PopPK and PBPK Models Guide Meropenem Dosing in Critically Ill Children with Augmented Renal Clearance

Background: Meropenem (MEM) is frequently prescribed for the empirical management of severe infections in the pediatric intensive care unit (PICU). Critically ill children exhibit substantial pharmacokinetic (PK) variability, and current dosing strategies remain inadequately evaluated, particularly in neonates, infants, and those with altered renal function. Methods: This study employed a dual modeling approach integrating population pharmacokinetic (PopPK) and physiologically based pharmacokinetic (PBPK) methodologies. Clinical data from two PICUs were utilized for PopPK model development and PBPK model evaluation. Both models were rigorously assessed using goodness-of-fit plots and prediction-based metrics. Monte Carlo simulations were subsequently conducted to calculate the probability of target attainment (PTA) for multiple dosing regimens across MICs of 0.25-16 mg/L. The pharmacodynamic target (PDT) was defined as maintaining unbound plasma concentrations above the MIC for 100% of the dosing interval (100% ƒT _{> MIC}), and dosing regimens were considered acceptable if the PTA exceeded 90% for efficacy while avoiding potential toxicity (Css ≥ 50 mg/L). Results: A total of 202 MEM plasma concentrations from 101 pediatric patients were analyzed. Marked inter-individual variability in MEM pharmacokinetics and pharmacodynamics was observed. Augmented renal clearance (ARC) was frequently identified in PICU patients. We simultaneously developed a two-compartment population pharmacokinetic model incorporating body weight and estimated glomerular filtration rate, and a whole-body physiologically based pharmacokinetic model scaled from adults with adjustments for transporter ontogeny and renal function. The PopPK model, by incorporating interindividual variability on clearance and volume of distribution, captured a wider range of drug exposures and demonstrated superior predictive performance, particularly in subgroups with high eGFR. The PBPK model showed higher precision in the low eGFR subgroup but slightly lower overall predictive accuracy. Both models identified ARC as a key driver of subtherapeutic exposure. Standard regimens were insufficient for preterm neonates when the MIC was ≥4 mg/L, and even the maximum label-recommended dose failed to achieve the pharmacodynamic target for infants older than 1 month when the MIC was ≥2 mg/L. Conclusions: Both PBPK and PopPK frameworks reliably predicted MEM pharmacokinetics in critically ill pediatric patients, with complementary strengths across renal function strata. Model-informed simulations highlighted the inadequacy of standard dosing under conditions of ARC or elevated MIC, supporting individualized, precision-guided dosing strategies based on age, eGFR, and pathogen MIC.