Phenobarbital does not alter phenytoin steady-state serum concentration or pharmacokinetics.

Phenytoin pharmacokinetics and biotransformation were studied with stable isotope tracer techniques in six patients before and after addition of phenobarbital. No significant (p less than 0.05) changes in phenytoin serum concentration, clearance, elimination half-life, volume of distribution, or clearance via production of p-hydroxyphenyl-phenylhydantoin or phenytoin dihydrodiol occurred after addition of phenobarbital. Thus, frequent phenytoin serum concentration determinations or a change in phenytoin dosing rate are probably not necessary after adding phenobarbital.

Carbamazepine increases phenytoin serum concentration and reduces phenytoin clearance.

Addition of carbamazepine to phenytoin monotherapy resulted in the following significant (p less than 0.05) changes: (1) increased mean phenytoin serum concentration; (2) decreased phenytoin clearance, due to decreased production of phenytoin dihydrodiol and p-hydroxyphenyl-phenylhydantoin; (3) increased phenytoin elimination half-life; and (4) increased drug-related toxicity. Close monitoring is required after addition of carbamazepine to phenytoin.

Vigabatrin for refractory complex partial seizures: multicenter single-blind study with long-term follow-up.

The irreversible GABA transaminase inhibitor vigabatrin (VGB) was given in a single-blind fashion to 89 patients with complex partial seizures (CPS) refractory to conventional drugs. The median number of CPS per month decreased from 11.0 to 5.0 after addition of VGB, and 51% of patients had a 50% or greater decrease in CPS frequency (p less than 0.001). Side effects (principally drowsiness, ataxia, and headache) occurred mainly during the initiation of therapy and decreased during therapy. After 12 weeks on VGB, side effects significantly interfered with functioning in only 13% of patients, and the efficacy:toxicity ratio warranted continued administration in 74% of patients. Coadministration of VGB resulted in a mean decrease of 20% in phenytoin serum concentration (p less than 0.001). Sixty-six patients with a favorable response to VGB during the single-blind study have been followed for a median of 16.7 months on VGB. No serious systemic or neurologic toxicity has been detected, and most patients have retained their initial favorable CPS control.

Methsuximide for complex partial seizures: efficacy, toxicity, clinical pharmacology, and drug interactions.

Methsuximide (MSM; Celontin) was administered for 8 weeks to 26 patients with complex partial seizures (CPS) refractory to phenytoin and carbamazepine and phenobarbital or primidone. A 50% or greater reduction in CPS frequency was obtained in eight patients. MSM therapy was continued chronically in these eight patients, and five continued to have a 50% or greater reduction in CPS frequency after 3 to 34 months of follow-up. Drowsiness, gastrointestinal disturbance, hiccups, irritability, and headache were the common side effects of MSM. No serious toxicity occurred. N-desmethylmethsuximide was the principal substance detected in plasma and had the following pharmacokinetic values: accumulation half-life, 49.7 hours; time to steady state, 10.4 days; elimination half-life, 72.2 hours; therapeutic range of plasma concentration, 10 to 30 mg per liter. Plasma concentrations of phenytoin and phenobarbital derived from primidone rose significantly (p less than 0.05) after addition of MSM.

Chronic benzodiazepine administration. IX. Attenuation of alprazolam discontinuation effects by carbamazepine.

Clinical studies suggest that carbamazepine may attenuate effects of alprazolam discontinuation. Since discontinuation of chronic alprazolam in a mouse model is associated with behavioral alterations and upregulation at the gamma-aminobutyric acidA (GABAA) receptor, we studied the effects of carbamazepine administration after alprazolam (2 mg/kg/day) discontinuation. Open-field activity was increased in mice 4 days after alprazolam discontinuation, but this effect was reduced significantly by continuous infusion of carbamazepine, 25 or 100 mg/kg/day. Benzodiazepine receptor binding in vivo was increased in cortex at 2 and 4 days after alprazolam discontinuation, and in hypothalamus at 4 days; with carbamazepine, 100 mg/kg/day, binding in both regions at these time points was similar to control values. Similar results were observed in cortex with benzodiazepine receptor binding in vitro. GABA-dependent chloride uptake was also increased at 4 days alprazolam administration. Treatment with carbamazepine attenuated (P less than 0.10) this increase. Carbamazepine alone after vehicle did not alter benzodiazepine binding or GABA-dependent chloride uptake. These results indicate that carbamazepine administration after alprazolam discontinuation attenuates behavioral and neurochemical alterations associated with discontinuation.

Alternatives to least squares linear regression analysis for computation of standard curves for quantitation by high performance liquid chromatography: applications to clinical pharmacology.

Standard curves and validation points for high-performance liquid chromatography (HPLC) determination of four drugs (carbamazepine and phenytoin at therapeutic drug monitoring concentrations and deuterium labeled carbamazepine and phenytoin at tracer dose concentrations) were computed using standard least squares linear regressions analysis and six alternative regression techniques (weighted 1/x, 1/y, 1/x2, 1/y2 least squares linear, log/log least squares linear, and robust). The coefficient of determination (R2) and the coefficient of prediction (R2pred) values for standard curves and the computed values for validation points did not differ significantly among the seven methods. The lower limit of quantitation (LLQ) values obtained with all six of the alternative regression methods were significantly (P < .01) lower than the LLQ values obtained with least squares linear regression analysis. The lowest LLQ values were obtained with 1/x2 and 1/y2 weighting and were threefold to tenfold less than the values obtained with unweighted least squares linear regression analysis (P < .001). The authors conclude that alternative regression analysis techniques (especially 1/x2 and 1/y2 weighting) offer significant advantages for clinical pharmacology studies when concentration values being measured by HPLC are near the LLQ of the method determined by unweighted least squares linear regression analysis. In other situations, alternative forms of regression analysis had no significant advantages in our study.

Performance of human mass balance/metabolite identification studies using stable isotope (13C, 15N) labeling and continuous-flow isotope-ratio mass spectrometry as an alternative to radioactive labeling methods.

Stable isotope labeling in therapeutic and subtherapeutic quantities of drug (15N2(13)C-phenobarbital) can be quantitated in biological matrices (urine) and high performance liquid chromatography (HPLC) peaks from urine using continuous-flow isotope-ratio mass spectrometry (CF-IRMS). Standard curves for 15N2(13)C-phenobarbital were reproducible and linear (R2 > 0.985) over the ranges of 3-100 micrograms/ml for whole urine (15N2 or 13C labeling) and 0.1-8.0 micrograms/mL for HPLC peaks derived from urine (15N2 labeling). The lower limit of quantitation values for urine drug concentration was 0.46-2.62 micrograms/mL in whole urine and 0.10-0.70 micrograms/mL in HPLC peaks. Validation samples quantitated with these standard curves yielded close to expected values. These data suggest stable isotope labeling and CF-IRMS may be used as an alternative to 14C labeling and radioactivity counting methods in mass balance/metabolite identification and other biomedical studies.

Performance of human mass balance studies with stable isotope-labeled drug and continuous flow-isotope ratio mass spectrometry: a progress report.

We propose performing human mass balance studies by administering stable isotope labeled (13C or 15N) drug and quantitating excess (above background) 13C or 15N in urine, serum, and feces by continuous flow-isotope ratio mass spectrometry (CF-IRMS). Theoretical calculations and empirical data (dynamic range, linearity, sensitivity, precision, accuracy) are presented to establish that commercially available CF-IRMS instruments can quantitate stable isotope labeled (one or two 15N or 13C labels) drug concentrations of 1.0 microg/mL or greater in urine, serum (15N), or feces. More than two 13C labels may be necessary to quantitate 1.0 microg/mL of drug in serum. Three volunteers received 650 mg of 15N13C2-acetaminophen, and urine was collected for 72 hours. Percent of administered label recovered in urine from the three subjects was 97.4, 78.9, and 95.4 for 13C and 90.3, 77.0, and 90.6 for 15N. Fecal recovery of label for one subject was 0.9% (13C2) and 1.1% (15N). Serum pharmacokinetic values obtained by counting 13C or 15N in one subject were as expected for acetaminophen. This method appears to be promising, and further validation is ongoing.

Studies with stable isotopes II: Phenobarbital pharmacokinetics during monotherapy.

Six healthy adults receiving no other medications were given tracer doses of 90 mg of stable isotope-labeled phenobarbital (PB) intravenously before, and four weeks after, and 12 weeks after beginning therapy. Serum samples were collected for 96 hours after each injection, and the concentration of stable isotope-labeled PB in each sample was determined by gas chromatographic mass spectrometry. The volume of distribution, elimination half-life, and total clearance of PB did not differ significantly on any of the three occasions measured. Phenobarbital clearance did not correlate significantly with total PB serum concentration. Clearances determined from single-dose studies before beginning PB therapy accurately predicted steady-state PB serum concentrations. Therefore, it is not necessary to adjust PB dosage for time-dependent or dose-dependent changes in clearance during monotherapy. In addition, clearance or serum concentration determined at one dosing rate directly predicts serum concentration at another dosing rate.

New pharmacokinetic methods: I. Estimation of mean serum concentration from trough serum concentration.

An equation is derived to estimate mean steady state serum concentration (Css) from trough steady state serum concentration (Cmin) which can be used for drugs with either linear or nonlinear pharmacokinetic properties. In 15 subjects receiving phenytoin monotherapy, estimated Css did not differ significantly from measured Css, while Cmin differed significantly (P less than .0001) from measured Css and estimated Css. Clearance (CL) and elimination half-life (t1/2) values determined by stable isotope tracer methods or by standard equations and measured Css or estimated Css did not differ significantly, while CL and t1/2 values calculated with standard equations and Cmin differed significantly (P less than .02) from values obtained by any of the other three methods. We conclude: 1) Cmin values and CL and t1/2 values calculated with Cmin values may differ significantly from Css values and CL and t1/2 values calculated with Css; 2) accurate estimates of Css and of CL and t1/2 can be obtained using our procedure to correct a Cmin value to an estimated Css value.

New pharmacokinetic methods. II: Determination of the presence or absence of product inhibition in drugs with nonlinear pharmacokinetics.

We show that for drugs metabolized by one enzyme the slope of a plot of serum concentration (C) versus 1/clearance (CL) is linear with a value of 1/Vmax in the presence of substrate saturation and may be linear (rarely) or curved (usually) with a slope always greater than 1/Vmax in the presence of substrate saturation and product inhibition (competitive, noncompetitive, or uncompetitive) when the serum concentration of product varies with the serum concentration of substrate. Serum concentration, CL, and Vmax were determined for a group of six subjects receiving phenytoin monotherapy using three approaches. With each approach: 1) a plot C versus 1/CL was linear (r greater than or equal to 0.738, P less than .01); 2) the slope of this regression line did not differ significantly (P greater than .30) from 1/Vmax. We conclude: 1) our method is a simple and useful method for determination of mechanism of a drug's nonlinear pharmacokinetics (substrate saturation versus substrate saturation and product inhibition), 2) phenytoin has nonlinear pharmacokinetics due to substrate saturation only in man.

New pharmacokinetic methods. III. Two simple test for "deep pool effect".

If a portion of administered drug is distributed into a "deep" peripheral compartment, the drug's actual elimination half-life during the terminal exponential phase of elimination may be longer than determined by a single dose study or a tracer dose study ("deep pool effect"). Two simple methods of testing for "deep pool effect" applicable to drugs with either linear or nonlinear pharmacokinetic properties are described. The methods are illustrated with stable isotope labeled (13C15N2) tracer dose studies of phenytoin. No significant (P less than .05) "deep pool" effect was detected.

Estimation of the elimination half-life of a drug at any serum concentration when the Km and Vmax of the drug are known: calculations and validation with phenytoin.

A method of estimating the elimination half-life of a drug at any selected serum concentration when the Km and Vmax of that drug are known is described. The method was validated in six patients by determining their Km and Vmax values for phenytoin, using data obtained at two serum concentrations and then using the Km and Vmax values to predict the elimination half-life of phenytoin at a third serum concentration. The predicted elimination half-lives at the third serum concentration did not differ significantly (t = 1.16, P = .30) from observed elimination half-lives determined with stable isotope tracer techniques.

Determination of the in vivo Km and Vmax of a drug with tracer studies.

A method is described for determining the in vivo Km and Vmax values for a drug, utilizing a variant of the Michaelis-Menten equation and the elimination-rate constants of tracer doses of the drug administered at two different serum concentration values. The new method is the most accurate yet devised; unlike older methods, it is dependent on assumptions that generally are true and can be validated empirically. Application of the method is demonstrated by a study of phenytoin in humans.

Studies with stable isotopes I: Changes in phenytoin pharmacokinetics and biotransformation during monotherapy.

Six patients were given tracer doses of 13C15N2-phenytoin (PHT) before and four and 12 weeks after beginning monotherapy. The following significant (P less than .05) changes occurred during monotherapy: (1) Apparent (from tracer doses) PHT total clearance by linear method decreased; (2) apparent PHT elimination half-life increased; (3) apparent mean PHT serum concentration per unit dose increased; (4) apparent rate of excretion of p-hydroxyphenyl-phenylhydantoin (p-HPPH) decreased; (5) apparent rate of excretion of PHT dihydrodiol increased; and (6) apparent PHT total clearance and elimination half-life and apparent p-HPPH rate of excretion were dose dependent. Phenytoin apparent pharmacokinetic and biotransformation values undergo a typical series of changes after beginning monotherapy at typical dosing rates, because PHT's dose-dependent pharmacokinetics result in differing apparent values as the serum concentration rises to steady state. Stable isotope methods are particularly suitable for investigating such phenomena.

Studies with stable isotopes III: Pharmacokinetics of tracer doses of drug.

Stable isotope labeled tracer doses of phenytoin (PHT) and phenobarbital (PB) were given intravenously before and four and 12 weeks after beginning monotherapy in two groups of six patients. Phenytoin demonstrated nonlinear pharmacokinetics, while PB demonstrated linear pharmacokinetics. Each of the 36 sets of tracer dose serum concentration versus time data points appeared linear during the elimination phase on semilog plots, and each demonstrated a high degree of linearity using semilog regression analysis (r2 = .977-.999, P less than .001, for PHT; r2 = .791-.996, P less than .005, for PB). We conclude tracer doses administered at steady-state serum concentration will exhibit linear serum concentration versus time relationships on semilog plots regardless of whether the steady-state serum concentration is in the linear or the nonlinear portion of a drug's dose versus steady-state serum concentration relationship. The mechanism and implications of this conclusion are discussed.

Staggered stable isotope administration technique for study of drug distribution.

By infusing intravenously a series of different stable isotope labeled forms of a drug at different times prior to performing a single collection of a body fluid via puncture (or tissue via biopsy), one can obtain the same information about a drug's distribution as would be obtained by infusing a single dose of drug and performing serial collections of fluid (or tissue). Application of this technique of "staggered stable isotope administration" is illustrated with a study of entry of phenobarbital into the cerebrospinal fluid of a dog. Advantages and disadvantages of this technique are discussed.

Pharmacokinetics: dose-dependent changes.

Stable-isotope tracer methods are described for answering the following questions about a drug: Does the drug exhibit dose-dependent changes in pharmacokinetic properties? What are the drug's Michaelis constant (Km) and maximum velocity (Vmax) for enzymatic biotransformation; Are the dose-dependent pharmacokinetic changes great enough to be clinically important? and Which routes of the drug's biotransformation are responsible for the drug's dose-dependent pharmacokinetic properties? Illustrative data are provided from tracer studies performed with a drug with dose-dependent pharmacokinetic properties, phenytoin, and a drug that does not exhibit dose-dependent pharmacokinetic properties, phenobarbital. The advantages and disadvantages of the described stable-isotope methods are discussed.

Bioavailability studies of drugs with nonlinear pharmacokinetics: I. Tracer dose AUC varies directly with serum concentration.

The authors show that for a drug cleared by one enzyme the area under the serum concentration-time curve from time 0 to infinity (AUC0-INF) of a test dose can be expressed as AUC0-INF = TD x F x (Km + C)/Vmax where TD is test dose size, F is fraction absorbed, C is drug serum concentration at the time of the study, and Km and Vmax are the Michaelis constant and maximum velocity of the enzyme. This equation predicts the AUC0-INF produced by a given tracer dose of drug will vary directly with C in drugs with nonlinear pharmacokinetic properties (i.e., drugs whose value for C approaches or exceeds Km) if C is held constant by administration of tracer and maintenance doses of drug. The AUC0-INF produced by intravenous tracer doses of 150 mg of 13C15N2-sodium phenytoin was determined in 15 subjects at 30 different values of C. AUC0-INF showed a high degree of direct linear correlation with C (AUC0-INF (ug x hr/mL) = 35.4 + 8.1 x C (ug/mL), r = 0.885, P < .0001). Consequences of this observation for relative bioavailability studies of drugs with nonlinear pharmacokinetic properties are discussed.

Absence of pharmacokinetic drug interaction of levetiracetam with phenytoin in patients with epilepsy determined by new technique.

Levetiracetam has recently been approved as an adjunctive medication for partial seizures and frequently will be added to phenytoin. The objective of this study was to determine the presence or absence of a pharmacokinetic drug interaction of levetiracetam with phenytoin. A stable isotope tracer technique using deuterium-labeled (D10) phenytoin and high-performance liquid chromatography with ultraviolet detection (rather than mass spectrometric detection) was employed. Tracer doses of D10-phenytoin were administered i.v. before and 12 weeks after adding levetiracetam to the regimen of 6 subjects on phenytoin monotherapy for epilepsy. Blood was collected for 96 hours after each infusion. The following pharmacokinetic parameters were determined for phenytoin: Cmax, Cmin, Cavo, AUC, CL, t 1/2, VD, and free (nonprotein bound) fraction. The ratio and the 90% confidence interval of the ratio of log-transformed mean values for phenytoin pharmacokinetic parameters before (denominator) and after (numerator) adding levetiracetam all fell within the range of 0.85 to 1.17 (two one-sided test). The authors conclude that the addition of levetiracetam did not bring about clinically important changes in phenytoin pharmacokinetic parameters and that it is not necessary to change the phenytoin dosing rate when levetiracetam is added to phenytoin.