Preliminary Results of a Topically Applied Bacteria-specific Antibiotic Gel to Improve Wound Healing

Introduction. Advances in molecular diagnostic medicine have allowed for more rapid, accurate, and comprehensive methods for identifying microorganisms in a chronic wound, which led to the development and use of a tailored topical antibiotic gel aimed at treating this bioburden. Objective. This is a retrospective chart review evaluating the authors’ early experience with the use of bacteria-specific antimicrobial gel therapy on chronic lower extremity wounds that have not responded to standard therapy. Materials and Methods. All patients in the study were treated with a topical gel along with standard of care modalities. Results. A total of 48 patients with 76 wounds (50/76 venous leg ulcers; 65.8%) were identified and analyzed. Of the 48 patients, 11 (22.9%) had complete wound closure at a mean of 101.6 days of treatment. The number of wounds decreasing in size improved from 45.3% to 77.6% after gel therapy. An analysis of all wounds showed an increase in size by 0.7% weekly with the topical gel; however, a mean weekly healing rate of 6.5% was seen when analyzing only the subset of wounds that decreased in size. Conclusions. Although a minor improvement of weekly healing rate was seen for a subset of the wounds, the overall wound closure rate was low.

Chronic wounds are an ever-growing health care concern, especially in the aging population. These wounds affect about 5% of the population > 80 years of age and up to 3% of the population over the age of 60.¹ Care for chronic wounds is estimated to cost 2% to 4% of the total health care budget in Western countries.^1,2 Standard wound care treatment modalities focus on facilitating healing by debridement, offloading, correcting arterial and/or venous insufficiency, topical therapy, compression therapy, nutrition therapy, and management or prevention of infection. There is continuous research on new modalities to improve overall healing rates of these challenging wounds. 

Chronic wounds are defined as those that do not progress through the healing process in the usual time frame, which could range from 1 to 3 months depending on the wound. There are many factors that inhibit the ability of a wound to heal: poor host environment, infection, biomechanics, poor arterial or venous status, and biofilm.^3-5 Biofilm has become an increasingly important topic in wound care, as it may be a major barrier to wound healing.^6-8

Biofilms are present in up to 60% of chronically open wounds versus only 6% of acute wounds.⁶ Biofilm is defined as an aggregate of microorganisms that live on a surface and are highly resistant to host immunity and antibiotics.^8-10 Bacteria within biofilms have lower metabolic rates, causing them to be more resistant to antibiotics as many antibiotics target metabolically active bacteria.^7,8,11 Biofilms have unique properties when compared to their planktonic counterparts; they adhere to surfaces, secrete substances for protection from the environment, exhibit varying states of metabolic activity and physiology, and grow and regenerate at very quick rates.^6,8,10,12 It only takes 10 hours for a biofilm to grow, and it can repopulate within 24 hours after debridement.^7,13 

Sharp debridement of a chronic wound is an essential treatment modality and serves to remove biofilm from a wound bed.^10,14-17 Along with debridement, topical antimicrobials have commonly been used as a conservative treatment modality. Several products aiming to prevent and treat mild or moderate local infections exist, including triple antibiotic ointments, iodine-based products, and silver-based products. The proven usefulness of these products remains unclear in the literature.^8,18-20 There is new interest in patient-specific topical antibiotic therapy for the treatment of chronic wounds and their biofilm. The literature suggests that these topical antimicrobials can be specifically tailored to a patient’s unique wound biofilm.^10,21-23 

Many bacteria within chronic wounds grow poorly on growth media or not at all. Only about 2% of all bacterial species can be cultured in the microbiology lab, and only some of these will grow within the given 24-hour period they are tested.^23-25 The use of polymerase chain reaction (PCR) within molecular diagnostic microbiology is being recognized as a more accurate and efficient method for identifying the bacterial composition of chronic wounds. By amplifying the unique deoxyribonucleic acid within a wound, the microorganisms can be accurately identified as well as quantified.²³ This method requires only a small amount of tissue and takes only a few hours. Identifying which of those bacteria or combination of bacteria are actual factors affecting healing is currently the subject of much research. There is literature suggesting the results of identification of the bacteria may help guide appropriate therapies including wound-specific antimicrobial gels.^21,26 When used in combination with routine wound debridement, medical optimization of the patient, and standard wound care modalities, these gels targeting the biofilm of chronic wounds have suggested a potential for improving wound healing.^21,26

The aim of this retrospective study is to evaluate the authors’ early experience with the use of bacteria-specific topical antimicrobial gel in the treatment of chronic lower extremity wounds at a tertiary wound care center. 

A retrospective analysis was performed for all patients (aged 18–97 years) who were treated with the bacteria-specific gel from August 2011 through October 2013 from MedStar Georgetown University Hospital (Washington, DC). This study was approved by the Biomedical Institutional Review Board of Georgetown University Medical Center. 

Patients were included if they received an application of the gel on a nonhealing wound during the selected time period. It was at the discretion of the clinician to determine who received the gel. Electronic medical records were reviewed for patient demographics, including age, sex, and race. Comorbidities were documented as well as etiology and location of the wound. Wound measurements were recorded at the initial presentation to the clinic, the start of the topical antibiotic gel therapy, and the stop date of the therapy. Wound measurements were obtained via a standard metric ruler performed by the clinician. Wounds were considered healed when there was full wound closure with 100% epithelialization and no remaining measurable wound. This was determined by the clinician evaluating the patient; patient and clinician were not blinded to the treatment. Total length of time of gel use was recorded as well as the change in wound size from the start to the end of treatment. 

Normally discarded tissue was collected from the wound during debridement and sent to a laboratory (PathoGenius Laboratories, Lubbock, TX) where the samples were analyzed through a PCR process. Bacteria present in the samples were speciated and the relative amounts were quantified. The laboratory then made recommendations for specific antibiotics to which the bacteria were susceptible, and a separate compounding pharmacy (Southeastern Medical Compounding, Savannah, GA) produced a gel cocktail that was sent back to the clinical site. The personalized topical antibiotic gel was composed of a singular or a combination of antibiotics that are approved by the United States Food and Drug Administration. The antibiotic gel was then applied once daily to the wound site with a gauze dressing by the patient, family member, or visiting nurse.  A secondary dressing was not standardized as this study was conducted retrospectively. Throughout the treatment period, when applicable, patients also received standard of care treatment as follows: offloading, medical management (eg, blood glucose control), compression therapy, nutritional support, oral antibiotic therapy (specific to either qualitative culture results or empirically selected), and vascular intervention. Oral antimicrobials were prescribed only when medically necessary and in accordance with standard practice if there were signs or symptoms of an acute infection, which includes redness, drainage, malodor, or warmth. It was not routine to start oral antimicrobials. Ulcerations were mechanically debrided in accordance with this institution’s standard of care during outpatient visits and at the discretion of the treating clinician. 

Patients were excluded for the following reasons: the antimicrobial gel was never applied after being dispensed, the gel was applied for fewer than 14 days, the patient was lost to follow-up, there were no accurate start or stop dates recorded for the gel, the patient underwent surgical intervention during the course of treatment (surgical debridement, split-thickness skin grafting, or amputation), and an allergic reaction to the gel. An allergic reaction to the gel was defined as a full body rash, with the antibiotic components of the gel then listed as allergies for a patient.

The success of the PCR-directed gel was defined by its ability to close the patient’s treated wound(s) and the time required to achieve that closure. An additional parameter analyzed was the rate of wound healing. A search identified 93 patients with 249 wounds that had application of the bacteria-specific antimicrobial gel from August 2011 to October 2013. After initial exclusions, 67 patients with 130 wounds remained (Figure 1). Major and minor outliers were then calculated, which excluded patients whose wounds measured > 48.3 cm² and < 1.0 cm². By calculating major and minor outliers for weekly healing rates, wounds were excluded if they were healing at a rate of ≥ 15% weekly or if they were increasing in size by ≥ 22% weekly. One patient was excluded for a wound on the abdomen, and there were 48 patients with 76 wounds left for final analysis (Figure 1).

Statistical analysis

All statistical analyses were performed using SPSS Statistics Version 21 (IBM, Armonk, NY). Outliers for initial wound size and initial rate of weekly wound healing prior to gel application were removed. T tests, chi-square tests, and Fisher’s exact tests were performed on all baseline variables to assess for confounders associated with the weekly wound healing rate prior to gel application (ie, patient comorbidities, wound etiology, wound location, gender, race, and age). In addition, chi-square and Fisher’s exact tests were performed to assess for confounding variables associated with wound outcome (ie, patient comorbidities, wound etiology, wound location, gender, race, and age in association with wound status, open or closed, after treatment concluded). Next, a paired sample t test was performed to compare the patient’s mean weekly rate of wound healing before gel application to the patient’s mean weekly rate of healing after gel application. 

The average age of patients in the study was 70 ± 16 years with an average body mass index in the obese category at 29.9 ± 8.5 (Table 1). Majority of patients (64.6%) were male.  African Americans made up 47.9% of the patient population with Caucasians accounting for another 31.3%. Multiple patient comorbidities were identified (Table 1). Of note, only 9 (18.8%) of the patients had diabetes; however, 27 (56.3%) were diagnosed with a venous disease, including varicose veins, venous insufficiency, or deep venous thrombosis. This is consistent with the wound etiology outlined below.  

Mean starting wound size was 16.3 cm² (± 28.3 cm²), and mean wound depth was 0.2 cm (0.1 to 1.1 cm). Etiology and anatomic location of the 76 wounds are noted in Table 2. Of the 48 patients, 19 (39.6%) had more than 1 lower extremity wound that was included in the study. An overwhelming majority of the wounds (50/76; 65.8%) were venous ulcers; only 7 (9.2%) were diabetic foot ulcers. Consistent with the etiologies, most wounds were located on the lower leg (35/76; 46.1%) or ankle (27/76; 35.5%). 

Sixty-four (84.2%) wounds had measurements and wound history documented prior to the start of the bacteria-specific antibiotic gel therapy. Wound duration ranged from 1 month to 5 years, with a median duration of 9.4 months. On average, wounds were treated with the antibiotic gel for 111.5 days. Of the 76 wounds, 17 (22.4%) were healed at the end of the antibiotic gel treatment with an average time to wound closure of 101.6 days. Prior to gel application, 29 (45.3%) wounds had been decreasing in size. After the start of therapy, the number of wounds decreasing in size increased to 59 (77.6%; Table 3). 

On average, wounds were increasing in size by a mean weekly rate of 0.7% before the gel was applied (Table 4). After gel application, this worsened to an overall weekly increase of 2.3%. Wounds that increased and those that decreased in size were then examined separately.  Wounds were increasing in size by 3.3% weekly prior to the gel; after gel application, the wounds continued to increase in size by 28.3% weekly.  Wounds that decreased in size were decreasing by 2.4% weekly prior to the gel and improved to achieve a weekly healing rate of 6.5% after implementation of the bacteria-specific antibiotic gel treatment. 

Race, wound duration, and patient age significantly correlate with the mean rate of weekly wound healing prior to gel application (P = .031, P < .001, and P < .001, respectively; correlations of -0.223, 0.035, and 0.62, respectively). However, race, wound duration, and patient age are not significantly correlated with the mean rate of weekly wound healing after antimicrobial gel treatment (P = .480, P = .486, and P = .382, respectively; correlations of 0.006, 0.005, and 0.038, respectively). There was no significant difference in the mean weekly rate of wound healing before (µ = 0.749; standard deviation [SD] = 4.25) as compared with after (µ = -0.750; SD = 23.3) bacteria-specific antimicrobial gel implementation (P = .625).

 Varying healing rates have been seen with numerous other topical antimicrobial therapies, including silver, honey, antibiotic ointments, and iodine products.^{8,18-20,27-30} The present study evaluated the use of a customized topical antimicrobial gel in chronic lower extremity wounds. These results suggest there was an overall improvement in the number of wounds that decreased in size, but that the weekly healing rate did not improve greatly after the initiation of the antimicrobial gel.

About half of the patients (48/93) and 30.5% (76/249) of the wounds that met initial eligibility criteria were left for analysis after exclusions were met. Majority of the wounds in this study were venous leg ulcers resistant to healing despite many previous treatments and local wound care. More than half of the wounds (36; 56.3%) had been increasing in size prior to the therapeutic gel, and, after the treatment, 59 (77.6%) had a decrease in overall size. Despite this, there was not a large improvement in the rate of wound healing. When all 76 wounds were included, they were in fact worsening by a weekly rate of 2.3%. In the study herein, only 17 (22.4%) wounds were healed at the end of the study period. 

In 2011, Dowd et al²¹ performed a retrospective, heterogeneous cohort study on 1378 patients who were treated before and after initiation of bacteria-specific antimicrobial therapy. The present results differ from this study,²¹ and the discrepancy can perhaps be attributed to differences in patient demographics, wound duration, wound size, and the years in which the study was conducted. The comparison study²¹ was unable to use wound duration or size as a data point, because the data were incomplete for these values. This makes it challenging to determine the average size or duration of the wounds in that study and could significantly affect healing if they are acute or small wounds. Several wounds in the study herein were months to years old, which would make any type of treatment more challenging and could have negatively impacted the results. In addition, this study was not a comparative study. Dowd et al²¹ had 3 cohorts that they were able to compare. Their results showed that 48.5% of patients healed in the standard of care group treated with systemic antibiotics, 62.5% healed in the group treated with systemic antibiotic therapy guided by molecular diagnostics, and 90.4% healed in the group treated with personalized topical antimicrobial therapy guided by molecular diagnostics. These results differ significantly from the present study that only saw 22.4% of patients heal with the use of bacteria-specific antimicrobial gel with a mean time to closure of 101.6 days compared with their²¹ documented 28 days.

Molecular bacterial diagnostics is a growing and evolving aspect of wound care. Classic culture technology relies on bacterial cells to grow and produce colonies on agar plates and often only grow 1 or 2 of the bacteria that may be present within a chronic wound.²⁷ Molecular biology technology using PCR techniques has been developed to allow for more accurate identification and quantification of organisms in a sample.²² This has led to bacteria-specific topical antibiotic therapy as introduced by Dowd et al²¹ and Wolcott et al.²⁶ Unfortunately, there is still no way to determine which bacteria or combination of bacteria may be contributing to continued infection, inflammation, or nonhealing. One would hope that as larger databases are collected, there will be a better understanding of what bacteria or combination of bacteria correlate with wound duration and/or nonhealing.²² As molecular diagnostics continue to grow, medical professionals will undoubtedly see improvements in patient-specific treatment options. 

There are significant limitations to this study, including its lack of generalizability, retrospective design, and reliance on medical records. The accuracy of data collected relies on proper documentation at the point of care. It also relies on the accuracy of wound measurements, which may vary depending on the clinician taking the measurement. Further, the determination to implement the bacteria-specific antibiotic gel therapy was solely based on clinician preference. Thus, significant patient selection bias exists. Criteria for the selection of patients who received this treatment may have varied among clinicians. 

Another limitation is the lack of control for other conditions and treatments in this study. A major factor would be how well-controlled a patients’ diabetes is as could be determined by hemoglobin A1_C. This was not collected
as a data point in this study and could significantly challenge the wound-healing process. In addition, while all patients received standard of care, the regimen was not standardized. Some patients may have been receiving oral antibiotic therapy in conjunction with the topical antimicrobial therapy. Oral antibiotics could have significantly influenced the healing rate. 

The results also lack generalizability as the majority of the wounds in the study were venous ulcers. Patients with venous disease who were deemed to have a surgically correctible disease were treated by various vascular surgeons with no standardized surgical treatment. While the vascular status was assessed at clinic visits and patients were sent to vascular surgery for referral when necessary, it was not factored into this study as a data point. Vascular status may have significantly delayed healing in particular patients, and this should be recognized as a limitation. 

Likewise, the type and number of bacteria species present in a wound may have influenced the healing potential. Some wounds may have had multiple bacteria present. While the compounded topical gel included 3 antibiotics that targeted a majority of the bacteria, it may have missed those that were clinically significant. Similarly, the healing rates may have been significantly affected by the recalcitrant nature of the wounds in this study as the institution used is a tertiary referral center for wound care.

 Bacterial molecular diagnostics shows growth and promise for its use in the treatment of chronic lower extremity wounds. Biofilm may play a critical role in the management of wounds, thus treatment focused at targeting biofilm should be more closely analyzed. The results reported suggest that bacteria-specific antimicrobial gel may marginally improve the wound healing rate in chronic nonhealing wounds but does not provide a robust effect. Future studies should be aimed at better understanding the role of biofilm in wound healing and in chronic wounds. Prospective studies evaluating the efficacy of various therapies addressing biofilm in chronic wounds will help define what role, if any, topical therapy could play. 

Anagha Kumar, MS, assisted in editing and statistical analysis. Kelly A. Powers, DPM, MS, and Charles Long, BS, assisted in data compilation. They received no compensation for their assistance. Consent was obtained from the above parties for the above-written statements.  

Affiliation: Center for Wound Healing and Hyperbaric Medicine, MedStar Georgetown University Hospital, Washington, DC 

Correspondence:
Caitlin S. Zarick, DPM
Center for Wound Healing and Hyperbaric Medicine
MedStar Georgetown University Hospital
3800 Reservoir Road NW
Washington, DC 20007
Caitlin.s.zarick@gunet.georgetown.edu

Disclosure: The authors disclose no financial or other conflicts of interest.