Evaluations on New Drugs Drugs 13: 81-123 (1977) © ADIS Press 1977

Sodium Valproate: A Review of its Pharmacological Properties and Therapeutic Efficacy in Epilepsy R.M. Pinder, R.N. Brogden, T.M. Speight and G.S. Avery Australasian Drug Informati6n Services, Auckland

Manuscript reviewed by : M.J. Eadie, Department of Medicine, University of Queensland, Royal Brisbane Hospital, Brisbane, Australia ; P.M. Jeavons, EEG Department, Dudley Road Hospital, Birmingham, England; J. W. Lance, Division of Neurology, The Prince Henry Hospital , Little Bay, N.S.W., Australia ; M.J. Parsonage, Neurology Department, General Infirmary, leeds, England; A. Richens, Department of Clinical Pharmacology, St. Bartholomew's Hospital, london, England; £. van der Kleijn, Department of Clinical Pharmacy, Sint Radboud Hospital, University of Nijmegen , Nijmegen , The Netherlands.

Table of Contents Summary 1. Pharmacodynamic Studies 1.1 Anticonvulsant Properties 1.2 Mechanism of Anticonvulsant Action 1.2.1 GABA Levels 1.2.2 GABA Metabolism 1.2.3 Interaction with Other Antiepileptic Drugs 1.3 Other CNS Effects 1.3.1 Effects on the EEG 1.3.2 Sedative Effects 1.3.3 Behavioural Effects 1.4 Effects on Blood Platelets 1.5 Toxicological Studies 1.5. 1 Acute Toxicity 1.5.2 Chronic Toxicity 1.5.3 Reproduction and Dysmorphology Studies 2. Pharmacokinetic Studies 2.1 Absorption 2.1.1 Half-Life 2.1.2 Plasma Levels and Therapeutic Effects 2.1.3 Delayed-Release Forms of Dipropylacetic Acid

:.............

82 85 85 88 88 89 90 90 90 91 91 91 92 92 92 92 92 93 94 94 95

Sodium Valproate

82

2.2 Distribution 2.2.1 Placental Transfer and Penetration into Breast Milk : ............ ................................ 2.2.2 Plasma Protein Binding 2.3 Metabolism and Excretion 3. Therapeutic Trials 3.1 Controlled Trials 3.2 Results by Seizure Type 3.2.1 Partial Seizures with Elementary Symptomatolog y .................................................. 3.2.2 Partial Seizures with Complex Symptomatology (Psychomotor or Temporal Lobe Epilepsy) 3.2.3 Partial Seizures Secondarily Generalised 3.2.4 Typ ical Absence Seizures (Petit Mat) 3.2.5 Atypical Absences 3.2.6 Myoclonic Epilepsy : 3.2.7 Tonic-Clonic Seizures (Grand Mall 3.2.8 Atonic or Akinet ic Seizures 3.2.9 Combinations of Individual Seizure Types 3.2.10 Miscellaneous Types of Seizure 3.3 Childhood Epilepsy c... .... ... .. .... .... .. .. .... .. .... 3.3.1 Febrile Convul sions 3.4 Sodium Valproate Compared with Other Drugs 3.5 Sodium Valproate Alone? 3.6 Tolerance and Long-Term Treatment 3.7 Status Epilepticus 3.8 Miscellaneous The rapeutic Trials 3.8.1 Dyskinesias 3.8.2 Liver Disease 4. Side-Effects 4.1 Gastro intestinal Effects 4.2 CNS Effects 4.3 Changes in Appetite 4.4 Hair Loss 4.5 Behavioural Changes 4.6 Laboratory Abnormalities 5. Drug Interactions 5.1 Other Antiepileptic Drugs 5.2 Enzyme Induction 6. Contraindications and Precautions 7. Overdosage 8. Dosage

Summary

95 96 97 97 97 98 99 100 102 104 104 104 104 106 107 107 108 108 III III III 112 I I3 113 113 113 114 114 I 14 115 115 115 116 I 16 I 16 117 117

I I8 118

Sodium valproate ' has a broad spectrum of anticonvulsant activity, but is structurally unrelated to conventional antiepileptic drugs. Its proposed mode of action is mediated through effects on the function of brain y-aminobutyric acid (GABA). However, the elevations in brain and cerebellar GABA, and the concomitant reductions in levels of cyclic guanosine monophosphate , occur in animals at dose levels which are unlikely to be achieved during the treatment of epileptic patients . 'Depakene ', 'Depakine', 'Epilim' , 'Ergenyl' , 'Eurekene', 'Labazene' (Reckitt-Labaz),

Sodium Valproate

83

Pharmacodynamic studies have shown that sodium valproate does not affect the autonomic nervous system. and it lacks significant cardiovascular. renal or respiratory effects. Its effects on the EEG are uncertain . but it does seem to reduce or abolish spike-andwave activity consistently in epileptic patients . Sodium valproate may have an intr insic hypnotic action, for a controlled trial in normal subjects has shown the drug to be significantly more sedative than placebo and additive with phenobarbitone. In clinical practice, however , the drug seems to produce increased awareness and alertness in many epileptic patients, possibly as a result of the reduced dosage of other antiepileptic drugs with sedative side-effects although the same effect is also noticed in patients who are on sodium valproate alone. Sodium valproate had no psychotropic properties during a controlled trial in patients with aggressive behavioural problems. Pharmacokinetic studies : Sodium valproate is rapidly absorbed after oral administration. reaching peak blood levels within I to 4 hours. A good correlation has not .yet been established between daily dose. plasma levels and therapeutic effects. due. among other reasons , to wide interindividual variations. The serum half-life is about 7 to 9 hours, but may be longer in cases of overdosage and shorter in epileptic patients receiving long-term therapy with some other antiepileptic drugs. Therapeutic blood levels appear to be about 50 to IOOllg/mlin most patients. associated with an adult daily dose of 1.200 to 1,500mg . Delayed-release forms of dipropylacetic acid may offer some advantages in terms of smoothing out fluctuations in inter-dose serum levels, but have to be administered with care due to large individual variations in release. Sodium valproate is rapidly distributed. reaching the brain of animals in a few minutes. but it is likely that most of the drug . which is strongly bound (90 % ) to human plasma Proteins. is restricted to the circulation and rapidly exchangeable extracellular water. The white matter is reached at a later period . Placental transfer occurs. and the drug produces dysmorphogenic effects in animals. Elimination is rapid, principally in the urine with minor amounts in the faeces and expired air . Remil excretion of unchanged sodium valproate is very small, and animal and human studies suggest that it is metabolised rapidly to a glucuronide conjugate and to products of omega-side-chain oxidation. Therapeutic trials: There are few controlled trials of sodium valproate in the treatment of epilepsy, and evidence of its efficacy comes mainly from open studies . In general. the drug has been given to patients already receiving but usually refractory to other antiepileptic medication , which in some cases has been subsequently reduced in dosage or withdrawn completely. Sodium valproate raises the plasma levels of other antiepileptic drugs , particularly phenobarbitoneand primidone, though long-term studies have suggested that these levels may return to normal if treatment is continued . Nevertheless, sodium valproate is effective alone in many patients, either as the drug first used in treatment or as a gradual replacement for previously ineffective medication. There is little evidence for the development of tolerance to its antiepileptic effects in long-term studies . Controlled trials have shown that sodium valproate is significantly superior to placebo when added to the previous antiepileptic medication of patients with various types of epilepsy. In one trial, the drug was indistinguishable from ethosuximide when either drug was given alone or added to other antiepileptic drugs in children with typical absence seizures. Sodium valproate is more effective in the generalised than in the partial epilepsies, and is particularly effective in patients with 3 cycle per second spike-and-wave discharges in the EEG . It may playa useful role in the management of grand mal, mixed grand mal and petit mal, drugrefractory temporal lobe epilepsy, and myoclonic epilepsy. In new patients with typical absence seizures . sodium valproate may become the drug of first choice. Its relative lack of sedative effects leaves intellectual performance unimpaired, which is important in childhood epilepsy. Infantile spasms and the Lennox-Gastaut syndrome respond somewhat less effectively to sodium valproate than they do to the benzodiazepines, though sodium valproate is

Sodium Valproate

84

more effective in myoclonic epilepsy where it may be combined with nitrazepam but not clonazepam. The drug may have a role in the prophylax is of febrile convulsions .

Side-effects: Sodium valproate is remarkabl y free of side-effects, in the general context of antiepileptic drugs. Gastrointestinal effects (nausea, vomiting , abdominal cramp , diarrhoea) are the most commonly reported reactions, occurring in about 9 to I6 % of adults and about 22 % of children. These effects have usually been transient and have only rarely required withdrawal of drug. CNS effects include transient drows iness and sedation, which can usually be corrected by a reduction in the dosage of other antiepileptic drugs given to the patient and may not therefore be strictly attributable to sodium valproate . Both increased and decreased appetite, with appropriate changes in weight, may occur, and there may be a temporar y loss of hair in some patients. Patients receiving sodium valproate , particularly children already on barbiturates, may show worsened behaviour with aggressiveness and hyperactivity which usually disappears on withdrawal of the barbiturate. However, most patients tend to feel more lively and alert. Sodium valproate inhibits the secondary phase of blood platelet aggregation, and this has been reflected in some isolated cases, mainly in children, of prolonged bleeding times and thrombocytopenia. Precautions: There are no specific contra-ind ications to sodium valproate , and there have been no reports of significant hepatic, cardiac or renal abnormalities attributable to the drug. It has been shown to be dysmorphogenic in animals (but no more so than phenytoin) and it probably should be used in women of child-bearing age only in severe cases of epilepsy or in those who are resistant to other treatment, though this is a matter for clinical judgement. There are no reports of dysmorphogenic effects in humans . Introduction of sodium valproate to existing medication may require concomitant reduction in the dosage of other drugs, especially phenobarbitone and primidone , due to the possibility of raised plasma levels and enhanced sedative effects. Sodium valproate may potentiate the effects of MAO inhibitors and thymoleptics, the dosages of which should be reduced accordingly. Caution should be exercised when the drug is administered with others that may affect blood coagulation or when surgery is contemplated in epileptic patients. Sodium valproate is partly eliminated in the form of ketone bodies, which may result in false positives when testing the urine of diabetic patients. Dosage: Sodium valproate may be introduced alone or added to existing medication, but dosage is individual. Treatment should begin with 600mg daily in divided doses, increasing by 200mg daily at 3-day intervals until control is achieved. This is generally within the range 1,000 to I,600mg daily. If adequate control has not been achieved after two weeks, the dose may be further increased by stages to a maximum of 2,600mg daily, though some severe cases may require doses in excess of this recommended maximum . An alternative drug may be gradually substituted, initially at low dosage, though combinat ion therapy should be considered only if individual drugs fail to produce adequate control. The same pattern of introduction of sodium valproate should be followed in patients already receiving other antiepileptic drugs . If increased sedation is observed, dosages of barbiturates or other drugs should be reduced as sodium valproate is increased. The dosage of both should be adjusted during the stabilisation period to give optimum control at the lowest combined dosage level. In some cases it may be possible to withdraw other medication entirely, allowing optimum control with sodium valproate alone. Initial dosage in children over 20kg should be 400mg/day irrespective of weight, in divided doses, with spaced increases until control is achieved. This is usually within the range of 20 to 30mg/kg of body weight per day. Children under 20kg and infants should be given 20mg/kg of body weight per day; in severe cases up to 50mg/kg/day. A dose of 50mg/kg should be exceeded only in patients in whom plasma levels are measured; plasma levels of 200pg/ml should be exceeded only with caution, and with monitoring of haematological function .

85

Sodium Valproate

Dipropylacetic acid (2-propylpentanoic acid, 2propylvaleric acid, valproic acid) was first synthesised almost a century ago. Serendipitous discovery of its anticonvulsant properties took another 82 years, when Meunier et at. (I 963) realised that several compounds of disparate structure were effective in animals when administered in dipropylacetic acid as solvent. Sodium valproate (fig. I), which lacks a nitrogen function and differs markedly in structure from other antiepileptic drugs, is the usually available form of the acid, though the magnesium salt has occasionally been used. The role of sodium valproate in the treatment of epilepsy has been reviewed (Simon and Penry, 1975).

Fig. 1. Chemical structure of sodium valproate .

was about as effective as phenobarbitone but less so than diazepam in suppressing ictal and interictal EEG phenomena. In another study in cats with cortical cobalt lesions and in normal animals (Mutani and Fariello, 1969), the drug also appeared to decrease the excitability of the cortex in response to electrical stimulation. In vitro studies of evoked potentials J. Pharmacodynamic Studies from excised guinea-pig cortex showed that sodium valproate slightly inhibited resting potentials and preThe pharmacodynamic effects of sodium valproate vented the development of penicillin-induced" potenare exerted principally upon the central rather than tials (Voskuyl et al., 1975). the autonomic nervous system . The drug does not inComparative studies (table II) with the lowfluence respiration or body temperature and it lacks frequency electroshock test or the maximal significant cardiovascular or renal effects (Shuto and electroshock test, both useful models for temporal Nishigaki, 1970; Swin yard , 1964). lobe or psychomotor epilepsy (Swinyard, 1964; Woodbury, 1972), suggest that sodium valproate is less potent than the benzodiazepines, succinimides, 1. I Anticonvulsant Properties phenobarbitone and phenytoin but has a more favourable therapeutic index than some of these The anticonvulsant properties of sodium valproate drugs. Sodium valproate also decreased the sponare summarised in table I. In general, the drug distaneous EEG signs of seizure discharge when given plays a broad spectrum of moderate activity against intraperitoneally to cats with hippocampal cobalt foci several types of chemically- or electrically-induced in the temporal lobe (Mutani et al., 1968). It convulsions in a variety of species (see also Frey and decreased the excitability of the hippocampus both in Loscher , 1976). The anticonvulsant potency and the presence and absence of a focal lesion, and limited therapeutic index for sodium valproate are within the the spread of the seizure activity . range exhibited by other clinically effective drugs Maximal electroshock seizures in animals are also (table II). a useful model for generalised tonic-clonic seizures in In cats with cortical cobalt or alumina lesions, the humans (Swinyard, 1969). Comparative studies sugexperimental models for partial focal seizures with gest that sodium valproate should have some activity elementary symptomatology (Swinyard , 1969; 1973; against generalised tonic-clonic seizures , though it is Woodbury, 1972), sodium valproate did not decrease less effective in this model than phenytoin, primidone focal discharges in the cortex but rather inhibited the or phenobarbitone (table 11). Sodium valproate spread of seizure activity from the focus (Fariello and - 200mg/kg (sc) did not inhibit bemegride-induced Mutani, 1970; Van Duijn and Beckmann, 1975). It tonic-clonic convulsions in alert cats , but the same

so

Table I. Ant iconvulsant properties of sod ium valproate in animals

Test '

Time to peak effect Ihrs)

ED,;o ' maximal electroshock lmg /kg)

Animal

mice rats rabb its cats

0

a.

Sw inyard (19641

Shuto and Nish igaki 119701

Frey and t.osche r (19761

ip

po

ip

iv

0 .5 1.0

0 .25 0 .5 0.5 0 .5

0 .5

0 .25

-

1.0 (pol

315 345 325

235 125 175

-

490 (pol

-

420 (pol

-

a Ql

mice rats rabb its cats

-

260 140 235 67

EDso ' low frequency m inimal elec troshock (mg/kgl

mice

330

158

EDso ' a.c, m inimal electroshock (mg/kgl

rats

260

165

-

165

ED. o' 70-85mg /kg penty lenetetrazole (sol

mice rats

265 1,020

.120- 122 263

220

120-122

EDso ' 15mg /kg pentylenetetrazole Ilv)

rabb its cats

-

180 33 2

290

305

LDso (mg/kg )2

mice rats rabb it s cats

1,700 1,530

1,060 790 1,200 56 5

1,197 1,519 1,650

838 1,045

766 946

750 (iv)

415 188 216 105

580

290

-

290 (pol

1,150

1,100

TDso (mg / kg)3

mice rats rabb its cats

605 318

-

1,165 570

-

3

<

Ql

'0

po

-

e

1 Lebreton et al. (1964) also reported that sod ium valproate (1OOmg/ kg , ip) was ineff ecti ve against seizures induced in rodents by str ychnine (2mg / kg , scI picrotoxin (5mg /kg , sc), coca ine (80mg /kg , ipl or thu jone (250mg /kg , ip). Frey and Loscher (197 6) found ED. o values of 200, 600 and 700mg/ kg Ipo) for sodium valproate against seizures induced in mice by picrotox in (4mg /kg , sc), bicuculline (5.5mg /kg , scl. and strychnine (1 .2mg/kg , scI 2 Dose fatal to 50% of animals w ithin 24 hours. 3 Dose producing minimal evidence of neurotoxicity (rotarod or chimney testl in 50% of animals at time of peak drug effect.

co

I""

'"

g>

Table II. Activity rankings of some anticonvulsant drugs in mice'

Maximal electroshock

Pentylenetetrazole seizures

Low-frequency minimal electroshock

order of potency

order of therapeutic index 2

order of potency

order of therapeutic index 2

order of potency

order of therapeutic index 2

Oxazepam Diazepam Primidone Phenyto in Phenobarbitone Nitrazepam Chlord iazepoxide Methsuximide Phenacemide Sodium valproate Trimethadione

Oxazepam Primidone Phenacemide Phenytoin Chlordiazepoxide Diazepam Phenobarbitone Sodium vslproete Trimethadione Methosuximide Nitrazepam

Clonazepam Nitrazepam Diazepam Oxazepam Chlordiazepoxide Phenobarbitone Phenacemide Sodium valproate Trimethadione Ethosux imide

Nitrazepam Oxazepam Clonazepam Diazepam Chlordiazepoxide Ethosuximide Phenacemide Sodium velproete Phenobarb itone Trimethadione

Diazepam Phenobarbitone Chlord iazepoxide Oxazepam Nitrazepam Phenyto in Phenacemide Sodium velproete Trimethadione

Oxazepam Diazepam Chlordi azepoxide Nitrazepam Phenacemide Sodium velproete Phenobarb itone Trimethadione Phenytoin

1 Data from the review of Simon and Penry (1975). 2 Ratio of median toxic dose to the median effective dose.

a. c· 3

s

-0

a III

CD

00 .....

88

Sodium Valproate

dose given intravenously gave marked protection (Van Duijn and Beckmann, 1975). The drug also inhibited the spread of seizure activity when bemegride was used to reactivate 'extinguished' cobalt foci in cats. Wiechert (1972) showed that rats pretreated with sodium valproate 2mmol/kg (ip) for 4 days were protected from the tonic-clonic seizures produced by intracisternal L-glutamate 7mg/kg. Dogs pretreated with the drug and also given 1Omg/kg (iv) did not develop status epilepticus and die, as did control animals, though isolated tonic-clonic seizures did occur. Higher doses of sodium valproate given before or during L-glutamate-induced status epilepticus generally gave complete protection or reversal. Pentylenetetrazole-inducedseizures in mice are the classical model for absence seizures in humans (Swinyard, 1969; 1973), and comparative studies show that sodium valproate is rated somewhat below the clinically useful succinimides and benzodiazepines but above phenobarbitone and trimethadione (table II). In the photosensitive baboon (Papio papio) model of absence seizures (Woodbury, 1972), sodium valproate blocked paroxysmal discharges induced by intermittent light stimulation up to 30 minutes after intravenous administration of doses up to 150mg/kg, and reduced their intensity for over I hour (Patry and Naquet, 1971). Audiogenic seizures in mice and carbon dioxide withdrawal seizures in rats, also useful models of absence seizures, were abolished by sodium valproate, with intraperitoneal ED so values of less than 200mg/kg and 146mg/kg respectively (Simler et al., 1973; Swinyard , 1964).

1.2 Mechanism of Anticonvulsant Action The mechanism of action of sodium valproate is unknown. It is possible, however, that , atleast in part and particularly at high doses, its anticonvulsant and neurotoxic actions are mediated through effects on the function of brain y-aminobutyric acid (GABA). Convulsions can be produced experimentally by compounds which either block the inhibitory synaptic action of GABA or inhibit the enzyme responsible for

its synthesis, glutamic acid decarboxylase (see review by Meldrum, 1975). Sodium valproate appears to raise brain and cerebellar levels of GABA in animals, possibly by inhibition of GABA degradative enzymes, but at dose levels which are unlikely to be achieved in clinical practice (Anlezark et al., 1976; Sawaya et al., 1975). It may inhibit the reuptake of GABA (Harvey, 1976).

1.2.1 GABA Levels Elevation of GABA levels in whole brain and the cerebellum have been correlated with the anticonvulsant effects of sodium valproate against several types of seizure in animals although distinction has not usually been made between the GABA pool involved with neurotransmission and the much larger one concerned with metabolism in the Krebs cycle. Godin et al. (1969) observed increases in rat brain GABA of 30 to 46 % I hour following 200 or 400mg/kg (ip) doses of sodium valproate, without any associated changes in the biosynthesis of GABA from glutamic acid. Simler et al. (J 973) found an 8 % increase in mouse brain GABA levels 15 minutes after sodium valproate (200mg/kg, im) and a 37 % increase 30 minutes after 400mg/kg, lasting for I and 3 hours respectively. The mice were protected from audiogenic seizures for 20 and 120 minutes respectively. In another epilepsy model in mice, sodium valproate was effective against seizures and in preventing the fall in GABA levels usually associated with withdrawal from alcohol dependence (Littleton and Griffiths, 1975). Audiogenic seizures in mice were markedly reduced by intraperitoneal sodium valproate 400mg/kg and abolished after 600mg/kg, and these doses were associated respectively with little change in brain GABA levels or an increase of 57 % (Anlezark et al., 1976). The demonstration of a clear anticonvulsant action of the drug, associated with insignificant changes in GABA levels in these latter experiments, makes it likely that other factors are involved. The low volume of distribution of sodium valproate (see section 2.2) suggests that it would be unlikely to enter cells and act on intraneuronal

Sodium Valproate

GABA. Nevertheless, sodium valproate blocked the convulsant actions in mice of the GABA antagonists picrotoxin and bicuculline at doses below or similar to those required to block pentylenetetrazole-induced seizures (Frey and Loscher, 1976). Sodium valproate specifically blocked the abnormal involuntary movements induced by I-O,4-dihydroxyphenyO piperazine in the rat, and this compound is believed to be an antagonist at central GABA receptors (Costall et al., 1976). Sodium valproate may be unique among antiepileptic drugs in that it both raises the levels of brain GABA and lowers those of cyclic guanosine monophosphate (cGMP) in animals (Kupferberg et al., 1975; Lust et al., 1976). Following intraperitoneal administration to mice, clonazepam 160J.1g/kg, phenytoin 20mgk/kg, and phenobarbitone 33mg/kg all reduced cerebellar cGMP by over 50 % but there were no changes in cerebellar GABA . In contrast, sodium valproate 400mg/kg increased cerebellar GABA by 56 % , decreased cerebellar cGMP by 61 to 70 %, and protected all the animals against maximal electroshock seizures at 30 minutes. Cerebellar GABA levels, and the less elevated levels in the cerebral cortex, paralleled anticonvulsant activity, returning to control levels at 3 hours . Both the convulsions induced in mice by isonicotinic acid hydrazide 200mg/kg, and the associated changes in GABA (-27%) and cGMP (+ 340%) levels in the cerebellum and cerebral cortex, were prevented by concomitant administration of sodium valproate 400mg/kg. 1.2.2 GABA Metabolism GABA is formed by decarboxylation of glutamic acid catalysed by the enzyme glutamic acid decarboxylase (GAD), and is then deaminated (GABA-transaminase; GABA-T) and converted to succinate (succinic semialdehyde dehydrogenase; SSA-DH). Sodium valproate inhibited both GABA-T and GAD activity in rat brain homogenates, the former more markedly than the latter , but no inhibition could be demonstrated in vivo (Godin et al., 1969). The same investigators subsequently showed

89

that sodium valproate acted as a powerful competitive inhibitor of GABA-T in the conversion of GABA to succinic semialdehyde and as a weak non-competitive inhibitor of GAD in the formation of GABA by transamination of a-ketoglutarate (Simler et al., 1973). Regional distribution studies in rodents have shown that sodium valproate is localised in those areas of the brain with high GABA-T activity (Ciesielski et al., 1975). Other workers have suggested that sodium valproate is a more potent competitive inhibitor of rodent brain SSA-DH than it is of GABA-T (Anlezark et al., 1976; Harvey et al., 1975; Sawaya et al., 1975). Only slight « 4 %) inhibition of GABA-T was observed with sodium valproate at concentrations (0.1 to ImM) likely to be present in human brains during therapy with the drug. The considerable excess ofGABA-T in the brain necessitates inhibition of greater than 50 % before functional changes or increases in GABA levels are apparent in animals (Anlezark et al., 1976). In contrast, SSA-DH was inhibited in vitro by about 15% at therapeutic concentrations of sodium valproate (0.1mM), whereas other antiepileptic drugs (diazepam, phenobarbitone , carbamazepine and phensuximide) required 5 to 10fold higher concentrations than those likely to occur in human brains in order to achieve this level of inhibition (Sawaya et al., I975). At the latter concentration of sodium valproate (5mM), which would likely follow doses of 600mg/kg in the mouse, inhibition of SSA-DH would be about 90% (Anlezark et al., 1976). The functional effects of inhibition ofSSA-DH are unknown, but they may be important in the absence of any changes in brain GABA content (Anlezark et al., 1976; Sawaya et al., 1975). Following synaptic release, GABA is inactivated by uptake into glia and nerve endings. A reduced conversion of GABA to succinate might critically modify energy metabolism in glia or in neurons involved in GABA-mediated synaptic inhibition , a functional effect which would not necessarily correlate with brain GABA levels (Sawaya et al., 1975). An alternative and complementary hypothesis is that sodium valproate acts to in-

Sodium Valproate

hibit GABA reuptake, as a result of the structural similarity between it and GABA (Harvey, 1976). 1.2.3 Interaction with Other Antiepileptic Drugs It has been suggested that sodium valproate has no antiepileptic action per se, but merely acts by raising blood levels of the other antiepileptic drugs (see section 5). This is difficult to disprove since in most patients the drug has been added to previous medication. However, in view of its anticonvulsant activity in animals and its beneficial effects alone in some patients or in association with phenytoin in the absence of elevated phenytoin levels, it is likely that sodium valproate has antiepileptic activity. Jeavons et al. (J 977b) have pointed out that control of fits is achieved in patients with a very long history of epilepsy previously not responsive to many antiepileptic drugs , and that very severe epilepsies respond (many having hourly seizures). Furthermore, many patients remain seizure-free when all other drugs have been withdrawn, and many respond to sodium valproate given alone from the start of therapy . These authors treated 108 patients with sodium valproate alone, and 85 were seizure free; 58 had received the drug alone from the start and of these 43 were seizurefree.

1.3 Other CNS Effects 1.3.1 Effects on the EEG The effects of sodium valproate on the EEG are unclear. Clinical trials have shown that improvement in seizure activity can both precede and parallel the reduction in paroxysmal activity noted on random EEG 's (Simon and Penry, 1975). Diffuse slowing of background activity was observed in the EEGs of epileptic children treated with 200 to 600mg daily (Miribel and Marinier , 1968). Higher doses tended to give less slow activity than did lower doses, and changes in background activity were independent of changes in paroxysmal activity. Slow waves in the temporo-occipital region or over the entire hemi-

90

sphere were observed in some patients with temporal lobe epilepsy but previously normal EEGs . Paroxysmal activity may be reduced in both absence seizures and photic seizures. Bergamini et al. (J 970) observed that spike-and-wave paroxysms disappeared in 22 patients with absence seizures treated with sodium valproate, although hyperventilation and photic stimulation continued to provoke spikeand-wave paroxysms in 5 of these patients. In a study in 3 adults with recurring absence seizures and on a stable regimen of anti-epileptic drugs, Penry et al. (J 976) measured 12-hour telemetered EEGs before and after the addition of 20mg/kg sodium valproate. Two of the 3 patients showed a significant reduction in spike-and-wave paroxysmal activity, associated with subjective improvement. This reduction could not be related to serum levels or half-lives of sodium valproate, which varied widely. One of the patients with subjective improvement continued to experience long-lasting spike-and-wave bursts in the EEG , despite a greater than 50 % reduction in seizure frequency. In a study in 37 patients with photosensitive epilepsy, 20 patients were given 800mg sodium valproate, 22 patients received 1,000mg and 13 received 1,200mg daily for several months (Jeavons et al., 1976). All had previously experienced seizures on exposure to flickering light and 10 had spontaneous seizures, associated with bilateral or local discharges of spike-and-wave or spike activity in the EEG. Photoconvulsive responses disappeared in I 9 patients receiving sodium valproate, including 4 in the I ,200mg group, 12 in the 1,000mg group, and 3 at the lowest dose level. Significant improvement was also observed at these dose levels in 5, 10 and I 3 patients respectively. Occipital spikes in 12 patients were still evident, however, at the highest dose level of sodium valproate, despite the abolition of photoconvulsive responses in 5 of them . In 72 % of 60 patients with 3 cycle per second spike-and-wave discharges in their EEG , the records became normal, and a normal EEG occurred in 4 of 6 patients with myoclonic atonic epilepsy whose fits were controlled (Jeavons et al., 1976, I977b).

91

Sodium Valproate

1.3.2 Sedative Effects Patients taking sodium valproate frequently complain of drowsiness, particularly if they are also receiving phenobarbitone (see section 4.2). However, the elevation by sodium valproate of phenobarbitone -levels in plasma (see section 5) makes it difficult to decide whether the drug has an intrinsic sedative action. Of 58 patients who received sodium valproate alone, none showed any signs of sedation (Jeavons et al., I977b). Animal studies have suggested that sodium valproate per se is more sedative than trimethadione or phenacemide but less so than phenobarbitone or phenytoin (Swinyard, 1964). In a double-blind study , 8 healthy volunteers were given single oral doses of sodium valproate 400mg, phenobarbitone 60mg, the two drugs together, or placebo (Scott et al., 1976). EEG recordings taken 1 to 11 /2hours after ingestion indicated a tendency for sodium valproate to produce higher sleep scores, that is with greater amounts of delta activity and sleep spindles and less alpha activity in the EEG . Sodium valproate was significantly more effective in this respect than either phenobarbitone or placebo, while the combined drugs were slightly more effective than sodium valproate alone. Only on sodium valproate (alone or with phenobarbitone) did any subject reach stage 3 sleep. In direct contrast to this study , Kugler et al. (I 973) reported that sodium valproate given by intravenous injection produced a rapid increase in drive as measured by EEG determinations of vigilance. The study was carried out in 10 healthy volunteers, who received a single dose of 200mg sodium valproate or placebo during a double-blind crossover trial. Effects appeared rapidly and were maximal after 30 minutes, and the drug was significantly more effective than placebo. 1.3.3 Behavioural Effects In a double-blind trial in 20 epileptic patients with aggressive behaviour problems, there was no significant difference between the psychotropic effects of sodium valproate or placebo, 300mg 3 times daily (Sonnen et al., 1975; see section 4.5).

_Animal studies have shown that sodium valproate has no effect on spontaneous motor activity at dose levels which produce signs of neurotoxicity (Swinyard, 1964). Other workers have suggested that no behavioural effects are evident until doses of 500 to 600mg/kg (ip) are reached, when animals show either increased (Anlezark et al., 1976) or decreased (Misslin et al., 1975) spontaneous motor activity. High doses of sodium valproate decrease the acquisition of conditioned responses but at low doses (IOOmg/kg) the drug facilitates the acquisition of conditioned responses with negative reinforcement. These dose levels are too high for such effects to be of clinical significance.

1.4 Effects on Blood Platelets Sodium valproate inhibits the secondary phase of platelet aggregation, and this may be reflected in altered bleeding times. This platelet function defect is unlikely to be of clinical significance unless patients receiving sodium valproate are given other drugs which affect coagulation, such as aspirin or warfarin , or are undergoing surger y. However, children may be particularl y susceptible. A prolonged bleeding time was reported in 4 of 5 patients who also showed coagulation abnormalities (Sutor and Jesdinsky-Buscher, 1974). Only 1 patient was thrombocytopenic, and the effect was therefore attributed to defective platelet function. In a subsequent report by the same authors (Sutor and Jesdinsky-Buscher, 1976), 8 of 12 children receiving daily doses of 0.45 to 2.lg sodium valproate showed prolonged bleeding times, and 2 had thrombocytopenia. Six of the 10 with coagulation abnormalities also had significant reductions in platelet adhesiveness. Espir et al. (I 976) also reported a single patient with thrombocytopenia and prolonged bleeding time; platelet counts were low but other tests of coagulation were normal. Other investigators have found abnormal haemostatic parameters in many children receiving therapeutic doses of sodium valproate (Gadner et al., 1976; Von Voss et al.,

Sodium Valproate

1976). Most of these patients have also been receiving other antiepileptic drugs , but thrombocytopenia has also been reported with sodium valproate alone (Winfield et al., 1976). There were no significant changes in the bleeding times of 23 epileptic patients (aged 4 to 22 years) taking sodium valproate 600 to 1,400mg daily (Richardson et aI., I976a,b). Inhibition of the secondary phase of platelet aggregation was observed in the plasma of 6 patients, 5 of whom were taking sodium valproate as the sole antiepileptic drug. None of the patients was thrombocytopenic. A study of plasma from 6 healthy adults showed that sodium valproate inhibited the secondary phase at concentrations which were 25 times those in the therapeutic range. Similar inhibition occurred at therapeutic concentrations, but in plasma from only 3 of the 6 subjects. Standard tests of coagulation were normal at all concentrations in plasma from all subjects, and there were no changes in platelet adhesiveness or platelet factor 3 release. Nevertheless, platelet dysfunction caused by sodium valproate cannot be ruled out with certainty by determination of bleeding time and platelet count. Thromboelastography may be a more reliable method for detecting haemorrhagic side-effects of sodium valproate (Gadner et al., 1976; Von Voss et al., 1976). 1.5 Toxicological Studies 1.5.1 Acute Toxicity The acute toxicity of sodium valproate in animals is low, ranging from 565mg/kg (ip) in cats to 1,700mg/kg (po) in mice (see table I). Human subjects have recovered uneventfully from doses as high as 36g (see section 7). The oral LDso appears to increase with animal size (surface area), that is toxicity was least in dogs and greatest in mice (Conine et al., 1976). 1.5.2 Chronic Toxicity In long-term studies, mice treated for 160 days (dosed for 5 days per week) with 400mg/kg sodium

92

valproate showed a statistically significantly higher incidence of fatalities than did control animals. Nine of 30 mice so treated with the drug died during the trial, for reasons unspecified. 1.5.3 Reproduction and Dysmorphology Studies There have been only 8 published reports of epileptic mothers coming to term while on sodium valproate, in none of which were congenital abnormalities apparent (Espir et aI., 1976; Whittle , 1976). In 6 further cases there was one soft tissue fetal abnormality (hair lip and cleft palate, with low birth weight); sodium valproate had been given to the mother for one week two months before delivery. Throughout the course of pregnancy, she had received phenobarbitone, sulthiamine and ethosuximide (Whittle, personal communication). Sodium valproate produced dose-related dysmorphogenic effects in mice, rats, and rabbits, principally kidney defects, cleft palates, encephaloceles, ablepharia and fusion of ribs and vertebrae ; the abnormalities varied according to species (Whittle, 1976). Their incidence at intermediate doses (J 50 to 200mg/kg/day) was about the same as that observed with phenytoin at similar dose levels. The lowest dose level, (75mg/kg/day), which is at the upper limit of therapeutic dose levels, produced a 4 % incidenceof major defects in mice compared with I % in controls and 7 % in animals treated with phenytoin (J 50mg/kg/day).

2. Pharmacokinetic Studies Sodium valproate is rapidly absorbed after oral administration, undergoes extensive distribution and metabolism, and is quickly eliminated principally in the urine (Meijer and Meinardi, 1976). The half-life is short, particularly in epileptic patients receiving chronic therapy with some antiepileptic drugs . A good correlation has not yet been established between daily dose, plasma levels and therapeutic effects, due, among other reasons, to wide interindividual variations.

Sodium Valproate

93

2.1 Absorption Sodium valproate was absorbed rapidly in man following single oral doses, reaching peak levels in the blood about I to 4 hours after ingestion (Ferrandez and Eymard, 1973; Loiseau et al., 1975; Mortel and Franck, 1976). Blood levels of the drug then fell rapidly over 10 hours and more slowly after 10 hours, and low levels were still present 24 hours after a single dose; other studies indicate a single phase first-order decay of plasma levels (Jordan, personal communication). The peak plasma level may be reached earlier in epileptic patients than in healthy volunteers (Richens et al ., 1976). The mean maximum serum levels after single oral doses of 400 or 800mg to non-epileptic hospitalised patients were about 25 to 70fJg/ml respectively (Loiseau et al., 1975). However, while increments in the single oral dose were generally associated with increases in the blood levels of sodium valproate, there is only one re-

port of a clear relationship between dose and blood levels (Ferrandez and Eymard , 1973). In general, the same single dose gives widely different blood levels in different individuals (Loiseau et al., 1975; Meijer and Hessing-Brand, 1973; Mortel and Franck , 1976; Schobben and van der Kleijn, I974a). This poor correlation between dose and blood levels may be due to variations in the interval between the last meal and the time of sampling. Sodium valproate has a short half-life (see section 2.1.1) in some patients, which may confound blood level correlations attempted on single sample determinations . Wide interindividual variations are also evident during chronic treatment of epileptic patients, though the overall relationship of dose to serum levels is reasonably consistent (Loiseau et al., 1975; Meijer and Hessing-Brand, 1973; Schobben and van der Kleijn, 1974a; Schobben et al., 1975). In epileptic patients receiving 300 or 400mg sodium valproate 3 or 4 times daily, the plasma levels rose rapidly to an

100

E

c,

.Ec: o "+:

e

80 .•0 • • • • . . • . . • •••. '0.

60

C Q)

'. Q

0:()

o

c:

o U

40

ell

E

o

o male 3yrs 3 x 150mg (previously 150 mg • male 4yrs 3 x 150mg • male 5yrs 3 x 150mg ... male 5yrs 3 x 300mg

III

ell

ii:

20

First dose

5

daily daily) daily daily daily

Q ~

10

'0

15

Days Fig. 2. Plasma concentration of sodium valproate in 4 epileptic patients during and after subchronic intermittent administration. The apparent steady-state concentration is illusory and has been calculated from accurately fixed times of treatment and sampling. In actual practice there are wide (40 to 140%) inter-dose fluctuations in plasma concentrations (after Schobben et al., 1975).

Sodium Valproate

apparent steady state within 24 hours, but also declined rapidly after the last dose (Loiseau et al., 1975; Schobben et al., 1975) (fig. 2). Plasma concentrations fluctuated considerably during the interdose intervals, varying between 40 and 140 % of the indicated 'steady-state concentrations'. Administration of sodium valproate with milk products did not significantly affect the maximum plasma concentration or the time at which the peak was reached, during sub-chronic treatment of epileptic children (Schobben et al., 1975). Absorption of sodium valproate is delayed if it is taken after a meal (Meinardi et al., 1975). If administered on an empty stomach it is transformed into free dipropylacetic acid and may cause local gastric irritation (Meijer and Meinardi, 1976).

2.1.1 Half-Life Following single-dose administration to volunteers or non-epileptic patients, the plasma half-life of sodium valproate appeared to be about 7 to 8 hours (Ferrandez and Eymard, 1973; Loiseau et al., 1975). A range of 6.67 to 15.77 hours was observed by Richens et al. (I 976) in 10 healthy volunteers, with one exceptional value of 4.05hrs and mean of 9hrs. The half-life is somewhat longer in epileptic patients receiving multiple doses of sodium valproate during pharmacokinetic studies , being about 10 hours in adults (Loiseau et al., 1975) and 8 to 15 hours in children (Schobben et al., 1975). Half-lives of sodium valproate ranged from 3.84 to 8.28 hours (mean 5.88 hours) in 8 patients who were receiving long-term treatment with phenytoin and phenobarbitone or primidone, and were considerably shorter than those in 10 non-drug-treated volunteers (Richens et al., 1976) (vide supra). Espir et al. (1976) also found a low half-life for sodium valproate in chronic patients; the range was 5.6 to 9.6 hours, with a mean of 7.4 hours. Phenytoin, primidone, phenobarbitone, carbamazepine and pheneturide appear to induce liver enzymes and thereby shorten the half-life of sodium valproate, but ethosuximide and the benzodiazepines apparently do not share this effect (Richens et al., 1976).

94

Long half-lives have also been observed, up to 30 hours, in cases of overdosage with sodium valproate (van der Kleijn et al., 1974).

2.1.2 Plasma Levels and Therapeutic Effects Because of its rapid elimination, plasma concentrations of sodium valproate fluctuate considerably. There is only a poor correlation between daily dose and plasma concentration, possibly due to interindividual differences in absorption or total body clearance. The proper management of epilepsy with the drug cannot be achieved by the simple choice of a medication regimen based on body weight , surface area or age of the patient (Schobben et al., 1975). Frequent administration and determination of individual plasma concentrations at accurately fixed times, appears to be the only reliable monitoring method . Therapeutic blood levels appear to be about 50' to 1001Jg/ml, since most patients with blood levels in this range seem adequately controlled. In a study of 700 serum samples from epileptic patients receiving 300 to 2,400mg sodium valproate daily, serum levels varied between 10 and 160IJgI ml and 75 % were in the range 40 to 70IJg/mi (Meijer and Hessing-Brand, 1973). Other groups have reported ranges of 70 to 80IJg/mi (Meinardi et al., 1974), 34 to 68IJg/mi (Richens et al., 1976),25 to 150IJg/mi (Schobben and van der Kleijn, I 974a), 60 to I05IJg/mi (Chard et al. , 1976) and 50 to 100IJg/mi (Loiseau et al., 1975; Vajda et al., 1976) in epileptic patients. The difficulties of predicting therapeutic response from serum levels of sodium valproate were shown by Meinardi et al. (I 974) and Chard (I 976) when epileptic patients with similar drug levels or doses showed vastly different degrees of improvement (table III) though others have found that higher levels are generally associated with improvement (Grant and Baret, 1976; Espir et al., 1976). Haigh and Forsythe (I975) and Jeavons et al. (I977b), in uncontrolled trials, did not find that serum levels were an aid to management. Furthermore, almost all patients have received other anti-epileptic drug s in addition to sodium valproate, and there is a need for controlled

Sodium Valproate

95

Table III . Relationship betw een dose, blood levels and therapeutic effect of sodium valproate in epileptic patients

Author

No. of pts

No improvement No. pts

Significant improvement

mean daily dose

mean serum levels (pg/mll

mean daily dose

mean serum levels (pg/mll

5

37mg/kg

85

No. pts

Chard (1976)

10

5

39mg/kg

63.4

Meinardi et al. (1974)

46

20

1,527mg

67 ±25

26

1,531mg

82 ±22

Meinardi et al. (1974)

40

20

1,578mg

76±28

20

1,400mg

68±30

studies with the drug alone in order to define precisely its therapeutic range . Levels of sodium valproate in saliva were poorly correlated with those in plasma, in contrast to the good correlation between saliva and plasma levels of many other antiepileptic drugs (Schmidt, 1976). This is due principally to the low pKa (4 to 6) of sodium valproate , and indicates that levels of the drug cannot be conveniently monitored in saliva.

2.1.3 Delayed Release Forms of Dipropylacetic Acid Depamide is the amide of dipropylacetic acid, and may satisfy some of the requirements for a precursor which would sustain the release of the active moiety and smooth out some of the inter -dose fluctuations in plasma levels associated with sodium valproate . Following a single oral dose of 600mg depamide to non-epileptic patients, plasma levels of dipropylacetic acid were maximal within 3 to 6 hours at 30 to 50J.lg/ml, which compare well with those attained after a similar dose of sodium valproate (Loiseau et al., I975). The half-life of dipropylacetic acid when given as depamide was almost twice as long at IS hours than when it was given as the sodium salt. Multiple dose administration to 33 patients of a mean oral dose of 22.2mg/kg depamide gave mean serum levels of 13.5 to 64J.1g/ml (mean 37.3), but fluctuations in daily serum levels still occurred in epileptic patients receiving chronic treatment with depamide . Levels of dipropylacetic acid continued to

vary by a factor of 2 as a function of time of ingestion in relation to meals (Loiseau et al., 1975). Enteric-coated or slow release formulations of dipropylacetic acid may offer some advantages over sodium valproate and depamide , especially in smoothing out some of the fluctuations in blood levels during the day (figs. 3a and b) (Monel and Franck, 1976; Reekers-Ketting et al., 1975). Other derivatives have included the 2,2-dipropyl ethanol and the 2,2-dipropylethyl esters, which were toxic in animals, but the compound glycerol tri-dipropylacetate seems to have many of the qualities of a safe slow-release preparation for dipropylacetic acid (Meijer and Meinardi, 1976). 2.2 Distribution Studies in epileptic patients (Schobben and van der Kleijn, 1975) have given calculated values for the apparent volume of distribution of sodium valproate as 0.15 to 0.40 1/kg, which correspond well to those in animals (Schobben and van der Kleijn, I974b). This is small compared to most other antiepileptic drugs, and distribution of sodium valproate is probably almost restricted to the circulation and rapidly exchangeable extracellular water. Brain concentra tions should be low, and relatively high plasma levels will be needed to maintain therapeutic effects (see section 2.1.1). Levels in the cerebrospinal fluid of over 700 epileptic patients receiving long-term treatment with sodium valproate were about one-tenth of those

Sodium Valproate

(40 to 70pg/mO in the serum (Meijer and HessingBrand, 1973). Autoradiographic studies in rodents have shown that sodium valproate is rapidly distributed , reaching grey matter structures of the brain within a few minutes . Following oral administration of 14C_ carboxy labelled drug to rats, radioactivity was present in blood, muscle, liver, kidneys and brain within 10 minutes (Eymard et al., 1971). Maximum levels in the brain occurred after 30 minutes , and paralleled the time of maximum protection against experimental seizures. In the mouse, radioactivity appeared in grey matter structures of the brain within I minute of intravenous administration of I4C-carboxy labelled sodium valproate (Schobben and van der Kleijn, I974b). Levels were maximal at 15 minutes, and were always about 3 times higher in the liver and kidneys than in the brain . Some retention of radioactivity was observed in the olfactory bulb, that could be attributed to sodium valproate. Regional distribution studies in mice initially showed that sodium valproate was homogeneously distributed in the brain , except for the cerebellum where the white matter showed more activity than the cortex (Schobben and van der Kleijn, I974b). The drug appeared to be localised preferentially in those parts of the brain with the greatest levels of activity of the GABA degradative enzyme GABA-T (Ciesieleski et al., 1975; see section 1.2). These include the caudate nucleus and putamen, nucleus accumbens, midbrain reticular format ion, substantia nigra and red nucleus. In rhesus monkeys (Schobben et al., 1977), sodium valproate appeared to accumulate in yellow ligaments of the brain, a feature of distribution observed for almost all antiepileptic drugs; which may suggest a common mode of antiepileptic action.

96

Sodium valproate penetrates into breast milk to the extent of less than 10 % . In a single nursing mother, Espir et al. () 976) reported that milk levels of the drug were about 7pg/ml at a time when serum levels were approximately 95pg/ml. In contrast, the same patient showed milk and serum levels of primidone of 6 and 12pg/ml, associated with drowsiness and sedation in the infant.

.E-

_ __ ____ DPA -

c:

-g e c:

80

DPA -

enteric coated slow release

Q)

o

c:

oo

4:

c..

o

E

::>

:D

(/)

o

15

10

5

Hours

1--------------------1 a ~Cl

.E-

c:

o

80

~ c: o o

60

4:

c..

o

E

2

................... normal DPA _ _ _ _ slow release DPA

100

-~

c:

;

/

. '

........

40 20

Q)

(/)

8

12

16

ttl

20

24

4

8

Drug Drug Drug Time (hour) L..-

2.2 .1 Placental Transfer and Penetration into Breast Milk Transfer of '4C-labelled sodium valproate across the placenta of pregnant rodents was reported by Eymard et al. (l 97 O. There are no data in humans, though sodium valproate is known to be dysmorphogenic in animals (see section 1.5.3).

.................. DPA _____ __ Dopamide

E

C, 100

.,..-

--'

Fig. 3. (a) Comparison of serum levels of di-n-propylacetate (DPA), generated by adm inistration of a single moming dose of 600mg of sodium valproate (DPAL depamide , enteric-coated sodium valproate and a slowrelease preparation of sodium valproate ; (b) fluctuations in serum levels during the day associated with the administration of sodium valproate as tablets or as a slow-release preparation (after Meijer and Meinardi, 1976).

b

97

Sodium Valproate

2 .2 .2 Plasma Protein Binding In human plasma obtained from healthy volunteers sodium valproate was bound strongly to plasma proteins (90 % ) but less so to serum albumin (60 % ), indicating that non-defatted albumin is unsatisfactory for these studies. The amount of binding to a- and yglobulins was negligible. {Jordan et al., 1976). The percentage of drug in the free. unbound form has been given as 5 to I 5 % in studies in 7 patients having serum levels of 45 to 240pg/ml (Espir et al., 1976). Sodium valproate also displaced phenobarbitone from protein binding sites in vitro {Jordan et al ., 1976). but the displacement was not significantly increased by increasing the concentration of valproate from 25 to 200pg/ml. Phenytoin was displaced from binding sites only at high levels of sodium valproate (I 50 to 200pg/mO. These displacements, particularly of phenobarbitone, may be of clinical significance when the ratio of sodium valproate to phenobarbitone levels in plasma is high (see section 5).

2.3 Metabolism and Excretion Elimination of sodium valproate is rapid in man (see section 2. I). Renal excretion of the unchanged drug is very small , however, and sodium valproate is probably metabolised rapidly in man as it is in animals (Schobben et al., 1975). The duration of anticonvulsant action of sodium valproate in rodents is prolonged by bilateral nephrectomy but unaffected by liver damage induced by carbon tetrachloride (Swinyard, 1964). Intravenous administration of 14C-Iabelled drug to mice led to the appearance of radioactivty within 15 minutes in the urine and bile (Schobben and van der Kleijn, I 974b), and urinary excretion of radioactivity after oral administration was 20 % in 3 hours and about 70 % in 24 hours (Eymard et al., 1971; Kukino and Matsumoto, 1971). About 80 % of the excreted radioactivity was present as a glucuronide conjugate (Kukino et aI., 1972). Mass spectrometric identification of two metabolites in rat urine showed them to be 2-n-propylgutaric acid, an end product in view of

its lack of further biotransformation when administered per se, and 2-n-propyl-5-hydroxypentanoic acid, a probable intermediate (Kuhara and Matsumoto, 1974). Faecal excretion of sodium valproate in rodents is minimal, varying between 3 % (Eymard et al., 197 I) and 7 to 13 % (Kukino and Matsumot, 1971; Kukino et al., 1972) in the 24 hours after oral administration. There is also some evidence that from 2 % (Eymard et al., 197 I) to 18 % (Kukino and Matsumoto, 197 I) of the administered drug is excreted in the expired air as carbon dioxide, though it is unclear whether decarboxylation precedes or follows a- or «i-oxidation. Enterohepatic circulation of sodium valproate was suggested by Eymard et al. (1971), who showed that about 7 % of the administered radioactivity appeared in rat bile within 7 hours of oral dosage. The bile contained 6 unidentified metabolites as well as unchanged sodium valproate. Maximum bile excretion of radioactivity occurred after 1 hour, whereas when bile from these rats was administered to untreated rats by duodenal catheter , maximum levels of radioactivity occurred in the bile after 2 hours .

3. Therapeutic Trials There are few controlled trials of sodium valproate in the treatment of epilepsy. and evidence of its efficacy comes mainly from open studies. In general, the drug has been given to patients already receiving but often refractory to other antiepileptic medication, which in some cases has been subsequently reduced in dosage or even withdrawn completely. Sodium valproate is more effective in the generalised than the partial epilepsies, which may require higher dosage. It is particularly effective in patients with 3 cycle per second spike-and-wave discharges in the EEG (Heathfield et al., 1976; Jeavons and Clark, 1974; Jeavons et al., 1977b; Lance and Anthony, 1976). It may playa useful part in the management of grand mal, mixed petit mal/grand mal, drug -refractory temporal lobe epilepsy and myoclonic epilepsy. Its relative lack of sedative effects

Sodium Valproate

98

daily doses of 1,200mg sodium valproate or placebo during an 8-week assessment period. Treatments were then crossed over gradually for a futher 8 weeks. Patients were resident in a special centre for epileptics, and existing antiepileptic medication was continued, except for some reduction of dosage of phenobarbitone and primidone in 3 patients with drowsiness. Mild and transient side-effects (nausea, drowsiness and ataxia) were noted in I 3 patients. 3.1 Controlled Trials Results were assessed 'blind', based upon nursing There have been 3 trials of the double-blind staff records. The frequency of major tonic-clonic crossover type, all of which have shown sodium seizures was reduced by 35 % in patients receiving valproate to be significantly superior to placebo when sodium valproate compared with the frequency on added to the previous antiepileptic medication of placebo, with a corresponding 57 % decrease in the patients with a variety of seizure types (Meinardi, frequency of minor seizures (psychomotor type or 1971; Richens and Ahmad, 1975; Suzuki et al., alterations of consciousness). Plasma levelsof sodium 1972). The results for each seizure type will be dis- valproate ranged from 34 to 68}lg/ml in 4 patients. The study of Suzuki et al. (1972) included 37 cussed in section 3.2, together with the results of unchildren with several different types of epilepsy, princontrolled trials . Meinardi (197 I) studied 42 hospitalised patients cipally absence seizures, of whom 35 (age 5 to 15 with several different types of drug-refractory epilep- years) completed the trial. Following a 4-week period sy, 39 of whom (2 to 38 years old) completed the trial when placebo was added to their usual antiepileptic with 3 dropping out because of side-effects due to medication, both inpatients and outpatients received sodium valproate. Placebo or sodium valproate was either sodium valproate and placebo ethosuximide or added to their usual antiepileptic medication follow- ethosuximide and placebo sodium valproate. Patients ing a base-line period of evaluation. The dose was were then crossed over to the alternate regimen, variable according to age, with children receiving up receiving mean daily doses of 851mg sodium valproto 30mg/kg/day and adults up to 1,800mg daily, ate (400 to 2,400mg) or 437mg ethosuximide (200 to given for 3 or 6 weeks with crossover. Results were I,200mg) over the two 4-week active drug periods. evaluated by a 'blind' physician, based on patient Previous antiepileptic medication was reduced or records and reports by nursing staff and patients, withdrawn where possible, resulting in 20 patients together with an indexing technique based on the being evaluated while on sodium valproate alone and number of seizures, their expected frequency and the I5 while on one or more additional antiepileptic length of treatment period. Results were similar by drugs. Patients kept a daily seizure record and each both the index and by clinical judgement, with 20 patient was assessed blind by an attending physician. patients (51 %) experiencing a greater than 33 % The order of drug administration had no effect on the reduction in seizure frequency during treatment with results, which were rated as excellent (greater than sodium valproate as compared with the placebo 75 % reduction in seizure frequency) in 25 patients periods. Blood levelsof sodium valproate were not re- (71 %) during treatment with sodium valproate as ported. compared to placebo. There was no significant In the trial of Richens and Ahmad (1975), 20 difference between the effects on seizure frequency of patients (age 19 to 54 years) with chronic un- sodiuin valproate and ethosuximide. In the group of controlled epilepsy, principally of the psychomotor 19 patients with absence seizures, sodium valproate type with occasional tonic-clonic seizures, received was superior when given alone in 4 of I5 patients and leaves intellectual performance unimpaired, which is important in children. In new patients with petit mal absence seizures, sodium valproate may be the treatment of first choice. However, its present relatively high cost, compared with many available antiepileptic drugs, is an important factor to take into account.

Sodium Valproate

99

Table IV. Results of a double-blind crossover comparison of sodium valproate and ethosuximide in 19 children with absence seizures (data from Suzuki et aI., 1972; see text for details of tr ial)

Treatment

Sodium valproate or ethosuximide given alone

W ith other antiepileptic drugs

Results

No. pts

Reduction in seizure frequency (no. pts) 100%

75-100% 3

All pat ients Sod ium valproate > ethosuximide Sodium valproate = ethosuximide Sod ium valproate < ethosuximide

15

10

4 6

4

All patients Sod ium valproate > ethosuximide Sodium valproate = ethosux imide

4

5

1 3

5 1

3

33 -75%

< 33 %'

-

3 1 2

Includes patients who deter iorated or were unchanged on drug treatment.

in I of 4 when given with other antiepileptic drugs, while ethosuximide was superior in 5 of 15 patients treated with no other drug. The remaining patients responded equally well or poorly to either drug (table IV).

3.2 Results by Seizure Type Simon and Penry (I 975) have already carried out a comprehensive analysis of 25 clinical trials with sodium valproate, which included the double-blind trials of Meinardi (I 97 I) and Suzuki et al. (I 972) and a total of 1,033 patients. Owing to withdrawals and to some patients having more than one type of seizure, their analysis included I, I 16 patients by seizure type (table V). These studies will not be considered further , except to add the overall results to those of the remaining trials not included in Simon and Penry's review (table VI). Not included are reports given at some symposia, for which full details of patients and methods were not available, and some studies which were essentially case reports. The present review considered 2,031 patients by seizure type, 899 (44 %) of whom experienced a reduction in seizure frequency of at least 75 % while on sodium valproate, with 509 (25 %) having a

reduction of 33 to 75% and 623 (31 %) being considered failures of treatment (fig. 4). The duration of therapy and the dosages varied widely between and within studies. Most adult patients received about I,200mg and children about 30 to 40mg/kg, daily in 3 or 4 divided doses, for periods ranging from a few weeks to several years. In only a few studies (Cavazzuti, 1975; Haigh and Forsythe, 1975; Hassan et al., 1976; Jeavons et al., I977a,b; Price, 1976) was sodium valproate given as the sole antiepileptic drug to substantial numbers of patients. In most trials the drug was introduced as an adjunct to other antiepileptic medication, which was reduced in dosage or withdrawn completely as clinical improvement was demonstrated . In the following sections, seizures have been classified according to the Clinical and Electroencephalographical Classification of Epileptic Seizures (Gastaut, 1970). Some types, such as simple and complex absences, or atonic and akinetic seizures, have been grouped together because they were so grouped in several reports containing large numbers of patients. The category.Combinations includes all patients (336) with 2 or more types of seizures not evaluated separately, the most common being absence and generalised tonic-clonic seizures (I 51 patients). The remaining 185 patients in this group include

Sodium Valproate

100

Table V. Results of clinical trials with sodium valproate. by seizure type (after Simon and Penry. 1975)

Seizure type

No. of patients 1.2.3

Reduction in seizure frequency 100-75%

Partial seizures with elementary symptomatology Partial seizures with complex symptomatology Partial seizures.secondarily generalised Absence (simple and complex) Atypical absences Myoclonic epilepsy Infantile spasms Tonic seizures Tonic-clon ic seizures Atonic akinetic seizures Combinations of above Others (not classified above) Total 4

48 127 17 218 67 35 19 5 279 39 236 26 1.116

22 44 3 140 32 21 7 1 147 10 79 3 509 (45.7%)

74-33% 17 37 7 51 12 7 2 2 58 9 76 6 284 (25.4%)

< 33% 9 46 7 27 23 7 10 2 74 20 81 17 323 (28.9%)

1 1.020 patients; 96 patients with two seizure types were included twice. 2 Includes 1 trial with the magnesium salt of dipropylacetic acid (Burin et al., 1972) 3 Data were taken from the following sources : Aoki et al., 1969 ; Balbi and Bravaccio. 1972; Bergamini et al., 1970; Bonnin. 1964; Broglia et al., 1972; Burin et al.• 1972; Campanella et al.• 1972; Cavazutti et al., 1971; de Biolley and Sorel. 1969a. b; Di Genio and Colosimo. 1972. 1973; Dumon-Radermecker. 1969 ; Forster . 1972; Huertas.1969; Jeavons and Clark. 1974; Kerfriden et al., 1970; Mairlot. 1970; Matthes and Schmutterer. 1971 ; Meinardi. 1971; Pazzaglia and Lugaresi. 1972; Scollo-Lavizzari and Corbat, 1970; Suzuki et al., 1972; Vidart et al., 1969 ; Volzke and Doose. 1973. 4 The present review included results in 2.031 patients by seizure type (see text for results f or each type!. 899 (44 %) of these patients experienced a greater than 75 % reduction in seizure frequency. 509 (25 %) had a reduction of 33 to 75 %. and 623 (31 %) were considered failures of treatment. 5 Jeavons and Clark (1974) have now reclassified their 17 patients from th is group as being in the complex absences group (see tables IX and X) (Jeavons. personal communication!.

those identified .as 'mixed' seizure types as well as small numbers of patients with identifiable combinations of seizure types. The heterogeneous group 'Miscellaneous' consists of 59 patients who did not fit clearly into the other classes, and includes such isolated cases as tonic seizures, drop attacks of focal origin , or seizures described as 'idiopathic'. Drug efficacy has been assessed on the basis of the percentage reduction in seizure frequency. Where patients did not fit easily into one of the categories, they were assigned to the next lower category. The number of seizure-free patients has not always been

given in trial reports, and they are therefore included in the group experiencing at least a 75 % reduction. 3.2 .1 Partial Seizures with Elementary Symptomatology The efficacy of sodium valproate in the treatment of focal motor seizures is limited, and it seems unlikely to replace the present drugs of choice except in patients who are resistant to standard treatment. In all cases the drug was added to the usual antiepileptic medication, most often to phenytoin and primidone, the dosages of which were reduced during the trial

Table VI. Characteristics of clinical studies of sodium valproate in the treatment of epilepsy

Author

Simon and Penry (1975)

Type of study '

Review (25 studies)

No. pts admitted

1.033

No. pts completing

1,020

Age of pts

Duration of therapy

5m-67yr

2w k-4yr

Average daily dose (range)

Child: 30-40 mg/kg (15-60) Adult : 1.200mg (200-2,400)

No. pts on sodium valproate alone at end of study 113

Barnes and Bower (1975)

UT

24

24

1-15yr

4-6m

20-80mg/kg

2

Cavazzuti (1975)

UT

47

47

10-42m

1yr

20mg /kg

47 2

Espir et al. (1976)

UT

92

73

5-58yr

6-12m

600-2,4oomg

Grant and Barot (1976)

UT

60

60

7-71yr

26-12wk

Child: 48mg /kg Adult : 33mg/kg

Haigh and Forsythe (1975)

UT

32

32

7m-18yr

8.7m

10- 175mg/kg (l,4oomg)

Harwood and Harvey (1976)

UT

106

94

6-69yr

1-24m

1.8oomg (200-3,200)

Hassan et aI., (1976a)

UT

90

85

0-4Oyr

4-16 m

400-2,4oomg

15

Heathfield et al. (1976)

UT

36

36

-

> 1yr

l ,2oomg (150-2.000)

0

4-29yr

6m-3 yr

600-1 ,6oomg

35

10m-47 yr

4m-3yr

Adult: 6oo-1.6oomg Child: 23-54mg /kg

73

4oo-4 ,ooomg

3

Jeavons et al. (1977a)

UT

63

63

UT

142

134

Lance and Anthon y (1976)

UT

60'

60

6-5Oyr

13m

Olive et al. (1969)

UT

23

23

11 12- 24m

3- 6m

40mg /kg (20-80)

Price (1976)

UT

70

67

-

14m

6oo-5.ooomg

672

1.2oomg

0

Richens and Ahmad (1975)

DBP

20

20

19- 54yr

8wk

Schafer and Kettner ] 1974)

UT

31

31

16-6Oyr

2-14m

900-1,2oomg

Schneble(1975)

UT

15

15

81/2-47yr

4m

l ,2oomg

1 UT = Uncontrolled t rial; DBP = Double-blind. placebo-controlled trial. 2 These patients initially received only sodium valproate .

3 4

3

<

'" if

a '"

CD

23

Jeavons et al. (1977b)

4

(J)

0

a.

E"

0

Includes some patients w ith more than one seizure type . Duration of each treatment period (crossover t rial).

'0

Sodium Valproate

102

period. The overall results of 16 uncontrolled trials (t 0 from Simon and Penry , 1975) and one controlled

tr ial (Richens and Ahmad, 1975) in a total of 121 patients are summarised in table VII and figure 4. Seizure frequency was reduced by at least 75 % in only 34 patients (28 %), and by 33 to 75 % in a further 48 patients (40 %). However , almost a third of the group failed to respond satisfactorily to sodium valproate . The controlled trial of Richens and Ahmad ( l 975) included 4 patients with focal motor seizures. Two experienced a reduction in seizure frequency of about 50 % , one was unchanged and the fourth patient was worse on the drug. Patients received a fixed dose of 1,200mg sodium valproate or placebo daily, during a double-blind trial with crossover at 8 weeks. In the largest uncontrolled trial , Harwood and Harvey (l 976) gave doses of 600 to 3,200mg daily to 29 patients with severe drug-refractory focal epilepsy; 22 patients (76 %) experienced a greater than 50 % reduction in seizures, with 7 becoming almost free from seizures for up to 2 years.

3.2 .2 Partial Seizures with Complex Symptomatology (Psychomotor or Temporal Lobe Seizures) The efficacy of sodium valproate in the treatment of temporal lobe epilepsy is similar to that in focal motor seizures. However, the often refractory nature of temporal lobe seizures to control by conventional antiepileptic drugs may lend sodium valproate a useful role in the treatment of this condition . In all but a few patients (Burin et al., 1972; Lance and Anthony , 1976) sodium valproate has been used as an adjunctive agent. Table VIII lists the results of 26 uncontrolled trials (I 6 from Simon and Penry, 1975) and 2 controlled trials (Richens and Ahmad, 1975; Meinardi, 1971 included in Simon and Penry, 1975). Excellent control of seizures was observed in 87 of 308 patients (28 %), but 129 patients (42 %) received little or no benefit from the drug (fig. 4). The two controlled trials included few patients with temporal lobe seizures. Meinardi (I 97 J) found that seizure frequency was reduced by at least 75 % in I of 5 patients, and by 33 to 75 % in 2 others, following doses of 400 to I,800mg daily for 3 to 6 weeks.

ReoduCllO" '" teilUfl! hllQuencv

o

100 to 15

o 1~

•

'00

Al1lypeo1.

(20 ) 11

r II

Nllh .,Iemenlltfy s.ymplom ill olagy

Pillt •• t S~llul P"'lIBI ~.,.."n w llh compleo. MCond",,,ly ,.-mpIOf'fllllology Qef1efahed

'fa eah 11] 1)

01

PartIal W'11Ul'@'

lll!fl\pot'&1 kltM! (loul (nvchomolorl .,11

IJ081

gl:Jl'ld

T '(poul

AI.,p.c.ll

absences (p081 't mall

.M..,Cf!S 16 61

mall t3S41

IJ

TOOOC: ·Clonoc: AIOftIC at

MI''''''"

oJ"nehc

15191

(1141

5g,~"dmo111

M1Uf~

to

33

33 (mcludes

110

Mu Pd ab · 01 comwnce and btNllllOn,\ IQftlc-eIQn'c of U'l lutM Ml lUl"

(polll m.al • gr.nd mill

11851

ch&ng'1I! i1nd delllt'lOfilltonl

~

'''h!l.C~II.aneou.

lnlilflhle

MtolU'"8

"PilSofnS 132)

ly~ (~91

U SlI

Fig. 4. Summary of results of clinical tri als of sodium valproate in various type s of epilepsy (see text for details). Figures in parentheses refer t o the total number of patients with each seizure type .

Sodium Valproate

103

Table VII. Results of treatment of partial seizures with elementary symptomatology (focal seizures) with sodium valproate Author

Simon and Penry (1975) Grant and Barot (1976) Espir et al. (1976) Harwood and Harvey (1976) Hassan et al. (1976a) Lance and Anthony (1976) Richens and Ahmad (1975) Schafer and Kettner (1974) Total

No. pts

48 2 16 29 11 7 4 4 121

Reduction in seizure frequency (no. pts) 100-75% 1

75-33%

< 33 % 2

22 1 (1)

17

7 (0) 2 (1) 2 (0)

15 5 1 2 3

9 1 11 7 4 4 2 1

48 (40%)

39 (32%)

53

34 (28%)

1 Figures in parentheses refer to seizure-free patients. 2 Includes those who were unchanged or deteriorated 3 These patients experienced a greater than 50% reduction (including some seizure-free patients), but the category was not farther broken down.

Table VIII. Results of treatment of partial seizures with complex symptomatology (temporal lobe or psychomotor seizures) with sodium valproate Author

No. pts

Reduction in seizure f requency (no. pts) 100-75% 1

75-33%

Simon and Penry (1975) Barnes and Bower (1975) Beaumanoir (1973) Espir et al. (1976) Grant and Barot (1976) Haigh and Forsythe (1975) Harwood and Harvey (1976) Hassan et al. (1976a) Heathfield et al. (1976) Lance and Anthony (1976) Richens and Ahmad (1975) Schafer and Kettner (1974)

127 5 4 14 45 5 65 18 7 13 2 3

44

37 2 1 43

13 (7)

16

21 3 (1) 3(1) 3 (0)

18 6 1 4 1 2

Total

308

87 (28%)

92 (30%)

< 33% 2 46 3 3 10 16 5 26 9 3 6 1 1 129 (42%)

1 Figures in parentheses refer to seizure-free patients. 2 Includes those who were unchanged or deteriorated. 3 These patients experienced a greater than 50% reduct ion (including some seizure-free patients) . but the category was not further broken down.

Sodium Valproate

Richens and Ahmad (I 975) included II patients with temporal lobe seizures in their study , but only 2 did not have concomitant tonic-clonic seizures; the other 9 are included in the combinations category of figure 4. The two patients were only poorly controlled by sodium valproate, but in the other 9 patients seizure frequency was reduced by at least 75 % in 4 patients and by 33 to 75 % in a further two. Two uncontrolled trials in large groups of patients have demonstrated the efficacyof sodium valproate in severe drug-refractory temporal lobe epilepsy. Grant ami Barot (1976) observed abolition of temporal lobe seizures over 26 weeks of treatment in 7 of 45 patients (adults and children), with reductions of greater than 50 % in the seizure activity of 19 patients (42%). Harwood and Harvey (1976) noted that seizures were reduced by at least 50 % in 39 of 65 patients (60 %) with virtual cessation of seizures in 21 (32 %) of these, though 4 patients actually deteriorated on sodium valproate. In a subsequent study in 9 new patients who had not previously received antiepileptic drugs, the same authors found that sodium valproate abolished temporal lobe seizures in 8 patients . 3 .2.3 Partial Seizures Secondarily Generalised There have been few studies of sodium valproate in the treatment of grand mal associated with focal features. In a collected total of 21 patients, the numbers of patients in each category (75 to I00 % reduction, 33 to 75 % , or less than 33 %) were respectively 4 (20%), 8 (39%) and 9 (41 %). The numbers are clearly too small for proper evaluation (Barnes and Bower, 1975; Simon and Penry, 1975) (fig. 4). 3.2.4 Typical Absence Seizures (Petit Mal) Sodium valproate is very effective in patients with typical absence seizures, particularly those characterised by 3 cycle per second spike-and-wave discharges in the EEG (Heathfield et al., 1976; Jeavons and Clark, 1974; Jeavons et al., 1977b; Lance and Anthony, 1976). The results of 29 studies (I9 from Simon and Penry, 1975) in 354 patients are sum-

104

marised in table IX and figure 4. There was at least a 75 % reduction in seizure frequency in about twothirds of the patients, and only 45 (13 %) patients failed to show significant improvement. The double-blind trials of Meinardi (197 l) and Suzuki et al. (1972) have already been described in detail (seesection 3.1). Of II institutionalised patients with severe drug-refractory absence seizures in Meinardi's study, 2 experiencedat least a 75 % reduction in seizure frequency and 5 a 33 to 75 % reduction; 4 patients were relatively unaffected by the drug. Suzuki et al. (I 972) included 19 patients with absence seizures, comparing the results of 4 weeks of treatment with sodium valproate to those of similar control or ethosuximide treatment periods. The results (see section 3. I and table IV) show that sodium valproate is significantly superior to placebo and indistinguishable from ethosuximide in terms of seizure control. 3.2.5 Atypical Absences Atypical absence seizures respond almost as readily to sodium valproate as do typical absence seizures (table X and figure 4), Half of the total group of 66 patients experienced at least a 75 % reduction in seizure frequency, with about one-third of these being free from seizures. Schneble (I 975) studied 15 patients with absence seizures of the pyknoleptic type, who had previously not responded to either sodium valproate or ethosuximide given alone. When the patients were given a combination of ethosuximide 750mg and sodium valproate I ,200mg daily, II became free from seizures, 3 were almost free, and one experienced a reduction in seizures of about 50 %. The clinical improvement was paralleled by EEG improvement in 12 patients. 3.2 .6 Myoclonic Epilepsy Clinical studies in this review occasionally failed to differentiate between patients with myoclonic epilepsy (bilateral massive epileptic myoclonus) and other types of myoclonic seizure, such as myoclonictonic-clonic seizures (which we have classified under Combinations) or myoclonic absences (which we

Sodium Valproate

105

Table IX. Results of treatme nt of typical absence seizures (simple and complex) with sodium valproate

Author

No. pts

Reduction in seizure frequency (no. pts) 100-75%'

75 -33%

< 33 %2

Simo n and Penry (1975) Barnes and Bower (19 75) Beaumanoir (1973) Grant and Barot (1976) Haigh and Forsythe (1975) Hassan et al. (1976a) Heathfield et al. (1976) Jeavon s and Clark (1974)3 Jeavons et al. (1977b) Lance and Anthony (1976) Rit z and Jacobi (19 73)

218 1 7 5 9 8 8 17 42 26 13

140 1 4 2 (1) 8(7) 4 (2) 8 14 (12) 40 (36 ) 15 (9) 7

51

27

2 1 3

1 2 1 1

1 4 4

3 1 7 2

Tot al

354

243 (68%)

66 (19 %)

45 (13%)

1 Figu res in parenthese s refer to seizure- free pat ients 2 Includes tho se who were unchanged or deteriorated 3 These patients were included in the atyp ical absence gro up by Simo n and Penry (see table V), but have subsequently been reclassified (Jeavons, personal commu nicat ion).

Table X. Results of treatmen t of atyp ical absence seizures with sod ium valproa te

Autho r

No. pts

Reduct ion in seizure f requency (no. pts) 100-75%'

Burin et al (1972)3 Campanella et al. (1972) Cavazzut i et al. (1971) de Biolley and Sorel (1969a) de Biolley and Sorel (1969b) Dumon-Radermecker (1969) Hassan et al. (1976a) Schneble (1975)

2 1 12 6 4 25 1 15

To tal'

66

75-33%

<33 %2

2 2 3 11 1 (0) 14 (11) 33 (50%)

3 1 1 7

13 (20%)

1 7 2 3 7

20 (30%)

1 Figures in parentheses refer to seizure-free pat ients 2 Includes tho se who were unchanged or deteriorated 3 Evaluated magnesiu m salt of dipropylacetic acid 4 This total excludes the 17 pat ients of Jeavo ns and Clark (1974), wh ich have been reclassified as comp lex absences (Jeavons, personal communication) and transferred to tab le IX.

106

Sodium Valproate

Table XI. Results of treatment of myoclonic epilepsy with sodium valproate

Author

No. pts

Reduction in seizure frequency (no. pts) 100-75%'

Simon and Penry (1975) Barnes and Bower (1975) Grant and Barot (1976) Hassan et al. (1976a) Jeavons et al. (1977b) Lance and Anthony (1976) Mortier and Michaelis (1975) Olive et al. (1969) Total

35

3 3 2 40 14

21 1 2 (1) 2 (11

7

36 (28)

2 3

3

8 (3) 2 (2)

1

1

101

73 (72%)

1

2

75-33%

1

< 33 %2 7 1

1 2

3 1

14 (14 %)

14 (14 %)

Figures in parentheses refer to seizure-free pat ients Includes those who were unchanged or deter iorated

have classified under Complex Absences). The studies listed in table XI refer strictly to bilateral massive epileptic myoclonus, and demonstrate that sodium valproate is effective in reducing seizures by over 75 % in almost three-quarters of patients (fig. 4). The numbers of patients are too small to allow a definite statement regarding the efficacy of the drug, but it does seem promising in myoclonic epilepsy. Jeavons et al. (l977b) regard it as the drug of choice for myoclonic epilepsy, especially in children. Sodium valproate may also be effective in the long-term treatment of progressive myoclonic epilepsy (Tomlinson, 1974). 3.2.7 Tonic-Clonic Seizures (Grand Mal) Sodium valproate appears to be effective against generalised tonic-clonic seizures. In 29 trials ( 18 from Simon and Penry, 1975), summarised in table XII, the drug was given to most of the 519 patients as an adjunct to other antiepileptic drugs, to most of which the patients were already refractory . 239 patients (46 %) experienced a reduction of 75% or more in seizure frequency, but the drug was ineffective in about one-third of patients (fig. 4). Few patients

received sodium valproate alone (Jeavons et al., I977a,b), and the controlled trials of Meinardi (1971) and Suzuki et al. (I972) included only a total of 7 patients with grand mal. There are few reports in the literature of sodium valproate aggravating or precipitating generalised tonic-clonic seizures (Grant and Barot, 1976; Heathfield et al., 1976). The possibility exists that the action of sodium valproate against generalised tonic-clonic seizures is mediated through its elevating effect on serum levels of other antiepileptic drugs, particularly phenobarbitone (see section 5. O. However, most patients do require dosage reduction of other drugs after introduction of sodium valproate, and, in the largest uncontrolled trial, Grant and Barot (1976) found that 29 of 59 patients experienced a greater than 50 % reduction in seizure frequency, associated in the total group with a reduction in the dosages of phenobarbitone of 61 % and of primidone of 31 %. These drugs were no longer required in a total of 10 patients, who remained well controlled on sodium valproate. Furthermore, Price (I 976) found sodium valproate alone to be very effective in patients with severe, drug-refractory generalised epilepsy. Jeavons et al. (1977a ,b) ob-

107

Sodium Valproate

Table XI/. Results of treatment of generalised tonic-clonic seizures (grand mall with sod ium valproate

Author

No. pts

Reduction in seizure frequency (no. pts) 100-75%'

Simon and Penry (1975) 8ames and Bower (1975) Espir et al. (1976) Grant and Baret (1976) Haigh and Forsythe (1975) Hassan et al. (1976a) Heathfield et al. (1976) Jeavons et al. (1977a) Jeavons et al. (1977b) Lance and Anthony (1976) Olive et al. (1969) Schafer and Kettner (1974)

279 2 51 59 11 10 3 35 28 23 4 14

147

Total

519

239 (46 %)

18 4 4 2 27 18 9 3 7

75-33% 58

(8)

14 3 23

(1) (1)

3

(20) (18) (3) (1) (7)

4 2 7 1 5 117 (23 %)

< 33 %2 74 2 37 18 7 3 1 4 8 7 2 163 (31 %)

1 Figures in parentheses refer to seizure-free pat ients 2 Includ ing those who were unchanged or deteriorated 3 These patients experienced a greater than 50 % reduct ion (inc luding some seizure-free pat ients} , but the category was not further broken down

tained complete control of tonic-clonic seizures with sodium valproate alone in 37 of 52 patients (see section 3.5).

3.2.8 Atonic or Akinetic Seizures Atonic and akinetic seizures have been grouped together in most of the trials which have reported them (table XIII). About one-third of patients experienced at least a 75 % reduction in seizure frequency during treatment with sodium valproate, of whom about half were seizure-free. However, 52 of 114 patients (46 % ) were not adequately controlled on the drug (fig. 4). In the large uncontrolled trial of Grant and Barot (1976) , almost half of the 29 patients with atonic-akinetic seizures experienced a greater than 80 % reduction in their frequency, but 10 patients were worse. Of the 32 patients reported by Jeavons et aI. (I 977b), 31 % improved by more than 80 %.

3.2.9 Combinations of Individual Seizure Types The efficacy of sodium valproate against generalised tonic-clonic (section 3.2.7) or absence seizures (sections 3.2.4 and 3.2.5) is also reflected in its action against mixtures of these two types (table XIV and figure 4). In a total of lSI patients, from 16 uncontrolled studies (I I from Simon and Penry , 1975), sodium valproate reduced seizures by at least 75 % in 65 patients (43 %) and was moderately effective in a further 42 patients (28 %) (fig. 4). Sodium valproate is somewhat less effective in the broad group of other mixed seizure types (table XV and figure 4). Of 185 patients , mainly with generalised tonic-clonic seizures combined with myoclonic, temporal lobe and/or focal seizures, about two-thirds experienced a greater than 33 % reduction in seizure frequency. In a double-blind crossover trial , Richens and Ahmad (I 975) included 9 patients with temporal lobe

Sodium Valproate

108

Table XIII. Results of treatment of aton ic or akinetic seizures w ith sodium valproate

Author

No. pts

Reduction in seizure frequency (no. pts) 100-75%'

Simon and Penry (1975) Grant and Barot (1976) Jeavons et al. (1977b) Lance and Anthony (1976) Total

1 2

39 29 32 14

10 13 (9) 10 (8) 6 (4)

114

39 (34 %)

9 5 6 3

20 11 16 5

23 (20 %)

52 (46%)

Figures in parentheses refer to seizure-free patients Includ ing those who were unchanged or deter iorated

Table XlV. Results of treatment of mixed absence and generalised tonic-clonic seizures (petit mal sodium valproate

Author

No. pts

Simon and Penry (1975) Beaumanoir (1973) Haigh and Forsythe (1975) Hassan et al. (1976a) Heathfield et al. (1976) Ritz and Jacobi (1973)

103

Total

151

1 2

<33%2

75-33%

8 4 21

7 8

grand mal) with

Reduction in seizure frequency (no. ptsl 100-75%'

75-33%

< 33 %2

40 4 2 (2) 10 (2) 7 2

28 3 8

35 1 2 3

3

3

65 (43%)

42 (28%)

44 (29%)

Figures in parentheses refer to seizure-free patients Including those who were unchanged or deter iorated

seizures associated with tonic-clonic seizures. After 8 weeks of treatment, sodium valproate was significantly superior to placebo, abolishing tonic-clonic fits in 5 patients and reducing minor fits by greater than 50 % in 3 patients.

patients with tonic seizures, seizures associated with degenerative cerebral lesions, 'idiopathic' epilepsy and those seizures of uncertain origin.

3.3 Childhood Epilepsy 3.2 .10 Miscellaneous Types of Seizure It is not possible to evaluate patients in this group because of the wide diversity of types of seizure involved. Table XVI lists those studies which included

In many trials with sodium valproate it is difficult to decide how many patients were children, though the age-range given or the use of paediatric dosages

Sodium Valproate

109

Table XV. Results of treatment of several seizure types (except for mixed absence and generalised tonic-clonic seizures) with sodium valproate Author

No. pts

Simon and Penry (1975)

133

Types of seizure

Reduction in seizure frequency (no. pts)

Mixed

Bames and Bower (1975)

6

Temporal lobe or myoclonic/grand mal

Haigh and Forsythe (1975)

3

Myoclonic/akinetic

Hassan et at, (1976a) Heathfield et al. (1976)

19

Mixed

100-75%'

75-33%

< 33 %2

39

48

46

3

2 3

6 (2)

5

Temporal lobe /grand mal

Richens and Ahmad (1976)

9

Temporal lobe /grand mal

4

Schafer and Kettner (1976)

10

Temporal lobe or focal or petit mal/grand mal

4 (4)

Total

1

2

185

11

2 4

56 (30 %)

4

4

67 (37 %)

2

62 (33 %)

Figures in parentheses refer to seizure-free patients. Including those who were unchanged or deteriorated.

often indicate that children were included. Table XVII has been constructed from the results of studies which specifically identified children; these results were also included, with the exceptions of the specifically paediatric epilepsies of infantile spasms and the Lennox-Gastaut syndrome, in the preceding sections on individual types of epilepsy in all age groups. Most of the children were selected for these trials because they were poorly controlled on other antiepileptic medication. Despite the obvious limitations of pooling the results of trials which have been conducted and reported in different ways , it is clear that sodium valproate is particularly effective in children with absence seizures or tonic-clonic seizures, reducing seizure frequency by at least 75 % in the great majority of patients. The remaining groups are too small for meaningful appraisal, but the efficacy of the drug seems more limited, particularly for temporal lobe epilepsy and for grand mal associated with focal

features. However, Jeavons et al. (J 977b) feel that sodium valproate is the drug of choice for myoclonic epilepsy in childhood or adolescence when it may be given safely with nitrazepam but rarely with clonazepam. Infantile spasms and the Lennox-Gastaut syndrome do not appear to respond as effectively to sodium valproate as they do to the benzodiazepines, particularly nitrazepam. Olive et al. (J 969), however, reported abolition of clinical and EEG signs in 5 of 10 infants with infantile spasms following treatment with sodium valproate, while 3 others showed no clinical signs but with continuing hypsarrhythmia in the EEG; all patients had relapsed after initial successful treatment with ACfH and hydrocortisone. In one of the largest trials in children, average doses of 21mg/kg/day (20 to 40mg/kg/day) were given to 80 patients who suffered either from seizures resistant to previous medication, idiopathic epilepsy or progressive myoclonus epilepsy (Siilanpaa and

Sodium Valproate

110

Table XVI. Results of treatment of miscellaneous t ypes of seizure with sod ium valproate

Author

Reduction in seizure f requency (no. pts)

No. pts

100-75% Simon and Penry (1975) Espir et al. (19 76) Heathf ield et al. (197 6) Olive et al. (1969) Richens and Ahmad (1975)

31 11

Total

59

4

75 -33%

19 7 3 2

8

42

4 9 4

< 33 %'

1 7

2 14 (24%)

14 (24 %)

31 (52%)

Including those who were unchanged or deter iorated 2 These pat ients experienced a greater than 50 % reduc tion (including seizure-free pat ients), but th e category was not furthe r broken down.

Table XVII. Results of treatment of childhood epilepsy w ith sod ium valproate

Type of seizure '

No. pts

Reduction in seizure frequency (no. pts) 100-75%

75 ·33%

< 33 %2

14

25

Tonic-clonic seitures (grand mal)

122

78 (64%)

Absences (peti t mal)

105

93 (89%)

6

6

Partial seizures with complex symp tomatology (temporal lobe)

47

14 (30%)

13

20

Partial seizures secondar ily generalised (focal /grand mal)

28

10 (36%)

6

12

Infantile spasms

32

12 (38%)

5

15

Lennox-Gastaut syndrome

62

23 (37 %)

14

25

Myoclonic4

17

16 (94%)

Others not specified alone>

78

39 (50%)

12

27

1 Data we re t aken from th e sources listed below. Most of the patients we re selected as being resistan t to treat ment w ith other antiepileptic drugs (Barnes and Bower , 19 75 ; Beaumanoir, 1973; Dumon-Radermecker, 1969 ; Espir et al. 1976; Forster, 1972; Grant and Baret, 1976; Heathfield et al., 1976 ; Haigh and Forstythe, 1975 ; Jeavons et al., 1977b ; Olive et al. 1969; Ritz and Jacob i, 1973; Simon and Penry, 1975 ; Volzke and Doose, 1973). 2 Including those who were unchanged or deter iorated . 3 Each category had too few patients to be analysed separately (includes myoclonic epilepsy , atyp ical and extended absences, atonic-akinetic seizures and mixed seizures). 4 Data taken only f rom Jeavons et al. (1977b)

Sodium Valproate

Donner, 1976). Sodium valproate was most effective in children with idiopathic epilepsy (petit mal absences, photogenic epilepsy , myoclonic petit mal), in seizures beginning in later childhood years, photosensitive and hyperventilation-sensitive seizures , and in progressive myoclonus epilepsy. Excellent results were also noted when the EEG showed symmetric generalised spike-slow wave complexes provoked by photostimulation or when the EEG was without generalised disturbance or focal findings. The best results were obtained in children with normal neurological and mental status. Overall, 69 % of the group experienced a greater than 50 % reduction in the frequency of seizures .

3.3 .1 Febrile Convulsions Sodium valproate may be useful in the prevention of febrile convulsions. Cavazzuti (J 975) studied 141 children aged 10 to 42 months, who had experienced at least one febrile convulsion without neurological deficit. Two groups of 47 patients received sodium valproate 20mg/kg daily or were untreated, while a third group of similar size received either phenobarbitone 5mg/kg daily or primidone 25mg/kg daily. In the untreated group, 55 % of patients experienced I to 4 febrile convulsions within I year, compared with only a single recurrence in about 4 % of each of the drug-treated groups. There were no side-effects in any group, though phenobarbitone prophylaxis of febrile convulsions has been previously associated with hyperactivity and irritability.

3.4 Sodium Valproate Compared With Other Drugs In general sodium valproate has been given to drug-refractory patients, whose dosage of other antiepileptic drugs has often been subsequently reduced or occasionally withdrawn. In one sense, therefore, it could be said that sodium valproate is superior to these other drugs. However, the only strict comparisons are those with ethosuximide in the treatment of absence seizures (Suzuki et aI., 1972), with

111

phenobarbitone and primidone in the prophylaxis of febrile convulsions (Cavazzuti, 1975), and with clonazepam in various types of epilepsy (Lance and Anthony, 1976). The double-blind trial of Suzuki et aI. (J 972) has already been described (section 3.1 and table IV). There was no significant difference between the antiepileptic effects of sodium valproate 500 to 2,400mg daily (mean 851 mg) and ethosuximide 200 to 1,200mg daily (mean 437mg), given to 19 children with typical absence seizures for periods of 4 weeks . Other antiepileptic medication was continued in 4 patients but was withdrawn completely in 15. In the prophylaxis of febrile convulsions, there was no significant difference between the effects of sodium valproate 20mg/kg/day and either phenobarbitone 5mg/kg/day or primidone 25mg/kg/day. This study by Cavazzuti (J 975) in 141 infants has already been described in section 3.3.1. Of 32 patients who were treated sequentially with clonazepam 3 to 12mg daily and sodium valproate 400 to I ,200mg daily, Lance and Anthony (J 976) reported that II responded better to c1onazepam, 10 responded better to sodium valproate, 10 responded to neither and I required both drugs. Qonazepam was superior in 4 patients with temporal lobe epilepsy, 5 patients with petit mal absences (I associated with grand mal and I with atonic seizures), and 2 with atonic seizures. Sodium valproate was more effective in 2 patients with myoclonus, 6 patients with petit mal absences (2 associated with grand mal and 2 with atonic seizures), and 2 patients with grand mal alone. In most patients, the previous antiepileptic medication was reduced in dosage or number of drugs. 3.5 Sodium Valproate Alone? In only a few studies was sodium valproate given

its the sole antiepileptic drug to substantial numbers of patients (see table VI). The evidence suggests that it is effective. Of 75 previously uncontrolled patients with partial and generalised epilepsy studied by Hassan et aI.

112

Sodium Valproate

(I 976a). only 2 (3 %) were seizure-free and 17 (23 %)

had occasional seizures when sodium valproate was added to their previous antiepileptic medication. However. of 15 other patients who received sodium valproate as the first and only drug (5) or as a replacement for other ineffectualsingle drugs (I 0). 8 patients (53 %) were seizure-free and 5 (33 %) had occasional seizures; the remaining 2 experienced at least a 50 % reduction in seizure frequency. In a further report. including another 25 patients, Hassan et al. (I 976b) demonstrated a reduction in seizure frequency of greater than 50 % in about 70 % of 80 patients with generalised seizures and 37 % of those with partial seizures. All but 6 patients had intractable, drug-resistant epilepsies. A number of patients reported increased alertness, increased appetite and improved school performance. In the double-blind trial of Suzuki et al. (1972). 10 of I5 children with typical absence seizures treated with sodium valproate alone (by withdrawal of other drugs) were seizure-free and 3 others experienced at least a 75% reduction in seizure frequency (see table IV). In contrast . Schneble (I 975) found sodium valproate alone. and ethosuximide alone. to be ineffective in 15 patients with atypical absence seizures (see section 3.2.5). Seizure control was achieved only when the two drugs were given together. Sodium valproate 20mg/kg/day was as effective over I year of treatment as phenobarbitone 5mg/kg/day or primidone 25mg/kg/day in the prophylaxis of febrile convulsions in 141 infants who had experienced at least I febrile convulsion without neurological deficit (Cavazzuti, 1975; sect. 3.3. I). Sodium valproate alone was given to 23 of 32 children with various types of drug-refractory epilepsy (Haigh and Forsythe. 1975). Effectivecontrol of seizures was obtained in 7 of 9 patients with petit mal absences and in 4 of I I with tonic-clonic seizures (some continued to receive other drugs), but sodium valproate was ineffective in 5 children with temporal lobe epilepsy and in 3 with myoclonic/akinetic seizures. In contrast. Harwood and Harvey (I 976) reported that 8 of 9 patients with temporal lobe epilepsy who were given sodium valproate as the drug of first

choice were seizure-free; the ninth patient. after an initial seizure-free period, went into status epilepticus. Price (I 976) gave sodium valproate alone in doses of 600 to 5.000mg daily for 14 months to 70 neurosurgical patients. These patients. principally with partial epilepsies (87 %). had an estimated seizure risk of at least 20 % . Ofthe highest risk group of 30 patients, who were having more than 5 seizures a month prior to the trial, 26 obtained complete seizure control with sodium valproate. Patients in the lower risk groups had no seizures on the drug . Three of the 70 patients were unable to tolerate sodium valproate because of side-effects. but two were fully controlled; the third patient remained seizure-free on a lower dose together with primidone and phenytoin. Of 35 patients with 'television epilepsy' who were treated from the start with sodium valproate alone. 27 were improved by more than 80 % . 20 becoming seizure free and having normal EEGs during photic stimulation (Jeavons et al.• I977a.b). 73 other patients received sodium valproate alone. 23 having been given the drug alone from the start. 23 patients with absences became seizure-free as did 17 with tonic-clonic seizures. 21 with myoclonic epilepsy. and 4 with myoclonic atonic epilepsy.

3.6 Tolerance and Long-Term Treatment Sodium valproate, either alone or with other antiepileptic drugs. has been given to patients for periods up to 4 years (see table VI). without any signs of chronic toxicity or the development of tolerance (Noronha and Bevan. 1976). Though the possibility exists that some other antiepileptic drugs may induce its metabolism (Richens et al., 1976; see section 2.1.1), there is no evidencethat the long-term efficacy is affected. Tolerance to antiepileptic drugs can occur with long-term treatment. particularly to those effective in petit mal (see Haigh and Forsythe, I 975). but it does not seem to be a problem with sodium valproate. Adults may be more likely to deteriorate during longterm treatment than children or adolescents

Sodium Valproate

(Heathfield et aI., 1976). Mortier and Michaelis (I 975), however, did report that 2 patients with

myoclonic jerks showed a diminished response to sodium valproate 900mg daily, after several weeks of treatment, even when the dosage was increased. After a 3-week drug-free period, sodium valproate was again effective in controlling the myoclonus. Jeavons et al. (1977b) reported temporary relapse as a quite common feature of sodium valproate treatment of febrile convulsions.

3.7 Status Epilepticus Sodium valproate seems unlikely to replace the present drugs of choice, diazepam and clonazepam , in the treatment of status epilepticus , but it should be tried in patients who are resistant to standard treatment . In a single patient with grand mal status who had failed to respond to intravenous diazepam and paraldehyde together with intramuscular phenytoin and phenobarbitone, sodium valproate 2,400mg daily for 5 days by intragastric drip produced rapid recovery (Manhire and Espir , 1974). In a mentall y retarded patient with Lennox-Gastaut syndrome, oral sodium valproate was effective in relieving status after 2 to 4 weeks of treatment in an intensive care unit; the patient's condition had previously failed to respond to 10 other antiepileptic drugs, including clonazepam and paraldehyde (Vajda et aI., 1976).. A child in status epilepticus, who did not respond to intravenous diazepam in doses totalling 45mg, responded within 48 hours to intragastric sodium valproate 30mg/kg/day (Barnes et aI., 1976). AIl three patients were subsequently maintained on oral sodium valproate without further episodes of status epilepticus.

3.8 Miscellaneous Therapeutic Trials

3.8.1 Dyskinesias On the basis of its elevating effects on brain GABA levels (see section 1.2), sodium valproate

113

1,500mg daily either alone or with GABA 24.5g daily for 21 days , was administered to 8 patients with Huntington's chorea, which is characterised by reduced levels of brain GABA (Shoulson et aI., 1976). Neither mode of treatment in this double-blind crossover trial was effective in ameliorating the motor signs of the disease, though combined treatment did raise central turnover of dopamine and serotonin. Sodium valproate was ineffective in alleviating choreiform movements in 5 patients, 3 of whom had hereditary Huntington's disease and 2 with severe chorea developing late in life. The drug was administered in doses of 600 to 1,200mg daily (Lenman et aI., 1976). However, the drug may be effective in ameliorating tardive dyskinesia in patients treated for long periods with antipsychotic drugs (Linnoila et aI., 1976). In a double-blind crossover study in 32 chronic psychiatric patients, oro -facial dyskinesias were totally or significantly relieved in I 7 cases by sodium valproate 900mg daily, but not at all by placebo. Involuntary movements of the extremities and dystonic spasms were also significantly relieved in 7 of 9 patients during the active treatment period , but 2 patients became worse. A significant improvement was noted in the psychiatric symptoms of 14 out of 32 patients , but 4 deteriorated during administration of sodium valproate. There was no correlation between the serum concentration of drug and its effect on the dyskinesias or on the psychiatric symptoms.

3 .8.2 Liver Disease Sodium valproate, I to 1.6g daily for adults and 800mg daily for children , was given to patients with viral hepatitis or liver cirrhosis for 30 or 60 days respectively (Carraz and Bernard , 1967). There was rapid improvement in the general condition of patients with viral hepatitis ; 60 of 100 showed no asthenia and none lost weight as a result of hepatitis . In cases of alcoholic cirrhosis there was significant improvement in general condition, increased appetite, and increased weight, while there was a marked reduction in hepatomegaly where this existed before

Sodium Valproate

114

Table XVIII . Overall incidence (%) of side-effects during sodium valproate treatment of epilepsy

Side-effects 1

Noronha and Bevan

Simon and Penry

(1976)

(1975)

Total (3,228)

Children (251)

Total (1,033)

22.3

16

Gastrointestinal

9.3

Drowsiness/ sedation (due to sodium valproate alone)

0.2

Drowsiness/ sedation (in association with other antiepileptic drugs)

4.9

4.0

Decreased appetite

0.7

0.8

Increased appetite

0.6

0.2

Temporary hair loss

0.5

Worsened behaviour

0.9

2.4

Others

1.4

1.6

1 Mental stimulation or excitation also occurs in many patients. While some complain of racing thoughts and an inability to sleep, the majority feel brighter and more alert when treated with sodium valproate. This occurs in patients treated with sodium valproate alone, as well as in those receiving other antiepileptic drugs.

treatment. There have been no further reports of these effects of sodium valproate, and controlled trials are clearly needed.

4. Side-Effects Sodium valproate is remarkably free of sideeffects, in the general context of antiepileptic drugs. Two surveys have been carried out (table XVIII), one in 1,033 patients participating in 25 clinical trials (Simon and Penry, 1975) and the other in 3,228 patients who had received daily doses of 100 to 3,OOOmg for periods up to 4 years (Noronha and Bevan, 1976). 4.1 Gastrointestinal Effects Nausea, vomiting, abdominal cramp, diarrhoea and indigestion are the most commonly reported ad-

verse effectsof sodium valproate, occurring in from 9 to 16 % of adults and over 22 % of children. In general, the effects have been transient , only rarely requiring withdrawal of drug (0.9 %) or limitation of dose (0.5 %). Gastrointestinal irritation by sodium valproate can be minimised in most patients by giving the drug with food or by slowly building up the dose from an initial low level.

4.2 eNS Effects The true incidence of drowsiness and sedation with sodium valproate is difficult to assess, since few patients have received the drug alone and it is known to increase the blood levels and hence the sedative effects of certain other antiepileptic drugs, particularly the barbiturates (see section 5). Of 58 patients who received only sodium valproate, none showed signs of sedation (Jeavons et al., I977b). Neverthe-

Sodium Valproate

less, sodium valproate does have an intrinsic sedative action (see section 1.3.2), and unexplained coma has occurred during clinical trials (see, for example, Harwood and Harvey, 1976). Noronha and Bevan (I 976) give the overall incidence of drowsiness and sedation as 5.1 %, of which only 0.2 % could be attributed directly to sodium valproate. Drowsiness and sedation are usually transient, and can generally be corrected by a reduction in dosage of the other antiepileptic drugs . Inco-ordination and ataxia have rarely been observed. Jeavons and Clark (I 974) reported 4 patients with ataxia, Haigh and Forsythe (I 975) observed I patient, and Richens and Ahmad (I 975) reported 3, but all were on additional antiepileptic medication, the withdrawal or reduction of which relieved the ataxia. Static tremor has been reported in some studies, but tends to occur at high doses and plasma levels of sodium valproate above 120pg/ ml (Espir et al., 1976; Harwood and Harvey, 1976; Price, 1976).

4.3 Changes in Appetitie Sodium valproate has variable effects on appetite. Both anorexia and increased appetite have been reported; for example, in 12 % and 9.5 % respectively of 164 patients (Kruse, (974) and in 25 % each of 24 patients (Barnes and Bower, 1975). Weight gain associated with increased appetite has been observed in 7 % of 90 patients (Hassan et al., 1976a,b) and 4 % of 116 patients (Volzke and Doose, 1973), while Price (I 976) reported weight gains of 5 to 20kg during 14 months of treatment in 6 of 70 patients. The overall incidences of either anorexia or increased appetite are very low, however, less than I % of all patients (table XVIII).

4.4 Hair Loss Loss and/or coarsening of hair has been observed in some patients receiving sodium valproate with

115

other antiepileptic drugs, but there are few reports of hair loss occurring in patients receiving it alone (Hassan et al., 1976b; Jeavons et al., I977b; Simon and Penry, 1975). The condition has usually been temporary, has not necessitated withdrawal of drug or reduction of dosage, and was of little concern to the affected patients. The loss of hair is usually slight (Barnes and Bower, 1975; Hassan et al., I976a,b ; Schlack, 1973; Volzke and Doose, 1913), and was confirmed dermatologically in only 6 of 15 patients who reported hair loss as a side-effect (Jeavons and Clark, 1974). Sporadic cases have been reported in general (see Simon and Penry , 1975), but Kruse (I 974) observed an incidence of 7 % in 171 patients and Jeavons and Clark (1974) of about 15 % in 63 patients.

4.5 Behavioural Changes Behaviour may be improved after sodium valproate, many patients being more lively and alert and better able to carry out their daily tasks . This may reflect reduced side-effects from the concomitant drugs which have been reduced in dosage, although mental stimulation and excitement can occur in the absence of any changes in drug dosage and the same effects can occur on sodium valproate alone (Haigh and Forsythe, 1975; Grant and Barot, 1976; Hassan et al., I976a,b; Jeavons and Clark, 1974; Lance and Anthony, 1976; Timpany , 1976). Worsened behaviour with aggressiveness and hyperactivity may also occur in children (Barnes and Bower, I975; Haigh and Forsythe, 1975; Hassan et al ., I976a,b ; Jeavons and Clark , 1974). In a double-blind trial in 20 epileptic patients with aggressive behaviour problems, there was no significant difference between the psychotropic effects of sodium valproate or placebo, 300mg 3 times daily (Sonnen et al., 1975). Five patients were rated as improved after 3 weeks on sodium valproate, 3 deteriorated and there was no change in 12 patients; after placebo, 6 were improved, 3 worse and I I unchanged.

Sodium Valproate

4.6 Laboratory Abnormalities Patients on sodium valproate and other antiepileptic drugs have experienced slight increases in serum alkaline phosphatase (Suzuki et al., 1972) and transient minor elevations of serum SGOT (Volzke and Doose, 1973). These have not persisted and there were no signs of organ damage. Relative lymphocytosis and slight thrombocytopenia have also been noted in isolated cases, with prolonged bleeding times (Espir et al., 1976; Sutor and JeskinskyBuscher, 1974; see section 1.4), and these effects may be particularly prevalent in children (Sutor and Jesdinsky-Buscher, 1976). Most of the patients have also been receiving other antiepileptic drugs, but thrombocytopenia has been attributed to sodium valproate alone (Winfield et al., 1976).

5. Drug Interactions 5. 1 Other Antiepileptic Drugs Sedative effects are observed less often in patients receiving sodium valproate alone than in those taking it in addition to other antiepileptic drugs, particularly phenobarbitone (Noronha and Bevan, 1976) and clonazepam (Jeavons et al., 1977b) . In many patients receiving phenobarbitone, phenytoin and primidone, it is frequently necessary to reduce their dosage upon introduction of sodium valproate. It is likely that the reason lies in the elevation by sodium valproate of the plasma levels of these antiepileptic drugs, though it is also possible that some others, such as ethosuximide and diazepam, raise the levels of sodium valproate (Loiseau et al., 1975). Phenytoin, phenobarbitone, primidone, carbamazepine and pheneturide appear to induce liver enzymes and may shorten the half-life of sodium valproate, an effect not shared by ethosuximide or the benzodiazepines (see sections 2.1.1 and 5.2). Several authors have reported elevated phenobarbitone levels following sodium valproate administration . Schobben et al. (I 975) found that steady state

116

levels of phenobarbitone rose by 35 to 200% in 3 children who had sodium valproate 450 to 900mg daily added to their usual dose of phenobarbitone (I 50mg daily). Richens et al. (I 976) observed significant increases of 17 to 48% (mean 27%) in serum levels of phenobarbitone in each of 7 patients receiving both drugs . A significant rise of 46 % (34 to 71 %) was noted in each of 6 patients, when 600mg sodium valproate was added to their daily dose of 300mg phenobarbitone (Vakil et al., 1976). In a trial in 94 patients ,one-third showed a rise in serum phenobarbitone levels of at least 30 % following sodium valproate (Harwood and Harvey , 1976). The effect of sodium valproate on phenytoin levels is less clear . It caused a dose-dependent rise in plasma phenytoin levels in 3 of 5 patients, associated with signs of phenytoin toxicity (Vajda et al ., 1976). However, Richens et al, (I 976) could detect no consistent changes in either serum levels or half-life of phenytoin in 6 patients who also received sodium valproate. A significant fall in plasma phenytoin was observed in 10 patients on long-term phenytoin when sodium valproate 600mg was added to their regimen (Vakil et al., 1976). Levels returned to normal when the dosage of sodium valproate was increased to 1 to 1.6g daily. Penry et al. (I 976) also observed a significant fall in plasma phenytoin in 3 patients given 20mg/kg sodium valproate. In 5 children receiving 6 to 10mg/kg phenytoin, levels rose from a mean of 8.4J.lg/ml to 2 1.6J.lg I ml within a few days of sodium valproate administration, but returned to normal after 1 to 3 months of combined therapy (Windorfer et al ., 1975). The possibility of antagonism between phenytoin and sodium valproate was suggested by Barnes and Bower (I 975), in a child who failed to respond to valproate whilst taking phenytoin but whose seizures fell by more than 90 % when phenytoin was with drawn. In a similar case, Coulthard (I975) reported that intractable childhood epilepsy was unaffected by 800mg sodium valproate daily added to a regimen of phenobarbitone (I OOmg), pyridoxine (I OOmg) and phenytoin (I20mg). The blood levels of sodium valproate was 51J.lg/ml, which rose to l l Opg/rnl

117

Sodium Valproate

over a 3-week period of phenytoin withdrawal associated with a 75 % decrease in seizure frequency. Primidone levels may also be affected by sodium valproate. Initial increases have been observed, from a mean of 6 .7~g/ml to 16~g/ml in 7 children, followed by a fall to normal values after I to 3 months of combined therapy (Windorfer et al., 1975). The mechanism of these changes in blood levels of other antiepileptic drugs is unknown. It may include enzyme inhibition, competition for protein binding sites or diminished clearance, but the most likely reason is that sodium valproate, being a fatty acid and therefore binding strong to hepatic microsomes, blocks the metabolism of phenobarbitone (and possibly other drugs) in the liver. Sodium valproate has no enzyme-inducing properties (see section 5.2), but may displace other drugs from plasma proteins when its plasma levels are high (see section 2.2.2>Jeavons et al. (l977b) reported five patients who developed absence status when clonazepam was combined with sodium valproate, and 9 of 12 patients showed severe drowsiness on this combination, although nitrazepam and sodium valproate could be combined with benefit and without undue sideeffects. 5.2 Enzyme Induction In contrast to many antiepileptic drugs, sodium valproate has no enzyme-inducing properties. Administration to rats of daily doses of IOOmg/kg for 7 days did not affect the activity of hepatic enzymes, and the enzyme-inducing effects of similar doses of phenobarbitone were unaltered by concomitant administration of sodium valproate (Jordan et al., 1976). The reduced half-life of sodium valproate in some epileptic patients receiving long-term treatment with some other antiepileptic drugs is evidence for induction of liver enzymes by these agents rather than of any inducing action of sodium valproate (Richens et al., 1976). Antiepileptic drugs which might shorten the half-life of valproate include phenobarbitone, pri-

midone, phenytoin, carbamazepine and pheneturide, but not ethosuximide or the benzodiazepines (Richens et al., 1976). Human studies suggest that it is unlikely that sodium valproate is an inducer of its own metabolism or that of other drugs. Recent reports (Vree and van der Kleijn, 1977) suggest that ethanol and sodium valproate may mutually inhibit their elimination rates; indicating interference with alcohol dehydrogenase. The threshold dose for sodium valproate to inhibit the elimination of ethanol was 90mg /kg in the dog and 45mg/kg in rhesus monkeys.

6 . Contraindications and Precautions General: There are no specific contraindications to sodium valproate, and there have been no reports of significant hepatic, renal or cardiac effects attributable to the drug. Minor gastric irritation, and, less frequently, nausea have been observed in some patients at the start of treatment ; should such symptoms persist, they can be relieved by standard medication. Transient hair loss has been noted in some patients. Tremor has been seen occasionally, usually at high doses. Prolongation of bleedingtime, sometimes with thrombocytopenia, has occurred rarely. Spontaneous bleeding or bruising, however, are indications for withdrawal of sodium valproate pending investigation. Platelet function should be monitored before major surgery is undertaken in patients receiving sodium valproate. Pregnancy: Sodium valproate, like certain other antiepileptic drugs, produces dysmorphogenic effects in animals. This possible hazard must be weighed against the benefits, if the drug is to be given to women of child-bearing age. It should probably be used only in severe cases of epilepsy or in those who are resistant to other treatment, though this is a matter for individual clinical judgement. Combined Medication: Sodium valproate is well tolerated in combination with other antiepileptic drugs, but it may enhance their sedative effects, particularly the barbiturates. Introduction of the drug to

Sodium Valproate

existing treatment may require a concomitant reduction in the dosage of other drugs. Absence status has been precipitated, as well as excessive drowsiness, when sodium valproate was given with clonazepam, but not with nitrazepam . Sodium valproate may, like other drugs, potentiate the effects of monoamine oxidase inhibitors and thymoleptics, and the dosage of these agents should be reduced accordingly. Blood platelet aggregation is inhibited by sodium valproate, which may prolong bleeding times , particularly in children and generally at high dose levels. Caution should be exercised when administering sodium valproate with other drugs which affect coagulation (aspirin, warfarin etc.). Diabetic Patients: Sodium valproate is partly eliminated by the renal route in the form of ketone bodies, which may result in false positives when testing the urine for the presence of ketone bodies. In addition, when treating diabetic patients with sodium valproate syrup, account should be taken of the 3.6g of sucrose contained in each 50ml of syrup . Pharmaceutical Preparation: Tablets of sodium valproate are hygroscopic; they must be kept in their protective foil until taken, and should be appropriately stored .

7. Overdosage Reports of accidental overdosage of sodium valproate are rare, but recovery after the ingestion of up to 36g has been uneventful following conservative management (Reckitt & Colman , unpublished data, 1974). Sodium valproate is absorbed very rapidly and gastric lavage may be of limited value. General supportive measures should be applied since the drug is eliminated almost entirely within 24 hours, principally in the urine. Particular attention should be made to the maintenance of adequate urinary output. Pillen (I 973) reported a suicide attempt by a 20year-old man who took Ig phenobarbitone and 300mg phenytoin in addition to 36g sodium valproate. The patient presented in a deep coma (responding

118

only to deeply painful stimuli) about 4 hours after ingestion of the drugs . EEG recording at 16 hours revealed diffuse slowing, compatible with the state of consciousness. The patient gradually became more alert and was discharged the following day. No measurements were made of serum levels of sodium valproate. Increased half-life of sodium valproate 00 hours) and unaltered rate of elimination of phenobarbitone was reported in a suicidal attempt with the drugs, with complete and rapid recovery (van der Kleijn et al., 1974).

8. Dosage Sodium valproate may be introduced alone or added to existing antiepileptic medication. It is available as scored white tablets (200mg) and as a cherryred flavoured syrup containing 200mg/5ml. The tablets should be swallowed whole, with water but not aerated water, and with or after food. In carbonated drinks, free valproic acid is liberated, which has an unpleasant taste the efficacy of the drug is unaffected. Monitoring of blood levels (not salivary levels) may enable the dose to be optimised, but only in some patients. New Adult Patients: Treatment should begin with 600mg daily in divided doses, increasing by 200mg/day at 3-day intervals until control is achieved. This is usually within the range I,000 to 1,600mg daily. If adequate control has not been achieved after two weeks, the dosage may be further increased in stages up to 2,600mg daily, though some severe cases may require doses in excess of this recommended maximum . An alternative drug may be gradually substituted , initially at low dosage, though combination therapy should be considered only if individual drugs fail to produce adequate control. Adult Patients ReceivingOther Antiepileptic Drugs: The same pattern of introduction of sodium valproate should be followed. If increased sedation is observed, the dosages of barbiturates or other drugs should be reduced as sodium valproate is increased. The dosage

Sodium Valproate

of each drug should be adjusted during the stabilisation period to give optimum control at the lowest combined dosage level. In some cases it may be possible to withdraw previous medication entirely, allowing optimum control with sodium valproate alone. Children : Dosage should be related to weight and not age. Initial dosage for children over 20kg should be 400mg daily, irrespective of weight, in divided doses with spaced increases until control is achieved. This is usually within the range 20 to 30mg/kg/day. Children weighing less than 20kg should be given 20mg/kg/day, and up to 50mg/kg/day in severe cases or in infants. A dose of 50mg/kg or more should be used only in patients in whom plasma levels are measured; plasma levels of 200pg/ml should be exceeded only with caution and with monitoring of haematological function.

References Anlezark , G .; Horton . R.W .; Meldrum . B.S. and Sawaya. M.C.B.: Anticonvulsant action of ethanolamine-Ovsulphate and di-npropylacetate and the metabolism of 'Y-aminobutyric acid (GABA) in mice with audiogenic seizures. Biochemical Pharmacology 25: 413 (1976). Aoki, Y.; Wada, T.; Takahashi. T.; Mitsutsuk a. K. and Toraiwa. S.: Clinical evaluation of dipropylacetate (DPA) . a new antiepileptic drug (In Japanese). Brain and Nerve (Tokyo) 21: 1297 (1969) . Balbi. R. and Bravaccio. F.: Studio preliminare clinico-elettroencefalografico sui trattamento dell'epilessia con il dipropilacetato di sodio (Depakin), Acta Neurologica (Napoli) 27: 479 (1972). Barnes. S.E.; Bland. D.; Cole. A.P . and Evans. A.R.: The use of sodium valproate in a case of status epilepticus. Developmental Medicine and Child Neurology 18: 236 (1976) . Barnes. S.E. and Bower. B.D.: Sodium valproate in the treatment of intractable childhood epilepsy. Developmental Medicine and Child Neurology 17: 175 (1975) . Beaumanoir. A.: Depakine in the treatment of epilepsy. Medicine et Hygiene 31: 756 (1973) . Bergamini. L.; Mutani , R.; Furlan. P.M. and Riccio. A.: Le Depakine dans Ie traitement de l'epilepsie essentielle ou idiopathique . Schweizerische Archiv fur Neurolog ie. Neurochirurgie und Psychiatrie 106: I (1970) .

119

Bonnin. J .: Premiers essais cliniques sur l'action antiepileptique du sel de de sodium de l'acide n-dipropylacetique. These de Doctorat en Medicine. Grenoble. 1964. Broglia, S.; Colombo. R.S. and Barberis. F.: L'acido di-n-propilacetico nella sindrome epilettica. La Clinica Terapeut ica 63: 165 (1972) . Burin. M.; Saks, G.; Gerber. B.I.; Meischenguiser. R. and Tallis. J.: Tratamiento de la epilepsia con dipropilacetato de magnesio . Semana Medica (Buenos Aires) 140: 678 (1972) . Campanella . G .; Cantini. A. and Damiani. A.: II sale sodico dell'acido n-dipropilacetico , 0 Depakin , nella terapie dell'epilessia. Rivista di Neurologia 42: 452 (1972). Carraz , G . and Bernard . L.: Effets hepatoprotecteurs et hepatoregenerateurs des molecules di-n-propylacetique, Therapie 22:631 (1967). Cavazzuti , G .B.: Prevention of febrile convulsions with dipropylacetate (Depakine), Epilepsia 16: 647 (1975) . Cavazzuti , G.B .: Cappella , L. and Gatti , G .: L'acido dipropilacetico (Depakin) nel trattamento della epilessia infantile. Neuropsichiatr ia Infantile 126: 650 (1971) . Chard , C.R .: A simple method for the determination of Epilim in serum . In Legg (Ed) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy, pp. 89-91 (MCS Consultants, Tunbridge Wells, 1976). Ciesieleski, L.; Maitre, M.; Cash , C. and Mandel. P.: Regional distribution and effect on cerebral mitochondrial respiration of the anticonvuls ive drug n-dipropylacetate . Biochemical Pharmacology 24: 1055 (1975). Conine, D.L.; Majors, K.R.; Lehrer , S. and Becker, B.A.: Acute toxicity of sodium 2-propylpentanoate, a compound whose toxicity decreases as animal size increases. Toxicology and Applied Pharmacology 36: 144 (1976). Costall , B.; Naylor , RJ . and Pinder, R.M.: Mechanisms of dyskinesia induction by 1-(J,4-dihydroxyphenyI) piperazine in the rat. Neuropharmacology 15: 203 (1976). Coulthard, M.G .: Sodium valproate in the treatment of intractable childhood epilepsy. Developmental Medicine and Child Neurology 17: 534 (1975) . de Biolley, D. and Sorel. L.: Premiers resultats en clinique d'un nouvel antiepileptique: Di-n-propylacetate de sodium (DPA) specialise sous la marque Depakine. Acta Neurologica et Psychiatrica Belgica 69: 909 (I 969a). de Biolley, D. and Sorel, L.: Resultats esperimentaux et cliniques d'un nouvel antiepileptique: Di-n-propylacetate de sodium (DP A) specialise sous la marque Depakine . Acta Neurologica et Psychiatrica Belgica 69: 914 (I 969b). DiGenio, R. and Colosimo, A.: Experienze cliniche su nu nuova farmaco ad azione antiepilettica : II di-n-propilacetate di sodio. La Clinica Terapeutica 63: 169 (1972) . DiGenio , R. and Colosimo , A.: Note prelimare sull'impigo clinico del dipropilacetate di sodio (Depakin) nell'epilessia infantile. Rivista di Neurologia 43: 54 (1973) .

Sodium Valproate

Dumon-Radermecker, M,: Sur les resultats obtenus avec Depakine dans les epilepsies resistantes aux aut res therapies. Acta Neurologica et Psychiatrica Belgica 69 : 90 I (I 969). Espir, M.L.E. : Benton . P.; Will . E.; Ha yes. M.1. and Walker. G .: Sodium valproate (Epilim) - some clinical and pharmacological aspects. In Legg (Edl .) Clinical and Pharmacological Aspects of Sod ium Valproate (Epilim) in the Treatment of Epilepsy, pp . 145-151 (MCS Con sultants. Tunbridge Well s. 1976) . Eymard , P.; Simiand , J .; Teoule , R.; Polverelli. M.: Werbenec, J .P. and Broil. M. Etude de la repartition et de la resorption de dipropylacetate de sodium marque au "C chez Ie rat. Journal de Pharmacologie (Paris) 2: 359 (I 97 Il. Fariello. R. and Mutani , R.: Modificazioni dell'attivita del focus epilettogeno corticomotorio da allumina indotte dal sale di sodio dell'acido n-diprop ylacetico (DPA). Acta Neurologica (Napoli) 25 : 116 (I 970) . Ferrandes. B. and Eymard , P.: Methode rapide d'anal yse quantitative du dipropylacetate de sodium dans Ie serum ou Ie plasma . Annales Pharmaceutiques Francais 31 : 279 (I 973) . For ster. C. Antikonvulsive Behandlung mit Dipropylessigsaure im Kindesalter. Miinchener Medizinische Wochenschrift 14: 399 (I 972) . Frey, H .H . and Loscher, W .: Di-n-propylacetic acid - profile of anticonvulsant activity in mice. Arzneimittel-Forschung 26: 299 (I 976) . Gadner, H. ; Bensel, I.; Grimm. B. and Riehm , H.: Beeinflussung der Hamostase durch Dipropylessigsaure. Monaschifte Kinderheilk 124: 448 (1976). Gastaut, H.: Clinical and electroencephalographical classification of epileptic seizures . Epilepsia II : 102 (1970). Godin. Y.; Heiner. L.; Mark. J . and Mandel . P.: Effects of d i-nprop ylacetate , an anticonvulsive compound. on GABA metabolism . Journal of Neurochemistry 16: 869 (1969). Grant, R.H .E . and Barot , M.H .: The use of sodium valproate (Epilim) in severely handicapped patients with epilepsy. In Legg (Ed .) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy . pp.14-21 (MCS Consultants , Tunbridge Wells , 1976). Haigh . D . and Forsythe. W .I.: The treatment of childhood epilepsy with sodium valproate. Developmental Medicine and Child Neurology 17: 743 (I975). Harvey. P.K.P.: Some aspects of the neurochemistry of Epilim . In Legg (Ed .) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy , pp .130-134 (MCS Consultants. Tunbridge Wells. 1976). Harvey. P.KP.: Bradford. H.F. and Davison. A .N.: The inhibitory effect of sodium n-dipropylacetate on the degradative enzymes of the GABA shunt. FEBS Letters 52: 521 (I975). Harwood. G . and Harvey. P.KP.: Results of a clinical trial on the use of Epilim in convulsive disorders. with special reference to its efficacy in temporal lobe attacks with focal features . In Legg (Ed .) Clinical and Pharmacological Aspects of Sodium

120

Valproate (Epilim) in the Treatment of Epilepsy. pp.40-43 (MCS Consultants. Tunbridge Wells. 1976). Hassan , M.N .: Laljee, H.e.K and Parsonage. M.1.: Experience in the treatment of resistant cases of epilepsy with sodium valproate (Epilim), In Legg (ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epileps y. pp . 23-39 (MCS Consultants. Tunbridge Wells . I 976a) . Hassan . M.N .: Laljee, H.e.K. and Parsonage. M.1.: Sodium valproate in the treatment of resistant epilepsy . Acta Neurologica Scandinavica 54: 209 (I976b). Heathfield , K.: Dunlop. D.; Karanjia, P. and Retsas , S.: The longterm results of treating thirty-six patients with intractable epilepsy with sodium valproate (Epilim) . In Legg (Ed .) Clinical and Pharmacological Aspects of Sodium Valproate (Epilirn) in the Treatment of Epilepsy. pp .165-170 (MCS Consultants. Tunbridge Wells . 1976). Huertas. X.: Contribution a I'etude d'un nouvel agent anticonvulsivant, I'acide n-dipropylacetique. dans un secteur pschiatri que. These de Doctorate en Medicine . Montpellier, 1969. Jeavons. P.M. and Clark, J .E.: Sodium valproate in the treatment of epilepsy . British Medical Journal 2: 584 (I 974) . Jeavons, P.M .; Clark . J .E. and Maheshwari, M.e. : The treatment of generalised epilepsies of childhood and adolescence with sodium valproate (Epilim) . Developmental Medicine and Child Neurology (1977b. in press). Jeavons. P.M .; Herrick. CE« Maheshwari, M.C.; and Harding. G.F.A.: Therapy and prophylaxis in photosensitive epilepsy . In Majkowski (Ed.) Post-traumatic epilepsy and pharmacological prophylaxis (Warsaw . 1977a . in press). Jeavons. P.M .; Maheshwari . M.e. ; Herrick. e.F. and Harding , G.F.A .: Epilim and photosensitivity. In Legg (Ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy . pp .56-60 (MCS Consultants. Tunbridge Wells. 1976). Jordan . B.1.; Shillingford. J .S. and Steed . KP.: Preliminary observations on the protein-binding and enzyme-inducing properties of sodium valproate (Epilim) . In Legg (Ed .) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy . pp .112-116 (MCS Consultants. Tunbridge Wells. 1976). Kerfriden, P.; Kerfriden, M. and Albe-Fessard, D.: La Depakine dans la traitement des enfants epileptiques. Observations longitudionales de 74 enfants. Nouvelle Presse Medicale 78: 1943 (I 970) . Kruse. R.: Symposium on dipropylacetic acid . Frankfurt. April 1974 (Labaz GmBH , Dusseldorf), Kugler . J .; Knorl, G . and Ernpt, J .: Die Behandlung therapieresistenter Epilepsien mit Dipropylacetat. Munchener Medizinische Wochenschrift 115: 1103 (I 973) . Kuhara, T . and Matsumoto. I.: Metabolism of branched medium chain length fatty acids . I-Omega oxidation of sodium dipropylacetate in rats . Biomed . Mass Spectrometry I: 291 (1974) .

Sodium Valproate

Kukino , K. and Matsumoto. I.: Stud ies on the absorption. excretion and biotransformation of a new anticonvulsant drug . dipropylacetic acid. Kurume Medical Journal 34: 369 (1971). Kukino, K.; Minuera, K.; Deguchi , T .; Ishii. A. and Takahari , H.: Studies on a new anticonvulsant drug. sodium dipropylacetate. Assay for metabolites and metabolic pathway. Journal of the Pharmaceutical Society of Japan 92: 896 (1972) . Kupferberg, H.J. : Lust. W .D. and Penryc J.K: Anticonvulsant activity of dipropylacetic acid (DPA) in relation to GABA and cGMP brain levels in mice. Federation Proceedings 34: 283 (1975) . Lance.T.W. and Anthony . M.: Sodium valproate and clonazeparn in the management of intractable epilepsy . Unpublished data . 1976 . [A preliminary report of this work appeared in Proceedings of the Australian Association of Neurologists 12: 55 (1975»). Lebreton . S.; Carraz, G .; Meunier. H. and Beriel, H.: Proprietes pharmacodynamiques de l'acide n-dipropylacetique. Deuxieme memoire sur les proprietes antiepileptiques. Therapie 19: 451 (1964) . Lenrnan. J .A.R .; Ferguson. LT.; Fleming . A.M.; Herzberg . L.; Robb, J .E. and Turnbull. M.J .: Sodium valproate in chorea. British Medical Journal 2: 1107 (1976) . Linnoila, M.; Viukari, M. and Hietala. 0 .: Effect of sodium valproate on tardive dyskinesia. British Journal of Psychiatry 129: 114 (1976). Littleton. J.M . and Griffiths. P.J. : The experimental approach to alcoholism. British Journal of Addiction 70: 99 (1975) . Loiseau. P.; Brachet , A. and Henry. P.: Concentration of dipropylacetate in plasma . Epilepsia 16: 609 (1975). Lust. W .D.; Kupferberg, H.J.; Passonneau. J.V. and Penry . J.K.: On the mechanism of action of sodium valproate: the relationship of GABA and cyclic GMP levels to anticonvulsant activity . In Legg (Ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy. pp.123-129 (MCS Consultants. Tunbridge Wells. 1976). Mairlot, F.: Observations cliniques sur l'effet de l'acide dipropylacetique dans les mainfestations epileptiques et caracterielles . Revue de Neuropsychiatrie Infantile 18: 269 (1970). Manhire, A.R . and Espir, M.: Treatment of status epilepticus with sodium valproate . British Medical Journal 3: 808 (1974) . Matthes . A. and Schmutterer.J .: KIinische Erfahrungen mit einen neuen Antiepileptikum: Dipropylessigsaure. Deutsche Medizinische Wochenschrift 96 : 63 (I 97 n Meijer. J .W .A . and Hessing-Brand , L.: Determination of lower fatty acids .' particularly of the antiepileptic dipropylacetic acid. in biological materials by means of micro-diffusion and gas chromatography. Clinica Chimica Acta 43: 215 (1973) . Meijer. J .W .A . and Meinardi , H.: Pharrriacokinetic studies on sodium valproate. In Legg (Ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy . pp.70-74 (MCS Consultants. Tunbridge Wells. 1976).

121

Meinardi, H.: Clinical trials of anti-epileptic drugs. Psychiatria, Neurologia, Neurochirurgia (Amsterdam) 74: 141 (1971). Meinardi, H.; Hanke . N.F.J . and van Beveren, J. : Sodium di-npropylacetate. Estimation of effective serum levels. Pharmaceutisch Weekblad 109: 45 (1974) . Meinardi, H.; van der K1eijn. E.; Meijer.J .W .A . and Rees. H.V .: Absorption and distribution of antiepileptic drugs . Epilepsia 16: 353 (1975). Meldrum. B.S.: Epilepsy and v-aminobutyric acid-mediated inhibition. International Review of Neurology 17: I (1975) . Meunier. G .; Carraz, G. ; Meunier. Y.; Eymard , P. and Airnard, M.: Proprietes pharmacodynamiques de l'acide ndipropylacetique. Therapie 18: 435 (1963) . Miribel. J. and Marinier, R.: Modifications electroen cephalographiques chez des enfants epileptiques traites par Ie Depakine. Revue Neurologie 119: 313 (1968) . Misslin, R.; Ropartz, P. and Mandel. P.: The effects of ndipropylacetate on the acquisition of conditioned behaviour with negative reinforcement in mice. Psychopharmacologia 44 : 263 (1975) . Mortel, L and Franck. G. : Etude pharmacocinetique du di-n-propylacetate de sodium sous les formes ordinaire et retard . Acta Neurologica et Psychiatrica Belgica 76: 165 (1976) . Mortier. W . and Michaelis. E.: Beeinflussbarkeit von Myok.Ionien durch Dipropylazetat (Ergen yl), Zeitschrift fur Elektroenzephalographie. Elektromyographie und verwandte Gebiete 6: 130 (1975) . Mutani . R.; Doriguzzi , T .; Fariello. R. and Furlan , P.M.: Azione antiepilettica del sale di sodio dell'acido n-dipropilacetico. Studio eperimentale sui gatto . Rivista di Patologia Nervosa e Mentale 89: 24 (1968) . Mutani, R. and Fariello . R.: Effetti dell'acido n-dipropilacetico (Depakine) sull'attivata del focus epilettogeno corticale da cobalto. Rivista di Patologia Nervosa e Mentale 90 : 40 (1969) . Noronha. M.J. and Bevan. P.L.T. : A literature review of unwanted effects with Epilim . In Legg (Ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy. pp.61-65 (MCS Consultants. Tunbridge Wells. 1976). Olive. D.; Tridon. P.; Weber. M.; Vidailhet, M. and Pierson. M.: Action du dipropylacetate de sodium sur certaines varietes d'encephalopathies epileptogenes du nourrison . Schweizerische Medizinische Wochenschrift 99:87 (1969) . Patry . G. and Naquet , R.: Action de l'acide dipropylacetique chez Ie Papio papio photcsensible, Canadian Journal of Physiology and Pharmacology 49: 568 (197 I). Pazzaglia, P. and Lugaresi, E.: \I Depakine (dipropilacetato di sodio) net trattamento delle epilessia ribelli agli altri farmaci. Studio EEG-clinico di 35 casi seguit per 3-12 mesi. Rivista di Neurologia 42: 187 (1972). Penry. J .K.; Porter. R.J .; Sato , S.; Reddenbough, J. and Dreifuss, F.E.: Effect of sodium valproate on generalised spike-wave

Sodium Valproate

paroxysms in the electroencephalogram. In Legg (Ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilirn) in the Treatment of Epilepsy. pp.158-163 (MCS Consultants. Tunbridge Wells . 1976). Pillen, E.: Case report on file with Labaz, Paris (1973). Price . D.1.E.: The advantage s of sodium valproate in neurosurgi cal pract ice. In L egg (Ed .l Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy. p.44-48 (MCS Consultants. Tunbridge Wells. 1976). Reekers-Ketting. J.1.; van der Kleijn, E.; Leliveld, B.A.; Schobben, A .F.A.M. and Vree, T.M. : Pharmacokinetics of di-propyl acetate (Depakine) in different dosage forms in man and rhesu s monkey. Pharmaceutisch Weekblad 110: 1232 (1975). Richardson . S.G.N .; Fletcher. DJ. and Jeavons . P.M.: Sodium valproate and platelet function . In Legg (Ed .) Clin ical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy , pp.119-122 (MCS Con sultants. Tunbridge Wells. I976a) . Richardson. S.G.N.; Fletcher , DJ.; Jeavons, P.M. and Stuart. J .: Sodium valproate and platelet function. British Medical Journal I: 221 (I 976b) . Richens, A . and Ahmad. S.: Controlled trial of sodium valproate in severe epilepsy . British Medical Journal 4: 255 (1975) . Richens , A.; Scoular, LT.; Ahmad . S. and Jordan. B.1.: Pharmacokinetics and efficacy of Epilim in patients receiving longterm therapy with other antiepileptic drugs. In Legg (Ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epiliml in the Treatment of Epilepsy . pp.78-88 (MCS Consultants. Tunbridge Wells. 1976). Ritz. A. and Jacobi. G .: Depakine in the treatment of epilepsy in children. Fortschritte der Medizin 91: 590 (1973) . Sawaya. M.C.B.; Horton. R.W . and Meldrum . B.S.: Effects of anticonvulsant drugs on the cerebral enzymes metabolising GABA . Epilepsia 16: 649 (1975) . Schafer , J. and Kettner , J .: Erfahrungen mit Dipropylacetat in der Behandlung therapieresistenter erwachsener Anfallskranker. Medizinische Welt 25: 561 (1974) . Schlack, H .G .: Ergenyl in the treatment of epilepsy. Clinical findings in children with various forms of epilepsy. Therapiewoche 24: 39 (1973) . Schmidt, D .: Salivary concentrations of antiepileptic drugs . Lancet 2: 639 (1976) . Schneble, H.: Behandlung atypischer Absencen mit einer Suecinimid-Dipropylacetat Kombination. Deutsche Medizinische Wochenschrift 100: 1564 (1975) . Sehobben. F. and van der Kleijn, E.: Determination of sodium din-propylacetate in plasma by gas-liquid chromatography. Pharmaceutisch Weekblad 109: 30 (l974a). Schobben, F. and van der Kleijn, E.: Pharmacokinetics of distribution and elimination of sodium di-n-propylacetate in mouse and dog. Pharmaceutisch Weekblad 109: 33 (l974b). Schobben. F.; van der Kleiin, E.; Vree , T .B.and Guelen , P.1.M.: Pharmacokinetics of 2-n-propyl pentanoate in man and

122

laboratory animals . Pharmaceutisch Weekblad , in press (1977) . Sehobben. F.; van der Kleijn, E. and Gabreels. F.1.M. : Pharmacokinetics of di-n-propylacetate in epileptic patients. European Journal of Clinical Pharmacology 8: 97 (1975) . Scollo-Lavizzari, G . and Corbat. F.: A clinical note on a new antiepileptic 'Depakine' . European Neurology 4: 312 (1970) . Scott . D.F .; Boxer. C.M . and Herzberg, J .L.: A study of the hypnotic effects of Epilim and its possible interaction with phenobarbitone. In Legg (ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy . pp.155-157 (MCS Consultants. Tunbridge Wells. 1976). Shoulson, I.; Kartzinel , R. and Chase . T.N .: Huntington's disease: Treatment with diprop ylacetic acid and gamrna-aminobutyric acid. Neurolog y 26: 61 (1976) . Shuto , K and Nishigaki, T. : Pharmacological studies on sodium diprop ylacetate. Anticonvulsant activities and general pharmacologic actions. Pharmacometrics 4: 937 (1970) . Sillanpaa , M. and Donner. M.: Experiences on the use of dipropylacetate in the treatment of childhood epilepsy . Acta Paediatrica Scandinavica 65: 209 (1976). Simler , S.; Ciesieleski, L.; Maitre . M.; Randrianarisoa, H. and Mandel . P.: Effect of sodium di-n-propylacetate on aud iogenic seizures and brain garnma-aminobutyric acid level. Biochemical Pharmacology 22: 170 I (1973) . Simon . D. and Penry . J.K: Sodium di-n-propylacetate (DPA) in the treatment of epilepsy . A review. Epilepsia 16: 549 (1975) . Sonnen . A.E.H .; Zelvelder, W .H . and Bruens, J.H .: A double blind study of the influence of dipropylacetate on behaviour. Acta Neurologica Scandinavica Suppl . 60 : 43 (1975) . Sutor, A.H . and Jesdinsky-Buscher, c.: Coagulation changes caused by dipropylacetic acid. Medizinische Welt 25: 447 (1974) . Sutor. A.H . and Jesdinsky-Buscher, c.: Alteration of haemostasis during anticonvulsive therapy with dipropylacetate (DP A). further observations. Fortschritte der Medizin 94 : 41 I (1976). Suzuki. M.; Maruyama. H.; Ishibashi. Y.; Ogawa . S.; Seki, T. ; Hoshino, M.; Maekawa, K ; Yo. T. and Sato, Y.: A doubleblind comparative trial of sodium diprop ylacetate and ethosuximide in epilepsy in children. with special empha sis on pure petit mal seizures (In Japanese) . Medical Progress (Japan) 82: 470 (1972). Swinyard, E.A.: The pharmacology of dipropylacetic acid sodium with special emphasis on its effects on the central nervous system . University of Utah College of Pharmacy. Salt Lake City. Utah.1964 . pp .I-25. Swinyard, E.A .: Laboratory evaluation of antiepileptic drug s. Review of laboratory methods . Epilepsia 10: 107 (1969 ). Swinyard. E.A .: Assay of antiepileptic drug activit y in experimental animals ; standard tests . In Mercier (Ed.) Anticonvulsant

Sodium Valproate

drugs (International Encyclopedia of Pharmacology and Therapeutics. Section 19. Volume O. pp.47-122 (Pergamon Press. London. 1973). Timpany , M.M.: The use of Epilim at school. In Legg, (Ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy. pp.152-154 (MCS Consultants. Tunbridge Wells. 1976). Tomlinson . E.B.: Progressive myoclonic epilepsy. The response to sodium di-n-propylacetate . Proceedings of the Australian Association of Neurologists II : 203 (1974) . Vajda. F.; Morris. P.; Drummer. O. and Bladin, P.: Studies on sodium valproate - a new anticonvulsan t. In Legg (Ed). Clinical and Pharmacological Aspects of Sodium Valproate (Epilirnl in the Treatment of Epilepsy. pp.92-100 (MCS Consultants. Tunbridge Wells. 1976). Vakil. S.D.; Critchley. E.M.R.; Philips. r.c., Fahim , Y.; Haydock . e.; Cocks. A. and Dyer. T. : The effect of sodium valproate (Epilim) on phenytoin and phenobarbitone blood levels. In Legg (Ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy. pp.75-77 (MCS Consultants. Tunbridge Wells. 1976). Van der Kleijn , E.; Guelen, PJ.M.; van Wijk. e.G.W.M.; Baars. I. and Vree, T.B.: Clinical pharmacokinetics of benzodiazepines, barbiturates and short chain fatty acids. In Levy (Ed.) Clinical Pharmacokinetics: A Symposium . p.79 (American Pharmaceutical Association. Washington D.e. . 1974). Van Duijn , H. and Beckmann. M.K.F.: Dipropylacetic acid (Depakine) in experimental epilepsy in the alert cat. Epilepsia 16: 83 (1975). Vidart , L.; Amstutz. e. ; Gasteau , E.; Geier . S.; Vidart , J.N. and Minh. T .V.: Resultats obtenus par Ie DPA a l'Hopital de Jour pour Epileptiques de Creteil. Annales de Medicine Psychologique I: 684 (1969). Vdlzke, E. and Doose. H.: Dipropylacetate (Depakine, Ergenyl) in the treatment of Epilepsy. Epilepsia 14: 185 (1973).

123

Von Voss. H.; Petrich. C.; Karch. D.; Schulz. H.V . and Gobel. U.: Sodium valproate and platelet function. British Medical Journal 3: 179 (( 976). Voskuyl , R.A.; Ter Keurs, H.E.DJ. and Meinardi, H.: Actions and interactions of dipropylacetate and penicillin on evoked potentials of excised prepiriforrn cortex of guinea pig. Epilepsia 16: 583 ((975). Vree , T.B. and van der Kleijn, E.: Interaction of 2-propylpentanoate (Depakine) with ethanol. Pharmaceutisch Weekblad (in press. 1977). Whittle . B.A.: Pre-clinical teratological studies on sodium valproate (Epilim) and other anticonvulsants . In Legg (Ed.) Clinical and Pharmacological Aspects of Sodium Valproate (Epilim) in the Treatment of Epilepsy. pp.105-1 10 (MCS Consultants. Tunbridge Wells. 1976). Wiechert . P.: Prufung von di-n-propylazetat auf antikonvulsive Wirkung im Tierexperiment. Zentralblatt fur Pharmazie III : 899 (1972). Windorfer, A.; Sauer. W . and Gadeke, R.: Elevation of diphenylhydantoin and primidone serum concentration by addition of dipropylacetate. a new anticonvulsant drug . Acta Paediatrica Scandinavica 64: 771 (1975). Winfield. D.A.; Benton. P.; Espir, M.L.E . and Arthur. L.J.H.: Sodium valproate and thrombocytopenia. British Medical Journal 2: 981 (1976). Woodbury. D.M.: Applications to drug evaluation. In Purpura, Tower. Woodbury and Walter (Eds). Experimental Models of Epilepsy. pp.557 -583 (Raven Press. New York. 1972).

Authors' address: R .M. Pinder. R.N. Brogden, T.M. Speight and G.S. Avery. Australasian Drug Information Services. P.O. Box 34-030 . Birkenhead, Auckland 10 (New Zealand).