Clinical Microbiology and Antimicrobial Chemotherapy. 2026; 28(1):131-144
Introduction. Selective digestive decontamination of the digestive tract (SDD) with non-absorbable antibiotics is considered a promising approach to reduce the risk of infections in critically ill patients. However, the widespread circulation of multidrug-resistant microorganisms, including carbapenemresistant Klebsiella pneumoniae and Pseudomonas aeruginosa, significantly reduces the efficacy of conventional SDD regimens based on colistin monotherapy. The key reason for failures is the wide «mutant selection window» – a range of antibiotic concentrations insufficient to suppress the growth of resistant subpopulations. The use of antimutant combinations of non-absorbable antibiotics may open new prospects for SDD optimization. Development and validation of a dynamic flow-through in vitro system for pharmacokinetic/ pharmacodynamic (PK/PD) modeling of clinical SDD regimens and evaluation of the antimutant activity of the combination of non-absorbable antibiotics colistin and rifaximin against extremely antibiotic-resistant strains of K. pneumoniae and P. aeruginosa.
The study included 10 clinical isolates of K. pneumoniae (producers of NDM, KPC, OXA-48, KPC+NDM) and 10 clinical isolates of P. aeruginosa (including 4 VIM carbapenemase producers). Minimum concentrations of colistin and rifaximin that prevent the selection of resistant mutants (MPC) were determined using agar plating with antibiotic. A single-chamber flow-through system was constructed that reproduces the pharmacokinetics of non-absorbable antibiotics during enteral administration. Clinical regimens of colistin, rifaximin, and their combination administration (every 8 hours for 7 days) were modeled.
MPC of colistin for K. pneumoniae strains ranged from 64 to 128 mg/L, for P. aeruginosa – 8 to 16 mg/L, which was 8–256 and 4–16 times higher than the MIC, respectively. MPC of rifaximin could not be determined within the tested concentrations for the vast majority of strains (MPC ≥256 mg/L). In the presence of rifaximin, the MPC of colistin decreased 4–8-fold for K. pneumoniae and 2–8-fold for P. aeruginosa. In static experiments, colistin (70 mg/L) and rifaximin (200 mg/L) did not prevent breakthrough growth of some strains within 24–48 hours, whereas the combination completely suppressed the growth of all strains for 7 days. In the dynamic flow-through system during monotherapy, adaptation of some microorganisms to colistin and rifaximin was observed; the combination of colistin and rifaximin ensured complete suppression of growth of all tested strains throughout the entire 7-day observation period.
The developed dynamic flow-through system is a convenient tool for preclinical evaluation of PK/PD parameters of antimutant combinations of non-absorbable antibiotics. The combination of colistin and rifaximin demonstrates high antimutant activity under conditions simulating SDD regimens against both K. pneumoniae and P. aeruginosa, which opens prospects for its clinical use in regions with high prevalence of antibiotic-resistant pathogens.