**3. Materials and methods**

Samples taken from January to August 2015 were collected from 4,341 patient samples admitted to the Laboratory of Institute of Public Health of Sarajevo Canton, Bosnia and Herzegovina (B&H) and they showed a total number of 653 methicillin-resistant *Staphylococcus aureus*, that is, out of 2279 found *Staphylococcus aureus* strains, 653 were methicillin-resistant *Staphylococcus aureus.* Those samples included nose swabs, throat, ear, eye, umbilicus swabs, wound and skin swabs. All MRSA isolates collected were identified by standard microbiological methods based on the demonstration of deoxyribonuclease, bound coagulase (rabbit plasma, bioMerieux, France) and free coagulase (Slidex Staph Plus, bioMerieux, France) [22]. Antibiotic susceptibility determination used the agar disk diffusion method in compliance with the guidelines of the Clinical Laboratory Standards Institute [23, 24]. The following twelve antibiotics were tested to the susceptibility of the *S. aureus* isolates: sulfamethoxazole-trimethoprim, oxacillin, cefoxitin, erythromycin, clindamycin, linezolid, ciprofloxacin, tetracycline, fusidic acid, rifampicin, vancomycin and gentamicin. Multiplex PCR was used for testing those MRSA strains for the presence of the mecA gene. Molecular analysis of the SCCmec cassette was conducted using a method earlier depicted by Oliveira et al. [25], with certain modification indicated by Budimir et al. [26]. Presence of the gene for PVL was found by use of PCR primers previously accounted for by Lina et al. [27]. In addition, all of these isolates passed analysis for epidemiological relatedness by pulsed-field gel electrophoresis (PFGE). Macrorestriction of chromosomal DNA used the restriction enzyme SmaI analysis, complying with previously described procedure. DNA fragments were separated using a CHEF-DR III electrophoresis system (Bio-Rad laboratories, Hercules, California, USA) at 6.0 V/cm for 2O h, with pulse times ramped from 5 s to 40 s. Gels used with ethidium bromide staining and they were photographed under UV illumination. PFGE patterns analysis was done by using GelCompare software (Applied Maths, Ghent, Belgium) according to the Tenover et al. scheme [28]. Isolates with indistinguishable band patterns were assigned to the same PFGE pattern type. Isolates that differed by ≤ three fragments were considered to be subtypes closely related i.e. of a given clonal group. The dendrogram was produced by use of 0, 5% optimization, a 3% band tolerance and the unweighted pair group method with arithmetic averages based on Dice coefficients.
