Clonal Dissemination of Klebsiella pneumoniae Carbapenemase KPC-3 in Long Beach, California

Carbapenems (CARBs) are the primary treatment strategy for serious infections caused by extended-spectrum beta-lactamase (ESBL)-producing Klebsiella pneumoniae. In a report by the National Healthcare Safety Network, 5% of K. pneumoniae isolates collected in the United States between January 2006 and October 2007 were resistant to CARBs (4). First detected in North Carolina in 1996, production of the KPC β-lactamase, a serine carbapenemase, in K. pneumoniae has disseminated in many states across the United States and is no longer confined to the northeast regions (5). Furthermore, KPC-producing strains are now emerging worldwide, affecting Israel, Asia, Europe, Canada, and Central and South America (3, 8, 9). The genes encoding KPC enzymes are usually located on plasmids and embedded on transposon Tn4401, providing impetus for its rapid dissemination (7). In this report, we describe the clinical and molecular characteristics of four infections caused by KPC-3-producing K. pneumoniae strains.

We collected potential ESBL-producing isolates at 12 hospitals located throughout southern California between 1 August 2007 and 31 March 2009. Initial ESBL screening was performed using each institution's standard of practice. Isolates were included in the study according to initial ertapenem susceptibility results as determined by the Etest (AB bioMérieux, Marcy l'Etoile, France), and those displaying ertapenem MICs that were ≥8 μg/ml were selected for further testing at an independent laboratory. KPC-producing strains were obtained at one study site, Long Beach Memorial Medical Center, which is a nonprofit, community-based, teaching hospital with 462 beds for adult patients. Institutional review board approval was obtained to collect patient data.

ESBL and carbapenemase production confirmatory tests (using a modified Hodge test [MHT]) were performed according to CLSI guidelines. Genes encoding ESBL, plasmidic AmpC, and carbapenemase (including serine- and metallo-β-lactamases) were detected using multiplex PCRs (1, 2, 6, 10). TEM- and SHV-encoding genes were amplified in separate reactions. All amplicons were sequenced, and nucleotide and deduced amino acid sequences were analyzed using the Lasergene software (DNAStar, Madison, WI) and compared with available sequences through BLAST (http://www.ncbi.nlm.nih.gov/blast/ ). Isolates were typed by pulsed-field gel electrophoresis. Genomic DNA was prepared in agarose blocks and digested with SpeI (New England Biolabs, Beverly, MA). Electrophoresis was performed on a CHEF-DR III apparatus (Bio-Rad, Richmond, CA) under the following conditions: 0.5× Tris-borate-EDTA (TBE), 1% agarose, 13°C, and 200 V for 24 h, with the switch time ramped from 5 to 90 s.

Of 232 Enterobacteriaceae isolates collected, 6 (2.5%) expressed ertapenem resistance. Among those strains, four (1.7%) K. pneumoniae isolates, which also displayed MICs of >8 μg/ml for imipenem and meropenem, were positive for carbapenemase production by MHT and bla _KPC by PCR. The remaining strains, one E. coli and one K. pneumoniae strain, were negative for carbapenemases in phenotypic and genotypic tests.

All four strains carried bla _KPC-3, and further PCRs showed that all strains carried genes encoding TEM-1 and SHV-11/SHV-36. Isolates were negative for all other beta-lactamase-encoding genes tested. These strains were resistant to CARBs, amikacin, aztreonam, piperacillin-tazobactam, cefepime, levofloxacin, and trimethoprim-sulfamethoxazole but susceptible to gentamicin (4 μg/ml), the polymyxins (≤0.5 μg/ml), and tigecycline (0.5 to 1.0 μg/ml). Molecular typing demonstrated that all strains were genetically related.

The clinical characteristics of patients infected with these isolates are described in Table 1. All patients (females with a mean age of 73.3 ± 16.5 years) presented with sepsis and transferred from two different long-term care facilities (LTCFs). Infections were separated by a mean of 26.7 ± 13.6 days. Most infections occurred at hospital admission, except with one patient, who developed infection on day 20 of hospitalization. Three patients had diabetes mellitus and had been hospitalized within the previous 9 months. Interestingly, although all KPC-producing strains were genetically related, it appears that patient 3 developed infection independently since her infection occurred at hospital admission, she was transferred from a different LTCF, and her prior hospitalization preceded our first case of infection with a KPC-producing strain. All other patients were related by medical care in a similar LTCF or hospital unit.

All patients experienced treatment failure at 72 h after initiation of empirical antibiotics, which were subsequently changed when culture and susceptibility results became available. Treatment failure was defined as an absence of clinical improvement, persistent positive cultures, a change in initial antibiotic treatment, or death within 14 days of the first positive culture. Microbiologic and clinical success was documented for all patients by the end of hospitalization. The mean duration of appropriate antibiotic therapy while patients were hospitalized was 8.8 ± 2.2 days. Three patients continued intravenous antibiotics after hospital discharge. The mean lengths of stay in the hospital and intensive care unit were 13.5 ± 9.3 and 4.3 ± 5.3 days, respectively. All patients survived their infections.

KPC-3-producing isolates have been previously documented in California (5); however, here we identified the clonal dissemination of KPC-3 producers in a southern California hospital recovered during a study evaluating clinical outcomes. The clinical information, although generated from a limited number of strains, supports our understanding of infections caused by KPC-producing isolates. With our limited therapeutic options, coupled with the escalating dissemination of CARB-resistant K. pneumoniae, vigilant surveillance, prompt detection, and implementation of effective infection control measures in hospitals and LTCFs are imperative.

• Copyright © 2010 American Society for Microbiology