The purpose of this study is to review patients with E. coli infections at the University of Pittsburgh Medical Center (UPMC) from September 1, 2006 to August 31, 2007 to determine if these infections have arisen in the community rather than in hospitals or nursing homes. The occurrence of such resistant isolates could be devastating if they were associated with bloodstream infection, such as sometimes accompanies a urinary tract infection, since antibiotic resistant E. coli is not suspected in isolates coming from the community. Therefore, the aims of this study are to:

1. Review whether extended-spectrum beta-lactamase (ESBL)-producing organisms (E. coli, Klebsiella species and Proteus species) are hospital-acquired, healthcare-associated, or community-associated.
2. Investigate the prevalence of ESBL-positive E. coli in foodstuffs (beef, poultry, turkey, and pork), as a potential source for ESBL producers in the community.
3. Compare the clonal relationship of the genome and resistance plasmids carried by the ESBL-producing isolates (healthcare-associated, community-associated, and of animal origin) and study the associations of ESBL-mediated resistance with resistance to other classes of antimicrobials.

The following variables will be followed: age, sex, occupation, hospital location at the time of positive culture (ER, medical ward, ICU, etc.), prior hospitalization or nursing home admission, receipt of outpatient dialysis, home care or other regular medical care (eg, outpatient chemotherapy), presence of invasive devices, receipt of antibiotics, including their type and whether they were adequate for the resistance profile of the organism, and prior positive microbiologic cultures. All sites are collecting the following information that was collected as part of the patient's clinical treatment. The following variables will be collected at all sites: time and location of positive cultures, underlying diseases and severity of illness, presence of urinary or intravascular devices, recent immunomodulative therapies or radiation therapy, physical exam findings, laboratory and radiographical data, antimicrobial usage within 30 days of onset of the infection, microbiological data and resistance patterns, choice of antibiotics once the organism is identified, bacteriological outcomes, laboratory results, demographic information, medications, clinical outcome, gender, height, weight, ethnicity, and past medical history. We will collect information prospectively for one year. The bacteria in the patient's cultures will be subcultured (after the diagnosis has been obtained since the microbiology lab would otherwise destroy the culture) and provided to the honest broker who will de-identify the specimen and link it to the medical information collected (which will also be de-identified by the honest broker). The following evaluation will be performed on these samples. The minimal inhibitory concentration of the antibiotic used in treatment will be performed by the E-Test method (AB Biodisk, Solna, Sweden). Specific mechanisms of antimicrobial resistance will be studied with emphasis on the presence of beta-lactamases produced by the bacteria. This evaluation will be performed by analytical isoelectric focusing techniques as well as PCR and gene sequencing analysis to determine genes encoding beta-lactamases. Biologic samples will be under the control of the principal investigator of this research project. All samples provided to the investigators are de-identified by the honest broker and will be coded with numbers. The information linking these code numbers to the corresponding subjects' identities will be kept in a separate, secure location that only the honest broker has access to. The investigators on this study will keep the samples indefinitely. Samples will be kept in the investigator's laboratory located in Scaife Hall, room 812, 3500 Terrace Street. At no time will the research investigators have access to any patient identifiers. Medical record information and bacteria samples from facilities outside of UPMC will be provided to the PI de-identified. At no time will anyone on the research team have access to patient identifiers. All information and bacteria are collected as part of the patient's clinical treatment.

Inclusion Criteria:

• All patients with E. coli, Klebsiella species, and Proteus species infections during the time period of the study will be reviewed. A community-associated E. coli isolate will be defined as one which was recovered from a clinical culture from a patient at UPMC who had no established risk factors for infection with an antibiotic resistant organism. Established risk factors are defined as:

  • Isolation of the organism two or more days after admission for hospitalization OR
  • History of hospitalization, surgery, dialysis, or residence in a long-term care facility within one year before the culture date OR
  • The presence of an indwelling catheter or percutaneous medical device (eg, tracheostomy tube, gastrostomy tube, or Foley catheter) at the time of the culture OR
  • Previous isolation of an antibiotic resistant organism.

Exclusion Criteria:

• None

ESBL infections