measure wound reepithelialization. Additionally, we used a relatively small sample size and burn size, and results may differ with larger burns. Thus, we cannot exclude the possibility this study was underpowered to detect smaller yet clinically significant differences among the groups in other study outcomes such as scar depth.

Conclusions

We found that partial thickness porcine burns treated with triple antibiotic ointment reepithelialized faster than similar burns treated with SSD and petrolatum ointment. Scar depth was not affected by topical ointment treatment. We confirmed that SSD does delay wound reepithelialization in the porcine model. Topical agents, such as triple antibiotic ointment or other topical antimicrobial ointments available on the market should be considered as the standard comparator for future research on burn wound healing.