LETTER

We write to report the detection of a vanB determinant in Enterococcus gallinarum in Australia and the second naturally acquired case (5). High-level acquired VanA-mediated glycopeptide resistance has been described for E. gallinarum, E. casseliflavus, and other enterococcal species (4, 6), and the natural acquisition of the VanB-mediated glycopeptide resistance has been commonly described for E. faecalis and E. faecium. However, acquisition of VanB-mediated resistance in two isolates of E. gallinarum has only recently been described in a Swiss hospital (5).

The current case occurred in a 20-year-old male patient with chronic methicillin-resistant Staphylococcus aureus osteomyelitis which had developed after resection and grafting for a tibial adamantinoma 2 years previously. Prolonged courses of both vancomycin and teicoplanin had been administered sequentially but discontinuously over a period of 18 months. Screening of a rectal swab obtained on readmission of the patient to the infectious diseases unit in February 2000 resulted in the isolation of a nonmotile enterococcus from an Enterococcosel agar (BBL, Cockeysville, Md.) plate containing 6 µg of vancomycin per ml (VRE agar). The isolate was unusual in that it required CO₂ for optimal growth but had typical Gram stain morphology and was pyrrolidonyl peptidase positive. The vancomycin MIC of 8 µg/ml and teicoplanin MIC of 0.5 µg/ml (E-test; AB Biodisk, Solna, Sweden) were consistent with a VanC phenotype.

Because of its unusual phenotype, the isolate was subjected to multiplex PCR as described preivously for the identification of E. faecalis and E. faecium (3) and detection of van gene determinants (1). Based on biochemical identification (positive results for o-nitrophenyl--D-galactopyranoside, arabinose, methyl-2-D-glucopyranoside, raffinose, and xylose and negative results for tellurite, melezitose, and sorbitol), the lack of E. faecalis and E. faecium genetic elements, and the demonstration of a vanC-1 determinant by PCR, the isolate was confirmed as E. gallinarum. However, a 330-bp element indistinguishable from the vanB determinant was also demonstrated.

The detection of motile enterococci which have acquired additional vancomycin resistance genes further supports the argument that identification of enterococci to the species level and determination of glycopeptide MICs do not necessarily predict the genotype (2, 5). Infections with E. gallinarum, E. casseliflavus, and E. flavescens are rare, and the Centers for Disease Control and Prevention does not recommend infection control initiatives for patients infected or colonized with these species (5, 7). However, the ability of these species to acquire glycopeptide resistance determinants warrants further attention, and we agree with the suggestion of Liassine et al. that resistance genotypes in "motile" enterococci should be closely monitored (5).