First, they took 20 different types of honey bought from supermarkets, small shops and beekeepers, then tested them to see if they were sterile. To their surprise, 18 contained viable pathogens — some of which were potentially nasty.
The gut of a healthy person could deal with these pathogenic bugs, but the vets concluded that putting these honeys on to a wound would be a bad idea, since they would introduce infection.
Next, the vets took swabs from infected equine wounds and grew cultures of 10 bacteria in the laboratory. They then selected 11 sterile honeys that contained absolutely no contamination and tested their effectiveness against the growth of these bacteria.
8 out of the 11 sterile honeys were effective in inhibiting the growth of all 10 bacteria in the lab — the best were heather honey and Manuka 20+ honey.
In conclusion, sterile honey does appear to have powerful antibacterial properties. The majority of honey available in the UK, however, is not sterile and should not be applied to a wound.