Relevance of vancomycin-intermediate susceptibility and heteroresistance in methicillin-resistant Staphylococcus aureus bacteraemia.

OBJECTIVES To assess the relevance of vancomycin-intermediate susceptibility (VISA) and heteroresistance (hVISA) in methicillin-resistant Staphylococcus aureus (MRSA) bacteraemia. METHODS We determined vancomycin MICs for 371 saved MRSA blood isolates (2002-03; 2005-06) by Etest and broth microdilution (BMD), screened for hVISA (Etest methods), determined the population analysis profile (PAP)/AUC for isolates with suspected reduced susceptibility (MICs >2 mg/L and/or hVISA-screen-positive versus Mu3 (hVISA control), and stratified patient characteristics and outcome according to susceptibility phenotype: VISA (PAP/AUC >1.3), hVISA (PAP/AUC 0.9-1.3), and susceptible (S-MRSA; PAP/AUC <0.9). RESULTS PAP/AUC revealed 6 (1.6%) VISA and 30 (8.1%) hVISA phenotypes. The Etest MIC was above the susceptibility cut-off (2 mg/L) for all VISA isolates, whereas the BMD MIC was within the susceptibility range in two (33.3%) instances. Eight hVISA isolates (26.7%) with MICs of 2 mg/L were hVISA-screen negative. SCCmec typing revealed SCCmec II in 100% of VISA, 86.7% of hVISA and 75.5% of S-MRSA isolates (P = 0.04). Prior vancomycin use was documented in 100% of VISA, 73.3% of hVISA and 52.2% of S-MRSA cases (P = 0.002). Outcome (compared in 243 vancomycin-treated patients with MICs of 2 mg/L) revealed longer time to clearance in VISA cases [12.1 ± 13.1 days versus 3.3 ± 3.9 (hVISA) and 3.7 ± 5.1 (S-MRSA); P = 0.001], more frequent endocarditis [33.3% versus 9.1% (hVISA; P = 0.1) and 4.2% (S-MRSA; P = 0.001)] and attributable mortality [33.3% versus 9.1% (hVISA; P = 0.1) and 8.4% (S-MRSA); P = 0.08]. CONCLUSIONS No adverse outcome was documented with hVISA phenotype, whereas VISA contributed to vancomycin treatment failure. VISA and hVISA appear to emerge in SCCmec II isolates among vancomycin-exposed patients and are better detected by Etest.