|Home | About | Journals | Submit | Contact Us | Français|
Extended-spectrum-β-lactamase (ESBL)-producing organisms have captured the attention of clinicians and laboratorians and are agents of nosocomial and community onset infections (J. D. Pitout and K. B. Laupland, Lancet Infect. Dis. 8:159-166, 2008). ESBLs in many enterobacteriaceae and in nonfermenting Gram-negative organisms have been described (K. Bush and G. A. Jacoby, Antimicrob. Agents Chemother. 54:969-976, 2010). We present the first case of a clinical isolate of multidrug-resistant Escherichia fergusonii expressing an extended-spectrum-β-lactamase (ESBL).
A 76-year-old Caucasian woman presented to the emergency department of a community hospital following a fall on ice. Investigations indicated she had sustained a fracture of her left hip and was admitted for a hip arthroplasty. A Foley catheter was inserted at the time of surgery and removed 48 h later. Three days after her surgery, the patient complained of urgency and dysuria. Urinalysis indicated the presence of moderate leukocyte esterase and microscopic pyuria. A culture was obtained, and an empirical 3-day treatment with trimethoprim-sulfamethoxazole was initiated for suspected catheter-associated cystitis.
Quantitative urine culture plated to CHROMAgar Orientation (BD, Mississauga, ON, Canada) media revealed >108 CFU/liter of an organism producing Escherichia coli-like pink colonies, which were subcultured for further identification because a brownish discoloration around the colonies suggested the possibility that the organism was not E. coli. Colonies were indole positive and oxidase negative. A Vitek 2 GN card (bioMérieux, Marcy L'étoile, France) was initially used for identification of the organism and provided a satisfactory (96.7%) identification of Escherichia fergusonii. Identification was confirmed using the API 20E strip (bioMérieux, Marcy L'étoile, France), which provided the code 5144113, corresponding to a 98.9% confident identification of E. fergusonii. The isolate was differentiated from E. coli by amygdalin, adonitol, and cellobiose fermentation and inability to ferment melibiose and sorbitol. Antimicrobial susceptibility, as determined by the Vitek 2 AST GN09 card (bioMérieux, Marcy L'étoile, France) and by broth microdilution using the CLSI-recommended method, is presented in Table Table1.1. The organism was multidrug resistant, with demonstrated high-level resistance to cephalosporins, fluoroquinolones, sulfonamides, monobactams, and aminopenicillins. Although the Vitek 2 system did not suggest the possibility of ESBL production, the antimicrobial susceptibilities suggested the ESBL mechanism of resistance. This was investigated using the CLSI disk diffusion method using ceftriaxone and ceftazidime disks with and without clavulanate (AB Biodisk, Solna, Sweden) (4). Clavulanate increased the inhibition zone diameter by 9 mm for ceftriaxone and 15 mm for ceftazidime. Although the clavulanate synergy method is not CLSI approved for this organism, PCR and sequencing of SHV, CTX-M, and TEM β-lactamase genes as described by Mulvey et al. (8) revealed the presence of TEM-1 and SHV-12, the later conferring the ESBL phenotype (2).
Despite in vitro resistance to sulfonamides, the patient's symptoms temporarily improved but recurred a week later. A second culture revealed the same organism. Because intravenous therapy was not indicated, a regimen of amoxicillin-clavulanate, 500 mg three times a day (TID), was given for 7 days. The treatment resulted in resolution of symptoms and microscopic pyuria. Subsequent urine samples a week and a month after treatment showed no evidence of bacteriuria. Retrospectively, the patient indicated that she had returned from a trip to Egypt 2 months previously, where she had experienced a bout of severe traveler's diarrhea which persisted in the form of 5 or 6 watery bowel movements daily for 6 weeks after her return. Although some improvement had been noted recently, loose stools were still present. Although investigations for routine enteric pathogens shortly after her return revealed no typical pathogens, a stool specimen obtained after her hospitalizations showed the presence of large quantities of the ESBL-producing E. fergusonii, suggesting a possible link between her recent travel and colonization with this organism.
E. fergusonii was formerly known by the vernacular name enteric group 10 (EG 10) and received official standing as E. fergusonii in 1985 (5). The organism has been isolated from clinically relevant specimens including those from abdominal abscesses, symptomatic urinary tract infections, infusion-related sepsis, and a patient with diarrhea, as well as animal sources and human stool (5, 11). One case of multidrug-resistant E. fergusonii causing cystitis is reported in the literature and had a similar pattern of antimicrobial resistance but was sensitive to expanded-spectrum cephalosporins and did not have an ESBL phenotype (11). Other isolates for which antimicrobial susceptibility is reported in the literature have variable susceptibility to quinolones, sulfonamides, aminopenicillins, and narrow-spectrum cephalosporins (3). The use of PCR and sequencing of ESBL genes in this case clearly prove the presence of the ESBL SHV-12, which would confer the same clinical and infection control consequences as an E. coli or Klebsiella pneumoniae strain with an ESBL phenotype. This ESBL in a clinically relevant isolate of E. fergusonii demonstrates both the pathogenic potential of this organism and the possibility of ESBL production going unrecognized because laboratories do not routinely screen this species for ESBL production (9). Although the CLSI-recommended screening and confirmation methods for ESBL production are recommended only for E. coli among the genus Escherichia, the method was successful in indentifying the ESBL in this isolate and has been used for Gram-negative fermentative rods other than those recommended by CLSI. Therefore, laboratories performing ESBL screening should consider applying them to all Escherichia species, given that the clinical and infection control implications are likely the same and many Escherichia species have been shown to have pathogenic potential, including E. vulneris, E. fergusonii, and E. hermannii (3, 5).
Evidence that travel may lead to colonization and infection with ESBL-producing E. coli has been shown in both anecdotal (6) and large retrospective (7) studies. In keeping with these observations, it seems likely that our patient became colonized by this pathogen in Egypt and subsequently became infected because of the urinary tract instrumentation involved in her surgery. The role that E. fergusonii may have played in her protracted traveler's diarrhea is unclear, although there have been reports of E. fergusonii causing gastroenteritis (1, 3, 10). It is also possible that there was an alternative cause of her diarrhea, which nevertheless resulted in a significant perineal colonization with the potential uropathogen, which subsequently contributed to her risk of catheter-associated cystitis.
Treatment of urinary tract infections caused by ESBL-producing organisms remains controversial, although both expert opinion and case reports have supported the use of amoxicillin-clavulanate for the treatment of uncomplicated cystitis caused by multidrug-resistant ESBL producers, which are commonly resistant to fluoroquinolones, sulfonamides, and furans (6, 9). This treatment may even provide a cure when in vitro testing shows that the organism is resistant to clavulanate-containing preparations, presumably because high concentrations of both the inhibitor and its companion antibiotic in the urine of patients with normal renal function are sufficient to inhibit the activity of the ESBL and potentiate the accompanying β-lactam (6). This report further substantiates the observation that clavulanate-containing antimicrobials are a viable option for the outpatient management of cystitis caused by ESBL producers.
Published ahead of print on 21 April 2010.