Effects of polytherapy with phenytoin, carbamazepine, and stiripentol on formation of 4-ene-valproate, a hepatotoxic metabolite of valproic acid.

The incidence of valproic acid hepatotoxicity has been reported to increase in patients who are receiving polytherapy. A minor valproic acid metabolite, 2-propyl-4-pentenoic acid (4-ene-VPA), formed by a cytochrome P450-mediated reaction, has been shown to be a potent inducer of microvesicular steatosis in rats. This study tested the hypothesis that formation of 4-ene-VPA would be increased in patients taking valproic acid with carbamazepine or with phenytoin but decreased with coadministration of an inhibitor of cytochrome P450 (the antiepileptic drug stiripentol in 300 to 1200 mg daily doses) in healthy subjects. Blood and urine samples in the studies were collected during a dosing interval at steady state. Valproic acid was assayed in plasma by capillary gas chromatography; valproic acid and 15 metabolites were measured in urine by gas chromatography/mass spectrometry. The formation clearance (CLf) of 4-ene-VPA was increased twofold in the valproic acid-carbamazepine and valproic acid-phenytoin groups. In the valproic acid/stiripentol studies, the CLf of 4-ene-VPA decreased by 32% in the 1200 mg/day stiripentol study. Similar findings were obtained at 600 and 300 mg/day stiripentol. These findings provide evidence supporting a role for cytochrome P450 in the formation of the hepatotoxic metabolite, 4-ene-VPA, in humans. The increased formation of 4-ene-VPA associated with carbamazepine and phenytoin is striking in relation to the epidemiologic finding of increased incidence of valproic acid-related hepatotoxicity during polytherapy with P450 inducers.

Human epileptic brain Na, K ATPase activity and phenytoin concentrations.

An abnormal flux of monovalent cations may be related to the epileptogenic process in man. One possible mechanism for deranged electrolyte metabolism in epileptic brain is an abnormality in sodium, potassium-dependent adenosine triphosphatase (Na, K ATPase). We found the activity of Na, K ATPase to be significantly less in epileptic human corfex than in nonepileptic cortex. Histological changes have been simultaneously evaluated in epileptic brain. A second membrane-bound enzyme, acetylcholinesterase (AChE), was also assayed as a marker for neuronal membranes and found not to correlate with the epileptogenicity of human brain. In addition, the concentrations of the anticonvulsant compound phenytoin have been determined in the serum and cerebral cortex of epileptic and nonepileptic patients. The ratio of phenytoin in cortex to serum concentration is significantly lower in epileptic patients than in nonepileptic controls.

Antiepileptic drug disposition during pregnancy.

A significant proportion of women with epilepsy have an increase in their seizure frequency during pregnancy. Multiple factors may be involved in this phenomenon, but changes in antiepileptic drug (AED) concentration appear to be the most significant. AED concentration declines as pregnancy progresses, due primarily to dynamic changes in plasma protein binding. Total concentrations of all first-line AEDs (carbamazepine, phenytoin, phenobarbital, and valproic acid) fall significantly during pregnancy, compared to baseline. Free or unbound drug concentrations, however, fall significantly only for phenobarbital. Valproate free concentrations actually increase by 25% by delivery. Women taking carbamazepine, phenytoin, or valproate may be relatively protected by adequate free concentrations of these compounds. When managing pregnant women with epilepsy, measurement of free AED concentrations and appropriate dose adjustment to maintain therapeutic ranges will permit more effective clinical management than using total concentration values.

Effect of doxepin on seizure frequency in depressed epileptic patients.

We performed a retrospective study of the effect of antidepressant treatment with doxepin on seizure frequency. From 47 patients treated with doxepin in 3 years, a confirmed seizure disorder and adequate follow-up (mean, 6.8 months) were available for 19 patients. Comparison of mean monthly seizure frequency during equal blocks of time before and during treatment with doxepin in doses of 5 to 400 mg/day demonstrated improved seizure control in 15, no change in 2, and increased seizures in 2 patients. The reduction in seizure frequency may have been a result of a direct antiepileptic effect of doxepin, an indirect effect because of improved affective state, drug interactions, or some combination of these mechanisms.

Serum prolactins in the diagnosis of epilepsy: sensitivity, specificity, and predictive value.

Serum prolactin levels rise after generalized tonic-clonic and partial complex seizures, but not after pseudoepileptic seizures. The criteria for a significant elevation in serum prolactin vary with individual investigators. The prevalence of pseudoseizures in the population studied determines the predictive value of serum prolactin determinations. In populations where most patients have epilepsy, a rise in serum prolactin is highly predictive for true epilepsy, but no increase in serum prolactin is not predictive for pseudoseizures.

Low and variable presence of valproic acid in human brain.

We studied the distribution of valproic acid (VPA) between brain (gray matter) and serum in 13 patients receiving chronic VPA therapy who underwent cortical resections for intractable seizures. Valproate concentration in cerebral cortex was remarkably low compared with either total or unbound valproate concentration in serum. The respective brain-to-serum partition ratios based on total and free drug in serum were 0.111 +/- 0.051 and 0.544 +/- 0.175. In comparison with other commonly used antiepileptic drugs, valproate has the distinction of exhibiting the lowest brain-to-blood partitioning. Moreover, the brain-to-serum concentration ratio varied over a four-fold range between patients. Some of this variability was related to variation in serum protein binding, as indicated by a modest correlation between the partition ratio and serum free fraction (r = +0.687). However, the brain-to-unbound concentration ratio still showed a three-fold variation. The variability in distribution of VPA between brain and blood is probably one of the underlying factors for the lack of a clearly definable therapeutic range of serum VPA concentration in epileptic patient populations.

Use of unbound drug concentrations to determine neonatal anticonvulsant exposure.

Unbound and total concentrations of several anticonvulsant drugs were measured by liquid chromatography in maternal and neonatal cord serum collected at birth from 16 women being treated for epilepsy and their newborns. Maternal and neonatal unbound drug concentrations agreed closely for phenobarbital (n = 6), phenytoin (n = 7), carbamazepine (n = 8), and its epoxide metabolite. Mean maternal total drug concentrations were higher than neonatal concentrations in the cases of phenobarbital, carbamazepine, its epoxide and diol metabolites. The differences were due to greater protein binding in maternal serum. Measurement of total anticonvulsant concentrations in newborns may be misleading, because of altered protein binding in the neonate. For the medications tested, neonatal and maternal exposures to unbound drug appear to be equivalent.

Human brain phenytoin: correlation with unbound and total serum concentrations.

Simultaneous brain (gray matter) and serum specimens from 18 patients treated with oral phenytoin, who underwent cortical resections for intractable seizures, were analyzed by HPLC. The correlation between brain phenytoin concentration and unbound (free) phenytoin (r = 0.98-0.99) was significantly better than the correlation between brain phenytoin and total serum phenytoin (r = 0.90-0.93). Phenytoin concentrations in tissue from epileptic foci were slightly lower than brain phenytoin concentrations in non-epileptic regions in the same patient (means 15.0 vs. 15.5 micrograms/g, P less than 0.05). The results support the value of monitoring unbound phenytoin in clinical situations where phenytoin binding is highly variable.

Pharmacokinetics of anticonvulsants in pregnancy: alterations in plasma protein binding.

Anticonvulsant levels decline as pregnancy progresses, even in the face of constant and, in some cases, increased dosages of medications. It has been suggested that this decline is responsible for the increase in seizure frequency seen in approximately one-third of the women with epilepsy who become pregnant. Changes in plasma protein binding may explain the declines in anticonvulsant concentrations during pregnancy. A prospective cohort study was designed to test this hypothesis. Carbamazepine, phenytoin and phenobarbital were studied. The mean total concentrations of all 3 drugs declined as pregnancy progressed, rising in the postpartum period. Free concentrations also declined, but did so significantly only for phenobarbital. The free fraction for all anticonvulsants studied rose significantly throughout pregnancy. Protein binding is significantly altered during pregnancy for all 3 drugs studied and appears to account for much of the decline in anticonvulsant concentrations seen in this condition. It is suggested that free rather than total drug concentrations be monitored in pregnant women with epilepsy.

Development of dosing guidelines for reaching selected target breath alcohol concentrations.

OBJECTIVE: This study evaluated gender-specific ethanol dosing protocols that were designed to result in one of two peak breath alcohol concentrations (BrACs)--0.07 or 0.10 g/2101. Inter- and intrasubject variability in BrAC were assessed and several possible methods for reducing variability in BrAC were evaluated. METHOD: Subjects (16 women, 16 men, ages 21-30 years) were studied after low (women 0.49 g/kg, men 0.53 g/kg consumed over 10 minutes) and high (women 0.81 g/kg, men 0.89 g/kg consumed over 20 minutes) ethanol doses, consumed following a 4-hour fast. All subjects were regular drinkers. RESULTS: Mean (+/-SD) peak BrACs actually achieved were 0.069+/-0.011 g/2101 after the low dose, and 0.105+/-0.014 g/2101 after the high dose. Mean values for peak BrAC, time to peak BrAC and area under the curve were not statistically significantly different between genders at either dose. BrACs varied by as much as twofold between subjects after equivalent gender and body weight adjusted doses. There was some reproducibility of ethanol pharmacokinetic parameters over dose and time in men, but not in women. CONCLUSIONS: The doses used resulted in equivalent mean ethanol exposures for women and men at each dose, with mean peak BrACs that closely approached the targets, but there was substantial inter- and intrasubject variability in ethanol pharmacokinetics.

Analysis of sertraline (Zoloft) and its major metabolite in postmortem specimens by gas and liquid chromatography.

We report methods for the analysis of sertraline and desmethylsertraline in postmortem biological fluids. The extraction method is based on a widely used procedure employing n-butyl chloride, and instrumental analysis is performed using GC/MS and HPLC with photodiode array detection. We report retention index, mass spectral, and UV-vis properties of the drug and its metabolite. Samples from three sertraline-related deaths were analyzed and revealed concentrations up to 10 times greater than the normal therapeutic levels, although two of the deaths were obviously the result of other causes. We also noted in one case that the drug concentrations in central and peripheral blood were very similar, suggesting that postmortem distribution may be uniform.

Three fatal drug overdoses involving bupropion.

Bupropion is a "second generation" antidepressant agent structurally related to the phenethylamines. Postmortem toxicology data are presented from three suicidal drug overdoses in which bupropion was detected. In two cases in which bupropion was the major toxicology finding, peripheral blood levels of the parent drug were 4.0 and 4.2 mg/L and total metabolite levels were 15 and 16.6 mg/L. Lethal doses in both cases were estimated by history to be less than 10 g.

Analysis of ketorolac in postmortem blood.

This paper describes analytical methods using high-performance liquid chromatography and gas chromatography-mass spectrometry (GC-MS) for the isolation of the nonsteroidal anti-inflammatory drug, ketorolac. The drug is isolated from postmortem blood using a batched solid-phase extraction method on Amberlite XAD-2 resin. Derivatization of ketorolac using diazopropane was necessary prior to GC-MS analysis. The methods were applied in the investigation of a death occurring shortly after the administration of an intramuscular injection of ketorolac tromethamine. Death was attributed to an adverse reaction to the drug, resulting in anaphylaxis and cardiac arrest.

An evaluation of the reliability of Widmark calculations based on breath alcohol measurements.

This study evaluated the reliability of Widmark calculations, based on breath ethanol reading (BrACs), for estimating the amount of alcohol consumed. A standard ethanol dose (males 0.51 g/kg; females 0.43 g/kg) was given to 115 college seniors, and BrACs were measured for two hours. Calculations of ethanol dose were performed using BrACs taken at 60, 75, 105, and 125 minutes after drinking. Mean calculated ethanol doses were lower than actual doses at each time point (P < 0.001). Mean underestimates were 13, 12, 15, and 14 mL of 100 proof vodka at 60, 70, 105, and 125 min after drinking. Calculated doses overestimated actual doses in 11, 10, 3, and 3 subjects at 60, 75, 105, and 125 min after drinking. The maximum overestimates were 13, 11, 6, and 8 mL of vodka at 60, 75, 105, and 125 min after drinking. At the 95% confidence level, the calculated dose at 105 and 125 min did not overestimate the true dose, but could underestimate it by as much as 30 mL vodka.

Problem of spurious correlation in the evaluation of steady-state carbamazepine levels using metabolite data.

The statistical problem of spurious correlation occurs when ratio variables with the same denominator are correlated. The correlation between two variables is also altered when heterogeneous groups are pooled. The problem is illustrated in the assessment of steady-state plasma levels of carbamazepine (CBZ), CBZ-10,11-epoxide (CBZ-epoxide), and 10,11-dihydro-10, 11-trans-dihydroxy-CBZ (CBZ-diol) in 12 epileptic patients on CBZ monotherapy and 12 epileptic patients on CBZ polytherapy. Expressing outcomes as dose/level ratios and metabolite level/parent drug level ratios is shown to be inappropriate, leading to spurious correlation. The amount of spurious correlation induced can be assessed by use of available statistical methodology, as is illustrated in this paper. More appropriate, alternative approaches are illustrated, using the data on the 24 patients with epilepsy.

Carbamazepine protein binding and disposition in pregnancy.

The pregnancies of five women with epilepsy treated with carbamazepine monotherapy were studied prospectively. Free and total serum concentrations of carbamazepine and its epoxide and dihydrodiol metabolites were analyzed at monthly intervals from the first trimester through 8 weeks postpartum. Assays were by high performance liquid chromatography, and free compounds were separated by ultrafiltration. The mean intrinsic clearance of carbamazepine (clearance of free drug corrected for changes in maternal body weight) did not change appreciably during pregnancy and the postpartum period. The mean free fractions of carbamazepine and the epoxide were elevated during pregnancy (0.25 and 0.50) compared with postpartum (0.22 and 0.43). Mean total maternal carbamazepine and epoxide concentrations were 40 and 48% higher than neonatal levels at birth, but maternal and neonatal free concentrations agreed closely. The ratio of epoxide to parent drug increased during pregnancy, as reported by other authors. Evidence is presented that this may be a result of inhibition of further biotransformation of the epoxide rather than of increased production. Two patients missed at least one dose of carbamazepine during labor, resulting in markedly reduced serum concentrations at delivery.

Variability of ethanol absorption and breath concentrations during a large-scale alcohol administration study.

Ethanol disposition was evaluated in 77 female and 97 male college seniors during an alcohol challenge study. All were regular drinkers who exceeded legal intoxication levels at least twice a month by history. A standard ethanol dose (females, 0.43 g/kg; males, 0.51 g/kg) was administered over 10 min, after a 4-hr fast, and breath alcohol concentrations (BrACs) were measured for 2 hr. Intersubject variability in BrACs was greatest early in the study, during ethanol absorption; the coefficient of variation decreased from 39% at 14 min to 14% at 125 min after the start of drinking. The time to peak BrAC varied from 10 to 91 min after the start of drinking (mean 39.6 min). Mean BrACs were significantly lower in females than males; mean peak BrACs were 0.054 g/210 liters in females and 0.058 g/210 liters in males (p = 0.031). The beta- and r-values for both genders were higher than those typically used in ethanol dose calculation formulas. Data are discussed to direct future research. The constants used in Widmark's formula need to be revised differentially for males and females in this population to reach specific target BrACs. Furthermore, substantial variability in absorption rates must be accounted for when assessing rising versus falling limb BrAC phenomena.

Valproic acid-phenytoin interaction.

The interaction between valproic acid (VPA) and phenytoin (DPH) was examined during therapeutic monitoring in an epileptic outpatient population. Gas-liquid chromatographic methods were used to measure DPH and VPA concentrations. (1) In 12 patients on stable DPH regimens, the mean DPH level declined from 19.7 to 15.3 microgram/ml when VPA was added (p less than 0.001). (2) In 20 patients receiving DPH and VPA, the median free fraction was 15.8%, compared to 9.1% free DPH in 40 patients receiving DPH only or DPH and phenobarbital (p less than 0.001). (3) Addition of VPA to a stable DPH regimen may result in a transient increased risk of DPH toxicity, followed by restabilization at the original free DPH level.

Single-dose pharmacokinetics and anticonvulsant efficacy of primidone in mice.

The pharmacokinetics and efficacy of the anticonvulsant primidone (PRM) and its active metabolites, phenobarbital (PB) and phenylethylmalonamide (PEMA), were studied after single-dose administration in mice. The half-life of PB is twice that of PRM and PEMA. The plasma/brain ratios provide evidence of poor penetration of PRM into brain. The results support our findings of negligible or absent PRM concentrations in the brains of patients on primidone therapy who were undergoing surgery for intractable epilepsy. The anticonvulsant properties of PRM, PB, and PEMA against maximal electroshock in mice were also studied with the use of the metabolic inhibitor SKF 525A. The half-life, potency, peak anticonvulsant effect, and effective dose curves of these compounds indicate that the anticonvulsant effect of short-term oral PRM administration in mice is from derived PB.