Efficacy of topical honey compared to systemic gentamicin for treatment of infected war wounds in a porcine model: A non-inferiority experimental pilot study

  • Måns Muhrbeck
    Correspondence
    Corresponding author at: Department of Surgery in Norrköping, and Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden.Present address: Department of Surgery, Vrinnevi Hospital, Norrköping 603 79, Sweden.
    Affiliations
    Department of Surgery in Norrköping, and Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden

    Department of Biomedical and Clinical Sciences, Center for Disaster Medicine and Traumatology, Linköping University, Linköping, Sweden
    Search for articles by this author
  • Andreas Wladis
    Affiliations
    Department of Biomedical and Clinical Sciences, Center for Disaster Medicine and Traumatology, Linköping University, Linköping, Sweden

    Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden
    Search for articles by this author
  • Maria Lampi
    Affiliations
    Department of Biomedical and Clinical Sciences, Center for Disaster Medicine and Traumatology, Linköping University, Linköping, Sweden
    Search for articles by this author
  • Peter Andersson
    Affiliations
    Department of Biomedical and Clinical Sciences, Center for Disaster Medicine and Traumatology, Linköping University, Linköping, Sweden

    Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden
    Search for articles by this author
  • Johan P.E. Junker
    Affiliations
    Department of Biomedical and Clinical Sciences, Center for Disaster Medicine and Traumatology, Linköping University, Linköping, Sweden

    Department of Biomedical and Clinical Sciences, Laboratory of Experimental Plastic Surgery, Linköping University, Linköping, Sweden
    Search for articles by this author
Open AccessPublished:October 20, 2021DOI:https://doi.org/10.1016/j.injury.2021.10.019

      Highlights

      • In armed conflicts, treatment of infected wounds constitutes a large portion of the workload. Honey could be a useful adjunct in this treatment.
      • Neither topically applied Manuka honey nor intramuscular gentamicin reduced S. aureus count in infected wounds in the porcine model used.
      • Wound area remained unchanged following treatment with topically applied Manuka but decreased with intramuscular gentamicin.
      • Systemic and local inflammatory responses were more persistent with topically applied Manuka honey than with intramuscular gentamicin.

      Abstract

      Background

      In armed conflicts, infected wounds constitute a large portion of the surgical workload. Treatment consists of debridements, change of dressings, and antibiotics. Many surgeons advocate for the use of honey as an adjunct with the rationale that honey has bactericidal and hyperosmotic properties. However, according to a Cochrane review from 2015 there is insufficient data to draw any conclusions regarding the efficacy of honey in treatment of wounds. We, therefore, decided to evaluate if honey is non-inferior to gentamicin in the treatment of infected wounds in a highly translatable porcine wound model.

      Material and methods

      50 standardized wounds on two pigs were infected with S. aureus and separately treated with either topically applied Manuka honey or intramuscular gentamicin for eight days. Treatment efficacy was evaluated with quantitative cultures, wound area measurements, histological, immunohistochemical assays, and inflammatory response.

      Results

      Topically applied Manuka honey did not reduce bacterial count or wound area for the duration of treatment. Intramuscular gentamicin initially reduced bacterial count (geometric mean 5.59*¸0.37 – 4.27*¸0.80 log10 (GSD) CFU/g), but this was not sustained for the duration of the treatment. However, wound area was significantly reduced with intramuscular gentamicin at the end of treatment (mean 112.8 ± 30.0–67.7 ± 13.2 (SD) mm2). ANOVA-analysis demonstrated no variation in bacterial count for the two treatments but significant variation in wound area (p<0.0001). The inflammatory response was more persistent in the pig with wounds treated with topically applied Manuka honey than in the pig treated with intramuscular gentamicin.

      Conclusion

      At the end of treatment S. aureus count was the same with topically applied Manuka honey and intramuscular gentamicin. The wound area was unchanged with topically applied Manuka honey and decreased with intramuscular gentamicin. Topically applied Manuka honey could consequently be non-inferior to intramuscular gentamicin in reducing S. aureus colonization on the wound's surface, but not in reducing wound size. The use of Manuka honey dressings to prevent further progression of a wound infection may therefore be of value in armed conflicts, where definite care is not immediately available.

      Keywords

      Abbreviations:

      ANOVA (analysis of variance), ATCC (American type culture collection), CFU/g (colony forming units per gram of tissue), °C (degree Celsius), CCD (charged-coupled device), CD68 (protein highly expressed in macrophages and other cells in the monocyte lineage), EDTA (ethylenediaminetetraacetic acid), EMEA (European agency for evaluation of medical products), FDA (US federal drug administration), H&E (haematoxylin and eosin staining), GSD (geometric standard deviation), IM (intramuscular), LMIC (low- and middle-income countries), MDR (multidrug-resistance), MIC (minimum inhibitory concentration), MGO (methylglyoxal, MRSA, methicillin-resistant S. aureus), PBS (phosphate-buffered saline), RPM (revolutions per minute), SAA (serum amyloid A), SD (standard deviation), UMF (Unique Manuka factor, a measurement of methylglyoxal concentration), WHO (World Health Organization), W/v (weight by volume)

      Introduction

      In armed conflicts infected wounds are common and contribute to a significant portion of the surgical workload [
      • Alga A.
      • Wong S.
      • Shoaib M.
      • Lundgren K.
      • Giske C.G.
      • von Schreeb J.
      • et al.
      Infection with high proportion of multidrug-resistant bacteria in conflict-related injuries is associated with poor outcomes and excess resource consumption: a cohort study of Syrian patients treated in Jordan.
      ,
      • Murray C.K.
      • Hinkle M.K.
      • Yun H.C.
      History of infections associated with combat-related injuries.
      ]. Current treatment protocols consist of repeated debridements, change of dressings, and administration of antibiotics to mitigate infection [
      • Giannou C.
      • Baldan M.
      ,
      Clinical guidelines: diagnostic and treatment manual. Paris.
      ]. Final wound closure is subsequently achieved with delayed primary closure, flaps, split-thickness skin grafts, or a combination thereof. Successful treatment of wounds in armed conflicts depends on supply chains that can provide enough dressing material and antibiotics, even in times of a surge of injured. In low- and middle-income countries (LMIC), these supply chains are fragile, and the onset of an armed conflict many times causes a complete disruption of essential supplies to health care facilities [
      • Leaning J.
      • Guha-Sapir D.
      Natural disasters, armed conflict, and public health.
      ,
      • Matowe L.W.
      • Waako P.
      • Adome R.O.
      • Kibwage I.
      • Omary M.
      • Bienvenu E.
      A strategy to improve skills in pharmaceutical supply management in East Africa: the regional technical resource collaboration for pharmaceutical management.
      ,
      United Nations
      Every-Woman-Every-Child.
      ]. Health care staff then have to rely on materials and antibiotics they have at hand. Locally available antibiotics can be of questionable origin, effectiveness and could even be harmful to patients [
      • Kelesidis T.
      • Falagas M.E.
      Substandard/counterfeit antimicrobial drugs.
      ].
      Furthermore, uncontrolled and non-systematic use of antibiotics significantly accelerates the globally increasing problem with antimicrobial resistance [
      • Laxminarayan R.
      • Duse A.
      • Wattal C.
      • Zaidi A.K.
      • Wertheim H.F.
      • Sumpradit N.
      • et al.
      Antibiotic resistance-the need for global solutions.
      ,
      World Health Organization
      ]. There is, therefore, much to gain in developing robust and effective treatment strategies that do not rely on antibiotics or expensive dressing materials. Many healthcare professionals, particularly in resource-scarce settings, have for many years advocated the use of honey dressings in the treatment of infected wounds [
      • Armon P.J.
      The use of honey in the treatment of infected wounds.
      ,
      • Saikaly S.K.
      • Khachemoune A.
      Honey and wound healing: an Update.
      ]. The rationale for the use of honey is that it empirically has been found to have bactericidal and hyperosmotic properties. There are some scientific reports to support this perception [
      • Cooper R.A.
      • Molan P.C.
      • Harding K.G.
      Antibacterial activity of honey against strains of Staphylococcus aureus from infected wounds.
      ,
      • Irish J.
      • Blair S.
      • Carter D.A.
      The antibacterial activity of honey derived from Australian flora.
      ,
      • Johnston M.
      • McBride M.
      • Dahiya D.
      • Owusu-Apenten R.
      • Nigam P.S.
      Antibacterial activity of Manuka honey and its components: an overview.
      ]. However, a Cochrane review from 2015 stated that: “it is difficult to draw overall conclusions regarding the effects of honey as a topical treatment for wounds due to the heterogeneous nature of the patient populations and comparators studied and the mostly low quality of the evidence” [
      • Jull A.B.
      • Cullum N.
      • Dumville J.C.
      • Westby M.J.
      • Deshpande S.
      • Walker N.
      • et al.
      Honey as a topical treatment for wounds.
      ]. We, therefore, decided to evaluate if topical honey is non-inferior to intramuscular (IM) gentamicin in the treatment of infected wounds in a well-established model for human wound healing, the porcine wound model [
      • Sullivan T.P.
      • Eaglstein W.H.
      • Davis S.C.
      • Mertz P.
      The pig as a model for human wound healing.
      ,
      • Junker J.P.E.
      • Lee C.C.Y.
      • Samaan S.
      • Hackl F.
      • Kiwanuka E.
      • Minasian R.A.
      • et al.
      Topical delivery of ultrahigh concentrations of gentamicin is highly effective in reducing bacterial levels in infected porcine full-thickness wounds.
      ,
      • Daly L.T.
      • Tsai D.M.
      • Singh M.
      • Nuutila K.
      • Minasian R.A.
      • Cameron C.Y.L.
      • et al.
      Topical Minocycline effectively decontaminates and reduces inflammation in infected porcine wounds.
      ]. Response to treatments was assessed during eight days with quantitative tissue cultures, wound surface area measurements, histological and immunohistochemical assays, as well as analysis of local and systemic inflammatory responses.

      Material and methods

       Treatments and bacteria

      Staphylococcus aureus (S. aureus, ATCC 29,213, Manassas, VA, US) was chosen as it is one of the most commonly detected bacteria in infected wounds in armed conflicts and is prone to develop antimicrobial resistance [
      • Alga A.
      • Wong S.
      • Shoaib M.
      • Lundgren K.
      • Giske C.G.
      • von Schreeb J.
      • et al.
      Infection with high proportion of multidrug-resistant bacteria in conflict-related injuries is associated with poor outcomes and excess resource consumption: a cohort study of Syrian patients treated in Jordan.
      ,
      • Eardley W.G.
      • Brown K.V.
      • Bonner T.J.
      • Green A.D.
      • Clasper J.C.
      Infection in conflict wounded.
      ,
      • Petersen K.
      • Riddle M.S.
      • Danko J.R.
      • Blazes D.L.
      • Hayden R.
      • Tasker S.A.
      • et al.
      Trauma-related infections in battlefield casualties from Iraq.
      ,
      • Lowy F.D.
      Antimicrobial resistance: the example of Staphylococcus aureus.
      ]. Several recent studies from armed conflicts have reported a high proportion of wounds being infected with methicillin-resistant S. aureus (MRSA) [
      • Alga A.
      • Wong S.
      • Shoaib M.
      • Lundgren K.
      • Giske C.G.
      • von Schreeb J.
      • et al.
      Infection with high proportion of multidrug-resistant bacteria in conflict-related injuries is associated with poor outcomes and excess resource consumption: a cohort study of Syrian patients treated in Jordan.
      ,
      • Nawfal Dagher T.
      • Al-Bayssari C.
      • Diene S.M.
      • Azar E.
      • Rolain J.M
      Bacterial infection during wars, conflicts and post-natural disasters in Asia and the Middle East: a narrative review.
      ]. Furthermore, S. aureus is able to develop a biofilm that limits the efficacy of antimicrobial treatment [
      • Reffuveille F.
      • Josse J.
      • Vallé Q.
      • Mongaret C.
      • Gangloff S.C.
      Staphylococcus aureus biofilms and their impact on the medical field.
      ].
      Manuka honey UMF 15+ (Comvita, Maidenhead, UK) from New Zealand was chosen as it has in previous studies been demonstrated to disrupt biofilms and has a bactericidal effect on S. aureus [
      • Johnston M.
      • McBride M.
      • Dahiya D.
      • Owusu-Apenten R.
      • Nigam P.S.
      Antibacterial activity of Manuka honey and its components: an overview.
      ,
      • Maddocks S.E.
      • Lopez M.S.
      • Rowlands R.S.
      • Cooper R.A.
      Manuka honey inhibits the development of Streptococcus pyogenes biofilms and causes reduced expression of two fibronectin binding proteins.
      ,
      • Campeau M.E.
      • Patel R.
      Antibiofilm activity of Manuka honey in combination with antibiotics.
      ]. Additionally, dressings with Manuka honey have been approved by the US Federal Drug Administration (FDA) for treatment of wounds [
      USFaD 510(k) premarket notification
      Food and drug administration.
      ].
      Gentamicin 40 mg/ml (Sanofi, Paris, France) was chosen as it has bactericidal properties against S. aureus, is generally available in LMIC and is recommended treatment for severe cutaneous infections by organizations that provide health care in armed conflicts [
      • Giannou C.
      • Baldan M.
      ,
      Clinical guidelines: diagnostic and treatment manual. Paris.
      ,
      • Tam V.H.
      • Kabbara S.
      • Vo G.
      • Schilling A.N.
      • Coyle E.A.
      Comparative pharmacodynamics of gentamicin against Staphylococcus aureus and Pseudomonas aeruginosa.
      ].

       Animals

      Two female Swedish native breed pigs (Sus scrofa domesticus, Persbo gård, Ransta, Sweden), both weighing 52 kg, were used for the study. Animals were socially housed and acclimatized for more than 72 h prior to any procedures. The pigs were kept in boxes measuring 2 × 2.5m2 with a light/dark cycle of 12 h/12 h and an ambient temperature of 18–20 °C. The pig's wellbeing and skin temperature were monitored at least three times daily. All animal experiments were performed under ethical approval from the regional animal ethics board (ID1418) and in strict accordance with guidelines postulated by the Medical Faculty at Linköpings University.

       Porcine wound model

      The porcine model was chosen as the pig's skin has been demonstrated to have similar cellular composition, morphology, and immunological properties as human skin [
      • Vardaxis N.J.
      • Brans T.A.
      • Boon M.E.
      • Kreis R.W.
      • Marres L.M.
      Confocal laser scanning microscopy of porcine skin: implications for human wound healing studies.
      ,
      • Lavker R.M.
      • Dong G.
      • Zheng P.S.
      • Murphy G.F.
      Hairless micropig skin. A novel model for studies of cutaneous biology.
      ]. Prior to any procedure, animals were sedated with IM administration of 3 µg/kg of zolazepam and tiletamine (Zoletil; Virbac A/S, Kolding, Denmark) and 10 µg/kg dexmedetomidine (Dexdomitor; Orion Pharma AB, Danderyd, Sweden). At the time of wound creation, inoculation, and termination (on days 0 and 10) general anaesthesia and analgesia were maintained with intravenous infusion of 3–7.5 mg/kg pentobarbital sodium (Pentobarbitalnatrium vet.; Apotek Produktion & Laboratorier AB, Kungens kurva, Sweden) in combination with 0.5–0.75 µg/kg Fentanyl (Fentanyl; Orion Pharma AB, Danderyd, Sweden) as needed. Animals were intubated with endotracheal tubes connected to automatic ventilators. Vital parameters were monitored by pulse oximetry, capnography, and oral thermometer. Signs of postoperative pain were treated with IM administration of 50–75 µg fentanyl and 40 mg meloxicam (Loxicom; N-Vet AB, Uppsala, Sweden). The dorsum of the pigs was waxed and shaved. Skin was disinfected with 10% povidone-iodine scrub (Jodopax vet.; Pharmaxim AB, Helsingborg, Sweden). The spine was marked with a pen to allow accurate placement of wounds. For the pig predetermined for IM gentamicin treatment, 20 wounds full thickness circular wounds down to the panniculus carnosus, each measuring 78.5 mm2, was created using a 10 mm punch biopsy tool (Acu-Punch; Acuderm Inc, Fort Lauderdale, FL, US). For the pig predetermined for topical honey treatment, 30 wounds were created in identical fashion (Fig. 1A). The 10 additional wounds created on the pig predetermined for honey treatment was due to the anticipated exclusion of wounds due to leakage of the applied honey. For both pigs, wounds were separated with 4 cm of healthy tissue to prevent cross-contamination between adjacent wounds. All wounds were inoculated with 108 colony-forming units (CFU) of S. aureus (ATCC 29,213, Manassas, VA, US) diluted in 1 ml saline by tunnelling through approx. 1 cm of adjacent health tissue with an 18-gauge needle (Fig. 1B). This method of inoculation was chosen to avoid leakage and cross-contamination. Wounds were covered with transparent and occlusive film (Tegaderm; Mölnlycke, Sweden) and an elastic dressing (Elastic bandage, Hansbo sport, Västra Frölunda, Sweden) (Fig. 1A, D). On days 0, 2, 4, 7, and 10 after inoculation, venous blood was drawn from ear or groin veins and collected in 2 ml EDTA tubes, and 3 ml polymer gel and lithium heparin tubes (BD Vacutainer; Becton, Dickinson and Company, Franklin Lakes, NJ, US) and oral temperature was measured. All wounds were macroscopically inspected and photographed with a ruler for later area calculations. On days 2, 4, 7, and 10, biopsies were taken with a 4 mm punch biopsy tool (Biopsy punch; Paramount Surgimed Ltd, New Delhi, India) and scalpel with a margin of approx. 3 mm and involving 50% of the wound circumference. Both biopsies included healthy underlying tissue. Punch biopsies were placed in dry, sterile cryogenic vials and frozen. Scalpel biopsies were immediately fixated in 4% neutral-buffered paraformaldehyde (HistoLab, Askim, Sweden). Biopsied wounds were subsequently excluded. 6–8 wounds were biopsied at each time point from the pig receiving topical honey treatment, and 5 wounds were biopsied at each time point from the pig receiving IM gentamicin. On day 10, the pigs were euthanized by intravenous injection of 400 mg/kg pentobarbital sodium (Pentobarbitalnatrium vet.; Apotek Produktion & Laboratorier AB, Kungens kurva, Sweden).
      Fig. 1
      Fig. 1Illustrations of animal procedures (A) Overview of wounds before inoculation (day 0). (B) Inoculation procedure. (C) Application of honey. (D) Wounds covered with occlusive and elastic dressing prior to terminating anaesthesia.

       Treatments

      For the pig with wounds predetermined for honey, treatment was initiated on day 2 after inoculation by completely filling wound cavities with approx. 2 ml undiluted honey per wound (Fig. 1C). On days 4 and 7 after inoculation, all wound cavities were superficially cleaned with cotton gauze to remove the honey and any wound secretion. Wound cavities were then immediately refilled with undiluted honey (Fig. 1C). This interval and method were chosen to mimic dressing change procedures in clinical settings. For the pig with wounds predetermined for IM gentamicin, treatment was also initiated on day 2 after inoculation with 5 mg/kg at 12 h intervals for the first day and subsequently once daily until day 10 when the experiment was terminated. Wound cavities were cleaned at the same time points and in an identical fashion as the honey-treated wounds. The administration route and length of gentamicin treatment were set to resemble treatment protocols used in LMIC and armed conflicts [
      • Giannou C.
      • Baldan M.
      ,
      Clinical guidelines: diagnostic and treatment manual. Paris.
      ]. Since no universally accepted guidelines for gentamicin treatment in pigs exists, dosage and interval were set after the recommendation from a gentamicin manufacturer (Biovet JSC, Peshtera, Bulgaria), European Agency for Evaluation of Medical Products (EMEA), and veterinarian's advice [
      Committee of veterinary medicinal products
      Gentamicin, summery report (EMEA/MRL/803/01-Final). London: The European agency for the evaluation of medicinal products.
      ]. Gentamicin was administrated to the right lateral side of neck or to the hind to avoid interaction with other IM medications, such as the anaesthetic, which was administrated to the left lateral side of neck.

       Quantitative S. aureus cultures in wound tissue

      The frozen 4 mm punch biopsies were thawed, weighed, homogenized, serially diluted, and plated overnight on Mannitol salt agar plates and counted for S. aureus colonies yielding log10 colony-forming units per gram of tissue (CFU/g).

       Control group for quantitative S. aureus cultures in wound tissue

      As a control, we used data from saline-treated wounds obtained from a previous study done by our group (unpublished data). The method and materials used in this study were identical to the present study, except that the wound cavities were filled with sterile saline 0.9% solution. No other treatment was given. 8–10 wounds on days 2, 4, 7, and 10 after inoculation were biopsied and cultured in the same manner as in the present study.

       Measurement of wound surface area

      Wound diameter was measured using ImageJ software 1.53f (National Institutes of Health, Bethesda, Maryland, US). The diameter was defined as border to border of macroscopically healthy skin. The largest diameter for all wounds was used for the calculation of area.

       Tissue preparation

      The fixed scalpel biopsies were dehydrated by immersion in a series of ethanol–xylene, embedded in paraffin, cut into 8–10 µm thick sections using a microtome (RM2255; Leica Biosystems, Wetzlar, Germany) and mounted on microscopy slides for subsequent staining.

       Histological examination and reepithelialization

      Slides from all wounds and time points were deparaffinized in a xylene-ethanol series, rehydrated and stained with Haematoxylin and eosin (H&E) (Sigma-Aldrich Sweden AB, Stockholm, Sweden), and mounted. Two evaluators independently assessed all slides for the appearance of neoepidermis in the central part of the wound. Representative images were chosen.

       Immunohistochemical analysis

      Staining for S. aureus was done with the Vectastin Elite ABC HRP Kit (Vector Laboratories Inc, Burlingame, CA, US). Following deparaffinization and rehydration, slides were incubated with normal goat serum (15 µl/ml; Vector Laboratories Inc., Burlingame, CA, US) in PBS, and thereafter with rabbit polyclonal antibodies for enterotoxins from S. aureus as validated immunogen (dilution 1:800, 4–5 mg/ml; Thermo Fisher Scientific Inc., Waltham, MA, US) for 30 min at room temperature [
      • Zhu X.
      • Liu D.
      • Singh A.K.
      • Drolia R.
      • Bai X.
      • Tenguria S.
      • et al.
      Tunicamycin mediated inhibition of wall teichoic acid affects Staphylococcus aureus and Listeria monocytogenes cell morphology, biofilm formation and virulence.
      ]. Following several rinses in PBS, slides were incubated with biotinylated goat anti-rabbit IgG antibodies (15 µl/ml; Vector Laboratories Inc, Burlingame, CA, US) for 30 min at room temperature. After subsequently washing in PBS, slides were incubated with an avidine-horseradish peroxidase complex and developed using Vector VIP (Vector Laboratories Inc., Burlingame, CA, US) as peroxidase substrate. The stained slides were examined using a light microscope (BX41; Olympus Corp., Tokyo, Japan) and images captured with a CCD camera (MTR3CDD, TouoTek Photonics Corp., Zhejiang, China).
      Staining for macrophages was done in a similar fashion with normal horse serum (15 µl/ml; Vector Laboratories Inc., Burlingame, CA, US), mouse CD68 antibodies with a subcellular fraction of human macrophages as validated immunogen (dilution 1:200, 0.2 mg/ml; Thermo Fisher Scientific Inc., Waltham, MA, US), biotinylated horse anti-mouse IgG antibodies (15 µl/ml, Vector Laboratories Inc., Burlingame, CA, US), and an avidine–horseradish peroxidase complex, using Vector VIP (Vector Laboratories Inc., Burlingame, CA, US) as peroxidase substrate [
      • Finkin S.
      • Yuan D.
      • Stein I.
      • Taniguchi K.
      • Weber A.
      • Unger K.
      • et al.
      Ectopic lymphoid structures function as microniches for tumor progenitor cells in hepatocellular carcinoma.
      ].

       Analysis of macrophages in wound tissue

      The CD68 stained slides from all wounds were examined using a light microscope (BX41; Olympus Corp.), and images were captured with a CCD camera (MTR3CDD, TouoTek Photonics Corp.). Two representative images from each treatment arm and time point were chosen by the authors. Chosen images were blinded, and the relative concentration of macrophages, 1–3 (low–high), was visually assessed by five independent experienced evaluators through a web-based questionnaire.

       Analysis of systemic parameters

      The EDTA tubes were immediately sent to Evidensia Valla Djursjukhus (Linköping, Sweden) for analysis of haemoglobin, total count of white blood cells, and neutrophils. The IDEEX ProCyte DX (IDEEX Laboratories Inc., Westbrook, Maine, US) was used for the analysis. The polymer gel and lithium heparin tubes were centrifuged for 10 min at 2000 rpm. The serum was then extracted and frozen at −20 °C for later analysis of serum amyloid A (SAA). SAA is an acute-phase protein, similar to C reactive protein, and will rapidly increase in the event of trauma, infection, or other systemic stressors in pigs [
      • Cray C.
      • Zaias J.
      • Altman N.H.
      Acute phase response in animals: a review.
      ]. SAA concentration was assessed using the PHASE RANGE Multispecies SAA ELISA Kit (Tridelta Development Ltd., Maynooth, Ireland). The manufacturer's protocol was followed. From each time point and treatment arm, 4 samples at a dilution of 1:500 were analysed. The absorbence of each sample and standards were read at 450 nm and also at 630 nm to adjust for background using a microplate spectrometer (Epoch2; BioTek, Winooski, Vermont, US). A constant was calculated by plotting the absorbence of the adjusted standards at 450 nm against their known concentrations on a standard graph. Concentrations of SAA for the samples were then calculated by multiplying the adjusted absorbence value with the constant. 1 out of 80 samples was excluded due to absorbence considerably deviating from other samples from the same time point (day 0) and treatment arm (topical honey). The ELISA was repeated with two samples with a dilution of 1:500 and one sample with a dilution of 1:1000 from each time point and treatment arm with confirming results.

       Statistical analysis

      Power calculation was done with G*Power 3.1 (Heinrich Heine Universität, Düsseldorf, Germany) [
      • Faul F.
      • Erdfelder E.
      • Lang A.G.
      • Buchner A.
      G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences.
      ]. The primary endpoint of reduction of bacteria in tissue after eight days of treatment was used for the power calculation. Reduction to ≤3 log10 CFU/g of tissue after eight days of treatment was considered successful. The rationale being that the potential of wound healing is significantly impaired, i.e., infected, with bacterial bioburden of >5 log10 CFU/g, and the bacterial bioburden of non-infected skin is considered to be 2–3 log10 CFU/g [
      • Robson M.C.
      • Lea C.E.
      • Dalton J.B.
      • Heggers J.P.
      Quantitative bacteriology and delayed wound closure.
      ,
      • Bendy R.H.
      • Nuccio P.A.
      • Wolfe E.
      • Collins B.
      • Tamburro C.
      • Glass W.
      • et al.
      Relationship of quantitative wound bacterial counts to healing of decubiti: effect of topical gentamicin.
      ]. Solitary treatment with antibiotics and topical honey have separately been demonstrated to reduce the bacterial bioburden in wounds from >5 log10 CFU/g to ≤3 log10 CFU/g within eight days [
      • Scher K.S.
      • Peoples J.B.
      Combined use of topical and systemic antibiotics.
      ,
      • Erbil B.
      • Ersoy G.
      • Ozkutuk A.
      • Akarca F.K.
      • Korkmaz T.
      • Demir O.F.
      • et al.
      The effects of oral antibiotics on infection prophylaxis in traumatic wounds.
      ,
      • Medeiros Vde F.
      • Azevedo I.M.
      • Rego A.C.
      • Egito E.S.
      • Araujo-Filho I.
      • Medeiros A.C.
      • et al.
      Antibacterial properties and healing effects of Melipona scutellaris honey in MRSA-infected wounds of rats.
      ]. We expected that a reduction to ≤3 log10 CFU/g of tissue would be seen in 95% of the wounds for both the honey and antibiotic treatment arms after eight days of treatment. With a non-inferiority margin of 0.21, to allow for error in the application and leakage of honey, we calculated that 19 samples were needed in each of the two treatment arms to demonstrate a non-inferiority at 90% power. Therefore, five samples were needed at each of the four time points to achieve a total sample size of 20 for each treatment arm and to allow for some margin of error. In total 20 wounds were created on the pig predetermined for IM gentamicin treatment and 30 wounds on pig predetermined for topical honey treatment. As previously mentioned, the difference in number of the wounds for each treatment arm was due to the expected leakage of the applied honey. Data was collected from all wounds.
      Statistical analysis was performed using GraphPad Prism 8.4 (GraphPad Software, San Diego, CA, US). Values are given as mean ± standard deviation (SD), except for the results on the quantitative cultures of S. aureus in wound tissue over time (Fig. 2) where instead geometrical mean *÷ geometric standard deviation (GSD) are given. This was done to include all observations and to diminish misleading visual effects caused by the log10 scale. Statistical comparisons between treatment groups were performed using a mixed-model ANOVA with Greenhouse-Geiser correction. P-values (two-tailed) less than 0.05 were considered significant.
      Fig. 2
      Fig. 2Quantitative S. aureus cultures from wound biopsies S. aureus infected wounds treated with either topical honey or intramuscular (IM) gentamicin were analysed for amount of bacteria, in colony-forming units/gram of tissue (CFU/g). IM gentamicin significantly reduced bacterial count after two days of treatment (day 4), but not in the following time points. Topical honey did not reduce bacterial count at any time point. Cultures from wounds treated with saline, from a previous study, were included as a reference (unpublished data). Clinical infection was defined as bacterial count >105 per gram of tissue [
      • Cray C.
      • Zaias J.
      • Altman N.H.
      Acute phase response in animals: a review.
      ,
      • Faul F.
      • Erdfelder E.
      • Lang A.G.
      • Buchner A.
      G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences.
      ]. Symbols and error bars represent geometrical mean and geometric standard deviation (GSD).

      Results

       Quantitative cultures of S. aureus in wound tissue

      Two days after inoculation, prior to initiation of treatment (day 2), all wounds exhibited S. aureus counts consistent with infection (>5 log10 CFU/g tissue), with geometric means 5.85 *÷ 0.55 (GSD) log10 CFU/g (n = 6) and 5.59 *÷ 0.37 (GSD) log10 CFU/g (n = 3) for wounds subsequently treated with topical honey or systemic gentamicin, respectively (Figs. 2 and 3). Treatment with topical honey did not reduce S. aureus counts (n = 6–8/time point) at any of the time points during the experiment. IM gentamicin administration resulted in a significant decline in S. aureus count after two days of treatment (day 4), geometric mean 5.59 *÷ 0.37 (GSD) to 4.27 *÷ 0.80 log10 CFU/g (n = 5), but thereafter counts increased and remained above the level of infection until terminal endpoint. There was no difference in the bacterial count at the terminal endpoint (day 10) when comparing topical honey treatment, IM Gentamicin treatment with saline treatment. In the mixed-model ANOVA no significant difference in variation of S. aureus counts between treatment with topical honey and IM gentamicin was observed (p = 0.56).
      Fig. 3
      Fig. 3Staining for S. aureus in wound tissue. Immunohistochemistry with polyclonal antibodies for enterotoxins from S. aureus on biopsies from S. aureus infected wounds treated with either topical honey or intramuscular gentamicin. Infiltration of S. aureus in wound tissue was observed in wounds from both treatment groups by day 4, indicated by arrows in the figure. Scale bars represent 500 µm.

       Reduction of wound surface area

      Two days after inoculation and prior to initiation of treatment (day 2) an increase of mean wound area was seen in all wounds, with means 78.5 mm2 to 118.2 mm2 ± 39.2 (SD) (n = 30) and 78.5 mm2 to 112.8 mm2 ± 30.0 (SD) (n = 20) for wounds later treated with topical honey or IM gentamicin, respectively (Figs. 4 and 5). Treatment with topical honey did not cause any changes in the size of the mean wound area, as it remained largely unchanged until the terminal endpoint (day 10), mean 112.2 mm2 ± 10.4 (SD) (n = 10). However, treatment with IM gentamicin resulted in a significant decrease in mean wound area observed at terminal endpoint (day 10), mean 112.8 mm2 ± 30.0 (SD) (n = 20) to 67.7 mm2 ± 13.2 (n = 5), compared to wounds treated with topical honey (p<0.0001).
      Fig. 4
      Fig. 4Surface area of wounds over time . S. aureus infected wounds were treated with either topical honey or intramuscular (IM) gentamicin. Wounds treated with IM gentamicin exhibit a significant decrease of wound area at terminal endpoint (day 10), compared to wounds treated with topical honey (p<0.0001). The largest diameter was used for calculation of area. The diameter was defined as border to border of macroscopically healthy skin. Symbols and error bars represent mean and standard deviation (SD).
      Fig. 5
      Fig. 5Appearance of wounds over time. S. aureus infected wounds were treated with either topical honey or intramuscular gentamicin. Scale bars represent 5 mm.

       Reepithelialization

      IM gentamicin, but not topical honey, treated wounds displayed ongoing reepithelialization at the terminal endpoint (day 10) (Fig. 6).
      Fig. 6
      Fig. 6Haematoxylin and eosin-staining of wounds S. aureus infected wounds were treated with either topical honey or intramuscular (IM) gentamicin. IM gentamicin, but not topical honey, treated wounds displayed ongoing reepithelialization at the terminal endpoint (day 10). Indicated by an arrow in the figure. Scale bars represent 500 µm.

       Local inflammatory response

      A tendency towards higher concentrations of macrophages in wound tissue was seen in wounds treated with topical honey from day 4 until the terminal endpoint (day 10) (Fig. 7, Table 1). For wounds treated with IM gentamicin, no substantial changes in the concentration of macrophages were observed until the terminal endpoint (day 10).
      Fig. 7
      Fig. 7Staining for macrophages in wound tissue. Immunohistochemistry with CD68 monoclonal antibodies for a subcellular fraction of macrophages on biopsies from S. aureus infected wounds treated with either topical honey or intramuscular gentamicin. Scale bars represent 500 µm.
      Table 1Visual assessment of macrophages in wound tissue. S. aureus infected wounds were treated with either topical honey or intramuscular (IM) gentamicin. Slides from wound biopsies were stained with CD68 monoclonal antibodies for a subcellular fraction of macrophages. Two slides per treatment group and time point were visually assessed by five independent experienced evaluators through a web-based questionnaire. Wounds treated with topical honey demonstrated a tendency towards higher concentration of macrophages in wound tissue from day 4 until terminal endpoint (day 10), compared to wounds treated with IM gentamicin.
      Concentration of macrophages

      Estimated, 1 (low) – 3 (high)
      Treatment initiatedTerminal endpoint
      Time points (days)2

      n = 10

      mean (SD)
      4

      n = 10

      mean (SD)
      7

      n = 10

      mean (SD)
      10

      n = 10

      mean (SD)
      Treatment
      Topical honey1.3 (0.5)2.2 (0.6)2.0 (0.9)2.9 (0.3)
      IM gentamicin1.9 (0.9)2.1 (0.6)1.7 (0.7)2.1 (1.0)

       Systemic inflammatory response

      From two days after inoculation and until the terminal endpoint (day 10) both pigs exhibited behaviour consistent with illness: loss of appetite, lethargy, and refraining from nesting. Consistent with an acute systemic inflammatory response, an increase in oral temperature, concentration of SAA, concentrations of total leukocytes, neutrophils, and a decrease in haemoglobin concentration was observed in both pigs two days after inoculation and prior to initiation of treatment (day 2) (Fig. 8A–E) [
      • Robertson C.M.
      • Coopersmith C.M.
      The systemic inflammatory response syndrome.
      ]. SAA increased from mean 23.5 µg/ml ± 78.1 (SD) to 947.1 µg/ml ± 58.2 and 94.7 µg/ml ± 159.4 (SD) to 967.4 µg/ml ± 131.6 for pigs receiving topical honey or IM gentamicin, respectively (Fig. 8B). Following treatment with topical honey, SAA remained elevated after two days of treatment (day 4), mean 969.7 µg/ml ± 58.2 (SD), and a slower decline of SAA was seen compared to treatment with IM gentamicin. For the pig that received IM gentamicin treatment, SAA concentration decreased after 2 days of treatment (day 4), 967.4 µg/ml ± 131.6 (SD) to 647.9 µg/ml ± 194.4 and returned to within normal range after five days of treatment (day 7), 342.0 µg/ml ± 194.3 (SD). These differences in changes of SAA concentrations were confirmed in the mixed model ANOVA (p = 0.04).
      Fig. 8
      Fig. 8Systemic immune response in pigs with S. aureus infected wounds. Wounds were treated with either topical honey or intramuscular gentamicin on separate pigs. Two days after inoculation (day 2), both pigs exhibit systemic measurements consistent with an acute systemic inflammatory response. This response was more persistent for the duration of treatment for the pig that received topical honey treatment. (A) Oral temperature was taken prior to tissue sampling at each time point. Normal temperature range modified from Soerensen et la. Acta Vet Scand 2015;57:5. (B) SAA was analysed from serum using the PHASE RANGE Multispecies SAA ELISA kit (Tridelta Development Ltd.). Normal range according to manufacturer. Symbols and error bars represent mean and standard deviation (SD). (C–E) Leukocytes, neutrophils, and haemoglobin were analysed from blood with the IDEEX ProCyte DX (IDDEX Laboratories Inc.). Normal ranges according to the manufacturer.
      Concentrations of total leukocytes and neutrophils normalized for both pigs after two days of treatment (day 4) (Fig. 8C, D). For the pig with wounds treated with topical honey, these concentrations subsequently increased until the terminal endpoint (day 10). For the pig with IM gentamicin treated wounds total leukocytes and neutrophils remained normal after 5 days of treatment (day 7), and then increased at the terminal endpoint (day 10). Concentrations of haemoglobin remained depressed for both pigs until the terminal endpoint (day 10) (Fig. 8E).

      Discussion

      Despite honey's established antimicrobial properties in vitro models, similar data from in vivo models and humans are scarce [
      • Jull A.B.
      • Cullum N.
      • Dumville J.C.
      • Westby M.J.
      • Deshpande S.
      • Walker N.
      • et al.
      Honey as a topical treatment for wounds.
      ,
      • Hixon K.R.
      • Klein R.C.
      • Eberlin C.T.
      • Linder H.R.
      • Ona W.J.
      • Gonzalez H.
      • et al.
      A critical review and perspective of honey in tissue engineering and clinical wound healing.
      ]. In a study by Medeiros et al., a significant reduction of the bacterial count could be demonstrated with Melipona scutellaris honey treatment compared to saline treatment of methicillin-resistant S. aureus (MRSA) infected wounds in rats [
      • Medeiros Vde F.
      • Azevedo I.M.
      • Rego A.C.
      • Egito E.S.
      • Araujo-Filho I.
      • Medeiros A.C.
      • et al.
      Antibacterial properties and healing effects of Melipona scutellaris honey in MRSA-infected wounds of rats.
      ]. In this study, inoculation was done through application of a gauze with 5 × 107 CFU of MRSA (ATCC 43,300) in the wounds, which were then closed for 24 h. Wounds were then opened, cavities drained, and treatment was initiated. The bacterial count was assessed only by quantitative cultures. No immunohistochemical analyses were done to assess bacterial infiltration in the tissue. Given the method of inoculation and the short period before initiation of treatment in Medeiros study, there is reason to suspect that the infection of MRSA was superficial and therefore accessible to topical treatment.
      In the present study, we choose to inoculate with a S. aureus strain without any acquired drug resistance (ATCC 29,213). The reason for not inoculating with a multidrug-resistant (MDR) bacterium, such as MRSA, was to avoid introducing a confounder that could have had a significant impact when establishing honey's efficacy in a complex wound model where previous data is lacking. The porcine model was chosen as it is a pragmatic approach to simulate the pathophysiology of infected traumatic wounds seen in humans without the interference of confounding factors such as co-morbidity, nutritional state, time since injury and mechanism of injury that otherwise could have influenced the results [
      • Sullivan T.P.
      • Eaglstein W.H.
      • Davis S.C.
      • Mertz P.
      The pig as a model for human wound healing.
      ,
      • Junker J.P.E.
      • Lee C.C.Y.
      • Samaan S.
      • Hackl F.
      • Kiwanuka E.
      • Minasian R.A.
      • et al.
      Topical delivery of ultrahigh concentrations of gentamicin is highly effective in reducing bacterial levels in infected porcine full-thickness wounds.
      ,
      • Tsai D.M.
      • Tracy L.E.
      • Lee C.C.
      • Hackl F.
      • Kiwanuka E.
      • Minasian R.A.
      • et al.
      Full-thickness porcine burns infected with Staphylococcus aureus or Pseudomonas aeruginosa can be effectively treated with topical antibiotics.
      ]. Furthermore, to better resemble infected wounds in armed conflict we choose a different inoculation method than the one used in Medeiros study. Inoculation was done by tunnelling through adjacent healthy tissue, wounds were not closed and more importantly treatment was initiated after 48 h. In our model, we did not observe any reduction in bacterial count in wounds treated with topical honey for the duration of the study (Fig. 2). This is most likely due to honey's inability to reach bacteria that had infiltrated in the tissue. Tissue infiltration of S. aureus was confirmed in the immunohistochemical analysis as we could identify growth of bacteria in the wound tissue and not only on the wound's surface (Fig. 3). For the wounds treated with IM gentamicin, a reduction in the bacterial count was observed after two days of treatment (day 4), which included the first day of treatment when gentamicin was administered twice compared to once daily for the remaining treatment. This reduction in the bacterial count was not sustained for the duration of the study, conceivably attributed to gentamicin's inability to reach bacteria in devitalized tissue, the dosage of antibiotics was too low or a combination thereof. Serum concentrations of gentamicin could have been measured to confirm that a therapeutic dosage was given. However, according to a report by EMEA the pharmacokinetics of gentamicin in pigs is poorly studied, and to our knowledge no established therapeutic concentrations exists [
      Committee of veterinary medicinal products
      Gentamicin, summery report (EMEA/MRL/803/01-Final). London: The European agency for the evaluation of medicinal products.
      ]. To mitigate this issue, we choose to give gentamicin in a dosage in the higher end of the recommendations from EMEA [
      Committee of veterinary medicinal products
      Gentamicin, summery report (EMEA/MRL/803/01-Final). London: The European agency for the evaluation of medicinal products.
      ].
      Consistent with the lack of reduction in bacterial count, we did not observe a reduction in wound surface area with topical honey treatment. Conversely, we did not observe an increase in wound surface area, which would be expected if honey did not have any effect on the bacterial activity. Honey could therefore have exerted an antimicrobial effect on the bacteria colonizing the surface of the wound, but not on the bacteria that had infiltrated the tissue. For wounds treated with IM gentamicin, we observed a reduction in wound surface area and signs of reepithelization (Figs. 5 and 6). This was probably because IM gentamicin was able to inhibit the growth of S. aureus in surrounding tissue and thereby enable wound repair. Our negative findings regarding the efficacy of honey on the healing of infected wounds are in line with the conclusion of the Cochrane review from 2015, as only one study was identified that could provide moderate evidence for honey's efficacy as treatment for infected wounds in humans [
      • Al-Waili N.S.
      • Saloom K.Y.
      Effects of topical honey on post-operative wound infections due to gram positive and gram negative bacteria following caesarean sections and hysterectomies.
      ]. In this randomized controlled trial, by Al-Waili et al., severely infected wounds following caesarean section or total abdominal hysterectomies were treated with either topical honey or antiseptic washes. Wounds treated with honey dressings exhibited a shorter time to eradication of bacterial infection, wound healing, and smaller scar formation [
      • Al-Waili N.S.
      • Saloom K.Y.
      Effects of topical honey on post-operative wound infections due to gram positive and gram negative bacteria following caesarean sections and hysterectomies.
      ]. However, both groups were treated with systemic antibiotics, and studies have indicated that the chosen antiseptic may have a negative effect on wound healing [
      • Leaper D.J.
      • Simpson R.A.
      The effect of antiseptics and topical antimicrobials on wound healing.
      ,
      • Atiyeh B.S.
      • Dibo S.A.
      • Hayek S.N.
      Wound cleansing, topical antiseptics and wound healing.
      ]. Therefore, it is unclear if Al-Waili et al. could demonstrate any additional benefit with topical honey compared to systemic antibiotics only.
      The WHO has stated that: “the global community must encourage sustainable investments in new medicines, diagnostic tools, vaccines, and alternative interventions. The majority of pharmaceutical companies are no longer researching a new antibiotic which is a global concern for human and animal health. Research and development are needed to produce new treatments that can be deployed against multi-drug resistant infections” [
      World Health Organization
      ]. As a result of the ever-spreading antimicrobial resistance, an increasing number of non-healing wounds globally are infected with MDR bacteria, causing morbidity and adding burden to already strained health care systems, especially in armed conflicts [
      • Eardley W.G.
      • Brown K.V.
      • Bonner T.J.
      • Green A.D.
      • Clasper J.C.
      Infection in conflict wounded.
      ,
      • Mulani M.S.
      • Kamble E.E.
      • Kumkar S.N.
      • Tawre M.S.
      • Pardesi K.R.
      Emerging strategies to combat ESKAPE pathogens in the era of antimicrobial resistance: a review.
      ,
      • Sahli Z.T.
      • Bizri A.R.
      • Abu-Sittah G.S.
      Microbiology and risk factors associated with war-related wound infections in the Middle East.
      ]. A recent cohort study of Syrian patients with conflict-related injuries treated in Jordan found MDR bacteria in excess of 70% of infected wounds and was associated with a poorer outcome and high resource consumption [
      • Alga A.
      • Wong S.
      • Shoaib M.
      • Lundgren K.
      • Giske C.G.
      • von Schreeb J.
      • et al.
      Infection with high proportion of multidrug-resistant bacteria in conflict-related injuries is associated with poor outcomes and excess resource consumption: a cohort study of Syrian patients treated in Jordan.
      ]. This highlights the need to identify treatment strategies that do not rely on antibiotics. Honey's antimicrobial properties, low cost, almost indefinite shelf life, and readily availability worldwide makes it an ideal candidate. Moreover, honey has been demonstrated to be efficient against MDR bacteria in vitro studies [
      • Girma A.
      • Seo W.
      • She R.C.
      Antibacterial activity of varying UMF-graded Manuka honeys.
      ,
      • Glasser J.S.
      • Guymon C.H.
      • Mende K.
      • Wolf S.E.
      • Hospenthal D.R.
      • Murray C.K.
      Activity of topical antimicrobial agents against multidrug-resistant bacteria recovered from burn patients.
      ].
      The antimicrobial properties of honey are not fully understood and will vary depending on floral source. However, most honey types share the following antimicrobial properties: 1. Hydrogen peroxide activity causing a breakdown of bacterial walls. 2. Hyperosmolarity that draws moisture from the wound and bacteria, causing a decrease of water activity (free water) to 0.56–0.59. 3. Acid pH ranging between 3.2 and 4.5, causing a hostile environment for bacteria. 4. Increase phagocytosis and bacterial destruction by stimulating macrophage activity and cytokines (IL-1β, IL-6 and TNF-α) production. 5. Promoting healing by oxygenation, angiogenesis, and fibroblast proliferation [
      • Saikaly S.K.
      • Khachemoune A.
      Honey and wound healing: an Update.
      ,
      • Hixon K.R.
      • Klein R.C.
      • Eberlin C.T.
      • Linder H.R.
      • Ona W.J.
      • Gonzalez H.
      • et al.
      A critical review and perspective of honey in tissue engineering and clinical wound healing.
      ].
      The hydrogen peroxide activity is a crucial component of honey's antimicrobial effect. Hydrogen peroxide is produced by breakdown of glucose by the enzyme glucose oxidase. Non-active glucose oxidase is a natural component of honey. Its activation is pH-dependant (5.5–8.0) and will not occur unless the honey (pH 3.2–4.5) is exposed to a more basic environment, such as a wound. Hydrogen peroxide eliminates bacteria through oxidation that cause DNA damage and change the permeability of the bacterial wall [
      • Saikaly S.K.
      • Khachemoune A.
      Honey and wound healing: an Update.
      ,
      • Hixon K.R.
      • Klein R.C.
      • Eberlin C.T.
      • Linder H.R.
      • Ona W.J.
      • Gonzalez H.
      • et al.
      A critical review and perspective of honey in tissue engineering and clinical wound healing.
      ,
      • Nolan V.C.
      • Harrison J.
      • Wright J.E.E.
      • Cox J.A.G.
      Clinical significance of Manuka and medical-grade honey for antibiotic-resistant infections: a systematic review.
      ]. Some bacteria, such as S. aureus, produce the enzyme catalase that breakdown hydrogen peroxide [
      • Foster T.
      • Baron S.
      Staphylococcus.
      ]. However, depending on the floral source of the honey, phytochemical components can add antimicrobial activity that is stable in the presence of catalase. This non-peroxide activity has been demonstrated in Manuka honey from New Zealand used in the present study [
      • Johnston M.
      • McBride M.
      • Dahiya D.
      • Owusu-Apenten R.
      • Nigam P.S.
      Antibacterial activity of Manuka honey and its components: an overview.
      ]. Manuka honey was chosen due to its established broad antimicrobial spectrum, including commonly found wound pathogens in armed conflicts [
      • Cooper R.A.
      • Molan P.C.
      • Harding K.G.
      Antibacterial activity of honey against strains of Staphylococcus aureus from infected wounds.
      ,
      • Eardley W.G.
      • Brown K.V.
      • Bonner T.J.
      • Green A.D.
      • Clasper J.C.
      Infection in conflict wounded.
      ,
      • Girma A.
      • Seo W.
      • She R.C.
      Antibacterial activity of varying UMF-graded Manuka honeys.
      ,
      • Nolan V.C.
      • Harrison J.
      • Wright J.E.E.
      • Cox J.A.G.
      Clinical significance of Manuka and medical-grade honey for antibiotic-resistant infections: a systematic review.
      ,
      • Patton T.
      • Barrett J.
      • Brennan J.
      • Moran N.
      Use of a spectrophotometric bioassay for determination of microbial sensitivity to Manuka honey.
      ]. The Manuka honey is a monofloral honey produced by the Apis mellifera honeybees that collect nectar from the flowers of the Manuka tree (Leptospermum scoparium). Manuka honey contains high concentrations of methylglyoxal (MGO), which has been demonstrated to be strongly antimicrobial, even in the presence of catalase [
      • Cooper R.A.
      • Molan P.C.
      • Harding K.G.
      Antibacterial activity of honey against strains of Staphylococcus aureus from infected wounds.
      ,
      • Irish J.
      • Blair S.
      • Carter D.A.
      The antibacterial activity of honey derived from Australian flora.
      ,
      • Johnston M.
      • McBride M.
      • Dahiya D.
      • Owusu-Apenten R.
      • Nigam P.S.
      Antibacterial activity of Manuka honey and its components: an overview.
      ,
      • Hixon K.R.
      • Klein R.C.
      • Eberlin C.T.
      • Linder H.R.
      • Ona W.J.
      • Gonzalez H.
      • et al.
      A critical review and perspective of honey in tissue engineering and clinical wound healing.
      ,
      • Carter D.A.
      • Blair S.E.
      • Cokcetin N.N.
      • Bouzo D.
      • Brooks P.
      • Schothauer R.
      • et al.
      Therapeutic Manuka honey: no longer So alternative.
      ]. The concentration of MGO will vary in different kinds of Manuka honey and is measured in Unique Manuka Factor (UMF, range 0–25+), which is equivalent to% (w/v) phenol in antimicrobial activity against S. aureus in an agar well diffusion assay [
      • Girma A.
      • Seo W.
      • She R.C.
      Antibacterial activity of varying UMF-graded Manuka honeys.
      ,
      • Allen K.L.
      • Molan P.C.
      • Reid G.M.
      A survey of the antibacterial activity of some New Zealand honeys.
      ]. However, how MGO exerts its antimicrobial effect is not fully understood. Studies using scanning electron microscopy have demonstrated loss of flagella and fimbriae in E. coli and B. subtilis, in the presence of MGO [
      • Rabie E.
      • Serem J.C.
      • Oberholzer H.M.
      • Gaspar A.R.
      • Bester M.J.
      How methylglyoxal kills bacteria: an ultrastructural study.
      ]. A shrinking and rounding of these rod-shaped bacteria were also observed [
      • Rabie E.
      • Serem J.C.
      • Oberholzer H.M.
      • Gaspar A.R.
      • Bester M.J.
      How methylglyoxal kills bacteria: an ultrastructural study.
      ].
      Consequently, higher UMF has been demonstrated to correlate with lower minimum inhibitory concentration (MIC) for flagella and fimbriae-carrying bacteria, such as E. coli and P. aeruginosa [
      • Tan H.T.
      • Rahman R.A.
      • Gan S.H.
      • Halim A.S.
      • Hassan S.A.
      • Sulaiman S.A.
      • et al.
      The antibacterial properties of Malaysian tualang honey against wound and enteric microorganisms in comparison to Manuka honey.
      ,
      • Sherlock O.
      • Dolan A.
      • Athman R.
      • Power A.
      • Gethin G.
      • Cowman S.
      • et al.
      Comparison of the antimicrobial activity of Ulmo honey from Chile and Manuka honey against methicillin-resistant Staphylococcus aureus, Escherichia coli and Pseudomonas aeruginosa.
      ]. The same correlation has not been demonstrated for S. aureus, which is lacking flagella and fimbriae and is coccidal in shape. It has therefore been suggested that Manuka honey exerts its antimicrobial effect on S. aureus through hydrogen peroxide or other phytochemical components, such as the protein bee defensin-1 [
      • Nolan V.C.
      • Harrison J.
      • Wright J.E.E.
      • Cox J.A.G.
      Clinical significance of Manuka and medical-grade honey for antibiotic-resistant infections: a systematic review.
      ]. The MGO concentration (UMF) may therefore have little relevance for the antimicrobial effect on S. aureus. In the present study, a MGO concentration of UMF 15+ was chosen as most FDA-approved Manuka honey dressings are in the range of UMF 13–18 [
      • Molan P.
      • Rhodes T.
      Honey: a biologic wound dressing.
      ]. However, it is worth noting that several in vitro studies have demonstrated that Manuka honey with lower UMF exhibit lower MIC for S. aureus [
      • Nolan V.C.
      • Harrison J.
      • Wright J.E.E.
      • Cox J.A.G.
      Clinical significance of Manuka and medical-grade honey for antibiotic-resistant infections: a systematic review.
      ].
      Our findings of unchanged wound size and S. aureus count in wound tissue suggest that topical Manuka honey may be beneficial to prevent further progression of the wound infection and subsequent complications. This notion is especially interesting in settings where antibiotics and surgical care are not immediately available, such as in armed conflicts. Manuka honey dressings could be used at the site of injury, at trauma stabilization points, and before transports to inhibit the progression of a potential wound infection until definite care can be undertaken. Manuka honey may also constitute a valuable adjunct to antibiotics and surgical debridement. In vitro and ex vivo studies have demonstrated increased antimicrobial activity against P. aeruginosa, S. aureus, and MRSA when using antibiotics combined with Manuka honey [
      • Roberts A.E.L.
      • Powell L.C.
      • Pritchard M.F.
      • Thomas D.W.
      • Jenkins R.E.
      Anti-pseudomonad activity of Manuka honey and antibiotics in a specialized ex vivo model simulating cystic fibrosis lung infection.
      ,
      • Jenkins R.
      • Cooper R.
      Improving antibiotic activity against wound pathogens with Manuka honey in vitro.
      ,
      • Liu M.Y.
      • Cokcetin N.N.
      • Lu J.
      • Turnbull L.
      • Carter D.A.
      • Whitchurch C.B.
      • et al.
      Rifampicin-Manuka honey combinations are superior to other antibiotic-Manuka honey combinations in eradicating Staphylococcus aureus biofilms.
      ]. Furthermore, a study by Jenkins and Cooper showed that subinhibitory concentrations of Manuka honey made MRSA susceptible to oxacillin again, i.e., reverse the resistance [
      • Jenkins R.E.
      • Cooper R.
      Synergy between oxacillin and Manuka honey sensitizes methicillin-resistant Staphylococcus aureus to oxacillin.
      ]. Essential to this synergistic effect of antibiotics and Manuka honey is Manuka honey's ability to interrupt biofilms produced by bacteria, thereby increasing the bacteria's susceptibility to antibiotics [
      • Maddocks S.E.
      • Lopez M.S.
      • Rowlands R.S.
      • Cooper R.A.
      Manuka honey inhibits the development of Streptococcus pyogenes biofilms and causes reduced expression of two fibronectin binding proteins.
      ,
      • Campeau M.E.
      • Patel R.
      Antibiofilm activity of Manuka honey in combination with antibiotics.
      ]. Honey's antimicrobial and antibiofilm activity could also be of benefit in the most fundamental part of the treatment of infected wounds in armed conflicts, the surgical debridement. Applying honey dressings at the end of the debridement may inhibit recolonization of pathogenic bacteria on the wound surface and subsequently hinder redevelopment of devitalized tissue. This could possibly result in a diminished need for repeated debridements and a shorter time to achieve wound closure. However, this needs to be extensively evaluated in vivo studies before honey can become relevant in the treatment of infected wounds in armed conflicts.

       Limitations

      The lack of previous studies regarding honey's efficacy as a treatment for infected wounds in humans and in vivo models poses restraints on designing a study that could have direct implications for how infected wounds are treated. It can be expected that the antimicrobial effect of honey previously reported from primarily in vitro models will be diminished in more complex in vivo models, such as the porcine model used, due to honey's inability to reach bacteria that have infiltrated into the tissue [
      • Cooper R.A.
      • Molan P.C.
      • Harding K.G.
      Antibacterial activity of honey against strains of Staphylococcus aureus from infected wounds.
      ,
      • Carter D.A.
      • Blair S.E.
      • Cokcetin N.N.
      • Bouzo D.
      • Brooks P.
      • Schothauer R.
      • et al.
      Therapeutic Manuka honey: no longer So alternative.
      ,
      • Brudzynski K.
      • Sjaarda C.
      Honey glycoproteins containing antimicrobial peptides, Jelleins of the Major Royal Jelly Protein 1, are responsible for the cell wall lytic and bactericidal activities of honey.
      ]. It would therefore have been rational to study honey's efficacy as an adjunct to surgical debridement or antibiotics. However, given the lack of existing data from more complex in vivo models, we found it essential to first quantitatively assess honey's efficacy without any confounding factors, such as surgical debridement or antibiotics. Based on the results from the power calculation and per the principles of the 3Rs (Replacement, Reduction, and Refinement), the lowest number of animals were used to obtain the needed data for a valid statistical comparison between topical honey and IM gentamicin [
      • Russell W.M.S
      • Burch R.L
      The principles of humane experimental technique.
      ]. Nevertheless, the use of only two animals poses limitations in the external validity of the study. The use of different numbers of infected wounds per treatment arm and pig could have affected our comparison of systemic inflammatory response for the two treatments. However, this should not have substantially impacted our primary outcome measure of reduction in bacterial counts per wound at terminal endpoint. The internal validity of the study is therefore robust, but the generalizability is limited.

      Conclusion

      In the porcine wound model used, intramuscular gentamicin was initially more effective than topically applied Manuka honey in reducing S. aureus in infected wounds. However, after eight days of treatment, S. aureus count was the same with both treatments. The wound area decreased with gentamicin and was unchanged with Manuka honey. Topically applied Manuka honey could therefore be non-inferior to intramuscular gentamicin in treating S. aureus colonization on the wound's surface, but not in reducing wound size. The use of Manuka honey dressings to prevent further progression of a wound infection prior to transports may therefore be of value in armed conflicts, where definite care is not immediately available.

      Acknowledgements

      We are immensely grateful to infectious disease consultant Anita Hällgren, MD PhD, for advice on choice of bacteria and antibiotics; Responsible veterinarian Anders Sandberg, DVM, for monitoring animal health; Danne Linghammar for technical assistance and animal care; Researcher Jonathan Rakar, PhD, for scientifical and technical assistance; Annika Starkenberg, MLS, and Kristina Briheim, MLS, for technical assistance; Elvira Lindholm, PhD-student, and Alexander Larsson, MD, for assisting in the assessment of concentration of macrophages; Statistician Lars Valter, MA, for statistical advice; Anna Nordström, DVM, and Per Wallgren, DVM PhD at the National Veterinary Institute, for experimental design; Inger Lilliehöök, DVM PhD at the Swedish University of Agricultural Science for experimental design and Evidensia Valla Animal Hospital for blood sample analysis. Linköping University provided open access funding.

      Funding

      This work was supported by grants from the Kamprad Family Foundation ( 20170287 ) and the Swedish state under the agreement between the Swedish government and the county councils, the ALF-agreement (LIO-700121).

      References

        • Alga A.
        • Wong S.
        • Shoaib M.
        • Lundgren K.
        • Giske C.G.
        • von Schreeb J.
        • et al.
        Infection with high proportion of multidrug-resistant bacteria in conflict-related injuries is associated with poor outcomes and excess resource consumption: a cohort study of Syrian patients treated in Jordan.
        BMC Infect Dis. 2018; 18: 233https://doi.org/10.1186/s12879-018-3149-y
        • Murray C.K.
        • Hinkle M.K.
        • Yun H.C.
        History of infections associated with combat-related injuries.
        J Trauma. 2008; 64: 221-231https://doi.org/10.1097/TA.0b013e318163c40b
      1. (Book)
        • Giannou C.
        • Baldan M.
        War surgery: working with limited resources in armed conflict and other situations of violence. 1. International Committee of the Red Cross, Geneva2019
      2. Clinical guidelines: diagnostic and treatment manual. Paris.
        (Book) Médecins sans frontières, 2019
        • Leaning J.
        • Guha-Sapir D.
        Natural disasters, armed conflict, and public health.
        N Engl J Med. 2013; 369: 1836-1842https://doi.org/10.1056/NEJMra1109877
        • Matowe L.W.
        • Waako P.
        • Adome R.O.
        • Kibwage I.
        • Omary M.
        • Bienvenu E.
        A strategy to improve skills in pharmaceutical supply management in East Africa: the regional technical resource collaboration for pharmaceutical management.
        Hum Resour Health. 2008; 6: 30https://doi.org/10.1186/1478-4491-6-30
        • United Nations
        Every-Woman-Every-Child.
        UN Commission on life-saving commodities for women and children. United Nations, New York2012 (https://www.unfpa.org" [Accessed 3 December 2019])
        • Kelesidis T.
        • Falagas M.E.
        Substandard/counterfeit antimicrobial drugs.
        Clin Microbiol Rev. 2015; 28: 443-464https://doi.org/10.1128/CMR.00072-14
        • Laxminarayan R.
        • Duse A.
        • Wattal C.
        • Zaidi A.K.
        • Wertheim H.F.
        • Sumpradit N.
        • et al.
        Antibiotic resistance-the need for global solutions.
        Lancet Infect Dis. 2013; 13: 1057-1098https://doi.org/10.1016/S1473-3099(13)70318-9
        • World Health Organization
        Antimicrobial resistance: global report on surveillance. World Health Organization, Geneva2014 (https://www.who.int/ [Accessed 10 January 2021])
        • Armon P.J.
        The use of honey in the treatment of infected wounds.
        Trop Dr. 1980; 10: 91https://doi.org/10.1177/004947558001000219
        • Saikaly S.K.
        • Khachemoune A.
        Honey and wound healing: an Update.
        Am J Clin Dermatol. 2017; 18: 237-251https://doi.org/10.1007/s40257-016-0247-8
        • Cooper R.A.
        • Molan P.C.
        • Harding K.G.
        Antibacterial activity of honey against strains of Staphylococcus aureus from infected wounds.
        J R Soc Med. 1999; 92: 283-285https://doi.org/10.1177/014107689909200604
        • Irish J.
        • Blair S.
        • Carter D.A.
        The antibacterial activity of honey derived from Australian flora.
        PLoS ONE. 2011; 6: e18229https://doi.org/10.1371/journal.pone.0018229
        • Johnston M.
        • McBride M.
        • Dahiya D.
        • Owusu-Apenten R.
        • Nigam P.S.
        Antibacterial activity of Manuka honey and its components: an overview.
        AIMS Microbiol. 2018; 4: 655-664https://doi.org/10.3934/microbiol.2018.4.655
        • Jull A.B.
        • Cullum N.
        • Dumville J.C.
        • Westby M.J.
        • Deshpande S.
        • Walker N.
        • et al.
        Honey as a topical treatment for wounds.
        Cochrane Database Syst Rev. 2015; (CD005083)https://doi.org/10.1002/14651858.CD005083.pub4
        • Sullivan T.P.
        • Eaglstein W.H.
        • Davis S.C.
        • Mertz P.
        The pig as a model for human wound healing.
        Wound Repair Regen. 2001; 9: 66-76https://doi.org/10.1046/j.1524-475x.2001.00066.x
        • Junker J.P.E.
        • Lee C.C.Y.
        • Samaan S.
        • Hackl F.
        • Kiwanuka E.
        • Minasian R.A.
        • et al.
        Topical delivery of ultrahigh concentrations of gentamicin is highly effective in reducing bacterial levels in infected porcine full-thickness wounds.
        Plast Reconstr Surg. 2015; 135: 151-159https://doi.org/10.1097/PRS.0000000000000801
        • Daly L.T.
        • Tsai D.M.
        • Singh M.
        • Nuutila K.
        • Minasian R.A.
        • Cameron C.Y.L.
        • et al.
        Topical Minocycline effectively decontaminates and reduces inflammation in infected porcine wounds.
        Plast Reconstr Surg. 2016; 138 (856e-): 68ehttps://doi.org/10.1097/PRS.0000000000002633
        • Eardley W.G.
        • Brown K.V.
        • Bonner T.J.
        • Green A.D.
        • Clasper J.C.
        Infection in conflict wounded.
        Philos Trans R Soc Lond B Biol Sci. 2011; 366: 204-218https://doi.org/10.1098/rstb.2010.0225
        • Petersen K.
        • Riddle M.S.
        • Danko J.R.
        • Blazes D.L.
        • Hayden R.
        • Tasker S.A.
        • et al.
        Trauma-related infections in battlefield casualties from Iraq.
        Ann Surg. 2007; 245: 803-811https://doi.org/10.1097/01.sla.0000251707.32332.c1
        • Lowy F.D.
        Antimicrobial resistance: the example of Staphylococcus aureus.
        J Clin Investig. 2003; 111: 1265-1273https://doi.org/10.1172/JCI18535
        • Nawfal Dagher T.
        • Al-Bayssari C.
        • Diene S.M.
        • Azar E.
        • Rolain J.M
        Bacterial infection during wars, conflicts and post-natural disasters in Asia and the Middle East: a narrative review.
        Expert Rev Anti Infect Ther. 2020; 18: 511-529https://doi.org/10.1080/14787210.2020.1750952
        • Reffuveille F.
        • Josse J.
        • Vallé Q.
        • Mongaret C.
        • Gangloff S.C.
        Staphylococcus aureus biofilms and their impact on the medical field.
        (editor)in: Enany S. The rise of virulence and antibiotic resistance in staphylococcus aureus. IntechOpen Ltd, London2017https://doi.org/10.5772/66380
        • Maddocks S.E.
        • Lopez M.S.
        • Rowlands R.S.
        • Cooper R.A.
        Manuka honey inhibits the development of Streptococcus pyogenes biofilms and causes reduced expression of two fibronectin binding proteins.
        Microbiology. 2012; 158: 781-790
        • Campeau M.E.
        • Patel R.
        Antibiofilm activity of Manuka honey in combination with antibiotics.
        Int J Bacteriol. 2014; 2014795281https://doi.org/10.1155/2014/795281
        • USFaD 510(k) premarket notification
        Food and drug administration.
        Dept of Health and Human Services, Rockville, MD: U.S2007
        • Tam V.H.
        • Kabbara S.
        • Vo G.
        • Schilling A.N.
        • Coyle E.A.
        Comparative pharmacodynamics of gentamicin against Staphylococcus aureus and Pseudomonas aeruginosa.
        Antimicrob Agents Chemother. 2006; 50: 2626-2631https://doi.org/10.1128/AAC.01165-05
        • Vardaxis N.J.
        • Brans T.A.
        • Boon M.E.
        • Kreis R.W.
        • Marres L.M.
        Confocal laser scanning microscopy of porcine skin: implications for human wound healing studies.
        J Anat. 1997; 190 (10.1046%2Fj.1469-7580.1997.19040601.x): 601-611
        • Lavker R.M.
        • Dong G.
        • Zheng P.S.
        • Murphy G.F.
        Hairless micropig skin. A novel model for studies of cutaneous biology.
        Am J Pathol. 1991; 138: 687-697
        • Committee of veterinary medicinal products
        Gentamicin, summery report (EMEA/MRL/803/01-Final). London: The European agency for the evaluation of medicinal products.
        (EMEA). 2001; (https://www.ema.europa.eu. [Accessed 25 September 2019])
        • Zhu X.
        • Liu D.
        • Singh A.K.
        • Drolia R.
        • Bai X.
        • Tenguria S.
        • et al.
        Tunicamycin mediated inhibition of wall teichoic acid affects Staphylococcus aureus and Listeria monocytogenes cell morphology, biofilm formation and virulence.
        Front Microbiol. 2018; 9: 1352https://doi.org/10.3389/fmicb.2018.01352
        • Finkin S.
        • Yuan D.
        • Stein I.
        • Taniguchi K.
        • Weber A.
        • Unger K.
        • et al.
        Ectopic lymphoid structures function as microniches for tumor progenitor cells in hepatocellular carcinoma.
        Nat Immunol. 2015; 16: 1235-1244https://doi.org/10.1038/ni.3290
        • Cray C.
        • Zaias J.
        • Altman N.H.
        Acute phase response in animals: a review.
        Comp Med. 2009; 59: 517-526
        • Faul F.
        • Erdfelder E.
        • Lang A.G.
        • Buchner A.
        G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences.
        Behav Res Methods. 2007; 39: 175-191https://doi.org/10.3758/bf03193146
        • Robson M.C.
        • Lea C.E.
        • Dalton J.B.
        • Heggers J.P.
        Quantitative bacteriology and delayed wound closure.
        Surg Forum. 1968; 19: 501-502
        • Bendy R.H.
        • Nuccio P.A.
        • Wolfe E.
        • Collins B.
        • Tamburro C.
        • Glass W.
        • et al.
        Relationship of quantitative wound bacterial counts to healing of decubiti: effect of topical gentamicin.
        Antimicrob Agents Chemother. 1964; 10: 147-155
        • Scher K.S.
        • Peoples J.B.
        Combined use of topical and systemic antibiotics.
        Am J Surg. 1991; 161: 422-425https://doi.org/10.1016/0002-9610(91)91103-p
        • Erbil B.
        • Ersoy G.
        • Ozkutuk A.
        • Akarca F.K.
        • Korkmaz T.
        • Demir O.F.
        • et al.
        The effects of oral antibiotics on infection prophylaxis in traumatic wounds.
        Ulus Travma Acil Cerrahi Derg. 2014; 20: 231-235https://doi.org/10.5505/tjtes.2014.63993
        • Medeiros Vde F.
        • Azevedo I.M.
        • Rego A.C.
        • Egito E.S.
        • Araujo-Filho I.
        • Medeiros A.C.
        • et al.
        Antibacterial properties and healing effects of Melipona scutellaris honey in MRSA-infected wounds of rats.
        Acta Cir Bras. 2016; 31: 327-332https://doi.org/10.1590/S0102-865020160050000006
        • Robertson C.M.
        • Coopersmith C.M.
        The systemic inflammatory response syndrome.
        Microbes Infect. 2006; 8: 1382-1389https://doi.org/10.1016/j.micinf.2005.12.016
        • Hixon K.R.
        • Klein R.C.
        • Eberlin C.T.
        • Linder H.R.
        • Ona W.J.
        • Gonzalez H.
        • et al.
        A critical review and perspective of honey in tissue engineering and clinical wound healing.
        Adv Wound Care. 2019; 8: 403-415https://doi.org/10.1089/wound.2018.0848
        • Tsai D.M.
        • Tracy L.E.
        • Lee C.C.
        • Hackl F.
        • Kiwanuka E.
        • Minasian R.A.
        • et al.
        Full-thickness porcine burns infected with Staphylococcus aureus or Pseudomonas aeruginosa can be effectively treated with topical antibiotics.
        Wound Repair Regen. 2016; 24: 356-365https://doi.org/10.1111/wrr.12409
        • Al-Waili N.S.
        • Saloom K.Y.
        Effects of topical honey on post-operative wound infections due to gram positive and gram negative bacteria following caesarean sections and hysterectomies.
        Eur J Med Res. 1999; 4: 126-130
        • Leaper D.J.
        • Simpson R.A.
        The effect of antiseptics and topical antimicrobials on wound healing.
        J Antimicrob Chemother. 1986; 17: 135-137https://doi.org/10.1093/jac/17.2.135
        • Atiyeh B.S.
        • Dibo S.A.
        • Hayek S.N.
        Wound cleansing, topical antiseptics and wound healing.
        Int Wound J. 2009; 6: 420-430https://doi.org/10.1111/j.1742-481X.2009.00639.x
        • World Health Organization
        Global action plan on antimicrobial resistance. World Health Organization, Geneva2017 (https://www.who.int. [Accessed 26 February 2019])
        • Mulani M.S.
        • Kamble E.E.
        • Kumkar S.N.
        • Tawre M.S.
        • Pardesi K.R.
        Emerging strategies to combat ESKAPE pathogens in the era of antimicrobial resistance: a review.
        Front Microbiol. 2019; 10: 539https://doi.org/10.3389/fmicb.2019.00539
        • Sahli Z.T.
        • Bizri A.R.
        • Abu-Sittah G.S.
        Microbiology and risk factors associated with war-related wound infections in the Middle East.
        Epidemiol Infect. 2016; 144: 2848-2857https://doi.org/10.1017/S0950268816000431
        • Girma A.
        • Seo W.
        • She R.C.
        Antibacterial activity of varying UMF-graded Manuka honeys.
        PLoS ONE. 2019; 14e0224495https://doi.org/10.1371/journal.pone.0224495
        • Glasser J.S.
        • Guymon C.H.
        • Mende K.
        • Wolf S.E.
        • Hospenthal D.R.
        • Murray C.K.
        Activity of topical antimicrobial agents against multidrug-resistant bacteria recovered from burn patients.
        Burns. 2010; 36: 1172-1184https://doi.org/10.1016/j.burns.2010.05.013
        • Nolan V.C.
        • Harrison J.
        • Wright J.E.E.
        • Cox J.A.G.
        Clinical significance of Manuka and medical-grade honey for antibiotic-resistant infections: a systematic review.
        Antibiotics. 2020; 9https://doi.org/10.3390/antibiotics9110766
        • Foster T.
        • Baron S.
        Staphylococcus.
        Medical microbiology. 4th ed. University of Texas, Galveston1996 (chapter 12)
        • Patton T.
        • Barrett J.
        • Brennan J.
        • Moran N.
        Use of a spectrophotometric bioassay for determination of microbial sensitivity to Manuka honey.
        J Microbiol Methods. 2006; 64: 84-95https://doi.org/10.1016/j.mimet.2005.04.007
        • Carter D.A.
        • Blair S.E.
        • Cokcetin N.N.
        • Bouzo D.
        • Brooks P.
        • Schothauer R.
        • et al.
        Therapeutic Manuka honey: no longer So alternative.
        Front Microbiol. 2016; 7: 569https://doi.org/10.3389/fmicb.2016.00569
        • Allen K.L.
        • Molan P.C.
        • Reid G.M.
        A survey of the antibacterial activity of some New Zealand honeys.
        J Pharm Pharmacol. 1991; 43: 817-822https://doi.org/10.1111/j.2042-7158.1991.tb03186.x
        • Rabie E.
        • Serem J.C.
        • Oberholzer H.M.
        • Gaspar A.R.
        • Bester M.J.
        How methylglyoxal kills bacteria: an ultrastructural study.
        Ultrastruct Pathol. 2016; 40: 107-111https://doi.org/10.3109/01913123.2016.1154914
        • Tan H.T.
        • Rahman R.A.
        • Gan S.H.
        • Halim A.S.
        • Hassan S.A.
        • Sulaiman S.A.
        • et al.
        The antibacterial properties of Malaysian tualang honey against wound and enteric microorganisms in comparison to Manuka honey.
        BMC Complement Altern Med. 2009; 9: 34https://doi.org/10.1186/1472-6882-9-34
        • Sherlock O.
        • Dolan A.
        • Athman R.
        • Power A.
        • Gethin G.
        • Cowman S.
        • et al.
        Comparison of the antimicrobial activity of Ulmo honey from Chile and Manuka honey against methicillin-resistant Staphylococcus aureus, Escherichia coli and Pseudomonas aeruginosa.
        BMC Complement Altern Med. 2010; 10: 47https://doi.org/10.1186/1472-6882-10-47
        • Molan P.
        • Rhodes T.
        Honey: a biologic wound dressing.
        Wounds. 2015; 27: 141-151
        • Roberts A.E.L.
        • Powell L.C.
        • Pritchard M.F.
        • Thomas D.W.
        • Jenkins R.E.
        Anti-pseudomonad activity of Manuka honey and antibiotics in a specialized ex vivo model simulating cystic fibrosis lung infection.
        Front Microbiol. 2019; 10: 869https://doi.org/10.3389/fmicb.2019.00869
        • Jenkins R.
        • Cooper R.
        Improving antibiotic activity against wound pathogens with Manuka honey in vitro.
        PLoS ONE. 2012; 7: e45600https://doi.org/10.1371/journal.pone.0045600
        • Liu M.Y.
        • Cokcetin N.N.
        • Lu J.
        • Turnbull L.
        • Carter D.A.
        • Whitchurch C.B.
        • et al.
        Rifampicin-Manuka honey combinations are superior to other antibiotic-Manuka honey combinations in eradicating Staphylococcus aureus biofilms.
        Front Microbiol. 2017; 8: 2653https://doi.org/10.3389/fmicb.2017.02653
        • Jenkins R.E.
        • Cooper R.
        Synergy between oxacillin and Manuka honey sensitizes methicillin-resistant Staphylococcus aureus to oxacillin.
        J Antimicrob Chemother. 2012; 67: 1405-1407https://doi.org/10.1093/jac/dks071
        • Brudzynski K.
        • Sjaarda C.
        Honey glycoproteins containing antimicrobial peptides, Jelleins of the Major Royal Jelly Protein 1, are responsible for the cell wall lytic and bactericidal activities of honey.
        PLoS ONE. 2015; 10e0120238https://doi.org/10.1371/journal.pone.0120238
      3. (Book)
        • Russell W.M.S
        • Burch R.L
        The principles of humane experimental technique.
        Methuen, London1959