OBJECTIVE： To provide reference for the formulation of primary medication regimen for antibacterial drug treatment of ICU patients with Escherichia coli infection. METHODS： Based on the surveillance report on E. coli resistance in hospitals issued by CHINET China bacterial drug resistance surveillance network in 2016， 19 third class A hospitals in China were collected as E. coli clinically isolated from ICU wards. Antibiotics with resistance rate of less than 40％ to E. coli and with high utilization rate in clinical practice were selected as the research objects， and a simulated drug delivery scheme was formulated. Monte Carlo simulation method was used to simulate the clinical effect of different dosage regimens on 10 000 cases among “patients with E. coli infection” in ICU wards. The target thresholds were ％fT＞MIC＞50％ （piperacillin/tazobactam， cefoperazone/sulbactam），％fT＞MIC＞40％ （meropenem）， fcmax/MIC＞10 （amikacin）. The cumulative response percentage （CFR） to the target threshold requires that CFR be greater than 90％ for the optimal regimen. The results were compared with those of 275 clinical ICU pationts. RESULTS： Four antibiotics were identified， namely cefoperazone/sulbactam， piperacillin/tazobactam， meropenem and amikacin； sixteen medication regimen were simulated， including 1 kind of cefoperazone/sulbactam “3.0 g， q8 h”； 3 kinds of piperacillin/tazobactam “2.25 g，  q6 h” “3.375 g， q8 h” and “3.375 g， q6 h”； 2 kinds of meropenem “0.5 g， q8 h” “1.0 g， q8 h”； 3 kinds of amikacin “0.4 g， q24 h” “0.6 g， q24 h” and “0.8 g， q24 h”. Their CFR values were higher than 90％， all of them could be regarded as primary medication regimen. The clinical results were basically consistent with the simulation results. CONCLUSIONS： Above medication regimen of piperacillin/tazobactam， cefoperazone/sulbactam， meropenem and amikacin can be used as initial empirical drug selection for patients with E. coli infection in ICU.