Use of Monte Carlo simulation to optimize antibiotic selection for bloodstream infections caused by Enterobacteriaceae in Shandong Province, China.

The increasing rates of resistance to β-lactams have made it more challenging for clinicians to select appropriate antibiotic treatment for bloodstream infections (BSIs) caused by suspected Enterobacteriaceae. The objective of this analysis was to determine the optimal dosage regimens of β-lactams for treatment of BSIs based on analysis of 19,334 Enterobacteriaceae collected from blood specimens. Monte Carlo simulation using pharmacokinetic parameters of infected patients was performed to determine the probability of overall pharmacokinetic/pharmacodynamic (PK/PD) target attainment (OPTA). E. coli, K. pneumoniae, and E. cloacae were the 3 most common species. Nine of the 16 tested regimens had optimal OPTAs (>90%) for Enterobacteriaceae overall (meropenem 2g q8h, 3 h infusion; meropenem 2g q8h, 0.5h; meropenem 1g q8h, 0.5h; piperacillin/tazobactam 4.5g q8h, 3h; ceftazidime 2g q8h, 3h; imipenem 0.5g q6h, 0.5h; imipenem 1g q8h, 0.5h; piperacillin/tazobactam 3.375g q6h, 0.5h; ceftazidime 2g q8h, 0.5h). Four other regimens had sub-optimal OPTAs of 80 to 90% (piperacillin/tazobactam 4.5g q8h, 0.5h; ceftazidime 1g q8h, 0.5h; cefepime 2g q12h, 3h; and cefepime 2g q12h, 0.5h). Although there are high antibiotic MICs among Enterobacteriaceae in Shandong Province, carbapenem- , ceftazidime- and piperacillin/tazobactam- based regimens provide the optimal treatment.