Kentaro Kikuchi, Chiyoko Motegi, Sayuri Osaki, Kotaro Matsumoto, Hiromichi Tsunashima, Tomohiro Kikuyama, Hikari Fujioka, Juri Kubota, Kozue Nagumo, Sho Ohyatsu, Tomoyuki Nariyama, Minoru Yoshida
We investigated the clinical characteristics of 30 inpatients with ESBL-producing Escherichia col ibacteremia (ESBL group) and 85 inpatients with non-ESBL-producing E. coli bacteremia (non-ESBL group) and evaluated the relationship between empiric therapy and prognosis. In the ESBL group, urinary tract infection was most frequently. Most patients were hospitalized, had a history of admission to geriatric health care facilities, or intravenous antimicrobial drug injection. The rate of patients to whom susceptible antimicrobial drugs were administered as empiric therapy was significantly lower in the ESBL group than non-ESBL group (50 vs. 100%, p=0.0001). The total antimicrobial drug administration period in the ESBL group with ineffective empiric therapy was significantly longer than non-ESBL group (14.1 ± 3.1 vs. 9.9 ± 3.7 days, p=0.03). The mortality rate in the ESBL group with ineffective empiric therapy was significantly higher than ESBL group with effective empiric therapy and non-ESBL group (26.7% vs. 0%, 8.2%, p<0.05). In conclusion, when treating patients with bacteremia of E. coli due to urinary tract infection, with a history of admission to geriatric health care facilities, admission to hospitals, or intravenous antimicrobial drug injection, empiric therapy under the assumption of ESBL-producing E. coli may be shortened the total antimicrobial drug administration period and reducing mortality.