Carbapenem resistance in Gram-negative bacteria is on the rise in the United States. A regional network was established to study microbiological and genetic determinants of clinical outcomes in hospitalized patients with carbapenem-resistant (CR) Klebsiella pneumoniae in a prospective, multicenter, observational study. To this end, predefined clinical characteristics and outcomes were recorded and K. pneumoniae isolates were analyzed for strain typing and resistance mechanism determination. In a 14-month period, 251 patients were included. While most of the patients were admitted from long-term care settings, 28% of them were admitted from home. Hospitalizations were prolonged and complicated. Nonsusceptibility to colistin and tigecycline occurred in isolates from 7 and 45% of the patients, respectively. Most of the CR K. pneumoniae isolates belonged to repetitive extragenic palindromic PCR (rep-PCR) types A and B (both sequence type 258) and carried either blaKPC-2 (48%) or blaKPC-3 (51%). One isolate tested positive for blaNDM-1, a sentinel discovery in this region. Important differences between strain types were noted; rep-PCR type B strains were associated with blaKPC-3 (odds ratio [OR], 294; 95% confidence interval [CI], 58 to 2,552; P < 0.001), gentamicin nonsusceptibility (OR, 24; 95% CI, 8.39 to 79.38; P < 0.001), amikacin susceptibility (OR, 11.0; 95% CI, 3.21 to 42.42; P < 0.001), tigecycline nonsusceptibility (OR, 5.34; 95% CI, 1.30 to 36.41; P = 0.018), a shorter length of stay (OR, 0.98; 95% CI, 0.95 to 1.00; P = 0.043), and admission from a skilled-nursing facility (OR, 3.09; 95% CI, 1.26 to 8.08; P = 0.013). Our analysis shows that (i) CR K. pneumoniae is seen primarily in the elderly long-term care population and that (ii) regional monitoring of CR K. pneumoniae reveals insights into molecular characteristics. This work highlights the crucial role of ongoing surveillance of carbapenem resistance determinants.


Antimicrobial resistance is “one of the greatest threats to human health worldwide” (1). The additional annual cost associated with treating antimicrobial-resistant pathogens in the United States is estimated to be between $21 billion and $34 billion (1). Carbapenem-resistant (CR) Klebsiella pneumoniae infections are especially worrisome, as very few treatment options remain for patients and outcomes are generally poor.
Central to the control of the problem of CR K. pneumoniae is the institution of effective regional surveillance networks. To address this, we formed a multicenter consortium to prospectively study the epidemiology, molecular characteristics, and clinical outcomes of infections due to CR K. pneumoniae in hospitalized patients in northeastern Ohio. Here, we report the initial findings of this consortium and describe the genomic analysis of a blaNDM-1-carrying strain that was discovered as a result of organized surveillance.


Establishment of the Great Lakes CR K. pneumoniae consortium.

A multicenter consortium was established that consists of three health care systems that include 18 hospitals serving more than 2 million people in northeastern Ohio. These included both community-based hospitals and tertiary-care referral centers ranging in size from 25 to >1,400 beds. Each health care system participating in this study is served by one or more clinical microbiology laboratories that process an average of 200,000 bacterial cultures from clinical samples yearly. None of the hospitals performed screening for asymptomatic carriage of CR K. pneumoniae during the study period.


We defined CR K. pneumoniae as isolates with decreased susceptibility to meropenem or imipenem (MIC, >1 μg/ml). Bacterial identification to the species level and routine antimicrobial susceptibility testing were performed with the MicroScan (Siemens Healthcare Diagnostics) or Vitek2 (bioMérieux) automated system supplemented by additional testing if required. In this study, MIC results were interpreted according to breakpoints defined by the Clinical and Laboratory Standards Institute (CLSI), with the exception of tigecycline and colistin, where criteria set forth by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) were used because of a lack of validated CLSI breakpoints. At each site, study personnel were notified by the local microbiology laboratory upon the identification of a CR K. pneumoniae isolate from a hospitalized patient. Clinical data were obtained from the medical record and entered into a centralized clinical database. The Institutional Review Boards of all of the health care systems involved approved this study.


All of the hospitalized patients from whom CR K. pneumoniae was isolated were included if their hospitalization began and ended within the study period, which lasted from 12/24/2011 until 3/1/2013. All of the analyses are based on the first culture positive for CR K. pneumoniae during the index hospitalization of each individual patient.

Definitions of infection.

A patient with a culture positive for CR K. pneumoniae was deemed to have an infection if CR K. pneumoniae was isolated from blood or any other sterile source. For patients with positive respiratory cultures, the criteria outlined by the American Thoracic Society and the Infectious Diseases Society of America were used (2, 3). For patients with positive cultures from urine or surgical wounds, the CDC National Healthcare Safety Network criteria were applied (4). Patients with cultures from nonsurgical wounds were considered infected only if the treating physician documented infection in the medical record and evidence of systemic inflammation on the day the positive culture was documented; this was defined as an abnormal systemic white cell count (either >10 × 103 or <4 × 103 cells/μl) and/or an abnormal body temperature (either >99.5°F or <96°F). All of the other culture episodes were designated colonizations.

Clinical data collection and definitions.

Clinical data collection included the following: demographics, prehospital origin and posthospital disposition, comorbidities, including the Charlson comorbidity index (5), CR K. pneumoniae treatment variables, previous antibiotic use, and culture site-specific variables. Previous antibiotic use was recorded as the class of antibiotic with documented exposure in the 14 days leading up to the day of the first positive culture. Infections were defined as nosocomial if the first CR K. pneumoniae-positive culture was obtained >48 h after hospital admission. Health care-associated infections were defined as infections in patients admitted from long-term care facilities or other hospitals who had a first CR K. pneumoniae-positive culture within the first 48 h. Patients with a first CR K. pneumoniae-positive culture within 48 h who were admitted from home were deemed to have community-acquired infections. Renal insufficiency was defined as any serum creatinine level of ≥2 mg/dl. The Pitt bacteremia score (PBS) was calculated as described previously (6). If patients were intubated and sedated and no further information on mental status was available, the mental status was coded as “stuporous” for the purpose of calculating this score only. This score has been validated as a predictor of the outcome of CR K. pneumoniae bloodstream infection (7).

Molecular detection of carbapenemases in CR K. pneumoniae.

All of the available CR K. pneumoniae isolates were sent to a central research laboratory in this consortium for further testing. PCR amplification of the blaKPC, blaNDM blaVIM, blaIMP, and blaOXA-48 genes was performed with established primers and methods (8, 9). Amplicons were sequenced at a commercial sequencing facility (MCLAB, San Francisco, CA) and analyzed as previously described (8, 9).

Molecular typing of CR K. pneumoniae.

Genetic similarity among isolates was investigated by repetitive extragenic palindromic PCR (rep-PCR) by using the DiversiLab strain typing system (Bacterial BarCodes, bioMérieux, Athens, GA). Isolates with ≥95% similarity were considered to be of the same rep-PCR type. Rep-PCR has previously been shown to have adequate discriminatory power for CR K. pneumoniae strain typing (10). Multilocus sequence typing (MLST) of the CR K. pneumoniae isolates of the predominant rep-PCR types was performed as described previously (11). The sequences of seven housekeeping genes (rpoB, gapA, mdh, pgi, phoE, infB and tonB) were compared to the MLST database (http://www.pasteur.fr/recherche/genopole/PF8/mlst).

Genome sequencing.

To generate a finished accurate CR K. pneumoniae blaNDM-1 genome, independent de novo assemblies were generated from sequence data from Pacific Biosciences and Illumina instruments. CR K. pneumoniae blaNDM-1 genomic DNA was sheared, SMRTbell libraries were generated, XL polymerase-bound complexes were loaded onto SMRT cells, and sequencing was performed with the RSII system as suggested by the manufacturer (Pacific Biosciences, Menlo Park, CA). Assemblies were generated with the Celera Assembler and then postprocessed with Quiver for realignment and polishing as described previously (12). In brief, we utilized the suggested hierarchical genome assembly process, which consisted of (i) selecting long sequencing reads as seed data, (ii) using seed sequences as references to recruit consensus shorter reads, (iii) assembling long seed data and short reads with the Celera Assembler adapted for SMRT sequencing data, (iv) polishing the assembly with Quiver to generate consensus sequences, and (v) manually trimming overlapping identical end regions to create circular finished chromosomes and plasmids.
For independent sequence comparison, a Nextera library was generated from the same genomic DNA and sequenced to an ∼100× depth on an Illumina MiSeq instrument; this was followed by assembly with MIRA as described previously (13). Draft annotation was performed with the RAST pipeline and finalized with PGAP.


Univariate logistic regression was used to estimate associations between CR K. pneumoniae strain types and microbiological and clinical variables. Cox proportional hazard modeling was used to estimate associations between potential predictors and time to death in the hospital. The time in days from the first positive culture to death in the hospital was displayed in a Kaplan-Meier plot. A log rank test was used to compare groups. Patients were censored at the time of discharge or on postculture day 30 if still in the hospital and alive. All of the analyses were performed with JMP software (SAS Inc., Cary, NC).

Nucleotide sequence accession numbers.

Sequences have been deposited in GenBank, and the accession numbers can be found via NCBI BioSample record number 2777842.



During the study period, 251 individual patients were included (Table 1 and Fig. 1). Patients were generally elderly (median age, 70 years; interquartile range [IQR], 58 to 81 years) and had several comorbid conditions (median Charlson score, 4; IQR, 2 to 6). Of note, 71 patients (28%) were admitted from home, whereas the remainder came from skilled-nursing facilities (SNFs) (49%) or long-term acute-care facilities (10%) or were transferred from other hospitals (13%).
TABLE 1 Demographic characteristics of the patients in this study
Patient characteristicAllInfectedColonized
Total no.251114137
Median age, yr (IQR)70 (58–81)67 (54–81)73 (64.5–83)
No. (%) female150 (60)67 (59)83 (61)
No. (%) who were:   
    White, non-Hispanic145 (58)63 (55)82 (60)
    Black, non-Hispanic95 (38)49 (43)46 (34)
    Hispanic7 (3)2 (2)5 (4)
    Other4 (2)04 (3)
No. (%) with:   
    Diabetes mellitus140 (56)66 (58)74 (54)
    Heart diseasea142 (57)62 (55)80 (58)
    Renal insufficiency65 (26)31 (27)34 (25)
    COPDb73 (29)30 (27)43 (31)
    Malignancy36 (14)18 (16)18 (13)
    Dementia54 (22)22 (19)32 (23)
    Immunocompromise25 (10)14 (12)11 (8)
        Prednisone therapyc12 (5)6 (5)6 (4)
        Solid-organ transplant10 (4)5 (4)5 (4)
        HIV infection2 (1)2 (2)0
        Hematopoietic stem cell  transplant1 (0)1 (1)0
Median Charlson comorbidity index (IQR)4 (2–6)4 (2–6)4 (2–6)
No. (%) from:   
    SNF123 (49)50 (44)73 (53)
    Home71 (28)34 (30)37 (27)
    Another hospital32 (13)20 (18)12 (8)
    Long-term acute-care facility25 (10)10 (9)15 (11)
Coronary artery disease and/or congestive heart failure.
COPD, chronic obstructive pulmonary disease.
Prednisone or equivalent corticosteroid at 10 mg/day.
FIG 1 Flow diagram outlining the characteristics of the patients in this study. In each box, the origin of admission for the index hospitalization, the first culture site, and whether the first culture represented infection or colonization are shown.

CR K. pneumoniae infection and colonization.

Urine (63%) was the predominant anatomic source of CR K. pneumoniae (Fig. 2), followed by respiratory (12%), wound (11%), blood (10%), and other (4%) samples. Patients culture sources grouped as “other” included abscess (n = 2), vascular catheter tip (n = 3), ascites (n = 2), bile (n = 2), and stool (n = 1). Antibiotic use prior to the first positive culture was common; only 29% had no documented antibiotic exposure in the preceding 14 days. CR K. pneumoniae infection was present in 45% of the patients. Of patients with CR K. pneumoniae infections, 43/114 (38%) had nosocomial infections, 48/114 (42%) had health care-associated infections, and 23 (20%) had community-acquired infections.
FIG 2 (A) Numbers of colonizations and infections per anatomical site of CR K. pneumoniae-positive cultures. Percentages are the proportions of patient with infections. (B) Selected antimicrobial susceptibility rates. Note that breakpoints defined by the CLSI were used for aminoglycosides and criteria set forth by the EUCAST were used for colistin and tigecycline. TMP/SMX, trimethoprim-sulfamethoxazole.
Most (74%) of the patients with CR K. pneumoniae colonizations had urinary tract colonizations, and half of the patients with infections had urinary tract infections. The use of a Foley catheter was documented in 53 and 39% of the patients with CR K. pneumoniae urinary tract colonizations and infections, respectively.
The antibiotic susceptibilities determined are summarized in Fig. 2. Colistin susceptibilities were determined for the first isolates of 114 patients; of these, 8/114 (7%) were nonsusceptible. In contrast, tigecycline nonsusceptibility was more common; 83/182 (45%) isolates tested were nonsusceptible to tigecycline. However, for most of these isolates (56/182, 31%) the MIC was 2 μg/ml (intermediate susceptibility). Three isolates that were colistin nonsusceptible were also nonsusceptible to tigecycline (intermediate, n = 1; resistant, n = 2).


The outcomes of the index hospitalizations are summarized in Table 2. Hospital stays were prolonged (median length of stay [LOS], 9 days; IQR, 5 to 17 days) and complicated. About half of the patients (51%) were admitted to the intensive care unit (ICU), and the median ICU LOS in those admissions was 6 days (IQR, 3 to 15 days). Most of the patients were discharged to SNFs (43%), and only 17% of the patients were discharged home. Of the 71 patients who were admitted from home, only 48% could be discharged home. Thirty-nine patients died, and an additional six were discharged to hospice care. In patients with CR K. pneumoniae infections versus colonizations, the hospital (including hospice care) mortality rate was 23% versus 14%.
TABLE 2 Outcome of first admission per patienta
Patient characteristicAllInfectedColonizedP valueb
Total no.251114137 
No. (%) at following location at time of culture:   0.48
    Emergency department66 (26)26 (23)40 (29) 
    Ward108 (43)50 (44)58 (42) 
    ICU77 (31)38 (33)39 (28) 
Median PBS (IQR)2 (1–4)2 (1–4)2 (0–4)0.10
Median no. of days to first positive culture (IQR)1 (0–4)1 (0–6)0 (0–4)0.08
No. (%) with CR K. pneumoniae isolated:    
    Within first 24 h124 (49)54 (47)70 (51) 
    On days 2–569 (27)29 (25)40 (29) 
    After day 558 (23)31 (27)27 (20) 
Median LOS, days (IQR)9 (5–17)10 (6–20)8 (5–15)0.11
No. (%) with ICU admission129 (51)65 (57)64 (47)0.13
Median no. of days in ICU (IQR)a6 (3–15)7 (3.5–15.5)6 (3–15)0.31
No. (%) with:    
    Surgery60 (24)38 (33)22 (16)<0.01
    Renal insufficiency91 (36)43 (38)48 (35)0.69
    Renal replacement therapy46 (18)24 (21)22 (16)0.33
    Mechanical ventilation85 (34)47 (41)38 (28)0.03
    Tracheostomy39 (16)17 (15)22 (16)0.86
No. (%) with following disposition:   0.05
    Death/hospice45 (18)26 (23)19 (14) 
    Home43 (17)19 (17)24 (18) 
    SNF107 (43)39 (35)68 (50) 
    Other56 (22)30 (26)26 (19) 
For patients with any ICU admission.
Fisher's exact test for binomial variables, t test for normally distributed continuous variables, and likelihood ratio test for distributions.
Of the 26 patients with bacteremia, 21 (81%) received either monotherapy (tigecycline [n = 4], colistin [n = 4], aminoglycosides [n = 3], or trimethoprim-sulfamethoxazole [n = 1]) or combination therapy (tigecycline-gentamicin [n = 5], amikacin-colistin [n = 3], or tigecycline-colistin [n = 1]). Five patients with bloodstream infections died before antibiograms were available and did not receive antibacterial therapy with activity against CR K. pneumoniae.
Antibacterial therapy active against CR K. pneumoniae was used in 65% of the patients with urinary tract infections and 60% of the patients with other sources of infection. Of note, 46% of the patients with urinary colonization were also treated with agents active against CR K. pneumoniae.
In patients with CR K. pneumoniae infection (n = 114), the source of infection was associated with the time to death in the hospital within 30 days of admission (Fig. 3). Variables associated with the time to death in the hospital in an age-corrected Cox proportional hazard model included severe acute disease, as defined by a PBS of ≥4 (hazard ratio [HR], 3.82; 95% CI, 1.49 to 11.11; P = 0.005) and a nonurinary source of infection (HR, 2.7; 95% CI, 1.04 to 8.49; P = 0.040). Receipt of antibiotics with in vitro susceptibility within 48 h of a positive culture was not associated with improved survival in this analysis (HR, 1.46; 95% CI, 0.41 to 4.13; P = 0.52).
FIG 3 Kaplan-Meier survival curves showing times from the first positive culture to death in the hospital for patients with CR K. pneumoniae infections by source of infection, censoring at 30 days, or time of discharge, whichever occurred first (P = 0.015 by log rank test).

Molecular strain typing and mechanisms of resistance.

Molecular strain typing by rep-PCR was performed for all of the 113 patients whose CR K. pneumoniae isolates were available, i.e., 52 patients with CR K. pneumoniae colonizations and 61 patients with CR K. pneumoniae infections. The two most common types were A (36/113, 29%) and B (52/113, 46%), which were both identified as belonging to MLST sequence type 258 (ST258). The remaining 22 isolates (19%) represented 16 different rep-PCR types, including strains belonging to ST15, ST17, ST299, and ST395.
Carbapenemase genes were detected in isolates from 109/113 patients, while none of the carbapenemase genes tested (blaKPC, blaNDM blaVIM, blaIMP, and blaOXA-48) were detected in the four isolates tested. Furthermore, these four isolates did not phenotypically express a carbapenemase, according to a modified Hodge test (data not shown); the carbapenem resistance of these isolates is likely due to a combination of extended-spectrum beta-lactamases (ESBLs) and changes in outer membrane proteins (14). All but one of the carbapenemase-harboring isolates carried either blaKPC-2 (51/109, 48%) or blaKPC-3 (57/109, 51%).
The remaining carbapenemase-harboring isolate tested positive for blaNDM-1, a sentinel finding in this geographic region. The patient in whom this blaNDM-1-positive CR K. pneumoniae strain was found was a male from Egypt who was admitted for cardiac surgery. Nine days postoperatively, he developed abdominal pain, turbid urine, a fever to 38.7°C, and leukocytosis. Urine analysis showed more than 25 white cells per high-power field, and his urine culture grew CR K. pneumoniae, as well as an ESBL-producing Escherichia coli strain that was susceptible to carbapenems. He was treated with a single dose of gentamicin and 7 days of meropenem. His symptoms resolved, and he was discharged 8 days after his first positive CR K. pneumoniae culture. He did not have a readmission during the study period. No other CR K. pneumoniae isolates obtained during the study period tested positive for blaNDM-1, suggesting that no transmission to other patients occurred (see below).
As ST258 is the most common CR K. pneumoniae sequence type worldwide, we further explored clinical and microbiological differences between the A and B types identified by rep-PCR (Table 3). Antibiotic susceptibility patterns were significantly associated with rep-PCR strain types; type A CR K. pneumoniae was more likely than type B to be tigecycline (odds ratio [OR], 5.34; 95% CI, 1.30 to 36.41; P = 0.018) and gentamicin (OR, 24.11; 95% CI, 8.39 to 79.38; P < 0.001) susceptible and less likely to be susceptible to amikacin (OR, 0.09; 95% CI, 0.02 to 0.31; P < 0.001). blaKPC was present in all of the type A and in all but one of the type B CR K. pneumoniae isolates. KPC-2 and KPC-3 were the two carbapenemases found. A significant association between the rep-PCR type and the KPC enzyme found was also discovered; type A CR K. pneumoniae isolates were associated with KPC-2, whereas type B isolates were associated with KPC-3 (OR, 294.00; 95% CI, 58.16 to 2552.68; P < 0.001).
TABLE 3 Comparison of patients with rep-PCR type A and B strains
Patient characteristicABORa (95% CI)P valueb
Total no.3952  
No. (%) with:    
    Any carbapenemase39 (100)51 (98)  
    KPC-236 (92)2 (4)294.00 (58.16–2552.68)<0.001
    KPC-33 (7)49 (96)0.00 (0.00–0.02)<0.001
No. (%) with isolate from:   0.135
    Urine19 (49)36 (69)  
    Blood11 (28)8 (15)2.61 (0.90–7.81) 
    Other9 (23)8 (15)2.13 (0.70–6.57) 
No. (%) infected22 (56)27 (51)1.20 (0.52–2.78)0.67
No. with susceptible isolates/no. with isolates tested (%)    
    Colistin32/35 (91)37/39 (95)0.57 (0.07–3.69)0.66
    Tigecycline23/25 (92)28/41 (68)5.34 (1.30–36.41)0.018
    Amikacin7/18 (39)42/48 (88)0.09 (0.02–0.31)<0.001
    Gentamicin30/38 (79)7/52 (13)24.11 (8.39–79.38)<0.001
    Trimethoprim-sulfamethoxazole9/36 (25)22/52 (42)0.45 (0.17–1.13)0.091
No. (%) from:    
    SNFs9 (23)25 (48) 0.043
    Home15 (38)15 (29)2.78 (0.99–8.16) 
    Other15 (38)12 (23)3.47 (1.21–10.55) 
No. (%) with strain present at admission14 (36)28 (54)0.48 (0.20–1.11)0.088
Median LOS (IQR)12 (6–32)8 (5–19.75)1.02c (1.00–1.04)0.043
No. (%) with following disposition:    
    Death/hospice11 (28)10 (19) 0.72
    Home7 (18)13 (25)0.49 (0.13–1.69) 
    SNF15 (38)20 (38)0.68 (0.22–2.02) 
    Other6 (15)9 (17)0.61 (0.15–2.29) 
OR for A versus B.
Univariate logistic regression.
OR per day increase in LOS.
On clinical grounds, patients with type B CR K. pneumoniae were more likely to be admitted from an SNF (OR, 3.09; 95% CI, 1.26 to 8.08; P = 0.013) and to have a shorter LOS (OR, 0.98/day increase; 95% CI, 0.95 to 1.00; P = 0.043) than patients with type A CR K. pneumoniae.

Genome sequencing of blaNDM-1-producing strain.

As the isolation of a blaNDM-1-producing CR K. pneumoniae strain represented a “sentinel event” in our community, the bacterial genome of this strain was sequenced in order to provide a reference sequence for the development of clinical diagnostics for these newly emerging organisms. In brief, the chromosome of this K. pneumoniae strain is 5,284,834 nucleotides (nt) in length, has a 57% GC content, and was determined to belong to ST395 (11), which is unique in our region.
Two plasmids were also found in this strain. Plasmid 1 was 94,656 nt long and present at 1.6× relative to the chromosome. Plasmid 2 carried the blaNDM-1 gene, was 118,033 nt long, and was present at 2.2× relative to the chromosome. Plasmid 2 is an IncF replicon with similarity to other IncF plasmids, like pKPN4 (accession no. CP000649.1). Notably, the blaNDM-1 gene environment is adjacent to a bleomycin resistance protein-encoding gene, a finding that is common in blaNDM-1-producing strains and is thought to impact the overall survival value of this strain (15).


Through this comprehensive effort, a clear picture of the origin, acute and chronic morbidity, hospitalization characteristics, and outcomes of hospitalized patients with CR K. pneumoniae in our region has emerged. Hospitalized patients with CR K. pneumoniae originate primarily from the long-term care population (1619). Skilled nursing and long-term acute-care facilities clearly play an important role in the epidemiology of CR K. pneumoniae. Whether this is simply a reflection of the overlap of risk factors for multidrug-resistant (MDR) organisms with risk factors for admission to long-term care facilities, such as comorbidities and invasive device use, or whether this is an indication of transmission within long-term care facilities remains to be seen (19).
Importantly, a substantial minority of the patients we studied were admitted from home. These patients may have had prior health care or long-term care exposure but were admitted from home for their index hospitalization during the study period. This has an impact for hospitals that screen patients for MDR organisms upon admission. It may not be sufficient to screen only those who are admitted from other hospitals.
Patients with CR K. pneumoniae are both chronically and acutely ill and have prolonged and complicated hospitalizations. Moreover, patients are unlikely to be discharged home, and the overall hospital mortality rate was 16%, with an additional 2% of the patients discharged to hospice care. In patients with CR K. pneumoniae infections, clinical factors associated with time to death in the hospital included expected variables such as the source of infection and severe acute illness, as previously described (7).
A troublesome trend was observed in the susceptibilities of CR K. pneumoniae to the “last-resort” antimicrobials colistin, tigecycline, and aminoglycosides. Colistin nonsusceptibility occurred in a small subset of the patients, and tigecycline and aminoglycoside nonsusceptibility was common. Most concerning was the fact that two patients presented with colistin- and tigecycline-resistant isolates. This is clear evidence that new antimicrobials directed against CR K. pneumoniae are urgently needed, as well as ongoing efforts to stem the rise in MDR pathogens.
The great majority of isolates belonged to ST258, which is the predominant strain of CR K. pneumoniae in the United States and worldwide (20). Within this sequence type, however, we were able to further identify two predominant separate strains by rep-PCR. Importantly, these two strains differed in their susceptibilities to aminoglycosides and tigecycline. In addition, differences in clinical outcomes were observed in patients possessing the two strains.
A very important finding in our cohort was the presence of a single patient during the study period who was infected with blaNDM-1-harboring CR K. pneumoniae. NDM-1 remains an uncommon mechanism of carbapenem resistance in the United States but is widespread in India, Pakistan, Bangladesh, and other parts of the world (21, 22). Fortunately, continued vigilance through this network has not detected further evidence of dissemination of blaNDM-1-harboring CR K. pneumoniae in our region.
Our study has several limitations. While we have used standardized definitions of infection, the distinction between infection and colonization remains a challenge. We suspect that at least some patients with infections were misclassified as having colonizations. However, when standardized definitions are used, the likelihood of the patients with colonizations being misclassified as having infections is low. In addition, our definition of CR K. pneumoniae may have missed some non-KPC-producing K. pneumoniae isolates with lower carbapenem MICs. The lack of active surveillance means that we are likely to underestimate the true number of CR K. pneumoniae-colonized patients. We categorized 20% of the infections as community acquired. However, data on prior health care exposures that may have preceded the current admission from home are not available.
The heterogeneity of antibiotic treatment in our cohort precluded comparative effectiveness analysis. The observed lack of association between the risk of death and CR K. pneumoniae-directed treatment in the first 48 h is probably secondary to both confounding by indication and other unmeasured covariables and the limited efficacy of currently available agents. Several studies have provided preliminary data suggestive of a benefit of combination therapy (23, 24).
Our findings illustrate the ongoing and increasing threat that CR K. pneumoniae poses to our vulnerable population and the clear benefits of real-time surveillance and molecular epidemiology in tracking the dissemination and clinical impact of resistance genes in specific pathogens.


This work was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award UM1AI104681 and by funding to D.V.D. and F.P. from the Clinical and Translational Science Collaborative of Cleveland, UL1TR000439 from the National Center for Advancing Translational Sciences (NCATS) component of the National Institutes of Health, and NIH Roadmap for Medical Research. In addition, this work was supported in part by the Veterans Affairs Merit Review Program (R.A.B.), the National Institutes of Health (grants AI072219-05 and AI063517-07 to R.A.B.), the Geriatric Research Education and Clinical Center VISN 10 (R.A.B.), the Research Program Committees of the Cleveland Clinic (D.V.D.), an unrestricted research grant from the STERIS Corporation (D.V.D.), NHGRI intramural funding (J.A.S.), and the NIH Director's Challenge Fund (J.A.S.).
The contents of this report are solely our responsibility and do not necessarily represent the official views of the NIH.
We thank all of the patients who are described in this paper and their families.


Spellberg B, Blaser M, Guidos RJ, Boucher HW, Bradley JS, Eisenstein BI, Gerding D, Lynfield R, Reller LB, Rex J, Schwartz D, Septimus E, Tenover FC, and Gilbert DN. 2011. Combating antimicrobial resistance: policy recommendations to save lives. Clin. Infect. Dis. 52(Suppl 5):S397–428.
Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell GD, Dean NC, Dowell SF, File TM Jr, Musher DM, Niederman MS, Torres A, and Whitney CG. 2007. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin. Infect. Dis. 44(Suppl 2):S27–72.
American Thoracic Society, Infectious Diseases Society of America. 2005. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am. J. Respir. Crit. Care Med. 171:388–416.
Horan TC, Andrus M, and Dudeck MA. 2008. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am. J. Infect. Control 36:309–332.
Charlson ME, Pompei P, Ales KL, and MacKenzie CR. 1987. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J. Chronic Dis. 40:373–383.
Chow JW and Yu VL. 1999. Combination antibiotic therapy versus monotherapy for gram-negative bacteraemia: a commentary. Int. J. Antimicrob. Agents 11:7–12.
Neuner EA, Yeh JY, Hall GS, Sekeres J, Endimiani A, Bonomo RA, Shrestha NK, Fraser TG, and van Duin D. 2011. Treatment and outcomes in carbapenem-resistant Klebsiella pneumoniae bloodstream infections. Diagn. Microbiol. Infect. Dis. 69:357–362.
Lascols C, Hackel M, Marshall SH, Hujer AM, Bouchillon S, Badal R, Hoban D, and Bonomo RA. 2011. Increasing prevalence and dissemination of NDM-1 metallo-beta-lactamase in India: data from the SMART study (2009). J. Antimicrob. Chemother. 66:1992–1997.
Viau RA, Hujer AM, Marshall SH, Perez F, Hujer KM, Briceno DF, Dul M, Jacobs MR, Grossberg R, Toltzis P, and Bonomo RA. 2012. “Silent” dissemination of Klebsiella pneumoniae isolates bearing K. pneumoniae carbapenemase in a long-term care facility for children and young adults in Northeast Ohio. Clin. Infect. Dis. 54:1314–1321.
Endimiani A, Hujer AM, Perez F, Bethel CR, Hujer KM, Kroeger J, Oethinger M, Paterson DL, Adams MD, Jacobs MR, Diekema DJ, Hall GS, Jenkins SG, Rice LB, Tenover FC, and Bonomo RA. 2009. Characterization of blaKPC-containing Klebsiella pneumoniae isolates detected in different institutions in the eastern USA. J. Antimicrob. Chemother. 63:427–437.
Diancourt L, Passet V, Verhoef J, Grimont PA, and Brisse S. 2005. Multilocus sequence typing of Klebsiella pneumoniae nosocomial isolates. J. Clin. Microbiol. 43:4178–4182.
Chin CS, Alexander DH, Marks P, Klammer AA, Drake J, Heiner C, Clum A, Copeland A, Huddleston J, Eichler EE, Turner SW, and Korlach J. 2013. Nonhybrid, finished microbial genome assemblies from long-read SMRT sequencing data. Nat. Methods 10:563–569.
Koren S, Harhay GP, Smith TP, Bono JL, Harhay DM, McVey SD, Radune D, Bergman NH, and Phillippy AM. 2013. Reducing assembly complexity of microbial genomes with single-molecule sequencing. Genome Biol. 14:R101.
Endimiani A, Perez F, Bajaksouzian S, Windau AR, Good CE, Choudhary Y, Hujer AM, Bethel CR, Bonomo RA, and Jacobs MR. 2010. Evaluation of updated interpretative criteria for categorizing Klebsiella pneumoniae with reduced carbapenem susceptibility. J. Clin. Microbiol. 48:4417–4425.
Dortet L, Nordmann P, and Poirel L. 2012. Association of the emerging carbapenemase NDM-1 with a bleomycin resistance protein in Enterobacteriaceae and Acinetobacter baumannii. Antimicrob. Agents Chemother. 56:1693–1697.
Ben-David D, Masarwa S, Navon-Venezia S, Mishali H, Fridental I, Rubinovitch B, Smollan G, Carmeli Y, and Schwaber MJ. 2011. Carbapenem-resistant Klebsiella pneumoniae in post-acute-care facilities in Israel. Infect. Control Hosp. Epidemiol. 32:845–853.
Endimiani A, Depasquale JM, Forero S, Perez F, Hujer AM, Roberts-Pollack D, Fiorella PD, Pickens N, Kitchel B, Casiano-Colon AE, Tenover FC, and Bonomo RA. 2009. Emergence of blaKPC-containing Klebsiella pneumoniae in a long-term acute care hospital: a new challenge to our healthcare system. J. Antimicrob. Chemother. 64:1102–1110.
Prabaker K, Lin MY, McNally M, Cherabuddi K, Ahmed S, Norris A, Lolans K, Odeh R, Chundi V, Weinstein RA, and Hayden MK. 2012. Transfer from high-acuity long-term care facilities is associated with carriage of Klebsiella pneumoniae carbapenemase-producing Enterobacteriaceae: a multihospital study. Infect. Control Hosp. Epidemiol. 33:1193–1199.
Won SY, Munoz-Price LS, Lolans K, Hota B, Weinstein RA, and Hayden MK. 2011. Emergence and rapid regional spread of Klebsiella pneumoniae carbapenemase-producing Enterobacteriaceae. Clin. Infect. Dis. 53:532–540.
Kitchel B, Rasheed JK, Patel JB, Srinivasan A, Navon-Venezia S, Carmeli Y, Brolund A, and Giske CG. 2009. Molecular epidemiology of KPC-producing Klebsiella pneumoniae isolates in the United States: clonal expansion of multilocus sequence type 258. Antimicrob. Agents Chemother. 53:3365–3370.
Kumarasamy KK, Toleman MA, Walsh TR, Bagaria J, Butt F, Balakrishnan R, Chaudhary U, Doumith M, Giske CG, Irfan S, Krishnan P, Kumar AV, Maharjan S, Mushtaq S, Noorie T, Paterson DL, Pearson A, Perry C, Pike R, Rao B, Ray U, Sarma JB, Sharma M, Sheridan E, Thirunarayan MA, Turton J, Upadhyay S, Warner M, Welfare W, Livermore DM, and Woodford N. 2010. Emergence of a new antibiotic resistance mechanism in India, Pakistan, and the UK: a molecular, biological, and epidemiological study. Lancet Infect. Dis. 10:597–602.
Centers for Disease Control and Prevention (CDC) 2010. Detection of Enterobacteriaceae isolates carrying metallo-beta-lactamase—United States, 2010. MMWR Morb. Mortal. Wkly. Rep. 59:750.
Tumbarello M, Viale P, Viscoli C, Trecarichi EM, Tumietto F, Marchese A, Spanu T, Ambretti S, Ginocchio F, Cristini F, Losito AR, Tedeschi S, Cauda R, and Bassetti M. 2012. Predictors of mortality in bloodstream infections caused by Klebsiella pneumoniae carbapenemase-producing K. pneumoniae: importance of combination therapy. Clin. Infect. Dis. 55:943–950.
Qureshi ZA, Paterson DL, Potoski BA, Kilayko MC, Sandovsky G, Sordillo E, Polsky B, Adams-Haduch JM, and Doi Y. 2012. Treatment outcome of bacteremia due to KPC-producing Klebsiella pneumoniae: superiority of combination antimicrobial regimens. Antimicrob. Agents Chemother. 56:2108-2113.

Information & Contributors


Published In

cover image Antimicrobial Agents and Chemotherapy
Antimicrobial Agents and Chemotherapy
Volume 58Number 7July 2014
Pages: 4035 - 4041
PubMed: 24798270


Received: 23 February 2014
Returned for modification: 2 April 2014
Accepted: 25 April 2014
Published online: 12 June 2014


Request permissions for this article.



David van Duin
Division of Infectious Diseases, University of North Carolina, Chapel Hill, North Carolina, USA
Federico Perez
Research Service, Louis Stokes Cleveland Department of Veterans Affairs Medical Center, Cleveland, Ohio, USA
Division of Infectious Diseases and HIV Medicine, Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA
Susan D. Rudin
Research Service, Louis Stokes Cleveland Department of Veterans Affairs Medical Center, Cleveland, Ohio, USA
Eric Cober
Department of Infectious Diseases, Cleveland Clinic, Cleveland, Ohio, USA
Jennifer Hanrahan
Department of Medicine, MetroHealth Medical Center, Cleveland, Ohio, USA
Julie Ziegler
Department of Medicine, MetroHealth Medical Center, Cleveland, Ohio, USA
Raymond Webber
Division of Infectious Diseases and HIV Medicine, Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA
Jacqueline Fox
Medicine Institute, Cleveland Clinic, Cleveland, Ohio, USA
Pamela Mason
Medicine Institute, Cleveland Clinic, Cleveland, Ohio, USA
Sandra S. Richter
Department of Microbiology, Cleveland Clinic, Cleveland, Ohio, USA
Marianne Cline
Department of Microbiology, Cleveland Clinic, Cleveland, Ohio, USA
Geraldine S. Hall
Department of Microbiology, Cleveland Clinic, Cleveland, Ohio, USA
Keith S. Kaye
Division of Infectious Diseases, Detroit Medical Center, Wayne State University, Detroit, Michigan, USA
Michael R. Jacobs
Department of Pathology, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA
Robert C. Kalayjian
Department of Medicine, MetroHealth Medical Center, Cleveland, Ohio, USA
Robert A. Salata
Division of Infectious Diseases and HIV Medicine, Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA
Julia A. Segre
National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland, USA
Sean Conlan
National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland, USA
Scott Evans
Department of Biostatistics and Center for Biostatistics in AIDS Research, Harvard School of Public Health, Boston, Massachusetts, USA
Vance G. Fowler Jr.
Division of Infectious Diseases, Duke University Medical Center, Durham, North Carolina, USA
Robert A. Bonomo
Research Service, Louis Stokes Cleveland Department of Veterans Affairs Medical Center, Cleveland, Ohio, USA
Division of Infectious Diseases and HIV Medicine, Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA
Department of Pathology, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA
Department of Pharmacology, Case Western Reserve University School of Medicine, Cleveland, Ohio, USA


Address correspondence to David van Duin, [email protected], or Robert A. Bonomo, [email protected].

Metrics & Citations



  • For recently published articles, the TOTAL download count will appear as zero until a new month starts.
  • There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
  • Citation counts come from the Crossref Cited by service.


If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

View Options

Figures and Media






Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy