Multiple-dose pharmacokinetic study with a slow-release carbamazepine preparation

Multiple-dose pharmacokinetic study with a slow-release carbamazepine preparation

Epilepsy Res., 6(1990) 126-133 126 Elsevier EPIRES 00329 Multiple-dose pharmacokinetic study with a slow-release carbamazepine preparation M. Reuna...

822KB Sizes 0 Downloads 72 Views

Epilepsy Res., 6(1990) 126-133

126

Elsevier EPIRES 00329

Multiple-dose pharmacokinetic study with a slow-release carbamazepine preparation M. Reunanen a, E. Heinonen b, M. Anttila b, P. J/irvensivu b, H. Lehto b and E. Hokkanen a aDepartment of Neurology, Universityof Oulu, Oulu (Finland), and bFarmosGroup Ltd., Research Center, Turku (Finland) (Received 6 October 1989; revision received 5 January 1990; accepted 8 January 1990)

Key words: Epilepsy; Carbamazepine; Pharmacokinetics; Slow release; Side effects

The pharmacokinetics, clinical efficacy and side effects of carbamazepine (CBZ) in the steady-state condition were studied using a slow-release preparation (SR), Neurotoi ® Slow, and a conventional preparation (C), Tegretol®. Eighteen adult epileptic patients under CBZ therapy were evaluated in this single-blind, randomized cross-over study. The previous daily CBZ dose was kept unchanged and divided into 2 daily doses during two 2 week study periods. At the end of each period blood samples were drawn at frequent intervals for 12 h after the administration of the morning CBZ dose. Serum concentrations of unchanged CBZ and its main metabolite, carbamazepine.10,1 l-epoxide (CBZE), were determined by HPLC. Peak concentrations of CBZ and CBZE were signficantly lower, and the time-lapse before CBZ reached its peak was significantly longer during SR treatment. The fluctuations in serum CBZ and CBZE were significantly lower during SR treatment. There was no significant difference in bioavailability between the 2 preparations. The number of epileptic seizures was 31 during SR and 57 during C treatment. Side effects were more common during C treatment. The occurrence of dizziness was significantly lower with SR treatment than with C treatment. We conclude that greater stability in serum CBZ and CBZE concentrations can be obtained by using an SR of CBZ, without reducing the bioavailability of the drug.

INTRODUCTION Carbamazepine (CBZ) has established its position as an important antiepileptic agent, especially in the treatment of partial seizures with or without secondary generalization 26. CBZ may, however, cause some side effects in clinical use, especially if the serum levels are too high. In long-term thera-

Correspondence to: Dr. Mauri Reunanen, Department of Neurology, Oulu University Central Hospital, 90220 Oulu, Finland.

py, side effects such as dizziness, nausea, fatigue, diplopia and headache may occur. These side effects have been associated with variations in the serum CBZ concentrations 27. Slow-release CBZ preparations (SR) have been developed in order to achieve more stable serum CBZ concentrations than those obtained with conventional CBZ preparations (C). The aim of this study was to compare the pharmacokinetics of a new SR (Neurotol ® Slow) with a C (Tegretol ®) at steady-state conditions in patients with partial seizures. In addition, the occurrence of epileptic seizures and side effects of the treatment were registered.

0920-1211/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

127

samples. Thus, 18 patients could be evaluated, 10 females and 8 males. Their characteristics are presented in Table I. Most of the patients had complex partial seizures (Table III). The mean CBZ dose was 644 + 200 mg (300-1200 rag) daily, divided into 2 doses as evenly as practically possible (mean 311 + 108 mg in the morning and 333 + 97 mg in the evening, difference 4.5%). The largest individual difference was in the dose of patient no. 18, who received 100 mg in the morning and 200 mg in the evening (Table I). Eleven patients were on CBZ monotherapy. In 7 patients adequate seizure control was not obtained by CBZ monotherapy and other antiepileptic drugs had to be added to the treatment (Table I). Besides antiepileptic drugs, one patient (no. 5) was receiving medication for asthma, one patient (no. 12) for arterial hypertension and hypothyreosis, and one patient for arterial hypertension and depression (no. 9). One patient (no. 10) was taking an antidepressant drug and one patient (no.

MATERIALS AND METHODS Patients Epileptic patients who h a d b e e n on stable C B Z t h e r a p y for at least 3 m o n t h s fulfilled the inclusion criteria. Patients with severe somatic or psychiatric disease, renal or hepatic insufficiency, and p r e g n a n t patients were excluded from the study. Prior to the study, E S R , hemoglobin, hematocrit, erythrocytes, leukocytes, serum creatinine, S G P T , S G O T , alkaline p h o s p h a t a s e , serum calcium a n d phosphate were determined. I n f o r m e d consent was obtained from the patients before the study, which was c o n d u c t e d according to the principles of the Declaration of Helsinki. T h e study comprised 21 epileptic patients with simple or complex partial seizures. O n e patient d r o p p e d out for personal reasons, one patient was excluded because of technical difficulties in the c h r o m a t o g r a p h i c analyses and one patient was excluded because of an insufficient n u m b e r of blood

TABLE I Characteristics of the patients in the study Patient no.

Age (years)

Sex

1

27

M

2 3

32 51

M F

4

34

F

5 6 7

43 38 37

F F M

8

55

M

9 10 11 12 13 14 15 16 17 18

49 62 39 58 56 35 40 49 32 36

F M F M M F F M F F

Mean + S.D.

42 10

Duration of antiepileptic treatment (months)

CBZ treatment (months)

Daily CBZ dose (mg) Morning dose

Evening dose

36 27 66 120

36 27 39 66

300 300 200 600

300 300 300 600

132 174 15 96 218 300 24 52 24 7 138 5 120 72

132 93 16 91 218 3 24 52 22 7 46 5 111 20

200 300 300 400 400 400 300 300 300 300 200 300 400 100

200 300 300 400 500 400 300 300 300 300 300 300 400 200

90 80

56 55

311.1 107.9

333.3 97.0

Other antiepileptic medication during the study clonazepam clonazepam valproate phenytoin phenytoin phenytoin valproate clonazepam -

128 17) a neuroleptic drug. The following commercially available CBZ products were used in this study: SR: Neurotol ® Slow 200 mg tablet (L/i~ikefarmos/Farmos Group Ltd., Finland); C: Tegretol ® 200 mg tablet (CibaGeigy, Switzerland). The design was a single-blind, randomized, cross-over trial. The drugs were given in identical containers for 2 weeks during each study period, and the observer did not know which drug was being administered Both were administered twice daily at 12 h intervals. The patients' previous daily CBZ dose was kept unchanged. Serum CBZ and carbamazepine-10,11-epoxide (CBZE) concentrations were determined from the morning sample (at 8.00 a.m., before the administration of the drugs) prior to the study and 1 week after initiation of each treatment period. Additionally, at the end of each treatment, a series of blood samples was drawn. In this series, the sample was drawn at the time of administzation of the drug, and subsequent samples were drawn after 0.5, I and 2 h and thereafter at intervals of 2 h up to 12 h (until 8.00 p.m.). During the study all epileptic seizures and subjective side effects reported by the patients were recorded. Special attention was paid to the occurrence of dizziness, fatigue, movement disorders and visual disturbances. Additionally, at the end of the study, the patients were asked which period they had preferred. Serum CBZ and CBZE concentrations were determined by high-performance liquid chromatography 15.

Pharmacokinetic calculations The pharmacokinetic parameters for CBZ and CBZE were calculated as follows: AUC0_ n h was calculated as the area under the concentrationtime curve by the trapezoidal rule over the 12 h interval; Cmax was determined as the highest observed concentration during the 12 h study interval; and tma x w a s determined as the time at which Cm~.~ occurred. When adjacent concentrations were equal to Cmax the median time was taken as tm.~; Cmin was taken as the lowest observed concentration during the study interval; C ss is defined as the time-averaged concentration at steady-state and was calculated by using the AUC0_~2h divided by the dosing interval (12 h). ADc~s is defined as

the deviation in the area-under-the-curve from C ~s (the areas above and below C~) and was calculated by the trapezoidal rule s. Furthermore, Cmax-Cmin/Css and ADc~JAUC0_ n h were used to describe the fluctuation of serum concentrations. M r is defined as the metabolite fraction and was calculated as the ratio of the area under the curve for the metabolite CBZE relative to that for the parent drug CBZ over the 12 h study period.

Statistical analysis The statistical comparisons of the pharmacokinetic parameters of CBZ and CBZE were performed using an analysis of variance for repeated measurements 12, extracting effects due to sequence, period and treatment, tma x values were compared by the Wilcoxon signed rank test. The differences in the number of epileptic seizures and the occurrence of side effects between the treatment periods were calculated with Wilcoxon's signed rank test and McNemar's test. The correlations between the clinical findings and the pharmacokinetic data were carried out with Spearman's rank test and the Mann-Whitney U test. Deviations from the mean were expressed by the standard deviation (+ S.D.). BMDP statistical software was used for the statistical analysis. RESULTS

Pharmacokinetic results Carbamazepine. CBZ was more rapidly absorbed from C than from SR. Table II shows the pharmacokinetic parameters of CBZ and CBZE during both treatments. The peak concentration of CBZ (Cmax) was higher (P < 0.05) for C than for SR, and tmax occurred later (P < 0.05) for SR than for C. Cmax -Cmi,/C ss was 28% smaller (P < 0.001) for SR than for C. ADcss (the area deviation from averaged steady-state serum CBZ levels) was 35% smaller (P < 0.001) during SR treatment than during C treatment (Fig. 1). Also, ADcsJ AUC0-12 h was significantly (P < 0.00001) smaller during SR. The AUC0_12 h, minimum concentration (Cmi,) and the metabolite fraction (Mr) did not differ significantly between the two preparations. When only patients on carbamazepine monotherapy (n

129 TABLE II

Mean + S.D. pharmacokinetic parameters of carbamazepine (CBZ) and carbamazepine-lO, l l-epoxide (CBZE) at steady-state after administration of CB Z twice a day as slow-release (SR) and conventional (C) tablets in 18 adult epileptic patients C s' = the time-averaged CBZ concentration at steady state; ADcss = the area deviations from (AUCcBzE/AUCcBz).

Parameters

CBZ

Mr

70.3 4.7 6.5 5.0 5.9 4.5 0.26 0.06

the metabolite fraction

CBZE

SR AUC0_~2 h (~g h/ml) tmax*(h ) Cmax~ug/ml ) Cmin(gg/ml) CSS0:~g/ml) ADc~s 0~g h/mi) Cmax-Cmin/C ss ADt,,,dAUCo_I., ~

CS~; M r =

C + 22.0 + 1.5 + 2.0 + 1.6 + 1.8 _+ 2.3 + 0.08 +- 0.02

0.15 + 0.06

75.5 3.5 7.3 5.1 6.3 7.0 0.37 0.10

+_ 20.6 + 1.1 + 1.8 + 1.6 _.+ 1.7 _+ 1.8 -+ 0.11 -+. 0.03

Stat. diff. P

SR

C

Stat. diff. P

0.147 0.011 0.026 0.724 0.147 0.0001 0.0001 <0.0001

10.2 + 5.3 * 1.0 + 0 . 5 0.7 + 0 . 4 0.9 +_0.4 0.6 + 0.5 0.27 +- 0.11 0.06 __ 0.02

11.5 + 5.4 * 1.1 + 0 . 5 0.8 + 0 . 4 1.0 +-0.5 0.9 + 0.6 0.30 __ 0.11 0.07 +_ 0.03

0.009

0.15 -+ 0.06

0.01 0.094 0.010 0.003 0.089 0.006

0.265

* tmax of CBZE could not be determined due to the plateau form of the curve.

- 11) were considered, ADcss was 36% smaller during SR (4.25 + 1.68 gg h/ml) than during C treatment (6.65 + 1.61/~g h/ml, P < 0.01). Cmax Cmin/Cs~was also smaller (P < 0.01) and tma~longer (5.0 + 1.9 h and 3.5 + 1.1 h, P < 0.05) during the SR treatment. The CBZ concentrations of the morning samples, taken 1 week after the beginning of each pe-

A

B

riod, were 6.0 + 1.9 gg/ml for SR and 5.3 +_ 1.0 gg/ml for C (the difference is not significant, P < 0.1). Carbamazepine-lO,11-epoxide. Cmax of CBZE was significantly (P < 0.01) smaller during SR than during C treatment (Table II). ADcss and ADc,dAUC0-~2 h were also significantly lower (P < 0.01) with SR (Fig. 2). AUC0_12h of CBZE was significantly lower (P < 0.01) during SR than during C treatment, tmax values could not be calculated because of the plateau form of the CBZE curve.

The CBZE concentrations in the morning samples were 0.97 _+ 0.73 gg/ml during SR and 0.83 +_ 0.39 gg/ml during C treatment (the difference was not significant).

Clinical observations

50~

TIME (h}

TIME (h)

Fig. 1. The steady-state concentration of carbamazepine in serum within 12 h interval during treatment with slow-release (A) and conventional (B) preparation. The daily carbamazepine dosage was divided into two doses. The fluctuation (ADcss) of concentration is significantly (P < 0.001) smaller during A than during B treatment.

During C treatment, patients had 12 generalized seizures, 25 complex partial and 8 simple partial seizures, totaling 56 seizures, whereas during SR treatment there were 2 generalized, 18 complex and 11 simple partial seizures, totaling 31 seizures (Table III). This difference did not, however, reach statistical significance (P = 0.093, Wilcoxon's signed rank test).

130 A

B

~. 11

10 09E; I

0.8. 07-

0

1 2

4

6

8

10

12

0

1 2

4

6

TIME(h) Fig.

2.

8

10

12

TIME (h)

T h e s t e a d y - s t a t e c o n c e n t r a t i o n o f c a r b a m a z e p i n e - 1 0 , 1 1 - e p o x i d e in s e r u m w i t h i n 12 h i n t e r v a l d u r i n g t r e a t m e n t w i t h slow-re-

lease ( A ) a n d c o n v e n t i o n a l (B) p r e p a r a t i o n , T h e f l u c t u a t i o n (ADc~s) o f c o n c e n t r a t i o n is s i g n i f i c a n t l y ( P < 0.01) s m a l l e r d u r i n g A t h a n during B treatment.

Side effects were reported 12 times during SR and 19 times during C treatment (Table IV). Seven patients complained of dizziness during C and

TABLE

only one patient during SR treatment (difference statistically significant, McNemar's test P = 0.034). Twelve patients reported no side effects

III

Number of epileptic seizures before and during both treatment periods SP - s i m p l e partial; C P = c o m p l e x p a r t i a l ; G P = p a r t i a l s e i z u r e s w i t h s e c o n d a r y g e n e r a l i z a t i o n ,

Patient no.

Before the study

During the study

2 weeks

Conventional (2 weeks)

SP

CP

m

GP

SP

CP

Slow.release (2 weeks) GP

SP

CP

1

m

u

~

~

~

m

GP

4

-

2

-

-

10

12

-

11

5

-

2

-

-

1

-

-

2

1

6

-

2

-

-

2

-

-

2

-

8

-

-

-

1

-

-

1

-

-

10

.

11

-

0

-

12

.

.

.

.

.

.

.

.

.

13

.

.

.

.

.

.

.

.

.

14

.

.

.

.

.

.

.

.

.

15 16

.

-

.

.

3

.

.

17 18

8 .

.

18 .

1

.

.

Total

8

M e a n _+ S . D .

1.2 + 2,2

.

. 2

. -

5

.

-

.

14

0

.

.

-

19 3.1 + 6 . 6

. 2

. -

10 -

25

.

. 10 .

-

.

11

18

. 12

. -

.

-

1.7 + 3.5

131 TABLE IV

Side effects reported by patients during treatment with carbamezepine (n = 18) SR Dizziness Fatigue Disturbances of vision Headache Disorder of coordination Nausea Mild gastric discomfort Total

1 ( 5.6%)* 4 (22.2%) 2 (11.1%) 2 (11.1%) 1 (5.6%) 1 (5.6%) 1 (5.6%) 12

C 7 (38.9%)* 4 (22.2%) 4 (22.2%) 1 (5.6%) 3 (16.7%) 19

*P = 0.034, McNemar's test,

during SR and 9 during C. The laboratory values taken before entrance into the study were within the normal range in all patients. At the end of the study the patients were asked which product they preferred: 8 patients had preferred SR and 5 C; 6 patients considered the products equal and I patient had no opinion (P ffi 0.4, Kruskal-Wallis test). Those patients who had most seizures during the trial preferred SR (the mean number of seizures in patients preferring SR was 2.4, and 0.3 in those preferring C). No significant correlations were found between fluctuations in CBZ concentrations and number of seizures during either treatment; there was, however, a clear positive, statistically significant (P < 0.05) correlation between the fluctuations in serum CBZE concentrations (ADcss) and occurrence of seizures during the trial during C (the correlation coefficient (r) was 0.479, P < 0.05) but not during SR (r = 0.385, P < 0.1). During SR there were no significant differences in the pharmacokinetic values for CBZ or CBZE between the patients reporting side effects and those without any, whereas during C the Cma~, Cmin and ADcs~ of CBZE of patients with side effects were much higher than those of patients without side effects, although the differences were not significant, despite displaying a clear trend (P = 0.07, 0.08 and 0.09 respectively). DISCUSSION Carbamazepine has established its positic i as an

antiepileptic drug, especially for the treatment of part'al seizures 26. CBZ is also effective in the treatment of trigeminal 7 and glossopharyngeal neuralgias zl, alcohol withdrawal symptoms 5 and acute mania ~s. The half-life of CBZ is rather long (20-65 h) in pharmacokinetic studies carried out with healthy volunteers 3. In patients on CBZ therapy, the halflife is markedly shorter (10-15 h). When used together with other antiepileptic drugs (primidone, phenytoin, etosuximide, phenobarbital) the halflife is still shorter, 5-13 h ~°. This is considered to be due to hepatic enzyme induction stimulating the elimination of CBZ. The autoinduction of CBZ is potentiated by other antiepileptic drugs. As the half-life becomes shorter, the fluctuations of serum CBZ concentrations appear. Dividing the daily dose into 3 or 4 doses decreases the fluctuation. In a study by Ghose et al. n the fluctuation was 73% when the CBZ dose was given once/day, but 60% or 33% respectively when given 2 or 3 times daily. Another method to reduce the fluctuations is to use SRs. To our knowledge our study is the first single-blind study comparing the pharmacokinetic properties of C and SR in patients who are on steady-state CBZ therapy. In our study SR provided slower absorption than C, with lengthening of tmax and a smoother rise in CBZ concentration. The peak concentration was also significantly lower. The fluctuations in serum CBZ were significantly (36%) smaller during the treatment with SR. Larkin et al. 17 compared the pharmacokinetics of a C (Tegretol, Ciba-Geigy Ltd.) and SR (Tegretol CR Divitabs, Ciba-Geigy Ltd.) in a doubleblind study, giving CBZ to healthy volunteers for 2 weeks. SR rendered 12% fewer fluctuations than C. Decrease in fluctuation has also been reported in open trials with some other slow or retard forms of CBZ L2z'z3. In our study, fluctuations in CBZE were also significantly less during SR, the difference was 20.7% (ADcsdAUC0_n h)" The differences in the fluctuations of CBZ and CBZE were also significant in patients with CBZ monotherapy. It is noteworthy that the bioavailability was not significantly different between the two preparations. In some slow-release CBZ preparations a

132 slow rate of absorption has been achieved, but with reduction of bioavailability by up to 25°~ 9'14'16 making it necessary to increase the daily dose of CBZ. Many investigators have tried to define a limit of CBZ concentration beyond which the side effects begin to emerge. According to HOppener et al. 13, this limit is about 32/xmol/l (8 gg/ml), and according to Riva et al. 2° over 38/~mol/l (9 gg/ml). In a study by Tomson 27, however, side effects occurred even at a CBZ concentration of 17 gmol/l (4 gg/ml). Thus it is difficult to determine a precise concentration level for the occurrence of side effects. Tomson 27 has, however, shown that there is a strong association between side effects and fluctuations in CBZ concentration. This is further supported by the results of our study, as the patients reported fewer side effects, especially dizziness, during the period on slow-release CBZ. Naturally the design of the study (single-blind) was not optimal for investigating such a subjective matter as side effects. The side effects have been shown to be significantly less when a slow-release preparation is used than with a conventional one, in a recent double-blind study by Persson et al.19. Good seizure control has been observed over a wide range of serum CBZ concentrations. In general, the plasma concentrations of patients with poor seizure control have not differed significantly from concentrations observed in patients with good response 4. In the study by Shorvon et al. 2s there were 10 epileptic patients on CBZ monotherapy who were seizure-free with serum concentrations below 17 ~mol/l (4/~g/ml). However, patients with intractable seizures often require rather high concentrations, ranging from 41 to 48 ~mol/l (9.6-11.3 ~g/ml). It may be that the therapeutic range varies with the severity of the epilep-

REFERENCES

1 Albani, M. and Sehringer-Mansour, B., Verbesserte therapeutische Effektivit/it yon Carbamazepin bei Kindern dutch Verwendung der Retard-Pr~iparation. In: J.W.A. Meijer (Ed.), 1. lnternationales Carbamazepin Slow-Release Symposium, Zuckschwerdt, Munich, 1985, pp. 32-39. 2 Albright, P.S. and Bruni, J., Effects of carbamazepine and

tic process 4. The fluctuations in serum CBZ concentrations may be of essential importance with reference to the control of seizures. This would be in agreement with the preliminary findings with other slow-release preparations 6'24. In our study, there were fewer epileptic seizures during the SR, but on the other hand 2 patients suffered a single generalized seizure during SR compared to one patient with a series of generalized seizures. Our study was of such a short duration that no conclusions can be drawn from the results with regard to correlation between the serum fluctuations of CBZ or CBZE and the epileptic seizures, although there was a significant correlation between the fluctuations of CBZE and epileptic seizures during the conventional preparation. The finding is interesting with regard to the known antiepileptic effect of CBZE 2. In concluaion, our study indicates that the use of slow-release CBZ preparations instead of conventional ones could be advantageous in the treatment of epilepsy. The SR clearly leads to reduced fluctuations in CBZ and CBZE and appears to decrease the occurrence of side effects of CBZ theraPY,

ACKNOWLEDGEMENTS We would like to thank Mr. J. Tuominen for his expert statistical help, Mrs. M. Love for precise data management, Mrs. P. Toikka for typing the manuscript and Farmos Group Ltd. for providing the investigational drugs. We are also grateful to Dr. E. Ben-Menachem for reviewing the manuscript and Dr. K. Battarbee for English language revision.

its epoxide metabolite on amygdala-kindled seizures in rats, Neurology, 34 (1984) 1383-1386. 3 Bertilsson, L., Clinical pharmacokinetics of carbamazepine, Clin. Pharmacokinet., 3 (1978) 128-143. 4 Bertilsson, L. and Tomson, T., Clinical pharmacokinetics and pharmacological effects of carbamazepine and carbamazepine-10,11-epoxide, Clin. Pharmacokinet., 11 (1986) 177-198. 5 Bj6rkvist, S.E., Isohanni, M., M~ikela, K. and M~ikinen,

133 L., Ambulant treatment of alcohol withdrawal symptoms with carbamazepine: a formal multicentre double blind comparison with placebo, Acta Psychiat. Scand., 53 (1976) 333-342. 6 Blankenhorn, V., Biilau, P., Kr~imer, G., Kreiten, K. and Stefan, H., Tegretol 400 retard versus Tegretol 200 in tier Behandlung schwerverlaufender partieller Epilepsien. Eine Einjahresstudie. In: G. Kriimer and H.C. Hopf (Eds.), Carbamazepin in der Neurologie, Georg Thieme, Stuttgart, 1987, pp. 254-258. 7 Blom, S., Trigeminal neuralgia. Its treatment with a new anticonvulsive drug (632883), Lancet, ii (1962) 839-840. 8 Boxenbaum, H., Pharmacokinetic determinants in the design and evaluation of sustained-release dosage forms, Pharm. Res., 2 (1984) 82-88. 9 Dupois, J.-P., D6rh6fer, G., Henriksen, O., Hulsman, J.R.A.J., Johannessen, S.I., Racine, A. and Theobald, W., Entwicklung des Tegretoi retard m kinetisch orientierte Konzeption. In: G. Krtimer and H.C, Hopf (Eds.), Carbamazepin in der Neurologie, Georg Thieme, Stuttgart, 1987, pp. 212-221. 10 Eichelbaum, M., Ekbom, K., Bertilsson, L., Ringberger, V.A. and Rane, A., Plasma kinetics of carbamazepine and its epoxide metabolite in man after single and multiple doses, Clin. Pharmacol., 8 (1975) 337-341. 11 Ghose, K,, Fry, D.E. and Christfides, J.A., Effect of dosage frequency of carbamazepine on drug serum levels in epileptic patients, Eur. J. Clin. Pharmacol., 24 (1983) 375-381. 12 Grizzle, J.E., The two-period change-over design and its use in clinical trial=, Biometrics, 21 (1965) 467-480. 13 H6ppener, R.J., Kuyer, A., Meijer, J.W.A. and Hulsman, J., Correlation between daily fluctuations of carbamazepine serum levels and intermittent side effects, Epilepsia, 21 (1980) 341-350. 14 Johannessen, S.I. and Hennxsen, O., Comparison of the serum concentration profiles of Tegretol and two new slowrelease preparations. In: P. Wolf, M. Dam, D. Janz and F.E. Dreifuss (Eds.), Adv. Epileptoi., 16 (1987) 421-424. 15 Juergens, U., May, T., Hillenkr6tter, K. and Rambeck, B., Systematic comparison of three basic methods of sample pretreatment for high-performance liquid chromatographic analysis of antiepileptic drugs using gas chromatography as a reference method, Ther. Drug. Monit., 6 (1984) 334-343. 16 Kr~imer, G., Besser, R. and Theisohn, M., Clinical pharmacokinetics of carbamazepine. Studies with different preparations in volunteers and epileptic patients. In: 17th Epilepsy Int. Congress, Jerusalem, 1987, p. 67.

17 Larkin, J.G., McLellan, A., Munday, M., Sunderland, E., Butler, E. and Brodie, M.J., A double-blind comparison of conventional and controlled-release carbamazepine in healthy subjects, Br. J. Clin. Pharmacol., 27 (1989) 313-322. 18 Okuma, T., Inanaga, K., Otsuki, S., Sarai, K., Takahashi, R., Hazama, H., Mori, A. and Watanabe, M., Comparison of the antimanic efficacy of carbamazepine and chlorpromazine: a double-blind controlled study, Psychopharmacology (Bed.), 66 (1979) 211-217. 19 Persson, L., Ben-Menachem, E., Bengtsson, E. and Heinonen, E., Differences in side effects between a conventional and a slow release preparation of carbamazepine, Epilepsy Res., 6 (1990) 134-140. 20 Riva, R., Albani, F., Ambrosetto, G., Contin, M., Cortelli, P., Perucca, E. and Baruzzi, A., Diurnal fluctuations in free and total steady-state plasma levels of carbamazepine and correlation with intermittent side effects, Epilepsia, 25 (1984) 476-481. 21 Rushton, J.G., Stevens, J.C. and Miller, R.H., Olossopharyngeal (vasoglossopharyngeai) neuralgia. A study of 217 cases, Arch. Neurol., 38 (1981) 201-205. 22 Siilke-Treumann, A., Rambeck, B., May, T. and Bonigk, H,E., Carbamazepin retard - - ntitzlich fiir die Therapie kindlicher Epilepsien? In: G. Kr~imer and H.C. Hopf (Eds.), Carbamazepin in der Neurologie, Georg Thieme, Stuttgart, 1987, pp. 240-245. 23 Schiifer, H., Mascher, J. and Bahr, A., Vergleich einer Carbamazepin-Retard-Formulierung gegen Carbamazepin-Standard-Tabletten bei epileptischen Patienten. In: J.W.A. Meijer (Ed.), 1. Internationales Carbamazepin Slow.Release Symposium, Zuckschwerdt, Munich, 1985, pp. 20-26. 24 Scollo-Lavizzari, G. and Bucher, C., Untersuchungen zur Epilepsietherapie mit Carbamazepin-Retard (Timonii 300 retard). In: J.W.A. Meijer (Ed.), 1. Internationales Carba. mazepin Slow.Release Symposium, Zuckschwerdt, Munich, 1985, pp. 58-61. 25 Shorvon, S.D., Galbraith, A.W., Laundy, M., Vydelingurn, L. and Reynolds, E.H., Monotherapy for epilepsy. In: S.I. Johannessen (Ed.), Antiepileptic Therapy: Advances in Drug Monitoring, Raven Press, New York, 1980, pp. 127-131. 26 Sillanp~iii, M., Carbamazepine. Pharmacological and clinical uses, Acta Neurol. Scand., 64, Suppl. 88 (1981) 64. 27 Tomson, T., Interdosage fluctuations in plasma carbamazepine concentration determine intermittent side effects, Arch. Neurol., 41 (1984)830-834.