Development of a simple, low-density array to detect methicillin-resistant Staphylococcus aureus and mecA dropouts in nasal swabs

Abstract

Detection of methicillin-resistant Staphylococcus aureus (MRSA) is important for prevention and control of MRSA infections, but the discovery of mecA dropouts and SCCmec junction sequences with homology to coagulase-negative staphylococci (CoNS) has challenged several real-time PCR tests. The objective of this study was to develop a user-friendly, gel element microarray test for MRSA detection, to estimate the analytical performance characteristics of the test on bacterial isolates, and to perform an initial evaluation of the test on nasopharyngeal swabs from patients known to have a high prevalence of S. aureus containing mecA dropouts. The assay limit of detection for the test was 250 fg (or less) of genomic DNA per amplification reaction (approximately 80 cell equivalents) and MRSA was consistently detected at a ratio of 1:12,000 in a non-target background. Of 87 bacterial isolates, the test accurately classified 86 (98.8%) overall, and correctly identified 14 mecA dropout specimens that were falsely positive in the BD GeneOhm MRSA test or BD GeneOhm StaphSR test. A retrospective analysis of 246 nasal swab samples acquired from a high-risk patient population (overall prevalence=10.8% by culture) resulted in 80.5% sensitivity (95% CI=68.4%, 92.6%) and 96.6% specificity. Of these 246 samples, 174 (71%) were positive for mecA, 86 (35%) were positive for S. aureus tufA and 46 (19%) were positive for a SCCmec junction sequence. To estimate method repeatability, 48 samples representing the full range of phenotypes, genotypes and microarray probe SNR values were tested in triplicate, with three discordant results for a concordance rate of 97.9% (141/144 tests). These data demonstrate that a very simple microarray test can identify mecA dropouts with high specificity in either cultured isolates or nasal swabs from a high-prevalence, high-risk patient population. However, the clinical sensitivity of the test will likely depend on local microbial ecology and the prevalence of mecA positive CoNS in any given patient population.