This was a point prevalence study conducted in a kindergarten in northern Taiwan. The study proposal was reviewed and approved by the National Defense Medical Center Institutional Review Board. Participant or parental informed consent was obtained in all cases. The kindergarten was a for-profit facility that the director had owned and operated for 28 years. Attendees were divided into the following four classes by age: toddler (2-younger than 4 years), older toddler (4-younger than 5 years), preschool child (5-younger than 6 years), and child attending the kindergarten prior to leaving for school (6-younger than 7 years). All classes shared a room for an average of one-half hour in the morning and indoor play areas.
All kindergarten attendees younger than 7 years of age, regardless of their medical history, were considered eligible to participate in this study. Participants and their guardians were approached by the same investigative team throughout the study period.
Cultures and questionnaires
After obtaining written consent, study personnel verbally administered a questionnaire to the guardians to collect demographic data and information on risk factors of all children and their household contacts. Risk factors for MRSA infection analyzed in this study included the following: (1) hospitalization, surgery, endotracheal intubation or antimicrobial therapy in the previous 12 months, (2) underlying chronic disorder (e.g., asthma, chronic lung disease, atopic dermatitis, heart disease or neurological disease) (3) presence of an indwelling venous or urinary catheter, or (4) household contact with an individual with an identified risk factor, (e.g., long-term care facility residence, intravenous drug abuse, recurrent skin infections, history of MRSA infection or colonization) or a worker in a health care environment in the 12 months preceding the culture .
A specimen for culture was obtained from both anterior nares of each enrolled child with a sterile dry cotton swab, pre-moistened with sterile water. The swab was immediately inoculated onto 5% sheep blood agar which was then incubated for 36 to 48 hours at 35°C.
Bacterial strains and antimicrobial susceptibility testing
Staphylococci were identified based on colonial morphology, catalase testing, tube-coagulase testing, DNase reaction, mannitol fermentation, tellurite reduction, and an oxidation-fermentation test. MRSA identification and antimicrobial susceptibility were determined according to the Clinical Laboratory Standards Institute (formerly known as the National Committee for Clinical Laboratory Standards) guidelines [10, 11]. In vitro macrolide-lincosamide-streptogramin-inducible (MLSi) phenotypes were detected by the double-disk diffusion assay . During the study period, all children colonized with CA-MRSA isolates were compared with consecutive control subjects with HA-MRSA infection. Susceptibility to penicillin-G, oxacillin, clindamycin, erythromycin, gentamicin, vancomycin, tetracycline, ciprofloxacin and trimethoprim/sulfamethoxazole was determined using the disc-diffusion method [10, 11].
Multilocus sequence typing (MLST)
MLST was performed by polymerase chain reaction (PCR) amplification and sequencing of seven housekeeping genes using primers designed by Enright et al . Each sequence was submitted to the MLST database website for assignment of the allelic profile and sequence type (ST).
Pulsed-field gel electrophoresis (PFGE)
All S. aureus nasal-colonization isolates were analyzed for epidemiologic relatedness by PFGE of chromosomal DNA performed using the enzyme SmaI (New England Biolabs, Beverly, Mass, USA). DNA was separated in 0.9% agarose gels at 14°C in 0.5× TBE buffer with a CHEF Mapper XA system (Bio-Rad Laboratories, Hercules, CA, USA) for 31.5 h, with initial and final switching times of 2 and 30 s, respectively. Gels were stained with ethidium bromide and photographed under UV illumination. The derived patterns were analyzed using GelCompar software (Applied Maths, Kortrijk, Belgium). Results were analyzed using the unweighted pair group method for arithmetic averages (UPGMA) and the Dice coefficient with 1.2% band tolerance . The strain types were designated in alphabetical order, and a new type, if identified, was designated consecutively. MRSA isolates sharing closely related PFGE profiles from one existing strain type were defined as its subtypes and labeled with suffixes of Arabic numbers.
Staphylococcal cassette chromosome mec (SCCmec) typing
SCCmec typing was performed by means of PCR using sets of region-specific primers as described elsewhere [15, 16].
PCR amplification of mecA, lukS-PV, lukF-PV, ermA, ermB, ermC and msrA
PCR for mecA was performed using relevant published sequences and temperature parameters . The PCR amplification of the lukS-PV, lukF-PV, and genes encoding Panton-Valentine leukocidin (PVL) components was performed as described elsewhere . The presence of MLS resistance genes (ermA, ermB, ermC and msrA) was determined according to previously described methods [19, 20].
Data collection and analyses were performed with the Fisher's exact test using SPSS software, version 10.0 (Statistical Package for Social Sciences; SPSS for Windows, Inc., Chicago, IL, USA). A P value of < 0.05 was considered to represent a statistically significant difference between tested groups.