Multicenter Evaluation of Epidemiological Typing of Methicillin-Resistant <i>Staphylococcus aureus</i> Strains by Repetitive-Element PCR Analysis

Rapid and efficient epidemiologic typing systems would be useful to monitor transmission of methicillin-resistant Staphylococcus aureus (MRSA) at both local and interregional levels. To evaluate the intralaboratory performance and interlaboratory reproducibility of three recently developed repeat-element PCR (rep-PCR) methods for the typing of MRSA, 50 MRSA strains characterized by pulsed-field gel electrophoresis (PFGE) (SmaI) analysis and epidemiological data were blindly typed by inter-IS256, 16S-23S ribosomal DNA (rDNA), and MP3 PCR in 12 laboratories in eight countries using standard reagents and protocols. Performance of typing was defined by reproducibility (R), discriminatory power (D), and agreement with PFGE analysis. Interlaboratory reproducibility of pattern and type classification was assessed visually and using gel analysis software. Each typing method showed a different performance level in each center. In the center performing best with each method, inter-IS256 PCR typing achieved R = 100% and D = 100%; 16S-23S rDNA PCR, R = 100% and D = 82%; and MP3 PCR, R = 80% and D = 83%. Concordance between rep-PCR type and PFGE type ranged by center: 70 to 90% for inter-IS256 PCR, 44 to 57% for 16S-23S rDNA PCR, and 53 to 54% for MP3 PCR analysis. In conclusion, the performance of inter-IS256 PCR typing was similar to that of PFGE analysis in some but not all centers, whereas other rep-PCR protocols showed lower discrimination and intralaboratory reproducibility. None of these assays, however, was sufficiently reproducible for interlaboratory exchange of data.