Epilepsy: A Comprehensive Textbook
2nd Edition

Although difficult to quantify, emotional stress is the most common factor (30% to 66%) identified by patients, mostly by women.^6,19,24,27 Since the time of Hughlings Jackson and Gowers, attention has been paid to the role of the emotions in the precipitation of epileptic seizures. Many reports provide evidence of an association between stressful life events or tension states and seizures.^{7,10,18,19,26} In one study,¹⁸ 58% of patients reported emotional stress, such as worry, anxiety, frustration, and anger, as the second most frequent precipitating or modulating factor for seizures. Another study⁷ examined the psychologic factors confronting psychiatric patients with epilepsy before the onset of epilepsy. No fewer than 20 of 51 patients had experienced a severe emotional disturbance shortly before the first attack, caused, for example, by a mother-in-law’s serious illness, the death of a mother, arrest by the Gestapo, a husband’s heart attack, retirement after 45 years with the same firm, severe financial difficulties, and frequent unemployment. In patients subjected to a stress interview with electroencephalogram (EEG) recordings, neuronal instability increases during procedures, as evidenced by a seizure or emergence or increase in epileptiform activity. The mechanisms whereby emotional factors may elicit seizures have yet to be determined. An activation of specific networks has been involved. Patients with generalized seizures and those with partial seizures seemed to be equally sensitive to emotional stress.¹⁹ However, patients with temporal lobe epilepsy would be expected to be more vulnerable to emotional activation of seizures than patients with complex partial seizures because the anatomic structures involved during complex partial seizures are those that handle normal emotional responses. In one study,¹⁰ stress represented 30% of seizure precipitants. Patients with temporal lobe epilepsy were the most likely to identify stress (46%) and patients with cryptogenic epilepsies were the least (15%). However, multiple factors may converge. Emotional disturbance may lead to sleep deprivation, noncompliance, excessive drinking, and even hyperventilation. Stress may be present during a period of time ranging from minutes to days, weeks, or years. Relaxation was the most commonly used technique for aborting seizures by 53.7% of the patients, which is consistent with the fact that 53% of the patients who could identify seizure precipitants reported that stress or tension could trigger their seizures.²⁵

Many patients have seizures only at night. In the 1950s, some neurologists erroneously considered sleep (“morpheic”) epilepsy as an entity. In several epileptic syndromes, such as idiopathic epilepsies, benign partial epilepsy of childhood with centrotemporal sharp waves, and symptomatic or cryptogenic frontal and temporal lobe epilepsies, seizures appear preferentially during sleep. Sleep deprivation is the second most often reported seizure precipitant in four studies.^{3,6,10,19,22,26} As sleep deprivation usually occurs during periods of overactivity or tension often associated with the use of stimulants and overhydration, its role is not always clinically clear.¹ Patients with idiopathic epilepsies frequently reported sleep deprivation as a significant precipitant.¹⁰ There is no gender distribution. Sleep deprivation is a common precipitant of seizures in juvenile myoclonic epilepsy (JME).⁶ In the adolescent and young adult population affected by JME, late nights of studying or socializing frequently result in myoclonic jerks and generalized tonic–clonic seizures (GTCSs). Sleep deprivation as an activator of EEG epileptiform discharges is commonly used as a diagnostic aid in epilepsy⁹ but has been recently reported as not affecting seizure frequency during video EEG monitoring¹⁷ (see Chapter 188). Sudden awakening is a major precipitant of JME. Provoked awakenings are more dangerous than spontaneous ones.
The effect is more noticeable when sleep is interrupted during unstable phases, such as rapid eye movement (REM) sleep during the early part of the night and phase 2 sleep at the end of the night.

Reports on the effects of fatigue and exercise on seizure frequency have been controversial. In Nakken’s study, which compared seizure precipitants in three twin registries (Norway, Denmark, and United States), tiredness is the third most frequently reported seizure precipitant.¹⁹ Physical activity was given as a precipitant factor by 6% of the Norwegian sample but accounted for only 0.3% and 0.7% of the reported precipitants by Danish and American populations, respectively. Some studies⁷ have suggested that aerobic exercise raises seizure thresholds, thereby conferring a protective effect. Hyperventilation in the resting patient is very different from that in response to exercise and rarely provokes seizures. This is a consequence of metabolic acidosis that occurs during exercise. Seizures induced by exercise are actually seizures occurring shortly after exercise. The risk is greater after continuous exercise than after intermittent exercise. The lower the pH and base excess values are after exercise and during the recovery phase, the greater the number is of EEG epileptiform abnormalities. Sustained hypoxia or hypoglycemia after vigorous physical exercise is probably also involved. Physical activity such as running or jumping is reported by 7% of the patients as a technique for aborting seizures.²⁴

The interactions between alcohol and epilepsy are complex. Chronic and acute effects of alcohol on the central nervous system are quite different and, sometimes, even opposite to one another.² These interactions are discussed more thoroughly in Chapter 268. It is commonly perceived that alcoholic beverages are a frequent cause of seizures in adult patients with epilepsy. Alcohol consumption was the fourth most frequent precipitant (5.7%) reported in one study.¹⁹ Given the usual reluctance to admit alcohol use, this proportion is clearly underestimated. Alcohol abuse is often associated with poor compliance and sleep deprivation. However, there is little experimental evidence that moderate ingestion of alcohol influences seizure occurrence.¹³ In fact, one study¹⁸ found that moderate alcohol intake does not trigger seizures and can even decrease EEG epileptiform abnormalities! Alcohol abuse may be accompanied by seizures in two situations.² On the one hand, chronically alcoholic patients with or without epilepsy experience attacks during bouts of heavy drinking but more frequently on withdrawal. On the other hand, seizures may be precipitated in nonalcoholic patients with epilepsy after excessive drinking. Seizures occur during the period of rapidly falling alcohol blood levels, especially when excessive alcohol intake is associated with insufficient sleep.¹⁸

Penetrating insights on compliance with antiepileptic drugs (AEDs) have been published.²³ It is generally accepted that about one third to one half of people receiving long-term therapy take their medication in ways that differ from the clinical prescription. In a small series involving 40 patients admitted with seizures, noncompliance was the most common potentially preventable precipitating factor (45%),²⁶ and patients with subtherapeutic AED levels despite good compliance was the second most common precipitating factor. Noncompliance is probably underestimated and the degree of compliance varies considerably, as does the motivation for the consequences of poor compliance. Despite some patients’ beliefs, occasional omission of one or two doses can be harmless. Because of a longer drug-free period, however, a greater risk exists when patients decide for themselves that they are cured and stop taking their medication. Two mechanisms explain the occurrence of seizures in such situations. First, a seizure can reflect the natural course of pharmaco-dependent disease. Seizures occur with different lag times, which vary according to the severity of epilepsy and with the drug (lag times are short with most AEDs but longer with valproate, probably because of a carry-over effect). Usually, a single seizure is observed, at least in moderate epilepsies, when AEDs are withdrawn, but status epilepticus is possible in severe epilepsies. Second, withdrawal seizures may also occur when barbiturates or benzodiazepines are stopped, as is observed in persons without epilepsy. These are generalized seizures, and they occur shortly after drug withdrawal.

Many classes of pharmacologic agents prescribed at therapeutic dosages have been implicated in lowering the seizure threshold: Antidepressants, antipsychotics, central nervous system (CNS) stimulants, hypoglycemic agents, antimicrobial agents, aminophylline, antihistaminics, ephedrine, steroids, and a wide variety of other drugs.^11,15,16,29

Significant metabolic derangements can result from diarrhea, constipation, acute infections, liver and renal failure, and diuretic intake, especially in the very young and very old. Hypernatremia or hyponatremia, hypocalcemia, and hypoglycemia can provoke seizures in patients with or without epilepsy. In patients with diabetes, seizures frequently occur with overdose of insulin or sulfonamides, and perhaps during hypoglycemia following a large meal.¹

Hyperventilation produces a respiratory alkalosis that causes central vasoconstriction and, if sufficiently marked, can modify the level of cerebral oxygen and glucose. Excessive breathing incommensurate with or without physical effort may eventually induce seizures in some patients.⁵ Hyperventilation easily elicits absence seizures in children and, less commonly, other types of seizures.¹² Involuntary hyperventilation may occur in the course of a patient’s daily activities due to anxiety, sobbing, or sexual activity.

Febrile convulsions are a special syndrome, characterized by seizures caused by a sudden rise in body temperature (see Chapter 57). They are seen only in children <5 years of age, usually <3 years of age. At any age, but mainly in the elderly, an acute febrile infection may provoke a seizure in the susceptible patient. Nevertheless, fever is often reported by patients as a precipitant factor of their seizures.^10,19

Catamenial epilepsy is defined as the occurrence of at least 75% of seizures per month within a period including the 3 days
preceding and the 4 days following menstruation.²⁰ While this is an uncommon condition, seizure frequency is reported to be affected by the menstrual cycle in 10% to 70% of women with epilepsy.²¹ Many women report an increase in seizure frequency around the time of their menses. Herzog described three distinct patterns of catamenial epilepsy: Perimenstrual, periovulatory, and inadequate luteal-phase catamenial seizures.¹⁴ Cyclic changes of ovarian hormones—estrogens (proconvulsivant) and progesterone (anticonvulsivant)—seem to play an essential role in the occurrence of catamenial seizures, as do possible variations in AED levels during the menstrual cycle. Natural progesterone therapy has proven to be effective and the use of neurosteroids in the future has to be explored.

Pregnancy has relatively little influence on seizure frequency. Altered disposition of AEDs and noncompliance seem to be the major reasons for increased seizure frequency (see Chapter 198).