Evaluation of Molecular Typing Methods for Methicillin Resistant Staphylococcus Aureus

Abstract

Effective control of methicillin resistant Staphylococcus aureus (MRSA) depends upon a thorough knowledge of its epidemiology. Typing of isolates can show the Infection Control Team whether there is an outbreak caused by spread of a single organism or, alternatively, a series of sporadic unrelated MRSA. The optimum typing method has not yet been defined. Bacteriophage typing, the mainstay for many years, is unreliable. Pulsed-field gel electrophoresis (PFGE) is highly discriminatory and is regarded as the “gold standard” but it is expensive, time consuming and requires expertise. Methods based on the polymerase chain reaction (PCR) may offer a compromise, as results are rapid, reasonably discriminatory and reproducible. The aim of this study was to systematically evaluate a selection of methods for molecular typing of MRSA, for use in the routine microbiology laboratory. This was carried by evaluating each of the molecular typing systems for typability, reproducibility, discriminatory ability, stability, cost effectiveness and turn around time. A total of 152 MRSA isolates were tested, comprising of two sets; outbreak/connected isolates and a diverse, non-epidemiologically connected set. A group of 36 isolates were from outbreaks from hospitals in Oxford, Aberdeen and Birmingham. The remaining 116 isolates were sporadic isolates from hospitals in the UK and epidemic MRSA stains (EMRSA 1-16). The methods evaluated were PFGE; repetitive element sequence based PCR methods involving amplification using repMP3 primer derived from Mycoplasama pneumoniae, Shine-Dalgarno-transposon 916 spacer amplification, inter-IS256 fragment amplification, and amplification of the 16S-23S rRNA intergenic spacer region; ribotyping and binary typing. Results were analysed using the program Gel Compar II (Applied Maths, Belgium). The degree of homology was determined by DICE coefficient and clustering correlation by UPGMA, with a 1.2% position tolerance. There are no standard guidelines for the interpretation of molecular typing generated profiles except those proposed by Tenover et al. Isolates were assigned groups according to two sets of criteria; a one band difference or a difference of 7 or more bands according to Tenover’s’ criteria. The resulting groups were used to determine reproducibility, epidemiological concordance and Simpson’s Index of diversity. PFGE was found to be the best method in terms of exhibiting excellent typability, reproducibility and epidemiological concordance, and good stability and discriminatory ability when Tenover’s criteria are applied. Of the Rep-PCR methods, RS-PCR was found to show excellent typability and epidemiological concordance and good reproducibility and stability. PFGE and RSPCR are the most suitable methods for use in the routine microbiology laboratory.