Clinical effectiveness of beta-lactams versus fluoroquinolones as empirical therapy in patients with diabetes mellitus hospitalized for urinary tract infections: A retrospective cohort study.

<h4>Background</h4>Diabetic patients are at risk of severe urinary tract infections (UTIs). Due to the emerging resistance rates to fluoroquinolones and β-lactams, we aimed to evaluate the effectiveness of β-lactams versus fluoroquinolones as empirical therapy for diabetic patients hospitalized for UTIs.<h4>Methods</h4>A retrospective cohort study was conducted in a medical center in Taiwan between 2016 and 2018. Patients with type 2 diabetes, aged ≥20 and hospitalized for UTIs were enrolled. Patients with UTI diagnosis within one year before the admission, co-infections at the admission, or ≥2 pathogens in the urine cultures were excluded. The primary outcome was empiric treatment failure.<h4>Results</h4>298 patients were followed for at least 30 days after the admission. Escherichia coli (61.07%) was the most common pathogen. The resistance rates of the pathogens to levofloxacin were 28.52% and 34.22% according to the historical Clinical and Laboratory Standards Institute (CLSI) breakpoints and the updated 2019 CLSI breakpoints, respectively. The resistance rates of ceftazidime and cefepime were 21.81% and 11.41%, respectively. Empirical β-lactams were associated with less treatment failure compared to fluoroquinolones (adjusted OR = 0.32, 95% CI = 0.17-0.60). Beta-lactams were associated with less treatment failure than fluoroquinolones when appropriatness was determined by the pre-2019 CLSI breakpoints but not the 2019 CLSI breakpoints.<h4>Conclusions</h4>In diabetic patients hospitalized for UTIs, β-lactams were associated with less empiric treatment failure compared to fluoroquinolones when the resistance rate to fluoroquinolone is higher than β-lactams. The updated 2019 CLSI breakpoint for fluoroquinolone was better than pre-2019 CLSI breakpoints to correlate with treatment outcomes for hospitalized UTIs in diabetic patients.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.