Abstract
Our aims are to assess various colistin dosing regimens against Pseudomonas aeruginosa (P. aeruginosa) infection in critically ill patients and to propose an appropriate regimen based on microbiological data. A Monte Carlo simulation was performed using the published colistin's pharmacokinetic parameters of critically ill patients, the published pharmacodynamic target from a mouse thigh infection model, and the minimum inhibitory concentration (MIC) results from a Vietnamese hospital. The probability of target attainment (PTA) of 80% and cumulative fraction of response (CFR) of 90% were used to evaluate the efficacy of each regimen. Of 121 P. aeruginosa laboratory datasets, the carbapenem-resistant P. aeruginosa (CRPA) and the colistin-resistant P. aeruginosa rates were 29.8% and 0.8%, respectively. MIC(50,90) were both 0.5 mg/L. The simulated results showed that at MIC of 2 mg/L, most regimens could not reach the PTA target, particularly in patients with normal renal function (Creatinine clearance (CrCl) ≥ 80 mL/min). At MIC of 0.5 mg/L and 1 mg/L, current recommendations still worked well. On the basis of these results, aside from lung infection, our study recommends three regimens against P. aeruginosa infection at MIC of 0.5 mg/L, 1 mg/L, and 2 mg/L. In conclusion, higher total daily doses and fractionated colistin dosing regimens could be the strategy for difficult-to-acquire PTA cases, while a less aggressive dose might be appropriate for empirical treatment in settings with low MIC(50/90).