In the present study, only 1 out of 189 horses from 10 different farms carried a typical representative of the LA-MRSA CC398 clone. The farms were all situated in a region surrounding an equine clinic where shortly before, high carriage rates had been detected in arriving patients . The fact that such a low presence of MRSA on farms was found in this study is rather surprising, and in contrast with the earlier described high carriage rate in horses arriving at the clinic. Possible factors responsible for the higher carriership in equine patients admitted to a clinic may be previous antimicrobial administration, stress due to transportation, transport in contaminated vehicles and direct contact with referring veterinarians who might carry MRSA. Large animal veterinarians are indeed considered to be at high risk of carrying the strains circulating in their main contact species [27–30].
The finding that only one horse was found to be MRSA positive was all the more unexpected since the samples were collected on rather large (n > 20) horse farms, which are more likely to harbor MRSA carriers  and the study was performed in a region where the LA-MRSA CC398 clone is highly prevalent in pigs, with high carriage rates detected at human, individual pig and farm level [31, 32]. Although contact of the horses in the current study with pigs was not assessed as such, it would not have been impossible for pig to horse transmission of CC398 to occur, especially given the long term environmental survival, potential airborne transmission of MRSA and the possibility of large animal veterinarians circulating between pig and horse farms [33–36]. In the current study, the only positive horse was a breeding mare housed on a breeding facility with consequent regular veterinary contact and a high number of horses present in the facility.
The low MRSA presence in the general horse population found here is in accordance with recent findings in the Netherlands where the detection of high CC398 prevalences in livestock (pig, veal calves, poultry), equine clinics and veterinarians [7, 37–40] also stands opposite to a study detecting no MRSA in the general horse population . Both Dutch and Belgian data thus seem to classify equine CC398 carriage as a primarily veterinary-care associated problem with high detection rates only being found in horses arriving and residing at veterinary clinics. This difference in MRSA prevalence found between horse farms and intensive food production animal facilities seems unlikely to be due to differences in host adaptation given the substantial MRSA carriage found in equine clinics [7, 26]. It could, however, be due to differences in husbandry between the examined horse farms and intensive animal husbandry in the same region. For instance, the animal density on horse farms is much lower and group medication (antimicrobials) is virtually non-existing. In fact, the conditions encountered in equine clinics concur much more with those in intensive animal production and may thus, partially, explain the high MRSA prevalence encountered in equine clinics. Common factors in the housing and husbandry of high prevalence sectors in the animal industry should probably be the first to scrutinize when examining potential risk factors for animal LA-MRSA carriage. In addition to the nature of equine husbandry, a second factor may counteract the spread of LA-MRSA in the general horse population. Indeed, horses appear to be mainly transient carriers eliminating MRSA quickly (< 3 weeks) when removed from potential contamination sources . If contaminated at clinic, they may thus quickly eliminate the bacterium when at home.