Optimizing mathematical models for the efficacy and precise regimens of meropenem in RICU patients: a real-world prospective study

Background It is inadequate to direct evaluation of meropenem pharmacokinetic targets and clinical efficacy. This study aimed to investigate the predictors of meropenem clinical effectiveness and to optimize dosing regimens precisely in the Department of Respiratory and Intensive Care Unit (RICU). Methods Patients with severe pneumonia using meropenem in RICU were included. Blood samples were assayed using two-dimensional high-performance liquid chromatography. Logistic regression analysis was conducted to identify predictors of efficacy. Population pharmacokinetics analysis was performed to optimize dosing regimens. Results 396 meropenem concentrations from 111 patients were measured. The joint predictor of total protein, platelet, C _peak , and APACHE Ⅱ score had a good predictive performance on the clinical efficacy ( P  < 0.001). Model-based simulation suggested to administer 2000 mg of meropenem every 8 hours at a 2-hour infusion for the patient with a minimum inhibitory concentration (MIC) of 8 mg/L and a creatinine clearance (CrCL) of 30 ml/min, or at 3-hour infusion for the patient with a MIC of 4 mg/L and a CrCL of 60 ml/min, respectively, which could achieve a probability of target attainment (PTA), with 100% of the time between doses at which the free fraction concentration remains above MIC (%fT > MIC) ≥ 80%. And only at specific MICs and pharmacodynamic targets, did PTAs of the prolonged infusion be significantly different ( P  < 0.05). Conclusion High C _peak caused potential unfavorable outcomes. 2000 mg of meropenem every 8 hours with 2-hour or 3-hour infusion was recommended for patients with severe pneumonia in RICU. The benefits of prolonged infusion were limited.