PLoS ONE (Jan 2019)

Antibiotic resistance of blood cultures in regional and tertiary hospital settings of Tyrol, Austria (2006-2015): Impacts & trends.

  • Peter Kreidl,
  • Thomas Kirchner,
  • Manfred Fille,
  • Ingrid Heller,
  • Cornelia Lass-Flörl,
  • Dorothea Orth-Höller

DOI
https://doi.org/10.1371/journal.pone.0223467
Journal volume & issue
Vol. 14, no. 10
p. e0223467

Abstract

Read online

Blood stream infections rank among the top seven causes of death of the general population. The aim of our study was to better understand the epidemiology of BSI in order to improve diagnostics and patient outcome. We used retrospective aggregated laboratory data of blood samples received from all public hospitals in Tyrol, Austria between 2006 and 2015. Microorganisms were categorized into obligatory, facultative, unusual pathogens and contaminants. The distribution, the cumulative incidence and antimicrobial susceptibility patterns were compared between the tertiary (TH) and regional peripheral hospitals (PH). Among 256,364 blood samples, 76.1% were from the TH The incidence of obligatory pathogens was 1.7 fold, and up to 3 times higher for facultative, unusual pathogens and contaminants in the TH and increased mainly due to an increase of E.coli, which was the most common isolated pathogen (n = 2,869), followed by Staphylococcus aureus (n = 1,439), Enterococcus sp. (n = 953) and Klebsiella sp. (n = 816). The distribution of obligatory pathogens differed between the hospital settings: In the TH Enterococcus sp. accounted for 40.8% and E.coli for 70.4%, respectively, whereas in the PH for 25.4% (p<0.0001) and 57.8%, respectively (p<0.0001) Antibiotic resistance of Gram negative bacteria and Staphylococcus aureus did not change during the observation period. Carbapenem resistance of Klebsiella sp. and vancomycin and linezolid resistance of Enterococcus faecium showed a non-significant increase since 2010 in the TH setting. We concluded that the incidence of BSI in TH was higher compared to PH. We observed higher contamination rates in the TH. We could not interpret the data of coagulase negative staphylococci due to lack of clinical data. We strongly recommend enhancement of training on blood culture sampling to decrease the rate of contamination. Due to differences in pathogen distribution and antimicrobial resistance between different hospital settings we recommend separate treatment guidelines for BSI by hospital setting.