Extremely low prevalence of UK Staphylococcus aureus isolates with reduced susceptibility to vancomycin

Sir, We note with interest that Livermore and colleagues have not detected any Staphylococcus aureus isolates with reduced susceptibility to vancomycin despite screening approximately 26 000 strains. Since April 2000, in our department, all isolates of S. aureus have been screened for reduced susceptibility to vancomycin. As of February 2001 3713 isolates, 39% of which were methicillin resistant, had been examined. Sixty-six per cent of the methicillinresistant S. aureus (MRSA) isolates were EMRSA groups 15 or 16. An agar incorporation method with brain–heart infusion agar (Oxoid, Basingstoke, UK) and multipoint inoculation of 10 cfu/spot was used, as recommended. Enterococcus faecalis NCTC 12697 (vancomycin MIC 4 mg/L) was included as a control. Plates were incubated at 37°C and examined at 24 h. A series of plates (0, 2, 4, 6 and 8 mg/L vancomycin) was chosen in preference to a single breakpoint plate method for several reasons: vancomycin MICs for strains displaying reduced susceptibility to glycopeptides are only twoto eight-fold above the values typical for susceptible S. aureus; using more than one breakpoint plate reduced the likelihood of recording inaccurate results arising from variations in media prepared in-house; uncertainty of the distribution of vancomycin MICs for our S. aureus clinical isolates; and current breakpoint testing recommendations are based on isolates obtained from non-UK populations. Only six of 3713 isolates (0.16%) yielded three or more colonies on plates containing 4 mg/L vancomycin. All of these were confirmed as fully susceptible to vancomycin by Etest on Iso-Sensitest agar; four and two strains had vancomycin MICs of 2 and 1.5 mg/L, respectively. Isolates of S. aureus with vancomycin MICs of 4 mg/L were not uncommon (15.9%), and we agree with Livermore and colleagues that there is little value in considering such isolates of major significance, as American workers have suggested. Our findings confirm that the current prevalence of S. aureus strains with reduced susceptibility to vancomycin, in the setting of a tertiary referral hospital, is extremely low. This is not an excuse for complacency, but rather a definitive baseline against which future surveillance results can be compared.