J Antimicrob Chemother 2013; 68: 487 – 489
doi:10.1093/jac/dks426 Advance Access publication 26 October 2012
Strategies for identification of carbapenemase-producing
Enterobacteriaceae
Patrice Nordmann* and Laurent Poirel
Service de Bactériologie-Virologie, INSERM U914 ‘Emerging Resistance to Antibiotics’, Hôpital de Bicêtre, Assistance Publique/Hôpitaux
de Paris, Faculté de Médecine et Université Paris-Sud, K.-Bicêtre, France
Downloaded from https://academic.oup.com/jac/article/68/3/487/778785 by guest on 23 February 2024
*Corresponding author. Service de Bactériologie-Virologie-Hygiène, Hôpital de Bicêtre, 78 rue de Général Leclerc, 94275 Le Kremlin-Bicêtre Cedex,
France. Tel: +33-1-45-21-36-32; Fax: +33-1-45-21-63-40; E-mail:
[email protected] The spread of carbapenem-hydrolysing b-lactamases in Enterobacteriaceae is becoming a major public health
issue. These b-lactamases hydrolyse almost all b-lactams, are plasmid-encoded and are easily transferable
among enterobacterial species. They are found in multidrug-resistant isolates. Detection of these isolates
from infected specimens first relies on careful recognition of any decreased susceptibility to carbapenems.
After this, rapid biochemical identification of carbapenemase producers using the novel Carba NP test should
be performed. Subsequently, molecular techniques can be used to identify carbapenemase genes if necessary
(epidemiology). Detection of carriers relies on a preliminary screening step, with stools or rectal swabs being
screened on selective culture media such as SUPERCARBA medium, which posseses the broadest spectrum
for detecting any type of carbapenemase producer.
Keywords: antibiotics, carbapenems, b-lactams, resistance, Gram-negatives
Introduction by any of the inhibitors that are in clinical use, but they can be
inhibited in vitro with compounds such as zinc chelators (EDTA,
Multidrug resistance is now emerging increasingly in Enterobac- for example). The OXA-48-type enzymes hydrolyse aminopenicil-
teriaceae among nosocomial and community-acquired infec- lins, ureidopenicillins and carbapenems at low levels, but do not
tions. One of the most important emerging resistance traits significantly hydrolyse broad-spectrum cephalosporins.1,2 Their
corresponds to the production of the carbapenem-hydrolysing activity is not affected by the inhibitors in clinical use, but they
b-lactamases, which confer resistance to almost all are inhibited by NaCl in vitro.2
b-lactams.1 Their current and extensive worldwide spread in
Enterobacteriaceae is an important source of concern since
these carbapenemases are, in most cases, combined with
non-b-lactam resistance mechanisms, therefore leading to Susceptibility breakpoints and the debate
multidrug-resistant isolates.1 The high rate of transmissibility of around detection
the carbapenemase genes, which are mostly located on self-
Both the US and European (EUCAST) breakpoints for carbape-
conjugative plasmids carrying other resistance determinants,
nems have been lowered recently to permit better recognition
explains the need to rapidly identify the carbapenemase produ-
of carbapenem-resistant isolates.3,4 The detection of carbapene-
cers in order to guide antibiotic therapy but also to prevent the
mase producers in clinical specimens is based first on the ana-
development of outbreaks.
lysis of susceptibility testing. According to the US guidelines
(CLSI) updated in 2012,3 the breakpoints for imipenem and
Categories of carbapenemases and their meropenem are susceptible (S) ≤1 and resistant (R) ≥4 mg/L,
and for ertapenem S ≤0.5 and R ≥2 mg/L. Ertapenem seems
properties
to be a good candidate for detecting most of the carbapene-
Carbapenemases in Enterobacteriaceae are currently mostly of mase producers since MIC values of ertapenem are usually
the KPC, VIM, IMP, NDM and OXA-48 types, and their detailed higher than those of other carbapenems.2 However, detection
properties have been extensively reported.1 KPC enzymes hydro- of carbapenemase producers based only on MIC values may
lyse all b-lactams (although they hydrolyse cephamycins at a lack sensitivity. According to the CLSI and EUCAST guidelines,
low level) and their activity is only inhibited partially in vitro breakpoints have to be considered only when reporting suscepti-
by clavulanic acid, tazobactam and boronic acid. The bility or resistance to carbapenems.3,4 Special tests for carbape-
metallo-b-lactamases (MBLs; IMP, VIM and NDM) hydrolyse all nemase detection are suggested for epidemiology and infection
b-lactams except aztreonam and their activity is not affected control issues only.3,4
# The Author 2012. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.
For Permissions, please e-mail: [email protected]
487
Leading article
We actually disagree with these guidelines, for the follow- is the Carba NP test.9 This biochemical test, applicable to isolated
ing reasons: (i) susceptibility to carbapenems is observed for bacterial colonies, is based on in vitro hydrolysis of the carbapenem
several carbapenemase producers; and (ii) there is a paucity of imipenem.9 Hydrolysis of imipenem is detected by a change in the
clinical successes of carbapenem-containing regimens for treat- pH value of the indicator (red to yellow/orange). This test is 100%
ing infections due to carbapenemase producers that are suscep- sensitive and specific, as are molecular techniques.9 It detects not
tible to carbapenems in vitro.1,2 Based on our own experience, we only all known carbapenemases (belonging to Ambler A, B and D
propose that, as a minimum, detection of carbapenemase activ- classes) in Enterobacteriaceae but should also identify virtually
ity should be performed on enterobacterial isolates exhibiting any new emerging carbapenemase, in contrast to molecular tech-
MIC values of ertapenem ≥0.5 mg/L or of imipenem or merope- niques. This rapid (,2 h), user-friendly and inexpensive technique
nem ≥1 mg/L, and also on any enterobacterial isolate that exhi- can be implemented in any laboratory worldwide. In addition, it
bits even a slight decrease in susceptibility to carbapenems does not require any specific equipment. We believe this technique
compared with a wild-type phenotype. This detection will be will soon become a reference technique since it fulfils the clinical
Downloaded from https://academic.oup.com/jac/article/68/3/487/778785 by guest on 23 February 2024
useful for treating patients and for preventing nosocomial outbreaks requirement of a rapid and low-cost identification method for
of carbapenemase producers (and therefore multidrug-resistant carbapenemase-producing Enterobacteriaceae.
isolates), whatever the carbapenem resistance level is.
Molecular tests for carbapenemase genes
Non-molecular tests for carbapenemase
production Molecular techniques remain the gold standard for the precise
identification of carbapenemase genes.2 Most of these techniques
A series of non-molecular tests have been proposed for detection of are based on PCR and may be followed by a sequencing step if a
carbapenemase activity.2 Some have good specificity and sensitiv- precise identification of the carbapenemase gene is needed (e.g.
ity, but none of them approaches 100%. The modified Hodge test VIM type, KPC type, NDM type or OXA-48 type).10,11 They are
(MHT), based on in vivo production of a carbapenemase, has been either single or multiplex PCR techniques. A PCR technique per-
used for years.3 Addition of zinc to the culture medium has been formed directly on colonies can give results within 4 –6 h (or less
shown to improve the sensitivity of this test.5 However, we believe when using real-time PCR technology) with excellent sensitivity
that this technique is time-consuming and shall be discarded. In and specificity. Similarly, other molecular techniques are useful
addition, MHT may lack sensitivity for detecting carbapenemase ac- for this purpose.12 The main disadvantages of the molecular-
tivity in Enterobacter spp. The added value of the inhibition-based based technologies are their cost, the requirement for trained
carbapenemase detection tests remains variable.2 These tests microbiologists and the inability to detect novel unidentified
are, for example, based on inhibition by tazobactam, clavulanic genes. Sequencing of the genes is interesting mostly for research
acid or boronic acid for detecting the production of Ambler class A and epidemiological purposes. Precise identification of the type
carbapenemases (KPC), and inhibition by EDTA or dipicolinic acid of carbapenemase is not actually needed for treating patients or
for detection of MBL activity. They are time-consuming and have for preventing outbreaks. We believe these molecular techniques
variable sensitivity and specificity. In addition, they require trained may be mostly used in reference laboratories.
microbiologists. High-level resistance to temocillin has been sug-
gested to be predictive of production of OXA-48, but the specificity
of this feature remains to be more extensively evaluated.2
Screening of colonized patients
Detection of carbapenemase activity can be done using a UV
spectrophotometer, which is available in many microbiology la- Preventing the spread of carbapenemase producers relies on the
boratories. It is based on several steps, including: (i) an 18 h accurate detection of colonized patients at an early stage of hos-
culture (which can be shortened in some cases to 8 h); (ii) a pitalization or on admission/discharge either to the hospital or to a
protein extraction step; and (iii) measurement of imipenem hy- specific unit. Screening should include as a minimum ‘at-risk’
drolysis using a UV spectrophotometer.6 We have shown that patients, such as those in intensive care units, transplant recipients
this spectrophotometry-based technique has 100% sensitivity and the immunocompromised, and those transferred from any
and 98.5% specificity for detecting any kind of carbapenemase ac- foreign hospital (unknown prevalence of carbapenemase produ-
tivity.6 This cheap technique can accurately differentiate carbape- cer carriage) or from non-foreign hospitals but known to face a
nemase producers from non-carbapenemase producers among high risk of carriage of carbapenemase producers.1 Geographical
carbapenem-non-susceptible isolates [outer membrane perme- regions with high prevalence of carbapenemase producers are in-
ability defect, overproduction of cephalosporinases or/and creasingly known.1 Since the reservoir of Enterobacteriaceae is
extended-spectrum b-lactamases (ESBLs)]. It can be implemen- mostly the intestinal flora, stools and rectal swabs are the most suit-
ted in any reference laboratory, but this technique still requires able specimens for performing such screening. Three screening
time. Recently, the use of mass spectrometry for detection of car- media are currently known, but mostly for carbapenem-resistant
bapenemase activity has been proposed, based on the analysis of rather than carbapenemase-producing organisms.
the degradation of a carbapenem molecule.7,8 Although this tech- The first marketed screening medium was the CHROMagar KPC
nique has to be further evaluated, matrix-assisted laser desorption medium, which contains a carbapenem (CHROMagar, Paris,
ionization time-of-flight mass spectrometry (MALDI-TOF) equip- France).13 It detects carbapenem-resistant bacteria only if they
ment is increasingly used in the diagnostic bacteriology laboratory. exhibit high-level resistance to carbapenems. Its main disadvan-
The most important and recent development for the accurate tage therefore remains its lack of sensitivity since it does not
identification of carbapenemase-producing Enterobacteriaceae detect carbapenemase producers exhibiting a low level of
488
Leading article JAC
carbapenem resistance, as observed for several MBL or OXA-48 pro- funding from INSERM. P. N. is a director of the INSERM U914 Emerging
ducers.1 The second screening medium also contains a carbapenem Resistance to Antibiotics.
(CRE Brilliance, Thermo Fisher Scientific, UK).14 It detects KPC and MBL
producers well, and most but not all OXA-48 producers.14,15 Finally,
one of the most recently developed screening media (SUPERCARBA) References
contains cloxacillin, zinc and ertapenem.16 It shows excellent sensitiv- 1 Nordmann P, Naas T, Poirel L. Global spread of carbapenemase-producing
ity and specificity for detection of any kind of carbapenemase produ- Enterobacteriaceae. Emerg Infect Dis 2011; 17: 1791–8.
cer (not only high-level carbapenem-resistant isolates).16 Compared 2 Nordmann P, Gniadkowski M, Giske CG et al. Identification and
with the two other media, it also shows improved sensitivity and spe- screening of carbapenemase-producing Enterobacteriacae. Clin
cificity for detecting all types of carbapenemase producers (including Microbiol Infect 2012; 18: 432– 8.
the OXA-48 producers) when present in low amounts in stools. Once 3 Clinical and Laboratory Standards Institute. Performance Standards for
carbapenem-resistant isolates are selected on SUPERCARBA Antimicrobial Susceptibility Testing: Twenty-first Informational Supplement
Downloaded from https://academic.oup.com/jac/article/68/3/487/778785 by guest on 23 February 2024
medium, we recommend use of the Carba NP test for detecting M100-S21. CLSI, Wayne, PA, USA, 2012.
carbapenemase activity. Then, if needed, molecular identification of 4 Leclercq R, Canton R, Brown DF et al. EUCAST expert rules in
the carbapenemase genes may be performed. antimicrobial susceptibility testing. Clin Microbiol Infect 2011;
doi:10.1111/j.1469-0691.2011.03703.x
Conclusion 5 Girlich D, Poirel L, Nordmann P. Value of the modified Hodge test for
detection of emerging carbapenemases in Enterobacteriacae. J Clin
Proposed algorithm for carbapenemase detection Microbiol 2012; 50: 477– 9.
We propose the following scheme for detection of carbapene- 6 Bernabeu S, Poirel L, Nordmann P. Spectrophotometry-based detection
of carbapenemase producers in Enterobacteriaceae. Diagn Microbiol
mase producers in Enterobacteriaceae.
Infect Dis 2012; 74: 88– 90.
(i) Infecting strains: Carba NP test on isolated colonies with
decreased susceptibility to carbapenems. If positive, molecular 7 Burckhardt I, Zimmermann S. Using matrix-assisted laser desorption
identification of the genes, mainly for epidemiological reasons. ionization-time of flight mass spectrometry to detect carbapenem
resistance within 1 to 2.5 hours. J Clin Microbiol 2011; 49: 3321– 4.
(ii) Screening of carriers: screening of carbapenem-resistant
isolates using e.g. SUPERCARBA medium, followed by the Carba 8 Hrabak J, Walkova R, Studentova V et al. Carbapenemase activity
NP test to be performed on selected colonies. If the latter test detection by matrix-assisted laser desorption ionization-time of flight
mass spectrometry. J Clin Microbiol 2011; 49: 3327– 32.
is positive, molecular identification of the genes, mainly for
epidemiological reasons. 9 Nordmann P, Poirel L, Dortet L. Rapid detection of carbapenemase-
The proposed scheme for detection of carbapenemase produ- producing Enterobacteriaceae. Emerg Infect Dis 2012; 18: 1503–7.
cers presents several advantages. It will lead to rapid identifica- 10 Poirel L, Walsh TR, Cuvillier V et al. Multiplex PCR for detection of acquired
tion of carbapenemase producers, allowing adequate antibiotic carbapenemase genes. Diagn Microbiol Infect Dis 2011; 70: 119–25.
stewardship. This is of the utmost importance due to the 11 Avlami A, Bekris S, Ganteris G et al. Detection of metallo-b-lactamase
current shortage of novel antibiotics. It will prevent the develop- genes in clinical specimens by a commercial multiplex PCR system.
ment of nosocomial outbreaks due to the multidrug-resistant J Microbiol Methods 2010; 83: 185– 7.
bacteria that represent one of the most worrisome medical 12 Cuzon G, Naas T, Bogaerts P et al. Evaluation of a DNA microarray for
threats. The Israeli experience with containment of the KPC out- the rapid detection of extended-spectrum b-lactamases (TEM, SHV and
breaks has demonstrated the possibility of obtaining frank CTX-M), plasmid-mediated cephalosporinases (CMY-2-like, DHA, FOX,
success by acting rapidly, and on a large scale.17 This strategy, ACC-1, ACT/MIR and CMY-1-like/MOX) and carbapenemases (KPC,
OXA-48, VIM, IMP and NDM). J Antimicrob Chemother 2012; 67: 1865– 9.
based on broad and rapid detection of carbapenemase produ-
cers, may also have a significant impact in preventing their 13 Moran-Gilad J, Carmeli Y, Schwartz D et al. Laboratory evaluation of
spread in the community. Finally, the first proposed steps the CHROMagar KPC medium for identification of carbapenem-non-
(screening, susceptibility testing and Carba NP test), excluding susceptible Enterobacteriaceae. Diagn Microbiol Infect Dis 2011; 70: 565–7.
the molecular techniques (needed for epidemiological purposes), 14 Withey S, Scopes E. A new screening medium for detection of
can be implemented immediately in any laboratory worldwide, carbapenem-resistant Enterobacteriaceae. In: Abstracts of the
since they are based on inexpensive and affordable techniques. Twenty-first European Congress for Clinical Microbiology and Infectious
Diseases, Milan, 2011. Abstract P862. European Society for Clinical
Microbiology and Infectious Diseases, Basel, Switzerland.
15 Girlich D, Poirel L, Nordmann P. Strategy of detection of
Funding carbapenemase-producing Enterobacteriaceae. In: Abstracts of the
This work was partially funded by grants from the INSERM (UMR914) and Twenty-second European Congress for Clinical Microbiology and Infectious
the University Paris Sud, France. Diseases, London, 2012. Abstract P2315. European Society for Clinical
Microbiology and Infectious Diseases, Basel, Switzerland.
16 Nordmann P, Girlich D, Poirel L. Detection of carbapenemase
Transparency declarations producers in Enterobacteriaceae by use of a novel screening medium.
This work was partially funded by grants from the Institut National de la J Clin Microbiol 2012; 50: 2761 –6.
Santé et de la Recherche Médicale (INSERM) (UMR914). An international 17 Schwaber MJ, Lev B, Israeli A et al. Containment of country-wide
patent form for the Carba NP test has been filed on behalf of INSERM outbreaks of carbapenem-resistant Klebsiella pneumoniae in Israeli hospitals
Transfert (Paris, France). P. N. and L. P. are employed by and receive via a nationally implemented intervention. Clin Infect Dis 2011; 52: 848–55.
489
