A Monte Carlo Simulation Approach for Beta‐Lactam Dosing in Critically Ill Patients Receiving Prolonged Intermittent Renal Replacement Therapy

Abstract

Cefepime, ceftazidime, and piperacillin/tazobactam are commonly used beta-lactam antibiotics in the critical care setting. For critically ill patients receiving prolonged intermittent renal replacement therapy (PIRRT), limited pharmacokinetic data are available to inform clinicians on the dosing of these agents. Monte Carlo simulations (MCS) can be used to guide drug dosing when pharmacokinetic trials are not feasible. For each antibiotic, MCS using previously published pharmacokinetic data derived from critically ill patients was used to evaluate multiple dosing regimens in 4 different prolonged intermittent renal replacement therapy effluent rates and prolonged intermittent renal replacement therapy duration combinations (4 L/h × 10 hours or 5 L/h × 8 hours in hemodialysis and hemofiltration modes). Antibiotic regimens were also modeled depending on whether drugs were administered during or well before prolonged intermittent renal replacement therapy therapy commenced. The probability of target attainment (PTA) was calculated using each antibiotic's pharmacodynamic target during the first 48 hours of therapy. Optimal doses were defined as the smallest daily dose achieving ≥90% probability of target attainment in all prolonged intermittent renal replacement therapy effluent and duration combinations. Cefepime 1 g every 6 hours following a 2 g loading dose, ceftazidime 2 g every 12 hours, and piperacillin/tazobactam 4.5 g every 6 hours attained the desired pharmacodynamic target in ≥90% of modeled prolonged intermittent renal replacement therapy patients. Alternatively, if an every 6-hours cefepime regimen is not desired, the cefepime 2 g pre-prolonged intermittent renal replacement therapy and 3 g post-prolonged intermittent renal replacement therapy regimen also met targets. For ceftazidime, 1 g every 6 hours or 3 g continuous infusion following a 2 g loading dose also met targets. These recommended doses provide simple regimens that are likely to achieve the pharmacodynamics target while yielding the least overall drug exposure, which should result in lower toxicity rates. These findings should be validated in the clinical setting.

Principles of Drug Dosing in Sustained Low Efficiency Dialysis (SLED) and Review of Antimicrobial Dosing Literature

“…Alternative regimens that met the PTA were 4.5 g every 6 h as an extended infusion (although a higher PTA was not obtained compared to conventional infusion time) or 16 g per day given as a continuous infusion. Piperacillin/tazobactam 16 g administered at the same time as the initiation of SLED was more likely to exceed the toxicity threshold than the other dosing regimens modelled [21].…”

“…Drug concentrations were assessed using methods such as high-performance liquid chromatography or mass spectrometry. More recently, there have been a few publications of in silico pharmacokinetic and pharmacodynamic analyses utilizing Monte Carlo simulations (MCS) and virtual subjects to simulate real world patient populations [21,22,24,26,27,33,35,38,47]. When performing these simulations, researchers can preset a desirable probability of target attainment (PTA) of having a drug concentration greater than a multiple of the MIC for a preset amount of time.…”

“…A MCS was performed in critically ill patients receiving SLED targeting free concentrations of piperacillin-tazobactam greater than four times the MIC of Pseudomonas aeruginosa for >50% the first 48 h of antibiotic therapy [21]. Four different models of dialysis were simulated [8 h per day (ultrafiltration rate/dialysate flow rate of 5 L/h) or 10 h per day (ultrafiltration rate/dialysate flow rate of 4 L/h)] of hemofiltration or hemodialysis.…”

Suboptimal Clinical Response Rates with Newer Antibiotics Among Patients with Moderate Renal Impairment: Review of the Literature and Potential Pharmacokinetic and Pharmacodynamic Considerations for Observed Findings

“…This consideration is also based on the finding that most nonimmunologic exposure–dependent adverse drug events occur after several days of therapy, and the risk of toxicity, even in the presence of high exposures, is typically low during the early days of therapy . As an example, Monte Carlo simulations of ceftazidime among critically ill patients showed the importance of “aggressive,” or nonadjusted initial dosing, even in the setting of renal impairment . This strategy increased the PTA of ceftazidime without increasing the probability of toxic concentrations in the first 48 hours of treatment.…”

“…23 As an example, Monte Carlo simulations of ceftazidime among critically ill patients showed the importance of "aggressive," or nonadjusted initial dosing, even in the setting of renal impairment. 26 This strategy increased the PTA of ceftazidime without increasing the probability of toxic concentrations in the first 48 hours of treatment. If renal function continues to change acutely, clinicians should consider collecting urine creatinine and output to assess a patient's kidney function.…”