Introduction
Wounds on the distal aspects of the limbs are common in horses and often involve extensive loss of skin, tendon damage, bone exposure, and an important degree of contamination that precludes primary closure and necessitates second-intention healing.1 Bacterial colonization and infection remain important factors capable of delaying healing of these types of wounds.2 Because the widespread use of systemic and topical antimicrobials has resulted in increasing numbers of antimicrobial-resistant bacterial strains, such as MRSA, Pseudomonas aeruginosa, vancomycin-resistant Enterococcus faecalis, and extended-spectrum β-lactamase–producing bacteria,2,3 the judicious use of dressings containing antiseptics or other components with antibacterial activity (eg, honey) rather than antimicrobial drugs is encouraged.4
For several available wound dressings and other compounds, the influence on microbial growth in contaminated wounds of horses has been evaluated either in a clinical situation or an experimental setting.5,6,7,8,9 However, large studies comparing the influence of multiple dressings or compounds on microbial growth are scarce. This type of evaluation can be performed with an in vivo experimental design, but such studies are often expensive and ethical issues may arise. In vitro studies10,11,12,13,14 performed to evaluate bacterial isolates exposed to various concentrations of an active compound or the effects of applying antimicrobial dressings on bacterial cultures grown in Petri dishes have provided preliminary information. However, factors such as interference from tissue proteins and the formation of tissue exudate that may dilute the active compound in a dressing are not accounted for in such experiments. Ex vivo wound models created with discarded human15,16,17,18 or porcine skin,19 some with simulated skin infection, and monitored for local immune responses.17 These models can be used to evaluate antimicrobial substances, but the wounds created have been either burn wounds or superficial traumatic lesions15,16,17,18,19 that would not be representative of typical equine wounds.
To the authors’ knowledge, there is no validated full-thickness wound model created with equine skin. However, an inoculated full-thickness perfused wound model created with nonviable equine skin has been used by our group to evaluate the antibacterial effects of various foams used with negative pressure wound treatment.20 The purpose of the study reported here was to use the same model to simultaneously quantitate the effects that various dressings used for treatment of equine wounds allowed to heal by second intention4 have on microbial growth. Dressings selected for the study contained activated charcoal, boric acid, cadexomer iodine, calcium alginate, manuka honey, nanoparticle silver, polyhexamethylene biguanide, or no active substance. We hypothesized that not all examined wound dressings would equally inhibit or reduce the number of bacteria in this wound model and that the effects of a given wound dressing would vary depending on the microorganism involved.
Materials and Methods
Equine perfused wound model
A wound model developed by our research group20 was adapted for use in the present study. Briefly, 16 cadavers of horses euthanized for clinical reasons other than colic or septicemia were used. Horses treated with antimicrobial drugs were excluded from the study. All owners consented to use of the cadaver for scientific purposes.
The horses were sedated with a mixture of butorphanol tartrate and detomidine hydrochloride, and general anesthesia was induced with midazolam and ketamine. An IV injection of an embutramide-containing euthanasia solution was used to euthanatize the horse. Shortly after this injection, heparin (150 U/kg, IV) was administered. After death was confirmed, the abdomen was aseptically prepared. A 25 × 38-cm musculocutaneous flap composed of the skin, subcutaneous tissues, external rectus sheath, and rectus abdominis muscle was collected lateral to the linea alba. The flap was placed in a sterile plastic container and perfused with sterile saline (0.9% NaCl) solution (0.5 mL/min) delivered via an 18-gauge catheter inserted into the superficial epigastric artery to prevent possible dehydration and simulate the possible dilutional effects of tissue exudate. The skin was sutured to the underlying muscles around the outer edges to maintain tension. Eight 3 × 3-cm full-thickness skin wounds were created by removing the skin, subcutaneous tissues, and external rectus sheath to reveal the underlying rectus abdominis muscle. As the skin wounds were being created, a bacterial stock solution of MRSA or P aeruginosa (concentration, 1 × 106 CFUs/mL in sterile PBS solution) was prepared. Both bacterial stocks were originally obtained from equine wounds and were cultivated and stored in a bacterial stock library at the University of Ghent. In 8 flaps, the 8 wounds were each inoculated with 1 mL of MRSA stock solution, and in the remaining 8 flaps, the 8 wounds were each inoculated with 1 mL of P aeruginosa stock solution. Perfusion was temporarily discontinued to avoid dilution of the bacterial stock solution and aid the bacterial colonization of the wounds. The flaps were incubated for 1 hour at 37°C and 5% CO2. After the incubation period, perfusion of the flap with saline solution was restored.
Dressing application
A sterile gauze pad was used to remove excess fluid from the surfaces of wounds, and the surrounding skin was decontaminated with 70% ethanol as a precaution to avoid accidental contamination of the wound by any commensal organisms. Each wound on a given flap was assigned a different dressing. For each of the 8 repetitions/microorganism, the location of each dressing was assigned according to a rotation to exclude bias associated with proximity to the epigastric artery. The dressings used were sterile products that contained activated charcoal,a boric acid,b cadexomer iodine,c calcium alginate,d manuka honey,e nanoparticle silver,f or polyhexamethylene biguanide.g Sterile nonadherent gauze dressingh was used as a control. The dressings were cut in an aseptic manner to fit the 3 × 3-cm wounds (Figure 1).
Photograph of a 25 × 38-cm musculocutaneous flap collected from the aseptically prepared abdominal region of 1 of 16 horses immediately after euthanasia and used in a study to quantitate the effects that various wound dressings have on microbial growth in a perfused equine wound model. The flap was placed in a sterile container, and an 18-gauge catheter was inserted into the superficial epigastric artery to perfuse the tissues with sterile saline (0.9% NaCl) solution during the observation period. Eight 3 × 3-cm wounds were created by removing the skin, subcutaneous tissues, and external rectus sheath, revealing the underlying rectus abdominis muscle. Each wound was inoculated with 1 × 106 CFUs of MRSA or Pseudomonas aeruginosa. After a 1-hour incubation period (when perfusion was temporarily stopped), a dressing containing activated charcoal, boric acid, cadexomer iodine, calcium alginate, manuka honey, nanoparticle silver, or polyhexamethylene biguanide or a control dressing (sterile nonadherent gauze) was applied to each wound according to a rotational design. Each dressing was cut in an aseptic manner to fit the wound surface. An 8-mm punch muscle biopsy specimen was obtained from each wound at baseline (0 hours; immediately after incubation and before application of dressings) and 6, 12, 18, and 24 hours after dressings were applied for MBL determination.
Citation: American Journal of Veterinary Research 82, 3; 10.2460/ajvr.82.3.189
Microbiological analysis
From each wound, an 8-mm punch muscle biopsy specimen was obtained immediately after incubation and prior to dressing application (baseline) and 6, 12, 18, and 24 hours after application of the dressings. After baseline, each dressing was lifted away for biopsy sample collection, and the same dressing was repositioned afterward. The biopsy specimens were stored in sealed polypropylene tubes at 4°C and processed ≤ 12 hours after collection to determine the number of CFUs per milligram of tissue. This was done by weighing the sample, subsequently homogenizing the sample with a disposable tissue grinder in 1 mL of PBS solution, and performing serial dilution of the samples.20 The diluted samples were spot plated and incubated for 24 hours at 37°C and 5% CO2 before the number of colonies was determined on a countable plate by 1 author (EDC). Columbia agar plates supplemented with colistin and nalidixic acid were used for MRSA-inoculated samples, and MacConkey agar plates were used for P aeruginosa–inoculated samples.
Statistical analysis
The MBL (mean number of CFUs per milligram of tissue for 8 repetitions/isolate for each dressing group) was determined and used as a measure of microbial growth. A lognormal distribution was acquired by transformation of the results to a log10 scale. If no microbial growth was detected, the missing value was replaced by multiplying the detection limit of the organism for the method used (0.4 for both MRSA and P aeruginosa) by 0.5.21 For wounds treated with each dressing type (ie, each dressing group), the MBL for a given isolate 6, 12, 18, and 24 hours after the dressing application was compared with that at baseline. The MBLs were also compared among groups at each time point. A 2-way repeated-measures ANOVA with Greenhouse-Geisser correction and a Tukey multiple-comparisons test were used for statistical analysis. The log-transformed MBL was set as the dependent variable. The bacterial species, dressing group, and sample acquisition time were set as fixed effects. Values of P ≤ 0.05 were considered significant. In addition, a post hoc power analysis was performed (α = 0.05). Statistical softwarei was used to perform all analyses.
Results
The MBLs for wounds inoculated with MRSA and P aeruginosa did not differ significantly at baseline, with a mean difference of −0.224 log10 CFUs/mg of tissue (95% CI, −1.134 to 0.687 log10 CFUs/mg). The MBLs of MRSA and P aeruginosa for the control dressing groups both increased during the observation period, with a nonsignificant mean difference of 0.496 log10 CFUs/mg (95% CI, −0.415 to 1.406 log10 CFUs/mg) between the control group MBLs for the 2 organisms at the 24-hour time point.
MRSA
The mean differences in MRSA MBL between dressing groups and the corresponding 95% CIs were summarized for each time point (Table 1). Comparison with results for the control group revealed that the MBLs for the cadexomer iodine group were significantly lower at all time points after baseline (Figure 2). The MBLs for the boric acid and nanoparticle silver groups were significantly lower than those for the control group at the 6− and 12-hour time points. The MBLs for the manuka honey group were significantly lower than those for the control group at the 18− and 24-hour time points.
Mean (95% CI) difference in MRSA MBL (log10 CFUs/mg of tissue) over time for wounds treated with various dressings in a perfused equine wound model.
| Comparison group | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Group | Time (h) | Calcium alginate | Boric acid | Activated charcoal | Control* | Cadexomer iodine | Manuka honey | PB | Nanoparticle silver |
| Calcium alginate | 0 | — | 0.327 (−0.517 to 1.171) | –0.041 (−0.698 to 0.616) | 0.215 (−0.521 to 0.951) | 0.042 (−0.664 to 0.747) | 0.172 (−0.903 to 1.246) | –0.206 (−0.887 to 0.476) | 0.091 (−0.905 to 1.087) |
| 6 | — | 2.318§ (1.047 to 3.589) | 0.429 (−0.706 to 1.562) | 0.487 (−0.967 to 1.941) | 4.192§ (3.029 to 5.355) | 1.218 (−0.330 to 2.765) | 1.061 (−0.332 to 2.455) | 2.115§ (0.917 to 3.312) | |
| 12 | — | 2.547† (0.326 to 4.767) | –0.078 (−0.876 to 0.720) | –0.155 (−0.954 to 0.644) | 5.456§ (4.539 to 6.372) | 1.424 (−0.144 to 2.992) | 0.496 (−0.614 to 1.606) | 1.382† (0.150 to 2.614) | |
| 18 | — | 1.710 (−0.00 to 3.424) | –0.082 (−0.974 to 0.810) | 3.132 × 10−-5(−0.710 to 0.711) | 4.963§ (2.923 to 7.003) | 1.232‡ (0.348 to 2.117) | 0.236 (−0.396 to 0.868) | 1.069 (−0.222 to 2.360) | |
| 24 | — | 1.492† (0.016 to 2.969) | –0.079 (−0.737 to 0.579) | 0.154 (−0.270 to 0.579) | 3.688‡ (1.606 to 5.770) | 0.929‡ (0.256 to 1.602) | 0.296 (−0.181 to 0.772) | 0.733† (0.061 to 1.404) | |
| Boric acid | 0 | — | — | –0.369 (−1.129 to 0.392) | –0.112 (−0.931 to 0.707) | –0.286 (−1.081 to 0.510) | –0.156 (−1.268 to 0.957) | –0.533 (−1.311 to 0.245) | –0.236 (−1.277 to 0.805) |
| 6 | — | — | –1.889§ (−2.820 to −0.959) | –1.831‡ (−3.188 to −0.473) | 1.874§ (0.889 to 2.859) | –1.100 (−2.565 to 0.365) | –1.257 (−2.542 to 0.029) | –0.203 (−1.238 to 0.832) | |
| 12 | — | — | –2.625† (−4.848 to −0.402) | –2.702† (−4.925 to −0.478) | 2.909† (0.678 to 5.141) | –1.122 (−3.503 to 1.259) | –2.051 (−4.307 to 0.206) | –1.164 (−3.445 to 1.117) | |
| 18 | — | — | –1.792† (−3.540 to −0.044) | –1.710 (−3.432 to 0.013) | 3.253‡ (0.9487 to 5.558) | –0.477 (−2.224 to 1.270) | –1.473 (−3.190 to 0.244) | –0.641 (−2.510 to 1.229) | |
| 24 | — | — | –1.571† (−3.060 to −0.083) | –1.338 (−2.815 to 0.139) | 2.196 (−0.038 to 4.429) | –0.563 (−2.054 to 0.927) | –1.196 (−2.673 to 0.280) | –0.759 (−2.250 to 0.731) | |
| Activated charcoal | 0 | — | — | — | 0.256 (−0.354 to 0.867) | 0.083 (−0.479 to 0.644) | 0.213 (−0.821 to 1.247) | –0.164 (−0.684 to 0.356) | 0.133 (−0.813 to 1.078) |
| 6 | — | — | — | 0.059 (−1.190 to 1.307) | 3.764§ (3.141 to 4.386) | 0.789 (−0.592 to 2.171) | 0.633 (−0.522 to 1.787) | 1.686§ (0.939 to 2.434) | |
| 12 | — | — | — | –0.077 (−1.001 to 0.847) | 5.534§ (4.521 to 6.547) | 1.503 (−0.088 to 3.093) | 0.574 (−0.598 to 1.747) | 1.460† (0.181 to 2.740) | |
| 18 | — | — | — | 0.082 (−0.887 to 1.051) | 5.045§ (2.990 to 7.1) | 1.314† (0.246 to 2.383) | 0.318 (−0.615 to 1.252) | 1.151 (−0.222 to 2.524) | |
| 24 | — | — | — | 0.233 (−0.428 to 0.895) | 3.767‡ (1.686 to 5.847) | 1.008‡ (0.214 to 1.802) | 0.375 (−0.307 to 1.057) | 0.812† (0.019 to 1.605) | |
| Control* | 0 | — | — | — | — | –0.174 (−0.840 to 0.493) | –0.043 (−1.104 to 1.017) | –0.421 (−1.060 to 0.219) | –0.124 (−1.103 to 0.856) |
| 6 | — | — | — | — | 3.705§ (2.437 to 4.974) | 0.731 (−0.874 to 2.336) | 0.574 (−0.891 to 2.040) | 1.628† (0.332 to 2.924) | |
| 12 | — | — | — | — | 5.611§ (4.597 to 6.624) | 1.579 (−0.011 to 3.170) | 0.651 (−0.522 to 1.824) | 1.537† (0.257 to 2.817) | |
| 18 | — | — | — | — | 4.963§ (2.922 to 7.004) | 1.232‡ (0.269 to 2.196) | 0.236 (−0.543 to 1.016) | 1.069 (−0.250 to 2.388) | |
| 24 | — | — | — | — | 3.534‡ (1.452 to 5.616) | 0.775† (0.098 to 1.451) | 0.142 (−0.343 to 0.626) | 0.579 (−0.096 to 1.254) | |
| Cadexomer iodine | 0 | — | — | — | — | — | 0.130 (−0.919 to 1.179) | –0.247 (−0.843 to 0.349) | 0.050 (−0.915 to 1.014) |
| 6 | — | — | — | — | — | –2.974§ (−4.368 to −1.581) | –3.131§ (−4.313 to −1.949) | –2.077§ (−2.917 to −1.238) | |
| 12 | — | — | — | — | — | –4.031§ (−5.648 to −2.415) | –4.960§ (−6.187 to −3.732) | –4.074§ (−5.399 to −2.749) | |
| 18 | — | — | — | — | — | –3.731§ (−5.785 to −1.676) | –4.727§ (−6.766 to −2.688) | –3.894§ (−6.031 to −1.757) | |
| 24 | — | — | — | — | — | –2.759† (−4.840 to −0.678) | –3.392‡ (−5.473 to −1.312) | –2.955‡ (−5.036 to −0.874) | |
| Manuka honey | 0 | — | — | — | — | — | — | –0.377 (−1.418 to 0.664) | –0.080 (−1.283 to 1.123) |
| 6 | — | — | — | — | — | — | –0.157 (−1.714 to 1.401) | 0.897 (−0.518 to 2.312) | |
| 12 | — | — | — | — | — | — | –0.928 (−2.606 to 0.750) | –0.042 (−1.771 to 1.687) | |
| 18 | — | — | — | — | — | — | –0.996† (−1.923 to −0.069) | –0.164 (−1.534 to 1.207) | |
| 24 | — | — | — | — | — | — | –0.633 (−1.329 to 0.063) | –0.196 (−0.999 to 0.607) | |
| PB | 0 | — | — | — | — | — | — | — | 0.297 (−0.658 to 1.251) |
| 6 | — | — | — | — | — | — | — | 1.053 (−0.161 to 2.268) | |
| 12 | — | — | — | — | — | — | — | 0.886 (−0.535 to 2.307) | |
| 18 | — | — | — | — | — | — | — | .833 (−0.471 to 2.137) | |
| 24 | — | — | — | — | — | — | — | 0.437 (−0.257 to 1.132) | |
| Nanoparticle silver | 0 | — | — | — | — | — | — | — | — |
| 6 | — | — | — | — | — | — | — | — | |
| 12 | — | — | — | — | — | — | — | — | |
| 18 | — | — | — | — | — | — | — | — | |
| 24 | — | — | — | — | — | — | — | — | |
There were 8 replicates (each on a musculocutaneous skin flap from an individual horse) for each dressing group at the described time points. Time 0 (baseline) samples were obtained after a 1-hour incubation period that followed inoculation of the wounds with 1 × 106 CFUs of the isolate; the described dressings were applied immediately after collection of the baseline samples. Dressings were cut in an aseptic manner to fit the wound surface. An 8-mm punch muscle biopsy specimen was obtained from each wound at each time point for MBL determination.
Sterile nonadherent gauze.
Symbols represent significant differences for the group in the left-hand column versus the comparison group at the specified time point (†P < 0.05, ‡P < 0.01, §P < 0.001).
— = Not applicable. PB = Polyhexamethylene biguanide.
Mean ± SD MRSA MBL for wounds of the 8 dressing groups described in Figure 1 at each sampling time. The results for each tested dressing group were compared with results for the control group at each time point by means of 2-way repeated-measures ANOVA with Greenhouse-Geisser correction and a Tukey multiple comparisons test. Significant (P < 0.05) differences are indicated (asterisk). AC = Activated charcoal. BA = Boric acid. CA = Calcium alginate. CI = Cadexomer iodine. MH = Manuka honey. NS = Nanoparticle silver. PB = Polyhexamethylene biguanide. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 82, 3; 10.2460/ajvr.82.3.189
Comparison of the MBLs between groups other than the control group indicated that the results for the cadexomer iodine group were significantly lower at all time points after baseline, compared with all other groups, except for the boric acid group at the 24-hour time point. The MBLs for the boric acid group were significantly lower, compared with those for the activated charcoal group at all time points after baseline and the calcium alginate group at the 6-, 12-, and 24-hour time points. The MBLs for the nanoparticle silver group were significantly lower, compared with those for the activated charcoal and calcium alginate groups at the 6-, 12-, and 24-hour time points. The MBL for the manuka honey group was significantly lower, compared with that for the polyhexamethylene biguanide group at the 18-hour time point and compared with those for the activated charcoal and the calcium alginate groups at the 18− and 24-hour time points.
Within-group comparisons of the MBL at each time point with the MBL at baseline revealed significant decreases in the cadexomer iodine group at the 6-hour (P < 0.001) and 12-hour (P = 0.008) time points. All other groups had MBL increases at each time point, compared with the respective baseline values. These increases were significant (P < 0.05) at all time points, except for the nanoparticle silver group at the 6-hour time point (P = 0.193).
P aeruginosa
The mean differences in P aeruginosa MBL between groups at each time point and the corresponding 95% CIs were summarized (Table 2). Comparison with the control group revealed that MBLs for the cadexomer iodine group were significantly lower at all time points after baseline (Figure 3). The MBLs for the boric acid group were significantly lower than those for the control group at the 12− and 24-hour time points. The MBL of the manuka honey group was significantly lower than that for the control group at the 12-hour time point.
Mean (95% CI) difference in Pseudomonas aeruginosa MBL (log10 CFUs/mg of tissue) over time for wounds treated with various dressings in a perfused equine wound model.
| Comparison group | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Group | Time (h) | Calcium alginate | Boric acid | Activated charcoal | Control* | Cadexomer iodine | Manuka honey | PB | Nanoparticle silver |
| Calcium alginate | 0 | — | 0.227 (−0.801 to 1.255) | –0.074 (−1.364 to 1.215) | 0.034 (−1.129 to 1.196) | 0.470 (−1.054 to 1.993) | –0.115 (−1.151 to 0.921) | 0.130 (−1.256 to 1.515) | 0.031 (−0.889 to 0.950) |
| 6 | — | 1.943§ (0.978 to 2.909) | –0.098 (−1.194 to 0.998) | 0.554 (−0.966 to 2.073) | 2.722§ (1.649 to 3.794) | 1.450‡ (0.482 to 2.419) | 0.578 (−0.531 to 1.686) | 1.881† (0.058 to 3.704) | |
| 12 | — | 3.101‡ (1.008 to 5.195) | –0.097 (−1.298 to 1.105) | 0.160 (−1.142 to 1.462) | 4.351§ (2.247 to 6.455) | 2.032‡ (0.581 to 3.482) | 1.307 (−1.636 to 4.249) | 1.516 (−0.070 to 3.101) | |
| 18 | — | 2.260 (−0.220 to 4.740) | 0.277 (−0.917 to 1.471) | 0.719 (−1.372 to 2.810) | 4.761‡ (2.116 to 7.406) | 1.514‡ (0.552 to 2.476) | 0.852 (−0.478 to 2.181) | 1.976‡ (0.467 to 3.485) | |
| 24 | — | 2.992§ (1.976 to 4.008) | 0.826 (−0.252 to 1.904) | 0.847 (−0.321 to 2.015) | 4.809‡ (2.060 to 7.557) | 2.288 (−0.093 to 4.670) | 1.109 (−0.880 to 3.097) | 2.054§ (0.862 to 3.246) | |
| Boric acid | 0 | — | — | –0.301 (−1.606 to 1.003) | –0.193 (−1.375 to 0.989) | 0.243 (−1.291 to 1.776) | –0.342 (−1.403 to 0.719) | –0.098 (−1.496 to 1.301) | –0.197 (−1.148 to 0.755) |
| 6 | — | — | –2.042‡ (−3.290 to −0.794 | –1.390 (−2.979 to 0.199 | 0.778 (−0.453 to 2.010 | –0.493 (−1.654 to 0.667 | –1.366† (−2.623 to −0.109 | –0.063 (−1.924 to 1.799 | |
| 12 | — | — | –3.198‡ (−5.284 to −1.112) | –2.942‡ (−5.055 to −0.828) | 1.250 (−1.268 to 3.767) | –1.069 (−3.236 to 1.097) | –1.795 (−4.941 to 1.352) | –1.585 (−3.810 to 0.639) | |
| 18 | — | — | –1.983 (−4.478 to 0.512) | –1.541 (−4.337 to 1.255) | 2.501 (−0.620 to 5.622) | –0.746 (−3.221 to 1.729) | –1.408 (−3.928 to 1.111) | –0.285 (−2.848 to 2.279) | |
| 24 | — | — | –2.166§ (−3.285 to −1.047) | –2.145§ (−3.347 to −0.943) | 1.817 (−0.933 to 4.566) | –0.704 (−3.089 to 1.682) | –1.883 (−3.880 to 0.113) | –0.938 (−2.163 to 0.287) | |
| Activated charcoal | 0 | — | — | — | 0.108 (−1.281 to 1.498) | 0.544 (−1.116 to 2.204) | –0.041 (−1.350 to 1.269) | 0.204 (−1.347 to 1.755) | 0.105 (−1.143 to 1.352) |
| 6 | — | — | — | 0.652 (−0.983 to 2.287) | 2.820§ (1.509 to 4.131) | 1.549† (0.299 to 2.798) | 0.676 (−0.657 to 2.009) | 1.979† (0.087 to 3.871) | |
| 12 | — | — | — | 0.257 (−1.018 to 1.531) | 4.448§ (2.351 to 6.544) | 2.129‡ (0.699 to 3.558) | 1.403 (−1.537 to 4.344) | 1.613† (0.045 to 3.181) | |
| 18 | — | — | — | 0.442 (−1.678 to 2.562) | 4.484‡ (1.827 to 7.141) | 1.237† (0.108 to 2.366) | 0.575 (−0.845 to 1.995) | 1.699† (0.121 to 3.276) | |
| 24 | — | — | — | 0.0212 (−1.225 to 1.267) | 3.983‡ (1.230 to 6.736) | 1.462 (−0.929 to 3.854) | 0.283 (−1.725 to 2.291) | 1.228 (−0.039 to 2.495) | |
| Control* | 0 | — | — | — | — | 0.436 (−1.157 to 2.029) | –0.149 (−1.337 to 1.039) | 0.096 (−1.376 to 1.567) | –0.003 (−1.110 to 1.103) |
| 6 | — | — | — | — | 2.168‡ (0.542 to 3.794) | 0.897 (−0.693 to 2.486) | 0.024 (−1.616 to 1.664) | 1.327 (−0.727 to 3.381) | |
| 12 | — | — | — | — | 4.191§ (2.067 to 6.315) | 1.872† (0.372 to 3.372) | 1.147 (−1.802 to 4.095) | 1.356 (−0.270 to 2.982) | |
| 18 | — | — | — | — | 4.042‡ (1.122 to 6.962) | 0.795 (−1.285 to 2.875) | 0.133 (−2.026 to 2.291) | 1.257 (−0.968 to 3.481) | |
| 24 | — | — | — | — | 3.961‡ (1.201 to 6.721) | 1.441 (−0.963 to 3.845) | 0.261 (−1.768 to 2.291) | 1.207 (−0.125 to 2.539) | |
| Cadexomer iodine | 0 | — | — | — | — | — | –0.585 (−2.121 to 0.952) | –0.340 (−2.057 to 1.376) | –0.439 (−1.937 to 1.058) |
| 6 | — | — | — | — | — | –1.271† (−2.504 to −0.038) | –2.144‡ (−3.463 to −0.825) | –0.841 (−2.727 to 1.045) | |
| 12 | — | — | — | — | — | –2.319† (−4.495 to −0.143) | –3.044 (−6.195 to 0.106) | –2.835‡ (−5.068 to −0.602) | |
| 18 | — | — | — | — | — | –3.247† (−5.890 to −0.604) | –3.909‡ (−6.586 to −1.232) | –2.785† (−5.500 to −0.071) | |
| 24 | — | — | — | — | — | –2.520 (−5.661 to 0.620) | –3.700† (−6.662 to −0.739) | –2.754 (−5.517 to 0.008) | |
| Manuka honey | 0 | — | — | — | — | — | — | 0.245 (−1.158 to 1.647) | 0.145 (−0.816 to 1.106) |
| 6 | — | — | — | — | — | — | –0.873 (−2.131 to 0.386) | 0.431 (−1.432 to 2.293) | |
| 12 | — | — | — | — | — | — | –0.725 (−3.693 to 2.243) | –0.516 (−2.238 to 1.206) | |
| 18 | — | — | — | — | — | — | –0.662 (−1.944 to 0.619) | 0.462 (−1.014 to 1.937) | |
| 24 | — | — | — | — | — | — | –1.180 (−3.859 to 1.499) | –0.234 (−2.642 to 2.174) | |
| PB | 0 | — | — | — | — | — | — | — | –0.099 (−1.450 to 1.252) |
| 6 | — | — | — | — | — | — | — | 1.303 (−0.5927 to 3.199) | |
| 12 | — | — | — | — | — | — | — | 0.209 (−2.784 to 3.202) | |
| 18 | — | — | — | — | — | — | — | 1.124 (−0.5320 to 2.779) | |
| 24 | — | — | — | — | — | — | — | 0.946 (−1.090 to 2.982) | |
| Nanoparticle silver | 0 | — | — | — | — | — | — | — | — |
| 6 | — | — | — | — | — | — | — | — | |
| 12 | — | — | — | — | — | — | — | — | |
| 18 | — | — | — | — | — | — | — | — | |
| 24 | — | — | — | — | — | — | — | — | |
See Table 1 for key.
Mean ± SD P aeruginosa MBL for wounds of the 8 dressing groups described in Figure 1 at each sampling time. See Figures 1 and 2 for key.
Citation: American Journal of Veterinary Research 82, 3; 10.2460/ajvr.82.3.189
Comparison of the MBLs between groups other than the control group indicated that results for the cadexomer iodine group were significantly lower than those for the activated charcoal and the calcium alginate groups at all time points after baseline; those for the manuka honey group at the 6-, 12-, and 18-hour time points; those for the nanoparticle silver group at the 12− and 18-hour time points; and those for the polyhexamethylene biguanide group at the 6-, 18-, and 24-hour time points. The MBLs for the boric acid group were significantly lower, compared with those for the activated charcoal and calcium alginate groups at the 6-, 12-, and 24-hour time points and that for the polyhexamethylene biguanide group at the 6-hour time point. The MBLs for the manuka honey group were significantly lower, compared with those for the activated charcoal and calcium alginate groups at the 6-, 12− and 18-hour time points. The MBLs for the nanoparticle silver group were significantly lower, compared with those for the activated charcoal group at the 6-, 12-, and 18-hour time points and those for the calcium alginate group at the 6-, 18-, and 24-hour time points.
On within-group comparisons of the MBL at each time point with the MBL at baseline, no significant decreases were identified. All groups except for the cadexomer iodine group had significant increases in MBL at various time points. The boric acid and polyhexamethylene biguanide groups had significantly increased MBLs at the 18-hour (P = 0.011 and P < 0.001, respectively) and 24-hour (P = 0.004 for both comparisons) time points, whereas the manuka honey and nanoparticle silver groups had significant increases at the 12-hour (P = 0.009 and P = 0.005, respectively), 18-hour (P < 0.001 and P = 0.001, respectively), and 24-hour (P = 0.020 and P < 0.001, respectively) time points. The calcium alginate and activated charcoal groups had significant (P ≤ 0.013 for all comparisons) increases at all time points after baseline.
The post hoc power analysis indicated a power of 0.99 for the comparisons between groups and also for within-group comparisons between baseline MBLs and those at subsequent time points. This indicated a very low probability of having detected false-negative results.
Discussion
Multiple dressings have been developed for human wound care,4 and several studies5,6,7,22,23,24 have evaluated their effects on wound healing and influences on microbial growth in clinical settings or in vivo experiments. However, comparative studies25,26 performed to evaluate multiple dressings under the same circumstances are limited, and the choice of a specific wound dressing is often based on personal preference. The results of the present study revealed significant differences in the effects of 8 different dressings used in equine wound management on the growth of MRSA and P aeruginosa, 2 bacterial pathogens frequently isolated from equine wounds.4,11
In our study, wounds in the cadexomer iodine dressing group had no significant increases in the MBLs of MRSA or P aeruginosa during the 24-hour observation period, and MBLs of MRSA were significantly reduced at the 6− and 12-hour time points, compared with the MBL for samples obtained immediately before the dressings were applied (ie, baseline). Comparison of the cadexomer iodine group with all other dressing groups, including the control (sterile nonadherent gauze) group, revealed that cadexomer iodine had a significantly greater antibacterial effect in MRSA-inoculated wounds at all sampling times, except for boric acid at the 24-hour time point. In P aeruginosa–inoculated wounds, cadexomer iodine had a significantly greater antibacterial effect, compared with activated charcoal, calcium alginate, and the control dressing at all time points after baseline and compared with manuka honey, nanoparticle silver, and polyhexamethylene biguanide at multiple time points. Iodine formulations have a long history in wound care and are known to be potent antiseptics.4,27,28,29,30,31 However, older iodine formulations are considered somewhat aggressive and might have detrimental effects on the surrounding tissues.29,30,32 More recent formulations such as povidone-iodine4,29,30,31,32 cadexomer iodine,4,28,29,32 and iodine and oxygen–releasing hydrogel dressings32,33 are regarded as safer than the older formulations because the slow release of iodine avoids excessively high local concentrations.4,31 There is also no known resistance to iodine formulations among bacteria, although some formulations may be rapidly deactivated by organic matter.4,28,29,30 In our study, we used cadexomer iodine, a starch polymer that is safe to use, modulates the inflammatory response, and is active against biofilms.27,32,34 Some studies4,27,32,34,35 have also found a positive effect on wound healing, more specifically on epidermal regeneration, with more epithelial cell layers, more rete ridges, and a thicker stratum corneum present in the tissue formed.35
Another interesting finding in the study reported here was the effect of boric acid on microbial growth, particularly for P aeruginosa. Indeed, only the boric acid group had MBLs of P aeruginosa that did not differ significantly from those for the cadexomer iodine group during the entire observation period. Our results were in general agreement with previous reports in the human medical literature that boric acid has activity against P aeruginosa36,37 in addition to other microorganisms such as MRSA and certain fungi.37,38 Results of some studies37,38,39,40 suggest that boric acid is a valuable antiseptic and promotes wound healing through anti-inflammatory and antioxidant effects. In veterinary medicine, it is used for the treatment of hoof abcesses,4 but little research is available regarding its use in wound care. Some authors have raised concerns regarding absorption and possible toxic effects of boric acid when used for a long period of time.4 However, as with all antiseptics, prolonged use is not advocated.4 The results of the present study indicated that it might be a useful alternative for treatment of contaminated wounds, but more in vivo research is needed to confirm this.
Comparison with results for the control group indicated that nanoparticle silver had a significant effect on MRSA, with lower MBLs in the first 12 hours, but results did not differ between the 2 groups after that time. In P aeruginosa–inoculated wounds, results for the nanoparticle silver group did not differ significantly from those for the control group at any time. These results were in contrast to the findings of other investigations4,8 showing that silver-containing formulations can successfully be used against MRSA and P aeruginosa in vitro. Although the reported prevalence of resistance against silver is low,4,41 this might have influenced the outcome of our study. The number of clinical studies on the use of silver-containing formulations is limited. A silver-containing semiocclusive foam dressing was found to have beneficial effects on wound healing in horses, especially during the first 30 days of treatment, although the complete wound healing time did not differ from that for a control dressing.7 One clinical study9 on the use of various topically applied treatments for bandaged wounds in horses found that application of silver sulfadiazine cream resulted in more exuberant granulation tissue formation, compared with a control (no topical treatment). In our study, a more recent nanoparticle silver formulation was used because it is reported to have faster and more prolonged antibacterial activity than silver sulfadiazine,16,23 but the latter property could not be confirmed in the present study.
In the present study, the manuka honey group had significantly lower MBLs of MRSA, compared with that for the polyhexamethylene biguanide group at the 18-hour time point and with those for the calcium alginate, activated charcoal, and control groups at the 18− and 24-hour time points. The manuka honey group also had significantly lower MBLs of P aeruginosa, compared with the control group at the 12-hour time point and the activated charcoal and calcium alginate groups at the 6-, 12-, and 18-hour time points. The topical use of honey in equine wound care has become popular in the past decade, and several studies5,6,24 have evaluated its efficacy for this purpose. There is little known resistance to the antibacterial activities of honey,4,5,6,11,12,13,14,24,42,43,44 which also contains antioxidants and has been shown to have anti-inflammatory and proinflammatory properties.42,44 However, the described antiseptic and immunomodulatory properties of topically applied honey are dependent on the type of honey used and how it is processed.4,5,6,11,13,24,42,43,44 We chose to use manuka honey because it has been reported to have greater activity against microbes attributed to its methylglyoxal content, which is graded by a unique manuka factor.4,5,6,11,24,43,44 The manuka honey used in our study had a unique manuka factor of 15, which is described as therapeutically acceptable.24 An in vitro study11 in which multiple honeys were tested against selected bacterial isolates from horses showed that the MIC values of manuka honey with a unique manuka factor of 20 are lower than those for manuka honey with a unique manuka factor of 10. Investigations of experimentally contaminated skin wounds created on the distal aspects of the limbs of horses have shown that manuka honey with a unique manuka factor of 20 reduces wound retraction5,6 and overall healing time.6 This may be attributable not only to antibacterial effects but also to proinflammatory properties stimulating the often-lacking initial inflammation in horses.5,6,11
The MBLs for both isolates in wounds of the activated charcoal, calcium alginate, and polyhexamethylene biguanide groups in the present study did not differ from those for the control group during the observation period. Within-group comparisons with the baseline MBLs revealed that the activated charcoal, calcium alginate, and control groups had significant increases in MBL for both isolates at all other time points throughout the observation period. This result was expected for the activated charcoal and calcium alginate groups as well as the control group because activated charcoal and calcium alginate are not known to have specific antibacterial properties, except for the possible absorption of bacteria into the dressing.4 The activated charcoal dressing is mainly used for odor control in patients with infected wounds, and the calcium alginate dressing is known for the stimulation of fibroblasts, which aids equine wound healing.4,45 The latter wound dressing is typically not applied to an infected wound bed unless used in combination with an antiseptic (eg, silver or polyhexamethylene biguanide formulations).4
The limited effect that polyhexamethylene biguanide dressings had on the MBLs of isolates in the present study was unexpected, as it is a well-known and often-used antiseptic in both human and veterinary medicine with activity against MRSA45 and P aeruginosa.46,47 Possible explanations included the short observation period, a too-low concentration of the active product in the dressing, or possible bacterial resistance,4,48 although to the authors’ knowledge, such resistance has not been reported to date.
With the choice of our perfused wound model, we aimed to mimic full-thickness open wounds in horses, including the presence of tissue proteins and wound exudate, for which dressings would commonly be applied. However, the model and therefore the study had several limitations. An important limitation was that the musculocutaneous flaps were fresh but nonviable, which limited the comparison to an in vivo situation because several variables such as immunologic responses and blood flow are excluded in such a model.48,49 Additionally, histologic examination was not performed to monitor the amount of tissue decay present at the different sampling times. Although the flaps showed no visible signs of dehydration, an objective measure of the decay would have added to the model. Moreover, the quality of the model could have been improved by replacing the saline solution with a polyionic and nourishing solution (eg, Dulbecco modified Eagle medium containing 10% fetal bovine serum) to increase tissue viability and delay decay.15 For these reasons, the results from our study cannot be extrapolated to a clinical situation. Moreover, the concentration of an active substance, its possible elution from or dilution in the dressing, and the ability to absorb exudate all depend on the choice of wound dressing. As a result, conclusions can be drawn only for the specific wound dressings used in the present study. Proximity to the epigastric artery might also influence elution of the active substance or its dilution in the dressing. To exclude any possible bias associated with the proximity to the epigastric artery, the dressings were applied according to a rotational design, and the experiment was repeated 8 times so that each dressing covered 1 of the 8 wound locations on the flap 1 time. To further improve the model, we would also propose evaluating sections of tested dressings by scanning electron microscopy or fluorescence microscopy to monitor microbial growth within them.10 Despite the described limitations, the perfused equine wound model allowed evaluation of the effects of multiple wound dressings on microbial growth under the experimental circumstances.
Our results supported the potential of the cadexomer iodine dressing to inhibit or reduce microbial growth in MRSA− or P aeruginosa–inoculated wounds. However, the effects of these dressings and their impact on wound healing should be further investigated with in vivo studies.
Acknowledgments
Supported by the European College of Veterinary Surgeons, Vetsuisse Faculty University of Zurich, ECVS Resident Research Grant 2017 (Large Animal). Funding sources did not have any involvement in the study design, data analysis and interpretation, or writing and publication of the manuscript. Manufacturers of the investigated dressings had no role in the study design; data collection, analysis, and interpretation; or the decision to submit the manuscript for publication.
The authors declare that there were no conflicts of interest.
Abbreviations
| MBL | Mean bacterial load |
| MRSA | Methicillin-resistant Staphylococcus aureus |
Footnotes
Carbonet (10 × 10 cm), Smith + Nephew, Watford, England.
Animalintex Poultice (40.5 × 20.5 cm), Robinson Healthcare, Carlton in Lindrick, England.
Iodosorb dressing (4 × 6 cm), Smith + Nephew, Watford, England.
Suprasorb A (5 × 5 cm), Lohmann & Rauscher, Rengsdorf, Germany.
Kruuse manuka AD (5 × 5 cm), Jørgen Kruuse A/S, Labngeskov, Denmark.
Acticoat Flex 3 (5 × 5 cm), Smith + Nephew, Watford, England.
Telfa AMD Island dressing (10 × 12.5 cm), Covidien, Dublin, Ireland.
Zorbopad (10 × 20 cm), Millpledge Veterinary, Clarborough, England.
Graphpad Prism, version 8.1.1, GraphPad Software, San Diego, Calif.
References
- 1. ↑
Theoret C. Physiology of wound healing. In: Theoret C, Schumacher J, eds. Equine wound management. 3rd ed. Hoboken, NJ: John Wiley & Sons Inc, 2016;1–13.
- 2. ↑
Dart AJ, Sole-Guitart A, Stashak TS, et al. Selected factors that negatively impact healing. In: Theoret C, Schumacher J, eds. Equine wound management. 3rd ed. Hoboken, NJ: John Wiley & Sons Inc, 2016;30–46.
- 3. ↑
Westgate SJ, Percival SL, Knottenbelt DC, et al. Microbiology of equine wounds and evidence of bacterial biofilms. Vet Microbiol 2011;150:152–159.
- 4. ↑
Jacobsen S. Update on wound dressings. In: Theoret C, Schumacher J, eds. Equine wound management. 3rd ed. Hoboken, NJ: John Wiley & Sons Inc, 2016;104–131.
- 5. ↑
Bischofberger AS, Dart CM, Perkins NR, et al. A preliminary study on the effect of manuka honey on second-intention healing of contaminated wounds on the distal aspect of the forelimbs of horses. Vet Surg 2011;40:898–902.
- 6. ↑
Bischofberger AS, Dart CM, Perkins NR, et al. The effect of short− and long-term treatment with manuka honey on second intention healing of contaminated and noncontaminated wounds on the distal aspect of the forelimbs in horses. Vet Surg 2013;42:154–160.
- 7. ↑
Kelleher ME, Kilcoyne I, Dechant JE, et al. A preliminary study of silver sodium zirconium phosphate polyurethane foam wound dressing on wounds of the distal aspect of the forelimb in horses. Vet Surg 2015;44:359–365.
- 8. ↑
Adams AP, Santschi EM, Mellencamp MA. Antibacterial properties of a silver chloride-coated nylon wound dressing. Vet Surg 1999;28:219–225.
- 9. ↑
Harmon CCG, Hawkins JF, Li J, et al. Effects of topical application of silver sulfadiazine cream, triple antimicrobial ointment, or hyperosmolar nanoemulsion on wound healing, bacterial load, and exuberant granulation tissue formation in bandaged full-thickness equine skin wounds. Am J Vet Res 2017;78:638–646.
- 10. ↑
Lipp C, Kirker K, Agostinho A, et al. Testing wound dressings using an in vitro wound model. J Wound Care 2010;19:220–226.
- 11. ↑
Carnwath R, Graham EM, Reynolds K, et al. The antimicrobial activity of honey against common equine wound bacterial isolates. Vet J 2014;199:110–114.
- 12. ↑
Cooper R, Molan P. The use of honey as an antiseptic in managing Pseudomonas infection. J Wound Care 1999;8:161–164.
- 13. ↑
Cooper RA, Molan PC, Harding KG. Antibacterial activity of honey against strains of Staphylococcus aureus from infected wounds. J R Soc Med 1999;92:283–285.
- 14. ↑
Kwakman PHS, de Boer L, Ruyter-Spira CP, et al. Medical-grade honey enriched with antimicrobial peptides has enhanced activity against antibiotic-resistant pathogens. Eur J Clin Microbiol Infect Dis 2011;30:251–257.
- 15. ↑
Stojadinovic O, Tomic-Canic M. Human ex vivo wound healing model. Methods Mol Biol 2013;1037:255–264.
- 16. ↑
Xu W, Hong SJ, Jia S, et al. Application of a partial-thickness human ex vivo skin culture model in cutaneous wound healing study. Lab Invest 2012;92:584–599.
- 17. ↑
Schaudinn C, Dittmann C, Jurisch J, et al. Development, standardization and testing of a bacterial wound infection model based on ex vivo human skin. PLoS One 2017;12:e0186946.
- 18. ↑
Stone R II, Wall JT, Natesan S, et al. PEG-plasma hydrogels increase epithelialization using a human ex vivo skin model. Int J Mol Sci 2018;19:3156.
- 19. ↑
Alves DR, Booth SP, Scavone P, et al. Development of a high-throughput ex-vivo burn wound model using porcine skin, and its application to evaluate new approaches to control wound infection. Front Cell Infect Microbiol 2018;8:196.
- 20. ↑
Van Hecke LL, Haspeslagh M, Hermans K, et al. Comparison of antibacterial effects among three foams used with negative pressure wound therapy in an ex vivo equine perfused wound model. Am J Vet Res 2016;77:1325–1331.
- 21. ↑
Antweiler RC, Taylor HE. Evaluation of statistical treatments of left-censored environmental data using coincident uncensored data sets: I. Summary statistics. Environ Sci Technol 2008;42:3732–3738.
- 22. ↑
Rigo C, Ferroni L, Tocco I, et al. Active silver nanoparticles for wound healing. Int J Mol Sci 2013;14:4817–4840.
- 23. ↑
Tian J, Wong KKY, Ho C-M, et al. Topical delivery of silver nanoparticles promotes wound healing. ChemMedChem 2007;2:129–136.
- 24. ↑
Tsang AS, Dart AJ, Sole-Guitart A, et al. Comparison of the effects of topical application of UMF20 and UMF5 manuka honey with a generic multifloral honey on wound healing variables in an uncontaminated surgical equine distal limb wound model. Aust Vet J 2017;95:333–337.
- 25. ↑
Berry DB II, Sullins KE. Effects of topical application of antimicrobials and bandaging on healing and granulation tissue formation in wounds of the distal aspect of the limbs in horses. Am J Vet Res 2003;64:88–92.
- 26. ↑
Eberlein T, Haemmerle G, Signer M, et al. Comparison of PHMB-containing dressing and silver dressings in patients with critically colonised or locally infected wounds. J Wound Care 2012;21:12, 14–16, 18–20.
- 27. ↑
Mertz PM, Oliveira-Gandia MF, Davis SC. The evaluation of a cadexomer iodine wound dressing on methicillin resistant staphylococcus aureus (MRSA) in acute wounds. Dermatol Surg 1999;25:89–93.
- 28. ↑
Leaper DJ, Durani P. Topical antimicrobial therapy of chronic wounds healing by secondary intention using iodine products. Int Wound J 2008;5:361–368.
- 29. ↑
Atiyeh BS, Dibo SA, Hayek SN. Wound cleansing, topical antiseptics and wound healing. Int Wound J 2009;6:420–430.
- 30. ↑
Bigliardi PL, Alsagoff SAL, El-Kafrawi HY, et al. Povidone iodine in wound healing: a review of current concepts and practices. Int J Surg 2017;44:260–268.
- 31. ↑
Wang L, Qin W, Zhou Y, et al. Transforming growth factor β plays an important role in enhancing wound healing by topical application of Povidone-iodine. Sci Rep 2017;20:991.
- 33. ↑
Moffatt CJ, Stanton J, Murray S. A randomised trial to compare the performance of Oxyzyme and Iodozyme with standard care in the treatment of patients with venous and mixed venous/arterial ulceration. Wound Med 2014;6:1–10.
- 34. ↑
Akiyama H, Oono T, Saito M, et al. Assessment of cadexomer iodine against Staphylococcus aureus biofilm in vivo and in vitro using confocal laser scanning microscopy. J Dermatol 2004;31:529–534.
- 35. ↑
Lamme EN, Gustafsson TO, Middelkoop E. Cadexomer iodine ointment shows stimulation of epidermal regeneration in experimental full-thickness wounds. Arch Dermatol Res 1998;290:18–24.
- 36. ↑
Nagoba BS, Selkar SP, Wadher BJ, et al. Acetic acid treatment of pseudomonal wound infections—a review. J Infect Public Health 2013;6:410–415.
- 37. ↑
Demirci S, Dogan A, Aydın S, et al. Boron promotes streptozotocin-induced diabetic wound healing: roles in cell proliferation and migration, growth factor expression, and inflammation. Mol Cell Biochem 2016;417:119–133.
- 38. ↑
Nzietchueng RM, Dousset B, Franck P, et al. Mechanisms implicated in the effects of boron on wound healing. J Trace Elem Med Biol 2002;16:239–244.
- 39. ↑
Güzel Y, Golge UH, Goksel F, et al. The efficacy of boric acid used to treat experimental osteomyelitis caused by methicillin-resistant Staphylococcus aureus: an in vivo study. Biol Trace Elem Res 2016;173:384–389.
- 40. ↑
Tepedelen BE, Soya E, Korkmaz M. Boric acid reduces the formation of DNA double strand breaks and accelerates wound healing process. Biol Trace Elem Res 2016;174:309–318.
- 41. ↑
Woods EJ, Cochrane CA, Percival SL. Prevalence of silver resistance genes in bacteria isolated from human and horse wounds. Vet Microbiol 2009;138:325–329.
- 43. ↑
Carter DA, Blair SE, Cokcetin NN, et al. Therapeutic manuka honey: no longer so alternative. Front Microbiol 2016;7:569.
- 44. ↑
Oryan A, Alemzadeh E, Moshiri A. Biological properties and therapeutic activities of honey in wound healing: a narrative review and meta-analysis. J Tissue Viability 2016;25:98–118.
- 45. ↑
Hendrickson DA. Management of superficial wounds. In: Auer JA, Stick JA, Kummerle JM, et al., eds. Equine surgery. 5th ed. St Louis: Elsevier Saunders, 2019;303–317.
- 46. ↑
To E, Dyck R, Gerber S, et al. The effectiveness of topical polyhexamethylene biguanide (PHMB) agents for the treatment of chronic wounds: a systematic review. Surg Technol Int 2016;29:45–51.
- 47. ↑
Sibbald RG, Coutts P, Woo KY. Reduction of bacterial burden and pain in chronic wounds using a new polyhexamethylene biguanide antimicrobial foam dressing-clinical trial results. Adv Skin Wound Care 2011;24:78–84.
- 48. ↑
Chindera K, Mahato M, Sharma AK, et al. The antimicrobial polymer PHMB enters cells and selectively condenses bacterial chromosomes. Sci Rep 2016;6:23121.
- 49. ↑
Assadian O, Assadian A, Stadler M, et al. Bacterial growth kinetic without the influence of the immune system using vacuum-assisted closure dressing with and without negative pressure in an in vitro wound model. Int Wound J 2010;7:283–289.
