Abstract
We reported a woman with urinary-tract infection caused by OXA-48 producing Klebsiella pneumoniae. Using molecular techniques, we showed that she might acquire this bacterium from another family member who lived in the same house. The two isolates, although different by pulsed-field gel electrophoresis and multilocus sequence typing, carried blaOXA-48 in a similar IncL/M plasmid. This case report shows that community spreading of carbapenemase producing Enterobacteriaceae occurs in the low endemic area not only in nosocomial setting but also in the community.
Introduction
C
Patient and Methods
Case report
An 18-year-old previously healthy woman presented to the emergency department in June 2014 with 1 day history of right abdominal pain and a burning sensation when urinating. On physical examination, she had normal blood pressure and no fever. Laboratory evaluation revealed the C-reactive protein level of 9.9 mg/L (normal range: <5) and slightly increased the leukocyte (11.4 × 103/mm3) and neutrophil count (7.3 × 103/mm3). Leukocytes (614/μl) and erythrocytes (283/μl) were found in one urine specimen. The diagnosis of uncomplicated urinary-tract infection (UTI) was made, and she was discharged with ciprofloxacin 500 mg once daily, planned for 5 days. In the culture of the urine, K. pneumoniae and Enterococcus faecalis were found in significant amount. K. pneumoniae showed reduced susceptibility to meropenem (disk diffusion zone diameter of 21 mm, minimum inhibitory concentration [MIC] of 0.5 mg/L) and was resistant to piperacillin/tazobactam and to the temocillin disk (both with diffusion zone diameter of 9 mm) according to EUCAST breakpoints version 5.0 (www.eucast.org). Moreover, this organism was resistant to trimethoprim–sulfamethoxazole and to cephalosporins such as cefotaxime (Table 1). The presence of CPE was suspected. Due to the finding of CPE, 4 days after her first visit, the patient was requested to be admitted in hospital. Despite the treatment with ciprofloxacin, it was noticed that she remained with symptoms. The treatment was switched from ciprofloxacin to meropenem (1 mg thrice daily), and amikacin (500 mg once daily), both given intravenously for 3 days. She showed rapid good evolution and was discharged from hospital after 3 days. The culture of the urine taken at the end of the treatment was negative.
S, susceptible; I, intermediate; R, resistance; according to EUCAST breakpoints version 5.0 (www.eucast.org).
During her hospitalization, further history was obtained. She denied any foreign travel or receiving any antibiotics in the last 3 months. Her anamnesis also revealed that another family member who lived in the same house had been hospitalized for several months for prostatectomy complicated with postoperative K. pneumoniae UTI. This K. pneumoniae also appeared to harbor OXA-48. The antibiotic susceptibility pattern of the isolates from these two people is shown in Table 1.
We could recover the K. pneumoniae isolate from this family member and sent it, together with the isolate from the patient, to the National Reference Centre where the isolates underwent multiplex polymerase chain reaction (PCR) targeting blaVIM, blaIMP, blaNDM, blaKPC, and blaOXA-48 and extended-spectrum β-lactamase (ESBL) genes. 2 Both isolates were confirmed by multiplex PCR to harbor blaOXA-48 gene. The multiplex PCR also showed the presence of CTX-M group 1 in the isolates from the patient and the family member; this explains the resistance to third generation cephalosporins (Table 1). In both isolates, blaOXA-48 was located on a 63-kb pOXA-48 plasmid. Plasmid extract using the Kieser method was transferred by electroporation in an Escherichia coli recipient strain (E. coli TOP10). E. coli transformants of both isolates harbored blaOXA-48 and showed high-level resistance to temocillin (MIC > 256 mg/L), piperacillin–tazobactam (MIC = 128 mg/L), and MIC to meropenem of 0.5 and 1 mg/L, respectively (vs. 0.064 mg/L in the recipient E. coli TOP 10). The isolates were typed by pulsed-field gel electrophoresis (PFGE) 16 and subjected to multilocus sequence typing (MLST) as described previously. 7 The Institut Pasteur database available at http://bigsdb.web.pasteur.fr/klebsiella/klebsiella.html was used for assigning sequence types (STs). The PFGE typing showed that the K. pneumoniae isolate from the patient belongs to type A (<80% homology) and from the family member to type ZU. The MLST showed that both isolates belonged to distinct clonal lineages (ST405 for the patient and ST15 for her family member). These STs have been occurred before in Belgium. 1 Plasmid incompatibility groups determined by a PCR-based replicon typing according to Carattoli et al. 4 and Galimand et al. 8 showed that pOXA-48 plasmid belongs to the IncL/M incompatibility group.
Discussion
The present case report justifies the concern of Nordmann et al. that CPE may spread in the community in the low endemic area. 12 In Belgium, the point prevalence of CPE isolates in Belgian hospitals was reported to be 0.28% in 2012 10 and in 2012–2013; however, around 9% of CPE isolates in Belgium were estimated to originate from ambulatory patients (www.wiv-isp.be/nsih/surv_carba/carbapenemase_nl.asp, last visited March 10, 2015).
In this case report, the K. pneumoniae isolates recovered from the patient and her family member living in the same household belong to two different clonal lineages. Perhaps, the dissemination of blaOXA-48 gene is likely due to the horizontal spread of a 63-kb conjugative IncL/M plasmid. Such finding was also shown in a recent study on the spread of OXA-48 producers in Belgian hospitals. 9 Yet, since the OXA-48-producing K. pneumoniae isolated from the patient and from the household member possessed distinct genotypes (distinct PFGE and MLST); we cannot confirm the household transmission. Noteworthy to mention that the spread of a single 63 kb self-conjugative IncL/M-type plasmid has also been observed among OXA-48-producing Enterobacteriaceae in some European countries with relatively high CPE burden.5,13 This pOXA-48 plasmid is likely to correspond to the common plasmid found harboring OXA-48 with known ability to easily conjugate between strains of Enterobacteriaceae. 11 The household transmission is thus not unlikely in the low endemic setting. In contrast, even if there is no relation between two OXA-48 producing K. pneumoniae isolates, it is likely that they both would harbor this common plasmid. Alternatively, we speculate that the young woman in this case report could have acquired this plasmid, for example, through food chain transmission. Unfortunately, no other household family members' screening was performed to substantiate this latter hypothesis in the present study.
Community-based spread of CPE may prove very difficult to document and to contain. Therefore, the scientific community should prioritize the identification of risk factors for community transmission of CPE and the development of strategies to prevent their dissemination. We believe that reducing global overuse of antibiotics, including in food production and over-the-counter use of antibiotics next to improvement of hand hygiene precaution in primary care and community, should be among these measures.
In conclusion, we present in this study a case report on possible intrahousehold transmission of multidrug-resistant Enterobacteriaeceae that illustrates community spread in the low setting area. Clearly, measures should be taken to cope with this threat.
Footnotes
Acknowledgment
The National Reference Centre for Monitoring of Antimicrobial Resistance in gram-negative bacteria (authors T.D.H. and Y.G.) is supported by the Belgian Ministry of Social Affairs through a fund within the Health Insurance System. No additional funding than for routine work was received by the other authors.
Disclosure Statement
No competing financial interests exist.
