In vitro synergy of colistin combinations against colistin-resistant Acinetobacter baumannii, Pseudomonas aeruginosa, and Klebsiella pneumoniae isolates

Abstract

Colistin resistance, although uncommon, is increasingly being reported among Gram-negative clinical pathogens, and an understanding of its impact on the activity of antimicrobials is now evolving. We evaluated the potential for synergy of colistin plus trimethoprim, trimethoprim-sulfamethoxazole (1/19 ratio), or vancomycin against 12 isolates of Acinetobacter baumannii (n = 4), Pseudomonas aeruginosa (n = 4), and Klebsiella pneumoniae (n = 4). The strains included six multidrug-resistant clinical isolates, K. pneumoniae ATCC 700603, A. baumannii ATCC 19606, P. aeruginosa ATCC 27853, and their colistin-resistant derivatives (KPm1, ABm1, and PAm1, respectively). Antimicrobial susceptibilities were assessed by broth microdilution and population analysis profiles. The potential for synergy of colistin combinations was evaluated using a checkerboard assay, as well as static time-kill experiments at 0.5× and 0.25× MIC. The MIC ranges of vancomycin, trimethoprim, and trimethoprim-sulfamethoxazole (1/19) were ≥128, 4 to ≥128, and 2/38 to >128/2,432 μg/ml, respectively. Colistin resistance demonstrated little impact on vancomycin, trimethoprim, or trimethoprim-sulfamethoxazole MIC values. Isolates with subpopulations heterogeneously resistant to colistin were observed to various degrees in all tested isolates. In time-kill assays, all tested combinations were synergistic against KPm1 at 0.25× MIC and 0.5× MIC and ABm1 and PAm1 at 0.5× MIC. In contrast, none of the tested combinations demonstrated synergy against any colistin-susceptible P. aeruginosa isolates and clinical strains of K. pneumoniae isolates. Only colistin plus trimethoprim or trimethoprim-sulfamethoxazole was synergistic and bactericidal at 0.5× MIC against K. pneumoniae ATCC 700603. Colistin resistance seems to promote the in vitro activity of unconventional colistin combinations. Additional experiments are warranted to understand the clinical significance of these observations.

Fig 1

Colistin population analysis profiles. (A) ●, A. baumannii ATCC 19606; ○, ABm1; ▽, AB1; △, AB2. (B) ●, K. pneumoniae ATCC 700603; ○, KPm1; ▽, KP3; △, KP4. (C) ●, P. aeruginosa ATCC 27853; ○, PAm1; ▽, PA4; △, PA5. The error bars indicate standard deviations.

Fig 2

In vitro evaluation of the bactericidal activity of colistin combinations at 0.5× MIC against A. baumannii ATCC 19606 (A), K. pneumoniae ATCC 700603 (B), P. aeruginosa ATCC 27853 (C), ABm1 (D), KPm1 (E), and PAm1 (F). ●, growth control; ○, colistin plus vancomycin; ▼, colistin plus trimethoprim; △, colistin plus trimethoprim-sulfamethoxazole. The error bars indicate standard deviations.