Clinical and microbiological determinants of severe and fatal outcomes in patients infected with Enterobacteriaceae producing extended-spectrum β-lactamase

L. Surgers, A. Boyd, P. Y. Boelle, V. Lalande, P. A. Jolivot, P. M. Girard, G. Arlet, C. Cambier, A. Homor, D. Decre, J. L. Meynard

Research output: Contribution to journalArticleAcademicpeer-review

8 Citations (Scopus)

Abstract

Although extended-spectrum β-lactamase (ESBL)-producing Enterobacteriaceae have become a worldwide public health concern, little is known regarding the clinical course of colonized or infected individuals. Our objective was to characterize the determinants of fatal outcomes related to ESBL-producing microorganisms at a large hospital in Paris, France. In 2012–2013, all consecutive patients with clinical samples testing positive for ESBL-producing Enterobacteriaceae at Saint-Antoine Hospital were identified. Patient clinical data were obtained at hospital entry, while information on intensive care unit (ICU) admissions and death were prospectively collected. Risk-factors for fatal 1-year outcomes were assessed using logistic regression. In total, 643/4684 (13%) ESBL-positive samples were observed, corresponding to 516 episodes (n = 206, 40% treated) among 330 patients. Most episodes were nosocomial-related (n = 347/516, 67%) involving Escherichia coli (n = 232/516, 45%) or Klebsiella pneumoniae (n = 164/516, 32%). Empirical antibiotic therapy was adequate in 89/206 (43%) infections, while the median length of hospital stay was 30 days [interquartile range (IQR) = 11–55] and 39/201 (19%) were admitted to the ICU. Overall, 104/241 patients (43%) with available data died within 1 year. In the multivariable analysis, 1-year death was associated with age >80 years (p = 0.01), concomitant comorbidity (p = 0.001), nosocomial-acquired infection (p = 0.002), and being infected rather than colonized (p < 0.001). In this series of patients with identified samples of ESBL-producing Enterobacteriaceae, hospital burden was large and 1-year mortality rates high. Understanding which patients in this setting would benefit from broad-spectrum empirical antibiotic therapy should be further examined.

Original languageEnglish
Pages (from-to)1261-1268
Number of pages8
JournalEuropean Journal of Clinical Microbiology and Infectious Diseases
Volume36
Issue number7
DOIs
Publication statusPublished - 1 Jul 2017

Cite this