Probable intrafamily transmission of a highly virulent CTX-M-3-producing Escherichia coli belonging to the emerging phylogenetic subgroup D2 O102-ST405 clone

Sir, Over the last decade, an explosive spread of CTX-M-type extended-spectrum β-lactamases (ESBLs) in Escherichia coli has occurred, both in hospital and community settings. Two phenomena may explain such an epidemic profile: the spread within bacterial strains of the plasmids bearing the antibiotic resistance genes and the spread of bacterial clones bearing the resistance-encoding plasmids. The O25b-ST131 clone (where ST stands for sequence type), belonging to the B2 phylogenetic group, has disseminated all over the world.1 Moreover, this clone, which is highly virulent, can capture a large panel of ESBL genes.1–3 The emergence of such clones constitutes a major concern for public health. We report two members of the same family successively admitted to hospital for a febrile urinary tract infection (UTI) caused by a new emerging ESBL-producing E. coli clone. Patient 1, a diabetic woman in her early 60s, was admitted to hospital for a pyelonephritis unsuccessfully treated with norfloxacin. She was obese and had a history of bariatric surgery 2 years ago. Therapy was switched to a combination of cefotaxime and amikacin. A urine culture taken at admission and screening for faecal carriage revealed an ESBL-positive E. coli resistant to ciprofloxacin and trimethoprim/sulfamethoxazole, but susceptible to cefoxitin, carbapenems, nitrofurantoin and aminoglycosides. The treatment was changed to ertapenem and the patient was managed by outpatient parenteral antimicrobial treatment for 2 weeks. She fully recovered and further urine cultures were sterile. Patient 2, the son of Patient 1 who was in his early 40s, was admitted 10 days after for a febrile prostatitis. He was obese and had been admitted to another hospital 2 months before for the onset of type 2 diabetes. A urine culture and screening for faecal carriage identified an ESBL-positive E. coli with the same antibiotype pattern as the isolate from the mother. The patient was initially prescribed ofloxacin, but this was changed to ertapenem. The patient was discharged on day 6, with 3 weeks of home parenteral antibiotic therapy. He fully recovered and subsequent urine cultures were sterile. The son lived independently, but had regularly visited his mother for dinner during the preceding 6 months and had used her toilet. Neither patient had domestic pets. A thorough analysis of one urine isolate and one faecal isolate from each patient was performed. The ESBL-producing isolates from the two patients were indistinguishable using enterobacterial repetitive intergenic consensus (ERIC)-1 and ERIC-2 PCR (Figure 1) and PFGE (data not shown) methods. The presence of blaCTX-M-3 was revealed by PCR and sequencing.4 Phylogenetic analysis by the PCR triplex method and multilocus sequence typing (MLST) using the Pasteur Institute scheme showed that these isolates belonged to the D2 phylogenetic subgroup I (ST221), corresponding to ST405 of the Achtman MLST scheme.4–6 The isolates exhibited the O102 type determined by allele-specific PCR, and harboured the virulence genes aer, hlyC, iha, fuyA, traT, sat and papGII, indicating the presence of the pathogenicity islands PAI ICFT073 and HPI.2,4 These similarities indicated clearly that the four isolates belonged to the same clone. When tested in a mouse model of septicaemia, the mother's urine isolate was highly virulent, killing all 10 of the inoculated mice, as did a classical extraintestinal virulent strain of E. coli.2 Typing of E. coli isolates using ERIC-1 PCR (a) and ERIC-2 PCR (b). λ, DNA molecular weight marker, Euroladder (Eurobio); lanes 1 and 2, E. coli isolates from urine and rectum of patient 1; lanes 3 and 4, E. coli isolates from urine and rectum of patient 2. Intrafamily transmissions of ESBL-positive E. coli responsible for UTIs are rarely described. A case of transmission between two family members of a CTX-M-15 ESBL-positive, E. coli O25b-ST131 strain was recently reported.7 We describe here two members of the same family successively admitted to hospital for a febrile UTI due to a CTX-M-3 ESBL-positive E. coli strain belonging to the D2 phylogenetic subgroup I-O102-ST405. Potential mechanisms for the observed strain sharing include host-to-host transmission and acquisition from an external source, such as a food supply or domestic pets. We cannot rule out a common source as we did not investigate the presence of ESBL-positive E. coli in food, but the temporal pattern of their UTIs suggests a transmission from the mother to the son and the development of the infection from the commensal faecal reservoir. Neither patient had a history of prior antimicrobial use and, despite one hospitalization for each patient in the past 2 years, independent acquisition of an ESBL strain from distinct sources (especially of the same pulsotype) seems unlikely. Although both patients had diabetes, an underlying risk factor for complications of UTIs, neither of them had a previous history of febrile UTI. The strain appears to be particularly virulent, as inferred from two lines of evidence, namely the number and types of virulence genes (seven genes coding for adhesins, siderophores, toxins and protectins) and the mouse lethality assay. Recent studies indicate that the present D2 O102-ST405 E. coli is an emerging clone, responsible, like the O25b-ST131 clone, for the worldwide spread of blaCTX-M genes.8,9 This clone appears to pose the double threat of multidrug resistance and substantial extraintestinal virulence, in addition to its colonization and transmission ability. This makes its emergence and dissemination particularly concerning. New approaches to prevention, detection and management will be needed for this and similar clones. This study was carried out as part of our routine work and partly funded by a grant from the Assistance Publique-Hôpitaux de Paris (Contrat d'Initiation à la Recherche Clinique 05103). None to declare.