Search

Several case reports have indicated that the selective serotonin re-uptake inhibitor (SSRI) fluoxetine increases phenytoin blood levels when given concurrently. The mechanism of this drug-drug interaction has been attributed to inhibition of CYP2C9-catalyzed hydroxylation of phenytoin to its major oxidative metabolite in humans, para-hydroxyphenyl phenyl hydantoin (HPPH). With a bank of human liver microsomes (HLM), four SSRIs (fluoxetine, norfluoxetine, sertraline, and paroxetine) were tested for inhibition of HPPH formation. Initially, the K(m) and V(max) values of phenytoin hydroxylation to HPPH were determined in the individual HLM samples. The average K(m) (n=8) was 9.7+/-2.9 microM. The V(max) varied fivefold, with an average value of 113+/-53 pmol HPPH/min/nmol CYP450. All of the SSRIs inhibited HPPH formation; resulting Ki values were 31.1+/-10.1 microM (fluoxetine) (n=5), 51.1+/-9.4 microM (norfluoxetine) (n=3), 52.2+/-21.5 microM (sertraline) (n=3), and 80.0+/-7.2 microM (paroxetine) (n=3). Sulfaphenazole (10 microM), utilized as a positive control for inhibition of HPPH formation, inhibited phenytoin hydroxylation (>95%) in all HLM samples. Diclofenac hydroxylation to 4'-OH diclofenac, a specific marker for CYP2C9 activity, was determined in HLM1-HLM6 and was highly correlated with HPPH formation in HLM1-HLM6, indicating that phenytoin hydroxylation in human liver microsomes is largely due to CYP2C9. This work presents direct evidence that the effect of fluoxetine on phenytoin blood levels may be explained by inhibition of CYP2C9-catalyzed phenytoin hydroxylation. In light of typical SSRI blood levels observed in patients, this study also suggests that the risk of a SSRI-phenytoin interaction is highest with fluoxetine and norfluoxetine, and less likely with sertraline and paroxetine.

Stable isotope analogues of phenytoin are useful for pulse dose pharmacokinetic studies in epilepsy patients. A simultaneous assay was developed to quantitate phenytoin (5,5-diphenylhydantoin) and its stable isotope analogue [13C3]-phenytoin (5,5-diphenyl-2,4,5-13C3-hydantoin) from plasma. Quantitation was achieved by GC-MS analysis of liquid/liquid extracted plasma samples, with [2H10]-phenytoin (5,5-di(pentadeuterophenyl)-hydantoin) as an internal standard. The total coefficients of variance (C.V.t) were < 7% for phenytoin (2.5-40 micrograms ml-1) and < 10.3% for [13C3]-phenytoin (0.1-6.0 micrograms ml-1). The accuracy of the assay varied from 87.8-100.1% (phenytoin, 2.5-40 micrograms ml-1) and 89.6-116.3% ([13C3]-phenytoin, 0.02-6.0 micrograms ml-1). The assay was tested under in vivo conditions by administration of a pulse dose of the stable isotope analogue to a single rat dosed to steady-state with fosphenytoin, a phenytoin prodrug. The results of the in vivo experiment demonstrate the usefulness of this assay for future pharmacokinetic studies in special population epilepsy patients.