Close to half of the 878 methicillin-resistant Staphylococcus aureus (MRSA) strains recovered between 1992 and 1997 from the pediatric hospital in Lisbon were bacteria in which antibiotic resistance was limited to β-lactam antibiotics. The other half were multidrug resistant. The coexistence of MRSA with such unequal antibiotic resistance profiles prompted us to use molecular typing techniques for the characterization of the MRSA strains. Fifty-three strains chosen randomly were typed by a combination of genotypic methods. Over 90% of the MRSA strains belonged to two clones: the most frequent one, designated the “pediatric clone,” was reminiscent of historically “early” MRSA: most isolates of this clone were only resistant to β-lactam antimicrobials and remained susceptible to macrolides, quinolones, clindamycin, spectinomycin, and tetracycline. They showed heterogeneous and low-level resistance to methicillin (MIC, 1.5 to 6 μg/ml), carried the ClaI-mecApolymorph II, were free of the transposon Tn554, and showed macrorestriction pattern D (clonal type II::NH::D). The second major clone was the internationally spread and multiresistant “Iberian” MRSA with homogeneous and high-level resistance to methicillin (MIC, >200 μg/ml) and clonal type I::E::A. Surprisingly, the multidrug-resistant and highly epidemic Iberian MRSA did not replace the much less resistant pediatric clone during the 6 years of surveillance. The pediatric clone was also identified among contemporary MRSA isolates from Poland, Argentina, The United States, and Colombia, and the overwhelming majority of these were also associated with pediatric settings. We propose that the pediatric MRSA strain represents a formerly widely spread archaic clone which survived in some epidemiological settings with relatively limited antimicrobial pressure.

The epidemic Iberian clone was among the index cases involved with the MRSA outbreak in 1993, and this was found to be endemic in a follow-up survey conducted in 1995, colonizing health care personnel and spreading to most hospital wards. A few isolates of another epidemic clone, the Brazilian MRSA, also were detected among 1995 isolates. A better understanding of the mechanism(s) of epidemicity of these rapidly spreading clones is urgently needed.