A number of new antiepileptic drugs act by indirect mechanisms and thus produce effects that may not best be measured by traditional blood studies of the drugs and their metabolites. Study of the indirect action of these drugs on GABA-mediated inhibition by microdialysis and nuclear MR spectroscopy has proved more relevant. These new investigative techniques may also prove valuable as compounds affecting glutamate or other excitatory neurotransmitters are developed.

Department of Neurology; New Haven, CT 06510, U.S.A. Connecticut VA Medical Center; West Haven, CT 06516, U.S.A.

Background: Two prospective observations of adults with symptomatic, localization-related (partial) epilepsy included 1,102 patients in VA multicenter studies (VA-118; and VA-264). Analyses assessed the likelihood of remaining seizure free for 12 and 24 months after initiating adequate antiepileptic drug therapy. Methods: Patients were grouped as having only secondarily generalized tonic-clonic seizures (GTC), only complex partial seizures (CPS), or both types (MIXED) at entry. The cumulative proportion of patients remaining seizure free with standard antiepileptic drug therapy was determined by actuarial life table methods. Results: At 12 months, 70% and 61% of GTC patients (VA-118 and VA-264, respectively) had no further GTC; 53% and 50% of MIXED, predominantly GTC patients had no further GTC, 21% and 28% of CPS patients had no further CPS and 98% and 91% were seizure free for GTC; 32% and 35% of MIXED, predominantly CPS patients had no further CPS, and 62% and 51% of patients with MIXED seizure types remained seizure free for CPS for 12 months after enrollment. Conclusions: The overall prognosis for control of seizures of any type for 12 months was best for those who had only GTC at entry (55% and 48%), worst for those who had only CPS at entry (23% and 26%), and intermediate for those with MIXED seizures at entry (32% and 25%) (all p < 0.0001). Prognosis can be based on the predominant seizure type in patients with multiple types.

Rapid turnaround testing of carbamazepine (CBZ) concentration helps the clinician assess patients efficiently. The Biotrack (BT) therapeutic drug monitor uses an automated turbidimetric latex agglutination assay of hemolyzed whole blood and reports the corresponding serum drug concentration within 3 min of sample application. We compared BT results for 53 patients with serum CBZ measurements made with the Abbott TDx fluorescence polarization immunoassay (FPIA) system. Mean BT concentration was 9.5 μg/ml (range, 2.8–18.3 μg/ml); one sample was reported “below 2.0 μg/ml” (FPIA = 1.1 μg/ml). Mean FPIA concentration of the 52 BT-quantifiable samples was 8.7 μg/ml (range, 2.8–18.0 μg/ml). Correlation of BT with FPIA was high (r = 0.96). BT results had a bias of 0.8 μg/ml [95% confidence intervals (CI) = 0.6 to 0.9 μg/ml]; median absolute error relative to FPIA was 0.6 μg/ml. Variations within the normal range for hematocrit, albumin, and creatinine had minimal effect on accuracy. As compared with a high-pressure liquid chromatographic (HPLC) assay, the BT assay showed less cross-reactivity than did FPIA to increased concentrations of the CBZ metabolite, carbamazepine-10,11-epoxide (CBZE). Assays performed on samples from 15 other patients not receiving CBZ were all <2.0 μg/ml. The BT assay is sufficiently accurate for clinical monitoring of CBZ concentrations

Summary: The Biotrack 516 is a simple, automated whole blood phenytoin (PHT) assay that reports corresponding total serum concentrations in 3 min. We compared Biotrack results in 58 patients with the total and unbound serum PHT concentrations measured by the standard TDx fluorescence polarization immunoassay. Correlation with total TDx concentration was high (r = 0.98); median absolute error was 1.4 μg/ml. Correlation of unbound PHT with Biotrack (r = 0.95) was comparable to correlation of unbound and total TDx (r = 0.94). The Biotrack assay is a promising method for clinical monitoring of PHT concentrations.

Population-based pharmacokinetic prediction algorithms have been developed for several medications. A fundamental assumption has been that the kinetics remain constant over time. Carbamazepine (CBZ), however, induces its own metabolism in a concentration- and time-dependent manner. A Bayesian estimation program is presented that models the changing catabolic enzyme activity, linearly related to hepatic microsomal enzyme concentration, along with the serum drug concentration. An Emax model is used for enzyme formation with respect to drug concentration: elimination of enzyme activity is modeled as a first-order process. This program was tested in 22 drug-naive outpatients begun on CBZ monotherapy. The 1 week concentrations were used to prospectively predict concentrations at 1 month of therapy and were very close to actual measurements: prediction bias (mean error of prediction) = -0.1 micrograms/mL and precision (median absolute error of prediction) = 1.2 micrograms/mL. Comparison estimates, made by assuming a constant concentration/dose ratio, had bias = 2.6 micrograms/mL (p < 0.001) and precision = 2.2 micrograms/mL (p = 0.01). We conclude that (1) CBZ autoinduction is not complete after 1 week of therapy and (2) the methodology permits accurate estimation of CBZ pharmacokinetics.

We report the first study of carbamazepine and carbamazepine-10,11-epoxide concentrations determined by using intracerebral microdialysis in three patients undergoing depth electrode studies for the evaluation of medically intractable epilepsy. Very small microdialysis catheters, affixed to and inserted with the depth electrodes, sampled drug concentration in the extracellular environment. We perfused artificial extracellular fluid continuously, and varied the perfusion rate to permit estimation of the absolute drug concentration in the extracellular space. Serum samples were obtained simultaneously. The relation between dialysate and extracellular concentration (recovery fraction) depended, in vivo but not in vitro, on the relative lipophilicity of the compounds, suggesting that diffusion of the drug within the brain is a major determinant of microdialysate drug concentration. When this is taken into account, the steady-state extracellular concentrations of these compounds closely mirror their unbound serum concentrations.

Intracerebral microdialysis has become a standard method for neurochemistry studies and is becoming recognized as an important new method for pharmacological studies. The technique permits repeated measurement of drug concentrations in the brain extracellular fluid with minimal disturbance of the neuronal environment. The sample volumes are exceedingly small, on the order of tens of microliters. This is countered by the purity of the samples, reducing the need for extraction prior to assay. We present a HPLC assay capable of reliably measuring the concentrations of the antiepileptic drug carbamazepine and its metabolites carbamazepine-10,11-epoxide and carbamazepine-10,11-trans-diol.

We report the first human study of phenytoin concentration using in vivo microdialysis, which permits sampling the extracellular environment of the brain. This technique has been applied to patients undergoing intracranial electrode investigation for intractable epilepsy. By varying the rate of perfusion (from 2.5 to 0.25 μl/min), it is possible to quantify the concentration of drug in the extracellular fluid (ECF), which reflects the concentration on the outer neuronal cell membrane. Samples were obtained from four catheters in two patients, in whom serum phenytoin (PHT) concentrations were held constant. Unbound serum concentrations were measured following ultrafiltration at 37°C. In one patient, with left and right hippocampal probes, steady state ECF/unbound serum ratios were 87 and 84% respectively. In the second patient, with hippocampal and frontal probes, ECF/unbound serum ratios were 87 and 85% respectively. Flow rate for 50% maximal recovery averaged 1.65 μl/min (1.5–1.7 μl/min). We found that steady state ECF PHT concentrations corresponded closely to unbound serum concentrations. No differences are observed between different sites within the brain. Flow rates needed for equilibration of dialysate with the extracellular space were slower than reported for carbamazepine, but faster than those we found for carbamazepine-epoxide and valproate.

Phenytoin (PHT) administration is complicated by saturation kinetics within the therapeutic range, causing marked changes in drug concentration with small changes in dose. The “half-life” increases with concentration, varying from 8–24 hr up to weeks, making it difficult to obtain the steady state levels needed by most prediction algorithms and nomograms. A Bayesian prediction program (Epidose) is presented which explicitly models PHT absorption and elimination kinetics in the non-steady state. The algorithm accounts for the interdependency of closely spaced sequential samples. Estimates of future PHT concentration were made on 20 hospital inpatients, most of whom were acutely ill and received other medications. Future (x = 4 day) PHT concentrations were predicted over a range from 4 to 22 μg/ml (mean 13.9 μg/ml) with a median absolute error of 1.0 μg/ml. These data demonstrate that the program can be used for accurate PHT concentration predictions in sick patients.

Valproate (VPA) is present in humans and is largely bound to protein. Only free drug is metabolized, and antiepileptic and toxic effects are probably related to free concentrations. By measuring serial free and total serum VPA levels after routine oral doses, we have determined individual in vivo protein binding parameters for 37 patients after a total of 49 separate drug administrations. Binding site concentrations and dissociation constants were fitted using a nonlinear algorithm. On sole VPA (n = 28) the mean dissociation constant was 91 mumol/L, and the mean concentration of binding sites was 1,176 mumol/L. Evidence suggests a second, nonsaturable binding site. Fraction of unbound VPA ranged from 5.4% at low levels up to 38.7%, rising with increasing total concentration. Concurrent therapy with phenytoin (n = 7) or carbamazepine (n = 8) did not cause displacement of VPA. Changes in free fraction were consistently observed during the interdose interval. The data demonstrate that the binding changes are not a factor in decreased VPA levels during coadministration of other antiepileptic drugs.

Adherence to prescribed drug dosing regimens declined substantially during the interval between clinic visits and drug level tests. Using microelectronic monitors to observe pill-taking habits, 20 patients averaged 88% compliance before and 86% compliance after the visit, but this dropped to 67% compliance a month later. These data indicate that spot drug levels do not represent long-term "steady-state" drug serum concentrations.