- Email: [email protected]
Educational antimicrobial susceptibility testing as a critical component of microbiology laboratory proficiency programs: American Proficiency Institute results for 2007–2011
Diagnostic Microbiology and Infectious Disease 75 (2013) 357–360
Contents lists available at SciVerse ScienceDirect
Diagnostic Microbiology and Infectious Disease journal homepage: www.elsevier.com/locate/diagmicrobio
Educational antimicrobial susceptibility testing as a critical component of microbiology laboratory proficiency programs: American Proficiency Institute results for 2007–2011 Ronald N. Jones a,⁎, Teri Glick b, Helio S. Sader a, Robert K. Flamm a, James E. Ross a, Paul R. Rhomberg a, Daniel C. Edson b a b
JMI Laboratories, North Liberty, IA, USA American Proficiency Institute, Traverse City, MI, USA
a r t i c l e
i n f o
Article history: Received 9 December 2012 Received in revised form 10 January 2013 Accepted 31 January 2013 Keywords: API Proficiency programs Microbiology Susceptibility tests Accuracy Quality assurance
a b s t r a c t External laboratory proficiency programs are an important requirement for test quality assurance (EQA) and compliance to regulatory guidelines (Clinical Laboratory Improvement Amendments and inspections). The American Proficiency Institute (API) regularly distributes EQA sample challenges (test events) including an Educational Sample (ES) for antimicrobial susceptibility testing. Beginning in 2007, API has sent 3 ES samples annually, each a well-characterized (molecular/phenotypic methods) strain having an interesting/emerging mechanism of resistance. Hundreds of USA laboratories, usually serving small- to medium-size hospitals and clinics, participate in the API ungraded ES test event. Analysis of responses is made and reported electronically as ES critiques addressing contemporary susceptibility testing issues that affect patient therapy. Seven Grampositive (+) and 8 Gram-negative (−) ES strains were tested over the 5 years (2007–2011) with organism identification (graded) accuracy of 95.3% (range, 91.0–99.2%) for Gram (−) and 97.0% (range, 94.2–100.0%) for Gram (+) challenges. Susceptibility testing categorical accuracy was generally greatest for the disk diffusion test (91.0/97.0%) compared to the MIC methods (commercial automated or manual) combined (89.9/96.1%, for Gram [−]/Gram [+], respectively). The most worrisome observations of these ES samples were as follows: 1) poor recognition of ESBL- and serine carbapenemase-producing strains (various types including Klebsiella pneumoniae carbapanemase) due to delayed application of Clinical and Laboratory Standards Institute [CLSI] guidelines; 2) overcalling of ESBL in organisms having wild-type non-ESBL enzymes (OXA series; OXA, 1/30) due to commercial system or participant interpretive error; and 3) occasional drugbug discords noted in nonfermentative Gram (−) bacilli. In conclusion, the API ES series of ungraded susceptibility testing challenges (accuracy was N90%) has been well received by subscribers and has provided detailed educational opportunities to improve laboratory testing performance. ES samples have delivered guidance to enable laboratories to rapidly comply with CLSI document changes of interpretive breakpoints such as those for β-lactams when testing Enterobacteriaceae and Pseudomonas aeruginosa; the program was sustained into 2012 and beyond to document quality of susceptibility tests in USA laboratories. © 2013 Elsevier Inc. All rights reserved.
1. Introduction External quality assurance (EQA) programs are important components of laboratory practice that are essential for maintaining clinical laboratory test accuracy. These EQA practices complement inspection and accreditation organizations mandated via professional societies (e.g., College of American Pathologists [CAP]) and the USA government (Clinical Laboratory Improvement Amendments [CLIA]). These USA programs have assessed the implementation and accuracy of antimicrobial susceptibility testing methods over 4 decades with published summaries dating from 1969 (Gavan, 1974; Gavan and ⁎ Corresponding author. Tel.: +1-319-665-3370; fax: +1-319-665-3371. E-mail address: [email protected] (R.N. Jones). 0732-8893/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.diagmicrobio.2013.01.027
King, 1970). The initial detailed publications, however, appeared beginning in 1982 covering the CAP Microbiology EQA surveys (Jones and Edson, 1982a; 1982b; Jones et al., 1982) and the most recent summarized testing events through 2005 (Jones and Fritsche, 2005; Pfaller and Jones, 2006). Special reports addressing quality control (QC) practices and specific test accuracy were also published (Jones, 1983; Jones and Edson, 1983). Interim analyses were reported by Doern et al. (1999) as well as by Jones and colleagues (Jones, 1992, 2001; Jones and Edson, 1983, 1985; Jones and Edson, 1988, 1991; Jones and Sommers, 1986). The American Proficiency Institute (API) is a federally approved EQA proficiency testing provider (since 1991) serving over 17,000 hospitals, clinics, and physician offices/laboratories (Edson et al., 2005). The API initiated a specific educational microbiology objective
358
R.N. Jones et al. / Diagnostic Microbiology and Infectious Disease 75 (2013) 357–360
(2007) that was designed to monitor and enhance antimicrobial susceptibility testing quality. This Educational Sample (ES) program distributes 3 organisms each year for species identification and testing of antimicrobials (both ungraded) by the routine method applied by the participating laboratories. The susceptibility test results are collected as categories (susceptible, intermediate, or resistant), and responses are analysed as an educational process to improve laboratory quality/education. A total of 15 ES challenges were assessed from 2007 through 2011, generally revealing high levels of test accuracy, yet documenting distinct areas for test improvement; summaries of the API ES Microbiology EQA Program are presented here.
2. Materials and methods
2.2. Participants and analysis Over 800 USA laboratories, usually serving hospitals and clinics, participate in each microbiology ES test event (N90% from commercial MIC methods or systems). Analysis of responses was made and reported electronically within a few weeks of deadlines as ES critiques addressing contemporary susceptibility testing issues that affect the quality of contemporary infection therapy. Categorical determinations of the participant results were classified by Clinical and Laboratory Standards Institute (CLSI), European Committee on Antimicrobial Susceptibility Testing (EUCAST), and US Food and Drug Administration breakpoint criteria, where appropriate (CLSI, 2012c; EUCAST, 2012). Reference methods of the CLSI (2012a,b) were used to characterized EQA ES samples, as well as specific molecular tests/ sequencing of resistance mechanisms.
2.1. Program description 3. Results and discussion As a component of a comprehensive Microbiology Proficiency Sample Program that has 43 distinct microbiology options, API offers the ES antimicrobial susceptibility testing samples (ungraded) every 4 months (e.g., 3 samples per year). This program was launched in early 2007 to circulate an organism sample, each a well-characterized (molecular/ phenotypic level methods) strain having an interesting or an emerging mechanism of resistance for educational purposes (see Table 1).
Table 1 lists the topics of the ES challenges for Enterobacteriaceae (4 species) with SHV-5, KPC-3 (2 events), OXA-series, and OXY-series β-lactamases; for nonfermentative Gram-negative (−) bacilli with multidrug-resistant (MDR) patterns; for S. aureus of the USA300 and USA100 types; for other Gram-positive (+) cocci (viridans group streptococci, E. gallinarum, S. agalactiae, and S. haemolyticus) with
Table 1 API Educational Sample (ES) topics for susceptibility testing (2007–2011) that focused on emerging resistance patterns and susceptibility testing problems/breakpoints. Year/sample no.
Species
Resistance educational topic
2007 ES-01
E. coli
Extended spectrum β-lactamase (SHV-5), an ESBL enzyme producing elevated MIC results for ceftazidime and monobactams. Monitored accuracy of ESBL detection by CLSI screening methods. Methicillin-resistant (MRSA) community-acquired strain (USA-300-0114) typical of emerging clonal type. Monitored accuracy of MRSA detection by CLSI methods for key β-lactam, macrolide, and fluoroquinolone resistances. Carbapenem-resistant (KPC-3) and typical of emerging resistance type in northeast USA and worldwide. Monitored detection accuracy of this enzyme type via current CLSI methods.
ES-02
S. aureus
ES-03
K. pneumoniae
2008 ES-01
S. mitis/oralis (S. peroris)
ES-02
K. oxytoca
ES-03
S. haemolyticus
2009 ES-01
P. aeruginosa
ES-02
S. pneumoniae
ES-03
A. baumannii
2010 ES-01
E. faecalis
ES-02
S. marcescens
ES-03
S. aureus
2011 ES-01
S. agalactiae
ES-02
E. gallinarum
ES-03
E. coli
High-level penicillin (β-lactam) resistance in a viridans streptococcus species associated with other antimicrobial resistances to clindamycin, macrolides, quinupristin–dalfopristin, fluoroquinolones, aminoglycosides, rifampin, tetracycline, and trimethoprim–sulfamethoxazole (TMP-SMX). Monitored accuracy of testing fastidious species. ESBL-like enzyme (OXY-series) having resistances to penicillins ± clavulanate, or sulbactam, or tazobactam, cephalosporins, and aztreonam. Monitored for detection of enzyme more commonly found in this not-uncommon Klebsiella spp. Multidrug-resistant (MDR) coagulase-negative staphylococcus (CoNS) with high monitored MIC results to methicillin, other β-lactams, macrolides, fluoroquinolones, and TMP-SMX. MDR strain only susceptible to amikacin, gentamicin, tobramycin, and polymyxins (colistin, polymyxin B). Monitored ability to recognize the few remaining treatment antimicrobials. An international quality control (QC) strain used to determine the technical accuracy of CLSI and other test methods. Results were compared to published QC ranges of MICs and zone diameters. Only minor levels of β-lactam resistances were present in this well-known strain. Acinetobacter harboring an OXA-23 carbapenemase producing a MDR organism emerging in the USA and worldwide. Only a few agents (aminoglycosides, ceftazidime, polymyxins, tigecycline, and TMP/SMX) were clearly active with their identification accuracy assessed. MDR strain with well-defined oxazolidinone resistance (target site mutations) that has appeared in recent years. Monitored multiple methods to detect a rare but emerging resistance that could decrease the utility of an important antimicrobial class. MDR strain with a KPC-3 serine carbapenemase (see mechanism in ES-03, 2007) having susceptibility to a few drugs (amikacin, gentamicin, and tigecycline). Appraised the use of recently published CLSI breakpoints for carbapenems and confirmatory methods (Modified Hodge Test). Healthcare-associated strain (USA 100) with a characteristic MDR pattern but retaining susceptibility to at least 10 antimicrobials including parenteral and oral agents. Monitored testing accuracy for commonly tested agents and methods. Group B streptococcus strains selected for elevated penicillin MICs and having high-level resistance to fluoroquinolones, macrolides, tetracyclines, and clindamycin. Monitored ability to recognize unusual resistances in β-hemolytic streptococci. Unusual enterococcus having a vancomycin resistance (VanA) with an intrinsic vanC1 gene. Monitored for misidentifications that were noted/documented across various commercial systems, usually naming the isolates an E. faecium or E. faecalis (only challenge not graded for identification, 2007–2011.) MDR pattern (resistant to penicillin with or without β-lactamase inhibitors, fluoroquinolones, aminoglycosides, and tetracyclines) with OXA-1/30 and TEM-1 enyzmes. Some participants erroneously diagnosed this organism as an ESBL producer by application of so-called expert systems. Monitored all key resistances in a communityacquired UTI sample.
R.N. Jones et al. / Diagnostic Microbiology and Infectious Disease 75 (2013) 357–360
MDR patterns; and for the S. pneumoniae quality control strain (ATCC 49619) having a penicillin nonsusceptibility. These topics were selected to address educational objectives of 1) recognizing emerging antimicrobial resistance mechanisms, 2) highlighting recent interpretive breakpoint criteria changes, 3) illustrating intermethod or system errors as applied to resistance screening/detection, and 4) emphasizing the important clonal, usually hospital-based, pathogens having MDR patterns. The objectives were met in a prospective manner as guided by international surveillance program results from the Jones Microbiology Institute (JMI Laboratories, North Liberty, IA, USA). Participant laboratories performed very well at an acceptable level of organism identification (Table 2). Identification of the Gram (+) species strains (97.3% acceptable) was slightly greater than that achieved for Gram (−) species (95.9%), but these rates were comparable to reports from other EQA surveys reported in the most recent years (Jones and Fritsche, 2005; Pfaller and Jones, 2006). Table 3 illustrates that the accuracy of susceptibility category results was generally high, but some resistance mechanisms were unrecognized due to various factors, mainly due to delayed implementation of modified/revised (CLSI, 2012c) breakpoints. Also, flaws in “expert” software systems found in some commercial products were discovered. However, only 6 pathogen/method accuracy rates fell below 90%, if these participant results were graded against a reference category. Finally, the most serious susceptibility testing interpretation and method detail concerns noted by the API ES Program samples were as follows: 1) ES-01 (2007) had significant numbers of laboratories not applying ESBL screening concentrations of ≥2 μg/mL for aztreonam or ceftriaxone or ceftazidime for the SHV-5–producing E. coli strain. 2) ES-02 (2007) provided educational content to update methods to recognize MRSA by oxacillin and cefoxitin disks, and the falsesusceptible rates were 2.0–6.7% for this strain and other β-lactams were often not reported as resistant. Inappropriate agents active only against urinary tract infection (UTI) isolates were reported (nitrofurantoin as an example). 3) ES-03 (2007) challenged laboratories with a KPC-3–producing serine carbapenemase in a K. pneumoniae that was found susceptible to imipenem (33.1–40.0%) and other carbapenems. This was due to yet to be published, modified (lower) breakpoints for marketed carbapenems; see ES-02 (2010). False-susceptible results were common during this time period. 4) ES-01 (2008) was a MDR S. peroris strain that had susceptibilities accurately assessed, but some laboratories reported disk diffusion susceptibility categories where no criteria were published in CLSI (2012c) tables for viridans group streptococci. 5) ES-02 (2008) was a K. oxytoca strain having an ESBLTable 2 Organism identification accuracy for API ES-series challenges for 2007–2011. Organism and percentage acceptable performance Gram-negative species ES-01 (2007) E. coli at 99.2% ES-03 (2007) K. pneumoniae at 94.0% ES-02 (2008) K. oxytoca at 91.0% ES-01 (2009) P. aeruginosa at 96.6% ES-03 (2009) A. baumannii at 94.4% ES-02 (2010) S. marcescens at 96.7% ES-03 (2011) E. coli at 99.2% Gram-positive species ES-02 (2007) S. aureus at 100.0% ES-01 (2008) S. mitis/oralis or S. peroris at 95.3% ES-03 (2008) S. haemolyticus at 95.7% ES-02 (2009) S. pneumoniae at 94.2% ES-01 (2010) E. faecalis at 97.8% ES-03 (2010) S. aureus at 99.1% ES-01 (2011) S. agalactiae at 98.9% ES-02 (2011) E. gallinarum at 56.8%a
g
g
95.9%
97.3%a
a Overall acceptable rate did not include the potentially ungraded sample ES-02 (2011).
359
Table 3 Categorical accuracy of disk diffusion (DD) and various MIC methods to determine the susceptibility of 15 API ES challenge strains (2007–2011). Year
Sample
Target susceptibility education topica
Categorical accuracy by method
2007
ES-01 ES-02 ES-03
96.9/92.7 93.1/92.3 92.9/89.5
2008
ES-01
ESBL (SHV-5) detection in E. coli MRSA (USA300-0114) Carbapenemases (KPC-3) in Enterobacteriaceae High-level penicillin resistance in streptococci (S. peroris) ESBL-like enzyme (OXY-2a) in K. oxytoca MDR S. haemolyticus MDR P. aeruginosa Control S. pneumoniae to assess QC MDR A. baumannii (OXA-23) MDR E. faecalis, linezolid-resistant MDR S. marcescens (KPC-3) MRSA (USA100 hospital strain) MDR S. agalactiae, fluoroquinoloneresistant Vancomycin-resistant E. gallinarum MDR E. coli from UTI
MIC/Disk Diffusion
2009
2010
2011
ES-02 ES-03 ES-01 ES-02 ES-03 ES-01 ES-02 ES-03 ES-01 ES-02 ES-03
a b c d
96.6/95.0 87.7/86.9 93.2/98.0 81.8b/96.9 99.4/98.9 94.6/97.1 97.3/98.8 85.5c/83.2c 97.1/99.3 96.2/97.6 88.9d/97.3 97.4/96.5
ESBL = Extended spectrum β-lactamase; MDR = multidrug-resistant. Driven to low rate by poor ceftazidime performance (only 23.3% accuracy). Slow adoption of current CLSI breakpoint criteria for ESBL and KPC-producing strains. Most influenced by poor fluoroquinolone test accuracy.
like β-lactamase (OXY-2a) and MIC results at ≥2 μg/mL for aztreonam and ceftriaxone, e.g., ESBL screen-positive phenotype. All methods and automated systems (even those with “expert” software programs) failed to recognize significant resistances per CLSI criteria. 6) ES-01 (2010), an E. faecalis with oxazolidinone target site mutational resistance, was noted to be accurately found as resistant (MIC, N8 μg/ mL), but less than 50% of participants tested linezolid. Moreover, this enterococcus was wrongly reported as ampicillin-nonsusceptible by 1.8–5.9% of participants. 7) ES-02 (2010) was the second Enterobacteriaceae (MDR S. marcescens) among ES challenges having a KPC-3 carbapenemase. Only 3 drugs were observed to be active (amikacin, gentamicin, tigecycline), and poor performance was documented with false-susceptible rates for carbapenems due to nonapplication of CLSI screening criteria for serine carbapenemases (CLSI M100-S20 U). Unacceptable numbers of laboratories continue to report falsesusceptible results (20.4–90.9%) for carbapenems and other broadspectrum β-lactam agents. And 8) ES-02 (2011) was an E. gallinarum with 2 glycopeptide resistance genes (vanC1 and vanA) that was generally misidentified to the species level; however, the key resistance was detected well by MIC methods (97.3%), but not by disk diffusion (88.9%). Overall, the quality and accuracy of antimicrobial susceptibility testing appear acceptable among the laboratories participating in the API ES Survey (2007–2011). However, the detection of clinically important emerging resistance mechanisms has been consistently compromised by the slow application of the modified, updated antimicrobial breakpoint criteria of the CLSI (2012c) or EUCAST (2012), or by the delayed adaption of the same breakpoints by commercial products or by reporting inappropriate agents for some infection sites, e.g., UTI agents for systemic infections. Clinical microbiology laboratory management and staff should keep apprised of evolving resistance patterns especially those mediated by βlactamases in Gram (−) bacilli (Ambler et al., 1991; Bonnet, 2004; Bush et al., 1995; Gales et al., 2001; Nordmann and Poirel, 2002; Smith-Moland et al., 2003) and of the contemporary challenges to the complete coverage of Gram (+) cocci by glycopeptides (vancomycin), lipoglycopeptides (daptomycin), and oxazolidinones such as linezolid (Mutnick et al., 2003). Regular reports from the API Program are
360
R.N. Jones et al. / Diagnostic Microbiology and Infectious Disease 75 (2013) 357–360
scheduled for publication to supplement participant critiques and to extend these summary analyses. References Ambler RP, Coulson AF, Frere JM, Ghuysen JM, Joris B, Forsman M, et al. A standard numbering scheme for the class A beta-lactamases. Biochem J 1991;276:269–70. Bonnet R. Growing group of extended-spectrum beta-lactamases: the CTX-M enzymes. Antimicrob Agents Chemother 2004;48:1-14. Bush K, Jacoby GA, Medeiros AA. A functional classification scheme for beta-lactamases and its correlation with molecular structure. Antimicrob Agents Chemother 1995;39:1211–33. Clinical and Laboratory Standards Institute (CLSI). Performance standards for antimicrobial disk susceptibility tests; approved standard: eleventh edition, M02-A11. Wayne, PA: CLSI; 2012a. Clinical and Laboratory Standards Institute (CLSI). Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; approved standard: ninth edition, M07-A9. Wayne, PA: CLSI; 2012b. Clinical and Laboratory Standards Institute (CLSI). Performance standards for antimicrobial susceptibility testing: 22nd informational supplement, M100-S22. Wayne, PA: CLSI; 2012c. Doern GV, Brueggemann AB, Pfaller MA, Jones RN. Assessment of laboratory performance with Streptococcus pneumoniae antimicrobial susceptibility testing in the United States: a report from the College of American Pathologists Microbiology Proficiency Survey Program. Arch Pathol Lab Med 1999;123:285–9. Edson DC, Dyke JW, Glick T, Massey LD. Identification of Escherichia coli O157:H7 in a proficiency testing program. Lab Med 2005;36:229–31. EUCAST Breakpoint tables for interpretation of MICs and zone diameters. Version 2.0, January 2012. Available at http://www.eucast.org/clinical_breakpoints/. Accessed January 1, 2012. Gales AC, Jones RN, Forward KR, Linares J, Sader HS, Verhoef J. Emerging importance of multidrug-resistant Acinetobacter species and Stenotrophomonas maltophilia as pathogens in seriously ill patients: geographic patterns, epidemiological features, and trends in the SENTRY Antimicrobial Surveillance Program (1997–1999). Clin Infect Dis 2001;32(Suppl 2):S104–13. Gavan TL. A summary of the bacteriology portion of the 1972 basic, comprehensive, and special, College of American Pathologists (CAP) Quality Evaluation Program. Am J Clin Pathol 1974;61:971–9. Gavan TL, King JW. An evaluation of the microbiology portions of the 1969 basic, comprehensive, and special College of American Pathologists proficiency testing surveys. Am J Clin Pathol 1970;54:514–20. Jones RN. Antimicrobial susceptibility testing (AST): a review of changing trends, quality control guidelines, test accuracy, and recommendation for the testing of beta-lactam drugs. Diagn Microbiol Infect Dis 1983;1:1-24.
Jones RN. Recent trends in the College of America Pathologists proficiency results for antimicrobial susceptibility testing: Preparing for CLIA '88. Clin Microbiol Newsl 1992;14:33–7. Jones RN. Methods preferences and test accuracy of antimicrobial susceptibility testing: updates from the College of American Pathologists and Microbiology Survey Program (2000). Arch Pathol Lab Med 2001;125:1285–9. Jones RN, Edson DC. The ability of participant laboratories to detect penicillin-resistant pneumococci. A report from the microbiology portion of the College of American Pathologists (CAP) surveys. Am J Clin Pathol 1982a;78:659–63. Jones RN, Edson DC. Interlaboratory performance of disk agar diffusion and dilution antimicrobial susceptibility tests, 1979–1981. A summary of the microbiology portion of the College of American Pathologists (CAP) surveys. Am J Clin Pathol 1982b;78:651–8. Jones RN, Edson DC. Special topics in antimicrobial susceptibility testing: test accuracy against methicillin-resistant Staphylococcus aureus, pneumococci, and the sensitivity of beta-lactamase methods. Am J Clin Pathol 1983;80(Suppl 4):609–14. Jones RN, Edson DC. Antibiotic susceptibility testing accuracy. Review of the College of American Pathologists Microbiology survey, 1972–1983. Arch Pathol Lab Med 1985;109:595–601. Jones RN, Edson DC. The identification and antimicrobial susceptibility testing of Neisseria gonorrhoeae, 1980–1987. Results from the College of American Pathologists Microbiology Surveys Program. Arch Pathol Lab Med 1988;112: 485–8. Jones RN, Edson DC. Antimicrobial susceptibility testing trends and accuracy in the United States. A review of the College of American Pathologists Microbiology Surveys, 1972–1989. Arch Pathol Lab Med 1991;115:429–36. Jones RN, Edson DC, Marymont JV. Evaluations of antimicrobial susceptibility test proficiency by the College of American Pathologists Survey program. A clarification of quality control recommendations. Am J Clin Pathol 1982;78:168–72. Jones RN, Fritsche TR. Accuracy for antimicrobial and antifungal susceptibility testing: report from the College of American Pathologists Microbiology Proficiency Survey Program for 2004–2005. In: 45th ICAAC, December 16–19, 2005. Abstr. D-1654. Washington, DC, USA. 2005. Jones RN, Sommers HM. Identification and antimicrobial susceptibility testing of Branhamella catarrhalis in United States laboratories, 1983–1985. Drugs 1986;31(Suppl 3):34–7. Mutnick AH, Enne V, Jones RN. Linezolid resistance since 2001: SENTRY Antimicrobial Surveillance Program. Ann Pharmacother 2003;37:769–74. Nordmann P, Poirel L. Emerging carbapenemases in Gram-negative aerobes. Clin Microbiol Infect 2002;8:321–31. Pfaller MA, Jones RN. Performance accuracy of antibacterial and antifungal susceptibility test methods: report from the College of American Pathologists Microbiology Surveys Program (2001–2003). Arch Pathol Lab Med 2006;130:767–78. Smith-Moland E, Hanson ND, Herrera VL, Black JA, Lockhart TJ, Hossain A, et al. Plasmidmediated, carbapenem-hydrolysing beta-lactamase, KPC-2, in Klebsiella pneumoniae isolates. J Antimicrob Chemother 2003;51:711–4.
Contents lists available at SciVerse ScienceDirect
Diagnostic Microbiology and Infectious Disease journal homepage: www.elsevier.com/locate/diagmicrobio
Educational antimicrobial susceptibility testing as a critical component of microbiology laboratory proficiency programs: American Proficiency Institute results for 2007–2011 Ronald N. Jones a,⁎, Teri Glick b, Helio S. Sader a, Robert K. Flamm a, James E. Ross a, Paul R. Rhomberg a, Daniel C. Edson b a b
JMI Laboratories, North Liberty, IA, USA American Proficiency Institute, Traverse City, MI, USA
a r t i c l e
i n f o
Article history: Received 9 December 2012 Received in revised form 10 January 2013 Accepted 31 January 2013 Keywords: API Proficiency programs Microbiology Susceptibility tests Accuracy Quality assurance
a b s t r a c t External laboratory proficiency programs are an important requirement for test quality assurance (EQA) and compliance to regulatory guidelines (Clinical Laboratory Improvement Amendments and inspections). The American Proficiency Institute (API) regularly distributes EQA sample challenges (test events) including an Educational Sample (ES) for antimicrobial susceptibility testing. Beginning in 2007, API has sent 3 ES samples annually, each a well-characterized (molecular/phenotypic methods) strain having an interesting/emerging mechanism of resistance. Hundreds of USA laboratories, usually serving small- to medium-size hospitals and clinics, participate in the API ungraded ES test event. Analysis of responses is made and reported electronically as ES critiques addressing contemporary susceptibility testing issues that affect patient therapy. Seven Grampositive (+) and 8 Gram-negative (−) ES strains were tested over the 5 years (2007–2011) with organism identification (graded) accuracy of 95.3% (range, 91.0–99.2%) for Gram (−) and 97.0% (range, 94.2–100.0%) for Gram (+) challenges. Susceptibility testing categorical accuracy was generally greatest for the disk diffusion test (91.0/97.0%) compared to the MIC methods (commercial automated or manual) combined (89.9/96.1%, for Gram [−]/Gram [+], respectively). The most worrisome observations of these ES samples were as follows: 1) poor recognition of ESBL- and serine carbapenemase-producing strains (various types including Klebsiella pneumoniae carbapanemase) due to delayed application of Clinical and Laboratory Standards Institute [CLSI] guidelines; 2) overcalling of ESBL in organisms having wild-type non-ESBL enzymes (OXA series; OXA, 1/30) due to commercial system or participant interpretive error; and 3) occasional drugbug discords noted in nonfermentative Gram (−) bacilli. In conclusion, the API ES series of ungraded susceptibility testing challenges (accuracy was N90%) has been well received by subscribers and has provided detailed educational opportunities to improve laboratory testing performance. ES samples have delivered guidance to enable laboratories to rapidly comply with CLSI document changes of interpretive breakpoints such as those for β-lactams when testing Enterobacteriaceae and Pseudomonas aeruginosa; the program was sustained into 2012 and beyond to document quality of susceptibility tests in USA laboratories. © 2013 Elsevier Inc. All rights reserved.
1. Introduction External quality assurance (EQA) programs are important components of laboratory practice that are essential for maintaining clinical laboratory test accuracy. These EQA practices complement inspection and accreditation organizations mandated via professional societies (e.g., College of American Pathologists [CAP]) and the USA government (Clinical Laboratory Improvement Amendments [CLIA]). These USA programs have assessed the implementation and accuracy of antimicrobial susceptibility testing methods over 4 decades with published summaries dating from 1969 (Gavan, 1974; Gavan and ⁎ Corresponding author. Tel.: +1-319-665-3370; fax: +1-319-665-3371. E-mail address: [email protected] (R.N. Jones). 0732-8893/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.diagmicrobio.2013.01.027
King, 1970). The initial detailed publications, however, appeared beginning in 1982 covering the CAP Microbiology EQA surveys (Jones and Edson, 1982a; 1982b; Jones et al., 1982) and the most recent summarized testing events through 2005 (Jones and Fritsche, 2005; Pfaller and Jones, 2006). Special reports addressing quality control (QC) practices and specific test accuracy were also published (Jones, 1983; Jones and Edson, 1983). Interim analyses were reported by Doern et al. (1999) as well as by Jones and colleagues (Jones, 1992, 2001; Jones and Edson, 1983, 1985; Jones and Edson, 1988, 1991; Jones and Sommers, 1986). The American Proficiency Institute (API) is a federally approved EQA proficiency testing provider (since 1991) serving over 17,000 hospitals, clinics, and physician offices/laboratories (Edson et al., 2005). The API initiated a specific educational microbiology objective
358
R.N. Jones et al. / Diagnostic Microbiology and Infectious Disease 75 (2013) 357–360
(2007) that was designed to monitor and enhance antimicrobial susceptibility testing quality. This Educational Sample (ES) program distributes 3 organisms each year for species identification and testing of antimicrobials (both ungraded) by the routine method applied by the participating laboratories. The susceptibility test results are collected as categories (susceptible, intermediate, or resistant), and responses are analysed as an educational process to improve laboratory quality/education. A total of 15 ES challenges were assessed from 2007 through 2011, generally revealing high levels of test accuracy, yet documenting distinct areas for test improvement; summaries of the API ES Microbiology EQA Program are presented here.
2. Materials and methods
2.2. Participants and analysis Over 800 USA laboratories, usually serving hospitals and clinics, participate in each microbiology ES test event (N90% from commercial MIC methods or systems). Analysis of responses was made and reported electronically within a few weeks of deadlines as ES critiques addressing contemporary susceptibility testing issues that affect the quality of contemporary infection therapy. Categorical determinations of the participant results were classified by Clinical and Laboratory Standards Institute (CLSI), European Committee on Antimicrobial Susceptibility Testing (EUCAST), and US Food and Drug Administration breakpoint criteria, where appropriate (CLSI, 2012c; EUCAST, 2012). Reference methods of the CLSI (2012a,b) were used to characterized EQA ES samples, as well as specific molecular tests/ sequencing of resistance mechanisms.
2.1. Program description 3. Results and discussion As a component of a comprehensive Microbiology Proficiency Sample Program that has 43 distinct microbiology options, API offers the ES antimicrobial susceptibility testing samples (ungraded) every 4 months (e.g., 3 samples per year). This program was launched in early 2007 to circulate an organism sample, each a well-characterized (molecular/ phenotypic level methods) strain having an interesting or an emerging mechanism of resistance for educational purposes (see Table 1).
Table 1 lists the topics of the ES challenges for Enterobacteriaceae (4 species) with SHV-5, KPC-3 (2 events), OXA-series, and OXY-series β-lactamases; for nonfermentative Gram-negative (−) bacilli with multidrug-resistant (MDR) patterns; for S. aureus of the USA300 and USA100 types; for other Gram-positive (+) cocci (viridans group streptococci, E. gallinarum, S. agalactiae, and S. haemolyticus) with
Table 1 API Educational Sample (ES) topics for susceptibility testing (2007–2011) that focused on emerging resistance patterns and susceptibility testing problems/breakpoints. Year/sample no.
Species
Resistance educational topic
2007 ES-01
E. coli
Extended spectrum β-lactamase (SHV-5), an ESBL enzyme producing elevated MIC results for ceftazidime and monobactams. Monitored accuracy of ESBL detection by CLSI screening methods. Methicillin-resistant (MRSA) community-acquired strain (USA-300-0114) typical of emerging clonal type. Monitored accuracy of MRSA detection by CLSI methods for key β-lactam, macrolide, and fluoroquinolone resistances. Carbapenem-resistant (KPC-3) and typical of emerging resistance type in northeast USA and worldwide. Monitored detection accuracy of this enzyme type via current CLSI methods.
ES-02
S. aureus
ES-03
K. pneumoniae
2008 ES-01
S. mitis/oralis (S. peroris)
ES-02
K. oxytoca
ES-03
S. haemolyticus
2009 ES-01
P. aeruginosa
ES-02
S. pneumoniae
ES-03
A. baumannii
2010 ES-01
E. faecalis
ES-02
S. marcescens
ES-03
S. aureus
2011 ES-01
S. agalactiae
ES-02
E. gallinarum
ES-03
E. coli
High-level penicillin (β-lactam) resistance in a viridans streptococcus species associated with other antimicrobial resistances to clindamycin, macrolides, quinupristin–dalfopristin, fluoroquinolones, aminoglycosides, rifampin, tetracycline, and trimethoprim–sulfamethoxazole (TMP-SMX). Monitored accuracy of testing fastidious species. ESBL-like enzyme (OXY-series) having resistances to penicillins ± clavulanate, or sulbactam, or tazobactam, cephalosporins, and aztreonam. Monitored for detection of enzyme more commonly found in this not-uncommon Klebsiella spp. Multidrug-resistant (MDR) coagulase-negative staphylococcus (CoNS) with high monitored MIC results to methicillin, other β-lactams, macrolides, fluoroquinolones, and TMP-SMX. MDR strain only susceptible to amikacin, gentamicin, tobramycin, and polymyxins (colistin, polymyxin B). Monitored ability to recognize the few remaining treatment antimicrobials. An international quality control (QC) strain used to determine the technical accuracy of CLSI and other test methods. Results were compared to published QC ranges of MICs and zone diameters. Only minor levels of β-lactam resistances were present in this well-known strain. Acinetobacter harboring an OXA-23 carbapenemase producing a MDR organism emerging in the USA and worldwide. Only a few agents (aminoglycosides, ceftazidime, polymyxins, tigecycline, and TMP/SMX) were clearly active with their identification accuracy assessed. MDR strain with well-defined oxazolidinone resistance (target site mutations) that has appeared in recent years. Monitored multiple methods to detect a rare but emerging resistance that could decrease the utility of an important antimicrobial class. MDR strain with a KPC-3 serine carbapenemase (see mechanism in ES-03, 2007) having susceptibility to a few drugs (amikacin, gentamicin, and tigecycline). Appraised the use of recently published CLSI breakpoints for carbapenems and confirmatory methods (Modified Hodge Test). Healthcare-associated strain (USA 100) with a characteristic MDR pattern but retaining susceptibility to at least 10 antimicrobials including parenteral and oral agents. Monitored testing accuracy for commonly tested agents and methods. Group B streptococcus strains selected for elevated penicillin MICs and having high-level resistance to fluoroquinolones, macrolides, tetracyclines, and clindamycin. Monitored ability to recognize unusual resistances in β-hemolytic streptococci. Unusual enterococcus having a vancomycin resistance (VanA) with an intrinsic vanC1 gene. Monitored for misidentifications that were noted/documented across various commercial systems, usually naming the isolates an E. faecium or E. faecalis (only challenge not graded for identification, 2007–2011.) MDR pattern (resistant to penicillin with or without β-lactamase inhibitors, fluoroquinolones, aminoglycosides, and tetracyclines) with OXA-1/30 and TEM-1 enyzmes. Some participants erroneously diagnosed this organism as an ESBL producer by application of so-called expert systems. Monitored all key resistances in a communityacquired UTI sample.
R.N. Jones et al. / Diagnostic Microbiology and Infectious Disease 75 (2013) 357–360
MDR patterns; and for the S. pneumoniae quality control strain (ATCC 49619) having a penicillin nonsusceptibility. These topics were selected to address educational objectives of 1) recognizing emerging antimicrobial resistance mechanisms, 2) highlighting recent interpretive breakpoint criteria changes, 3) illustrating intermethod or system errors as applied to resistance screening/detection, and 4) emphasizing the important clonal, usually hospital-based, pathogens having MDR patterns. The objectives were met in a prospective manner as guided by international surveillance program results from the Jones Microbiology Institute (JMI Laboratories, North Liberty, IA, USA). Participant laboratories performed very well at an acceptable level of organism identification (Table 2). Identification of the Gram (+) species strains (97.3% acceptable) was slightly greater than that achieved for Gram (−) species (95.9%), but these rates were comparable to reports from other EQA surveys reported in the most recent years (Jones and Fritsche, 2005; Pfaller and Jones, 2006). Table 3 illustrates that the accuracy of susceptibility category results was generally high, but some resistance mechanisms were unrecognized due to various factors, mainly due to delayed implementation of modified/revised (CLSI, 2012c) breakpoints. Also, flaws in “expert” software systems found in some commercial products were discovered. However, only 6 pathogen/method accuracy rates fell below 90%, if these participant results were graded against a reference category. Finally, the most serious susceptibility testing interpretation and method detail concerns noted by the API ES Program samples were as follows: 1) ES-01 (2007) had significant numbers of laboratories not applying ESBL screening concentrations of ≥2 μg/mL for aztreonam or ceftriaxone or ceftazidime for the SHV-5–producing E. coli strain. 2) ES-02 (2007) provided educational content to update methods to recognize MRSA by oxacillin and cefoxitin disks, and the falsesusceptible rates were 2.0–6.7% for this strain and other β-lactams were often not reported as resistant. Inappropriate agents active only against urinary tract infection (UTI) isolates were reported (nitrofurantoin as an example). 3) ES-03 (2007) challenged laboratories with a KPC-3–producing serine carbapenemase in a K. pneumoniae that was found susceptible to imipenem (33.1–40.0%) and other carbapenems. This was due to yet to be published, modified (lower) breakpoints for marketed carbapenems; see ES-02 (2010). False-susceptible results were common during this time period. 4) ES-01 (2008) was a MDR S. peroris strain that had susceptibilities accurately assessed, but some laboratories reported disk diffusion susceptibility categories where no criteria were published in CLSI (2012c) tables for viridans group streptococci. 5) ES-02 (2008) was a K. oxytoca strain having an ESBLTable 2 Organism identification accuracy for API ES-series challenges for 2007–2011. Organism and percentage acceptable performance Gram-negative species ES-01 (2007) E. coli at 99.2% ES-03 (2007) K. pneumoniae at 94.0% ES-02 (2008) K. oxytoca at 91.0% ES-01 (2009) P. aeruginosa at 96.6% ES-03 (2009) A. baumannii at 94.4% ES-02 (2010) S. marcescens at 96.7% ES-03 (2011) E. coli at 99.2% Gram-positive species ES-02 (2007) S. aureus at 100.0% ES-01 (2008) S. mitis/oralis or S. peroris at 95.3% ES-03 (2008) S. haemolyticus at 95.7% ES-02 (2009) S. pneumoniae at 94.2% ES-01 (2010) E. faecalis at 97.8% ES-03 (2010) S. aureus at 99.1% ES-01 (2011) S. agalactiae at 98.9% ES-02 (2011) E. gallinarum at 56.8%a
g
g
95.9%
97.3%a
a Overall acceptable rate did not include the potentially ungraded sample ES-02 (2011).
359
Table 3 Categorical accuracy of disk diffusion (DD) and various MIC methods to determine the susceptibility of 15 API ES challenge strains (2007–2011). Year
Sample
Target susceptibility education topica
Categorical accuracy by method
2007
ES-01 ES-02 ES-03
96.9/92.7 93.1/92.3 92.9/89.5
2008
ES-01
ESBL (SHV-5) detection in E. coli MRSA (USA300-0114) Carbapenemases (KPC-3) in Enterobacteriaceae High-level penicillin resistance in streptococci (S. peroris) ESBL-like enzyme (OXY-2a) in K. oxytoca MDR S. haemolyticus MDR P. aeruginosa Control S. pneumoniae to assess QC MDR A. baumannii (OXA-23) MDR E. faecalis, linezolid-resistant MDR S. marcescens (KPC-3) MRSA (USA100 hospital strain) MDR S. agalactiae, fluoroquinoloneresistant Vancomycin-resistant E. gallinarum MDR E. coli from UTI
MIC/Disk Diffusion
2009
2010
2011
ES-02 ES-03 ES-01 ES-02 ES-03 ES-01 ES-02 ES-03 ES-01 ES-02 ES-03
a b c d
96.6/95.0 87.7/86.9 93.2/98.0 81.8b/96.9 99.4/98.9 94.6/97.1 97.3/98.8 85.5c/83.2c 97.1/99.3 96.2/97.6 88.9d/97.3 97.4/96.5
ESBL = Extended spectrum β-lactamase; MDR = multidrug-resistant. Driven to low rate by poor ceftazidime performance (only 23.3% accuracy). Slow adoption of current CLSI breakpoint criteria for ESBL and KPC-producing strains. Most influenced by poor fluoroquinolone test accuracy.
like β-lactamase (OXY-2a) and MIC results at ≥2 μg/mL for aztreonam and ceftriaxone, e.g., ESBL screen-positive phenotype. All methods and automated systems (even those with “expert” software programs) failed to recognize significant resistances per CLSI criteria. 6) ES-01 (2010), an E. faecalis with oxazolidinone target site mutational resistance, was noted to be accurately found as resistant (MIC, N8 μg/ mL), but less than 50% of participants tested linezolid. Moreover, this enterococcus was wrongly reported as ampicillin-nonsusceptible by 1.8–5.9% of participants. 7) ES-02 (2010) was the second Enterobacteriaceae (MDR S. marcescens) among ES challenges having a KPC-3 carbapenemase. Only 3 drugs were observed to be active (amikacin, gentamicin, tigecycline), and poor performance was documented with false-susceptible rates for carbapenems due to nonapplication of CLSI screening criteria for serine carbapenemases (CLSI M100-S20 U). Unacceptable numbers of laboratories continue to report falsesusceptible results (20.4–90.9%) for carbapenems and other broadspectrum β-lactam agents. And 8) ES-02 (2011) was an E. gallinarum with 2 glycopeptide resistance genes (vanC1 and vanA) that was generally misidentified to the species level; however, the key resistance was detected well by MIC methods (97.3%), but not by disk diffusion (88.9%). Overall, the quality and accuracy of antimicrobial susceptibility testing appear acceptable among the laboratories participating in the API ES Survey (2007–2011). However, the detection of clinically important emerging resistance mechanisms has been consistently compromised by the slow application of the modified, updated antimicrobial breakpoint criteria of the CLSI (2012c) or EUCAST (2012), or by the delayed adaption of the same breakpoints by commercial products or by reporting inappropriate agents for some infection sites, e.g., UTI agents for systemic infections. Clinical microbiology laboratory management and staff should keep apprised of evolving resistance patterns especially those mediated by βlactamases in Gram (−) bacilli (Ambler et al., 1991; Bonnet, 2004; Bush et al., 1995; Gales et al., 2001; Nordmann and Poirel, 2002; Smith-Moland et al., 2003) and of the contemporary challenges to the complete coverage of Gram (+) cocci by glycopeptides (vancomycin), lipoglycopeptides (daptomycin), and oxazolidinones such as linezolid (Mutnick et al., 2003). Regular reports from the API Program are
360
R.N. Jones et al. / Diagnostic Microbiology and Infectious Disease 75 (2013) 357–360
scheduled for publication to supplement participant critiques and to extend these summary analyses. References Ambler RP, Coulson AF, Frere JM, Ghuysen JM, Joris B, Forsman M, et al. A standard numbering scheme for the class A beta-lactamases. Biochem J 1991;276:269–70. Bonnet R. Growing group of extended-spectrum beta-lactamases: the CTX-M enzymes. Antimicrob Agents Chemother 2004;48:1-14. Bush K, Jacoby GA, Medeiros AA. A functional classification scheme for beta-lactamases and its correlation with molecular structure. Antimicrob Agents Chemother 1995;39:1211–33. Clinical and Laboratory Standards Institute (CLSI). Performance standards for antimicrobial disk susceptibility tests; approved standard: eleventh edition, M02-A11. Wayne, PA: CLSI; 2012a. Clinical and Laboratory Standards Institute (CLSI). Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; approved standard: ninth edition, M07-A9. Wayne, PA: CLSI; 2012b. Clinical and Laboratory Standards Institute (CLSI). Performance standards for antimicrobial susceptibility testing: 22nd informational supplement, M100-S22. Wayne, PA: CLSI; 2012c. Doern GV, Brueggemann AB, Pfaller MA, Jones RN. Assessment of laboratory performance with Streptococcus pneumoniae antimicrobial susceptibility testing in the United States: a report from the College of American Pathologists Microbiology Proficiency Survey Program. Arch Pathol Lab Med 1999;123:285–9. Edson DC, Dyke JW, Glick T, Massey LD. Identification of Escherichia coli O157:H7 in a proficiency testing program. Lab Med 2005;36:229–31. EUCAST Breakpoint tables for interpretation of MICs and zone diameters. Version 2.0, January 2012. Available at http://www.eucast.org/clinical_breakpoints/. Accessed January 1, 2012. Gales AC, Jones RN, Forward KR, Linares J, Sader HS, Verhoef J. Emerging importance of multidrug-resistant Acinetobacter species and Stenotrophomonas maltophilia as pathogens in seriously ill patients: geographic patterns, epidemiological features, and trends in the SENTRY Antimicrobial Surveillance Program (1997–1999). Clin Infect Dis 2001;32(Suppl 2):S104–13. Gavan TL. A summary of the bacteriology portion of the 1972 basic, comprehensive, and special, College of American Pathologists (CAP) Quality Evaluation Program. Am J Clin Pathol 1974;61:971–9. Gavan TL, King JW. An evaluation of the microbiology portions of the 1969 basic, comprehensive, and special College of American Pathologists proficiency testing surveys. Am J Clin Pathol 1970;54:514–20. Jones RN. Antimicrobial susceptibility testing (AST): a review of changing trends, quality control guidelines, test accuracy, and recommendation for the testing of beta-lactam drugs. Diagn Microbiol Infect Dis 1983;1:1-24.
Jones RN. Recent trends in the College of America Pathologists proficiency results for antimicrobial susceptibility testing: Preparing for CLIA '88. Clin Microbiol Newsl 1992;14:33–7. Jones RN. Methods preferences and test accuracy of antimicrobial susceptibility testing: updates from the College of American Pathologists and Microbiology Survey Program (2000). Arch Pathol Lab Med 2001;125:1285–9. Jones RN, Edson DC. The ability of participant laboratories to detect penicillin-resistant pneumococci. A report from the microbiology portion of the College of American Pathologists (CAP) surveys. Am J Clin Pathol 1982a;78:659–63. Jones RN, Edson DC. Interlaboratory performance of disk agar diffusion and dilution antimicrobial susceptibility tests, 1979–1981. A summary of the microbiology portion of the College of American Pathologists (CAP) surveys. Am J Clin Pathol 1982b;78:651–8. Jones RN, Edson DC. Special topics in antimicrobial susceptibility testing: test accuracy against methicillin-resistant Staphylococcus aureus, pneumococci, and the sensitivity of beta-lactamase methods. Am J Clin Pathol 1983;80(Suppl 4):609–14. Jones RN, Edson DC. Antibiotic susceptibility testing accuracy. Review of the College of American Pathologists Microbiology survey, 1972–1983. Arch Pathol Lab Med 1985;109:595–601. Jones RN, Edson DC. The identification and antimicrobial susceptibility testing of Neisseria gonorrhoeae, 1980–1987. Results from the College of American Pathologists Microbiology Surveys Program. Arch Pathol Lab Med 1988;112: 485–8. Jones RN, Edson DC. Antimicrobial susceptibility testing trends and accuracy in the United States. A review of the College of American Pathologists Microbiology Surveys, 1972–1989. Arch Pathol Lab Med 1991;115:429–36. Jones RN, Edson DC, Marymont JV. Evaluations of antimicrobial susceptibility test proficiency by the College of American Pathologists Survey program. A clarification of quality control recommendations. Am J Clin Pathol 1982;78:168–72. Jones RN, Fritsche TR. Accuracy for antimicrobial and antifungal susceptibility testing: report from the College of American Pathologists Microbiology Proficiency Survey Program for 2004–2005. In: 45th ICAAC, December 16–19, 2005. Abstr. D-1654. Washington, DC, USA. 2005. Jones RN, Sommers HM. Identification and antimicrobial susceptibility testing of Branhamella catarrhalis in United States laboratories, 1983–1985. Drugs 1986;31(Suppl 3):34–7. Mutnick AH, Enne V, Jones RN. Linezolid resistance since 2001: SENTRY Antimicrobial Surveillance Program. Ann Pharmacother 2003;37:769–74. Nordmann P, Poirel L. Emerging carbapenemases in Gram-negative aerobes. Clin Microbiol Infect 2002;8:321–31. Pfaller MA, Jones RN. Performance accuracy of antibacterial and antifungal susceptibility test methods: report from the College of American Pathologists Microbiology Surveys Program (2001–2003). Arch Pathol Lab Med 2006;130:767–78. Smith-Moland E, Hanson ND, Herrera VL, Black JA, Lockhart TJ, Hossain A, et al. Plasmidmediated, carbapenem-hydrolysing beta-lactamase, KPC-2, in Klebsiella pneumoniae isolates. J Antimicrob Chemother 2003;51:711–4.