306. Can We Use Data from Escherichia coli Strains Isolated One Month prior to an Escherichia coli Urinary Tract Infection to Guide Empiric Antibiotic Treatment?
Session: Poster Abstract Session: Antimicrobial Therapy: Clinical Studies
Friday, October 21, 2011
Room: Poster Hall B1
  • Escherichia.coli.predictive.for.urinary.tract.infection.pdf (226.7 kB)
  • Background: We investigated whether E. coli strains isolated in the month preceding an E. coli urinary tract infection (UTI) were predictive for the UTI strain in women with recurrent UTIs.

    Methods: We used E. coli strains that originated from the NAPRUTI study: two randomized controlled trials comparing cranberries (in premenopausal women) or lactobacilli (in postmenopausal women) with trimethoprim/sulfamethoxazole (TMP/SMX) in the prevention of recurrent UTI. During 15 months patients monthly collected urinary samples. When patients experienced symptoms indicating a UTI, an additional sample was collected. Urine was cultured and resistance patterns were determined in uropathogens. We randomly selected 47 patients that experienced an E. coli UTI and from whom an E. coli was isolated in the month prior to this UTI. Pulsed-field gel-electroforesis (PFGE) was performed on the E. coli strains. Results were stratified according to menopausal status, presence of complicating host factors and study arm.

    Results: 160 of the 445 patients (36%) experienced an E. coli UTI. In the month prior to this UTI, E. coli was isolated in 76 patients (48%). Of the 47 patients selected for PFGE, 33 (70%) showed similar E. coli strains at both time points. No significant differences were observed between pre- and postmenopausal women (76% vs 67% respectively), between groups with and without complicating host factors (67% vs 72%) or between the study arms: 86% (cranberries) vs 72% (TMP/SMX) vs 63% (lactobacilli). 83% (39/47) of the E. coli isolates had similar antimicrobial susceptibility patterns. In two patients with similar E. coli PFGE-results, susceptibility patterns differed. On the other hand, different PFGE-results showed similar susceptibility patterns in 8 patients.

    Conclusion: Seventy  percent of the E. coli strains isolated in the month prior to an E. coli UTI had similar PFGE-results, independent of patient characteristics and received intervention. Eighty-three percent of E. coli strains showed similar susceptibility patterns. Hence, data from E. coli strains isolated in the month prior to a symptomatic E. coli UTI can be used to make informed choices for empiric antibiotic treatment.

    Subject Category: B. Bacterial pathogenesis, studies in animal models, molecular pathogenicity

    Marielle Beerepoot, MD1, Casper den Heijer, MD2, Suzanne Geerlings, MD, PhD1, Jan M. Prins, Professor, MD3 and Ellen Stobberingh, PhD2, (1)Department of Internal Medicine, Division of Infectious Diseases, Tropical Medicine and AIDS, and Center for Infection and Immunity Amsterdam (CINIMA), Academic Medical Center, Amsterdam, Netherlands, (2)Department of Medical Microbiology, Maastricht University Medical Center, Maastricht, Netherlands, (3)Internal Med., Infectious Diseases, Tropical Med. and AIDS, Ctr. for Infection and Immunity Amsterdam (CINIMA), Academic Med. Ctr., Amsterdam, Netherlands


    M. Beerepoot, None

    C. den Heijer, None

    S. Geerlings, BMS, AstraZeneca, Astellas: Scientific Advisor, Consulting fee

    J. M. Prins, None

    E. Stobberingh, None

    Findings in the abstracts are embargoed until 12:01 a.m. EST Thursday, Oct. 20 with the exception of research findings presented at IDSA press conferences.