Fluoroquinolones as an alternative treatment for Klebsiella pneumoniae liver abscess and impact on hospital length of stay.

• Fluoroquinolones had similar clinical efficacy to β-lactams in Klebsiella pneumoniae liver abscess (KPLA). • KPLA patients treated with fluoroquinolones had a shorter duration of intravenous antibiotics. • KPLA patients treated with fluoroquinolones had a shorter hospital length of stay. Klebsiella pneumoniae liver abscess (KPLA) is an endemic disease in East Asia. Patients with KPLA usually require prolonged intravenous (i.v.) β-lactam therapy and hospitalisation. Fluoroquinolones have high oral bioavailability and the potential to shorten the duration of i.v. therapy. The aim of this study was to investigate the feasibility of fluoroquinolones as an alternative treatment for KPLA in Taiwan. Consecutive patients with KPLA in a medical centre in Taiwan between July 2012 and August 2019 were retrospectively enrolled. Clinical characteristics and outcomes were compared between cases treated with β-lactams and fluoroquinolones. A multivariate logistic regression model and propensity-score adjusted analysis were performed to identify independent risk factors for prolonged hospitalisation. A total of 234 patients with KPLA were identified during the study period. Most patients received β-lactams (n = 199; 85.0%), whilst only 35 (15.0%) received fluoroquinolones as the major therapy. Fluoroquinolones had similar clinical efficacy to β-lactams even in critically ill patients. Patients treated with fluoroquinolones had a shorter i.v. antibiotics duration (18.9 ± 7.6 days vs. 28.5 ± 14.7 days; P < 0.001) and hospital length of stay (LOS) (20.9 ± 8.3 days vs. 29.5 ± 16.2 days; P < 0.001) than patients treated with β-lactams. Major therapy with fluoroquinolones was an independent protective factor for hospital LOS > 14 days in all patients and for hospital LOS > 21 days in critically ill patients. In conclusion, fluoroquinolones were an effective alternative treatment for KPLA that resulted in a shorter duration of i.v. therapy and hospital LOS. [ABSTRACT FROM AUTHOR]