Novel insights into pharmacokinetics of intraperitoneal cefepime during automated peritoneal dialysis using Monte Carlo simulations: INTRAPERITONEAL CEFEPIME DURING SHORT-DWELL EXCHANGES IN AUTOMATED PERITONEAL DIALYSIS

Introduction:
The International Society for Peritoneal Dialysis (ISPD) guidelines currently recommend intraperitoneal (IP) administration of cefepime, a fourth-generation cephalosporin, as a first-line monotherapy for infectious peritoneal dialysis-associated peritonitis (PDAP). Despite this, data on the pharmacokinetics (PK) of cefepime and its optimal dosing in automated peritoneal dialysis (APD) remain limited. This study aimed to characterize the PK of cefepime in APD patients and to predict the probability of achieving pharmacokinetic/pharmacodynamic (PK/PD) targets during short versus long-dwell periods using Monte Carlo simulations.

Methods:
Eight adult APD patients without recent infections were enrolled. APD was initiated with Physioneal® 40 Glucose 1.36% or 2.27%, using five short cycles totaling 8 h. Cefepime (2g) was injected once into a preheated (37°C) 5 L peritoneal dialysis fluid (PDF) bag and administered IP using a HomechoicePro Cycler®. After 8 h, the PDF was switched to 1.5 L Icodextrin for a single 16 h long dwell. This protocol enabled cefepime delivery during the short dwell cycles, followed by an antibiotic-free long dwell phase to promote passive back-diffusion into the peritoneal cavity.
Blood, peritoneal fluid, and urine samples were collected at set intervals to evaluate the PK profile. Bioanalysis used high-performance liquid chromatography (HPLC). Results were assessed against the minimum inhibitory concentration (MIC) breakpoints for PDAP-causing bacteria defined by the EUCAST 2025 (v 15.0). Concentration thresholds of interest were 4 and 8 mg/L.
A population PK model was developed using nonlinear mixed effects modelling based on all samples, followed by Monte Carlo simulations to generate cefepime PK profiles for different IP dosing regimens. Separate simulations were conducted, assuming a 3-fold and 10-fold increase in transfer rate of cefepime between plasma and peritoneal fluid to account for PDAP. A fixed protein binding of 16.4% for cefepime in plasma was applied. A %fT>MIC (percentage of the dosing interval during which free cefepime levels exceed MIC) of ≥ 50% was considered the target associated with maximal bactericidal activity.

Results:
Among the eight patients, two maintained plasma cefepime concentrations above the MIC of 8 mg/L, and one maintained such levels in the dialysate. Additionally, plasma concentrations exceeded 4 mg/L in all patients, while dialysate concentrations exceeded 4 mg/L in seven out of eight patients throughout the 24-hour period.
A one-compartment model for both plasma and dialysate concentrations best described the PK profiles. Plasma-dialysate transfer of cefepime was described by a bidirectional clearance of 1.16 L/h [0.929-1.39] (point estimate and 95% CI). Systemic cefepime clearance was 3.04 L/h (2.36-3.72).
For patients without peritonitis, Monte Carlo simulations indicate that 1 g/day IP is sufficient to achieve the target of %fT>MIC ≥ 50% target for a MIC of 4 mg/L in both plasma and dialysate without accumulation, while 2 g/day is needed for a MIC of 8 mg/L in over 90% of patients. To reach the 8 mg/L target, 1.5 g/day IP was predicted to be sufficient in PDAP.

These Monte Carlo simulations demonstrate that a single short-dwell administration can achieve effective plasma and dialysate / IP cefepime concentrations in APD patients with PDAP.