Association of Cefepime Trough Levels With Clinical Efficacy and Neurotoxicity in Patients With Febrile Neutropenia

Cefepime, a fourth-generation cephalosporin, is widely prescribed for the treatment of serious bacterial infections. Cefepime has a more narrow spectrum and has relatively lower minimal inhibitory concentration (MIC) values compared to meropenem. Therefore cefepime is widely endorsed, and preferred over a carbapenem antibiotic, as a first line agent for the management of febrile neutropenia (FN) in hematologic patients.

Several studies have shown adequate exposure using the recommended 3x2 g dosing strategy for cefepime in most FN patients. The pharmacokinetic/pharmacodynamic (PKPD) target of fT100%>MIC, which is frequently used for BL AB in critically ill patients, is achieved in most patients. Subsets of patients have been shown to be at risk for underdosing, especially those with a higher than normal creatinine clearance, obese or cachectic patients and those infected with bacteria with a MIC >4 mg/L. Despite adequate cefepime exposure, the association with clinical outcome still needs to be confirmed on a large scale. An association between suboptimal PKPD target attainment and clinical outcome was shown on a large scale for β-lactam antibiotics. These data need to be validated for the specific antimicrobials agents and in subsets of patients. Another study suggested a cefepime PKPD target for clinical outcome of at least fT>MIC of 68% to 74% based on simulations using 2 existing population PK models. In a mixed population of critically and non-critically ill patients with Gram negative blood stream infections this threshold was a significant predictor of in-hospital mortality. A recent study, undertaken in patients with nosocomial pneumonia treated with cefepime, even reported clinical failure despite achievement of fT100%>MIC, they suggest a fCmin/MIC ratio of at least 2.1 as a better predictor for clinical failure in Gram negative bacterial pneumonia patients. Up until today, there are no explicit data linking PKPD target attainment with clinical outcome in FN patients.

With respect to cefepime toxicity, there is a lot of controversy on the upper limit that should not be exceeded in clinical practice. Cefepime toxicity is mainly neurological, going from mild confusion to seizures or encephalopathy. In the first study that assessed the relationship between cefepime PK and associated neurotoxicity, trough levels >22 mg/L were found to be associated with a 50% probability of neurological toxicity. However, a more recent study showed, using a non-parametric approach, that estimated rates of neurotoxicity using this threshold are very high and discordant with rates seen in clinical practice. Hence this threshold seems not representative for the clinical practice. Other PK parameters (Cmax, AUC, ...) and covariates (CrCl, Age, dose, duration of therapy, co-medication, …) probably also influence cefepime related neurotoxicity. Recently, an attempt to redefine cefepime's threshold for neurotoxicity was published in a retrospective cohort study. No neurotoxicity was observed below 35 mg/L for all samples (steady-state levels during continuous infusion, trough and non-trough levels during intermittent infusion). However, when looking only at trough levels, they also advise avoiding levels >20 mg/L until more information from prospective studies is available.

Moreover, until now there has only been one study investigating the carryover of cefepime in samples taken via a central venous catheter. This study suggested possible carryover for triple lumen catheters and peripherally-inserted central catheters. Due to the small sample size and the in vitro nature of this research, these results need to be confirmed.

The aim of this research is two-fold. Firstly, the aim is to investigate whether there is significant carryover of cefepime in samples taken via a central venous catheter. Next, the aim is to assess prospectively the relation between cefepime exposure and efficacy or safety in a FN patient population. As literature on these topics is scarce this research will further explore the exposure-response relationship in order to define a therapeutic window for maximal clinical cure and minimal neurotoxicity that can be used in future research and clinical practice.