New vancomycin disk diffusion breakpoints for enterococci

J. M. Swenson, M. J. Ferraro, D. F. Sahm, P. Charache, D. J. Hardy, R. C. Moellering, W. R. Wilson, F. C. Tenover

Research output: Contribution to journalArticlepeer-review

40 Scopus citations


Since 1988, when the first vancomycin-resistant enterococcus was described, several descriptions of failures of disk diffusion breakpoints to detect low-level vancomycin resistance (MICs, 8 to 32 μg/ml) have been published. A four-laboratory collaborative study was undertaken to establish more accurate breakpoints for the disk test. Mueller-Hinton agar was used to perform dilution testing (in three laboratories) and disk diffusion testing (in all laboratories). Results were determined at 18, 24, and 48 h, and zones of inhibition were read using both transmitted and reflected light. One hundred organisms (35 Enterococcus faecalis, 55 E. faecium, and 10 E. gallinarum or E. casseliflavus isolates) were selected to represent vancomycin-susceptible and -resistant phenotypes. Interlaboratory agreement of agar dilution MICs was better at 24 h (91 to 94% within ±1 dilution) than at 18 h (76% within ± 1 dilution). Therefore, 24-h agar dilution MIC results were used as the reference. For disk diffusion, it was critical to note the presence of a haze or colonies inside the zone when interpreting the test, since this correlated better with the results of the agar dilution test. The presence of a haze or inner colonies was best detected by reading the zones with transmitted light and incubating the plates for a full 24 h. When plotted against 24-h agar dilution MICs, breakpoints of ≤14 mm (resistant), 15 to 16 mm (intermediate), and ≥17 mm (susceptible) resulted in 58 minor errors (14.5% of total values) and 5 very major errors (2.2% of resistant values or 1.3% of total values). No major errors were seen. Results of repeat testing using a common lot of Mueller-Hinton agar showed 52 minor errors (13.3%) and 4 major errors (4.2% of susceptible values or 1.0% of total values) but no very major errors. It is recommended that any haze or colonies within the zone be taken into account when determining zones of inhibition and that an MIC test be performed for strains with intermediate zones if vancomycin is being considered for treatment.

Original languageEnglish (US)
Pages (from-to)2525-2528
Number of pages4
JournalJournal of clinical microbiology
Issue number10
StatePublished - 1992
Externally publishedYes

ASJC Scopus subject areas

  • Microbiology (medical)


Dive into the research topics of 'New vancomycin disk diffusion breakpoints for enterococci'. Together they form a unique fingerprint.

Cite this