Although colistin methanesulfonate (CMS) has been used extensively in critically ill patients infected with multidrug-resistant organisms, the optimum dosing regimen remains to be determined. Herein, we examined the pharmacokinetics of three different dosing regimens of CMS, 3 million units every 8 h (regimen A), 4.5 million units every 12 h (regimen B), 9 million units every 24 h (regimen C) and evaluated the bactericidal activity of serum containing various concentrations of colistin against Pseudomonas aeruginosa with a minimum inhibitory concentration (MIC) of 1 microg/ml. the means +/- SE serum C(max )of colistin for regimens A, B, and C were 3.34+/-0.35, 2.98+/-0.27, and 5.63+/-0.87 microg/ml, respectively. All serum samples containing colistin >4 microg/ml (serum concentration/MIC >4) eliminated P. aeruginosa whereas only 40% of samples containing colistin <4 microg/ml resulted in complete bacterial killing. these findings indicate that the currently used dosing regimens might not provide the most effective therapy with CMS and justify administering larger dosages in longer intervals.