Pharmacokinetics/pharmacodynamics analysis and establishment of optimal dosing regimens using unbound cefmetazole concentration for patients infected with Extended-Spectrum β-lactamase producing Enterobacterales (ESBL-E).

Study objective: Establish methods for measuring cefmetazole (CMZ) concentrations conduct a pharmacokinetic/pharmacodynamic (PK/PD) analysis using unbound CMZ concentrations for extended-spectrum β-lactamase producing enterobacterales (ESBL-E) and investigate optimal dosing regimens for not undergoing hemodialysis (non-HD) and undergoing hemodialysis (HD) patients.

Design: Prospective observational study.

Patients: Included patients treated with CMZ who provided written informed consent and were admitted to the Tokyo Bay Urayasu Ichikawa Medical Center between August 2021 and July 2022.

Measurements: Total and Unbound CMZ concentration was measured by high-performance liquid chromatography (HPLC) with solid-phase extraction and ultrafiltration.

Setting: Determining the CMZ dosing regimen involved modified creatinine clearance (CL_CR ) with measured body weight (BW) using the Cockcroft-Gault equation. For non-HD patients, blood samples were collected during at least three points. For patients undergoing HD, 1 g was administered via intravenous infusion, or rapid intravenous injection after HD, or 30 min before the end of HD. Blood samples were collected before HD (pre-HD), and 1 and 3 h after starting HD and post-HD. All blood samples were collected at steady-state. Patient information was collected from electronic medical records. An unbound PK model was constructed for the non-HD patients. A nomogram was constructed using Monte Carlo simulations with a 90% probability of target attainment at 70% free time above the minimum inhibitory concentration (MIC). For the HD patients, a nomogram was used to determine the optimal dosing regimen for each HD schedule.

Main results: CMZ measurement methods were established. A model analysis of unbound PK in 37 non-HD patients incorporated creatinine clearance (CL_CR ) using the Cockcroft-Gault equation, albumin (ALB) for clearance and body weight (BW) for the volume of distribution. In Monte Carlo simulations, nomograms corresponding to the MIC (known and unknown) were generated for each covariate. Using the nomogram, non-HD patients with an ESBL-E MIC of 8 mg/L, a BW of 60 kg, an ALB of 25 g/L, and a CL_CR of 60 mL/min required administration of 2 g every 6 h (1- and 3-h infusions). Unbound PK model parameters were calculated for 7 HD patients, and the optimal dosing regimens following PK/PD were determined for each HD schedule. In HD patients, the regimen after and during HD was established using a treatment that was effective up to an ESBL-E MIC of 4 mg/L.

Conclusions: The nomogram for CMZ regimens established by PK/PD analysis of measured CMZ concentrations enables optimal CMZ dosing for ESBL-E-infected patients.