Monitoring of antiepileptic drugs using microdialysis : methodological and clinical aspects

Abstract

In therapeutic monitoring of antiepileptic drugs (AEDs) the level of unbound drug at the sampling site should ideally equal the unbound concentration at receptor sites. Microdialysis enables direct sampling of the unbound drug concentration and avoids some of the difficulties encountered with ex vivo protein binding. Following cathether application, the patient can be continuously monitored during several days, avoiding the trauma associated with repeated venepuncture.

The general aim of this thesis was to evaluate the potential of subcutaneous microdialysis as a tool for therapeutic drug monitoring (TDM) of AEDs by studies of AED pharmacokinetics in plasma, subcutaneous extracellular fluid (ECF) and subdural cerebrospinal fluid (CSF).

The reliability and safety of subcutaneous microdialysis sampling of valproate (VPA), carbamazepine (CBZ), phenobarbital (PB) and phenytoin (PHT) was investigated in vitro and in vivo. In vitro studies showed no binding of VPA to the equipment. The binding of CBZ and PB was limited and predictable while the binding of PHT was extensive and highly variable, which precluded in vivo testing. In vivo subcutaneous microdialysis studies showed a close to 100% recovery for VPA. The CBZ recovery, when corrected for binding to the plastic tubing, was 81%. In single case studies, subdural CSF and subcutaneous ECF levels of CBZ and PB were similar and when corrected for binding to the plastic tubings they were also close to protein unbound concentrations in plasma.

The time course of the distribution of VPA to subdural CSF in relation to subcutaneous ECF and plasma after a single oral dose in three patients undergoing subdural EEG monitoring because of drug resistent epilepsy was studied. VPA was rapidly distributed to subdural CSF in concentrations marginally lower than subcutaneous microdialysate and unbound plasma concentrations.

The relation between VPA induced tremor and VPA concentrations measured as unbound and total concentrations in plasma and in subcutaneous microdialysate was investigated in a dose matched casecontrol study. VPA levels did not differ significantly between tremor cases and controls, irrespective of which sampling method that was used. The degree of tremor was weakly correlated with VPA concentrations and none of the concentration measures used was superior to the other in terms of correlation to tremor.

When comparing unbound and total plasma VPA and subcutaneous microdialysate VPA after a single dose and in steady state, the unbound plasma fraction was significantly higher (p< 0.001) in the steady state group compared to the single dose group. The unbound VPA fraction was also significantly higher in steady state compared to spiked samples both at low and high total VPA concentrations (p<0.001). The results provide indirect evidence of displacement of VPA from plasma proteins by product(s) of VPA biotransformation.

Microdialysis was used to estimate topiramate (TPM) concentrations in subdural CSF and subcutaneous ECF in relation to plasma levels in a single case study of a patient undergoing subdural EEG monitoring because of drug resistent epilepsy. The subcutaneous ECF and subdural CSF TPM concentrations were similar and corresponded well (r=0.98). Total TPM concentrations in plasma were 75 and 80 % of subcutaneous and subdural levels respectively and, assuming a TPM protein binding of -13% they were close to unbound TPM levels in plasma.

It is concluded that subcutaneous microdialysis has the potential to be useful for TDM of VPA, CBZ, PB and TPM and may be of benefit in situations when frequent long term monitoring of unbound drug concentrations and a minimal trauma is particularly desirable.