Initially, the nares of 10 pigs were cultured. All were negative for MRSA. At a later stage, the perineum of 30 pigs was cultured; 1 was positive for MRSA. The regional pig farmers were screened (throat and nares) during a monthly professional meeting that happened to be on the farm of the MRSA-positive family, at the time of investigation. With the exception of this meeting, the farmers had no further epidemiologic links, other than being from the southeastern region of the Netherlands. Six (23%) of the 26 farmers were colonized with MRSA.

As mentioned above, all MRSA isolates were resistant to digestion with restriction-endonuclease SmaI, when typing with PFGE was attempted. To ensure that we did not falsely classify a pig-related staphylococcal species as MRSA, the identification of all isolates was confirmed by testing for the presence of a S. aureus–specific DNA element as well as the MecA gene, according to the methods of Reischl et al. (3). To compare the MRSA isolates, we performed random amplified polymorphic DNA analysis with primers Eric II (5´-AAG TAA GTG ACT GGG GTG AGC G-3´), RW3A (5´-TCG CTC AAA ACA ACG ACA CC-3´), D14307 (5´-GGT TGG GTG AGA ATT GCA CG-3´) and spa-typing.

Overall, 3 different MRSA strains were identified. The isolates of the girl (case-patient A), her parents, and the pig from their farm were identical with random amplified polymorphic DNA and belonged to spa-type t108. Furthermore, one of the regional pig farmers screened during the meeting, the pig farmer from a different region (case-patient B), the young boy (case-patient C), as well as his father and the nurse who treated the boy, were colonized with the same strain (Table). Three of the regional pig farmers shared spa-type 567. The isolate from the remaining MRSA-positive regional farmer showed a spa-type not previously described (Table).

Recently, MRSA has been found in horses and in persons who take care of them (4). Human carriage has also been linked to colonized companion cats and dogs (5,6). While Lee et al. (7) reported an MRSA isolation frequency of 0.6% in major food animals, but did not find MRSA in 469 samples from pigs, Armand-Lefevre et al. (8) described S. aureus (methicillin-susceptible and -resistant) carriage among pigs and pig farmers. Although the authors showed that both farmers and pigs carried methicillin-sensitive S. aureus and MRSA and that both groups shared certain multilocus sequence typing, the isolates came from separate, nonrelated collections.

Here we demonstrate transmission of MRSA between an animal and human (pig and pig farmer), between family members (pig farmers and their families), and between a nurse and patient in the hospital. The unexpected high frequency of MRSA among the group of regional pig farmers (>760× higher than in the general Dutch population) indicates that their profession might put them at risk for MRSA colonization. Overall, we found 3 different MRSA strains, including a new spa-type. Therefore, we expect that multiple strains are present in the pig population and the pig farmers. The strain with spa-type t108 appears to be more prevalent and widespread, given that the strain spread from animal to human, between family members, between patient and nurse, and among pig farmers from different regions.

Further research on a larger scale is needed to see if these observations hold true in other regions. If so, pig farming poses a significant risk factor for MRSA carriage in humans that warrants screening wherever pig farmers or their family members are admitted to a hospital.