Background
Studies comparing colistin and polymyxin B minimum inhibitory concentrations (MICs), as well as their correlation and agreement, using non-commercial broth microdilution methods, remain limited. This study evaluated the in vitro susceptibilities of two polymyxins against clinically important Gram-negative bacteria.
Methods
Between 2018 and 2022, 891 Enterobacterales, 100 Acinetobacter baumannii, and 100 Pseudomonas aeruginosa clinical isolates from a tertiary hospital were tested for MICs of colistin and polymyxin B using broth microdilution. MIC correlation, categorical agreement (CA), essential agreement (EA), and directional discrepancies were assessed.
Results
Non-resistant rates of colistin for Enterobacterales, A. baumannii, and P. aeruginosa were 90.2 %, 99 %, and 98 %, respectively; those of polymyxin B were 90.5 %, 99 %, and 100 %. Carbapenem-resistant Enterobacterales had higher resistance to both agents (26 %). Colistin and polymyxin B showed identical MIC₅₀/MIC₉₀ values against Enterobacterales (0.25/2 mg/L) and A. baumannii (0.25/1 mg/L). Regarding P. aeruginosa, MIC₅₀/MIC₉₀ values were 1/1 mg/L and 0.5/1 mg/L, respectively. MICs were highly correlated (Spearman’s ρ=0.669, p < 0.001), with CA and EA ≥ 96 % across species. Using colistin to predict polymyxin B susceptibility yielded low very major error (VME) and major error (ME) rates. Conversely, elevated VMEs occurred when using polymyxin B to infer colistin susceptibility, particularly in A. baumannii (100 %) and P. aeruginosa (50 %), though these reflected very small numbers of colistin-resistant isolates (A. baumannii, n = 1; P. aeruginosa, n = 2).
Conclusions
Colistin and polymyxin B showed comparable in vitro activity and high MIC correlation. Colistin reliably predicted polymyxin B susceptibility, but caution is needed when inferring colistin susceptibility from polymyxin B results.
扫码关注我们
求助内容:
应助结果提醒方式:
