Therapeutics

Systematic review and meta-analysis

Impact of honey as a topical

treatment for wounds
remains unclear
10.1136/eb-2013-101331

Rose Cooper
Cardiff School of Health Sciences, Cardiff Metropolitan University,
Cardiff, UK
Correspondence to: Professor Rose Cooper, Cardiff School of Health
Sciences, Cardiff Metropolitan University, Llandaff Campus, Western
Avenue, Cardiff, CF5 2YB, UK; rcooper@cardiffmet.ac.uk

Commentary on: Jull AB, Walker N, Deshpande S. Honey as a topical

treatment for wounds. Cochrane Database Syst Rev 2013;(2):CD005083.

Context
In some countries, honey has been used continuously in the topical treatment of wounds for thousands of years, in others it lost favour and was
reintroduced into modern clinical practice only after licensed wound care
products containing medical-grade honey became available within the
last decade. A systematic review of honey as a topical treatment for
wounds completed in 2008 has recently been updated by Jull and
colleagues.

Methods
Databases were searched from 2008 to June 2012 and two authors independently selected randomised controlled trials (RCTs) and quasi-RCTs
for review. Studies involving patients of any age with wounds treated
topically with honey and any other dressing or component were included.
They included 15 with acute wounds, 7 with chronic wounds and 3 with
mixed acute and chronic wounds. The data were required to include
either time to complete healing or the proportion of patients with completely healed wounds. The data were extracted using a standardised
form, checked for risk of bias using the Cochrane Collaboration tool by
each of the two authors, analysed and summarised.

Findings
Six new studies were found to supplement the 19 included in the last
review; the number of participants increased from 2554 to 2987
(median=87, range=30900). A total of 46 studies were excluded. The
risk of bias was judged mainly to be high or unclear with regard to blinding arrangements and, to a lesser effect, in allocation generation and
concealment. Diversity in wound type and study methodologies precluded meta-analysis. The authors concluded that there was insufcient
evidence to guide clinical practice in some areas, and did not recommend
purchase of honey dressings. They found that honey used in conjunction
with compression did not signicantly improve healing rates of venous
leg ulcers (relative risk 1.15, 95% CI 0.96 to 1.38). They noted prolonged
healing in partial-thickness burns and full-thickness burns treated with
honey in comparison to early excision and grafting (mean difference
13.6 days, 95% CI 10.02 to 17.18 days), and also in leishmaniasis when

used with meglumine antimoniate (relative risk 0.72, 95% CI 0.51 to

1.01). Yet in partial-thickness burns honey reduced healing time compared to conventional dressings (mean difference 4.68 days, 95% CI
4.28 to 5.09 days). Recommendations to improve future trial design
were formulated.

Commentary
This review included many studies completed before widened access to
modern, licensed wound care products containing medical grade honey.
Quality standards for honeys destined for clinical use have been established in order to discriminate them from honeys destined for consumption.1 Also characteristics of a honey depend on bee species, the oral
source, harvesting and storage conditions,2 and different honeys may
have different modes of antibacterial action.3 The honeys included in this
update came from countries where the producing bee was probably Apis
mellifera, but honeys produced in Asian countries may have come form
other species. Floral source was not always specied. Five RCTs utilised
manuka and jarrah honey, so a case could have been made to analyse
those separately.
A valid criticism of the clinical studies included was their small size
and design. The decision to judge risk evaluations as unclear where the
investigator had provided information on allocation concealment but
failed to explain whether envelopes were opaque seemed rather harsh.
Lack of blinding participants and healthcare providers yielded the greatest risk of bias. It is difcult to disguise the smell, consistency and colour
of honey and it may be that this will always be an unachievable goal in
clinical trials with honey.
Most trials utilised conventional comparators (eg, silver sulfadiazine,
povidone iodine, hydrogel or usual care), but sugar, boiled potato skins,
amniotic membrane or honey supplemented with vitamins and polyethylene glycol 4000 were unusual. As new clinical evidence becomes available from better-designed RCTs using modern honey dressings, evidence
from former trials may become less inuential.
Cochrane reviews provide a valuable means to evaluate clinical evidence, but including only studies that report complete wound healing as
a primary outcome is restrictive. The exclusion of one study of malignant
wounds4 because wound healing was not achieved suggests that outcomes, such as decreased wound size and increased survival time, were
not important. Future reviewers might consider using criteria that allow
more diverse outcomes to be evaluated.
Competing interests RC has consulted for Arnold & Palmer, Flen
Pharma, BSN Medical, Crawford Healthcare, Molylnke, Bayer Consumer
Healthcare. She has lectured for KCI, Advancis Medical, SAWC and the
College of Practitioners of Phytotherapy. Additionally, she has provided
educational presentations for North American Center for Continuing
Medical Education (NACCME). Lastly, her institution received grants from
the Waterloo Foundation and the Sir Halley Stewart Trust.
References
1. Molan P, Hill C. Chapter 5: quality standards for medical grade honey. In: Cooper R,
Molan P, White R, eds. Honey in modern wound management. Aberdeen, UK:
Wounds UK, 2008:6379.
2. Molan P. The antibacterial nature of honey: I. The nature of the antibacterial
activity. Bee World 1992;73:528.
3. Kwakman PHS, te Valde AA, de Boer L, et al. Two major medicinal honeys have
different mechanisms of bactericidal activity. PLoS ONE 2011;6:e17709.
4. Lund-Nielsen B, Adamsen L, Kolmos HR, et al. The effect of honey-coated
bandages compared with silver-coated bandages on treatment of malignant wounds
a randomised study. Wound Repair Regen 2011;19:66470.

Evid Based Med February 2014 | volume 19 | number 1 |

11

Copyright of Evidence Based Medicine is the property of BMJ Publishing Group and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.