Elsevier

Journal of Infection

Volume 72, Issue 2, February 2016, Pages 152-160
Journal of Infection

Comprehensive clinical and epidemiological assessment of colonisation and infection due to carbapenemase-producing Enterobacteriaceae in Spain

https://doi.org/10.1016/j.jinf.2015.10.008Get rights and content

Highlights

  • OXA-48-producing Enterobacteriaceae were associated with community-onset, healthcare acquisition.

  • Metallo-β-lactamase (MBL) producers occurred more frequently in ICU patients.

  • OXA-48-producers caused urinary tract infection more frequently than MBL producers.

  • Only 10% of infections were bacteraemic.

  • Non-active targeted therapy was independently associated with increased mortality.

Summary

Background

Most available information on carbapenemase-producing Enterobacteriaceae (CPE) is usually associated with specific types of infection or patient or with descriptions of outbreaks. The aim of this study was to comprehensively analyse the clinical epidemiology, clinical features and outcomes of colonisation and infections due to CPE in Spain.

Methods

A multicentre prospective cohort study was carried out in 34 Spanish hospitals from February to May 2013. All new patients testing positive for CPE in clinical samples were included. Logistic regression was used to identify predictors of mortality.

Results

Overall, 245 cases were included. The most frequent organism was Klebsiella pneumoniae (74%) and the carbapenemases belonged to the OXA-48 (74%), metallo-β-lactamase (MBL) (24%) and KPC (2%) groups. Acquisition was nosocomial in 145 cases (60%) and healthcare-associated (HCA) in 91 (37%); 42% of the latter were nursing home residents, in whom OXA-48-producing K. pneumoniae ST405 predominated. MBLs and OXA-48 predominated in ICU and medical patients, respectively. Overall, 67% of patients had infections. The most frequent infections identified in this study were urinary tract (43%) and skin structure (21%) infections, and 10% of infections were bacteraemic. Crude mortality was 20%. Inappropriate antibiotic therapy was independently associated with an increased risk of death (OR = 3.30; 95% CI: 1.34–8.11).

Conclusions

We found some differences in the epidemiology of CPE depending on the type of carbapenemase produced. Although a low proportion of CPE infections were bacteraemic, active antibiotic therapy was a protective factor for reducing mortality.

Introduction

Carbapenemases are β-lactamases with the ability to hydrolyze carbapenems. In Enterobacteriaceae, these enzymes are largely plasmid-mediated. According to the classification scheme of β-lactamases currently in use, the different carbapenemases found in these bacteria belong to molecular class A (Klebsiella pneumoniae carbapenemase [KPC]), class B or metallo-β-lactamases (MBL) (including VIM, IMP and NDM, among others), and class D (OXA-48 and OXA-48-like enzymes).1, 2 Carbapenemase-producing Enterobacteriaceae (CPE) are rapidly spreading worldwide1 due to both the vertical dissemination of mobile genetic elements harbouring the genes that encode these enzymes and to the transmission of successful epidemic clones.1 Their spread is a serious public health problem because the therapeutic options for infections caused by CPE are usually very limited, or even lacking.1 In fact, a recent systematic review found that infections due to CPE were associated with increased mortality.3

Despite the importance of CPE, the epidemiological and outcome data available to date are associated with studies investigating outbreaks, patients drawn from specific populations, specific types of infection (particularly bacteraemia), and/or specific carbapenemase types; moreover, these data appear to come from countries with a high prevalence of CPE.1, 3, 4, 5 More specifically, the mortality rates reported among patients with bacteraemia due to CPE range from 43 to 96%, but the available data for other types of infection are very scarce and heterogeneous.3 While these studies are important, there is a need for more comprehensive knowledge concerning the epidemiological and clinical features of CPE infections. This study consisted of a nationwide multicentre study in Spain to investigate aspects associated with CPE infection including the importance of non-nosocomial acquisition, whether carbapenemase types are associated with different epidemiological features and the outcome of different types of infection caused by CPE.

Section snippets

Study design and sites

This analysis forms part of a multicentre prospective cohort study performed at 83 Spanish hospitals between February 1st and May 31st 2013, which analysed the clinical and molecular epidemiology of CPE. Briefly, all new patients were eligible for inclusion if any Enterobacteriaceae isolate with diminished susceptibility to carbapenems (either meropenem and/or ertapenem MICs of >0.125 mg/L and/or imipenem MIC of >1 mg/L, or inhibition zone diameters of <25 mm for meropenem and/or ertapenem

Results

Enterobacteriaceae with diminished susceptibility to carbapenems were isolated from 702 patients in the 83 participating centres; 379 isolates produced a carbapenemase. Overall, 79 patients were excluded because the isolates were obtained only from screening samples and 55 because their hospitals did not participate in clinical data collection, so that 245 cases of CPE colonisation or infection from 34 hospitals were finally included in this study. The number of cases per participant hospital

Discussion

We describe here the epidemiological, clinical and outcome features of patients with colonisation and infection due to CPE in Spain, where the overall prevalence of CPE is low but increasing, and is primarily associated with OXA-48 and VIM producers (although some hospital outbreaks caused by KPC producers have also been reported12, 13, 14). Some important data include: (a) more than one third of cases were community-onset but healthcare-associated, many occurring in nursing home residents; (b)

Funding

This work was supported following research programs from the Instituto de Salud Carlos III, Ministerio de Economía y Competitividad: grant number PI12/1242 from the “Fondo de Investigación Sanitaria”; the Antibiotic Resistance Surveillance Programme of the Spanish Centro Nacional de Microbiología; and the Spanish Network for Research in Infectious Diseases (REIPI RD12/0015) – co-financed by European Development Regional Fund “A Way to Achieve Europe” ERDF. It was also supported from funding

Conflicts of interest

GB was speaker for Pfizer, Janssen-Cilag, Novartis and Astellas and received research support for Pfizer. LMM was speaker for Merck, Pfizer, Janssen-Cilag, and Astra-Zeneca and received research support from Merck, Wyeth, and Janssen-Cilag and Astra-Zeneca. AP was a consultant for Merck and Pfizer, has served as speaker for AstraZeneca, Merck and Pfizer and has received research support from Merck and Pfizer. JRB was speaker for Merck, AstraZeneca, Astellas, Novartis and Pfizer, and was

Acknowledgements

Members of GEIH-GEMARA (SEIMC) and REIPI participating in this study: Ángel Zaballos (Centro Nacional de Microbiología, Majadahonda, Madrid); Rafael Cantón and Laura Martínez-García (Hospital Universitario Ramón y Cajal, Madrid); Ana María Fleites and Carlos Rodríguez-Lucas (Hospital Universitario Central de Asturias); Ma Isabel Sánchez-Romero (Hospital Universitario Puerta de Hierro, Majadahonda, Madrid); Luisa García-Picazo (Hospital El Escorial, Madrid); Esteban Aznar and Carolina Campelo

References (24)

  • American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis

    Crit Care Med

    (1992)
  • The European Committee on Antimicrobial Susceptibility Testing

    Breakpoint tables for interpretation of MICs and zone diameters

    (2015)
  • Cited by (0)

    p

    These authors contributed equally to the work.

    q

    Other participants from GEIH-GEMARA (SEIMC) and REIPI are listed in the Acknowledgements section.

    View full text