A rapid and sensitive method for excluding the presence of methicillin-resistant Staphylococcus aureus (MRSA) in clinical samples was developed and evaluated. The method utilised an MRSA-selective enrichment broth for 16 h, followed by PCR quantification of the nuc gene. Samples below a quantitative PCR threshold were reported as MRSA-negative. Broths from PCR-positive samples were subcultured for MRSA isolation. Clinical samples (n = 334) in a constructed high prevalence population were analysed in parallel with a selective plating method. The new broth-PCR assay increased the number of positive samples by 35% (49 vs. 66), and 94% of negative samples were reported within 24 h. To reduce costs and workload, 665 clinical samples were grown separately in enrichment broth and then pooled in the PCR step. The broth-PCR assay increased the number of MRSA positive samples from 11 to 15 compared with selective plating. Most (89%) of the culture-negative samples were also PCR-negative and could be reported within 24 h. The growth of 25 European EMRSA strains was tested in the selective enrichment broth. On average, the MRSA strains showed a 300 000-fold increase in CFU, compared with 30-fold for the eight methicillin-sensitive Staphylococcus aureus strains tested.