Antimicrobial treatment of monomicrobial phenotypic carbapenem-resistant Klebsiella pneumoniae bacteremia: Two are better than one.

Abstract

Infections caused by carbapenem-resistant Klebsiella pneumoniae (CRKP) are emerging worldwide. The optimal treatment for CRKP infections is challenging for clinicians because therapeutic agents are greatly limited. A retrospective study of CRKP monomicrobial bacteremia was conducted at a medical center between 2010 and 2016. The use of at least one or more drugs with invitro activity against the blood isolates was defined as appropriate combination therapy. The logistic regression model and propensity score analysis was used to assess clinical effects of therapeutic strategies. The 30-day crude mortality was the primary end point. Two hundred and three patients were eligible and the 30-day mortality rate was 37.9% (77 patients). As compared with monotherapy, empirical (11.6 vs. 57.3%, p<.001) or definitive (26.5% vs. 48.6%, p=.001) combination antibiotic therapy showed a lower 30-day mortality rate independently. The propensity score analysis showed that those receiving combination therapy had less clinical (p≤.001) or microbiological failure (p=.003) and a lower 30-day mortality rate (p<.001). Among various regimens of definitive therapy, the 30-day mortality rate was the lowest among patients with appropriate combination therapy 23.6%, (p<.001; by log rank test). The primary outcome was similar in those with definitive carbapenem-containing and carbapenem-sparing combination regimens (p=.81). The presence or absence of carbapenemase production did not affect the mortality rate (p=.26). Combination therapy, regardless of carbapenem-containing or carbapenem-sparing regimens, was associated with a favorable outcome.