Carbapenem-resistant Enterobacteriaceae: what has happened, and what is being done

The Enterobacteriaceae is a large, heterogenous family of gram-negative rods consisting of more than 30 genera and hundreds of species. This family of bacteria is the most common cause of bacterial disease with both hospital- and community-acquired infections involving all body sites.¹ Historically, most members of the family were susceptible to a wide spectrum of antibiotics, including broad-spectrum cephalosporins and penicillins, monobactam, and carbapenem antibiotics.²

In the early 1980s, the first extended spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae were reported in Germany. The original isolate produced a variant of the SHV-1 beta-lactamase present in Klebsiella pneumoniae that extended the spectrum of beta-lactam antibiotics that were hydrolyzed by the modified enzyme to include all penicillin, cephalosporin, and monobactam antibiotics. Similar ESBLs derived from TEM-1 and TEM-2 beta-lactamases present in Escherichia coli were soon reported, as well as a third class of ESBLs, CTX-M (named for the ability to hydrolyze cefotaxime). CTX-M-producing Enterobacteriaceae are now the most common ESBLs in many geographic areas.^3,4

To date, more than 200 ESBLs have been described, and bacterial strains carrying these enzymes have been isolated throughout the world.^3,5,6 Because most enzymes are encoded on plasmids, they are readily spread from organism to organism, are found in many species of Enterobacteriaceae as well as other gram-negative rods such as Pseudomonas and Acinetobacter, and are associated with resistance genes for other classes of antibiotics.^3,5,6 The carbapenems are the empiric treatment of choice for these organisms.⁷

Carbapenem overuse—and its consequences

Just as the widespread use of extended spectrum cephalosporins selected for the emergence of ESBL-producing Enterobacteriaceae, carbapenem overutilization was instrumental in selecting the next generation of resistant gram-negative bacteria. It is interesting to note that fewer than 15 years ago the isolation of carbapenem-resistant Enterobacteriaceae (CRE) was almost certainly the result of testing errors. Unfortunately, these resistant organisms are now established in many hospitals worldwide,^8,9 and, as was observed with ESBLs, the spectrum of carbapenemases is broad and heterogeneous. The Ambler molecular classification scheme divides beta-lactamases into four classes (A-D), with carbapenemases found in classes A, B, and D.¹⁰

The most common carbapenemases are the class A KPC (Klebsiella pneumoniae carbapenemase) type enzymes, the class B metallo-ß-lactamases (including NDM, VIM, and IMP enzymes), and the class D OXA enzymes.^11-14 The term KPC is a misnomer because the enzyme is not restricted to K. pneumoniae; it is also found in other Klebsiella species, Escherichia coli, Serratia marcescens, Enterobacter species, Citrobacter species, Salmonella enterica, and other Enterobacteriaceae. Likewise, the other carbapenemases are widely distributed among the Enterobacteriaceae as well as other gram-negative genera including Acinetobacter and Pseudomonas. The widespread prevalence of carbapenemases is because the resistance genes are encoded on plasmids or mobile genetic structures.

The KPC enzymes were first described in the North Carolina area but have spread throughout North America and more recently worldwide. NDM (New Delhi metallo-ß-lactamase) spread from the Indian subcontinent to Europe, the Middle East, and Asia.⁹ VIM, IMP, and OXA are currently less common than KPC and NDM and are regionally distributed with sporadic outbreaks in Europe, North Africa, Asia, and North America. With the exception of KPC, CRE in the United States are primarily associated with patients who have traveled to other countries; however, it should be recognized that much like the spread of ESBL strains and KPC strains, it is only a matter of time before the other carbapenemase-producing strains are common in U.S. hospitals. Clearly the opportunity exists today to introduce effective control measures to prevent the spread of these bacteria.

Getting control—strategies and best practices

The U.S. Centers for Disease Control and Prevention (CDC) has issued guidelines for the control of CRE in acute-care and long-term care facilities.⁸ Eight core measures have been recommended:

• Hand hygiene with the use of alcohol-based hand rubs or soap and water before and after contact with a patient;
• Contact precautions for patients identified as colonized or infected with CRE or patients at increased risk (for example, previously colonized with CRE, transferred from a hospital unit or outside facility with active CRE disease, recent antibiotic exposure). Contact precautions include hand hygiene and the donning of gown and gloves before entering the patient’s room and removal before exiting;
• Healthcare personnel education about CRE, methods of transmission, and proper use of hand hygiene and contact precautions;
• Use of devices such as central venous catheters, endotracheal tubes, and urinary catheters should be minimized or, when required, used appropriately;
• Patients should be in private rooms or, if that is not feasible, cohorted with other colonized patients. Staff should also be restricted to care of CRE patients if possible, although this would imply the presence of a large colonized or infected population, which is a bad omen;
• Prompt laboratory notification of positive lab results for CRE to both the patient care area and infection control personnel. For the patient care area this should be telephone notification; printed results are insufficient. For the infection control personnel, this can be accomplished by use of printed reports that are available at a specific time each day, when a member of the infection control team visits the laboratory;
• Antimicrobial stewardship—structured guidance for the responsible selection and use of antimicrobial agents—is not restricted to carbapenems alone. All broad-spectrum antibiotics should be used appropriately, providing narrow focused therapy for susceptible organisms and for the proper duration;
• CRE screening. This should include both the use of appropriate standardized susceptibility test methods to recognize CRE and the establishment of a screening program to detect colonization in asymptomatic patients.

The challenges we face—and some lessons we are learning

The difficulty in controlling the spread of CRE in a hospital setting was illustrated by Snitkin et al,¹⁴ who described their efforts to terminate a KPC outbreak at the National Institutes of Health (NIH) Clinical Center. Despite the use of sophisticated, real-time sequencing techniques to characterize and track the outbreak strains and rigorous infection control practices, the outbreak strains persisted for more than six months, infecting 18 patients and resulting in six deaths. Whole genome sequencing identified the complexity of this outbreak, with three distinct episodes of transmission, illustrating the challenges involved with eliminating the organism. Sandora and Goldmann¹⁵ correctly pointed out when assessing this outbreak that resistant organisms “are mainly on the hands of caregivers who do not practice effective hand hygiene after every contact with patients and their environment.” Any infection control program is only as effective as the weakest link.

I found the NIH experience particularly troubling because, as a former member of the clinical staff at the NIH, I have first-hand knowledge of the dedicated efforts employed by the entire medical staff and infection control team to control not only this outbreak, but also the practices used in daily preventive efforts for the hospital population. When efforts such as these have limited effectiveness, it illustrates the difficult battle that confronts us.

Critical to the control of ESBL and CRE strains is recognition of resistant isolates when recovered in clinical specimens. Although this is simple in concept, manual and automated susceptibility testing methods were historically insensitive because many of the resistant organisms had relatively low levels of resistance; that is, the Minimum Inhibitory Concentration (MIC) values were elevated compared with fully susceptible organisms but still within the susceptible category. In recent years, both Clinical Laboratory Standards Institute (CLSI) and European Committee on Antimicrobial Susceptibility Testing (EUCAST) have modified the interpretive breakpoints to classify ESBL and CRE strains more accurately, although this has not been fully successful for all commercial susceptibility test systems.^16,17 Enterobacteriaceae with KPC enzymes have decreased susceptibility or resistance to all ß-lactam antibiotics.

In contrast, the metallo-ß-lactamases (e.g., NDM, VIM, IMP) have variable activity against the penicillins, cephalosporins, and carbapenems and no activity against aztreonam. Additionally, the OXA-type enzymes can have weak activity against the carbapenems and extended spectrum cephalosporins.¹¹ The most reliable method for detecting carbapenemase production, including in organisms with low-level resistance, is with DNA-based multiplex assays. Several assays have been primarily used with isolated organisms to confirm carbapenem resistance.^4,17-20 In addition, an assay has been used successfully to screen rectal swabs from ICU patients in Europe.

The carbapenemase multiplex assays are not without limitations. The carbapenemase resistance gene must be included in the multiplex targets, which is a challenge as the number of genes continues to increase. Additionally, assays that target detection of carbapenemase genes will not be useful for recognition of resistance mediated by other mechanisms, such as permeability barriers that prevent uptake of the antibiotics by the bacteria or efflux mechanisms that pump the antibiotic out of the cell before it can affect the organism. Selective culture media, such as commercially produced chromogenic agars, have been used as a general screening method for detection of carbapenem-resistant organisms in clinical specimens such as rectal swabs.^19,21,22 Screening agars appear to have good sensitivity for detection of drug-resistant bacteria, although comprehensive evaluations of the agars are lacking and not all commercial agars are equivalent.²² The decision to use a particular medium should be guided by an internal validation study. Additionally, optimum sensitivity of agar-based screening requires use of an enrichment broth and confirmatory testing of suspected resistant organisms, so definitive results may not be available for three to four days after the specimen is collected.

In summary, the widespread distribution of multidrug-resistant Enterobacteriaceae is a challenge for both physicians and laboratory scientists. Physicians have a limited selection of antibiotics available for treating infections caused by these organisms, and laboratory scientists are handicapped by the diagnostic tools available. Although carefully performed in vitro susceptibility tests are the foundation for detecting these resistant organisms, use of selective agars and multiplex molecular tests are important tools for screening asymptomatic patients. The combination of screening high-risk patients and use of established infection control practices are critical for preventing the introduction and spread of these organisms in a hospital population.