Plasmid DNA analysis, biotyping, and antimicrobic susceptibility as subtyping tests for Klebsiella pneumoniae and Klebsiella oxytoca.

We compared plasmid DNA analysis, biotyping by Vitek, and disk diffusion antimicrobic susceptibility as subtyping tests of Klebsiella pneumoniae and Klebsiella oxytoca. The 92 tested isolates were from alternate, culture-positive patients over 6 months. No outbreak or cluster of infections was recognized during this interval. Plasmid DNA was detected in 85% of the isolates. Each isolate except one had a reproducible absence of plasmid DNA or a reproducible plasmid DNA profile on repetitive testing. Restriction endonuclease enzyme analysis of plasmid DNA was necessary to distinguish differences among some isolates that had only large plasmids. Isolates with only large plasmids represented 18% of the collection. Of the 78 isolates with plasmid DNA, all but two were considered different from one another by plasmid DNA analysis. Biotyping and antimicrobic susceptibility testing were not highly reproducible. In addition, biotyping did not demonstrate a sufficient variety of patterns among the isolates for subtyping purposes. We conclude that plasmid DNA analysis is very useful as a subtyping test for isolates of K. pneumoniae and K. oxytoca. Neither biotyping nor antimicrobial susceptibility as performed in our laboratory had sufficient discriminatory power and reproducibility for subtyping these organisms.