Diversilab typing for management of Enterobacter cloacae outbreaks at a neonatal ICU

Abstract number: P1682

Visser C.E., de Wever B., Rijnsburger M., Spanjaard L., van Toledo L., Savelkoul P., Duim B.

Objectives: In April 2007 the neonatal ICU observed a possible outbreak of Enterobacter cloacae. Concern was raised when several children presented with clinically significant infections due to E. cloacae.

Methods: To establish the relation between these infections all clinical isolates were collected and all patients were screened for the presence of E. cloacae in throat and rectal swabs. At approximately the same time a new typing system was introduced in our laboratory: Diversilab (bioMerieux). This system enables automated genotyping of bacteria, yeasts and moulds, applying kits for amplification of repetitive extragenic palindromic (REP-PCR) sequences. PCR fragments are analysed on an Agilent® 2100 Bioanalyzer. Web-based software stores sample fingerprints, normalizes the data for comparison and analyses similarity between fingerprints. To validate Diversilab typing, extracted DNA of all E. cloacae strains was also typed by AFLP as gold standard.

Results: Diversilab and AFLP data were concordant and showed identical strains in 5 patients and unrelated strains in 3 patients. Hygienic measures were enhanced in the following weeks and consequently the number of E. cloacae-positive patients decreased. In July a sudden increase in E. cloacae-positive cultures was observed, and we hypothesised that the outbreak strain persisted in a niche on the neonatal ward. Again strains collected from both surveillance and clinical cultures were typed with Diversilab and AFLP. Both typing systems showed that the hypothesis was refuted. An unrelated E. cloacae strain was responsible for the second outbreak. This strain was present in 5 patients. One patient, epidemiologically related to the first outbreak, tested positive for the outbreak strain of the first episode. One patient, who was present on the ward during both episodes, tested positive for an unrelated E. cloacae strain in both episodes. We concluded that both outbreaks on the neonatal ICU were not the result of one single persisting strain, but rather based on the successive introduction of unrelated E. cloacae strains. In October 2007 again an increase of E. cloacae positive cultures was observed and typing showed no new outbreak, as all but 2 strains were unrelated.

Conclusion: The analysis of the presented outbreaks demonstrated that the Diversilab is a very fast, robust and reproducible tool to determine clonal outbreaks of E. cloacae and typing of bacterial strains is crucial for adequate outbreak management.

Session Details

Date: 19/04/2008
Time: 00:00-00:00
Session name: 18th European Congress of Clinical Microbiology and Infectious Diseases
Location: Barcelona, Spain
Presentation type:
Back to top