Antiepileptic Drugs, 5th Edition

Oxcarbazepine

45

Chemistry, Biotransformation, and Pharmacokinetics

Meir Bialer PhD, MBA

David H. Eisenberg Professor of Pharmacy, Department of Pharmaceutics, School of Pharmacy, Faculty of Medicine, The Hebrew University of Jerusalem, Jerusalem, Israel

Oxcarbazepine (OXC), 10,11-dihydro-10-oxo-carbamazepine or 10-oxo-carbazepine (Figure 45.1) is a 10-keto analog of carbamazepine (CBZ). OXC is a new antiepileptic drug (AED) that has been approved worldwide for the treatment of different kinds of partial-onset seizures and generalized tonic-clonic seizures (1, 2, 3, 4, 5). OXC was developed as a “second-generation” and follow-up compound to CBZ (1,2). The main advantage of OXC is its nonoxidative metabolic pathway, which implies lower induction potential and fewer drug interactions (3, 4, 5). Whereas CBZ undergoes oxidative metabolism to carbamazepine-10,11-epoxide (CBZ-E), OXC is rapidly and extensively reduced by cytosolic enzymes in the liver to its monohydroxylated derivative (MHD) (2) (Figure 45.1). Whereas CBZ metabolism to CBZ-E is mediated by cytochrome P450 (CYP) isoforms CYP3A4 and CYP2C8 (6) and is highly susceptible to induction and drug interactions, the biotransformation of OXC to MHD is catalyzed by reductases that are much less subject to enzyme induction (7, 8, 9). Thus, OXC can be regarded a soft drug analog of CBZ owing to its lack of oxidative metabolism (10). However, OXC is not only a soft drug analog of CBZ but a prodrug of MHD because in humans it undergoes an extensive presystemic first-pass conversion to MHD. Therefore, in humans MHD is in essence the active entity of OXC. In a study conducted in rats, no marked difference in anticonvulsant activity was found between CBZ, OXC, and racemic MHD (11,12).

CHEMISTRY AND METABOLIC SCHEME

Synthetic and Analytical Chemistry

OXC is a neutral lipophilic compound (molecular weight 252.3) with a melting point of 215°C to 216°C and low water solubility. The compound is prepared by hydrolysis of 10-methoxycarbazepine in dilute hydrochloric or sulfuric acid (13). The starting material, 10-methoxycarbazepine, is produced by reacting 10-methoxy-5H-dibenz[b,f]azepine with phosgene in toluene to give the 10-methoxy-5H-dibenzo[b,f]azepine-5-carbonyl chloride, which is then converted in ethanol with ammonia to the amide (13).

Methods of Determination

OXC is a prodrug of MHD, which is the principal active entity of OXC. Consequently, the published methods of determination for OXC actually measure plasma levels of MHD and in some cases OXC as well. The first methods were nonstereoselective high-performance liquid chromatography (HPLC) methods with a limit of quantification (LOQ) of 2 µmol/L or 0.5 mg/L (14,15). von Unruh et al. described a nonstereoselective gas chromatographic assay for OXC and its metabolites (16). Flesch et al. were the first to describe a stereoselective (enantioselective) method for the simultaneous determination of the two enantiomers, (S)-MHD and (R)-MHD, with an LOQ of 0.1 mg/L (17). Volosov et al. published an enantioselective HPLC method for monitoring the enantiomers of MHD and its metabolite carbamazepine-10,11-trans-dihydrodiol (DHD; Figure 45.1) (18). This method had a precision better than 15% for all analytes and a LOQ of 0.1 mg/L (serum) and 0.2 mg/L (urine) for each MHD enantiomer, and 0.4 mg/L (urine) for each of the DHD enantiomers. No enantiomeric interconversion occurred during the procedure of this assay. This method allows reliable determination of the MHD and DHD enantiomers in human urine

P.460

and was used recently in enantioselective pharmacokinetic studies of MHD in humans and dogs (19, 20, 21).

Metabolic Scheme

OXC undergoes rapid presystemic metabolic 10-keto reduction, mediated by cytosol arylketone reductase, to MHD (Figure 45.1), which is then partly conjugated with glucuronic acid and partly biotransformed to DHD before excretion in the urine (22, 23, 24, 25, 26). The first-pass reduction of OXC to MHD in humans is stereoselective, resulting in approximately a 1:4 area under the concentration-time curve (AUC) ratio between (R)-MHD and (S)-MHD (17,19,24,27). Because the phase II metabolite DHD has two chiral centers at positions 10 and 11, its trans-configuration exists in two enantiomeric forms, (R)-DHD and (S)-DHD (Figure 45.1) (19).

ABSORPTION

Bioavailability

OXC given orally as a tablet is almost completely absorbed in humans, and peak plasma concentrations of OXC and MHD occur 1 to 3 hours and 4 to 6 hours after dosing, respectively (3,7). The peak plasma concentration of (racemic) MHD was approximately five times higher than that of OXC, and the AUC of OXC was less than 4% of that of MHD (7,25,26,28,29). Oral doses of racemic MHD produced the same plasma profile as OXC (3).

P.461

When OXC was taken with a fat- and protein-rich breakfast, the AUC of MHD increased by 16% and the peak plasma concentration by 23% (30). In one study, 12 healthy (6 male and 6 female) adult volunteers received OXC orally (300-mg tablet) and racemic MHD intravenously (i.v.; 250 mg infused over 30 minutes) (27). After i.v. administration of racemic MHD, the mean [± standard deviation (SD)] AUC values of (R)-MHD and (S)-MHD were 120 ± 26 µmol/L/hr and 167 ± 37 µmol/L/hr., respectively (27). After oral administration of OXC to the same healthy subjects, the enantiomeric ratio of the mean AUC values of (S)-MHD (241 ± 55 µmol/L/hr) over (R)-MHD (64 ± 20 µmol/L/hr) equaled 3.8 (27). An enantiomeric AUC ratio of 4.9 was obtained after oral administration of OXC (600 mg) to 12 healthy Chinese subjects (19). These two studies demonstrate an enantiospecific metabolic reduction of the prochiral carbonyl group of OXC in its biotransformation to MHD (19,27). The ratio (normalized to the dose) of the total AUC values for both MHD enantiomers obtained after oral (OXC) and i.v. (racemic MHD) administration (27) to the same subjects gives a mean absolute bioavailability (oral availability) value of 89% (Table 45.1). The oral availability of MHD after oral and i.v. administration (400 mg) of racemic MHD to six dogs was 78% ± 21% and 79% ± 27% for (R)-MHD and (S)-MHD, respectively (20).

Formulations and Routes of Administration

OXC is commercially available (Trileptal; Novartis, Summit, NJ) as a regular film-coated, divisible tablet at dosage strengths of 150, 300, and 600 mg and as an oral suspension at a concentration of 60 mg/mL. Because of its water insolubility, there are no parenteral preparations of OXC. However, MHD is being developed in its racemic form as a new AED for parenteral administration to supplement OXC oral therapy because of its better water solubility compared with OXC and CBZ.

DISTRIBUTION

Calculations of MHD volume of distribution (Vβ) based on the only published study where racemic MHD was administered intravenously (250 mg) to six healthy subjects yielded mean values of 11.7 L and 13.8 L for (R)-MHD and (S)-MHD, respectively (27) (Table 45.1). Higher Vβ

P.462

values were obtained after i.v. administration (400 mg) of racemic MHD to six dogs: 25 ± 6 L for (R)-MHD and 47 ± 14 L for (S)-MHD (20). The apparent volume of distribution (Vβ/F) of OXC obtained after its oral administration (600 mg) to eight healthy subjects was very high and variable, 12.5 ± 12.9 L/kg (26) (Table 45.1).

The plasma protein binding of MHD is approximately 40% and is constant at clinically relevant concentration ranges of 20 to 150 µmol/L (31). OXC is approximately 60% bound at a plasma concentration range of 0.2 to 11.4 µmol/L. OXC and MHD are excreted in breast milk, with a milk-plasma concentration ratio of 0.5 (32,33). OXC and MHD cross the placenta. The transfer of OXC through the perfused placenta was quicker than the transfer of antipyrine, whereas the transfer of MHD was slower (34). OXC is biotransformed to some extent in human placenta in vitro, suggesting that the placenta also might be a metabolic site for OXC in vivo (34). As neutral lipophilic compounds, OXC and MHD pass rapidly through biologic membranes, including the blood-brain barrier (1).

ROUTES OF ELIMINATION

OXC undergoes rapid and extensive metabolism to MHD by a stereoselective biotransformation mediated by a cytosolic, nonmicrosomal, and noninducible arylketone reductase. MHD is eliminated from the body by metabolism, and the metabolic scheme of OXC and MHD is depicted in Figure 45.1 (24). Each of the formed MHD enantiomers is excreted in the urine or undergoes glucuronide conjugation or subsequent oxidation to the respective DHD enantiomers (19,24) (Figure 45.1). After oral administration (600 mg) of OXC to 12 healthy subjects, approximately 27% of the molar dose of OXC was recovered (free and conjugated) in the urine (within 48 hours after dosing) as the enantiomers of MHD: (R)-MHD 4.0% ± 2.1% and (S)-MHD 22.6% ± 8.2% (19). The urinary recovery of the DHD enantiomers accounts for less than 1%, mostly in unconjugated form.

After oral administration (400 mg) of ¹⁴C-labeled MHD to two healthy subjects, most of the dose (94.6% and 97.1%) was excreted in the urine within 6 days after dosing. Fecal excretion in the two studied subjects accounted for 4.3% and 1.9%, respectively (24). In this study, 28% and 71% of the dose was excreted in the urine as free and conjugated MHD enantiomers, respectively (24). Analysis of the diastereoisomeric MHD glucuronide shows that approximately 6% is accounted for by (R)-MHD glucuronide and approximately 45% by (S)-MHD glucuronide (Figure 45.1). Approximately 4% and 9% of the dose appeared in the urine as sulfate and glucuronide conjugates, respectively, of 10-hydroxy carbamazepine (24). Thus, the involvement of the CYP isozyme family in the metabolism of OXC is quite minimal and is limited to the formation of DHD. Consequently, to date there have been no reports of genetic polymorphism in OXC metabolism.

CLEARANCE AND HALF-LIFE

The half-life of OXC is 1 to 3.7 hours, and its AUC after oral administration to healthy subjects amounted to 2% to 4% of the AUC for MHD, indicating that OXC has a high oral clearance value of 2.4 ± 1.1 L/hr/kg (25,26,28,29). After multiple dosing to six epileptic patients (daily dose, 1.5 to 2.4 g), the oral clearance (mean ± SD) of OXC was 1.6 ± 0.5 L/hr/kg and its half life was 3.1 ± 1.5 hours (26). After i.v. administration (250 mg infused over 30 minutes) of racemic MHD to 12 healthy subjects, the clearance (calculated from the AUC data) and half-life of (R)-MHD and (S)-MHD were 4.2 ± 0.9 L/hr and 9.0 ± 1.5 hours, and 3.0 ± 0.7 L/hr and 10.6 ± 2.6 hours, respectively (27) (Table 45.1). After oral administration of OXC (300 mg) to the same 12 healthy subjects, the half-life values for (R)-MHD and (S)-MHD were 15.8 ± 2.8 hours and 11.2 ± 1.5 hours, respectively (27).

In another study, the half-life of MHD enantiomers obtained after oral administration of OXC (600 mg) to 12 Chinese healthy subjects was 11.9 ± 3.3 hours and 13.0 ± 4.1 hours for (R)-MHD and (S)-MHD, respectively (19). The renal clearance of the two MHD enantiomers ranged between 0.7 to 1.1 L/hr, with no enantioselectivity (19,27). After i.v. administration of racemic MHD to healthy subjects, 12% and 16% of the dose was excreted in the urine as (R)-MHD and (S)-MHD, respectively (27). After oral administration of OXC to healthy subjects, 4% and 23% of the dose was excreted in the urine as (R)-MHD and (S)-MHD, respectively (19,27).

When MHD is given i.v., the AUC of (S)-MHD is 40% higher than that of (R)-MHD because its clearance is significantly smaller than that of its enantiomer. After oral administration of OXC, the enantioselective pharmacokinetics of MHD are much more profound because of enantioselective presystemic metabolic ketoreduction of the prochiral carbonyl group of the OXC molecule (Figure 45.2). The enantiomeric serum concentrations ratio of (S)-MHD over (R)-MHD increases from a value of 3 at 1 hour after dosing to 5.8 at 48 hours after dosing (19) (Figure 45.2). The observation that the ratio between the (S)- and (R)-enantiomers in serum increased over time shows that differences in elimination (clearance) of the enantiomers contribute to the higher serum levels of (S)-MHD. However, differences in clearance are too small (40%) to explain the striking difference (400%) in the AUC values of MHD enantiomers after oral administration of OXC. Indeed, the observation that at the first sampling time (1 hour after dosing) the concentration of (S)-MHD was three times greater than that of (R)-MHD (Figure 45.2) strongly suggests that the differences in their kinetic profiles are related mainly to

P.463

stereoselectivity in the formation clearance of MHD after oral administration of OXC.

There are no reports on MHD clearance in comedicated epileptic patients, children, or the elderly. The half-life of MHD was similar in patients and healthy subjects (7). In healthy subjects aged 60 to 82 years, the AUC and peak plasma concentration values of MHD were significantly higher than in younger adults, probably because of the diminished creatinine clearance in older persons (35). MHD concentrations in children 6 to 18 years of age are similar to those observed in adults, but those reported in children 2 to 5 years of age have been lower (36,37).

Hepatic impairment has no effect on the pharmacokinetics of OXC and MHD. However, MHD plasma levels increased significantly in patients with creatinine clearance of 30 mL/min. In this group of patients, the dosage of OXC should be reduced by 50% and the dosage titration should be prolonged (38).

RELATIONSHIP BETWEEN SERUM CONCENTRATION AND DOSE

Studies in healthy subjects and epileptic patients showed a linear, proportional relationship between daily doses of OXC and serum concentrations of MHD (3,39). After oral administration of MHD (600 mg) to healthy subjects, serum levels of (S)-MHD and (R)-MHD ranged from 1 to 4.5 mg/L and 0.2 to 1 mg/L, respectively (19). In other (nonstereospecific) studies, the peak plasma concentration of MHD and OXC obtained after oral administration of a single 600-mg dose of OXC to healthy subjects ranged from 5.4 to 8.9 mg/L (MHD) and 1.1 to 1.7 mg/L (OXC) (26,28,29,37,40) (Table 45.1). After multiple dosing OXC (300 mg twice daily) to 24 young (18 to 32 years of age) and old (60 to 81 years of age) healthy subjects, the accumulation of MHD was found to be more than expected based on linear pharmacokinetics (35). No significant difference was observed between male and female volunteers. The following MHD plasma levels (mean ± SD) were obtained in this study: young men, 8.5 ± 2 mg/L; elderly men, 12 ± 1.2 mg/L; young women, 9.3 ± 1.1 mg/L; and elderly women, 11.7 ± 2.0 mg/L. The peak-to-trough fluctuation of (racemic) MHD ranged from 22% to 43% (35).

According to the manufacturer's prescribing information, treatment with OXC was well tolerated in epileptic patients receiving dosages of 1,200 mg/day. At dosages of 2,400 mg/day, more than 65% of the patients discontinued treatment mainly because of central nervous system-related side effects. Consequently, assuming linear kinetics, it is expected that OXC daily doses of 1,200 mg will yield MHD steady-state plasma levels in a range of 12 to 25 mg/L.

RELATIONSHIP BETWEEN SERUM CONCENTRATION AND EFFECT

In a review article on therapeutic drug monitoring of major AEDs, the target range of MHD plasma concentrations associated with antiepileptic effect was reported to be 5 to 50 mg/L or 20 to 200 µmol/L (41).

CONCLUSION

OXC is a soft drug analog of CBZ and a prodrug to MHD. Thus, MHD is the active entity in OXC therapy. Because of species differences in the pharmacokinetics of OXC, this finding was discovered in the late stages of OXC development. Otherwise, it might be worthwhile from a pharmacokinetic and biopharmaceutic standpoint to develop MHD (in its racemic or stereospecific form) rather than OXC. The two enantiomers of MHD showed similar median effective dose (ED₅₀) values in animal models for anticonvulsant activity

P.464

(11,12). The stereoselective pharmacokinetics of MHD in dogs (20,21) and humans (19,27) and the different pharmacokinetic profile and exposure between animals and humans suggest that the in vivo anticonvulsant activity of MHD be assessed in terms of median effective concentration (EC₅₀) rather than ED₅₀. Only after a comparative analysis of the EC_50Sand the toxicologic profile of the two MHD enantiomers, as well as a thorough stereospecific pharmacokinetic-pharmacodynamic evaluation, can a decision be made regarding the possible development of an individual MHD enantiomer as a new drug candidate.

REFERENCES

1. Dam M, Ostergaard LH. Oxcarbazepine. In: Levy RH, Mattson RH, Meldrum BS, eds. Antiepileptic drugs,4th ed. New York: Raven Press, 1995:987-995.
2. Grant SM, Faulds D. Oxcarbazepine: a review of its pharmacology and therapeutic potential in epilepsy, trigeminal neuralgia and affective disorders. Drugs1992;43:873-888.
3. Lloyd P, Flesch G, Dieterle W. Clinical pharmacology and pharmacokinetics of oxcarbazepine. Epilepsia1994;35[Suppl 3]: S10-S13.
4. Perucca E. The new generation of antiepileptic drugs: advantages and disadvantages. Br J Clin Pharmacol1996;42:531-543.
5. Baruzzi A, Albani F, Riva R. Oxcarbazepine: pharmacokinetic interactions and their clinical relevance. Epilepsia1994;35[Suppl 3]:S14-S19.
6. Kerr BM, Thummel KE, Wurden CJ, et al. Human liver carbamazepine metabolism: role of CYP3A4 and CYP2C8 in 10,11epoxide formation. Biochem Pharmacol1994;47:1969-1979.
7. Schachter SC. Oxcarbazepine. In: Eadie MJ, Vajda FJE, eds. Antiepiletic drugs: pharmacology and therapeutics. Handbook of experimental pharmacology.Berlin: Springer-Verlag, 1999: 319-330.
8. Spina E, Pisani F, Perucca E. Clinically significant pharmacokinetic drug interactions with carbamazepine. Clin Pharmacokinet1996;31:198-214.
9. Leach JP, Blacklaw J, Jamieson V, et al. Mutual interaction between remacemide hydrochloride and carbamazepine: two drugs with active metabolites. Epilepsia1996;37:1100-1106.
10. Bodor N. Soft drugs. In: Encyclopedia of human biology.San Diego: Harcourt Brace Jovanovich, 1991;7:101-107.
11. Kubova H, Mares P. Anticonvulsant action of oxcarbazepine, hydroxycarbamazepine, and carbamazepine against metrazol-induced motor seizures in developing rats. Epilepsia1993;34: 188-192.
12. Schmutz M, Ferrat T, Heckendorn R, et al. GP47779, the main human metabolite of oxcarbazepine (Trileptal), and both enantiomers have equal anticonvulsant activity.Epilepsia1993;34 [Suppl 2]:122.
13. U.S. Patent 3,642,775, W. Schindler assigned to Ciba Geigy, 1972.
14. Elyas AA, Goldberg VD, Patsalos PN. Simple and rapid micro-analytical high-performance liquid chromatographic technique for the assay of oxcarbazepine and its primary active metabolite 10-hydroxycarbazepine. J Chromatogr Biomed Appl1990;528:473-479.
15. Hartley R, Green M, Lucock MD, et al. Solid phase extraction of oxcarbazepine and its metabolites from plasma for analysis by high performance liquid chromatography.Biomed Chromatogr1991;5:212-215.
16. von Unruh GE, Paar WD. Gas chromatographic assay for oxcarbazepine and its main metabolites in plasma. J Chromatogr Biomed Appl1985;345:67-76.
17. Flesch G, Francotte E, Hell F, et al. Determination of the R-(-) and S-(+) enantiomers of the monohydroxylated metabolite of oxcarbazepine in human plasma by enantioselective high-performance liquid chromatography. J Chromatogr1992;581:147-151.
18. Volosov A, Bialer M, Xiaodong, et al. Simultaneous stereoselective high-performance liquid chromatographic determination of 10-hydroxycarbazepine and its metabolite carbamazepine-10,11-trans-dihydrodiol in human urine. J Chromatogr B Biomed Sci Appl2000;738:419- 425.
19. Volosov A, Xiaodong S, Perucca E, et al. Enantioselective pharmacokinetics of 10-hydroxycarbazepine after oral administration of oxcarbazepine to Chinese subjects. Clin Pharmacol Ther1999;66:547-553.
20. Volosov A, Sintov A, Bialer M. Stereoselective pharmacokinetic analysis of the antiepileptic 10-hydroxycarbazepine in dogs. Ther Drug Monit1999;21:219-223.
21. Volosov A, Yagen B, Bialer M. Comparative stereoselective pharmacokinetic analysis of 10-hydroxycarbazepine after oral administration of its individual enantiomers and racemic mixture to dogs. Epilepsia2000;41:1107-1110.
22. Feldmann KF, Dorhofer G, Faigle JW, et al. Pharmacokinetics and metabolism of GP 47779, the main human metabolite of oxcarbazepine (GP 47680) in animals and human volunteers. Adv Epileptol1981:89-96.
23. Feldmann KF, Brechbuhler S, Faigle JW, et al. Pharmacokinetics and metabolism of GP 47680, a compound related to carbamazepine, in animals and man. Adv Epileptol1978:290-294.
24. Schutz H, Feldmann KF, Faigle JW, et al. The metabolism of 14C-oxcarbazepine in man. Xenobiotica1986;8:769-778.
25. Tarata A, Galimberti CA, Mani R, et al. The pharmacokinetics of oxcarbazepine and its active metabolite 10-hydroxycarbazepine in healthy subjects and epileptic patients treated with phenobarbital and valproic acid. Br J Clin Pharmacol1993;36:366-368.
26. Dickinson RG, Hooper WD, Dunstan PR, et al. First dose and stead-state pharmacokinetics of oxcarbazepine and its 10hydroxy metabolite. Eur J Clin Pharmacol1989;37:69-74.
27. Flesch G, Czendlik C, Ehrhart F, et al. Pharmacokinetics of two enantiomers of the monohydroxy derivative of oxcarbazepine in man. Eur J Pharm Sci1999;8:xxii(abstr).
28. Keranen T, Jolkkonen J, Jensen PK, et al. Absence of interaction between oxcarbazepine and erythromycin. Acta Neurol Scand1992;86:120-123.
29. Keranen T, Jolkkonen J, Klosterskov-Jensen P, et al. Oxcarbazepine does not interact with cimetidine in healthy volunteers. Acta Neurol Scand1992;85:239-242.
30. Degen PH, Flesch G, Cardot JM, et al. The influence of food on the disposition of the antiepileptic oxcarbazepine and its major metabolites in healthy volunteers. Biopharm Drug Dispos1994; 15:519-526.
31. Patsalos PN, Elyas AA, Zakrewska JM. Protein binding of oxcarbazepine and its primary active metabolite, 10-hydroxy carbazepine, in patients with trigeminal neuralgia. Eur J Clin Pharmacol1990;39:413-415.
32. Bulau P, Paar WD, von Unruh GE. Pharmacokinetics of oxcarbazepine and 10-hydroxycarabazepine in the newborn child of an oxcarbazepine-treated mother. Eur J Clin Pharmacol1988;34: 311-313.
33. Bar-Oz B, Nulman I, Koren G, et al. Anticonvulsant and breast feeding: a critical review. Pediatr Drugs2000;2:113-126.

P.465

34. Pienmimaki P, Lampeal E, Hakkola J, et al. Pharmacokinetics of oxcarbazepine and carbamazepine in human placenta. Epilepsia1997;38:309-316.
35. van Heiningen PNM, Eve MD, Oosterhuis B, et al. The influence of age on the pharmacokinetics of the antiepileptic agent oxcarbazepine. Clin Pharmacol Ther1991;50:410-419.
36. Sallas W, Hossain M, D'Souza J. Population pharmacokinetic analysis of oxcarbazepine (Trileptal) in children with epilepsy. Epilepsia1999;40[Suppl 7]:102(abstr).
37. Schachter SC. The next wave of anticonvulsants: focus on levetiracetam, oxcarbazepine and zonisamide. CNS Drugs2000;14: 229-249.
38. Rouan MC, Lecaillon JB, Godbillon J, et al. The effect of renal impairment on the pharmacokinetics of oxcarbazepine and its metabolites. Eur J Clin Pharmacol1994;47:161-167.
39. Augusteijn R, van Parys JAP. Oxcarbazepine (Trileptal, OXC): dose concentration relationship in patients with epilepsy. Acta Neurol Scand Suppl1990; 133:37.
40. Jung H, Noguez A, Mayet L, et al. The distribution of 10hydroxy carbazepine in blood compartments. Biopharm Drug Dispos1997;18:17-23.
41. Glauser TA, Pippenger CE. Controversies in blood-level monitoring: reexamining its role in the treatment of epilepsy. Epilepsia2000;41[Suppl 8]:S6-S15.