Conversion to colistin susceptibility by tigecycline exposure in colistin-resistant Klebsiella pneumoniae and its implications to combination therapy.

OBJECTIVES: This study investigated the effect of tigecycline exposure on susceptibility of colistin-resistant Klebsiella pneumoniae isolates to colistin and explored the possibility of antibiotic combination at low concentrations to treat colistin-resistant K. pneumoniae isolates. METHODS: Twelve tigecycline-resistant (TIR) mutants were induced in vitro from wild-type, colistin-resistant, and tigecycline-susceptible K. pneumoniae isolates. Antibiotic susceptibility was determined using the broth microdilution method. The deduced amino acid alterations were identified for genes associated with colistin resistance, lipid A biosynthesis, and tigecycline resistance. Expression levels of genes were compared between wild-type stains and TIR mutants using quantitative real-time polymerase chain reaction (PCR). Lipid A modification was explored using MALDI-TOF mass spectrometry. Time-killing assay was performed to assess the efficiency of combination therapy using low concentrations of colistin and tigecycline. RESULTS: All TIR mutants except one were converted to be susceptible to colistin. These TIR mutants had mutations in the ramR gene and increased expression levels of ramA. Three genes associated with lipid A biosynthesis, lpxC, lpxL, and lpxO, were also overexpressed in TIR mutants, although no mutation was observed. Additional polysaccharides found in colistin-resistant, wild-type strains were modified in TIR mutants. Colistin-resistant K. pneumoniae strains were eliminated in vitro by combining tigecycline and colistin at 2 mg/L. In this study, we found that tigecycline exposure resulted in reduced resistance of colistin-resistant K. pneumoniae to colistin. Such an effect was mediated by regulation of lipid A modification involving ramA and lpx genes. CONCLUSION: Because of such reduced resistance, a combination of colistin and tigecycline in low concentrations could effectively eradicate colistin-resistant K. pneumoniae strains.

Inappropriate empirical antibiotic therapy does not adversely affect the clinical outcomes of patients with acute pyelonephritis caused by extended-spectrum ß-lactamase-producing Enterobacteriales.

Extended-spectrum ß-lactamase-producing Enterobacteriales (ESBL-PE) are often associated with inappropriate empirical therapy (IAT). The aim of this study was to investigate whether IAT of acute pyelonephritis (APN) caused by ESBL-PE is related to adverse outcomes. A retrospective cohort study was performed at a tertiary-care hospital from 2014 through 2016. Patients who had APN caused by ESBL-PE and were definitely treated with appropriate antibiotics for at least 7 days were enrolled. IAT was defined as when inappropriate empirical antibiotics were given 48 h or longer after initial diagnosis of APN. Primary endpoint was treatment failure defined as clinical and/or microbiologic failure. Secondary endpoints were length of hospital stay and recurrence of APN. Propensity score matching was used to adjust heterogeneity of each group. Among 175 eligible cases, 59 patients received IAT and 116 patients received appropriate empirical antimicrobial therapy (AT). Treatment failure was observed in five (8.4%) patients and nine (7.8%) patients in each group, respectively. After matching, the treatment failure rate was similar between both groups (adjusted odd ratio [aOR] 1.05; 95% confidence index [CI] 0.26-4.15). The length of hospital stay (median 11 days in the IAT group versus 11 days in the AT group; P = 0.717) and absence of recurrence within 2 months (90.3% in IAT and 86.7% in AT; P = 0.642) were also similar. IAT did not adversely affect the clinical outcome. In this regard, clinicians should be more cautious about indiscriminate prescription of broad-spectrum antibiotics such as carbapenem empirically for treatment of APN possibly caused by ESBL-PE.