Epilepsy is defined as recurrent, unprovoked seizures. Children with febrile convulsions and acute symptomatic seizures do not have epilepsy and ordinarily do not require maintenance antiepileptic drug (AED) therapy. Management of febrile seizures and symptomatic seizures is discussed elsewhere in this issue. Abnormal results on electroencephalograms (EEGs) occur in approximately 15% to 20% of all children, febrile seizures appear in 5%, and 0.5% to 1.0% of children have epilepsy. Abnormal EEGs are much more frequent in the population than epilepsy, and most children with abnormal EEGs do not have epilepsy. However, once the diagnosis of epilepsy is reasonably well established, one must determine whether treatment is indicated, and, if so, which treatment is necessary and which to advise.
Management of seizures in the primary care setting is normally through drug therapy. It means taking medication every day and this is surprisingly difficult. This means that management of the child with epilepsy is more than prescribing an AED. The general pediatrician is called on to deal with a variety of issues, including side effects and toxicity, medication compliance, and school and other social issues. One must focus on the child with epilepsy, rather than just the epilepsy itself; otherwise, compliance problems and quality of life issues may nullify the value of modem AEDs. This article reviews the management of the child rather than the management of seizures.
The natural history refers to patient outcomes with or without treatment. Gowers and other 19th century neurologists suspected that each seizure increased the likelihood of more seizures; therefore, they thought it best to start treatment early (although they had no effective treatment other than bromides). Modern epidemiologic studies suggest the opposite. Most patients have only a few seizures, with or without treatment. There is little reason to begin treatment after just one seizure, even if that seizure was unprovoked. The challenge is to identify those likely to have chronic epilepsy and to start their treatment earlier than the treatment of those who would have a good prognosis regardless. Thus, treatment must be individualized.
Patients with epilepsy can be divided into two groups. Those with idiopathic epilepsy have no neurologic problems other than the seizures and complications of seizure treatment; those with symptomatic epilepsy have signs of brain disease, such as developmental delay, hemiparesis, microcephaly, and tuberous sclerosis. The prognosis of idiopathic epilepsy is much better than that of symptomatic epilepsy. Patients with symptomatic epilepsy enter seizure remission less often and are more likely to experience a relapse if treatment is withdrawn.
The prognosis for seizure remission is generally good. A British study of epilepsy in general practice found that 80% to 90% of new cases went into remission, and 68% had a 5-year seizure-free remission within 9 years.1 Approximately onethird of patients in remission experienced a relapse when medication was withdrawn. Remission rates were equal in children and adults. A study of epilepsy patients managed by a specialist yielded similar results - 253 of 281 children attained remission.2 Ninety percent went at least 1 year without a seizure and 57% went 3 years. Almost 80% achieved remission in the first year. Of the 44 who experienced a relapse, the conditions of all but 1 came under subsequent control. Remission is less likely when the seizures begin in infancy and when the epilepsy is symptomatic rather than idiopathic. The more years that seizures continue, the less the chances of remission.3
Prognosis extends beyond seizure control. Follow-up studies of childhood epilepsy show significant risk of adult psychosocial dysfunction (being unmarried, living alone, or being unemployed).4,5 Absence epilepsy, a generally favorable form of epilepsy, has much worse psychosocial outcomes than juvenile rheumatoid arthritis. One study showed that former seizure patients were 19 times more likely to have an unwanted pregnancy, 6.4 times more likely to have had school behavior problems, and 3.7 times more likely to not finish high school than patients with juvenile rheumatoid arthritis.6 The quality of life of children with epilepsy is also worse than that of asthmatic children.7
Epilepsy and school can cause problems. Teachers may fear that they will be called on to perform cardiopulmonary resuscitation and assume that staring in school always indicates seizures. They may suggest that home schooling is best for the child until the seizures are controlled. One may speak of children having school phobia and schools having "seizure phobia." Parents often ask schools not to push children with epilepsy academically and to keep them out of sports. These attitudes contribute to adult psychosocial dysfunction.
Epilepsy increases the risk of death. This is due to accidental death, including drowning, and sudden unexplained death, a late complication of chronic epilepsy.5,8,9 Nonaccidental sudden death is rare and is more likely in symptomatic man in idiopathic epilepsy. Sudden death in epilepsy is most common between ages 15 and 44 years, although epilepsy is less common in this age group than in younger children.9 Some of these deaths are associated with heart disease.10 The risk of death should not be exaggerated. More than 1,500 of this physician's patients have had video telemetry monitoring during which seizures were recorded under controlled conditions. There were more than 5,000 seizures, and none of the patients died or required cardiopulmonary resuscitation. Children with epilepsy, even those with pirre absence epilepsy, have more injuries of all sorts, including burns." Most of these injuries occur after starting treatment and almost all are preventable.
Treatment is advised if its estimated benefits exceed the likely risks or costs. Epilepsy treatment in primary care is drug treatment. Nondrug treatments are limited to special circumstances. The ketogenic diet has a small place in epilepsy treatment, but must be organized and supervised at centers experienced with this therapy. Parents may ask about it as a form of "natural" and drugfree therapy. However, it is a complex but useful therapy that can cause major side effects.12 It may be difficult for most parents. Likewise, epilepsy surgery can be useful, and even curative, but is relevant only for certain patients who fail standard therapy.
Before any therapy is started, parents should be prepared for future seizures and advised how to manage them at home, when to call the physician, and when to bring the child to the clinic or emergency department. Elsewhere in this issue, Bebin discusses the acute management of seizures in the home, clinic, and emergency department. Parents must know that children cannot swallow their tongues and that nothing should be inserted into the mouth during a seizure. Tongue depressors and other objects may injure the palate or teeth, and fingers in the mouth are usually bitten. The child should be turned to one side during the seizure to help manage secretions. Children should be taken to the emergency department only when the convulsion lasts longer than 5 minutes, is associated with high fever, or causes injury, or when there are multiple convulsions in the same day. It is usually reasonable to wait until several seizures have occurred before starting treatment. Such delay does not jeopardize long-term outcomes.13 Instead, it may be beneficial because parents can become more convinced of the need for treatment, more careful about giving medication, and more accepting of drug side effects, which are generally mild and temporary.
There are many potent AEDs available today. Elsewhere in this issue, Russell and Parks review the commonly prescribed AEDs in children. This physician recommends that primary care physicians prescribe only those AEDs approved by the Food and Drug Administration for pediatric use and with which they have experience. Specialists should be the ones to initiate off-label prescriptions. This reduces the primary care physician's risk and may mitigate legal problems in the event of complications. This strategy limits the general pediatrician's choices to phénobarbital, primidone, phenytoin, carbamazepine, valproate sodium, ethosuximide, and benzodiazepines.
Differences between patients are greater and more important than differences between drugs. This physician uses carbamazepine and valproate sodium most. Carbamazepine is used for seizures that have obvious focal components. When possible, twice daily dosing is best at 7 am and 7 pm. When seizures do not respond to carbamazepine given every 12 hours, then replacing the standard carbamazepine with a delayed-release formulation is useful as a trial. This physician prefers Carbatrol (Shire Richwood, Inc., Florence, KY) because it comes in a sprinkle formulation. The capsule can be opened and the granules sprinkled onto soft food. This can be given even to small babies. Phenytoin, phénobarbital, and mysoline are alternatives to carbamazepine.
This physician prefers valproate sodium for generalized seizures. It is a broad-spectrum AED, also useful for focal seizures, and is available in a sprinkle formulation. This physician rarely gives it more than twice a day. Valproate sodium should be used carefully in small children. Those who have inborn errors of metabolism or take enzyme-inducing drugs, such as phénobarbital, phenytoin, and carbamazepine, are at greater risk for organ toxicity than older children. Primary care physicians are advised not to give valproate sodium to children younger than 1 year of age. Some infants with epilepsy respond to valproate sodium, but the prognosis for this condition may be poor. So, those children should be referred to a specialist. Phénobarbital is recommended for children younger than 6 months of age. Phénobarbital, carbamazepine, and phenytoin are ineffective for absence seizures. Ethosuximide and valproate sodium are the drugs of choice here. However, ethosuximide will not control motor seizures, which accompany absence epilepsy in 30% to 40% of cases.
Treatment should begin gradually. Because full doses make patients feel bad, they often stop the medication. The AED should be started at approximately one-third of the final target dose and increased every 4 to 5 days. When seizures are frequent, this physician gives a dose of parenteral lorazepam, sends the child home, and increases carbamazepine or valproate sodium more quickly. This usually produces temporary drowsiness.
The AED formulation is important. Contrary to the thinking of many pharmacists and parents, tablets are acceptable for children and have advantages over liquid preparations. Liquids produce rapid absorption, produce high blood levels, generally have to be given more frequently, and are often measured carelessly. Some tend to settle if not shaken as they are used. Tablets and sprinkles are preferable to liquids and capsules. When liquid preparations are used, the dosing should be stipulated in milliliters, not by the teaspoon or fractions of teaspoons, to ensure accuracy.
The frequency of follow-up visits depends primarily on seizure control and side effects. The child should be seen approximately 1 month after the initiation of treatment. At this time, the physician should inquire about further seizures and should try to separate other activities the parents might perceive as seizures, such as staring or acting strangely. Parents may pay closer attention to the child's sleeping and may actually start sleeping with the child. This should be discouraged. Jerks while sleeping can be a concern, but are rarely significant. It is important to review the drug dose and schedule. The physician should ask specific questions about who gives the drug, at what time of day, and side effects. It is also important to ask about the child's general health, well-being, school performance, and social adjustment. Time must be set aside for the invariable questions that arise after the family has had the opportunity to talk with relatives, friends, and teachers. Fear causes some mothers to sleep with the child constantly and makes them unable to discipline the child or deny his or her wishes. This can distort the child's psychosocial development.
The physical examination should focus on the well-being of the child and a search for signs of toxicity, rash, or other adverse effects of medication. Laboratory studies may be necessary, as discussed below.
If the child is doing well, subsequent visits should be scheduled at 3- to 6-month intervals. If definite and unprovoked seizures occur, the AED dose should be increased by 15% to 25% and the next visit should be scheduled sooner, perhaps within 2 months. The dose can be increased again as long as there are no signs of dose-related side effects, such as drowsiness or slurred speech. The dose should not be increased until at least 10 to 14 days have passed since the last change to determine the effect of the previous increase. The physician should be satisfied with success. The dose should not be increased if seizures are controlled, or because an AED blood level is "too low." If seizures persist after the drug is increased to the point of producing side effects, another AED should be phased in or consultation with a specialist should be sought. If seizures stop and the child remains in remission for 2 years, the drug may be slowly tapered for the child with idiopathic epilepsy who is not driving. If the child is driving or has symptomatic epilepsy, this physician recommends 3 years before attempting to withdraw the drug. The risk of relapse is higher in patients with symptomatic epilepsy, and a daytime seizure while driving can be catastrophic.
MANAGING AED SIDE EFFECTS
AEDs cause worry more often than major problems. Death or irreversible complications are extremely rare. Acetaminophen and aspirin cause far more American deaths than all of the AEDs combined. The main problems are subtle disturbances of higher nervous system function: sedation and impaired learning or concentration, irritability, and sleep disturbance. These often worsen if the dose is increased, but can occur at low doses and blood levels. A child with persistent problems taking one AED may do well with another, especially one from a different drug family. Systemic intolerance, manifested by rashes and organ toxicity, is a risk during the first months of treatment with any AED. Skin rashes, sometimes associated with hepatic and other organ toxicity in the aromatic AED hypersensitivity syndrome,14 occur more frequently with carbamazepine, phenytoin, and lamotrigine.
Each AED has its own side effect profile. One study reported 245 side effects among the 244 patients completing the study.15 The two drugs studied (valproate sodium and carbamazepine) had a similar incidence of side effects, although specifics varied. Somnolence, fatigue, headache, and rash were more common in patients taking carbamazepine, whereas weight gain, increased appetite, gastrointestinal disturbances, and alopecia were more common in those taking valproate sodium. Some patients had two or three side effects, but perhaps only half were truly drug related and none were life threatening. One child among the 260 subjects had significant liver abnormalities and was found to have infectious hepatitis.
Drug interactions can be troublesome because AEDs interact with other drugs. The most common question concerns AEDs and oral contraceptives. Phénobarbital, phenytoin, and carbamazepine all accelerate the metabolism of oral contraceptives and require the use of high estrogen preparations. Valproate sodium and ethosuximide do not. Combinations of carbamazepine with erythromycin, isoniazid, and verapamil hydrochloride often cause problems. Comprehensive reviews of this subject are available.1617 When there is doubt, a knowledgeable pharmacist can help. It is important to remember the risk of ingestion of AEDs by small children. Parents must store the medication carefully.
Organ toxicity from AEDs is rare, and usually occurs in the first 4 months of use. It is unlikely in a child who is eating well, playing, and gaining weight. Barbiturates and benzodiazepines require little laboratory monitoring for organ toxicity. Monitoring of phenytoin, carbamazepine, and valproate sodium treatment may be more extensive, but must be used judiciously.
Neutropenia is often detected in children taking carbamazepine. If neutropenia is the only abnormality, more than 95% of children have a normal complete blood count at a later date. In the meantime, parents may become terrified and tests may be repeated every few days, which is excessive. Small transaminase increases (less than threefold) in children who are eating, active, and seem well are usually due to enzyme induction.18 In such cases, serum 7-glutamyl-transferase is usually increased to a greater extent and the liver synthetic function (albumin, prothrombin time, and partial thromboplastin time) is normal. Monitoring of well-being by weight and general examination and inquiring about sleep disturbances, behavior, and schoolwork are more important than screening laboratory tests.
The main benefit of laboratory monitoring of AED blood levels is to detect noncompliance. Undetectable serum levels mean that the drug was not given or was not absorbed in the gut. If the drug is detectable later, one can assume that the child had not received the AED. The patient's serum drug level does not necessarily correlate with whether he or she is getting the "right" amount of the drug. More than half of patients do best with drug levels above or below the conventional therapeutic range.19 The serum drug level that has controlled seizures is high enough and the level that produces no side effects is not too high.20 The goal of treatment is to produce remission of seizures, not to reach a particular drug level.
COMPLIANCE WITH DRUG THERAPY
One should anticipate compliance problems. Only approximately half of patients with chronic diseases take their medications as prescribed, and physicians cannot accurately predict which patients are compliant.21 Rarely do parents report noncompliance because this is due to more than just forgetfulness and ignorance. Families often decide that a drug is "too strong" or give half the prescribed dose.22 The parent may increase the dose for reasons not related to its indications (eg, an extra dose is given when the child appears nervous). Parents may alternate between underdosing and overdosing. Unintentional noncompliance is associated with forgetful, busy parents who do not supervise medication taking or who cannot deal with a rebellious or difficult child. In this physician's experience, patients who are noncompliant are more likely to have school behavior problems than patients with acceptable drug levels. Pillboxes are the best way to control forgetfulness.
Intentional noncompliance often occurs if there has been poor communication between the physician and the family. The noncustodial parent, relative, or others can sabotage the management if they do not understand the basis for the chosen treatment. The pediatrician should ask at each follow-up visit whether the child is getting the medication as prescribed, who gives the medication, and whether anyone has reservations about the use of the drug and its side effects. Persistent worries deserve a nonurgent specialist consultation. Many parents are trapped between fears that the child will die or become brain damaged by seizures, and fear that the AED will make the child into a "zombie." This ambivalence makes the parents unable to enforce regular medication taking. Simply talking about these fears is helpful. Full compliance, regular visits, and sensible restrictions reduce the risks of serious problems considerably.
Physician behavior influences compliance. Families who feel that the physician is sympathetic and interested are more likely to be compliant or to admit their failures. Patients with regularly scheduled appointments are more compliant than those simply told to contact the physician when there are problems.23 One should strive for a therapeutic alliance among the physician, the family, teachers, and the child. This requires communication and continuity of care. It can promote optimism and compliance with medication and restrictions.
PROBLEMS, ACTIVITIES, AND RESTRICTIONS
Children and families should be treated, rather than seizures. This means anticipating common problems, even if the seizures remit. One can expect parents to report minor disturbances of behavior or sleeping and to be worried about trivial rashes after treatment has begun. Often these symptoms resolve spontaneously with no intervention. The most common causes for treatment failure are parents who do not return for follow-up, children who do not receive the medication as prescribed, children whose spells are not epileptic, parents who use emergency departments as physicians and have no continuity of care, and children switched repeatedly from one AED to another because the parents perceive adverse side effects on any drug. The most common physician errors are failure to increase the dose of medication when seizures continue despite "therapeutic levels," and acceptance of strange behavior as being seizures. Compliance problems are common and may be detected by AED blood levels.
The first year of management is critical. During this period, the parents, teachers, and the child develop attitudes that may become irreversible and may interfere with therapy and later psychosocial adjustment. For the child who is in school, teacher anxiety, teasing by peers, and taking medications while at school are possible issues that must be dealt with. Most teachers feel inadequately informed about epilepsy and want more information if their class is to include children with epilepsy. A letter to the school is often helpful, especially if the school is asked to administer a mid-day medication.
Families must be careful about water safety. In this physician's experience, more than a dozen patients have died from drowning, usually in the bathtub at home. Some have been found in lakes and rivers, apparently having had a seizure while playing near water. The pediatrician should counsel parents accordingly. Showers are recommended for any child who bathes alone.
Patients may fall from heights and suffer serious harm. One child fell from a tree and was left paraplegic. A reasonable rule is no heights above 6 feet. Open fires are a problem, including gas stoves and heaters.
Seizures while driving a car, a motorcycle, or a tractor can produce a fatal injury. Seizures while riding bicycles are common; children must wear helmets and stay off busy streets." Patients should refrain from bicycle or horseback riding until 3 months have passed without their having a daytime seizure and then they should wear a helmet.
Seizures are extremely rare while running or actively exercising. None of this physician's patients have been injured by seizures while running or playing, and several have played on basketball teams despite only partially controlled seizures. Parents may ask to have their children excused from exercise. This often indicates an excessive fear of seizures. A few patients have seizures activated by exercise,24 but, as a rule, "seizures" associated with exercise are due to heart disease, not epilepsy. Patients with epilepsy are not advised to participate in contact sports.
Drugs That May Trigger Seizures
Sleep deprivation may precipitate seizures, even in patients in remission. This can especially be a problem in teenagers, and sleep deprivation should be avoided. An adolescent who stays out all night at the prom may have an early morning seizure.
Certain drugs should be avoided because they can activate seizures25 (Table). They are not absolutely contraindicated, but should be used cautiously and for strong indications. Drugprovoked seizures are dose-dependent. Restriction of Imipramine hydrochloride to a total dose of 1.0 mg/kg/d and methylphenidate hydrochloride to 0.75 mg/kg/d usually avoids this problem in children with epilepsy.
Children with epilepsy are as likely to get other illnesses as are normal children. There is often a mistaken tendency to attribute symptoms to the seizure diagnosis or the drug treatment. The pediatrician must continue the role as the primary caregiver, especially when a specialist also observes the child. Parents should be dissuaded from calling the specialist first for every health problem.
General pediatricians can manage most uncomplicated cases of epilepsy. They need to have a friendly relationship with an epilepsy center or a neurologist. A telephone consultation often suffices to initiate management until consultation is available. For nonurgent matters, email is a much better means of communication.
Comprehensive epilepsy centers, available in almost every state, can provide neurologic, psychiatric, and psychological consultation. They may offer more sophisticated interpretations of EEGs and neuroimaging scans. These tests do not "speak for themselves," and hospitals performing fewer than 250 pediatric EEGs a year may not produce technically adequate studies. EEG-video telemetry is useful in diagnosis, and for detecting pseudo-seizures. It is not useful for evaluating spells that occur less than two to three times a week.
Referral is indicated if the diagnosis is uncertain, the seizures are frequent or the handicap severe, or if the seizures persist despite treatment. Referral should be considered if the parents are anxious about the diagnosis and /or treatment of epilepsy, and especially if the seizures persist after 1 year of treatment. Unprovoked seizures in the first year of life are often associated with a poor prognosis and referral is usually made for such patients. Infants may be given phénobarbital if the diagnosis of epilepsy is confident.
Several sources of information can be useful to the pediatrician. Taylor's book20 on pediatric epilepsy and that of Dodson and Pellock26 are helpful and concise. On the Internet, PubMed (http: / / www.ncbi.nlm.nih.gov / PubMed / ) provides a good, rapid, free online service from the U.S. National Library of Medicine. When keywords are entered, a search is generated and relevant articles are produced. Another good on-line source of epilepsy information is the Washington University Epilepsy Center's web site (http:/ /www.neuro.wustl.edu/epilepsy/ information/). Important new developments occur each year and rapid access to pertinent information is helpful for the busy pediatrician.
Pediatricians can satisfactorily manage epilepsy in children if the diagnosis is confident and if continuity of care and compliance are obtained. The health hazards of epilepsy are overstated, and good outcomes can be expected for most children. If the child has definite motor seizures with an obvious neurologic handicap, parents should be given the option of AED treatment. This type of patient is more likely to develop chronic, refractory epilepsy. If the child is neurologically normal or the diagnosis is uncertain, it is better not to begin therapy after the initial seizure. Simple precautions can reduce the likelihood of injury from seizures, regardless of whether an AED is prescribed. The child and the family are the focus of treatment, not the seizures. There is much more to treating diabetes than prescribing insulin, and much more to the management of epilepsy than prescribing AEDs.
1. Cockerell OC, Johnson AL, Sander JWAS, et al. Prognosis of epilepsy: a review and further analysis of the first nine years of the British National General Practice Study of Epilepsy, a prospective population-based study. Epilepsia. 1997;38:31-46.
2. Hauser E, Freilinger M, Seidl R, Groh C. Prognosis of childhood epilepsy in newly referred patients. / Child Neurol. 1996;11:201-204.
3. Semah F, Picot M-C, Adam C, et al. Is the underlying cause of epilepsy a major prognostic factor for recurrence? Neurology. 1998;51:1256-1262.
4. Jalava M, Sillanpaa M, Camfield C, Camfield P. Social adjustment and competence 35 years after onset of childhood epilepsy: a prospective controlled study. Epilepsia. 1997;38:708-715.
5. Sillanpaa M, Jalava M, Kaleva O, Shinnar S. Long-term prognosis of seizures with onset in childhood. N Engl J Med. 1998;338:1715-1722.
6. Wirrell EC, Camfield CS, Camfield PR, et al. Long-term psychosocial outcome in typical absence epilepsy: sometimes a wolf in sheeps' clothing. Arch Pediatr Adolesc Med. 1997;151:152-158.
7. Austin JK, Huster GA, Dunn DW, et al. Adolescents with active or inactive epilepsy or asthma: a comparison of quality of life. Epilepsia. 1996;37:1228-1238.
8. Ficker DM, So EL, Shen SK, et al. Population-based study of the incidence of sudden unexplained death in epilepsy. Neurology. 1998;51:1270-1274.
9. Hauser WA, Annegers JF, Elveback LR. Mortality in patients with epilepsy. Epilepsia. 1980;21:399-412.
10. Natelson BH, Suarez RV, Terrence CF7 Turizo R. Patients with epilepsy who die suddenly have cardiac disease. Arch Neurol. 1998;55:857-860.
11. Wirreì EC, Camfield PR, Camfíeld CS, et aï. Accidental injury is a serious risk in children with typical absence epilepsy. Arch Neurol. 1996;53:929-932.
12. Swink TD, Vining EPG, Freeman JM. The ketogenic diet: 1997. Adv Pediatr. 1997;44:297-329.
13. Camfield C, Camfield P, Gordon K Dooley J. Does the number of seizures before treatment influence ease of control or remission of childhood epilepsy? Not if the number is 10 or less. Neurology. 1996;46:41-44.
14. Shear NH, Spielberg SP. Anticonvulsant hypersensitivity syndrome: in vitro assessment of risk. Clin Invest. 1988;82:1826-1832.
15. Verity CM, Hosking G, Easter DJ. A multicentre comparative trial of sodium valproate and carbamazepine in paediatric epilepsy. Dev Med Child Neurol. 199537:97-108.
16. Loiseau P. Treatment of concomitant illnesses in patients receiving AEDs. Drug Safety. 1998;19:495-510.
17. Anderson GD. A mechanistic approach to antiepileptic drug interactions. Ann Pharmacother. 1998;32:554-563.
18. Aiges HW1 Daum F, Olson M, et al. The effects of phénobarbital and diphenylhydantoin on liver function and morphology. / Pediatr. 1980;97:22-26.
19. Schmidt D, Haenel F. Therapeutic levels of phenytoin, phénobarbital and carbamazepine: individual variation in response to seizure frequency and type. Neurology. 1984;34:1252-1255.
20. Taylor MP. Managing Epilepsy in Primary Care. Oxford, England: Blackwell; 1996.
21. Goldberg AI, Cohen G, Rubin A-HE. Physician assessments of patient compliance with medical treatment. Soc Sei Med. 1998;47:1873-1876.
22. Donovan JL, Blake DR Patient non-compliance: deviance or reasoned decision-making? Soc Sei Med. 1992;34: 507-513.
23. Buck D, Jacoby A, Baker GA, Chadwick DW. Factors influencing compliance with antiepileptic drug regimes. Seizure. 1997;6:87-93.
24. Nakken KO, Loyning A, Loyning T, et al. Does physical exercise influence the occurrence of epileptiform EEG discharges in children? Epilepsia. 1997;38:279-284.
25. Zaceara G, Muscas GC, Messori A. Clinical features, pathogenesis and management of drug induced seizures. Drug Safety. 1990;5:109-151.
26. Dodson WE, Pellock JM, eds. Pediatric Epilepsy, Diagnosis and Therapy. New York: Demos; 1993.
Drugs That May Trigger Seizures