Transfer of the blaKPC genes encoding the Klebsiella pneumoniae carbapenemase (KPC) are increasingly responsible for emerging carbapenem resistance. The modified Hodge test (MHT) is recommended for the detection of KPC. We compared MHT with a real-time polymerase chain reaction (PCR) assay targeting common subtypes of blaKPC, using previously described forward and reverse primer sequences. The PCR product was detected using SYBR Green (Applied Biosystems, Foster City, CA) and confirmed by melt curve analysis. PCR was positive in 96% (52/54) of isolates that were MHT+, 90% (28/31) of MHT– isolates were PCR–, and the results were strongly correlated (P = .0001; Fisher exact test). The PCR assay is a sensitive, specific, and rapid test for detecting blaKPC genes. It could help optimize patient care by reducing the time taken to institute appropriate antimicrobial therapy and so help improve patient outcomes.
Carbapenems are highly efficacious drugs for treating infections with extended-spectrum β-lactamase–producing gram-negative bacteria.1 Previously, resistance to carbapenems has been rare; however, the emergence of transmissible carbapenem resistance is now a growing concern.1–9 An increasingly common mechanism of carbapenem resistance is the class-A, Klebsiella pneumoniae carbapenemase (KPC).2,8,10–13 KPCs have been reported in K pneumoniae and in Klebsiella oxytoca, Pseudomonas aeruginosa, Escherichia coli, Proteus mirabilis, Citrobacter freundii, Enterobacter spp, Serratia spp, and Salmonella spp.2,8,14–21 The blaKPC genes that encode KPCs are present on transferable plasmids and are flanked by transposable elements, thus allowing for the gene to move from plasmid to the bacterial chromosome and back.8,17,22,23 This potential to disseminate resistance to several classes of β-lactam antibiotics has been demonstrated in several reported outbreaks with high mortality rates.9,13,21,24,25 Laboratory detection of KPC in clinical isolates is, therefore, crucial to limit this spread.
Automated susceptibility testing methods do not reliably detect KPC-mediated resistance.7,15 In 2009, the Clinical and Laboratory Standards Institute (CLSI) guidelines (M100) recommended use of the modified Hodge test (MHT) to definitively identify KPC producers.26 Alternatively, the blaKPC genes have been detected by conventional and real-time polymerase chain reaction (PCR) methods.15,20,27–29 However, there are limitations to using PCR: (1) It is more technically challenging and prone to inhibition.30 (2) It is target-specific and may miss new variants of KPC owing to genetic mutation. Carbapenemase-like patterns of nonsusceptibility leading to false-positive MHT results have also been described in association with endemic CTX-M production and AmpC hyperproduction.31,32
In this study, we compared the detection of the blaKPC genes by an in-house real-time PCR assay with the detection of KPC by the current CLSI-recommended standard of practice,26 the MHT. We also tested isolates with discordant MHT and PCR results for the production of AmpC β-lactamase.
Materials and Methods
Selection of Test Isolates
Clinical isolates of Enterobacteriaceae that had been analyzed with the VITEK2 automated system (BioMérieux, Durham, NC) for antimicrobial susceptibility and were intermediate to resistant to imipenem or ertapenem were subsequently tested by using the MHT. Stocks of these isolates were deidentified and stored in tryptic soy broth with 20% glycerol at −20°C. These included isolates that tested MHT+ and MHT–. We selected 61 MHT+ isolates and 18 MHT– isolates at random from among the stored isolates to perform DNA extraction. A reference K pneumoniae strain (American Type Culture Collection, Manassas, VA [ATCC] BAA-1705) was selected as the positive control and a second reference K pneumoniae strain (ATCC 700603) was selected as the negative control for the PCR assay.
Detection of KPC by the MHT
The MHT was performed as previously described.26 Briefly, a 0.5 McFarland suspension of E coli, ATCC 25922, was used to prepare a lawn culture on a Mueller Hinton agar plate (Becton Dickinson, Sparks, MD), and a 10-μg ertapenem susceptibility disk (Becton Dickinson) was placed in the center of the test area. Test isolates were subcultured onto sheep blood agar plates (Becton Dickinson) to establish pure cultures. The isolate was then streaked in a straight line from the edge of the disk to the edge of the plate and was incubated overnight at 35°C in ambient air. After 16 to 24 hours of incubation, the plate was examined for a cloverleaf-shaped indentation at the intersection of the test organism and the E coli ATCC 25922 within the zone of inhibition. The presence of a cloverleaf-shaped indentation was considered MHT+.
DNA Extraction From Cultured Bacterial Colonies
Fresh, well-isolated test colonies grown on sheep blood agar plates following overnight incubation were used for DNA extraction using the Qiagen EZ1 Virus Mini Kit, version 2.0, on the Qiagen Biorobot automated DNA extraction platform (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. Briefly, a 0.5 McFarland-standard bacterial suspension was prepared in saline. A final elution volume of 90 μL containing bacterial DNA was extracted from 400 μL (2.4 × 108 colony-forming units) of the suspension and stored at −20°C before using them for PCR.
blaKPC Detection by Real-Time PCR
The forward primer sequence (5′-ATG TCA CTG TAT CGC CGT C-3′) and reverse primer sequence (5′-CTC AGT GCT CTA CAG AAA ACC-3′) used in this assay have been described by Tibbetts et al.20
The amplicons were detected using the intercalating nucleic acid dye SYBR Green (Invitrogen, Eugene, OR) and were confirmed by melt curve analysis. The blaKPC primers were diluted to a concentration of 25 pmol/μL. The final 25 μL of the PCR mixture contained 5 μL of the extracted DNA and 20 μL of PCR buffer composed of 2 μL each of the forward and reverse primers and 12 μL of the SYBR Green, nucleotides, and Taq polymerase solution (Invitrogen) and 4 μL of nuclease-free water (Ambion, Austin, TX). The PCR was performed under the following conditions: stage 1, 95°C for 5 minutes (“hot-start” step) with optics off; stage 2, 95°C for 25 seconds with optics off, 60°C for 30 seconds, 72°C for 30 seconds with optics on, for 30 cycles, followed by a melt curve analysis from 65°C to 95°C at 0.2°/second. A sample was considered positive by PCR if it crossed the preset threshold within 30 cycles and the melting temperature was within 0.1°C of the positive control as determined by melt curve analysis.
The AmpC disk test was performed in the manner previously described.33 Briefly, AmpC disks containing tris(hydroxymethyl)aminomethane (Tris)-EDTA (Becton Dickinson) were prepared in house by applying 20 μL of a 1:1 mixture of saline and 100 μL Tris-EDTA (Sigma-Aldrich, St Louis, MO) to sterile filter paper disks, allowing the disks to dry, and storing them at 2°C to 8°C. The surface of a Mueller Hinton agar plate (Becton Dickinson) was inoculated with a lawn of cefoxitin-susceptible E coli ATCC 25922 according to the standard disk diffusion method. Immediately before use, AmpC disks were rehydrated with 20 μL of saline. Colonies of K pneumoniae and Enterobacter cloacae that had differing test results by the MHT and PCR were applied to a disk. A 30-μL cefoxitin disk (Becton Dickinson) was placed on the inoculated surface of the Mueller Hinton agar. The inoculated AmpC disk was then placed almost touching the antibiotic disk with the inoculated disk face in contact with the agar surface. The plate was then inverted and incubated overnight at 35°C in ambient air. After incubation, plates were examined for an indentation or a flattening of the zone of inhibition, indicating enzymatic inactivation of cefoxitin (positive result), or the absence of a distortion, indicating no significant inactivation of cefoxitin (negative result).
Described blaKPC sequences are available online through the GenBank database (Accession No. AF297554; http://www.ncbi.nlm.nih.gov/Genbank/). The ABI Prism sequence detection system (Applied Biosystems, Foster City, CA) was used to amplify and detect the blaKPC amplicons (246 base pairs). We selected 5 positive amplicons and 2 negative samples for direct DNA sequencing using the BigDye Terminator 3.1 cycle sequencing kit (Applied Biosystems) according to the manufacturer’s protocols. Sequences were detected by the ABI 3100 Capillary Electrophoresis Instrument with Data Collection version 2.1 (Applied Biosystems), and analysis of the sequences was performed with ABI DNA Sequencing Analysis Software, version 5.1.1 (Applied Biosystems).
The specificity of the primers for the detection of blaKPC genes was evaluated by the BLAST program (National Center for Biotechnology Information; available at http://blast.ncbi.nlm.nih.gov/Blast.cgi). No matches to the primers’ sequences were found other than those for the blaKPC genes. Frozen stocks of 76 isolates were available for DNA extraction, including 55 K pneumoniae, 17 E cloacae, 2 E coli, and 2 Serratia marcescens. Of the K pneumoniae isolates selected, 48 were found to be resistant to carbapenems by the VITEK2 automated antibiotic susceptibility testing system, while 7 were found to be susceptible Table 1. In addition, there were 4 E cloacae and 1 S marcescens isolates that were found to be resistant by the VITEK2 system, while 13 E cloacae and 2 E coli isolates were found to be susceptible to carbapenems. In total, there were 61 MHT+ isolates and 15 MHT– isolates.
Initial MHT and PCR Results
The PCR assay was positive in 50 of the MHT+ test isolates (50/61 [82%]) and was negative in 13 MHT– isolates (13/15 [87%]). Eleven isolates that had been reported as MHT+ were negative by PCR, including 8 K pneumoniae and 3 E cloacae isolates Table 2. AmpC testing was performed on all 11 of these isolates. One isolate of S marcescens and 1 of E cloacae were negative by MHT and positive by PCR. The results of the MHT and PCR were strongly correlated (P = .0001; Fisher exact test). The threshold temperature for the melt curve analysis for the DNA sequences that were amplified by the PCR was between 91°C and 92°C. We did not encounter a single sample that was reportable as equivocal by PCR.
Repeated MHT, Repeated PCR, and AmpC Results
The MHT and PCR were repeated for the 11 isolates that had discrepant results. Following the repeated MHT, 9 previously MHT+ isolates were reidentified as MHT–, including 6 K pneumoniae and 3 E cloacae isolates. Two K pneumoniae isolates remained MHT+ in repeated testing and were subsequently detected as PCR+ in the repeated test. The AmpC test was positive in both of these K pneumoniae isolates.
AmpC was detected in 4 isolates, including 3 K pneumoniae and 1 E cloacaeTable 3. Of these, the results of 1 K pneumoniae isolate remained MHT+/PCR–. The results of 1 K pneumoniae isolate changed from MHT+/PCR– to MHT–/PCR+, while the results of the third K pneumoniae isolate changed from MHT+/PCR– to MHT+/PCR+. The results for the E cloacae isolate changed from MHT+/PCR– to MHT–/PCR–. AmpC was not detected in 7 organisms for which the results had changed from MHT+/PCR– to MHT–/PCR–, including 5 K pneumoniae and 2 E cloacae isolates.
Following this repeated testing, there were 51 isolates that were positive by the MHT and PCR, while 21 isolates were negative for both Table 4. Based on these results, the sensitivity of the PCR was 98% (51/52), the specificity was 88% (21/24), and the results of the MHT and PCR were strongly correlated (P = .0001; Fisher exact test).
By including only K pneumoniae and E cloacae in the analysis, there were 48 isolates that were positive by the MHT and PCR, while 21 isolates were negative for both. This yielded a sensitivity of 98% (48/49) and specificity of 91% (21/23) for the PCR, with the results of the MHT and PCR being strongly correlated (P = .0001; Fisher exact test). Furthermore, with only isolates of K pneumoniae included in the analysis, 46 (98%) of 47 isolates were positive by the MHT and PCR, and 7 (88%) of 8 isolates were negative by the MHT and PCR, with a strong correlation between the results of the MHT and PCR (P = .0001; Fisher exact test).
The DNA sequence of 5 randomly selected amplicons was confirmed to correspond to the blaKPC genes using a BLASTn analysis. The protein sequence encoded by all 5 was also confirmed to correspond to a part of the KPC enzyme structure using a BLASTp analysis. Two nonamplified DNA samples were negative.
The class-A carbapenemase enzyme, KPC, is an emerging mechanism of resistance to all classes of β-lactams.2,8,10–13 The laboratory detection of KPC among clinical isolates is crucial to limit its spread.9,13,21,24,25 Automated susceptibility testing methods have not been shown to reliably detect KPC-mediated resistance.7,15
In 2009, the CLSI guidelines (M100) recommended the MHT to definitively identify KPC producers.26 This is a phenotypic test to detect carbapenemase production as the cause of reduced susceptibility to carbapenems. The procedure was first described by Lee et al34 in 2001 and was shown to have 100% sensitivity and specificity for detecting KPC. This complex test requires skilled laboratory personnel to set up and trained operators to interpret the results. Consequently, the test suffers from subjectivity in interpretation of the results and might not be optimally sensitive or specific. Being a phenotypic test, it also has a lengthy turnaround time, with reportable results not available for up to 24 hours. Hence, the test is more useful as a confirmatory test rather than as a screening tool.
In accordance with the current CLSI guidelines, clinical samples of E cloacae, S marcescens, E coli, and P aeruginosa are not routinely tested for the presence of KPC by the MHT in our laboratory. The carbapenem resistance detected in these organisms has been considered to have different causes, such as loss of the porin channel combined with the presence of an AmpC β-lactamase.35,36 Carbapenemase-like patterns of nonsusceptibility leading to false-positive MHT results have also been described in association with endemic CTX-M production and AmpC hyperproduction.31,32
Among the 13 isolates that had initially tested positive by the MHT and negative by PCR in our study, 10 were negative by MHT on repeated testing, including 6 K pneumoniae, 3 E cloacae and 1 P aeruginosa. The 2 K pneumoniae isolates that remained MHT+ and had discrepant PCR results were both positive for AmpC. Two possibilities exist that may explain these 2 MHT+/KPC PCR– isolates. First, as reported by Schechner et al,30 KPC PCR could be falsely negative due to inhibitory substances in the reaction or technical inexperience. However, the second and most probable reason could be the presence of an up-regulated AmpC enzyme, extended-spectrum β-lactamase, primarily CTX-Ms with mutation as described by Carvalhaes et al32 and Pasteran et al.37
The blaKPC genes encoding the KPCs are present on plasmids.8,17,22,23 During the past few years, conventional and real-time PCR methods have been described to detect these genes.15,20,27–29 A PCR assay, such as ours, offers the advantage of faster genotyping and a shorter turnaround time (4 hours for the PCR, compared with >24 hours for the MHT). There is also less subjectivity in interpreting the results. However, the complexity of these tests requires skilled and trained operators as well. Because these assays are specific for the particular target sequence, they cannot be used to monitor the emergence of novel variants. Furthermore, the presence of the target blaKPC genes does not always correlate with the phenotypic expression of KPC, nor is it able to quantify the level of expression. These assays are not able to detect mechanisms of resistance apart from KPC.
During the preparation of this manuscript, new CLSI guidelines were published (CLSI, M100-S20U, 2010) with 2-fold lower minimum inhibitory concentrations (MICs) for each category (susceptible, intermediate, resistant) for the Enterobacteriaceae for meropenem, imipenem, and ertapenem.38 Furthermore, it is suggested that the use of these breakpoints will obviate the need for confirmatory testing on specimens if only for infection control purposes. However, Daikos et al39 have shown that K pneumoniae isolates with MICs to imipenem as low as 0.12 μg/mL still possess carbapenemases. Although these were metallo-β-lactamases rather than KPCs, they would have been detected using MHT. In addition, some of the current automated system platform configurations will make it difficult if not impossible to implement these new guidelines in the next year or two. In our laboratory, we have detected MHT+ isolates with MICs to imipenem as low as 1 to 2 μg/mL and confirmed as KPC with PCR. Despite these new recommendations, we will continue to confirm carbapenemase activity using MHT or PCR.
The PCR assay shows strong correlation with the results of the MHT while offering comparable sensitivity and specificity. These findings, combined with the more rapid turnaround time of PCR assays, suggest this assay might be more suitable as an initial screening test for detecting KPC-mediated carbapenem resistance. This rapid identification of KPC would help optimize selection and commencement of appropriate antibiotic therapy in a timely manner and so might be of value in improving clinical outcomes.
Upon completion of this activity you will be able to:
recognize emerging mechanisms for resistance to β-lactam antibiotics.
cite the current Clinical and Laboratory Standards Institute standard for laboratory detection of Klebsiella pneumoniae carbapenemase–mediated carbapenem resistance.
discuss the advantages and disadvantages of the currently recommended method for detection of carbapenem resistance.
The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module.
The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.
Questions appear on p 643. Exam is located at www.ascp.org/ajcpcme.