OUP user menu

Predominant characteristics of CTX-M-producing Klebsiella pneumoniae isolates from patients with lower respiratory tract infection in multiple medical centers in China

Shuchang An, Jichao Chen, Zhanwei Wang, Xiaorong Wang, Xixin Yan, Jihong Li, Yusheng Chen, Qi Wang, Xiaoling Xu, Jiabin Li, Jingping Yang, Hui Wang, Zhancheng Gao
DOI: http://dx.doi.org/10.1111/j.1574-6968.2012.02586.x 137-145 First published online: 1 July 2012


From February 2010 to July 2011, 183 of 416 presumptive Klebsiella pneumoniae isolates with reduced susceptibility to third-generation cephalosporins from patients with lower respiratory tract infection were collected from seven tertiary hospitals in China. Phenotypic and genotypic methods were employed to characterize 158 extended-spectrum β-lactamase (ESBL)-producers. Among the 158 isolates analyzed, 134 (84.8%) harbored blaCTX-M, within which the most predominant ESBL gene was CTX-M-14 (49.4%), followed by CTX-M-15 (12.0%) and CTX-M-27 (10.8%). Also, 120 (75.9%) harbored blaSHV. One novel SHV variant, blaSHV-142 with T18A and L35Q substitutions, was identified. Ninety-one isolates carried blaTEM-1. An isolate containing blaTEM-135 was first identified in Klebsiella spp. blaKPC-2 was detected in 5 isolates. More than one ESBL combination was detected in 18 isolates (11.4%). Fifty-four (34.2%) isolates demonstrated the multidrug resistant (MDR) phenotype. Seventy-four sequence types (STs) were identified, which showed large genetic background diversity in ESBL-producing K. pneumoniae isolates from the six areas. This is the first report on the high prevalence of CTX-M-27 in China with the possible transmission of a single clone (ST48). The correlated surveillance of organisms with MDR phenotype should be investigated in future.

  • extended-spectrum β-lactamase
  • CTX-M
  • Klebsiella pneumoniae
  • multilocus sequence typing


Extended-spectrum B-lactamase (ESBL)-producing Enterobacteriaceae, especially Klebsiella pneumoniae and Escherichia coli, have been shown to have a significant impact on treatment options and clinical outcome in inpatients and outpatients (Tumbarello et al., 2007; Meier et al., 2011). Further, ESBL-producing bacteria have been shown to cause higher morbidity, mortality, and fiscal burden (Jean & Hsueh, 2011; Dhillon & Clark, 2012). The typical characteristic of ESBLs is their ability to hydrolyze oxyiminocephalosporins and aztreonam while being inhibited by β-lactamase inhibitors (Paterson & Bonomo, 2005). As the first types of ESBL derived from the non-ESBL blaSHV-1 and blaTEM-1 were reported, CTX-M-type ESBLs are now actually the most frequent types worldwide and are clustered in five subgroups (Falagas & Karageorgopoulos, 2009). Furthermore, some ESBLs exhibiting inhibitor resistance properties have also been identified in gram-negative bacteria (Nüesch-Inderbinen et al., 1997). So far, there are 124 CTX-M variants, 143 SHV variants, and 196 TEM variants, and many other types of ESBLs have been reported worldwide (http://www.lahey.org/studies).

The prevalence of ESBL-producing bacteria and their antimicrobial resistance profiles vary worldwide (Dhillon & Clark, 2012). Further, the widespread use of broad-spectrum antibacterial agents is an independent risk factor for infection with ESBL-producers (Pfaller & Segreti, 2006). Because of various antibiotic prescription patterns in different regions and increasing internal travel and trade in China, continuous surveillance studies and epidemiologic data on the prevalence of genotypes of ESBLs in different areas are of great needs. To date, the predominant ESBLs in Enterobacteriaceae are CTX-M- and SHV-type, with other ESBL enzymes were less often encountered (Chanawong et al., 2002; Yu et al., 2007; Liu et al., 2009; Zhang et al., 2009). The aim of this investigation was to clarify the current phenotypes, genotypes, and the genetic characteristics of blaCTX-M/SHV/TEM-producing K. pneumoniae isolates originating from patients with lower respiratory tract infection in seven tertiary hospitals in China.

Materials and methods

Clinical strains

From February 2010 to July 2011, 416 consecutive nonduplicate clinical K. pneumoniae isolates were collected from seven tertiary hospitals in Beijing Xicheng District (n = 109), Beijing Haidian District (n = 45), Fujian Province (n = 71), Anhui Province (n = 64), Hebei Province (n = 52), Liaoning Province (n = 40), and Inner Mongolia Autonomous Region (n = 35) in China. The lower respiratory tract infection was defined as described elsewhere (Li et al., 2011). Species identification was initially carried out by each of the hospital microbiological laboratories using their own protocols. The presumptive ESBL phenotype was screened by reduced susceptibility to ceftriaxone, cefotaxime, and aztreonam with automated systems or the disk diffusion methods using the Clinical and Laboratory Standards Institute (CLSI) criteria (Clinical & Laboratory Standards Institute, 2010). Upon arrival at the referral laboratory, the identification of all isolates was confirmed by sequencing analysis of the rpoB gene coding for the β-subunit of K. pneumonia RNA polymerase (Diancourt et al., 2005). The patients' clinical data such as demographics (age, sex) and the hospital units where they had received medical service were also reviewed. This study was approved by Peking University People's Hospital Ethics Committee (Federal-wide Assurance 00001384).

Confirmation of ESBL production and antimicrobial susceptibility testing

All presumptive ESBL-producing isolates were subjected to the confirmation test for ESBL production by the double-disk synergy test (Clinical & Laboratory Standards Institute, 2010). Minimum inhibitory concentrations (MICs) to 21 antimicrobial agents (ampicillin, ampicillin/sulbactam, piperacillin, piperacillin/tazobactam, cefazolin, cefuroxime, cefuroxime axetil, ceftriaxone, ceftazidime, cefepime, cefotetan, aztreonam, imipenem, meropenem, amikacin, gentamicin, tobramycin, ciprofloxacin, levofloxacin, nitrofurantoin, and trimethoprim-sulfamethoxazole) were performed using the VITEK 2 system (bioMe′rieux, France) with the AST-GN09 card. The susceptibility to cefotaxime refers to the confirmation test. All results were interpreted based on the CLSI guidelines (Clinical & Laboratory Standards Institute, 2010). Multidrug resistant (MDR) phenotype was determined as described previously (Tumbarello et al., 2007). The disks used for confirmation test were obtained from Beijing Tiantan Biological Products Corporation (China). Escherichia coli (ATCC 25922) and K. pneumoniae (ATCC 700603) were used as quality control strains.

PCR amplification of β-lactamase genes and sequence analysis

Plasmid DNA was extracted and purified by the alkaline lysis method using a commercial plasmid DNA purification kit (Tiangen Biotech Co., Ltd, China). Detection of genes encoding blaSHV/TEM/CTX-M groupI/CTX-M groupIV enzymes was performed by PCR with the primers listed in Table S1, Supporting information. The specific PCR assay of CTX-M group II, III, and V was implemented with the relevant primers (Nagano et al., 2003; Pitout et al., 2004; Chmelnitsky et al., 2005). Further amplification for K. pneumoniae carbapenemase (KPC) in speculative isolates (MIC of imipenem or meropenem of ≥ 2 µg mL−1), another pair of primers was used (Yigit et al., 2001). PCR products were subjected to bidirectional nucleotide sequencing using an automated DNA sequencer (ABI 3730XL, Weiterstadt, Germany). The nucleotide sequences or deduced protein sequences were analyzed via both Basic Local Alignment Search Tool (blast) program (http://www.ncbi.nlm.nih.gov/BLAST) and web site on the nomenclature of ESBLs (http://www.lahey.org/studies).

Multilocus sequence typing (MLST)

The genetic relationship between all qualified K. pneumoniae isolates was determined by MLST with seven housekeeping genes (Diancourt et al., 2005). Chromosomal DNA was obtained by the alkaline lysis method using a commercial genomic DNA purification kit (Tiangen Biotech Co., Ltd) according to the manufacturer's instructions. Allele sequences and sequence types (STs) were verified at the http://www.pasteur.fr/recherche/genopole/PF8/mlst/Kpneumoniae.html web site. The phylogenetic relationships among the different STs were established according to a dendrogram generated using the unweighted pair group method with arithmetic mean (UPGMA) algorithms and eBURST analysis (http://pubmlst.org/perl/mlstanalyse).

Statistical analysis

Categorical variables were evaluated by the chi-square test. Values are expressed as percentages of the group from which they were derived (categorical variables). Two-tailed tests were used to determine statistical significance. P value of < 0.05 was considered significant. All statistical analysis was performed by the spss statistics program, version 16.0, for Windows (SPSS Inc., IBM).


Identification of the K. pneumoniae isolates

In total, 183 ESBL-producers were screened. Of the 183 study isolates, seven isolates were negative for ESBL production by the double-disk synergy test and 18 isolates were excluded by the rpoB confirmatory test. Thus, 158 K. pneumoniae was included for further characterization. The frequency of occurrence of ESBL-producers in the six areas ranges from 28.8% to 64.4% (Fig. S1).

Clinical characteristics of the K. pneumoniae isolates

Complete medical records were available for review from 133 of 158 patients. As shown in Table S2, 107 (80.5%) isolates were collected from the general wards except for 26 (19.5%) of which were collected from the intensive care units (ICU). Most of the patients (84/133) were over 60 years old and were predominantly male (90 males vs. 43 females). Ninety percent isolates were collected more than 48 h after hospitalization.

Antibiotic susceptibility tests

All isolates were resistant to ampicillin, cefazolin (MICs ≥ 64 µg mL−1), and manifested 100% resistance to ceftriaxone (MIC range 8–≥ 64 µg mL−1) (Table 1). The resistance rates to drugs with lower overall resistance rate were 26.6%, 22.2%, 10.1%, 8.2%, and 3.8%, to amikacin, cefepime, piperacillin/tazobactam, cefotetan, and imipenem, respectively. All isolates were resistant to cefotaxime with the zone diameters of ≤ 22 mm except for one of 24 mm. A total of 54 of the 158 isolates (34.2%) were classified as MDR (Table 2).

View this table:
Table 1

Antimicrobial resistances in 158 ESBL-producing Klebsiella pneumoniae isolates from seven tertiary hospitals in China

No. of isolates (%)
blaCTX-M116 (73.4)116 (100)91 (78.4)106 (91.4)9 (7.8)116 (100)116 (100)116 (100)9 (7.8)116 (100)43 (37.1)
blaSHV15 (9.5)15 (100)15 (100)15 (100)2 (13.3)15 (100)12 (80)12 (80)0 (0)15 (100)13 (86.7)
blaCTX-M14+CTX-M153 (1.9)3 (100)3 (100)3 (100)0 (0)3 (100)3 (100)3 (100)0 (0)3 (100)2 (66.7)
blaCTX-M + blaSHV15 (9.5)15 (100)14 (93.3)14 (93.3)4 (26.7)15 (100)15 (100)15 (100)1 (6.7)15 (100)12 (80)
Not detect genotype9 (5.7)9 (100)6 (66.7)7 (77.8)1 (11.1)9 (100)7 (77.8)7 (77.8)3 (33.3)9 (100)4 (44.4)
Total (%)158 (100)158 (100)129 (81.6)145 (91.8)16 (10.1)158 (100)153 (96.8)153 (96.8)13 (8.2)158 (100)74 (46.8)
blaCTX-M116 (73.4)20 (17.2)56 (48.3)4 (3.4)25 (21.6)81 (69.8)35 (30.2)55 (47.4)45 (38.8)66 (56.9)78 (67.2)
blaSHV15 (9.5)2 (13.3)11 (73.3)0 (0)7 (46.7)10 (66.7)10 (66.7)9 (60)9 (60)9 (60)14 (93.3)
blaCTX-M14+CTX-M153 (1.9)3 (100)3 (100)0 (0)0 (0)2 (66.7)0 (0)1 (33.3)1 (33.3)1 (33.3)3 (100)
blaCTX-M + blaSHV15 (9.5)9 (60)12 (80)1 (6.7)7 (46.7)15 (100)7 (46.7)11 (73.3)11 (73.3)6 (54.5)9 (60)
Not detect genotype9 (5.7)1 (11.1)6 (66.7)1 (11.1)3 (33.3)8 (88.9)3 (33.3)4 (44.4)4 (44.4)7 (77.8)8 (88.9)
Total (%)158 (100)35 (22.2)88 (55.7)6 (3.8)42 (26.6)116 (73.4)55 (34.8)80 (50.6)70 (44.3)89 (56.3)112 (70.9)
  • AMP, ampicillin; SAM, ampicillin/sulbactam; PIP, piperacillin; TZP, piperacillin/tazobactam; CFZ, cefazolin; CXM, cefuroxime; FUR, cefuroxime axetil; CRO, ceftriaxone; CAZ, ceftazidime; FEP, cefepime; CTT, cefotetan; ATM, aztreonam; IMP, imipenem; AMK, amikacin; GEN, gentamicin; TOB, tobramycin; CIP, ciprofloxacin; LEV, levofloxacin; FT, nitrofurantoin; SXT, trimethoprim-sulfamethoxazole. The antimicrobial resistances to meropenem were similar to imipenem.

  • Three isolates harboring bla KPC-2 were included.

  • One isolate harboring bla KPC-2 was included.

View this table:
Table 2

Antimicrobial resistances in representative blaCTX-M-harboring Klebsiella pneumoniae isolates

No. of isolates (MIC range µg mL−1)
Antimicrobial agentsNo other bla (n = 11)With other non-ESBL bla (n = 54)With other ESBL bla (n = 9)No other bla (n = 0)With other non- ESBL bla (n = 12)With other ESBL bla (n = 7)No other bla (n = 1)With other non- ESBL bla (n = 15)With other ESBL bla (n = 1)
TZP0 (≤ 4–8)1 (≤ 4–≥ 128)0 (≤ 4–8)0 (≤ 4–8)1 (≤ 4–≥ 128)0 (8)2 (≤ 4–≥ 128)1 (≥ 128)
ATM3 (≤ 1–≥ 64)12 (≤ 1–≥ 64)6 (≤ 1–≥ 64)12 (16–≥ 64)7 (≥ 64)1 (16)13 (2–≥ 64)1 (≥ 64)
CRO11 (8–≥ 64)54 (8–≥ 64)9 (8–≥ 64)12 (≥ 64)7 (≥ 64)1 (32)15 (16–≥ 128)1 (≥ 64)
CAZ2 (≤ 1–≥ 64)10 (≤ 1–≥ 64)5 (≤ 1–≥ 64)7 (4–≥ 64)6 (8–≥ 64)0 (4)12 (4–≥ 64)1 (≥ 64)
FEP1 (≤ 1–≥ 64)5 (≤ 1–≥ 64)3 (≤ 1–≥ 64)6 (2–≥ 64)7 (32–≥ 64)0 (2)2 (≤ 1–≥ 64)1 (≥ 64)
CIP2 (≤ 0.25–≥ 4)19 (≤ 0.25–≥ 4)4 (≤ 0.5–≥ 4)6 (≤ 0.25–≥ 4)5 (0.5–≥ 4)1 (≥ 4)13 (1–≥ 4)1 (≥ 4)
LVX2 (≤ 0.25–≥ 8)18 (1–≥ 8)4 (1–≥ 8)4 (≤ 0.25–≥ 8)5 (1–≥ 8)0 (1)12 (1–≥ 8)1 (≥ 8)
GEN8 (≤ 1–≥ 16)41 (≤ 1–≥ 16)8 (≤ 1–≥ 16)5 (≤ 1–≥ 16)6 (≤ 1–≥ 16)0 (≤ 1)10 (≤ 1–≥ 16)1 (≥ 16)
TOB0 (≤ 1–4)16 (≤ 1–≥ 16)3 (≤ 1–≥ 16)1 (≤ 1–≥ 16)0 (≤ 1–8)0 (8)4 (≤ 1–≥ 16)0 (4)
AMK0 (≤ 2)13 (≤ 2–≥ 64)3 (≤ 2–≥ 64)0 (≤ 2–4)0 (≤ 2)0 (4)2 (≤ 2–≥ 64)0 (≤ 2)
IMP0 (≤ 1)0 (≤ 1)0 (≤ 1)0 (≤ 1)0 (≤ 1)0 (≤ 1)0 (≤ 1)0 (≤ 1)
No. of MDR phenotype2 (18.2%)14 (25.9%)4 (44.4%)4 (33.3%)5 (71.4%)1 (100%)10 (66.7%)1 (100%)
  • TZP, piperacillin/tazobactam; ATM, aztreonam; CRO, ceftriaxone; CAZ, ceftazidime; FEP, cefepime; CIP, ciprofloxacin; LEV, levofloxacin; GEN, gentamicin; TOB, tobramycin; AMK, amikacin; IMP, imipenem.

  • Four isolates showing carbapenems resistance were not included.

  • P < 0.05 vs. subgroup CTX-M-14 with other non-ESBL bla.

  • P < 0.01 vs. subgroup CTX-M-14 with other non-ESBL bla.

β-Lactamase genes characterization

All 158 isolates yielded purified plasmids and harbored β-lactamase genes by PCR. Sequence analysis revealed that blaCTX-M, blaSHV, and blaTEM were present in 134, 120, and 92 isolates, respectively. A total of 149 (94.3%) isolates harbored one or more ESBL genes. Of 134 CTX-M producers, 78 carried the blaCTX-M-14, which was the most common type of ESBLs in seven hospitals, 19 isolates carried blaCTX-M-15, 17 blaCTX-M-27, 12 blaCTX-M-3, 4 blaCTX-M-55, 2 blaCTX-M-65, 2 blaCTX-M-24, 2 blaCTX-M-24a, 1 blaCTX-M-38, and 1 blaCTX-M-98. No group II, III, and V blaCTX-M have been detected. Sequencing of blaSHV PCR products indicated that 15 of 120 clinical isolates had blaSHV-12 and 7 blaSHV-5. Other ESBL genes were blaSHV2a (n = 3), blaSHV-2 (n = 2), blaSHV-27 (n = 2), and blaSHV-38 (n = 1). The most prevalent non-ESBL blaSHV was SHV-11 (n = 45, 28.5%), which commonly coexisted with other ESBLs except for 2 isolates. Other non-ESBL blaSHV were blaSHV-1 (n = 23), blaSHV-108 (n = 5), blaSHV-28 (n = 4), blaSHV-36 (n = 3), blaSHV-1a (n = 1), blaSHV-26 (n = 1), blaSHV-32 (n = 1), blaSHV-33 (n = 1), blaSHV-60 (n = 1), blaSHV-103 (n = 1), blaLEN (n = 1), and blaLEN-22 (n = 1). One novel SHV variant, of which the deduced protein sequence showed the combination of T18A and L35Q (according to the ABL numbering scheme) substitution in relation to blaSHV-1, named SHV-142, was detected (Fig. 1). Nearly, all of the blaTEM encoded TEM-1 except for one isolate carrying SHV-2a and TEM-135 with a single point mutation in CDS, T396G (data not shown). Seventeen (10.8%) isolates were detected to have two ESBL genes, and 1 (0.6%) isolate was detected to have three ESBL genes (Fig. 1). Five of 6 isolates with resistances to carbapenems also coded the blaKPC-2.

The UPGMA dendrogram, sequence types (STs), and genotypes of 155 ESBL-producing Klebsiella pneumoniae isolates from the seven tertiary hospitals in China. Dendrogram shows the majority of 155 isolates were un-related. The black vertical line shows the cut-off (similarity >70%). Letters in the strain IDs correspond to the source hospitals. The seven tertiary hospitals were Peking University People's Hospital (X), the First Affiliated Hospital of Anhui Medical University (A), the Second Hospital of Hebei Medical University (H), the Second Affiliated Hospital of Dalian Medical University (D), Fujian Province Hospital (F), the Central Hospital of China Aerospace Corporation (C), and the Third Affiliated Hospital of the Inner Mongolia Medical College (Baotou, B). Three isolates were non-typeable (the isolate H5 coding CTX-M-14+SHV-11+TEM-1, the isolate F4 coding CTX-M-27, and the isolate X2 coding TEM-1). Five isolates producing blaCTX-M-27+SHV-1 in the pane were collected from the same hospital (C). ACTT700603 was as the referral strain.

Antimicrobial susceptibilities of representative blaCTX-M-harboring K. pneumoniae isolates

An analysis of MICs and resistance patterns of the predominant blaCTX-M-14 (49.4%), blaCTX-M-15 (12%), and blaCTX-M-27 (10.8%) subtypes is shown in Table 2. With non-ESBL bla, the subgroups CTX-M-15 and CTX-M-27 exhibited higher proportion of resistance to aztreonam and ceftazidime than that of subgroup CTX-M-14 (P < 0.05). The MIC ranges to aztreonam and ceftazidime among 12 isolates in subgroup CTX-M-15 were 16–≥ 64 µg mL−1 and 4–≥ 64 µg mL−1, respectively. Also, the MIC ranges to aztreonam and ceftazidime in subgroup CTX-M-27 were 2–≥ 64 µg mL−1 and 4–≥ 64 µg mL−1, respectively, while both of which among 54 isolates in subgroup CTX-M-14 were ≤ 1–≥ 64 µg mL−1. The subgroup CTX-M-15 exhibited higher level of resistance to cefepime than that of subgroup CTX-M-14 (P < 0.01), and the MIC range in subgroup CTX-M-15 was 2–≥ 64 µg mL−1. As for subgroup CTX-M-27, it exhibited higher proportion of resistance to ciprofloxacin and levofloxacin than that of subgroup CTX-M-14 (P < 0.01), and the MIC range to ciprofloxacin in subgroup CTX-M-27 was 1–≥ 4 µg mL−1, while it was ≤ 0.25–≥ 4 µg mL−1 in subgroup CTX-M-14. The proportion of MDR in subgroup CTX-M-27 was higher than that in subgroup CTX-M-14 (P < 0.01). Nevertheless, when other ESBL bla (except for blaKPC-2) were present, subgroup CTX-M-14 showed significant increase in resistance to aztreonam and ceftazidime (P < 0.05).


To investigate the genetic relationship between the 158 clinical isolates, MLST was performed for all isolates. ST patterns for three isolates were not obtained because of the deletion or insertion of oligonucleotide in the gene (tonB) sequence coding for periplasmic energy transducer. Of the 155 isolates, 74 STs were identified, and the most prevalent ST was ST11 (n = 19), followed by ST48 (n = 9), ST37 (n = 7), ST17 (n = 7), ST15 (n = 6), ST340 (n = 6), ST23 (n = 5), and so forth (Fig. 1). The UPGMA dendrogram showed that there were only a few blaCTX-M-14-producing isolates exhibiting genetic relationships (Fig. 1). Analysis of STs by eBURST showed three clonal complexes (CCs) CC11 (n = 34), CC709 (n = 32), CC37 (n = 18), and other singletons (data not shown). This result also indicated the majority of 155 isolates were unrelated among the six geographical areas. Twenty-nine new STs in six hospitals except for Inner Mongolia were identified. The MLST results showed a large genetic background diversity in these ESBL-producing K. pneumoniae isolates from the six geographical areas in China. Interestingly, five isolates producing blaCTX-M-27 with the same patterns (ST48) were originated from patients in the same hospital.

Nucleotide sequence accession number

The nucleotide sequences of the novel blaSHV-142 and blaTEM-135 have been deposited in the GenBank nucleotide database under accession number JQ029959 and JQ060998, respectively.


ESBL-producing K. pneumoniae strains are frequently associated with nosocomial outbreaks, especially in ICU settings (Falagas & Karageorgopoulos, 2009; Shu et al., 2010). Senior, critical, or immunocompromised statuses are important risk factors for such infections (Falagas & Karageorgopoulos, 2009). In the present study, nearly two-thirds of the patients were over 60 years old, and the majority of isolates came from patients in common wards, where the respiratory expectoration was of significance contributing to the dissemination of ESBL-producing organisms. Since the first report of ESBLs in 2002 (Chanawong et al., 2002), blaCTX-M has been predominant in mainland (Yu et al., 2007; Liu et al., 2009). In this multicentre study, the prevalence of ESBL production in K. pneumoniae has been demonstrated to be about 40%. Of 158 ESBL-producers, the isolates harboring ESBL genes and blaCTX-M-14 were 94.3% and 49.4%, respectively, and were shown to increase 10% and 9% to those in another large-scale study (Yu et al., 2007), respectively. The proportion of blaCTX-M increased 12% compared to the percentage (72.3%) described in a report of southern China three years ago (Liu et al., 2009) and doubled the percentage reported nine years ago (Li et al., 2003). Because the usage of plasmid-based amplification method in this study and the potential false-negative products owing to the unbinding on some novel bla, the detection of β-lactamase genes may have been underestimated. Although there are some differences in the source of the isolates in our study as compared to the studies mentioned above, our results clearly suggest the increasing prevalence of blaCTX-M in K. pneumoniae in China.

CTX-M-type ESBLs exhibit powerful activity against cefotaxime and ceftriaxone but generally not against ceftazidime, and several variants with enhanced ceftazidimase activity have been reported (Poirel et al., 2002; Bonnet et al., 2003; Rossolini et al., 2008). In this study, it was observed that the isolates harboring CTX-M-15 or CTX-M-27 alone exhibited higher resistance rates to ceftazidime and aztreonam than that in subgroup CTX-M-14 without other ESBLs (Table 2). Further, a high percentage of isolates harboring blaCTX-M-27 demonstrated the MDR phenotype. To our knowledge, this is the first study about the high prevalence of CTX-M-27 in Enterobacteriaceae in China. This warrants for an active surveillance to monitor these resistant bacteria.

The overall resistance rates to the tested β-lactam antimicrobial agents were over 30% except for cefepime, piperacillin/tazobactam, and cefotetan in this study. As shown in Table 2, only 9.3% isolates harboring CTX-M-14 alone showed resistance to cefepime, but 50% isolates harboring CTX-M-15 exhibited resistance (P < 0.01), and a 100% resistance rate when CTX-M-15 coexisted with other ESBLs. Nevertheless, piperacillin/tazobactam show only 10.1% resistance rate in vitro, although the proportion increased to 26.7% when the isolates contained two types of ESBLs(blaCTX-M + blaSHV) (Table 1). Several clinical intervention studies also supported that piperacillin/tazobactam may contribute to preventing the ESBL-producing K. pneumoniae outbreaks (Lee et al., 2007; Tangden et al., 2011). These properties highlight the value of piperacillin/tazobactam as empirical therapy for infections by suspected organisms possessing a single ESBL (especially the blaCTX-M). Although the overall sensitivity to cefotetan was nearly 90% in this study, its clinical utility has been limited because of its severe adverse events (Martin & Laber, 2006). There were 34.2% isolates that met the MDR criteria in our study. The lowest resistance rate among 158 isolates to non-β-lactam agents was still as high as 26.6% (to amikacin). Therefore, therapeutic options for ESBL-producing K. pneumoniae infections will become increasingly limited.

In this survey, the most prominent non-ESBL blaSHV was identified to be SHV-11 (28.5%). Interestingly, a survey in Korea indicated that the incidence of blaSHV-12 was more predominant in K. pneumoniae strains carrying the chromosomal blaSHV-11 (19.3%) than in strains carrying the blaSHV-1 (2.0%) (Lee et al., 2006). SHV-12 is classified as group 2be and sometimes shows high-level resistance to third-generation cephalosporins and resistance to β-lactamase inhibitors (Nüesch-Inderbinen et al., 1997). It is currently not known why this overabundance of SHV-12 had occurred, but the high prevalence of blaSHV-11 in our study certainly warrants further surveillance. Two isolates carrying the novel SHV-142 together with CTX-M-14 were detected. Both isolates showed slight MICs increase to gentamicin and ciprofloxacin to isolates harboring CTX-M-14 alone (data not shown). Five isolates coding blaSHV-108 were detected, and they all showed the MDR phenotype (data not shown). The data indicated the isolates co-harboring SHV-108 showed high MIC values to non-β-lactam antibiotics. This is the first report of the occurrence of SHV-60, SHV-103, and SHV-108 in China. blaTEM-1 was detected in 91 isolates but one encoding TEM-135, which was sporadically reported in Neisseria gonorrhoeae strains (Ohnishi et al., 2010). In this study, 6 (3.8%) carbapenem-resistant isolates were detected and five of them were with blaKPC-2. Lower breakpoints of the carbapenems do not completely exclude the possibility of resistant KPC isolate to be called susceptible (Bulik et al., 2010). This suggests that KPC producers have been underestimated in this study. Nine (5.7%) isolates no blaCTX-M/SHV/TEM ESBL was detected (Table 1). These isolates may have produced another ESBL, which was not determined in this study or might have given positive results for ESBL activity.

Among 155 isolates, only a small number of isolates showed clonal relationships (> 70% similarity) by the MLST methods. ST-11 and CC11 were the most predominant, present in 19 (12.3%) and 34 (21.9%) isolates, respectively. As for the predominate ESBL, CTX-M-14-producing K. pneumoniae strains of the main STs 37, 5, 505, 11, 23, 1, 22, and 48 were scattered in six geographical areas, exhibiting a multiclonal distribution. ST340 and ST15 as two major CTX-M-15-producing K. pneumoniae epidemic clones were dispersed in three independent areas. Three SHV-12 clones, ST722, ST340, and ST709 were also dispersed in three areas. These data indicate that the predominant ESBL-producing K. pneumoniae isolates from lower respiratory tract might acquire ESBL genes independently. Several different clones might contribute to the horizontal transfer of ESBL genes among K. pneumoniae in China. It is worthy of note that clone ST-48 harboring CTX-M-27 coupled with SHV-1 was detected in one hospital, especially that 4 isolates were detected from the same ward, suggesting the possible single clone dissemination. These findings confer the concern of various multiresistant pathogens and present new epidemiological and clinical challenges.

In conclusion, although some ESBL genes may be missed by this basically plasmid encoded method, our study clearly indicates the high prevalence of blaCTX-M and large phylogenetic diversity in ESBL-producing K. pneumoniae. The consequent surveillance of multiple ESBL-producing organisms with MDR phenotype is of paramount importance.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Fig. S1. Distribution of 158 ESBLs-producing K. pneumoniae isolates from seven tertiary hospitals located in six geographical areas (five red areas and the yellow Beijing area) in China.

Table S1. Primers used for PCR amplification of β-lactamase genes.

Table S2. The demographics and origination of 133 K. pneumoniae isolates from seven tertiary hospitals in China (no. of patients).


This project was supported by the National Science and Technology Major Project of the Ministry of Science and Technology of China (Grant No. 2009ZX10004-016) and National High-Tech R&D program (Grant No. 2006AA02Z4A9). We would like to gratefully appreciate Dr Dakun Wang, Senior Scientist, Precision Therapeutics, for kindly helping the English version.


View Abstract