Journal of the Royal Society of Medicine Volume 84 March 1991
Modern treatment strategies for patients with epilepsy:
a
159
review
J S Duncan DM MRCP Institute of Neurology, National Hospital for Neurology and Neurosurgery, Queen Square, London WCIN 3BG and Chalfont Centre for Epilepsy Keywords: epilepsy; antiepileptic drugs
Summary Of patients who develop epilepsy 70-80% will become seizure free, the remaining 20% are the most difficult to treat satisfactorily. Eighty per cent of patients are best treated with a single drug. Stepwise treatment plans for the treatment of newly diagnosed patients and for the evaluation of patients with chronic epilepsy are suggested. There is a lack of consensus regarding the rate at which to taper antiepileptic drugs being discontinued in patients with active epilepsy. There are arguments for making drug changes rapidly; these arguments and strategies for managing the withdrawal of individual drugs are presented.
Introduction About one in 40 people will have two or more nonfebrile seizures at some time and one in 200 people endure chronic epilepsy'. Despite the availability of potent antiepileptic drugs, many of the latter are likely to remain intractable and present a challenge to the physician.
Antiepileptic drug prescribing strategy About 70-80% of patients developing epilepsy may expect to become seizure free with optimal antiepileptic drug therapy. Approximately 80% will be best controlled with a single drug, and 10-15% with a combination of two agents. Poor compliance, drug interactions and long-term toxicity are all more likely to occur if more than one drag is prescribed2-4. The goal of therapy should be complete seizure control with a single drug taken once or twice a day, without sideeffects. Precise clasification of the types of seizure and of the epilepsy, from clinical and EEG data, and careful recording of seizures and side-effects is essential if rational management decisions are to be made. If control is difficult to achieve, the maximally tolerated dose of each drug that is used should be explored, but a balance needs to be struck between side-effects and control of seizures, and drugs that do not contribute to seizure control should be discontinued. In patients in whom treatment appears to be ineffective, the diagnosis of epilepsy and compliance with therapy should be reviewed. Consideration should also be given to the presence of a progressive cerebral disorder, such as a tumour or metabolic defect, and the patient investigated accordingly. A step-wise treatment strategy for naive patients There follows a programme of treatment that may be followed from the time of diagnosis, and continued as far as necessary. The issues of when to start and stop drug treatment have been reviewed recently6 and are not considered further.
Table 1. Antiepileptic drugs for different types of seizures Seizure type
Generalized Tonic-clonic, tonic & clonic Simple absences
Complex absences 1 ) Atonic Infantile spasms Myoclonic
First-line drugs
Phenytoin Carbamazepine Sodium valproate Ethosuximide Sodium valproate f Sodium valproate Clonazepam/Clobazam ACTH/steroids
Clonazepam Sodium valproate Clonazepam
Partial Phenytoin Simple, complex & secondarily generalized Carbamazepine Sodium valproate Second-line drugs: Phenobarbitone Clonazepam Clobazam Primnidone Acetazolamide Vigabatrin In each section drugs are listed in order of appearance
(1) Any precipitating factors for seizures, such as fever in young children, excessive fatigue, alcohol and drug abuse, and photo-sensitivity should be identified and the patient and relatives counselled about their avoidance.
(2) The reasons for, goals and limitations of antiepileptic drug treatment, the likely duration of therapy and the need for regular tablet taking should be explained. (3) A patient should be commenced on a small dose of one of the first line antiepileptic drugs for their type of seizure (Table 1) and dose increments made if seizures continue and side-effects do not occur, with increments being guided by the measurement of serum drug concentrations. When employed as a guide to dosing, measurement of serum concentrations of phenytoin are particularly useful. Assay of concentrations of carbampine and phenobarbitone are moderately helpful, and valproate levels are of little utility in this regard. Zealous adherence to quoted therapeutic ranges of serum antiepileptic drug concentrations is inappropriate. These data should always be subordinate to the clinical picture of whether the patient continues to have seizures and/or doserelated side-effects from antiepileptic drugs, or not. The barbiturates, phenobarbitone and primidone, are now regarded as second line drugs because ofthe
0141-0768/91/ 030159-04/$02.00/0 © 1991 The Royal
Society of Medicine
160
Journal of the Royal Society of Medicine Volume 84 March 1991
greater risk of chronic adverse side-effects from these drugs, compared to other agents. For similar reasons, phenytoin is a less attractive initial choice than is carbamazepine or sodium valproate. Phenytoin, sodium valproate, phenobarbitone, clobazam and clonazepam may generally be given once or, at most, twice daily. Acetazolamide, ethosuximide and vigabatrin may be given twice daily. Few adult patients can tolerate more than 400 mg of carbamazepine as a single dose, without acute dose-related side-effects occurring in the subsequent 1-2 h. For patients requiring large doses, the slow release preparation of carbamazepine is a useful alternative, allowing twice daily ingestion, and avoiding high peak serum concentrations. (4) If seizures continue despite a maimally tolerated dose of a first line antiepileptic drug, the diagnosis of epilepsy and its putative aetiology should be reconsidered, and compliance checked with counselling, tablet counts and measurement of the serum drug concentration. (5) Another first line drug should then be commenced, built up to an optimal dose and the initial agent then withdrawn. The ideal rate of drug withdrawal in this situation is controversial; phenytoin and sodium valproate probably may be safely withdrawn over a few days, but carbamazepine, barbiturate and benzodiazepine withdrawal should probably be over a period of weeks6. (6) The dose of the second drug, taken alone, should then be adjusted to optimum, as was the initial agent. (7) If seizures continue despite a maimally tolerated dose of all the individual first line drugs the next step is to try a combination of two first line drugs for that seizure type (eg ethosuximide and sodium valproate for generalized absences or phenytoin and carbamazepine for partial seizures). The chances of duotherapy controlling seizures when monotherapy has been unsuccessful is of the order of 10_15%2,3,7.
(8) Should a combination of two first line agents be unhelpful, the drug which appears to have the most effect, and least side-effects, should be continued and the other antiepileptic drug replaced by a second line drug (Table 1). The list of second-line drugs is not exhaustive and other agents may also be considered (eg carbamazepine for intractable atonic seizures).
(9) If the second line drug is effective, withdrawal of the initial agent should be considered. If prescription of a second line agent is unhelpful, it should not be continued. (10) The use of a novel antiepileptic drug may be considered. As a general rule, the use of such drugs should only be as part of a formal organized trial, with very accurate documentation of seizures and side-effects. (11) The above scheme will generally take a number of months or even years to work through. If satisfactory control cannot be obtained with drugs and the patient has partial seizures, consideration should be given to surgical treatment of the epilepsy. Such evaluations should be carried out in a specialized
unit. The most commonly performed operation is a temporal lobectomy and, amongst the minority of patients that are suitable for surgery, 50% may expect to become seizure free and a further 25% to have a marked improvement in seizure control9'2.
Antiepileptic drug treatment in patients with chronic epilepsy Patients who suffer intractable, chronic epilepsy should generally be treated with a single, or at most two, antiepileptic drugs that have been shown to have some efficacy, at doses that do not give rise to intolerable side-effects. This is a difficult area and referral for a specialist evaluation is appropriate. The general principles are as follows: (1) The diagnosis and history of the epilepsy should be reviewed. This will usually necessitate obtaining and reviewing old records, in addition to interviewing the patient and a witness to the seizures. The types of seizure and the epilepsy should then be classified on the basis of clinical features and the EEG. In addition, two specific questions should be addressed. Is there evidence for pseudoseizures? Is there an identifiable aetiology? In many series of patients with supposed chronic epilepsy, 10-15% had pseudoseizures. The aetiology of a seizure disorder may be identified in 40-50% of patients. Imaging of the brain with CT or MRI scanning is appropriate at this time, MRI being more sensitive, looking for an underlying structural lesion'3 4.
(2) Check serum concentrations of antiepileptic drugs and question compliance. (3) Review past and present antiepileptic drug treatment for efficacy and side-effects. Has the patient had a good trial with a maximally tolerated dose of all the major antiepileptic drugs? (4) Select the antiepileptic drug that is most likely to be efficacious and with the least side-effects. This information needs to be determined from a detailed drug and seizure history. Adjust the dose ofthis drug to the optimum. The drug to be retained will usually be chosen on the grounds that it has not been used to its full potential, and/or that it appears to have had a definite beneficial effect. In some cases, for example, patients with primary generalized epilepsy, there is a clear indication to use a particular agent. Other antiepileptic drugs may then be withdrawn if their prescription has not aided seizure control, and they may be giving rise to adverse effects.
(5) Attempt to reduce and discontinue the other antiepileptic drugs, recognizing that seizures may worsen at this time. Frequently, however, reduction of the number of antiepileptic drugs results in the patient feeling better and improved seizure control4"5'i8. Management of the withdrawal of an antiepileptic drug is detailed below. (6) If seizures remain uncontrolled, consider other first-line drugs and second-line agents, and surgical treatment, as discussed for the treatment of naive
patients. Withdrawal of antiepileptic drugs in patients with active epilepsy Once it has been decided to discontinue a particular antiepileptic drug, and the doses of remaining
Journal of the Royal Society of Medicine Volume 84 March 1991
antiepileptic drugs have been optimized, the question of the rate of tapering arises. A very wide range of rates of drug reduction has been used in studies of polytherapy reduction4" 531, with duration ranging from 7 days to 12 months. This impression of a lack of consensus was confirmed by a recent survey of the practice of consultant neurologists32. There are several arguments in favour of making rapid antiepileptic drug reductions rather than slow changes, provided that the former does not carry an increased risk of an exacerbation of seizures. These arguments may be summarized as: (1) Patients may be supervised very closely over a short period of time. (2) The strategy is more likely to be followed.
(3) The effect of the drug change on seizures, medication side-effects and serum concentrations of concomitant antiepileptic drugs may be rapidly determined. (4) Potentially useful drug treatments can be explored quickly, and the patient established on optimal therapy as soon as is possible.
In a prospective, double-blind study of the withdrawal of phenytoin, carbamazepine and sodium valproate from the polytherapy of patients with intractable epilepsy, phenytoin was withdrawn in 50 mg decrements every 2 or 5 days, carbamazepine was discontinued at 200 mg every 2 or 5 days, and sodium valproate at 200 mg every 2 or 4 days6. It was concluded that:
(1) The rapid tapering of phenytoin and valproate appeared to be generally problem free in these patients with active epilepsy, who were under close supervision. The optimal rates of carbamazepine reduction remained uncertain. Withdrawal of carbamazepine was often associated with an increase in seizures and its discontinuation should be approached with caution, even if its use does not appear to have helped seizure control. It should not be taken that withdrawal of phenytoin and valproate will always be free of problems as there are clearly many patients who would have a marked increase in seizure should these antiepileptic drugs be stopped, however slowly. (2) Whilst the acute effects of a change of drug may be rapidly determined, the longer term consequences need more prolonged observation, particularly if seizures occur only infrequently.
(3) If a marked increase in seizures occurs on reduction and withdrawal of an antiepileptic drug, and serum concentrations of concomitant drugs have not fallen, the increase in seizures is likely to continue until the withdrawn drug is recommenced. In the event of a serious increase in seizures, diazepam should be given and the withdrawn drug reintroduced. This action would be expected to restore the seizure frequency to the baseline level, immediately. (4) In cases of failed antiepileptic drug withdrawal, the withdrawn drug may be recommenced at the dose that was being taken prior to the increase in seizures, and then increased further as necessary. In the case
of reintroduction of phenytoin or phenobarbitone, a loading dose might be necessary. (5) A marked increase in seizures on discontinuation of a particular antiepileptic drug suggests that that drug is efficacious and it may be that a higher dose, than that which was being given at the baseline, will improve seizure control. On the other hand, a lower dose may be equally effective.
(6) If a drug change does result in an increase in seizures this can be managed safely, and the clinician and the patient can be reasonably confident that, with appropriate treatment, there will be no long lasting deterioration of control of the seizure disorder. Holistic issues In addition to the pharmacological aspects, it is important to consider the implications and consequences of intractable epilepsy, which are often more devastating than the seizures themselves. Not least, people with intractable epilepsy benefit enormously from somebody taking a personal interest in them and their problems, and from advice on practical issues, such as safe cooking with a microwave oven, and a discussion of reasonable expectations and limitations, with regard to prognosis and the prospects of independent living, a social and leisure life and of employment. Conclusion Patients whose epilepsy is intractable are the most difficult to treat. Careful attention may improve these patients' lot, with improved control of seizures and reduction of the toxic effects of antiepileptic drug therapy. Acknowledgments: I am grateful to colleagues who have collaborated in investigations and to the Brain Research Trust, Ciba-Geigy Pharmaceuticals and the Merrell Dow Research Institute for support. References 1 Duncan JS, Shorvon SD. Prognosis of epilepsy. Br J Hosp Med 1986;35:254-9 2 Reynolds EH, Shorvon SD. Monotherapy or polytherapy for epilepsy? Epilepsia 1981;22:1-10 3 Schmidt D. Two antiepileptic drugs for intractable epilepsy with complex partial seizures. J Neurol Neurosurg Psychiatry 1982;45:1119-24 4 Thompson PJ, Trimble MR. Anticonvulsant drugs and cognitive functions. Epilepsia 1982;23:531-44 5 Chadwick D, Reynolds EH. When do epileptic patients need treatment? Starting and stopping medication. BMJ 1985;290:1885-8 6 Duncan JS, Shorvon SD, Trimble MR. The withdrawal of phenytoin, carbamazepine and sodium valproate in patients with active epilepsy. Epilepsia 1990;31:324-33 7 Shorvon SD, Reynolds EH. Unnecessary polypharmacy for epilepsy. BMJ 1977;1:1635-7 8 Mattson RH, Cramer JA, Collins JF, et aL Comparison of carbamazepine, phenobarbital, phenytoin and primidone in partial and secondarily generalized tonic clonic seizures. N Engi J Med 1985;313:145-52 9 Falconer MA, Taylor DC. Surgical treatment of drugresistant epilepsy due to mesial temporal sclerosis. Arch Neurol 1968;19:353-61 10 Jensen I. Temporal lobe surgery around the world. Acta Neurol Scand 1975;52:354-73 11 Duncan JS, Sagar HJ. Seizure characteristics, pathology, and outcome after temporal lobectomy. Neurology 1987;37:405-9
161
162
Journal of the Royal Society of Medicine Volume 84- March 1991
12 Polkey CE. Neurosurgery. In: Laidlaw J, Richens A, Oxley J, eds. A textbook of epilepsy, 3rd edn. Edinburgh: Churchill Livingstone, 1988:484-510 13 Lesser RP, Modic MT, Weinstein-MA, et al. Magnetic resonance imaging (1.5 Tesla) in patients with intractable focal seizures. Arch Neurol 1986;43:367-71 14 Franceschi M, Triulzi F, Ferini-Strambi L. Focal cerebral lesions found by magnetic resonance imaging in cryptogenic nonrefractory temporal lobe epilepsy patients. Epilepsia 1989;30:540-6 15 Shorvon SD, Reynolds EH. Reduction in polypharmacy for epilepsy. BMJ 1979;2:1023-5 16 Theodore WH, Porter RJ, Removal of sedative-hypnotic antiepileptics from the regimens. of patients with intractable epilepsy. Ann Neurol 1983;13:320-4 17 Callaghan N, O'Dwyer R, Keating J. Unnecessary polypharmacy in patients with frequent seizures. Acta Neurol Scand 1984;69:15-19 18 Albright P, Bruni J. Reduction of polypharmacy in epileptic patients. Arch Neurol 1985;42:797-9 19 Jeavons PM, Clark JE, Maheshwai MC. Treatment of generalized epilepsies of childhood and adolesoence with sodium valproate ('Epilim'). Deu Med Child Neurol 1977;19:9-25 20 Milano Collaborative Group for Studies on Epilepsy. Long term intensive monitoring in the difficult patient. Preliminary results of 16 months of observationsusefulness and limitations. In: Gardner-Thorpe C, Janz D, Meinardi H, Pippenger CE, eds. Antiepileptic drug monitoring. Tunbridge Wells: Pitman, 1977: 197-213 21 Callaghan N, O'Callaghan M, Duggan B, Feely M. Carbamazepine as a single drug in the treatment of epilepsy. J Neurol Neurosurg Psychiatry 1978;41:907-10
22 Schobben AFAM. Pharmacokinetics and therapeutics in epilepsy. Nijmegan: Stichting Studentenpers, 1979: 251-8 23, Maheshwari MC, Padmrini R. Role of carbamazepine in reducing polypharmacy in epilepsy. Acta Neurol Scand 1981;64:22-8 24 Gannaway DJ, Mawer CE. Transfer from multiple to single antiepileptic drug therapy. Lancet 1981;i:217 25 Covanis A, Gupta AK, Jeavons PM. Sodium vaiproate: monotherapy and polytherapy. Epilepsia 1982;23:693-720 26 Fischbacher E. Effect of reduction of anticonvulsants on wellbeing. BMJ 1982;28&.423-4 27 Bennett HS, Dunlop T, Ziring P. Reduction of polypharmacy for epilepsy in an institution for the retarded. Dev Med Child Neurol 1983;25:735-7 28 Schmidt D. Reduction of two drug therapy in intractable epilepsy. Epilepsia 1983;24:368-76 29 Roman EJ, Lambert JB, Buchanan N, Barrah N. -Rationalization of therapy in severe epilepsy' Aust N Z J Med 1983;13:601-4 30 Alvarez N, Hazlett J. Seizure management with minimal medications in institutionalized mentally retarded epileptics:A prospective study. First report after 4% years of follow-up. Clinical Electroencephalography 1983;14:164-73 31 Schmidt D, Richter K. Alternative single anticonvulsant drug therapy for refractory epilepsy. Arch Neurol 1986;19:85-7 32 Duncan JS, Shorvon SD. Rates of antiepileptic drug reduction in active epilepsy-current practice. Epilepsy Res 1987;1:357-64
(Accepted 28 September 1990)
Modern treatment strategies for patients with epilepsy:
a
159
review
J S Duncan DM MRCP Institute of Neurology, National Hospital for Neurology and Neurosurgery, Queen Square, London WCIN 3BG and Chalfont Centre for Epilepsy Keywords: epilepsy; antiepileptic drugs
Summary Of patients who develop epilepsy 70-80% will become seizure free, the remaining 20% are the most difficult to treat satisfactorily. Eighty per cent of patients are best treated with a single drug. Stepwise treatment plans for the treatment of newly diagnosed patients and for the evaluation of patients with chronic epilepsy are suggested. There is a lack of consensus regarding the rate at which to taper antiepileptic drugs being discontinued in patients with active epilepsy. There are arguments for making drug changes rapidly; these arguments and strategies for managing the withdrawal of individual drugs are presented.
Introduction About one in 40 people will have two or more nonfebrile seizures at some time and one in 200 people endure chronic epilepsy'. Despite the availability of potent antiepileptic drugs, many of the latter are likely to remain intractable and present a challenge to the physician.
Antiepileptic drug prescribing strategy About 70-80% of patients developing epilepsy may expect to become seizure free with optimal antiepileptic drug therapy. Approximately 80% will be best controlled with a single drug, and 10-15% with a combination of two agents. Poor compliance, drug interactions and long-term toxicity are all more likely to occur if more than one drag is prescribed2-4. The goal of therapy should be complete seizure control with a single drug taken once or twice a day, without sideeffects. Precise clasification of the types of seizure and of the epilepsy, from clinical and EEG data, and careful recording of seizures and side-effects is essential if rational management decisions are to be made. If control is difficult to achieve, the maximally tolerated dose of each drug that is used should be explored, but a balance needs to be struck between side-effects and control of seizures, and drugs that do not contribute to seizure control should be discontinued. In patients in whom treatment appears to be ineffective, the diagnosis of epilepsy and compliance with therapy should be reviewed. Consideration should also be given to the presence of a progressive cerebral disorder, such as a tumour or metabolic defect, and the patient investigated accordingly. A step-wise treatment strategy for naive patients There follows a programme of treatment that may be followed from the time of diagnosis, and continued as far as necessary. The issues of when to start and stop drug treatment have been reviewed recently6 and are not considered further.
Table 1. Antiepileptic drugs for different types of seizures Seizure type
Generalized Tonic-clonic, tonic & clonic Simple absences
Complex absences 1 ) Atonic Infantile spasms Myoclonic
First-line drugs
Phenytoin Carbamazepine Sodium valproate Ethosuximide Sodium valproate f Sodium valproate Clonazepam/Clobazam ACTH/steroids
Clonazepam Sodium valproate Clonazepam
Partial Phenytoin Simple, complex & secondarily generalized Carbamazepine Sodium valproate Second-line drugs: Phenobarbitone Clonazepam Clobazam Primnidone Acetazolamide Vigabatrin In each section drugs are listed in order of appearance
(1) Any precipitating factors for seizures, such as fever in young children, excessive fatigue, alcohol and drug abuse, and photo-sensitivity should be identified and the patient and relatives counselled about their avoidance.
(2) The reasons for, goals and limitations of antiepileptic drug treatment, the likely duration of therapy and the need for regular tablet taking should be explained. (3) A patient should be commenced on a small dose of one of the first line antiepileptic drugs for their type of seizure (Table 1) and dose increments made if seizures continue and side-effects do not occur, with increments being guided by the measurement of serum drug concentrations. When employed as a guide to dosing, measurement of serum concentrations of phenytoin are particularly useful. Assay of concentrations of carbampine and phenobarbitone are moderately helpful, and valproate levels are of little utility in this regard. Zealous adherence to quoted therapeutic ranges of serum antiepileptic drug concentrations is inappropriate. These data should always be subordinate to the clinical picture of whether the patient continues to have seizures and/or doserelated side-effects from antiepileptic drugs, or not. The barbiturates, phenobarbitone and primidone, are now regarded as second line drugs because ofthe
0141-0768/91/ 030159-04/$02.00/0 © 1991 The Royal
Society of Medicine
160
Journal of the Royal Society of Medicine Volume 84 March 1991
greater risk of chronic adverse side-effects from these drugs, compared to other agents. For similar reasons, phenytoin is a less attractive initial choice than is carbamazepine or sodium valproate. Phenytoin, sodium valproate, phenobarbitone, clobazam and clonazepam may generally be given once or, at most, twice daily. Acetazolamide, ethosuximide and vigabatrin may be given twice daily. Few adult patients can tolerate more than 400 mg of carbamazepine as a single dose, without acute dose-related side-effects occurring in the subsequent 1-2 h. For patients requiring large doses, the slow release preparation of carbamazepine is a useful alternative, allowing twice daily ingestion, and avoiding high peak serum concentrations. (4) If seizures continue despite a maimally tolerated dose of a first line antiepileptic drug, the diagnosis of epilepsy and its putative aetiology should be reconsidered, and compliance checked with counselling, tablet counts and measurement of the serum drug concentration. (5) Another first line drug should then be commenced, built up to an optimal dose and the initial agent then withdrawn. The ideal rate of drug withdrawal in this situation is controversial; phenytoin and sodium valproate probably may be safely withdrawn over a few days, but carbamazepine, barbiturate and benzodiazepine withdrawal should probably be over a period of weeks6. (6) The dose of the second drug, taken alone, should then be adjusted to optimum, as was the initial agent. (7) If seizures continue despite a maimally tolerated dose of all the individual first line drugs the next step is to try a combination of two first line drugs for that seizure type (eg ethosuximide and sodium valproate for generalized absences or phenytoin and carbamazepine for partial seizures). The chances of duotherapy controlling seizures when monotherapy has been unsuccessful is of the order of 10_15%2,3,7.
(8) Should a combination of two first line agents be unhelpful, the drug which appears to have the most effect, and least side-effects, should be continued and the other antiepileptic drug replaced by a second line drug (Table 1). The list of second-line drugs is not exhaustive and other agents may also be considered (eg carbamazepine for intractable atonic seizures).
(9) If the second line drug is effective, withdrawal of the initial agent should be considered. If prescription of a second line agent is unhelpful, it should not be continued. (10) The use of a novel antiepileptic drug may be considered. As a general rule, the use of such drugs should only be as part of a formal organized trial, with very accurate documentation of seizures and side-effects. (11) The above scheme will generally take a number of months or even years to work through. If satisfactory control cannot be obtained with drugs and the patient has partial seizures, consideration should be given to surgical treatment of the epilepsy. Such evaluations should be carried out in a specialized
unit. The most commonly performed operation is a temporal lobectomy and, amongst the minority of patients that are suitable for surgery, 50% may expect to become seizure free and a further 25% to have a marked improvement in seizure control9'2.
Antiepileptic drug treatment in patients with chronic epilepsy Patients who suffer intractable, chronic epilepsy should generally be treated with a single, or at most two, antiepileptic drugs that have been shown to have some efficacy, at doses that do not give rise to intolerable side-effects. This is a difficult area and referral for a specialist evaluation is appropriate. The general principles are as follows: (1) The diagnosis and history of the epilepsy should be reviewed. This will usually necessitate obtaining and reviewing old records, in addition to interviewing the patient and a witness to the seizures. The types of seizure and the epilepsy should then be classified on the basis of clinical features and the EEG. In addition, two specific questions should be addressed. Is there evidence for pseudoseizures? Is there an identifiable aetiology? In many series of patients with supposed chronic epilepsy, 10-15% had pseudoseizures. The aetiology of a seizure disorder may be identified in 40-50% of patients. Imaging of the brain with CT or MRI scanning is appropriate at this time, MRI being more sensitive, looking for an underlying structural lesion'3 4.
(2) Check serum concentrations of antiepileptic drugs and question compliance. (3) Review past and present antiepileptic drug treatment for efficacy and side-effects. Has the patient had a good trial with a maximally tolerated dose of all the major antiepileptic drugs? (4) Select the antiepileptic drug that is most likely to be efficacious and with the least side-effects. This information needs to be determined from a detailed drug and seizure history. Adjust the dose ofthis drug to the optimum. The drug to be retained will usually be chosen on the grounds that it has not been used to its full potential, and/or that it appears to have had a definite beneficial effect. In some cases, for example, patients with primary generalized epilepsy, there is a clear indication to use a particular agent. Other antiepileptic drugs may then be withdrawn if their prescription has not aided seizure control, and they may be giving rise to adverse effects.
(5) Attempt to reduce and discontinue the other antiepileptic drugs, recognizing that seizures may worsen at this time. Frequently, however, reduction of the number of antiepileptic drugs results in the patient feeling better and improved seizure control4"5'i8. Management of the withdrawal of an antiepileptic drug is detailed below. (6) If seizures remain uncontrolled, consider other first-line drugs and second-line agents, and surgical treatment, as discussed for the treatment of naive
patients. Withdrawal of antiepileptic drugs in patients with active epilepsy Once it has been decided to discontinue a particular antiepileptic drug, and the doses of remaining
Journal of the Royal Society of Medicine Volume 84 March 1991
antiepileptic drugs have been optimized, the question of the rate of tapering arises. A very wide range of rates of drug reduction has been used in studies of polytherapy reduction4" 531, with duration ranging from 7 days to 12 months. This impression of a lack of consensus was confirmed by a recent survey of the practice of consultant neurologists32. There are several arguments in favour of making rapid antiepileptic drug reductions rather than slow changes, provided that the former does not carry an increased risk of an exacerbation of seizures. These arguments may be summarized as: (1) Patients may be supervised very closely over a short period of time. (2) The strategy is more likely to be followed.
(3) The effect of the drug change on seizures, medication side-effects and serum concentrations of concomitant antiepileptic drugs may be rapidly determined. (4) Potentially useful drug treatments can be explored quickly, and the patient established on optimal therapy as soon as is possible.
In a prospective, double-blind study of the withdrawal of phenytoin, carbamazepine and sodium valproate from the polytherapy of patients with intractable epilepsy, phenytoin was withdrawn in 50 mg decrements every 2 or 5 days, carbamazepine was discontinued at 200 mg every 2 or 5 days, and sodium valproate at 200 mg every 2 or 4 days6. It was concluded that:
(1) The rapid tapering of phenytoin and valproate appeared to be generally problem free in these patients with active epilepsy, who were under close supervision. The optimal rates of carbamazepine reduction remained uncertain. Withdrawal of carbamazepine was often associated with an increase in seizures and its discontinuation should be approached with caution, even if its use does not appear to have helped seizure control. It should not be taken that withdrawal of phenytoin and valproate will always be free of problems as there are clearly many patients who would have a marked increase in seizure should these antiepileptic drugs be stopped, however slowly. (2) Whilst the acute effects of a change of drug may be rapidly determined, the longer term consequences need more prolonged observation, particularly if seizures occur only infrequently.
(3) If a marked increase in seizures occurs on reduction and withdrawal of an antiepileptic drug, and serum concentrations of concomitant drugs have not fallen, the increase in seizures is likely to continue until the withdrawn drug is recommenced. In the event of a serious increase in seizures, diazepam should be given and the withdrawn drug reintroduced. This action would be expected to restore the seizure frequency to the baseline level, immediately. (4) In cases of failed antiepileptic drug withdrawal, the withdrawn drug may be recommenced at the dose that was being taken prior to the increase in seizures, and then increased further as necessary. In the case
of reintroduction of phenytoin or phenobarbitone, a loading dose might be necessary. (5) A marked increase in seizures on discontinuation of a particular antiepileptic drug suggests that that drug is efficacious and it may be that a higher dose, than that which was being given at the baseline, will improve seizure control. On the other hand, a lower dose may be equally effective.
(6) If a drug change does result in an increase in seizures this can be managed safely, and the clinician and the patient can be reasonably confident that, with appropriate treatment, there will be no long lasting deterioration of control of the seizure disorder. Holistic issues In addition to the pharmacological aspects, it is important to consider the implications and consequences of intractable epilepsy, which are often more devastating than the seizures themselves. Not least, people with intractable epilepsy benefit enormously from somebody taking a personal interest in them and their problems, and from advice on practical issues, such as safe cooking with a microwave oven, and a discussion of reasonable expectations and limitations, with regard to prognosis and the prospects of independent living, a social and leisure life and of employment. Conclusion Patients whose epilepsy is intractable are the most difficult to treat. Careful attention may improve these patients' lot, with improved control of seizures and reduction of the toxic effects of antiepileptic drug therapy. Acknowledgments: I am grateful to colleagues who have collaborated in investigations and to the Brain Research Trust, Ciba-Geigy Pharmaceuticals and the Merrell Dow Research Institute for support. References 1 Duncan JS, Shorvon SD. Prognosis of epilepsy. Br J Hosp Med 1986;35:254-9 2 Reynolds EH, Shorvon SD. Monotherapy or polytherapy for epilepsy? Epilepsia 1981;22:1-10 3 Schmidt D. Two antiepileptic drugs for intractable epilepsy with complex partial seizures. J Neurol Neurosurg Psychiatry 1982;45:1119-24 4 Thompson PJ, Trimble MR. Anticonvulsant drugs and cognitive functions. Epilepsia 1982;23:531-44 5 Chadwick D, Reynolds EH. When do epileptic patients need treatment? Starting and stopping medication. BMJ 1985;290:1885-8 6 Duncan JS, Shorvon SD, Trimble MR. The withdrawal of phenytoin, carbamazepine and sodium valproate in patients with active epilepsy. Epilepsia 1990;31:324-33 7 Shorvon SD, Reynolds EH. Unnecessary polypharmacy for epilepsy. BMJ 1977;1:1635-7 8 Mattson RH, Cramer JA, Collins JF, et aL Comparison of carbamazepine, phenobarbital, phenytoin and primidone in partial and secondarily generalized tonic clonic seizures. N Engi J Med 1985;313:145-52 9 Falconer MA, Taylor DC. Surgical treatment of drugresistant epilepsy due to mesial temporal sclerosis. Arch Neurol 1968;19:353-61 10 Jensen I. Temporal lobe surgery around the world. Acta Neurol Scand 1975;52:354-73 11 Duncan JS, Sagar HJ. Seizure characteristics, pathology, and outcome after temporal lobectomy. Neurology 1987;37:405-9
161
162
Journal of the Royal Society of Medicine Volume 84- March 1991
12 Polkey CE. Neurosurgery. In: Laidlaw J, Richens A, Oxley J, eds. A textbook of epilepsy, 3rd edn. Edinburgh: Churchill Livingstone, 1988:484-510 13 Lesser RP, Modic MT, Weinstein-MA, et al. Magnetic resonance imaging (1.5 Tesla) in patients with intractable focal seizures. Arch Neurol 1986;43:367-71 14 Franceschi M, Triulzi F, Ferini-Strambi L. Focal cerebral lesions found by magnetic resonance imaging in cryptogenic nonrefractory temporal lobe epilepsy patients. Epilepsia 1989;30:540-6 15 Shorvon SD, Reynolds EH. Reduction in polypharmacy for epilepsy. BMJ 1979;2:1023-5 16 Theodore WH, Porter RJ, Removal of sedative-hypnotic antiepileptics from the regimens. of patients with intractable epilepsy. Ann Neurol 1983;13:320-4 17 Callaghan N, O'Dwyer R, Keating J. Unnecessary polypharmacy in patients with frequent seizures. Acta Neurol Scand 1984;69:15-19 18 Albright P, Bruni J. Reduction of polypharmacy in epileptic patients. Arch Neurol 1985;42:797-9 19 Jeavons PM, Clark JE, Maheshwai MC. Treatment of generalized epilepsies of childhood and adolesoence with sodium valproate ('Epilim'). Deu Med Child Neurol 1977;19:9-25 20 Milano Collaborative Group for Studies on Epilepsy. Long term intensive monitoring in the difficult patient. Preliminary results of 16 months of observationsusefulness and limitations. In: Gardner-Thorpe C, Janz D, Meinardi H, Pippenger CE, eds. Antiepileptic drug monitoring. Tunbridge Wells: Pitman, 1977: 197-213 21 Callaghan N, O'Callaghan M, Duggan B, Feely M. Carbamazepine as a single drug in the treatment of epilepsy. J Neurol Neurosurg Psychiatry 1978;41:907-10
22 Schobben AFAM. Pharmacokinetics and therapeutics in epilepsy. Nijmegan: Stichting Studentenpers, 1979: 251-8 23, Maheshwari MC, Padmrini R. Role of carbamazepine in reducing polypharmacy in epilepsy. Acta Neurol Scand 1981;64:22-8 24 Gannaway DJ, Mawer CE. Transfer from multiple to single antiepileptic drug therapy. Lancet 1981;i:217 25 Covanis A, Gupta AK, Jeavons PM. Sodium vaiproate: monotherapy and polytherapy. Epilepsia 1982;23:693-720 26 Fischbacher E. Effect of reduction of anticonvulsants on wellbeing. BMJ 1982;28&.423-4 27 Bennett HS, Dunlop T, Ziring P. Reduction of polypharmacy for epilepsy in an institution for the retarded. Dev Med Child Neurol 1983;25:735-7 28 Schmidt D. Reduction of two drug therapy in intractable epilepsy. Epilepsia 1983;24:368-76 29 Roman EJ, Lambert JB, Buchanan N, Barrah N. -Rationalization of therapy in severe epilepsy' Aust N Z J Med 1983;13:601-4 30 Alvarez N, Hazlett J. Seizure management with minimal medications in institutionalized mentally retarded epileptics:A prospective study. First report after 4% years of follow-up. Clinical Electroencephalography 1983;14:164-73 31 Schmidt D, Richter K. Alternative single anticonvulsant drug therapy for refractory epilepsy. Arch Neurol 1986;19:85-7 32 Duncan JS, Shorvon SD. Rates of antiepileptic drug reduction in active epilepsy-current practice. Epilepsy Res 1987;1:357-64
(Accepted 28 September 1990)
