Multicentre study of risk factors for mortality in patients with acinetobacter bacteraemia receiving colistin treatment

[[abstract]]Colistin remains a last-line antibiotic for the treatment of multi-drug-resistant Acinetobacter species. However, mortality rates are high in patients with Acinetobacter infection receiving colistin treatment. This multi-center study evaluated whether colistin susceptibility, additional antimicrobial agents, or other prognostic factors influenced the clinical outcomes of patients receiving colistin treatment for Acinetobacter bacteremia. This retrospective study enrolled 122 adults receiving colistin treatment for monomicrobial Acinetobacter bacteremia at six medical centers over 8 years. Clinical information, antimicrobial susceptibility, and colistin resistance determinants were analysed. The primary outcome measure was 14-day mortality. Among 122 patients, 18 and 104 were infected with colistin-resistant (ColR) isolates (minimal inhibitory concentrations [MICs] >/=4 mg/L) and colistin-susceptible (ColS) isolates (MICs </=2 mg/L), respectively. Patients infected with ColR and ColS isolates were not significantly different in Charlson comorbidity index, invasive procedures, sources of bacteremia, disease severity, and 14-day mortality rates (44.4% vs. 34.6%, P=0.592). No specific additional antimicrobial agent was independently associated with higher or lower mortality. Coronary artery disease, higher APACHE II score, and bacteremia caused by A. baumannii were independent risk factors associated with 14-day mortality. The mechanisms of colistin resistance were associated with amino acid variants in the pmrCAB operon. Finally, we identified previously unreported A. nosocomialis amino acid variants related to colistin resistance. In conclusion, colistin susceptibility and colistin combination antimicrobial treatment were not associated with decreased 14-day mortality in patients with Acinetobacter bacteremia receiving colistin treatment