Detection of Healthcare-Related Extended-Spectrum Beta-Lactamase-Producing Escherichia coli Transmission Events Using Combined Genetic and Phenotypic Epidemiology.

BACKGROUND

Since the year 2000 there has been a sharp increase in the prevalence of healthcare-related infections caused by extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli. However, the high community prevalence of ESBL-producing E. coli isolates means that many E. coli typing techniques may not be suitable for detecting E. coli transmission events. Therefore, we investigated if High-throughput MultiLocus Sequence Typing (HiMLST) and/or Raman spectroscopy were suitable techniques for detecting recent E. coli transmission events.

METHODS

This study was conducted from January until December 2010 at Erasmus University Medical Center, Rotterdam, the Netherlands. Isolates were typed using HiMLST and Raman spectroscopy. A genetic cluster was defined as two or more patients carrying identical isolates. We used predefined definitions for epidemiological relatedness to assess healthcare-related transmission.

RESULTS

We included 194 patients; strains of 112 patients were typed using HiMLST and strains of 194 patients were typed using Raman spectroscopy. Raman spectroscopy identified 16 clusters while HiMLST identified 10 clusters. However, no healthcare-related transmission events were detected. When combining data from both typing techniques, we identified eight clusters (n = 34 patients), as well as 78 patients with a non-cluster isolate. However, we could not detect any healthcare-related transmission in these 8 clusters.

CONCLUSIONS

Although clusters were genetically detected using HiMLST and Raman spectroscopy, no definite epidemiological relationships could be demonstrated which makes the possibility of healthcare-related transmission events highly unlikely. Our results suggest that typing of ESBL-producing E. coli using HiMLST and/or Raman spectroscopy is not helpful in detecting E. coli healthcare-related transmission events.