Department of Infectious Disease/Clinical Bacteriology, University of Gothenburg Department of Paediatrics, University of Gothenburg, Gothenburg, Sweden Department of Paediatric Pneumology and Immunology, Charité Medical University, Berlin, Germany Department of Microbiology Sandro Pertini Hospital Paediatric Allergy Unit, Sandro Pertini Hospital Medical Statistics and Information Technology, Fatebenefratelli Association for Research, Isola Tibernia, Rome, Italy.
Clin Microbiol Infect ABSTRACT: In recent years, Staphylococcus aureus has become a common bowel colonizer in Swedish infants. We aimed to identify host factors that determine such colonization. Stool samples from 100 Italian and 100 Swedish infants were obtained on seven occasions during the first year of life and cultured quantitatively for S. aureus. In a subgroup of infants in each cohort, individual strains were identified by random amplified polymorphic DNA analysis. Colonization at each time-point was related to delivery mode, siblings in family and antibiotic treatment. In total, 66% of the Italian and 78% of the Swedish infants had S. aureus in their stools on at least one time-point (p 0.08) and 4% of Italian and 27% of Swedish infants were positive on at least six of the seven time-points investigated (p 0.0001). Most infants analysed regarding strain carriage harboured a single strain in their microbiota for several months. The S. aureus stool populations in colonized infants decreased from 10(7) to 10(4) colony-forming units/g between 1 week and 1 year of age in both cohorts. In multivariate analysis, the strongest predictor for S. aureus colonization was being born in Sweden (OR 3.4 at 1 week of age, p 0.002). Having (an) elder sibling(s) increased colonization at peak phase (OR 1.8 at 6 months, p 0.047). Antibiotic treatment was more prevalent among Italian infants and correlated negatively with S. aureus colonization at 6 months of age (OR 0.3, p 0.01). To conclude, S. aureus is a more common gut colonizer in Swedish than Italian infants, a fact that could not be attributed to feeding or delivery mode.