The Evidence Supporting the Use of Honey as a Wound Dressing

Some clinicians are under the impression that there is little or no evidence to support the use of honey as a wound dressing. To allow sound decisions to bemade, this seminar article has covered the various reports that have been published on the clinical usage of honey. Positive findings on honey in wound care have been reported from 17 randomized controlled trials involving a total of 1965 participants, and 5 clinical trials of other forms involving 97 participants treated with honey. The effectiveness of honey in assisting wound healing has also been demonstrated in 16 trials on a total of 533 wounds on experimental animals. There is also a large amount of evidence in the form of case studies that have been reported. It has been shown to give good results on a very wide range of types ofwound. It is thereforemystifying that there appears to be a lack of universal acceptance of honey as a wound dressing. It is recommended that clinicians should look for the clinical evidence that exists to support the use of other wound care products to compare with the evidence that exists for honey.

There is a rapidly increasing interest in the use of honey as a wound dressing, although common clinical opinion would appear to be that there is no evidence to support its use as a wound dressing. Even where reviews of clinical evidence for the use of honey have been published, a negative impression is often obtained from consulting these, as the conclusions stated are that the evidence is of low quality and/or that there is a need for more evidence. But the myriad of advertisements for modern wound dressings possibly blinds people to the fact that only small, poor-quality trials exist to support the use of these products. For example, if the PubMed database is searched for evidence to support the use of nanocrystalline silver dressings, it can be seen that there is in fact very little published evidence. A recent systematic review of publications on the use of advanced dressings in the treatment of pressure ulcers has found that their generalized use in the treatment of pressure ulcers is not supported by high quality evidence. In evidence-based medicine, decisions should be made on the basis of the available evidence: where randomized controlled trials of the highest quality have not been conducted, it is necessary to consider evidence of a lower quality. It is for these reasons that this perspective article has been written, to allow clinicians to see the large amount of evidence that exists for the effectiveness of honey as a wound dressing. By comparing this with the evidence for other wound-care products, clinicians can then judge for themselves the relative merits of honey as a treatment option for wounds.

The literature cited was found by searching the PubMed, BIOSIS, and ISIWeb of Science databases for the term honey. Literature not included in the databases but found from citations in papers were included in this search. Excluded were papers where honey was used in a mixture with other therapeutic substances, the 31 case studies found for single cases of wound treatment with honey, the 21 papers found giving brief reports on the use of honey on wounds, and papers that were expressions of opinion rather than reports of treatment of wounds with honey. Conference presentations were also excluded from this consideration.

CLINICAL EVIDENCE

Many randomized controlled trials have been carried out comparing honey with various other wound treatments. Other clinical trials have been conducted where the form of the trial has been other than a randomized controlled trial. In some of these, the results for the group of patients treated with honey were compared retrospectively with those from the control treatment. In others, the patients were crossed over to treatment with honey after a period of the treatment normally used for that type of wound. Some of the case studies reported for single cases have also involved a comparative study. In these, the patient has had multiple wounds, so honey could be used to treat wounds on one side of the body and normal wound treatment used on those on the other side.

There have also been many non-comparative studies reported on the use of honey as a wound dressing. Because many of these cases were not responding to standard treatment for quite some time before dressing with honey was commenced, these provide evidence that is somewhat like that from a crossover trial, although these studies involved no reverse change in treatment such as would be done in a crossover trial. Some of these studies have been with multiple cases.

EVIDENCE FROM ANIMAL EXPERIMENTS

Many studies have been performed on the effectiveness of honey in promoting the healing of standardized wounds created on experimental animals. These experiments not only have allowed there to be much more closely comparable controls in trials but also have allowed histological examination of the healing wounds to provide additional data besides the usual measurements of decrease in wound size and time to heal.

DISCUSSION

The evidence presented in this article amply demonstrates that honey, the oldest wound dressing material known to medicine, can give positive results where the most modern products are failing. Because people generally are unaware of the historical usage of honey as a wound dressing, or know only of its ancient usage, its clinical usage is presumed to be a new development or something that has been rediscovered. However, a look at the reference list at the end of this article will reveal reports of clinical usage published in the 1950's,10,11 1960's,12 1970's,13-16 and 1980's 17-23 as well as the rapidly increasing number since its apparent rediscovery.

Limitations of Evidence Presented

The evidence presented here that supports the use of honey in wound care includes evidence from many clinical trials. However, none of the findings from these trials would be considered to be evidence of the very highest level, because even though they may have been randomized controlled trials, they have not been double blind. It is near impossible to conduct a double blind trial of honey as a wound dressing, because of the difficulty of keeping obscured from the patients that a material as recognizable as honey is being used. Even if honey is applied in the form of a manufactured dressing, its aroma is immediately recognized, even a single blind randomized trial may be difficult to conduct.

However, there are trials and case studies in which the honey and the comparative treatment were used simultaneously on the same patient thereby offering a degree of control. These demonstrate that positive results achieved with honey are not merely a placebo effect. One of these was a prospective randomized controlled trial of honey on split-thickness skin graft donor sites. On patients in this trial who had single donor sites (3 groups of 14 patients), half of the donor site was treated with honey and half with the comparative treatment.On patients with 2 donor sites (3 groups of 15 patients), 1 of the donor sites was treated with honey and 1 with the comparative treatment. (Honey was compared with 3 controls, saline-soaked gauze, paraffin gauze, and a hydrocolloid.) In that trial, the significantly faster healing rates and lower pain scores achieved with honey compared with saline-soaked gauze and paraffin gauze clearly would have been due to physical effects of the honey and not to psychosomatic effects. Unlike with the trial with the skin graft donor sites where the wounds being compared were of a standard nature, there is a possibility the wounds given different treatment for comparison may not have been identical when treatment was started.

The most convincing evidence for the results with honey not being due to a placebo effect comes from the many studies that demonstrated the effectiveness of honey on standard wounds inflicted on experimental animals. Although the participants in these trials may well have been able to detect by smell that honey was being used, they would not have had any psychosomatic effects on healing resulting from beliefs that natural products would be more effective, or from hearing via the news media of the effectiveness of honey in wound treatment.

Does honey give good results in individual cases studied because those wounds received more attention, or the prior treatment was less than ideal? There are cases where honey has worked even on wounds that had received prior specialist attention. They changed to healing from non-healing only when treatment with honey was commenced. In many of these cases, the wounds were not responding to best practice with modern dressings, although a recent systematic review of the evidence for the efficacy of modern wound dressings in the treatment of pressure ulcers has concluded that there is no evidence that these are any better than saline-soaked gauze.

Supporting Evidence From In Vitro Studies

Further evidence to support the use of honey as a wound dressing comes from laboratory studies that have clearly demonstrated that honey has bioactivities that would be beneficial in wound care. In work with cultures of leukocytes, honey has been shown to stimulate cytokine production bymonocytes.25,26 The release of cytokines is what initiates the tissue repair process as well as the immune response to infection. Also, stimulation by honey of other aspects of the immune response, the proliferation of B- and T-lymphocytes and the activity of phagocytes, has been shown. Additional to this work with cells in culture, it has been demonstrated that honey stimulates the production of antibodies in mice in response to antigens from Escherichia coli. These findings suggest that part of the effectiveness of honey in clearing and preventing infection in wounds that is so widely seen in the clinical evidence may be due to enhancement of the body’s own immunity as well as being due to the antibacterial activity of honey.

The number of publications on laboratory studies showing that honey has antibacterial activity with a very broad spectrum is very large. But what is often not taken into account is that honeys can vary as much as 100-fold in the potency of their antibacterial activity. More recent publications have reported on the sensitivity of various species of bacteria to honey, with antibacterial potency near the median level found in surveys of large numbers of samples. This level is a little below that of the various honey wound-care products now on sale manufactured from Leptospermum Manuka honey, but there are other wound-care products manufactured from honeys not selected to have high levels of antibacterial activity. Laboratory studies with Leptospermum honey with antibacterial potency near the median level have shown the MIC (minimum inhibitory concentration, ie, the concentration down to which honey could be diluted by wound exudate and still prevent bacterial growth) to be 2% to 3% for Staphylococcus aureus,32 3.3% to 4% for coagulase-negative staphylococci,33 5.5% to 9% for pseudo-monads,34,35 2.7% to 3% for MRSA,36 and 3.8% to 5% for VRE.36 The effectiveness of honey in clinical usage in clearing infection with MRSA and VRE has been reported. The slow clearance of infection, or failure to clear infection, in some of the cases reported may well reflect the use of honey with a low antibacterial potency. For example, this may have been the case in the randomized controlled trial where honey was found to be less effective than early tangential excision followed by autologous skin grafting in controlling infection in the treatment of burns. The same author, publishing results comparing the MIC values for various types of honey available locally, reported that the MIC for the most potent honey against S aureus was 20% to 25%,43 which means that the honey had only about a tenth of the antibacterial potency of the Leptospermum honey used in wound care products now on sale.

Dressing Techniques

Another reason for variability in results may have been that the honey dressings were not being kept in place on the wound in some cases. The difficulty of achieving this was commented on. If the honey is flushed out of the dressing by wound exudate, then its various bioactivities cannot be having any effect on the wound. A case that may be an example of this is where infection in a leg ulcer was reported to recur when compression was commenced. Here it was noted that there was a problem with dressings adhering, which is a clear indication that honey has been flushed out of the dressing by wound exudate. A similar occurrence was reported where honey-impregnated tulle dressings were being used. These have very little absorbency, so honey is easily flushed from them. It was noted in this case that the dressings became saturated with exudate within 1 hour. In another case where poor progress was occurring with honey, it was found that much better progress with healing occurred when more frequent changes of the dressings were made.

Anti-Inflammatory Activity of Honey

It has been noted that if sufficient honey is kept in place, by applying it by way of impregnated dressings and changing these frequently enough, then its anti-inflammatory activity will reduce the amount of exudate and thus remove the need for frequent dressing changes. There is a very large amount of evidence for honey having significant anti-inflammatory activity. As well as the evidence that has come from the many clinical observations summarized in this review, there is evidence from histological observation of biopsy samples taken in a clinical trial of honey on burns and from biochemical assays of indicators of inflammation in other clinical trials on burns. One of these biochemical studies was in the form of a randomized controlled trial with 60 patients, comparing honey with silver sulfadiazine, and it was demonstrated that honey decreased oxidative stress by mopping up the free radicals arising from burns. There is also histological evidence for the anti-inflammatory activity of honey from some of the studies on experimental animals. In some of the experimentally induced burns, there was no infection evident, yet honey still brought about a decrease in inflammation. This indicates that the anti-inflammatory activity of honey is a direct action and not a secondary consequence of removal of infection through its antibacterial activity. This is confirmed also by honey giving a positive result in the standard guinea pig wrist stiffness test for anti-inflammatory activity. That honey has a direct anti-inflammatory activity is also indicated by the finding that honey was as effective as prednisolone in a trial on induced colitis in rats,54 and by its being found to give a highly significant (P < .001) reduction in peritoneal adhesions following surgery on the cecum and ileum in another trial on rats. A laboratory study also demonstrated a direct anti-inflammatory activity in honey, as honey was shown to significantly (P < .001) decrease the amount of reactive oxygen intermediates released from monocytes in culture that had been stimulated with Escherichia coli lipopolysaccharide.

CONCLUSIONS

There is a large body of evidence to support the use of honey as a wound dressing for a wide range of types of wounds. Its antibacterial activity rapidly clears infection and protects wounds from becoming infected, and thus it provides a moist healing environment without the risk of bacterial growth occurring. It also rapidly debrides wounds and removes malodor. Its anti-inflammatory activity reduces edema and exudate and prevents or minimizes hypertrophic scarring. It also stimulates the growth of granulation tissue and epithelial tissue so that healing is hastened. Furthermore, it creates a non-adherent interface between the wound and the dressing so that dressings may be easily removed without pain or damage to newly regrown tissue.

The barrier to using honey that has existed for many clinicians who have been constrained to using only licensed products has been removed now that honey is available in the form of various sterile products licensed for use in wound care. To practice evidence based medicine, clinicians involved in wound care thus should check what evidence exists for other wound dressing products they may be considering using and weigh this up against the evidence that exists to support the use of honey.