Anticonvulsant-induced aplastic anemia: increased susceptibility to toxic drug metabolites in vitro

A 53-yr-old man sequentially developed aplastic anemia from phenytoin and carbamazepine. Both compounds undergo metabolism to potentially toxic arene oxide intermediates. We tested the hypothesis that the patient's adverse reactions were due to a defect in detoxification of such metabolites by challenging his peripheral lymphocytes with drug metabolites generated by a murine hepatic microsomal system in vitro. The patient's cell viability was normal in the absence of drugs. However, his cells showed greater toxicity from both phenytoin and carbamazepine metabolites than did controls. Toxicity was dependent on microsomes and NADPH. Intermediate toxicity was noted in cells from the patient's mother. The results provide the first evidence for a role of arene oxide drug metabolites in aplastic anemia in humans and suggest that enhanced susceptibility to toxicity may be based on an inherited abnormality in metabolite detoxification.

Dose-dependent effect of cimetidine on phenytoin kinetics

Eight normal subjects were given 250 mg intravenous phenytoin alone and with 3-day regimens of oral cimetidine, 400 mg at bedtime, 1200 mg a day, and 2400 mg a day in a randomized crossover fashion. Plasma samples for phenytoin and cimetidine, and urinary concentrations for phenytoin and 5-(4-hydroxyphenyl)-5-phenylhydantoin (HPPH) were measured by HPLC. All cimetidine regimens decreased phenytoin clearance, and there was no difference between the 400-mg bedtime dose and the 1200-mg a day regimens. There was, however, a difference between the 400-mg and 1200-mg and the 2400-mg regimens. There was no linear correlation between steady state cimetidine plasma concentrations and the decrease in phenytoin clearance. Urinary HPPH/phenytoin ratios decreased with all cimetidine treatments, but the differences were not significant. Phenytoin toxicity may result when cimetidine is added to existing regimens of this anticonvulsant.

Phenytoin dosage requirements and pharmacokinetic variables

The relationships between phenytoin dose, pharmacokinetic variables, patient data, and serum phenytoin concentrations were studied. One hundred sixty-eight adult epileptic patients who were receiving phenytoin were randomly selected and studied retrospectively. The method of Ludden et al. or a Bayesian forecasting technique was employed to estimate the patients' pharmacokinetic values for maximum rate of drug metabolism (Vmax) and the Michaelis-Menten constant (Km). Resulting steady-state serum concentrations were estimated. The daily doses of phenytoin necessary to produce steady-state serum phenytoin concentrations of 10 and 20 micrograms/ml were also determined in patients whose values were definable. Analysis of variance was used to test possible correlations between patient demographic data, pharmacokinetic values, and doses. The majority of patients (85.6%) failed to achieve concentrations between 10 and 20 micrograms/ml when receiving phenytoin sodium 300 mg daily. Patients receiving more than one phenytoin dosage regimen had significant but weak correlations between Vmax and Km. The data suggest that low Km and Vmax values occur concurrently. Initial phenytoin dose based on patients' weights or body surface areas may be useful in determining initial dosage requirements, but estimated pharmacokinetic values for Vmax and Km provide the best guide for dosage adjustment.

The fate of orally administered [4-14C]phenytoin in two healthy male volunteers

A recovery study was conducted to determine whether phenytoin (DPH), like the barbiturates, is metabolized via the recently discovered N-glucosidation pathway. Virtually 100% of the ingested 14C-labelled doses in two subjects could be accounted for in the excreta within 5 days, with 35% in feces and 65% in urine. Radioactivity in the urine was entirely due to free and conjugated 5-(4-hydroxy-phenyl)-5-phenylhydantoin (p-HPPH) and the dihydrodiol, and that in the feces mostly due to the unmetabolized drug. There was no indication of phenytoin N-glucoside being excreted in either the urine or feces of either subject, although one of the subjects was known to possess a particularly strong N-glucosidation capacity for barbiturates. The other subject was a poor metabolizer of debrisoquine and sparteine. Nevertheless, the DPH disappearance from serum and the DPH metabolite excretion in urine were virtually alike in these two subjects, indicating that the debrisoquine 4-hydroxylating and DPH hydroxylating capacities may be separable entities.

Low-melting phenytoin prodrugs as alternative oral delivery modes for phenytoin: a model for other high-melting sparingly water-soluble drugs

Phenytoin is a high-melting, weakly acidic, and sparingly water-soluble drug. Because of these physicochemical properties, phenytoin is subject to erratic bioavailability in a variety of dosage forms both in its acidic as well as sodium salt forms. A homologous series of 3-acyloxymethyl derivatives of phenytoin (acetyl through decanoyl) were synthesized and various physicochemical properties measured. The prodrugs were more readily soluble in various metabolizable glycerol esters such as tributyrin, trioctanoin, and triolein than phenytoin. The solubility of the prodrugs in the various organic vehicles studies was closely correlated to the melting point of the prodrug: the lower the melting point the greater the solubility. The cleavage rates of the prodrugs in plasma and tissue homogenates followed a parabolic relationship with chain length. The prodrug, 3-pentanoyloxymethyl-5,5-diphenylhydantoin when administered in tributyrin gave superior oral phenytoin bioavailability in rats when compared with sodium phenytoin administered as an aqueous solution.

Hypermetabolism of phenytoin as a cause of treatment failure

Hypermetabolism of phenytoin is not frequently recognized as a cause of treatment failure. We report the case of a 37-year-old male in whom detailed pharmacokinetic investigation revealed that hypermetabolism, rather than lack of compliance or poor absorption, was responsible for low plasma levels of phenytoin. An increase of his daily dose of phenytoin to 800 mg resulted in adequate plasma levels and good seizure control. Additional studies with two model drugs metabolized by the liver--aminopyrine and antipyrine--showed that he was also a fast metabolizer for these substrates, suggesting a nonspecific induction of hepatic drug metabolizing enzymes. Low plasma phenytoin levels should not be systematically ascribed to lack of compliance, and increased phenytoin metabolism should be considered as an occasional cause of treatment failure.

Hydrogen bonding interaction of diphenylhydantoin and 9-ethyladenine

A hydrogen-bonded complex of diphenylhydantoin (DPH) and 9-ethyladenine (EtAd) crystallizes from 2,4-pentanedione with the asymmetrical unit consisting of two DPH molecules, one EtAd molecule, and one solvent molecule. The crystal structure was solved by direct methods and refined to a residual of R = 0.054. Structure determination reveals that one DPH hydrogen-bonds to EtAd in a Watson-Crick scheme while the second DPH N(3)--H bonds to EtAd N(3) to form a 2:1 DPH-EtAd complex. Comparisons are made with barbiturate-adenine complexes and with an earlier postulation of a 1:1 DPH-EtAd complex derived from NMR and IR data. The 2,4-pentanedione molecule adopts the keto-enol configuration with an asymmetrical intramolecular hydrogen bond.

Phenytoin II: in vitro-in vivo bioequivalence standard for 100-mg phenytoin sodium capsules

A bioequivalence study was undertaken using an oral solution, a fast-dissolving capsule and a slow-dissolving phenytoin sodium capsule. The AUC, tmax and Cmax correlated with in vitro dissolution data. The results of the present studies substantiate the presence of two types of phenytoin sodium products on the market. On the basis of these studies, in vitro specifications for fast- and slow-dissolving phenytoin sodium capsules as well as the in vivo bioequivalence requirements for these two types of products are recommended.

Phenytoin elimination in newborns

We measured the apparent half-life (t50%) of phenytoin (PHT) 30 times in 16 infants (aged 2 to 36 days) who had seizures. During the first week of life, the t50%s ranged from 6 to 140 hours. After the first week, the concentration dependence of PHT elimination was demonstrated; the t50% was related to the initial concentration (Ci). The t50% also decreased with postnatal age. Controlling for a Ci of 18 mg per liter, the average t50% decreased threefold between the first (57.3 +/- 48.2 hours) and fourth weeks (19.7 +/- 1.31). In newborns, both age-related changes and the concentration dependence of PHT elimination can cause PHT levels to decrease when constant doses are given.

Decreased phenytoin level during antineoplastic therapy: a case report

We report a case of interaction between anticonvulsant and antineoplastic drugs in one male patient with seizures from brain tumour. The patient was treated with phenytoin (PHT), phenobarbital (PB), and an antineoplastic protocol based on a combination regimen with carmustine (BCNU), vinblastin (VLB), methotrexate (MTX), and radiotherapy. Plasma concentrations of PHT fell from 9.4 to 5.6 micrograms/ml 24 h after VLB administration, and remained low for at least 10 days. During this period, partial seizures occurred. Plasma concentrations of PB were unchanged during the period of observation. It is suggested that impaired absorption of PHT, caused by VLB or MTX or both, is responsible for this interaction.

The effect of diphenylhydantoin (phenytoin) on the sequential stages of intestinal folate absorption

The drug diphenylhydantoin (phenytoin) (DPH) is thought to interfere with the bioavailability of dietary folate through an effect on intestinal folate deconjugation and/or monoglutamate folate transport. In order to determine whether DPH inhibition occurs in the sequential steps of folate deconjugation, uptake, or reduction-methylation, the effect of the drug on the intestinal absorption of hexaglutamate folate (PteGlu6), pteroylmonoglutamate folate (PGA), and N-5-methyltetrahydrofolate (CH3FH4) was studied. Folate absorption was directly quantified by the method of triple lumen tube perfusion in 12 subjects serving as their own controls. All 12 received PGA mixed with and without DPH (20 micrograms/ml), while 6 of the 12 subjects received hexaglutamate and 6 received reduced methylated folate with and without added DPH. With this model DPH was shown not to impair significantly folate absorption by any action upon the process of folate deconjugation, absorption, or reduction-methylation. The previously reported association between DPH intake and reduced levels of serum folate remains unexplained by these studies.

Bioavailability of phenytoin on single and multiple oral doses of two dosage forms in normal subjects

The extent and rate of absorption of phenytoin (PHT) from tablet and powder were studied in four healthy adult volunteers. It was demonstrated by urinary and fecal excretion that the almost all quantity of PHT in tablet was absorbed through the gastrointestinal tract, and the observed values of the estimated free concentration (Cest.f) estimated from mixed saliva concentration of PHT in the multiple dose were in fair agreement with the calculated values of that by using computer simulation in case of tablet. On the contrary, the variations were observed in Cest.f using therapeutic dose of PHT powder. The values of Cest.f at steady-state in tablet administration were higher than those in powder administration. The absorption ratio of PHT powder was low and variable, and decreased upon increase of dose. The ratio calculated from the Cest.f values of both dosage forms at steady-state were in good correspondence to the observed values of PHT excreted in feces.

High intravenous phenytoin dosage requirement in a newborn infant

A term neonate was being treated with intravenous phenytoin. To maintain a serum level above 10 micrograms per milliliter and abolish seizure activity, it was necessary to carry out repeated serum concentration measurements, administer several loading doses, and administer an unusually large maintenance dose (25 mg per kilogram per day), divided into a short dosing interval (6 hours). Declining serum levels from postnatal days 8 to 13 on a constant dose of 9 mg per kilogram per day suggested that the rate of phenytoin metabolism was gradually increasing; rapid elimination was documented on day 18 by a half-life measurement of 8.8 hours from three samples. The changing pharmacokinetics were attributed to maturation of oxidative metabolism of phenytoin, concurrent phenobarbital administration, or both. The need for additional loading doses and maintenance dose increases must be guided by serum concentration measurements to obtain maximum benefit with minimal risk of toxicity.

Effect of cholestyramine and colestipol on the absorption of phenytoin

The hypocholesterolemic cation resins, cholestyramine and colestipol, have variable effects on the absorption parameters of a number of lipophillic, anion drugs. Because of the unpredictable nature of this interaction, we have assessed in human volunteers the effect of these resins on the rate and total absorption of phenytoin. Our results indicate that these resins do not affect the absorption parameters of phenytoin and that special care does not have to be exercised when these agents are co-administered.

[Monitoring therapy by analysis of the drug concentration of saliva]

The distribution of a drug between blood and saliva depends on its physicochemical properties, e.g. binding to plasma proteins, apparent dissociation constant and lipid solubility. The clinical value of measuring salivary drug concentrations in therapeutic drug monitoring is demonstrated using anticonvulsant therapy in children as an example. For carbamazepine and phenytoin there is a close and constant saliva/serum ratio over a wide range of concentrations, which is influenced by the salivary flow rate only to an insignificant degree. Salivary concentrations of carbamazepine account for about 40%, of phenytoin for about 10% of the serum concentrations. In contrast, salivary levels of primidone and phenobarbital are significantly influenced by the rate of saliva flow. In resting saliva primidone levels slightly exceed those in serum and fall significantly below the corresponding serum values during forced stimulation of salivary flow. For phenobarbital in resting saliva the mean saliva/serum ratio is 0.3 and increases significantly during forced stimulation. Provided the conditions of sample collection are standardized saliva is suitable for monitoring primidone and phenobarbital therapy, too.

Reduction of phenytoin clearance caused by cimetidine

The effect of cimetidine on the disposition kinetics of phenytoin was investigated in 7 healthy volunteers. Each subject received a single intravenous dose of phenytoin on two occasions, in the control state, and during concurrent treatment with cimetidine 1 200 mg/day for 6 days. A slight but statistically significant decrease both in the rate of elimination and total body clearance of phenytoin was observed during the administration of cimetidine. The effect is probably due to inhibition of metabolism.

Metabolism of phenytoin in teratogenesis-susceptible and -resistant strains of mice

Phenytoin (PHT) is teratogenic in A/J but not in C57BL/6J mice. Teratogenesis in F1 hybrid offspring of these two strains is dependent on the susceptibility of the maternal parent. Hepatic microsomes from untreated pregnant A/J females produced more of both the phenolic and diol metabolites than did microsomes from pregnant C57BL/6J females. This difference disappeared when the animals were pretreated with phenobarbital or phenytoin. It seems unlikely that the genetic difference in susceptibility to PHT-induced teratogenesis can be explained on the basis of maternal metabolism of the drug.

Folic acid and the protective action of diphenylhydantoin against maximal electroshock in rats after single doses

Conflicting evidence regarding the antagonistic action of folic acid (FA) on the efficiency of the antiepileptic drug diphenylhydantoin (DPH) led us to study this problem in rats. From the results it is concluded that the enteral absorption of particularly higher doses of DPH is slightly hampered by FA. When the dose of DPH is high enough, this diminished absorption will not manifest itself in impaired protection. On the basis of DPH levels in brain tissue it is concluded that also on brain level there is some antagonistic action of FA on DPH protection. Here too, a sufficiently high level of DPH prevents this antagonistic action from becoming evident. We suggest that only in patients in whom the DPH treatment is marginal, FA treatment may hamper the therapeutic antiepileptic effect.