Comparative effectiveness

In three prospective, randomised, controlled clinical trials, honey was found to help heal superficial burns quicker than polyurethane film (OpSite™, Smith and Nephew), a dressing commonly used for providing a moist healing environment; and quicker than silver sulfadiazine (SSD) 1% ointment, the current ‘gold standard’ dressing for preventing infection in burn wounds (Pruitt, 1987). In a study comparing honeyimpregnated gauze with the polyurethane film, the mean times to healing in each group (n = 46) were 10.8 days and 15.3 days respectively (p<0.001). In addition, significantly fewer honey-dressed wounds became infected (p<0.001; Subrahmanyam, 1993). In the first of the two studies that compared honey-impregnated gauze with silver sulfadiazineimpregnated gauze (n = 52 patients in each group), 87% of the wounds treated with honey healed within fifteen days, compared with 10% of those treated with SSD (p<0.001; Subrahmanyam 1991). In this study, a statistically significant difference (p<0.001) was found in the clearance of bacteria from the burns. In the forty-three out of fifty-two cases that presented positive swab cultures on admission in the group treated with honey, thirty-nine (91%) became sterile within seven days. In the comparison (SSD) group, only three (7%) of forty-one wounds with positive swab cultures became sterile: evidence of the antibacterial effect of honey in vivo.

In the second burns trial (twenty-five patients in each group), 100% of the wounds treated with honey healed within twenty-one days, compared to twenty-one (84%) of those treated with SSD (p<0.001; Subrahmanyam, 1998). In addition to the significant difference found in burn wound healing, biopsies of the treated areas showed greater histopathological evidence of reparative activity. This was seen in 80% of wounds treated with honey dressing compared to 52% of the wounds treated with SSD (p<0.005; noted in biopsy samples from the wound margins after seven days of treatment). Regarding the clearance of bacteria from burns, in twenty-three of the twenty-five cases treated with honey that had positive swab cultures on admission in the group, fifteen (65%) of the wounds became sterile in seven days and twenty-two (96%) in twenty-one days. By comparison, of the twenty-two wounds with positive cultures treated with SSD, sixteen (73%) became sterile in seven days, and nineteen (86%) in twenty-one days (p<0.001). This is further evidence of the antibacterial effect of honey in vivo.

Although these trials showed that honey offered better control of infection than standard treatment, a trial on moderate burns where half of the total burn area was full-thickness showed that control of infection was better with early tangential excision followed with autologous skin grafting than with honey treatment (Subrahmanyam, 1999). In two groups (n = 25) of young adults, 34% of swab cultures were positive for the group treated with honey, compared with 10% of the group treated with early tangential excision (p<0.05). Antibiotics were needed for 32±18 days in the honey-treated group compared with 16±3 days in the excision group (p<0.001). These findings relate to the need to debride eschar as it serves as a ‘reservoir’ of potential pathogens in the burn. The mean blood volume replaced was less with the honey treatment (21% ± 15%, compared with 35% ± 12%, p<0.01) and skin grafting was required on only eleven patients of the group treated with honey.

In recent reports where selected honey was used on an infected wound following surgical treatment of hidradenitis suppurativa (Cooper et al, 2001) and infected skin lesions from meningococcal septicaemia (Dunford, 2000a), the antibacterial activity gave rise to rapid clearance of infection and healing of the wounds. In both of these studies, it had not been possible to achieve healing with the many systemic antibiotics and modern dressing materials previously tried over a long period of time. Good infection control was reported in a crossover study of nine infants with large infected surgical wounds (Vardi, 1998). Honey was used on the wounds after they failed to heal following at least fourteen days of treatments with intravenous antibiotics (a combination of vancomycin and cefotaxime, subsequently changed according to bacterial sensitivity), fusidic acid ointment, and wound cleaning with aqueous 0.05% chlorhexidine solution. Marked clinical improvement was seen in all cases after five days of treatment; all wounds were closed, clean, and sterile after twenty-one days of honey application. A prospective, randomised controlled trial on severe post-operative wound infections following caesarean section or abdominal hysterectomy was conducted to compare dressing with honey (n = 26) to washing wounds with 70% ethanol and applying povidone-iodine (n = 24). Both groups received systemic antibiotics according to culture and sensitivity. In the group treated with honey, infection was rapidly eradicated (6±1.9 days vs 14.8±4.2 days), wounds healed faster (10.7±2.5 days vs 22±7.3 days), post-operative scars were less than half the size, and the period of hospitalisation was less than half of that for the patients in the control group (9.4±1.8 days vs 19.9±7.4 days: p<0.05 for each parameter: Al-Waili and Saloom, 1999). This study was of particular interest as all patients were treated with appropriate antibiotics, yet the topical application of honey still proved to be effective in reducing bioburden. This might be due to low local tissue levels of antibiotic from poor perfusion of the wound. It is of interest that in vitro studies have shown a synergy between honey and common antibiotics in multidrug resistant Pseudomonas spp (Karayil et al, 1998). It would, therefore, appear to justify the combination of systemic antibiotics with use of topical antibacterials (such as honey) in wounds where poor perfusion and drug resistance might compromise healing.

A trial on patients with dehisced abdominal wounds following caesarean section, showed healing in less than half the time (mean length of stay in hospital 4.5 days, range two to seven days) when the wound was dressed with honey, compared retrospectively with the usual treatment of wound care (cleansing with hydrogen peroxide solution, Dakin’s solution, and packing with saline-soaked gauze) and subsequent re-suturing (mean length of stay in hospital 11.5 days, range nine to eighteen days: Phuapradit and Saropala, 1992).

Back to top