Comparison of antibacterial activities of polymyxin B and colistin against multidrug resistant Gram negative bacteria

Doymaz M. Z., Karaaslan E.

INFECTIOUS DISEASES, vol.51, no.9, pp.676-682, 2019 (SCI-Expanded) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 51 Issue: 9
  • Publication Date: 2019
  • Doi Number: 10.1080/23744235.2019.1640386
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.676-682
  • Bezmialem Vakıf University Affiliated: Yes


Background: Polymyxin B and colistin have similar structures except for one amino acid. Usually, physicians choose either polymyxin B or colistin for treatment of infections caused by multidrug-resistant (MDR) organisms. The preference is based on previous experience. Not much data are found in the literature comparing the two drugs against the same microorganisms. In this study, we compared in vitro antimicrobial activities of the two polymyxins against a panel of highly resistant and susceptible microorganisms. Methods: Eighty-nine clinical isolates (27 Klebsiella pneumoniae, 31 Acinetobacter baumannii and 31 Pseudomonas aeruginosa) were tested in broth microdilution assays. Time-kill curve experiments were carried out on selected isolates. Results: Significantly lower MICs for polymyxin B than for colistin were found against all species tested including K. pneumoniae (p < .02), A. baumannii (p < .001) and P. aeruginosa (p < .01). The low MICs caused a change in categorical interpretations of only two K. pneumoniae and two P. aeruginosa. Similar results were obtained in time-kill curve experiments with both susceptible and resistant clinical isolates. Conclusions: Significantly lower MICs were found for polymyxin B against three of the most critical MDR species. Even though differences in categorical interpretations were not striking, lower MICs might be a critical consideration in clinical management of select cases where the concentration of these toxic antibiotics matters because of underlying co-morbidities. These results provide support to previous suggestions that re-consideration of breakpoint interpretations for polymyxins might be needed.