Epilepsy is a recurring behavioral, sensory, or subjective experience caused by an abnormal electrical discharge in the brain. Though abnormal changes accompany seizures of any kind, they may not be reflected in the electroencephalogram (EEG) between seizures and sometimes epileptic EEG discharges occur in the absence of clinically apparent seizures. Generalized seizures always cause a disturbance in consciousness and the EEG is virtually always abnormal during such seizures. Partial seizures may not be associated with unconsciousness, with any abnormal movement, or with an epileptiform, scalprecorded EEG, even during the episode.
It may not always be clear how much of a patient's behavior can be attributed to epilepsy and there may be disagreement among competent neurologists as to the relation of abnormal behavior to the seizure state. In such cases, a therapeutic trial of anticonvulsants may appear to resolve the medical problem, but the pragmatic definition of epilepsy as abnormal thought and behavior that resolves with the use of antiepileptic medications is medically and scientifically weak. Mental symptoms in mania and other disorders can improve during therapy with antiepileptic drugs in patients who do not have epilepsy.
The deviations of thought and behavior that may occur in epileptics can be the result of electrical epileptic discharges, the cerebral response to abnormal discharges, the destruction of tissue that has led to the epileptic condition, the
P.2
medication used to treat the epilepsy, serendipitous concurrence of unrelated conditions, or some combination of these.
Idiopathic Versus Symptomatic Epilepsy
The term symptomatic epilepsy is applied to seizure disorders with an identifiable cause, such as encephalitis, tumor, trauma, or known metabolic disease. In these conditions, seizures are considered to be a symptom of another disease. The term idiopathic epilepsy literally means an unto itself, without other disease of the nervous system. The concept idiopathic epilepsy is artificial. The epilepsy is surely the result of a physiological dysfunction, but the biochemical or neuroanatomical locus of the abnormality is not yet known. A high proportion of the relatives younger patients with idiopathic epilepsy have abnormal EEGs, yet most of these persons never have clinical seizures.Idiopathic in younger patients probably should refer to inherited epilepsy, which usually begins in childhood or adolescence; hence, using a positive family history to diagnose genetically transmitted epilepsy as phenotypic expression may be incomplete: history is helpful in making a reliable diagnosis. It is not essential to have a positive family history. The clinical course, seizure type, and EEG characteristics are sufficient for this diagnosis.
The assumption that the brains of patients with idiopathic epilepsy childhood onset are normal, except for the epilepsy, may be incorrect. The thinking processes and behavior of children with idiopathic epilepsy may not be normal even when the seizures have completely stopped. Thirteen of 20 children diagnosed as having benign Rolandic epilepsy (focal seizures arising near the Rolandic fissure) showed language impairment with difficulties in 2 or more of 12 standardized language tests. Language dysfunction, as revealed by these tests, correlated with learning difficulties at school, though the overall intelligence quotient (IQ) was average (Staden et al., 1998).
Some patients who have a genetic tendency to epilepsy also show other symptoms of brain dysfunction and have a history events known to be associated with brain damage. Acquired brain damage may promote expression of the genetic trait, and to the extent that it does, the resulting seizures are symptomatic of brain damage. It is also true that even in symptomatic epilepsy, the mechanism by which identifiable lesions cause seizures is not well understood. Whether epilepsy is idiopathic or symptomatic, caused by a gene tumor, the lesion, i.e., tumor or abnormal gene, is ever-present but seizures are intermittent. There are cerebral mechanisms that prevent the spread of discharges, among them hyperpolarization and inhibition around the focus. Inhibition can involve sites far removed from the epileptic focus, even in the opposite hemisphere. The inhibiting control mechanisms are varied and many have separate genetic determinants.
P.3
The impact of this neurophysiologic inhibition surrounding the epileptic focus on human behavior has never been assessed but some of the behavioral-cognitive distortions that epileptics encounter may originate in the activation of such natural “epilepsy control mechanisms.” When these “control” mechanisms fail, the epileptic activity spreads and a clinical seizure occurs.
Clinical Genetic Studies
For certain forms of generalized epilepsy, EEG studies have demonstrated a dominant mode of inheritance with incomplete penetrance and expressivity (Lennox et al., 1940; Metrakos and Metrakos, 1961; Bray and Weiser, 1964). Until the late 1980s, genetic influences on partial seizures were thought to be of minimal importance. Since then it has become clear that inheritance plays a major role in the pathogenesis of partial, focal seizures. Criteria for benign partial seizures, include the absence of any evidence brain dysfunction other than the seizures themselves (no cognitive, motor or sensory deficit), a family history of epilepsy, onset seizures after 2 years age, stereotyped brief seizures, frequent nocturnal occurrence, spontaneous remission in adolescence, and EEGs with spikes that have a distinctive morphology and localization on a normal background of electrical activity. The two most common the benign, focal, partial epilepsies are Rolandic and occipital epilepsy (Holmes, 1993).
Benign Rolandic epilepsy of childhood with centrotemporal spikes is the most common partial epilepsy syndrome in children. The onset of seizures is between the ages of 3 and 13 years old. The typical presentation is a partial seizure with parasthesias and tonic or clonic activity of the lower face associated with drooling and dysarthria. Seizures occur commonly at night may become secondarily generalized. They are usually infrequent and are easily controlled. The EEG shows characteristic high-voltage sharp waves in the centrotemporal regions, which are activated with drowsiness and sleep. Atypical cases common. The long-term medical prognosis is excellent with essentially all children entering long-term remission by mid-adolescence (Wirrel, 1998).
Benign occipital epilepsy of childhood is a partial epilepsy syndrome with elementary visual symptomatology frequently associated with other ictal phenomena. Seizures are usually followed by postictal headache and are often associated with interictal occipital rhythmic paroxysmal EEG activity that appears only after eye closure. Early-onset benign occipital epilepsy may consist of brief, infrequent attacks and prolonged status epilepticus. There is ictal deviation of the eyes or head both and vomiting, starting in children between the ages of 3 and 7 years old. The benign focal epilepsies are commonly associated with migraine (Andermann and Zifkin, 1998).
Mothers of epileptics are more frequently epileptic than fathers of epileptics
P.4
(Annegers et al., 1976). This may not reflect the influence of some sex-linked genetic trait but rather may stem from the adverse effect of epilepsy and anticonvulsant medications on the male libido and sexual performance. A similar situation exists among the parents of schizophrenics (see pp. 101–103).
Twin studies have buttressed the view of the dominant inheritance idiopathic epilepsy (Lennox, 1951; Inouye, 1960; Marshall et al., 1962). The same studies indicate a role for brain damage as a factor in the expression of epileptic gene.
No study has shown a concordance rate for seizures or even epileptiform electroencephalographic abnormalities in monozygotic twins greater than 85 percent. Why do monozygotic twins not have the theoretically expected concordance rate of 100 percent? Acquired brain damage in one twin might explain nonconcordance in some monozygotic pairs. This could lower the seizure threshold enough to allow expression of the inherited epileptic gene.
Acquired brain damage can give rise to epilepsy in patients who may or may not have a genetic tendency to express epilepsy. The rates of epilepsy vary among social groups. The high rate of epilepsy among African-Americans (Shamansky and Glaser, 1979) may reflect social factors that influence the prevalence of acquired brain damage. There appears to be a disproportionate number of preterm and low-birth weight infants among the poor black population, especially males. Contributory factors may include poor prenatal care, smoking, drinking and drug use, malnutrition, teenage pregnancy, low income, low educational status (Reed and Staley, 1977). Fatal child abuse is more common among African-Americans (Herman-Giddens et al., 1999) and lead exposure has been demonstrated in up to 18 percent of African-American children the recent past (Mahaffey et al., 1982; Farfel, 1985; Pirkle et al., 1998). Boys have more developmental brain problems in childhood than girls do (Rapin and Allen, 1988), including epilepsy (Gissler et al., 1999).
Complex Partial Seizures—Temporal Lobe Origin
Most commonly, the discharges that produce complex partial seizures originate in deep, medially placed limbic nuclei in the temporal lobe, which consists of the amygdala and hippocampus. They may also arise from virtually any other part of the limbic system, not all of which is located in temporal lobe. In addition, other areas of the nervous system may be the source these discharges; they may arise from frontal, occipital, parietal, diencephalic, or upper brain stem regions and then can spread through one or both temporal lobes. When electrical activity spreads to the temporal lobes, the characteristic electroencephalographic
P.5
abnormality is temporal spiking (Fig. 1-1), unilaterally or bilaterally. There are no pathognomonic electroencephalographic configurations that make a diagnosis of complex partial seizures absolutely certain, and the resting EEG may be normal.
|
Figure 1-1 The EEG record of a 46-year-old man with psychomotor seizures. Note the well-localized, left anterior temporal spikes. L Inf E, left inferior frontal; Ant T, anterior temporal; LT, left temporal; LTO, temporal occipital; R Inf E, right inferior frontal; R Ant T, right anterior temporal; RT, temporal; RTO, right temporal occipital. |
Subjective Feelings
Manifestations of complex partial seizures temporal origin include subjective experiences, automatisms, and postural changes. The subjective feelings (cognitive, psychosensory, affective symptoms) include forced, repetitive, and disturbing thoughts, alterations of mood, sensations impending disaster and anxiety, as well inappropriate familiarity or unfamiliarity (déjà vu, jamais vu). Some patients have episodes of depersonalization, dream-like states, or sensations like those occurring in alcoholic intoxication. Metamorphopsias include visual distortions such as macropsia and micropsia. Auditory distortions, olfactory and gustatory hallucinations, abdominal symptoms (“butterflies in the stomach,” epigastric rising sensation) are some of the more common sensory experiences. Consciousness may be partially lost during complex partial seizures.
P.6
To the extent that consciousness is impaired, the seizures are really generalized but the use of the term “partial” has been retained, perhaps inappropriately in some cases.
Automatisms
The automatisms of complex partial seizures are much more difficult to recognize as ictal events than the tonic-clonic stages of grand mat seizures. They tend to be repetitive and often consist of such oral activities as lip-smacking, chewing, gagging, retching, or swallowing. Some patients may perform a variety of complicated acts that seem to blend with normal behavior. Usually the behavior is inappropriate. The repetition of a phrase over and over again the buttoning and unbuttoning of clothing are common. A few patients may assume bizarre positions resembling those of catatonic schizophrenia. These positions are held for varying periods of time. Some patients have fugue states. Fortunately, outbursts of directed, aggressive behavior are extremely rare, but when these outbursts lead to violence, they often present neurologists and psychiatrists with difficult medicolegal questions concerning the responsibility of an individual for his actions. Though the behavior of patients during complex partial seizures tends to be automatic, it may be influenced by environmental factors. Such an influence can be seen in the case of a patient who had a complex partial seizure while he was waiting in line at the hospital pharmacy to have his anticonvulsant prescription filled. It was a hot day and he was frustrated by the long wait. During the seizure, or possibly during postictal confusional period following it, he shoved past the people who were in front of him until he came to the pharmacist's window where he stood, mumbling incoherently, until his wife led him away. After a minute or two of confusion, his sensorium cleared.
Sometimes it is quite difficult to distinguish behavior caused by complex partial seizures from episodic aberrations caused by postictal confusion, intoxication, dissociation, anxiety, obsessive-compulsive disorder, pavor nocturnus, narcolepsy, migraine, transient global amnesia, depression, panic disorder, delerium, postconcussion state, conversion disorder, or psychosis. To distinguish epilepsy from these, several questions regarding the patient's history are helpful: (1) Does the patient have a history of generalized seizures? (2) Is there positive family history of epilepsy? (3) Is there a past history events known to be associated with seizures such as brain injury? (4) Does the patient describe subjective alterations typical of complex partialseizures? (5) Has he been observed performing any of the characteristic automatisms? (6) Was he confused during the episode? (7) Is the patient's memory for events that occurred during the episode impaired? Though memory for the early events of the seizure may seem relatively well preserved, in epilepsy such memories are nearly always incomplete or incorrect. (8) Did the patient experience a postictal depression?
P.7
Though called depression the term describes sensorium, not just mood. Postictal depression is almost always present following a partial complex seizure of temporal lobe origin. It may be mild, and close questioning can be necessary to elicit a positive history. After the seizure, this depression may be manifest as only a brief period of fatigue, when the patient may wish to lie down, perhaps complaining of a headache or of not feeling quite normal. (9) Has he had other lapses during which he engaged in nearly identical behavior? Motor activity tends to be stereotyped in complex partial epilepsy. (10) Was the episode brief, lasting seconds to minutes? Although longer episodes can occur in epilepsy, they are unusual. In rare cases of complex partial epilepsy, there may be prolonged episodes of abnormal behavior lasting for hours or days, though epilepsy is very rarely manifested solely as a prolonged behavioral disturbance. When it is, it tends to be of frontal lobe origin. When considering episodic and especially patterned behavioral abnormalities, even when they are prolonged, the clinician should include some form of associated seizure disorder in his differential diagnosis. It can be extremely difficult to distinguish dissociative episodes from true complex partial seizures.
In addition to the preceding 10 features that may help in establishing a diagnosis of complex partial epilepsy, the EEG and the response to anticonvulsant therapy may be useful. The EEG may help to confirm the suspicion that an active seizure state exists. It also helps determine whether a disorder is diffuse, lateralized, or focal in origin. The characteristic interictal abnormality complex partial seizures is an anterior temporal spike focus. Repeating the test during sleep or using nasopharyngeal sphenoidal leads will reveal abnormalities in some patients, but the incidence of normal interictal records remains relatively high.
Depending on the criteria used for diagnosis of epilepsy, normal sleep records may be seen in roughly one-third of epileptic patients. For this reason, activating agents such as metrazole and megimide have been used to induce electroencephalographic changes in patients suspected of being epileptic. Though this is often successful, results must be interpreted cautiously, given that these drugs, even in low doses, may induce epileptiform changes some nonepileptic persons. These chemical activation methods are rarely used anymore.
If a normal interictal EEG does not rule out complex partial epilepsy, neither can the diagnosis of epilepsy be sustained merely on the basis an abnormal record, given that 10 to 15 percent of the general population have abnormal EEGs. These abnormalities, however, are usually just slowing and are not of a paroxysmal or an epileptiform type (see pp. 265–266). Prolonged EEG recording is now widely used to try and capture a complex partial seizure (Bridges and Ebersole, 1985). If a seizure occurs during the recording the seizure causes a change in the patient's state of consciousness, epileptiform abnormalities are almost always revealed. If no electrical epileptiform activity
P.8
accompanies the episode, episode is probably not epileptic and may be a pseudoseizure.
The use of a pharmacological response to establish a clinical diagnosis is difficult at best, and often impossible, yet a trial of anticonvulsant therapy, starting with phenytoin, carbamazepine, or valproic acid may sometimes provide useful diagnostic information. In such therapeutic trials, increasing doses should be given until either the seizures stop or signs of an overdose develop. The determination of the serum concentration anticonvulsants is a useful guide to the adequacy of a therapeutic trial. Nystagmus is usually the first sign toxicity with phenytoin but not always. Other signs of toxicity include slurred speech, ataxia, lethargy, difficulty concentrating, and dysmnesia.
Complex Partial Seizures—Extratemporal Origin
Complex partial seizures of extratemporal origin are even more frequently misdiagnosed than those of temporal origin. Common reasons for diagnostic errors are failure to recognize the epileptic origin of the attacks or to appreciate localizing clinical seizure characteristics. Complex partial seizures of frontal lobe origin are brief, frequent attacks that begin and end suddenly. Complex motor automatisms may strongly suggest pseudoseizures. There be kicking and thrashing, pelvic thrusting, masturbatory actions, raucous shouting, and verbalization. Episodes of complex partial frontal status epilepticus frequently develop during which the bizarre manifestations continue for hours and even days. The automatisms may be semipurposeful or purposeful. Interictal and sometimes ictal EEGs are often normal or nonspecifically abnormal. The stereotypy of the attacks suggests the diagnosis that may only be made with certainty in some cases by depth electrode study (Spencer, 1981, 1983; Spencer et al., 1982; Williamson and Spencer, 1985, 1986; Williamson et al., 1985a). Thomas et al. (1999) observed 10 patients with nonconvulsive status epilepticus of frontal lobe origin. They noted that patients with unilateral foci manifested euphoria, disinhibition, distractability, and decreased word fluency. All these patients were alert and oriented even if they had a brief motor convulsion. In contrast, the patients with bilateral foci appeared confused and disoriented during the episode.
Occipital seizures can mimic those of temporal lobe origin. Scalp EEGs are often misleading. The initial clinical symptoms are the most important clue to correct diagnosis of occipital seizures and include elemental visual symptoms (e.g., colors, lights), visual loss, the sensation of the eyes being pulled or sensations of movement in the absence detectable movement, rapid forced
P.9
eye blinking or eye flutter, and contralateral eye deviation (Williamson and Spencer, 1986; Williamson et al., 1992).
Electroencephalography and the Diagnosis of Epilepsy
The term epileptiform has been applied to any paroxysmal discharge recorded on EEG that contains spikes or sharp waves, either localized generalized. Epileptiform discharges are usually associated with epilepsy, but it is also possible to have an epileptiform EEG and not epilepsy (Peterson et al., 1968; Trojaborg, 1968; Zivin and Ajmone-Marsan, 1968).
If an epileptiform EEG cannot fully establish the diagnosis of epilepsy, neither does the finding of a normal interictal EEG rule out the possibility epilepsy. Ajmone-Marsan and Zivin (1970) found epileptiform activity in only 56 percent of epileptics. In subsequent recordings an additional 26 percent of patients showed such activity. Fully 18 percent of the patients had at least three consecutively negative recordings and only 30 percent had epileptiform activity in all recordings.
The age of the patient at the time of EEG and at the onset seizures influenced the incidence of epileptiform tracings; these are much more common in children than adult epileptics. The type of epilepsy influenced the rate positivity. Epileptiform EEGs were seen in almost 95 percent of patients with “absence” attacks. However, the authors of this important study failed to state how they established the diagnosis of epilepsy and this failure probably underlies one of their more surprising findings: that 98 percent patients with complex partial seizures had an epileptiform EEG and, therefore, the authors claimed that complex partial seizures correlate better with the EEG than any other clinical seizure type! This conclusion is probably unwarranted.
Even though a high correlation between complex partial epilepsy and positive EEG has been confirmed many times (Gibbs and Gibbs, 1952;Glaser, 1967; Currie et al., 1971), this finding flies in the face of the common clinical experience that epileptiform EEG abnormalities in complex partial seizures are discovered less often than in patients with most other clinical types of epilepsy. Since the diagnosis of epilepsy is very much dependent on the EEG, high correlation of epileptiform patterns with complex partial seizures in these reports may reflect the great difficulty that clinicians face in establishing this diagnosis solely on the basis of history and observations. Clinicians are often dependent on the EEG for a positive diagnosis of complex partial epilepsy.
When consciousness is not impaired during a seizure, EEG abnormalities may not be present even during epileptic attacks, but the EEG provides a much better
P.10
reflection of epileptic activity when the seizure impairs consciousness. While a patient is in an altered state of consciousness, the EEG provides “gold standard” (albeit an imperfect standard) for differentiating true from pseudoseizures.
External scalp recordings during the seizure show epileptic activity in only 20 percent of the simple partial sensory seizures. In partial motor seizures, epileptic activity in scalp recordings during the seizure reveal epileptic abnormality in only 30 percent of the seizures (Devinsky et al., 1988). The site of origin of simple, partial seizures is the surface the brain. Engel (1986) said, “In temporal lobe epilepsy patients in whom we have data from depth electrodes, probably 90 percent have no scalp or sphenoidal changes at a time when the depth electrodes show discharges restricted to one hippocampus and the patient may be having some experiential phenomena.” Smells, tastes, and odd sensations are experiential phenomena. Confusion and loss of consciousness indicate a degree of epileptic spread that can virtually always be detected with an EEG.
Sphenoidal leads give essentially the same information that nasopharyngeal leads are supposed to provide, but with much less of a problem caused by artifactual activity. The difficulty of positioning the electrodes and the need for a surgical procedure to insert them has resulted in their use only a few specialized clinics.
Sleep activation is least complicated in its interpretation when natural sleep is recorded. Many laboratories encourage a patient to drift off into sleep after obtaining an initial recording in the waking state. Most EEG activation occurs in the drowsiness or slow-wave phases during which it appears that there is enhanced synchrony that results from a lower cerebral resistance to the synaptic spread of convulsive potentials. Thus, in slow-wave sleep, discharges that are limited during wakefulness to deep structures such as the limbic structures now spread to cortical regions.
Sleep deprivation per se provides an additional degree of activation compared with sleep unrelated to deprivation (Mattson et al., 1965). For this reason sleep records are usually done after a period of sleep deprivation.
The sampling problem inherent in any standard electroencephalography lasting 45 to 60 minutes is obvious. For this reason 24-hour ambulatory cassette EEGs have been introduced. This is the EEG equivalent of Holter monitoring for cardiac arrhythmias; and it has proven to be quite useful, nearly doubling the incidence of epileptiform abnormalities as compared with standard EEGs in epileptic patients. In the initial tests of cassette technique, the patients studied were known to have some epileptiform features by means of the much more elaborate cable telemetry with video recording of the patient and his EEG for 24 to 72 hours (Ebersole and Leroy, 1983a, b).
Liporace et al. (1998) compared the usefulness of a sleep deprived EEG of ordinary duration with a computer-assisted, 16-channel, 24-hour ambulatory
P.11
EEG in 46 patients with historical information consistent with epilepsy but whose initial screening EEG had been normal or nondiagnostic. Both the sleep deprived and ambulatory EEGs improved detection of epileptiform discharges by a similar amount (24% vs. 33%) but the ambulatory EEG was superior because it captured actual seizures in 15 percent of the patients that the sleep deprived EEG had missed.
Sedatives should not be used to induce sleep as a means of uncovering an epileptic focus. They introduce artifacts and some may suppress epileptic discharges.
The EEG manifestations of complex partial seizures may be minimal or not pathognomonic, and even with telemetry cassette monitoring it may be difficult to distinguish seizures from pseudoseizures. The postictal elevation of serum prolactin promised to provide a biochemical marker of complex partial seizures as it does for generalized tonic-clonic seizures (Prichard et al., 1983, Lin et al., 1997) but the prolactin level does not rise following most complex partial seizures (Alving, 1998) and it may be elevated in some patients with pseudoseizures (Collins et al., 1983). Even when seizures elevate serum prolactin leads, the prolactin response attenuates with repeated seizures (Malkowicz et al., 1995). These considerations and the long delay in obtaining results have limited the usefulness of serum prolactin as a diagnostic tool in epilepsy.
A variety of new techniques have entered neurologic practice that can be useful as adjuncts to the EEG in the diagnosis of epilepsy. For the most part these techniques cannot establish or rule out the diagnosis of epilepsy if the EEG is normal but they can all be helpful in localizing the possible sites of origin of epileptic discharges. These include computerized tomography (CT), magnetic resonance imaging (MRI), and the dynamic, functional techniques: single photon emission computerized tomography (SPECT), positron emission tomography (PET), functional MRI (fMRI), and nuclear magnetic resonance (NMR) techniques.
Computerized tomography and MRI are static techniques. They provide information about the structure of the brain. Computerized tomography utilizes X-rays and is especially sensitive in identifying calcification hemorrhage but is not sensitive for distinguishing tumors and scars from normal brain. Magnetic resonance imaging uses a magnet that energizes the hydrogen atoms in water and detects the wave of energy derived from them. The digitalization the energy wave provides an image.
These two static techniques produce pictures of the brain as if it were a tomato that has been sliced and each slice is photographed. Magnetic resonance imaging is superior for identifying tumors, strokes, and scars for this reason it is preferred for discovering the cause of epilepsy. For instance, MRI showed epileptogenic lesions in 38 of 300 consecutive patients with initial seizures, 17 of whom had tumors (King et al., 1998). Fifty of 250 patients who were evaluated
P.12
for intractable partial seizures were shown to have space-occupying lesions with MRI, 35 of which were neoplastic (Boon et al., 1991). Volumetric analysis of MRI identified the atrophic, epileptic hippocampus in 17 of 110 cases of chronic temporal lobe epilepsy (Watson et al., 1997). Even in this specialized group of epileptics evaluated for surgery, 29 of 110 had primary generalized epilepsy. Magnetic resonance imaging findings in 26 percent of 341 patients with chronic, refractory epilepsy were normal (Li et al., 1995).
Without wishing to denigrate the important role of neuroimaging in diagnosing surgically treatable epilepsy, we want to emphasize the fact that epileptics, usually after their first seizure and often years of intractable focal seizures, have normal MRI imaging studies. For this reason, dynamic imaging techniques have been tried to localize the site of origin the epileptic discharges.
Of the techniques that provide an index of metabolic activity the brain, two have proven to be clinically helpful in epilepsy: PET and SPECT. These have shown that during the interictal period there is a reduction in metabolic activity in the region of focus. During ictus, metabolic activity becomes intense in the focus. A zone of interictal hypometabolism is commonly seen patients with complex partial seizures of medial temporal origin but is less characteristic of seizures originating elsewhere in the brain. The epileptogenicity hypometabolic zones must be proven electrographically. A hypometabolic zone on PET or SPECT without electroencephalographic definition does not support a diagnosis of epilepsy (Engel, 1991).
It is difficult to obtain PET scans and interpret the results during spontaneous partial seizures. Ictal scans can be more easily obtained with SPECT. Single photon emission computed tomography has been used during both intracranial and extracranial scalp EEG recording and has been found to be more accurate during than between seizures in localizing the source of epilepsy. During seizures, blood perfusion (measured by SPECT) increases. Between seizures, blood perfusion decreases focally (Boundy et al., 1996; Spanaki 1999).
Abnormal increases and decreases in perfusion lack the specificity of EEG spikes. Not all focal changes in blood perfusion are caused by epilepsy. Regional increases and decreases in perfusion cerebral activity can accompany the use of the brain.
Dynamic imaging tests have provided a means of studying neural systems that subserve a variety of cortical functions and show great promise for illuminating brain-behavior relationships. For example, dyslexics show deficient phonological task-related activity in the left posterior superior temporal gyrus, angular gyrus, and extrastriate cortex. These regions show increased activity during phonological activity in normals. Dyslexics show increased phonological task-related activity in the left inferior frontal gyrus, whereas normals do not. The increased activity in the frontal gyrus during testing dyslexics may be a
P.13
compensatory response to the failure of more posterior areas (Shaywitz et al., 1998). Though such use-related hyperactivity in the frontal lobes dyslexics is abnormal, it is not caused by epilepsy. Schorner et al. (1987), compared the clinical usefulness of MRI, PET, and SPECT techniques in localizing lesions.
Abnormal Electroencephalogram Without Epilepsy
During the laboratory evaluation of a psychiatric patient, it is not uncommon to find an abnormal EEG when there is no evidence of overt seizures. As one might expect, the relationship between an abnormal EEG and behavior disturbance in nonepileptic patients can be difficult to define. All abnormal behavior comes from the brain. Disorders of behavior that are best handled by psychiatrists are generally not associated with EEG abnormalities. An abnormal is generally a hallmark of a neurologic condition. Many conditions can cause abnormal EEGs besides epilepsy, but if the pattern of abnormality suggests epilepsy, how sure can a clinician be that a patient with behavior disturbance and an abnormal EEG does not have a seizure disorder? The criteria for such a judgment are almost never explicitly stated.
The electroencephalographic abnormalities seen in some “psychopathic” patients with a history of aggressive behavior may possibly be secondary to the kind of brain damage that is properly the concern neurologists. There considerable evidence correlating electroencephalographic abnormalities with certain psychiatric symptoms (Tucker et al., 1965). This underscores the likelihood that neurological disease of the brain has behavioral consequences.
The incidence of EEG abnormality among “psychopaths” and children with serious behavior disorders is over 50 percent. In schizophrenic patients, the incidence is 24 to 40 percent (Ellingson, 1955; Hill, 1963; Williams, 1969). This indicates that there are neurologic components of these conditions. The diagnosis of epilepsy or psychosis cannot be made on the basis of electroencephalographic criteria alone, and the meaning of electroencephalographic abnormalities in the absence of seizures or clinical symptoms cannot always be determined.
Pseudoseizures
The differentiation of hysterical seizures and true epilepsy is very difficult. Despite common wisdom to the effect that tongue biting, incontinence, injuries, and seizures during sleep do not occur in pseudo-grand mal seizures, there are no firm clinical criteria for the diagnosis of hysterical seizures. Self-harm may
P.14
occur with hysterical attacks (Ferriss, 1972; Rossen, 1974; Standage, 1975; Cohen and Suter, 1982) and urinary incontinence is common (Freud, 1949; Riley and Brannon, 1980).
Cohen and Suter (1982) found that of 51 patients with pseudoseizures, two reported self-injury, 13 reported urinary incontinence, 12 tongue lacerations, and 2 reported spells in sleep. The diagnosis of hysterical seizures was established by initiating and terminating an attack with suggestion saline injection during EEG monitoring without any change in the record during the attack. This activation test induced urinary incontinence twice and tongue laceration once.
Using an EEG technique with suggestion to induce seizures, Lesser and his colleagues (1983) found that 50 of 79 patients referred to the Cleveland Clinic for differentiation between psychogenic and epileptic attacks had pseudoseizures. Five (10%) of these had both psychogenic seizures and true epilepsy. Eight patients had epileptiform records but either no psychogenic seizure was recorded or a definite decision could not be made. Among patients with known epilepsy, between 10 and 20 percent have experienced psychogenic seizures (Desai et al., 1979; Ramani 1980; King 1982).
Devinsky et al. (1996) found that of 387 consecutive admissions to an epilepsy center for evaluation, 25 percent of the patients had nonepileptic (pseudo) seizures and 20 percent of these patients had true epilepsy. In both the epileptics who had no pseudoseizures and the epileptics who had pseudoseizures, convulsive movements, staring, and automatisms were the most common manifestations. The nonepileptic seizure patients' convulsive movements lasted longer (5 min vs. 80 sec); they showed shaking without loss of consciousness, apparent blackouts, stiffening, and loss of consciousness without EEG changes; they were usually alert or slightly drowsy after the event whereas the epilepsy patients were usually more lethargic. The authors stressed that when a patient with history of well-characterized seizure patterns begins to have different types of phenomena, one should suspect pseudoseizures but the EEG was the “gold standard” for diagnosis. The clinical state during ictus, the psychiatric state of the patient and/or the history of abuse in childhood fail to differentiate epileptic from pseudoseizures. The clinical neurologic observations (without EEG) during spells were often misleading. Several of the patients with pseudoseizures showed clinical characteristics that were identical to the epilepsy patients. Various provocative tests e.g., saline injection (Ney et al., 1996) or stress interviews (Cohen and Howard, 1991) have been used to diagnose pseudoseizures too, but, as with placebo interventions, there are always some false positives and some negatives.
The history of sexual abuse in childhood is rife among patients with pseudoseizures but is also common among patients with true complex partial seizures (Alper et al., 1997). Physical and sexual abuse may be more likely to be directed
P.15
at damaged children. Serious physical abuse of children that can cause brain damage and can generate epilepsy can start very early in life. It has been documented by videotape at 3 months of age (Southall et al., 1997). Thus, true epilepsy and the experience of child abuse can go together.
Arnold and Privitera (1996) confirmed that the history of early emotional trauma did not distinguish between true and pseudoseizures. They found no difference in DSM-IIIR diagnoses between patients with pseudoseizures and those with documented epilepsy. The main psychiatric diagnoses in both groups were major depression, post-traumatic stress disorder (PTSD), alcohol dependence, and panic disorder.
Total dependence on the EEG for differentiating pseudo and true seizures has been criticized. Using long-term video EEG monitoring, Henry and Drury (1997) studied 145 epileptic patients, each of whom had interictal EEG spikes. There was no correlation between clinical seizure phenomena and epileptic EEG activity during what appeared to be real clinical seizures in 8 percent. The normal EEGs in these cases were false negatives.
Pseudoseizures are usually manifestations of dissociation. Dissociative disorders are under diagnosed. In prior editions of this book, for example, we did not even use the term dissociation. Yet dissociative disorders may comprise 10 percent of psychiatric populations (Coons, 1998). A very high proportion patients with pseudoseizures attempt suicide (Roy, 1979), so the diagnosis of pseudoseizures should raise the question of depression. This possibility should be evaluated and depression treated, if present.
Because of the difficulty in establishing a diagnosis dissociation, number of questionnaires have been developed to identify dissociative episodes, one of which is the Dissociative Experiences Scale (DES). One hundred thirty-two patients with pseudoseizures were compared sex and age-matched patients with complex partial seizures. The DES did not distinguish the two groups. depersonalization/derealization factor of the DES was elevated in patients who reported childhood abuse. Though some of these had pseudoseizures, others complex partial epilepsy (Alper et al., 1997).
Why do some abused children develop dissociative conversion reactions, including pseudoseizures, and others not? Good (1993) has pointed out that dissociative symptoms and disorders (amnesia, fugue, depersonalization, multiple personality, automatisms, and furors) can be induced by a variety of medications, drugs of abuse, medical illnesses, and other conditions that affect cerebral function. All of these are the result of abnormal cerebral activity. Past experiences, learning, and recall also affect cognition.
There may be a great similarity in the pathogenesis of dissociative symptoms related to abuse and epilepsy in that dissociative symptoms are the result of memories, permanently stored in the brain that “boil up” at times to recruit and dominate the functions of the cortex (Teicher et al., 1993). Similarly, but by a
P.16
different electrophysiological mechanism, an epileptic focus spreads to recruit into its abnormal rhythms the activity of normal brain cells. Both epilepsy and dissociation result from permanent conditions of the brain that intermittently cause clinical syndromes. Though they are not identical disorders, anticonvulsant medications can have a stabilizing influence on both, especially when affective symptoms are also present (Tucker, 1998). Many anticonvulsants excellent mood stabilizers. These include carbamazapine, valproic acid, lamotrigine, to- piramate, and gabapentin (GHACMI, 1998). Yet lithium, the major mood stabilizer, must be used cautiously in epileptics as it can cause cognitive impairment, worsening of epilepsy, and aggressiveness (Schiff et al., 1982).
There is evidence that mood disorders, sexual abuse, and pseudoseizures are linked. Wyllie et al. (1999) performed psychiatric evaluation on 34 children with pseudoseizures (average age 14 years). Eleven (32%) had mood disorders and two others (11%) had psychosis. Eleven patients (32%) admitted to having suffered sexual abuse and this was especially frequent (64%) in the subgroup with mood disorders.
It is tempting to speculate that the sexual abuse of a child with a mood disorder is more likely to give rise pseudoseizures. The special vulnerability of some abused children to the development of pseudoseizures could be explained if the experience of abuse and mood disorder interact to produce pseudoseizures. The beneficial effect of some anticonvulsant mood stabilizers on mood disorders would also explain why anticonvulsant mood stabilizers are beneficial in some patients with pseudoseizures.
Therapy
The major principles of therapy for epilepsy are simple. Initially, a single drug should be used at a moderate dosage. The dosage should then increased until either the seizures are controlled or signs of toxicity appear. If it is not possible to control seizures at nontoxic doses, a second drug should be added. When epilepsy is not controlled by two antiepileptic drugs (AEDs) prescribed together in therapeutic doses, it is very unlikely that any combination of AEDs will provide complete control. It is probably better to recognize this fact than prescribe excessive numbers of AEDs at toxic doses (Kwan and Brodie, 2000).
The development and utilization of tests to determine the blood levels the various medications, particularly phenytoin, carbamazapine, valproate, phenobarbital, primidone, and ethosuximide has mitigated this problem. The determination of serum levels has been great help in determining whether a patient is taking his medicine as prescribed and whether more or less need be administered. Aside from the usual symptoms of toxicity with phenytoin (cerebellar symptoms, nystagmus, ataxia, sedation, euphoria) when the serum level of the
P.17
drug is above 20 µg/ml, various psychiatric symptoms such as inattentiveness, confusion, depression, and psychosis may develop. In some instances, seizures increase in frequency. Milder behavioral symptoms that are commonly described by patients with high phenytoin blood levels include feelings of lower energy levels and initiative decreased sociability ability to concentrate.
Occasionally, complete seizure control is not possible without some degree of toxicity. In such cases, the functional capacity of the patient should determine what an acceptable degree of seizure control is. Among adult epileptics, grand mal seizures are generally the easiest to control, and complex partial seizures are the most difficult.
Between the introduction of phenytoin (Dilantin) in the 1930s and the 1980s, only a few new drugs, carbamazepine (Tegretol), valproic acid (Depakote), and clonazepam (Klonapin), became available for the treatment of generalized motor, focal, and complex partial seizures. (Smith et al., 1987) Carbamazepine has essentially the same spectrum of clinical usefulness as phenytoin, and it also has very similar side effects (Troupin, 1976). Dipropylacetate (valproic acid, Depakote) is a universally effective anticonvulsant, most useful for generalized seizures (mainly uncomplicated tonic-clonic spells, absence, and myoclonic). It has a long track record and has, during more than 2 decades of use, achieved a wide vogue as a first-line drug. (Delgado-Escueta et al., 1983; Wilder 1983; Mattson et al., 1992).
Clonazepam, a benzodiazepine, is helpful in myoclonic seizures. It has broader efficacy, but sedation is common at therapeutic levels and therefore markedly limits its usefulness.
Several new anticonvulsants are now available. These include lamotrigine (Lamictal) (Ducac and Kamins, 1997), gabapentin (Neurontin), felbamate (Felbatol), topiramate (Topamax), and tiagabine (Gabatril). These are expensive drugs and marketing efforts have often emphasized that it is not necessary to obtain their blood levels, thus lowering the cost of use.
The claim by pharmaceutical companies that their new anticonvulsants do not require blood level monitoring may be behind the fact that there are few published therapeutic ranges for the newer drugs. Without knowing the blood levels, it is not possible to determine if a patient with poor seizure control has taken his or her medication. The introduction of these agents has been a major advance in the control of epilepsy. However, lack blood level criteria and the cost of these drugs have been impediments to their use.
Blood Levels
Understanding of pharmacokinetics has tremendously improved our ability to use the older anticonvulsant drugs. This was of great practical value to
P.18
practicing physician. Buchthal et al. (1960) and other researchers demonstrated that a good correlation exists between serum levels of phenytoin and phenobarbital, seizure control, and symptoms of overdose. The capacity to determine serum levels of these older anticonvulsants represents, in our opinion, the single most important advance in seizure therapy since the introduction of phenytoin.
When serum phenytoin and phenobarbital levels were first measured in the Epilepsy Center at Yale, almost half of the patients had levels that were below the therapeutic range, though all patients were prescribed what was considered an adequate dose. Noncompliance was the main reason for this, but serial monitoring of serum levels was of great help in dealing with the problem. Over course of 2 years, 48 patients followed at the Epilepsy Center experienced a steadily decreasing number of seizures when their anticonvulsant doses were adjusted to yield therapeutic levels in their sera, with the frequency of seizures per year in the group falling from 210 to 75 (McElligott, 1974).
In some instances, individual differences in drug metabolism explain an inadequate amount of medication. The standard dosage phenytoin is usually 300 mg/day, but this is often too low. In rare cases, 500 mg is too low, and in two patients whom we have seen personally, 700 mg was necessary to produce serum levels of 10 to 20 µg/ml. In addition, other drugs can affect anticonvulsant blood levels; isoniazide, chloramphenicol, dicumerol, disulfiram, or sulthiame, for example, can markedly increase phenytoin levels (Goodman et al., 1990). Anticonvulsant agents such as carbamazepine and valproic acid can lower phenytoin serum levels (Mattson, 1976). Other anticonvulsants interact as well. Lamotrigine, for example, must be used cautiously with valproic acid as it can raise the blood levels of valproate to the toxic range.
It has been the practice of most physicians to prescribe anticonvulsants in divided doses, and patients have been advised to take their medications three and even four times a day. Pharmacokinetic analysis based on blood levels indicates that this is usually an irrational way of prescribing medication in adults (though children may require it) and often works to the detriment of seizure control given that patients are more likely to forget their medication when it must be taken so frequently (Cramer et al., 1989). In addition, midday doses can be inconvenient and embarrassing for adults at work for children school. The half-life of an anticonvulsant is the time it takes for the serum level of an anticonvulsant to drop half its steady-state level after it has been discontinued (see Table 1-1). Drugs with half-lives of more than 20 hours seldom need to be given more than once a day. As can be seen in Table 1-1, phenytoin, phenobarbital, and ethosuximide have half-lives of over 24 hours. A long acting form of carbamazapine (Carbatrol) is available and can be given at 12-hour intervals. Gabapentin and valproic acid require more frequent dosing.
Anticonvulsants differ markedly in their effect on behavioral functions. There
P.19
are many anticonvulsants that are useful in psychiatric practice as mood stabilizers. Some of these approach the efficacy of lithium in treating affective psychosis. These drugs are preferred in patients who suffer both epilepsy and emotional disorders. Other anticonvulsants that can be extremely effective in controlling epilepsy have been reported to cause dysphoric states and even frank psychosis. These drugs are to be avoided in epileptics who have concomitant emotional disorders (Tables 1-2a and 1-2b).
|
Table 1-1 Commonly Used Anticonvulsants |
||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||
Measurements of blood levels have also provided information on how to begin therapy rationally. For some time it has been known that, when phenytoin is started orally, it takes 5 to 10 days for the blood level to rise the therapeutic
P.20
P.21
P.22
range. When a patient is given 1 to 2 mg/kg of phenobarbital orally, it takes almost 3 weeks to achieve steady-state levels in the therapeutic range. This takes 4 days when double the ordinary dose is given (Aird and Woodbury, 1974). To achieve therapeutic levels even more rapidly, 10 mg/kg of phenytoin or phenobarbital administered orally results in levels of 15 to 30 µg/ml within a few hours. Given intravenously it takes minutes. Though all anticonvulsant drugs may cause some degree of sedation at therapeutic levels, only the barbiturates and clonazepam regularly induce it.
|
Table 1-2a Antiepileptic Drugs That May Be Helpful in Dysphoria/Psychosis |
||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||
|
Table 1-2b Antiepileptic Drugs That May Worsen or Cause Dysphoria/Psychosis |
||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||
Surgery in Epilepsy
There are very few seizure disorders that will not respond to some combination of medications, but a small number of unfortunate individuals suffer from severe seizures that cannot be controlled by medical therapy. For such cases, there are several major surgical approaches, and they all seem to be effective in certain cases. Engel (1996) reported on the surgical results for refractory, partial epilepsy on 3579 patients from 100 specialized surgery-epilepsy centers (1986–1990) 69 percent were seizure free and 22 improved after anterior temporal lobectomy.
Anterior temporal lobectomy usually involves removing the amygdala, anterior part of the hippocampus, entorhinal cortex, and a small portion temporal pole, leaving the lateral temporal cortex intact. Although other types of surgery for epilepsy are less frequent, the success rates lesionectomy (n = 293) and hemispherectomy (n = 190) were similar. Corpus callosal sections — 563) led to improvement in 60 percent but only 7.6 became seizure free. Callosal section was particularly effective for drop attacks but not for generalized or partial seizures.
In general, epilepsy surgery has produced impressive results and it represents an important intervention for a small number of patients who have not responded to anticonvulsants. With the growth of specialized multidisciplinary centers for the treatment and study of epilepsy has come an increasing sophistication in the evaluation and selection of appropriate patients for surgery. Routinely, the evaluation for surgery now includes the following very costly testing: continuous video/EEG monitoring, MRI, often specialized MRI studies and volumetric analysis, PET and ictal SPECT, neuropsychological testing, and carotid amobarbital injection to determine hemispheric dominance (Polkey, 1993).
The reason for such extensive testing is that the site of origin the seizures is difficult to pinpoint. Berkovic et al. (1995) claimed that if the MRI scan showed a visible lesion, removing the lesion provided good control of epilepsy. Careful EEG, SPECT, and other studies might not be necessary. The MRI cannot be relied upon exclusively, however, because seizures do not always arise in or
P.23
immediately adjacent to the lesion. Still, patients with a discrete lesion are the most likely to benefit from surgical intervention.
Another surgical approach proposes performing en bloc anterior temporal lobectomy as seizures so frequently seem to either arise there or spread the anterior temporal region (including the hippocampus) before a complex partial seizure is clinically apparent. Recent results for anterior temporal lobectomy indicate an 80 percent reduction of seizure frequency (Sperling, 1996). Engel (1996) described a syndrome of Mesial Temporal-Lobe epilepsy, which consists of early age of onset, increased incidence of febrile convulsion, and an increased incidence of Wada test memory asymmetries. Hermann et al. (1997) studied 107 patients with mesial temporal sclerosis and found that they showed generalized cognitive impairment. Verbal memory was impaired in those with left side lesions, but attention and executive functions remained intact.
It is clear that the more discrete lesion, better outcome from surgery. Eliashav (1997) studied 60 patients who had anterior temporal lobe resection for intractable seizures because of glial tumors, hamartomas, or vascular malformations (followed for a mean period of 8.4 years). Of the patients, 80 percent became seizure free. Only three patients had a late recurrence of seizures. A prolonged history of seizures was associated with a poorer outcome. The tumor reoccurrence rate was only 3.3 percent. At follow-up, 67 percent were working but 59 percent noted some form of psychosocial improvement. Similar good results have been reported for resection of discrete frontal lobe lesions with a 67 percent seizure-free rate following surgery (Laskowitz et al., 1995). If the seizure focus is not localized, surgery less successful. There are few contraindications to surgery though some centers are unwilling operate on patients with a history of psychosis or low IQ. Most the reports excellent results represent the conclusions of the groups who have performed evaluations that preceded surgery, but a recent randomized case-controlled study of 80 patients with intractable partial seizures was reported. Half of the had surgery, the other half were followed for 1 year with only medical treatment. Fifty-eight percent of the operated patients were seizure-free as compared with only 8 percent in the medical group at end of a year (Wiebe and Blume, 2001).
Surprisingly little attention has been given to the psychiatric status of patients following epilepsy surgery. Blumer and his colleagues (1998)found that after surgery, 39 percent of 44 patients experienced either de novo psychiatric complications (6 psychotic, 6 dysphoric, 2 depressed) or exacerbation of a preoperative dysphoria (3 patients). This exceptional psychiatric morbidity had not been reported previously. Cognitive functioning assessed before and after surgery with neuropsychological testing is no substitute for psychiatric evaluation. More than half the readmissions to hospitals following epilepsy surgery are for psychiatric disorders (Wilson et al., 1999).
P.24
Alternative procedures may provide new therapeutic avenues for epileptics who have been unresponsive to medical therapy and for whom resective surgery is not feasible. These include bilateral stimulation of the centromedian thalamic nuclei (Velasco et al., 1995) and stimulation of the vagus nerve (Multicenter Study, 1995). Both of these procedures have improved generalized and partial seizures but the role of each in epilepsy control has yet to be fully explored. In the distant past, there were claims for cerebellar stimulation the purpose of seizure control. Though some dramatic cases of improvement were reported (Cooper et al., 1973), overall results were not favorable and cerebellar stimulation has been abandoned.
Epileptogenesis, Cognitive, and Behavioral Deterioration
Epilepsy has been inaccurately blamed for many of the problems that can develop in epileptics, including dementia. It is true that epilepsy can cause deterioration somewhat indirectly: traumatic brain injuries sustained during major motor seizures can cause intellectual losses (Trimble and Cull, 1989) the drugs that are used to treat epilepsy can depress cognitive function, usually reversibly. Neither of these is especially common. Postictal depression can look like dementia if it occurs often enough. This form of cognitive decline is linked to seizure frequency and responds to control as does nonconvulsive status epilepticus, a rare form of temporary confusion.
There is strong clinical evidence that seizures themselves do not cause deterioration: febrile seizures and electroconvulsive therapy (ECT) are benign. Even though Schiottz-Christensen and Bruhn (1973) showed that there was a performance IQ difference of 7 points on the Wechsler Intelligence Scale between the febrile seizure and control groups, in such a study, it is difficult to determine if brain damage is a cause or result of febrile seizure. All the evidence that seems to support the hypothesis that febrile seizures are harmful the development of the brain do not distinguish preexistant brain damage as cause epilepsy and intellectual deficit from epilepsy as the cause of and brain damage.
There is overwhelming evidence that febrile seizures are benign. Hauser et al. (1977) followed 657 patients with fever-induced seizures for more than 8000 person-years to assess risks for subsequent afebrile seizures. When patients with profound, preexisting neurological deficits were excluded, only 3 percent of the remaining 632 patients developed recurrent afebrile seizures. The characteristics of the febrile seizures that increased risk of subsequent afebrile seizures were those that suggest that brain damage had preceded the initial febrile seizure: the presence of focal features and the prolonged duration the seizure.
P.25
Thus, this study casts doubt upon the epileptogenicity of brief, generalized febrile seizures that occur in children who are otherwise normal.
Nelson and Ellenberg (1976) reported that recurrent nonfebrile epileptic seizures had occurred by 7 years of age in only 2 percent 1706 children who had experienced at least one febrile seizure. Of those whose prior neurologic or developmental status was abnormal and whose first seizure longer than 15 minutes, multiple or focal epilepsy developed at a rate of 9.2 percent as compared with 0.5 percent in a control group without febrile seizures. Of those whose preseizure neurological status was normal, epilepsy developed in only 1.1 percent, which is only slightly greater than that for children with no febrile seizures. Farwell, et al. (1994) and Berg (1997) have reemphasized the benignity of uncomplicated febrile seizures.
The overwhelming evidence that febrile seizures do not cause cognitive or behavioral problems (Nelson and Ellenberg, 1977), must be added to those reports that question the harmful effect of febrile seizures on inducing subsequent nonfebrile seizures (epilepsy).
Behavioral disturbances and learning disorders in epileptic children may occur, not from seizures but the use of antiepileptic medications.Farwell et al. (1990) have indicated that these disturbances may not be reversible. The question of whether, when, and how to treat febrile seizures remains unsettled. Why use potentially harmful drugs for a benign condition (Bourgeois, 1998)?
Repeated cerebral stimulation in animals can change the brain (Goddard et al., 1969) and provides a theoretical rationale for attempting prompt complete seizure control to prevent further epilepsy. Animals with electrodes implanted in their brains were stimulated briefly each day with subconvulsant currents. During the first week there was no behavioral change or EEG after-discharge but during the second week, stimulation produced minimal focal seizures and during the third week, bilateral seizures. Some animals developed spontaneous seizures. This is the kindling phenomenon and is thought to underlie the development of independent “mirror” foci on the previously normal side the brain opposite the primary focus, though not by everyone (Goldensohn, 1984).
The closest clinical approach to kindling is the use of ECT in mental illnesses.
Epilepsy seldom begins after ECT (114 per 100,000). Although this rate is greater than in an age-adjusted nonpsychiatric cohort, it probably can be explained by individual vulnerabilities in the psychiatric population and cannot be attributed to ECT (Devinsky and Duchowny, 1983).
There are benign forms of epilepsy that children “outgrow” whether or not they are treated with AEDs (Astradsson et al., 1998) and some of these seizures are focal (Holmes, 1993). Epilepsy is not a long-term problem in these familial conditions, thus proving that seizures may not be epileptogenic.
In summary, febrile seizures, ECT, and familial forms of focal epilepsy in
P.26
humans are not epileptogenic. There is little proof that the phenomenon of kindling exists in humans despite its demonstration animals and the occasional development of a mirror focus in humans. Perhaps the cause epilepsy and the status of the brain in which it occurs are more important than the seizure itself in determining future epileptogenicity.
A loss of neurons and glial scarring in the hippocampus, particularly Ammon's horn, has been described in the brains of epileptics. This related to anoxia and to a variety of behavioral cognitive abnormalities in epileptics (Margierson and Corsellis, 1966; Ounstead et al., 1966). If epilepsy gives rise to such lesions, why do not all epileptics demonstrate them? Most brains of epileptics are normal, microscopically.
To measure intellectual deterioration in epileptics it would be necessary to control many factors, such as the age of the patient at onset of seizures, the duration of epilepsy, the clinical form epilepsy (including etiology and presence or absence of brain damage), the presence other diseases, the social class and intelligence of the patient before disease onset, the drug used, its dosage and the blood level, the frequency of seizures, length time between the last seizure and the utilization of tests that reflect the functions the patient that appear to have been affected. It hardly need be said that the perfect study has yet to be done.
A prospective study of the stability IQ in 12 epileptic children tested within 2 weeks of the initial diagnosis and yearly for 4 years revealed no overall differences over time or in comparison with nonepileptic siblings. Eleven percent of the patients did experience a persistent drop in IQ of 10 points or more. In these, an early age of onset of seizures and the number drugs to which the patient became toxic best predicted changes in IQ. This suggested that total seizure control, especially in younger children, should not be achieved at the price of repeated episodes drug toxicity (Bourgeois et al., 1983).
The importance of AED toxicity in the pathogenesis behavioral disorders and poor seizure control in patients with intractable epilepsy was highlighted by a study of 69 such patients of whom more than half were benefitted by withdrawal of sedative-hypnotic AEDs with respect to symptoms drug toxicity and seizure control (Theodore and Porter, 1983). The improvement of patients with intractable epilepsy on tests of cognitive, perceptual, motor, and memory functions has been specifically related to withdrawal of barbiturates and an overall reduction in the number of AEDs but not to seizure frequency (Giordani et al., 1983). Armon et al. (1996) reported some disturbing but reversible cognitive changes in patients taking valproate for more than a year. Martin et al. (1999) reported that topiramate caused significant declines in attention and word fluency, but gabapentin lamotrigine did not. One problem in assessing the harmful effects of AEDs on cognition is that IQ tests do not assay frontally mediated functions and the batteries of psychological tests used
P.27
to establish the cognitive capacity of epileptics usually feature IQ and omit all tests of frontal function, except word fluency, which is often depressed.
On the Wechsler Adult Intelligence Scale (WAIS), most epileptic patients have scores in the normal range, but the scores do tend to cluster around the lower end of the range (Rodin, 1968). Epilepsy often occurs in adults and children with learning disorders (Branford et al., 1998a, b). In most cases, epilepsy and learning disorders arise independently from a brain lesion such as congenital cerebral dysgenesis (micro-or macroscopic) (Gressen, 1998). The seizures that derive from such static lesions can worsen over the years and may be resistant to AEDs (Branford et al., 1998a, b); however, the learning disorder is not attributable to epilepsy but to the brain condition that caused epilepsy.
The etiology of cognitive dysfunction in patients with seizure disorders will vary. It is also apparent that ordinary cognitive tests such as measures of IQ are not the best measures of function and quality life for seizure disorder patients. Perrine, et al. (1995) developed a quality of life scale for patients with seizure disorders and tested 257 patients. The patient's mood was the strongest predictor of the patient's quality of life, explaining 47 percent the variance. Other factors that were pertinent, but not as important, were psychomotor speed, verbal memory, and language ability. The comprehensive evaluation treatment of the patient with seizure disorders must take into account the patient's executive functioning and his emotional state.
There is no doubt that in a minority of epileptic patients, deterioration intellect does occur. It is most often encountered in patients with incompletely controlled seizures (Chaudry and Pond, 1961; Rodin, 1968) but there is no convincing evidence that epilepsy per se causes deterioration. When it occurs, it is reasonable to assume that deterioration reflects progression of the condition that causes epilepsy the toxic effects of AEDs, or both (Trimble, 1987). In some cases the postictal depressions of patients with frequent seizures produce reversible pseudodementia.
Anoxia has been frequently suggested as the basic mechanism in epileptic deterioration, assuming that deterioration in epileptics is the result of epilepsy. There is no doubt that tissue anoxia can occur in all forms of epilepsy, and it could be an important factor in deterioration. One would expect anoxia to be especially important in generalized motor seizures, as opposed to complex partial seizures because in generalized motor seizures excessive muscular activity and apnea coincide with a generalized increase in neuronal firing rate, which creates a cerebral metabolic demand for oxygen that cannot be met. Guerrant et al. (1962) observed the opposite. Complex partial seizures were more likely to precede dementia and psychosis than grand mal seizures. Caplan et al. (1998) also failed to demonstrate more psychopathology or deterioration in patients who had generalized epilepsy.
Epileptic deterioration may not be caused by epilepsy at all. Seizures of temporal
P.28
origin have often masqueraded as the cause of a syndrome: seizures increase, hemiparesis advancement and emotional deterioration, intellectual decline. The existence of this syndrome has been one of the most substantial supports for the theory that seizures themselves, especially complex partial seizures of temporal lobe origin, damage the brain. This syndrome is actually caused by a focal inflammation in the temporal lobe. The origin of the inflammation is mysterious and may be autoimmune (Andrews et al., 1997). Seizures and cognitive decline are both the result of the inflammation. Seizures do not cause the intractability of seizures in that condition nor do cause dementia and hemiparesis. Yet it may appear as though seizures themselves have caused more seizures, hemiparesis, and cognitive decline. When the electrical focus is surgically removed and examined by neuropathologists, the characteristic subacute inflammation secures the diagnosis of Rasmussen's encephalitis. There is no other manner of establishing that diagnosis with certainty.
The diagnosis of epileptic deterioration, that is, deterioration caused by epilepsy, has often been misapplied. The dementia of cerebral dygenesis, brain tumors, or degenerative diseases that cause both epilepsy and dementia do not recover if epilepsy is controlled as both epilepsy and dementia are the resulting symptoms of the primary condition (Hart et al., 1998).
The impairment of recent memory that follows electroconvulsive therapy (ECT) occurs even when muscular activity during the ECT-induced seizure is abolished with succinycholine and oxygenation is artificially maintained. Yet there is no good evidence that this transient deficit induced by ECT can become permanent. There is no evidence that ECT induces any type of brain damage as shown by autopsy, CT/MRI, and neuropsychological studies of patients who had received ECT (Devanand et al., 1994). In fact, all the changes that causes are transient: cognitive changes and spotty memory loss for events around the time of ECT administration. Devanand and his colleagues also looked at animal studies of electroconvulsant stimulation. These also showed no neuronal loss consequent to seizures; (there was neuronal loss after 1.5–2 hours of continuous seizure activity in primates, but this could be prevented with appropriate oxygenation and the use of muscle relaxants). This indicates that brain damage in prolonged status epilepticus is the result of an imbalance between cerebral oxygen need and the ability of the body to provide it. When muscles are paralyzed they do not engage in convulsive activity and consume less oxygen, leaving more for the brain. There is only some evidence from animal studies (Meldrum, 1978) that the mere passage of electricity or the transient disruption blood-brain barrier associated with seizures causes any type of brain damage.
In conclusion, there is little evidence that epilepsy causes dementia. When increasing seizures, dementia, or both occur in epileptic patients, some other etiology is likely. Therefore, any epileptic with cognitive decline likely to have (1) a deterioration of the condition that causes his epilepsy; (2) anticonvulsant
P.29
toxicity; or (3) frequent seizures. Investigation should include MRI, anticonvulsant blood levels, and EEG.
Psychosis
There is an increased incidence of serious psychopathology in patients with seizure disorders that ranges from 20 to 40 percent (Trimble, 1991). Conversely, seizure disorders are 3 to 7 times more common in groups of chronically hospitalized psychotic patients than in the general population (McKenna et al., 1985; Diehl, 1989). Dodrill and Batzel (1986) reviewed the literature of the association of psychopathology and epilepsy concluded that there is a positive relation between the two. The advent of DSM-III in 1980 led to many advances standardizing the diagnosis of psychiatric disorders. Victaroff (1994) used a standardized interview to assess the DSM-IIIR lifetime diagnoses of 60 patients with complex partial seizures. Thirteen percent had a history of psychosis.
In an older study designed to test the hypothesis that patients with complex partial seizures have more psychiatric disorders than have patients with grand mal epilepsy or general medical chronic illnesses, Guerrant et al. (1962) found that psychosis was present in 20 percent of patients with complex partial epilepsy but in only 4 percent of those with grand mal. The incidence psychosis in the general population is about 1 percent (Srole et al., 1962).
Standard psychometric testing by Guerrant's group failed to confirm greater abnormality in patients with complex partial seizures. This discrepancy between the clinical impression of psychosis on the one hand, and conventional psychological test results, on the other, has been observed by several other investigators (Small et al., 1962; Stevens, 1966). It is likely that the standard psychological tests (IQ) assess aspects of cognitive functions that are different from the psychotic features that were uncovered during the clinical interviews.
Rodin et al. (1976) reported the greatest prevalence of serious behavioral abnormalities in temporal lobe seizure patients with more than one clinical seizure type. Patients with only CPE were not different from other seizure patients with respect to psychotic tendencies.
Though there had been many anecdotal reports linking schizophrenia and epilepsy (and occasionally even indicating their nonassociation), it was not until publication of a large study by Slater and Beard (1963) that a clear relationship between a psychosis resembling schizophrenia and epilepsy was established. Of 69 psychotic epileptics in this study, 80 percent showed evidence of temporal lobe dysfunction on the basis of a history complex partial seizures or temporal lobe spiking on the EEG or both. The mean age of onset of psychosis was 30 years. The mean duration of the onset epilepsy before the psychosis was 14 years. Seizures varied from rare to frequent and there no
P.30
relationship between psychosis and the dosage of anticonvulsant drugs. The low incidence of schizophrenia in the first-degree relatives these patients strongly indicated that they were suffering from something other than classic schizophrenia. Though this group did show, at various times, all the cardinal features of schizophrenia, the psychoses in the epileptics deviated from norms for schizophrenia in some respects. Affective responsiveness was often preserved to an extent unusual in schizophrenia. In the later stages of the development psychosis, the patient's personality was sometimes left essentially undamaged, which is rarely the case in the later stages of schizophrenia.
Bredkjaer et al. (1998) confirmed Slater and Beard's findings. The prevalence of schizophrenia-like psychoses was significantly higher in both male and female epileptics than in the general Danish population. These authors used a national patient register and the Danish Psychiatric Register for a record-linkage study of 67,116 people with epilepsy. The relationship between epilepsy and psychosis was evident even after people with learning disability or substance abuse were excluded as subjects. This study impressively supported the notion of an association between epilepsy and subsequent schizophrenia-like psychosis.
It may be extremely difficult for a clinician to distinguish ordinary schizophrenia from the schizophrenia-like psychosis of epilepsy. In both conditions, the psychosis may begin in the second or third decade, and the course varies. Remission may occur in either, though both tend to be chronic. Personality and affect tend to be less abnormal in epileptic psychosis, but symptomatically the two psychotic conditions may be virtually indistinguishable. The family history and a positive medical history of epilepsy aids in the diagnosis. In ordinary schizophrenia there is a high incidence of serious psychopathology in the immediate family of affected patients, with 10 to 15 percent the first-degree relatives diagnosed as schizophrenic. By contrast, the incidence of schizophrenia in the families of patients with epileptic psychosis does not exceed that the general population (Slater and Glithero, 1963). A history of epilepsy is, of course, the main distinguishing feature, being positive in epileptic psychosis and negative in ordinary schizophrenia.
The EEG may be of some help in the differential diagnosis, given that electroencephalographic abnormalities, particularly those related to the anterior temporal regions, are implicated in epileptic psychosis. There is an increase frequency of abnormal EEGs in schizophrenics, but normal EEGs are not unusual in epileptic psychotics. Indeed, the phenomenon of forced normalization (normalized EEG in a psychotic epileptic) emphasizes the inverse relationship between the manifestations of epilepsy and psychosis. Some authors have commented on the tendency of some patients to become psychotic when seizures are controlled and vice versa. This is the clinical equivalent of the EEG phenomenon called forced normalization (Pond, 1957; Flor-Henry, 1969; Mignone, et al., 1970; Reynolds, 1971; Standage and Fenton, 1975; Pakalnis, 1987; Trimble, 1989; Krishnamoorthy and Trimble, 1999).
P.31
Epilepsy and psychosis could be related in at least three ways. They could vary inversely, psychosis appearing as seizures are controlled. Psychosis and seizures could be completely unrelated to each other, both independent manifestations of the same brain damage. They could vary directly, both psychosis and seizures worsening and clearing in parallel. Each of these situations seems to obtain in some cases.
There are two theories that could explain an inverse relationship between seizures and psychosis. (1) schizophrenia-like psychosis may result from the suppression of the epileptic focus and (2) AEDs may cause psychosis.
P.32
not appear to be a factor (Fischer et al., 1965). Other anticonvulsant drugs that are well known to cause psychosis include vigabatrin, tiagabine, felbamate, bromides, tridione and phenurone.
Lesions in the temporal lobe and its limbic projections probably predispose an individual to a schizophrenia-like psychosis. This hypothesis has been supported by other studies performed over the past 50 years of psychotic epileptics (Pond, 1957; Glaser et al., 1963; Flor-Henry, 1969, 1972). Andermann et al. (1999) related psychosis in epileptics to brain pathology. Kanemoto et al. (1996) related epileptic psychosis to hippocampal sclerosis visible on the MRI, and Sachdev (1998) concluded that structural brain abnormalities such as cortical dysgenesis underly both psychosis and epilepsy in patients with both disorders. The presence of tissue pathology in the temporal lobe even seemed to explain transient postictal psychosis in a group of epileptic patients (Devinsky et al., 1995). Tumors in the temporal lobe can cause psychosis even without epilepsy (Malamud, 1967) as can encephalitis. The temporal lobes are likely to cause psychosis when they are damaged, whether or not the lesions give rise to epilepsy. When temporal lobe disease causes epilepsy, it is most likely to cause complex partial seizures (CPE) of temporal lobe type, hence the clinical association of CPE and psychosis.
Psychosis could be the direct result of epilepsy through the mechanism kindling. Kindling occurs in animals. Repetitive subthreshold stimuli, particularly to areas of the limbic system, eventually lead actual seizures; prior these kindled seizures there are often marked behavioral changes. Kindling has never been documented in humans but the idea that an epileptic focus could “kindle” a psychosis has remained an attractive hypothesis to explain the occurrence of psychosis in patients with seizure disorders (Adamec, 1990). Kindling has also served as an attractive hypothesis for speculation about bipolar disorder and its relation to seizure disorders. Post (1992), drawing from the animal studies, speculated that there is a similar kindling process that goes on early in the course of bipolar disorder that then sensitizes other parts brain and particularly the temporal lobes that create not only symptoms similar to temporal lobe epilepsy, but make the bipolar disorder amenable to treatment with anticonvulsant drugs (Post, 1992, 1996; Atre-Vaidya, 1997).
Psychosis that occurs after a seizure is part of the postictal state and as such is a direct outgrowth of epilepsy. In specialized epilepsy monitoring units (Blumer et al., 1995; Kanner, 1996), it is customary to lower or stop anticonvulsants while the patient is being clinically and electroencephalographically studied. Kanner et al. (1996) noted postictal psychiatric events in 13 patients, 10 of whom had psychotic episodes of short duration (meantime, 66.5 hours) that disappeared either spontaneously or responded to psychotropic medication. The patients who continued to have seizures after the studies, even when the AED was resumed, also continued to have postictal psychiatric events. Ketter et al. (1994)
P.33
noted similar phenomena when withdrawing patients' AEDs for a drug trial of a new anticonvulsant medications.
These studies indicating that at least some of the psychotic episodes suffered by epileptics are postictal events that directly related to the previous seizure provide some clues to the treatment of psychosis in epileptics. One of the first efforts should be directed to the most effective seizure control possible with antiepileptic medications. Clinically, this often means raising the anticonvulsant blood levels to the upper limits of therapeutic window and, at times, even beyond. If the effective use of one anticonvulsant does not diminish the psychosis, the anticonvulsant should be changed. Persistance of psychosis indicates adding an antipsychotic neuroleptic. To date, there have been no large-scale studies of the use of neuroleptic medications to control psychotic symptoms associated with seizures (McConnell, 1998). Most of the reports are anecdotal, but there is now considerable experience in the use of neuroleptic medications in combination with anticonvulsants.
The older phenothiazines, butyrophenones, thioxanthenes, and pimozide have all been used in small doses for this purpose. Patients with seizure disorders can be very sensitive to medication changes and there may be an initial worsening of the mental or epileptic condition when psychotropic medications are introduced, changed, or increased. Consequently, most changes in medication should be made slowly. The new antipsychotic medications that are called atypical agents such as clozapine, quetiapine, rispiradone, olanzapine, and sertindole, have all induced clinical improvement. All seem to have very little tendency lower the seizure threshold, except for clozapine (McConnell, 1998).
The evidence is strong that links limbic tissue dysfunction (unrelated to epilepsy) to psychosis and affective disturbance. There is some evidence that disease in the dominant hemisphere is more likely to produce psychosis than disease in the nondominant hemisphere (Flor-Henry, 1969; Taylor, 1975). We are unconvinced. The evidence that the dominant hemisphere is the source of psychological problems has been challenged by a careful psychological study of 27 children with either pure left (13) or pure right hemisphere (14) temporal lobe epilepsy. This study revealed no left-right differences in WISC, Halstead-Reitan, Achievement Test, and Personal Inventory scores. Cognitive, personality, school problems were encountered in 10 (5 with left and 5 right foci) those showed lower neuropsychological test functioning than did the normally adjusted children (Camfield et al., 1984).
In conclusion, the evidence supports notion that there is a schizophrenialike psychosis in epilepsy that affects a minority of epileptics. Most these have complex partial seizures. The postictal state, anticonvulsant use, the lesion in the brain that is causing epilepsy, and the neurophysiological response to a discharging focus in the brain may all cause or contribute to development of a schizophrenia-like psychosis in epileptics.
P.34
Affective Disorders
Although much of the literature on behavior disturbance in seizure disorders focuses on psychosis, it has become evident that affective disturbances are much more common (Schmitz et al., 1999). Many disorders of the brain have an affective component that is more than an emotional reaction to the illness (Silver, 1990; Popkin and Tucker, 1994; Blumer et al., 1995). Depressive symptoms can occur as a manifestation of an ictal event, during the postictal phase seizure (Robertson, 1998), as a medication effect, and probably manifestation of brain injury.
Blumer et al. (1995) described an interictal mood disorder that consists of labile depressive symptoms (depressed mood, anergia, insomnia, and pain), labile affective symptoms (fear, anxiety), and paroxysmal irritability euphoric moods. Blumer's report needs to be confirmed by others but his emphasis on the high prevalence of depression and manic-like symptoms in epilepsy is supported by other data.
The suicide rate is very high in epilepsy (Gehlert, 1994), and attempts are also more common (Matthew, 1981). Barraclough (1987) reported a 25-fold increase in suicide risk for patients with temporal lobe epilepsy. A higher incidence of every sort psychopathology is usually found in university hospital populations and specialized epilepsy study centers as patients in these settings are usually highly selected and tend to be the more difficult unusual cases. Lower incidence figures are usually associated with community surveys of normally distributed populations. Even so, the relationship between epilepsy and depression seems real.
Patients with seizure disorders and comorbid depression show high levels of hostility, guilt, and self-criticism as well as greater impairment of function (Guze, 1994; Robertson, 1994). The relationship to type of epilepsy or syndrome is unclear and some have noted a preponderance of depressive symptoms with left-sided lesions (Mendez et al., 1994; Victoroff, 1994). Others have associated the depressive symptoms with partial seizures, male gender, and left epileptogenic focus (Strauss, 1992; Septien, 1993; Altshuler, 1999). To explain depression some have invoked the anticonvulsant therapy itself. Phenobarbital, vigabatrin, and combinations of other anticonvulsants have all been associated with depressive feelings (Brent et al., 1990; Mendez, 1993; Bauer, 1995) (see Tables 1-4a and 1-4b).
In a large epilepsy care center, Blumer (1997) found that half of all patients with chronic epilepsy experience an intermittent and polysymptomatic affective disorder. In comparison, fewer than 10 percent suffered from psychosis and these were the very patients who had the most severe affective disorders. Could the schizophrenia-like psychosis of epilepsy actually be a form affective psychosis? Altschuler et al. (1999) did a 10-year follow-up study of 49 patients who had undergone surgery for refractory temporal lobe seizures. The incidence
P.35
of affective disorder in these patients was quite high: 77 percent had a prior history of depression; 10 percent developed depression for the first time after surgery; and 50 percent showed complete remission of depression after surgery. Forty-seven percent had no recurrence of depression during the post-surgery follow-up.
Using a combination of tricyclic antidepressant and selective serotonin reuptake inhibitor, Blumer (1997) reported a good or excellent response in 15 of 22 refractory cases of psychosis with epilepsy. The affective disorders associated with seizures responded to antidepressant treatment though smaller doses and slower titrations were felt to be necessary. The reason for being cautious in using antidepressants is that they all lower seizure threshold to some degree. Even in patients without previous seizure disorders the tricyclic antidepressants and monamine oxidase inhibitors can lower the seizure threshold (Trimble, 1978). Bupropion, maprotiline, clomipramine, and amoxapine cause approximately 2-3 percent of patients to have seizures; maprotiline induces seizures in 15 percent. Venlafaxine causes seizures in only 0.2 percent and the SSRIs approximately 0.1 percent (Maxmen, 1995). There seems to be no evidence in favor of one antidepressant over another with relation to efficacy antidepressant effect, but the low potential of the new SSRIs to lower the seizure threshold would make them seem to be a logical first step in the treatment of depression in epileptics (Bryan et al., 1983).
As data on the effects of antidepressants hepatic cytochrome p450 system that is responsible for metabolizing many drugs emerge, the question of drug interactions is an important consideration in using anticonvulsants and antidepressants. One can potentially raise the blood level of the other. Some the antidepressant drugs also may actually have a direct anticonvulsant effect in some cases. Favale et al. (1995)treated 17 patients with complex partial seizures with fluoxetine as an adjunct to their anticonvulsants and noted the complete cessation of seizures in six of the patients and a 30 percent decrease seizure frequency in the rest. This has also been reported with some of older antidepressants as well. Some patients with affective disorders and seizures may respond to treatment with carbamazepine alone for both conditions (Carried, 1993; Varney et al., 1993). Several studies have observed that psychotherapeutic and psychological efforts have been helpful in reducing seizure frequency as well as increasing coping skills and compliance with treatment (Fenwick, 1992; Mathers, 1992;Regan, 1993).
Anxiety Disorders
Seizure disorders and anxiety have many similarities, particularly panic attacks. Both are paroxysmal, sudden in onset, often with no precipitants, with marked feelings of fear, anxiety, or both, as well autonomic and physical
P.36
symptoms. Both result from abnormalities of the brain, and both conditions can respond to benzodiazepines. One of the most frequent sources of neuropsychiatric consultation by another physician is the patient who has episodic anxiety attacks with some depersonalization but who has not responded to initial treatment. Could such a patient have epilepsy? Many have postulated that there is a subgroup of panic attacks that are related to epilepsy and particularly temporal lobe pathology (Dantendorfer et al., 1995). Others have postulated a relationship between dysfunction of parietal-frontal lobes and panic attacks (McNamara et al., 1990; Alemayehu et al., 1995). In epidemiological studies, Neubgebauer et al. (1993) noted a suggestive overlap between seizures and anxiety disorders, but Spitz (1991)found no relation between complex partial seizures and anxiety disorders. Seizures can cause repetitive obsessive thoughts and anxiety thus presenting as an atypical obsessive-compulsive disorder.
Seizure disorders can also cause recurrent memories, flashbacks, etc. that mimic many of the features post-traumatic stress disorder; and some patients with seizure disorder can develop agrophobic symptoms during seizures. However, patients with panic disorder usually do not manifest a disturbance of consciousness; there is usually a positive family history of panic disorder; there are no automatisms in panic disorder. However, the differential diagnosis can be difficult and diagnosis usually relies on a careful history clinical examination and findings. To capture an episode on EEG, prolonged recording may be necessary. Weilburg et al. (1995) studied patients with atypical panic attacks with ambulatory EEG monitoring and found a significant number who showed focal paroxysmal EEG changes. Some of these patients had normal standard EEGs. The treatment of anxiety symptoms in patients with seizure disorders is similar to the treatment of anxiety disorders in clinical practice. The symptoms often respond to psychological treatments, benzodiazepines, and antidepressants.
Sexual Behavior
Some have denied that there is any change in the sexual functioning of epileptics (Jensen et al., 1990; Duncan 1997). Others have found hyposexuality in men (Murialdo, 1995) and some have described hyposexuality in women (Demerdash, 1991). Hyposexuality has been ascribed to the effects of epilepsy, resolving with successful anticonvulsant treatment (Silveira et al., 2001), to the effects of the brain damage that causes epilepsy, and to the medications used to treat epilepsy (Bergan et al., 1992; Isojarvi and 1995). Gastaut Collomb (1954) reported hyposexuality in two-thirds of 36 male temporal lobe epileptics but found that hyposexuality was infrequent in patients with other types of epilepsy. A striking feature of this hyposexuality was poverty sexual drive rather than impotence. The condition seemed to follow the onset of seizures.
P.37
In general, male sexual drive does not develop in complex partial epilepsy that begins in childhood, and sexual interest diminishes after the onset of complex partial epilepsy in adults. A study by Taylor (1969) of 100 complex partial epileptics before and after temporal lobectomy confirmed the findings of Gastaut and Collomb. Only 14 patients had a satisfactory sexual adjustment, preoperatively. Postoperatively, the sexual adjustment in these 14 was still normal; it improved in 22 other patients; and worsened another 14. Fifty patients maintained the same poor adjustment. Those patients who improved also experienced the greatest relief from seizures. Since virtually every AED can cause impotence, the improved sexual function of those patrients with the best surgical results may simply reflect the benefit of a reduction in AED dosage.
The relationship between epilepsy and impotence has also been studied. Hierons and Saunders (1966) reported 15 cases of impotence unrelated to diminished libido in patients with temporal lobe lesions. It is not clear whether their report reflected the selection of patients studied or accurately represented the psychosexual adjustment of all patients with temporal lobe epilepsy.
It is not uncommon for sexual problems to appear in any chronically ill patient. But the experimental evidence that relates sexual function to the limbic system is a further reason to expect some sort of sexual dysfunction in complex partial epilepsy. Destructive lesions in the amygdala have given rise to indiscriminate hypersexuality (Kluver and Bucy, 1939) stimulation of limbic structures has given rise to erections (MacLean and Ploog, 1962). “Sexual seizures” have been noted in temporal lobe epileptics but are quite rare (Currier et al., 1971). Sexual automatisms have more recently been identified with frontal foci (Spencer et al., 1983).
Personality Disturbance
Many have postulated a specific epileptic personality. Such traits and behaviors as preoccupation with philosophical and religious concerns, dependency, humorlessness, circumstantially, hypergraphia, hyposexuality, viscosity, and paranoia have all been cited as typical of the epileptic personality (Waxman, 1975; Bear, 1977; Hermann, 1981). These observations rest on case reports. When large groups of epileptics are surveyed with standardized instruments, these specific traits do not seem to hold up well (Stevens, 1975; Menges, 1982; Stark-Adamec, 1985). With regard to specific personality type, one systematic study has been done by Mendez (1993) who found a wide range of personality disorders in an epilepsy clinic. These included borderline, explosive, and dependent. When these patients were compared to epileptic patients without a personality disorder, the former had more auras with psychic symptoms (fear, depersonalization, etc.) and fewer generalized tonic-clonic convulsions. Others
P.38
have postulated a relationship to types of borderline personality disorder also (Andrulonis, 1982; Gunderson, 1989), as well to increased episodic and impulsive behavior.
Violence, The Temporal Lobes, and The Limbic System
One behavior that has often been thought to arise from an epileptic disorder of the brain, particularly the limbic system, is explosive rage and violence. This idea stems in part from animal studies where sites the limbic system are stimulated and the animal either has a rage attack or becomes aggressive. As many epileptic patients have temporal lobe foci, it has been postulated that there may be an association of epilepsy and violence. Studies prison populations have also shown a high incidence of EEG abnormality (Gunn and Bonn, 1971). Nonetheless, violence and aggression are very complex behaviors; the vast majority of epileptics are not violent, and the vast majority criminals neither epileptic nor excessively violent. Explanations that attribute violence or aggressive behavior to a specific anatomical site or disease are overly simplistic.
The limbic system is in constant interaction with the frontal lobes, striatum, thalamus, and the hypothalamus, as well the neuroendocrine immune systems (Mesulam, 2000). This constant interplay with other brain regions complicates the relationship between limbic system and violent behavior must be kept in mind throughout this chapter.
The term limbic system refers to the ring of deep, centrally located nerve cells of gray matter and connections between the hemispheres in the medial portions of the brain that play a role in emotions. Phylogenetically, many of the areas designated as the limbic system are among the oldest portions of cortex; in lower creatures these structures largely subserve smell and have traditionally been called the rhinencephalon. But since all regions designated “limbic” are not related to olfaction, and since other brain regions in addition the limbic system play a role in emotional functioning, the term has been criticized (Brodal, 1981).
Papez (1937) first pointed out that the limbic system was related to emotion and behavior and visceral reactivity in humans. He predicted that following stimulation of the hippocampus, there could be prolonged active electrical discharges, which resulted in very little spread to neocortical areas on the surface of the brain that would spread between and among limbic system's other components. He predicted that these reverberating circuits of discharge within the limbic system would produce marked alterations in the subjective emotional life of an individual. In effect, Papez proposed an anatomical and physiological
P.39
substrate for the intense affective reactions and instincts that are customarily the domain of much psychiatric theory and research.
At the time Papez published his paper on the limbic system, Freudian theory was in wide vogue. The idea of a phylogenetically ancient, deep, central portion of the nervous system that influenced behavior and thought not under conscious, neocortical control was consistent with some Freudian concepts of instinctual drive. The reasoning went something like this: sense of smell in lower animals seems to be closely associated with memory, instinct, and emotion, for it is often smell that alerts an animal to danger and provokes fear, flight, or fighting, as well sexual arousal and mating. Smell memory in such animals must be related functions given that it is important for lower animals to remember the associations of particular smells. The autonomic nervous system must be closely related to the limbic system because such autonomic responses as pupillary dilation, piloerection, increased heart rate, and increased blood flow to skeletal muscles occur in response to environmental circumstances which an animal must fight or flee or prepare for mating. Though the sense of smell is no longer as important to human life, the limbic system in man is still involved with emotions and memory, disturbances of the limbic system disrupt them.
Anatomy
The gray matter components encompassed by the term limbic system include those in the anterior and medial portions of temporal lobe those outside the temporal lobe. Limbic components in the lobe include the amygdala, hippocampus (both the gyrus hippocampus and its medial portion, hippocampal formation, which is sometimes called Ammon's horn), and the uncus. Limbic components outside the temporal lobe include mamillary bodies, anterior nucleus of the thalamus, gyrus cingulus, nuclei septum, portions of the midbrain tegmentum, and supracallosal gyri. The major tracts interconnecting these regions include the fimbria, fornix, mamillothalamic tract, anterior commissure, stria terminalis, stria medullaris, median forebrain bundle, and diagonal band of Broca (see Figs. 1-2 and 1-3). In order to conceptualize this system, it may be helpful to recall that many of the medial structures brain have the shape of a large “C”—with one end in the anterior temporal lobe and the other in or near septal region. Among limbic components that have this form are: (1) the gyrus cingulus; (2) fimbria-fornix-mamillary body pathway; (3) the stria terminalis, which connects the amygdala and the septal area; and (4) the supracallosal gyrus longitudinal striae, which connect the hippocampus region with the septal region. Other tracts with a curved shape are the media forebrain bundle, which connects the septal nuclei with midbrain tegmentun, and the stria medullaris, which connects the septal region with the
P.40
habenula (Fig. 1-2). The anterior commissure is a tract that laterally connects the right and left amygdala. The diagonal band of Broca also runs laterally to connect the septum with amygdala. The tracts that connect gray matter of the limbic system generally contain both afferent and efferent fibers (Figs. 1-2, 1-3, 1-4). The richness of interconnections among regions the limbic system can only partly be appreciated by the account above; actually not all the interconnections are known.
|
Figure 1-2 The limbic system. AC, anterior commissure; ANthal, anterior nucleus of the thalamus; CG, central gray matter of the midbrain; DBB, diagonal band broca; G, Gudden's deep tegmental nucleus; HAB, habenula; HYPO, hypothalamus; IPN, interpeduncular nucleus; LMA NAUTA, lateral midbrain area of nauta; lat olf stria, lateral olfactory stria; med olfac str, medial olfactory stria; MFB, median forebrain bundle; MAM B, mamillary bodies; stria med, stria medullaris; stria term, stria terminalis; tub, tuber cinereum; OLF BULB, olfactory bulb; NUC SEPTUM, Septum. |
Physiological Psychology
There are rich interconnections between the frontal lobes and the limbic system. In general, the frontal lobe inhibits limbic system (Porrino et al., 1981; Goldman-Rakic et al., 1984; Jay et al., 1995; Thierry 2000). Lesions in
P.41
P.42
either the frontal lobe or in the fibers that link and limbic regions would tend to disinhibit the limbic system, permitting expression of some primitive thoughts and behaviors.
|
Figure 1-3 The limbic system: anterior coronal section. |
|
Figure 1-4 The limbic system: posterior coronal section. |
Efforts to determine the function of the various components limbic system have involved stimulation and ablation studies in animals to some extent in man. These have yielded evidence that these components influence memory, learning, emotional states (including anxiety, rage, placidity, and alertness), visceral and endocrine responses, behavior (particularly aggressive), oral, and sexual activity. It is not possible to define the function of each of the components of the limbic system because none them acts as a center for a particular function. All of the limbic system is more or less associated with all of the functions listed above. As Papez predicted, there is a strong tendency after electrical stimulation of any of the limbic components for prolonged electrical discharges to persist and to spread throughout the limbic system with comparatively little involvement of the neocortex (MacLean, 1952, 1954).
The specific aspects of behavior elicited by stimulation and ablation studies in the limbic system are of strong theoretical interest to both psychiatrists and neurologists given that the behavioral and cognitive alterations produced by experiments in animals closely resemble human responses that involve cognition and behavior. Stimulation of the hippocampus cats results in apparent bewilderment and anxiety together with intense attention to something the animal seems to sense in the environment. Such phenomena have been considered alerting and defensive reactions, resulting perhaps from hallucinations induced by the stimulation. Amygdalar stimulation may produce similar reactions. Bilateral hippocampal destruction leads to recent memory loss with prevention of new learning in both animals and man. Destruction of other components the limbic system also produces deficits in recent memory. Bilateral ablation of the anterior gyrus cingulus and bilateral division of the fornices produce similar deficits.
Stimulation of portions the limbic system produces rage reactions in animals. Similar reactions have been seen after stimulation of the midbrain gray matter or the placement of destructive lesions in septum. Stimulation amygdala in animals has provoked reactions that have been interpreted as reflecting feelings of fear. Sensations fear have also been described in conscious human beings while this region was stimulated during surgery. Chewing, gagging, licking, retching, swallowing, bladder contractions, respiratory, pulse, and blood pressure increases, and increased secretion of adrenocorticotropic hormone (ACTH) have all been produced by amygdalar stimulation, as well stimulation elsewhere in the limbic system. Such phenomena are quite similar to the manifestations of certain forms epilepsy. This clinical similarity and the characteristic anterior temporal spikes seen in the EEG of patients with complex partial epilepsy led some clinicians to apply the term limbic epilepsy to
P.43
such seizures (Fulton, 1953). The bizarre behavioral alterations (e.g., docility, loss of natural fear, compulsive oral activity, and heightened indiscriminate sexual activity) noted by Kluver and Bucy (1939) after bilateral removal of the anterior temporal lobe, amygdaloid nuclei, and overlying hippocampal cortex provided further evidence for the limbic system's role in these functions. Similar changes have been noted in man after bilateral temporal lobectomies, which, if performed somewhat caudal to the amygdala, also produce profound memory loss, particularly for recent events.
Considering the relationship between limbic system and emotions, it might be predicted that diseases involving limbic components would cause emotional disorders. It also seems to be so in other conditions that involve limbic components (Gibbs, 1952; Malamud, 1967; Glaser and Pincus, 1969; Himmelhoch et al., 1970).
In patients being examined for epilepsy, electrical stimulation of the amygdala and the hippocampus after placement of chronic, implanted electrodes has produced brief alterations that mimic complex partial seizures and persist only during the passage of current and the limited after-discharge (Stevens et al., 1969). After such stimulation, however, mood and thought disturbances of psychotic proportions may persist for hours.
Aggression and Violence as an Epileptic-like Discharge of the Limbic System
A wide range of behaviors have been related to the limbic system, but perhaps aggression, and particularly aggression in epileptics, has received the most attention (Fenwick, 1986). The syndrome of episodic dyscontrol manifested by irrational or unprovoked expressions of rage presumably reflects epileptic-like discharges in the limbic system (Drake et al., 1992; van Elst 2000). DSM-IV has included a category similar to episodic dyscontrol called Intermittent Explosive Disorder, and although this condition has been related to affective disorder, it does respond to anticonvulsant drugs (McElroy, 1999). Though some violence may result from epileptic-like discharges in the limbic system, this is rare. Many varied sources of violence in patients with seizures have been reported:
P.44
were associated with spiking discharges in the amygdala and that the behavior could be reproduced by stimulation. In only one patient was there a clear indication of an ictal basis violent behavior from the surface EEG recordings (Smith, 1980).
P.45
patients who had only complex partial epilepsy were “uniformly intelligent and conforming children and none of them had rage outbursts at any time.” Perhaps the most important of the studies showing no association between complex partial epilepsy and violence was that of Rodin (1973), who evaluated 57 patients with complex partial epilepsy and observed them during their seizures. There were no instances of ictal or postictal aggression. A review 700 case histories of patients in his epilepsy clinic revealed 34 who had committed aggressive acts. The presence or absence of complex partial epilepsy in these 34 patients was not a relevant variable.
The contradictions in these reports may be related to varying definitions of aggression, violence, and complex partial seizures to the selection of patients. Fighting among children, particularly boys, is certainly not unusual. Physicians who have cared for large numbers of seizure patients would agree that violence and aggressive acts do occur in patients with complex partial seizures, but they would disagree as to whether the incidence of violence in this form epilepsy is more or less frequent than in the epileptic general population.
Gunn and Bonn (1971) found no difference in the kind of violent behavior manifested by epileptic and nonepileptic prisoners. In a comparative study of epileptic prisoners and hospitalized epileptic nonprisoners, Gunn (1974) noted great similarities in medical and social factors psychiatric symptoms; the main difference was more drinking in the group of prisoners. This again raises questions about the hypothesis that violent behavior and seizures per se are related.
Gunn and Fenton (1971) noted a relatively high incidence of epilepsy in prison populations but decided that automatic behavior during actual seizures was a rare explanation for crimes committed by epileptic patients. This study supports Stevens and Herman (1981) implies that nonictal violence occurs more frequently in brain-damaged prisoners with or without complex partial seizures.
Episodic Dyscontrol
Episodic dyscontrol is a descriptive syndrome manifested by outbursts of rage and violence that seem to be motiveless. It is thought arise from an epileptic discharge that originates in the brain (often the temporal lobe), and can be controlled with anticonvulsants. The concept of episodic dyscontrol does not provide a comprehensive explanation for more than few isolated instances of violence (Drake et al., 1992; van Elst 2000). Only one-third of murderers have evidence of temporal lobe abnormality (Blake et al., 1995), and only 5 of 97 incarcerated juvenile delinquents seemed to have an episode of violence that could have been the direct result of a complex partial seizure (Lewis et al., 1982). All five of these youths had other episodes of violence that were clearly
P.46
not caused by seizures. Any benefit derived from the use of anticonvulsants in treating people with episodic violent behavior may be related to the stabilization of their mood rather than the control of putative epilepsy.
Violent behavior has been reported in patients with lesions of the limbic system secondary to temporolimbic epilepsy (Mark and Ervin; 1970;Ashford et al., 1980; St Hilaire et al., 1980; Devinsky and Bear, 1984) focal lesions elsewhere, e.g., a patient with hamartoma of the ventromedial thalamus (Reeves and Plum, 1969). These reports seem to support the concept that violent behavior can arise from cerebral stimulation or destruction. However, there is an alternate explanation: because limbic circuits and limbic impulses are dampened by the frontal lobe, especially the orbitofrontal cortices, loss of the frontal fibers that reach the limbic system and inhibit it may allow expression of unacceptable limbic impulses. Lesions in the frontal lobe could do this. Indeed, the case of Phineas Gage has been interpreted this way (Damasio et al., 1994). Patients with orbitofrontal injuries have also been found to display disinhibition, impulsivity, lack of empathy that have justified the appellation “acquired sociopathy” (Tranel, 1994). Lesions in the temporal lobe, by disrupting inhibiting influences that arise in the frontal lobes and travel to limbic system, could also disinhibit behavior.
Conclusion
Acquired damage to the cerebral cortex is often a factor in epilepsy that clinically relevant. This is true even in epilepsy that genetically determined. There are forms of epilepsy that so difficult to differentiate from other causes of abnormal thought and behavior that the EEG can be an indispensable diagnostic tool, even though it is flawed by false negatives and, less often, by positives. By expanding EEG recording times and using other diagnostic tools, SPECT, PET, fMR, and MRI among them, diagnostic accuracy has been improved. Pseudoseizures of psychogenic origin, usually representing dissociative reactions of formerly physically and sexually abused patients, are especially difficult to distinguish from complex partial seizures without EEG confirmation. Therapy, in the form of anticonvulsant drugs and invasive procedures such as surgical ablation, disconnection, and vagal stimulation has become more effective. Some therapies that have proven effective in controlling epilepsy either not controlled, exacerbated, or created cognitive and behavioral abnormalities.
There have been several adverse circumstances associated with epilepsy that may be a direct or indirect result of continuing seizures. Seizures may be epileptogenic, leading to a further vulnerability of the patient seizures. Seizures may cause intellectual deterioration, psychosis, affective disorders, anxiety disorders,
P.47
personality disorders, disordered sexual behavior, and episodic dyscontrol, a form of violence. This chapter explores the relationship epilepsy to each of these.
In general, the lesions of the brain that have given rise to epilepsy are probably the major factor in continuing seizures, intellectual deterioration, and psychosis, though epilepsy and its treatment may also play a role in some cases. Much of the schizophrenia-like psychosis encountered among epileptics is probably affective and may be a manifestation of the response brain to presence of an active, discharging epileptic focus. Hyposexuality, encountered in a minority of epileptics, is probably less a manifestation epilepsy than it is a result of the brain abnormality that also caused epilepsy, though epilepsy and its treatment also may play a role. Violence is rarely direct manifestation of epilepsy. More commonly, epileptogenic lesions that damage the frontal inhibition of limbic activity cause behavioral vulnerabilities that lead to violence.
Perhaps no other neurological disorder confirms so clearly the role of the brain in psychopathology. Hallucinations, affective changes, panic attacks, sexual disturbances, and other behavioral changes can occur in epileptics. Consequently, the diagnosis of epilepsy must be part every clinical differential diagnosis in neuropsychiatry, especially if the behavioral disorder runs an episodic course.
References
Adamec, R. E., Does kindling model anything clinically relevant? Biol Psychiatry 27:249, 1990.
Adkins, J. C., S. Noble. Tiagabine: A review of its pharmacodynamic and pharmokinetic properties and therapeutic potential in the management of epilepsy. Drugs 55:437, 1998.
Aird, R. B., D. M. Woodbury. The Management of Epilepsy. C. C. Thomas, Springfield, IL, 1974.
Ajmone-Marsan, C., L. S. Zivin. Factors related to the occurrence of typical paroxysmal abnormalities in the EEG records of epileptic patients. Epilepsia 11:361, 1970.
Alemayehu S., G. K. Bergey, E. Barry, et al. Panic attacks as ictal manifestations of parietal lobe seizures. Epilepsia 36:824, 1995.
Alper, K., O. Devinsky, K. Perrine, et al. Dissociation in epilepsy and conversion nonepileptic seizures. Epilepsia 38:991, 1997.
Altschuler, L., R. Rausch, S. Delrahim, et al. Temporal lobe epilepsy, temporal lobectomy, major depression. J Neuropsychiary Clin Neurosci 11:436, 1999.
Alving J. Serum prolactin levels are elevated also after pseudo epileptic seizures. Seizure 7:85, 1998.
Andermann, F, B. Zifkin. The benign occipital epilepsies of childhood: an overview of the idiopathic syndromes and of the relationship to migraine. Epilepsia 39 Supl 4: S9-23, 1998.
P.48
Andermann, L. F, G. Svard, H. J. Meencke, et al. Psychosis after resection of ganglioglioma or DNET: evidence for an association.Epilepsia 40:83, 1999.
Andrews, P. I., J. O. McNamara, D. V. Lewis. Clinical and electroencaphelographic correlates in Rasmussen's encephalitis. Epilepsia38:189, 1997.
Andrulonis, P. A., B. C. Glueck, C. F. Stroebel, et al. Borderline personality subcategories. J Nerv Ment Dis 170:670, 1982.
Annegers, J. F., W. A. Hauser, L. R. Elveback, et al. Seizure disorders in offspring of parents with a history of seizures—A maternal-paternal difference. Epilepsia 17:1, 1976.
Ardinger H. H., J. F. Atkin, R. D. Blackston et al. Verification of the fetal valproate syndrome phenotype. AM. J. Med. Genet 29: 171, 1988.
Armon, C. C. Shin, P. Miller, et al. Reversible parkinsonism and cognitive impairment with chronic valproate use. Neurology 47:626, 1996.
Arnold, L. M., M. D. Privitera, Psychopathology and trauma in epileptic psychogenic seizure patients. Psychosomatics 37:438, 1996.
Ashford J. W., S. C. Schultz, F. O. Walsh. Violent automatism in a partial complex seizure. Arch Neurol 37:120, 1980.
Astradsson, A., E. Olafsson, P. Ludvigsson, et al. Rolandic epilepsy: an incidence study in Iceland. Epilepsia 39:884, 1998.
Atre-Vaidya, N., M. Taylor. The sensitization hypothesis and importance Of psychosensory features in mood disorder, J Neuropsychiatry Clin Neurosci 9:525, 1997.
Barraclough, B. M. The suicide rate in epilepsy. Acta Psychiatr Scand 76:339, 1987.
Bauer, J., C. E. Elger. Anticonvulsive drug therapy. Nervenarzt 66:403, 1995.
Bear, D. M., P. Fedio. Quantitative analysis of interictal behavior in temporal lobe epilepsy. Arch Neurol 34:454, 1977.
Berg, A., S. Shinnar, A. S. Darefsky, et. al. Predictors of recurrent seizures. Arch Pediatr Adolesc Med 151:371, 1997.
Bergen D., S. Daugherty, E. Eckenfels. Reduction of sexual activity in females taking antiepileptic drugs. Psychopathology 25:1, 1992.
Berkovic, S. F., A. M. Mclntosh, R. M. Kalnins, et al. Preoperative MRI predicts outcome of temporal lobectomy. Neurology 45:1358, 1995.
Blake, P. Y., J. H. Pincus, C. Buckner. Neurologic abnormalities in murderers. Neurology 45:1641, 1995.
Blumer, D. G. Montouris, B. Herman. Psychiatric morbidity in seizure patients on a neurodiagnostic monitoring unit. J Neuropsychiatry7:445, 1995.
Blumer, D., Antidepressant and double antidepressant treatment for the affective dissorder of epilepsy. J Clin Psychiatry 58:3, 1997.
Blumer, D., Wakhlu, K. Davies, et al. Psychiatric outcome of temporal lobectomy for epilepsy: incidence and treatment of psychiatric complications. Epilepsia 39:478, 1998.
Boon, P. A., P. D. Williamson, I. Fried, et al. Intracranial, intraaxial space occupying lesions in patients with intractable partial seizures: an anatomoclinical neuropsychological and surgical correlation. Epilepsia 32:467, 1991.
Boundy, K. L., C. C. Rowe, A. B. Black, et al. Localization of temporal lobe epileptic foci with iodine-123 iododexetimide cholinergic neuro receptor SPCT. Neurology 47:1014, 1996.
Bourgeois, B. F. Antiepileptic drugs, learning and behavior in childhood epilepsy. Epilepsia 39:913, 1998.
Bourgeois, B.F.D., A. C. Prensky, H. S. Palkes, et al. Intelligence in epilepsy: a prospective study in children. Ann Neurol 14:438, 1983.
P.49
Branford, D., S. Bhaunik, F. Duncan. Epilepsy in adults with learning disorders. Seizure 7:473, 1998.
Branford, D., R. A. Collacott. A followup study of adults with learning disabilities and epilepsy. Seizure 7:469, 1998.
Bray, P. F, W. C. Wiser. Evidence for a genetic etiology of temporal-central abnormalities in focal epilepsy. N Engl. J Med. 27.1:926, 1964.
Bredkjaer, S. R., P. B. Mortensen, J. Parnas. Epilepsy and non-organic, non-affective psychosis. National Epidemiologic Study. Br J Psychiatry 712:235, 1998.
Brent, D. A., P. K. Crumrine, R. Varma, et al. Phenobarbital treatment and major depressive disorder in children with epilepsy: a naturalistic follow-up. Pediatrics 85:1086, 1990.
Bridgers, S. L. J. S. Ebersole. The clinical utility of the ambulatory cassette EEG. Neurology 35:116, 1985.
Brodal, A. Neurological Anatomy in Relation to Clinical Medicine, 3rd ed. Oxford University Press, New York, 1981.
Bruehl C., O. W. Witte. Cellular activity underlying altered brain metabolism during focal epileptic activity. Ann Neurol 38:414, 1995.
Bryan, G. E., E. E. Marsh, B. Jabbari, et al. Incidence of seizures in patients receiving tricyclic and tetracyclic antidepressants. Ann Neurol 14:153, 1983.
Caplan, R., S. Arbelle, W. Megharious, et al. Psychopathology in pediatric complex partial and primary generalized epilepsy. Dev Med Child Neurol 40:805, 1998.
Carried, P. B., V. Provitera, B. lacovetti, et al. Mood disorders in epilepsy. Acta Neurol (Napoli) 15:62, 1993.
Chaudry, M. R., D. A. Pond. Mental deterioration in epileptic children. J Neurol Psychiat 24:213, 1961.
Cohen, L. M., G. F. Howard. Provocation of pseudoseizures by psychiatric interview during EEG and video monitoring. Int J Psychiatry Med 22:131, 1991.
Cohen, R. J., C. Suter. Hysterical seizures: suggestion as a provocative EEG test. Ann Neurol 11:391, 1982.
Collins, W. C., O. Lanigan, N. Callaghen. Plasma prolactin concentrations following epileptic and pseudo seizures J Neurol Neurosurg Psychiatry 46:505, 1983.
Coons, P. M. The dissociative disorders. Psychiatric Clin North Am 21:637, 1998.
Cramer, J., R. H. Mattson, M. L. Prevey, et al. How often is medication taken as prescribed? A novel assessment technique. JAMA261:3273, 1989.
Currie, S., KWG. Heathfield, R. A. Henson, et al. Clinical course and prognosis of temporal lobe epilepsy: a survey of 61 patients. Brain94:173, 1971.
Currier, R. D., S. C. Little, J. F. Suess, O. J. Andy. Sexual seizures. Arch Neurol 25:260, 1971.
Damasio, H., T. Grabowski, R. Frank. The return of Phineas Gage: clues about the brain from the skull of a famous patient. Science264:1102, 1994.
Dantendorfer K., M. Amering, W. Baischer, et al. Is there a pathophysiological and therapeutic link between panic disorder and epilepsy?Acta Psychiatr Scand 91:430, 1995
Delgado-Escuta, A., R. Mattson, L. King, et al. The nature of aggression during epileptic seizures. New Eng. J. Med. 305:711, 1981.
Delgado-Escueta, A. V., D. M. Tremain, G. O. Walsh. The treatable epileptics. N Engl J Med 308:1508, 1983.
Demerdash A., M. Shaalan, A. Midani, et al. Sexual behavior of a sample females with epilepsy. Epilepsia 3282, 1991.
P.50
Desai, B. T., R. Porter, J. K. Penry. The psychogenic seizure by video tape analysis: A study of 42 attacks in six patients (Abs.).Neurology 29:602, 1979.
Devanand, D. P., A. J. Dwork, E. R. Hutchinson, et al. Does ECT alter brain structure? Am J Psychiatry 151:957, 1994.
Devinsky, O., M. S. Duchowny. Seizures after convulsive therapy: a retrospective case survey. Neurology 33:921, 1983.
Devinsky O., D. Bear. Varieties of aggressive behavior in temporal lobe epilepsy. Am J Psychiatry 141:651, 1984.
Devinsky O., Kelley, R. J. Porter, et al. Clinical and electroencephalographic features of simple partial seizures. Neurology 38:1347, 1988.
Devinsky O., H. Abramson, K. Alper, et al. Postictal psychosis: a case control series of 20 patients and 150 controls. Epilepsy Res20:247, 1995.
Devinsky, O., S. Sanchez-Villasenor, B. Vasquez, et al. Clinical profile of patients with epileptic and nonepileptic seizures. Neurology46:1530, 1996.
Diehl, L. W. Schizophrenic syndromes in epilesies. Psychopathology 22:65, 1989.
Dodrill, C. B., L. W. Batzel. Interictal behavior fetures of patients with epilepsy. Epilepsia 27 (Suppl 2) S64, 1986.
Drake, M. E., A. Pakahis, B. B. Phillips. Neuropsychological and psychiatric correlates of intractable pseudoseizures. Seizure 1:11, 1992.
Dulac, O., A. Kaminska. Use of lamotrigine in Lennox-Gastaut and related epilepsy syndromes. J Child Neurol 12 (Suppl 1):23, 1997.
Duncan S., J. Blacklaw, G. H. Beastall, et al. Sexual function in women with epilepsy. Epilepsia 38:1074, 1997.
Ebersole, J. S., R. J. Leroy. An evaluation of ambulatory cassette EEG monitoring. II. Detection of interictal abnormalities. Neurology33:8, 1983a.
Ebersole, J. S., R. J. Leroy. Evaluation of ambulatory cassette EEG monitoring. III. Diagnostic accuracy compared to intensive inpatient EEG monitoring. Neurology 33:853, 1983b.
Eliashev, S. D., S. DeWar, I. Wainwright, et al. Long term follow up after temporal lobe resection for lesions associated with chronic seizures. Neurology 48:1383, 1997.
Ellenberg, J. H., K. B. Nelson. Febrile seizures, tested intelligence and learning disorder. Neurology 27:342, 1977.
Ellingson, R. J. Incidence of EEG abnormality among patients with mental disorders of apparently nonorganic origin. Critical review. Am J Phychiatry 111:263, 1955.
Engel, J. Discussion: epilepsy and behavior. Epilepsia 27 (Suppl 2):62, 1986.
Engel, J. PET scanning in partial epilepsy. Can J Neurol Sci 18 Suppl 4:588, 1991.
Engel, J. S. Henry, T. R, M. W. Risinger, et al. Presurgical evaluation for partial epilepsy: relative contributions of chronic depth electrode recordings versus FDG-PET and scalp-sphenoidal ictal EEG. Neurology 40:1670, 1990.
Engel, J. S. Surgery for seizures, N Engl J Med 334:647, 1996.
Falconer, M. Clinical, radiological and EEG correlations with pathological changes in the temporal lobes and their significance in surgical treatment. In: Temporal Lobe Epilepsy. M. Baldwin, P. Bailey, eds. C Thomas, Springfield, IL, 1958, p. 306.
Farfel, M. R. Reducing lead exposure in children. Annu Rev Public Health 6:333, 1985.
Farwell, J. R., G. Blackner, S. Sulzbacher, et al. First febrile seizures. Characteristics of the child, the seizure, and the illness. Clin Pediatr 33:263, 1994.
Farwell, J. R., Y. J. Lee, D. G. Hirtz, et al. Phenobarbital for febrile seizures:effects on intelligence and on seizure recurrence. N Engl J Med 322:364, 1990.
P.51
Favale, E., V. Rubino, P. Mainardi, et al. Anticonvulsant effect of fluoxetine in humans. Neurology 45:1926, 1995.
Fenwick, P., Is dyscontrol epilepsy? In What is Epilepsy? Eds. Trimble, M., E. Reynolds, Churchill Livingstone, 1986, pp. 139-160.
Fenwick, P. B. Antiepileptic drugs and their psychotropic effects. Epilepsia 33 (Suppl 6):33, 1992.
Ferrie, C. D., R. O. Robinson, P. Panayiotopoulos. Psychotic and severe behavioral reactions with vigabatrin: a review. Acta Neurol Scand 93:1, 1996.
Ferris, G. S. The recognition of non-epileptic seizures. South Med J 52:1557, 1972.
Fischer, M., G. Korskjaer, E. Pedersen. Psychotic episodes in Zarondan treatment. Effects and side effects in 105 patients. Epilepsia6:325, 1965.
Flor-Henry, P. Psychosis and temporal lobe epilepsy. Epilepsia 10:363, 1969.
Flor-Henry, P. Ictal and interictal psychiatric manifestations in epilepsy: specific or nonspecific? Epilepsia 13:773, 1972.
Freud, S. Collected Papers, Vol. 2. Hogarth, London, 1949.
Fulton, H., Discussion. Epilepsisa 2:77, 1953
Gastaut, H., H. Collomb. Etude du comportement sexuel chez les epileptiques psychomoteurs. Ann Med Psychol 112:657, 1954.
Gehlert, S. Perceptions of control in adults with epilepsy. Epilepsia 35:81, 1994.
Ghaemi, S. N., J. J. Katrow, S. P. Desai, F. K. Goodwin. Gabapentin treatment of mood disorders: a preliminary study. J Clin Psychiatry.59:426, 1998.
Gibbs, F., Ictal and nonictal psychiatric disorders in temporal lobe epilepsy. J Nerv Ment Dis 113:522, 1951.
Gibbs, F. A., E. C. Gibbs. Atlas of Electroencephalography, 2nd Ed., Addison-Wesley, Reading, MA, 1952.
Gilman, A., Rall, T, Nies, et al. The Pharmacological Basis of Therapeutics, 8th ed., Pergoamon Press, NY, pp. 436-462.
Giordani, B., J. C. Sackellares, S. Miller, et al. Improvement in neuropsychological performance in ptients with refractory seizures.Neurology 33:489, 1983.
Gissler, J., M. R. Jarvelin, P. Lonhiala, et al. Boys have more health problems in childhood than girls: follow-up of the 1987 Finnish birth cohort. Acta Paediatr 88:310, 1999.
Glaser, G. H. The problem of psychosis in psychomotor temporal lobe epileptics. Epilepsia 5:271, 1964.
Glaser, G. H. Limbic epilepsy in childhood. J Nerv Ment Dis 144:391, 1967.
Glaser, G., J. Pincus Limbic encephalitis. J Nerv Ment Dis 144:391, 1969.
Glaser, G. H., R. J. Newman, R. Schafer. Interictal psychosis in complex partial temporal lobe epilepsy: an EEG-psychological study. In:EEG and Behavior, G. H. Glaser, ed., Basic Books, New York, 1963, p. 345.
Goddard, G. V., D. C. Mclntyre, C. K. Leech. A permanent change in brain functioning resulting from daily electrical stimulation. Exp Neurol 25:295, 1969.
Goldensohn, E. S. The relevance of secondary epileptogenesis to the treatment epilepsy: kindling and the mirror focus. Eplepsia 25 Suppl 2:156, 1984.
Good, M. I. The concept of an organic dissociation syndrome: what is the evidence? Harvard Rev Psychiatry 1:145, 1993.
Goldman-Rakic, P., L. Selemon, M. Schwartz. Dual pathways connecting the dorsolateral prefrontal cortex with the hippocampal formation and parahippocampal cortex in the rhesus monkey. Neuroscience 12:719, 1984.
Gressen, S. P. Mechanisms of cerebral dysgenesis. Curr Opin Pediatr 10:556, 1998.
P.52
Gunderson, J. G., A. F. Frank, E. F. Ronningstam, et al. Early discontinuance of borderline patients from psychotherapy. J Nerv Ment Dis177:38, 1989.
Gunn J., J. Bonn. Griminality and violence in epileptic prisoners. Br J Psychiatry 118:337, 1971.
Gunn, J., G. Fenton. Epilepsy, automatism and crime. Lancet 1:171, 1971.
Gunn, J., Social factors and epileptics in prison. Br J Psychiatry 124:509, 1974.
Gunn, J., Epileptic homicide. Br J Psychiatry 132:510, 1978.
Gunn, J., Violence and epilepsy. New Eng J Med 306:299, 1982.
Gutnick, M. J., D. A. Prince. Effects of projected cortical epileptiform discharges on neuronal activities in ventrobasal thalamus of the cat; ictal discharge. Exp Neurol 46:418, 1975.
Guze, B. H., M. Gitlin The neuropathologic basis of major affective disorders: neuroanatomic insights. J Neuropsychiatry Clin Neurosci6:114, 1994.
Hancock, E., J. P. Osborn. Vigabatrin in the treatment of infantile spasms tuberous sclerosis: literature review. J Child Neurol 14:71, 1999.
Hart, Y. M., F. Andermann, Robitaille, et al. Double pathology in Rasmussen's Syndrome: a window on the etiology? Neurology 50:731, 1998.
Hauser, W. A., J. F. Annegers, L. T. Kurland. Febrile convulsions: Prognosis for subsequent seizures. Neurology 27:341, 1977.
Henry, T. R., I. Drury, Non-epileptic seizures in temporal lobectomy candidates with medically refractory seizures. Neurology 48:1374, 1997.
Hermann, B. P., P. Riel. Interictal personality and behavioral traits in temporal lobe generalized epilepsy. Cortex 17:125, 1981.
Hermann, B. P., M. Seidenberg, J. Schoenfeld, et al. Neuropsychological characteristics of the syndrome of mesial temporal lobe epilepsy.Arch Neurol 54:369, 1997.
Herman-Giddens, M. E., G. Brown, S. Verbiest, et al. Under ascertainment of child abuse mortality in the United States. JAMA 282:463, 1999.
Hierons, R., M. Saunders. Impotence in patients with temporal lobe lesions. Lancet 2: 761, 1966.
Hill D. The EEG in psychiatry. In: Electroencephalography, J.D.N. Hill, G. Parr, eds. Macmilan, Oxford, 1963, p. 368.
Himmelhoch, J., J. Pincus, G. Tucker. Behaviorial features of subacute encephalitis. Brit J Psychiat 116:531, 1970.
Holmes, G. L. Benign focal epilepsies of childhood. Epilepsia 34 Suppl 3:49, 1993.
Inouye, E. Observations on forty twin index cases with chronic epilepsy and their cotwins. J Nerv Ment Dis 130:401, 1960.
Isojarvi, J. L, M. Repo, A. J. Pakarinen Carbamazepine, phenytoin, sex hormones and sexuality. Epilepsia 36:366, 1995.
Ito, M. Antiepileptic drug treatment of West syndrome. Epilepsia 39 (Suppl 5):38, 1998.
Jay, T, J. Glowinski, A. Thierry. Inhibition of prefrontal cortex excitatory response by the mesocortical DA system. Neuroreport 6:1845, 1995.
Jensen, P., S. Jensen, P. Sorensen, et al. Sexual dysfunction in male and emale patients with epilepsy. Arch Sex Behav 19:1, 1990.
Johnson, M. V, G. W. Golstein. Selective vulnerability of the develoing brain to lead. Curr Opin Neurol 11:689, 1998.
Kalynchuke, L. E., J. P. Pinel, D. Treit, et al. Persistence of the interictal emotionality produced by long-term amygdala kindling in rats.Neuroscience 85:1311, 1998.
Kanner, A. M., S. Stagno, P. Kotagai, et al. Postictal psychiatric events during prolonged video-EEG monitoring. Arch Neurol 53:258, 1996.
P.53
Kanemoto, K., J. Takeuchi, J. Kawasaki, et al. Characteristics of temporal lobe epilepsy with mesial temporal sclerosis, with special reference to psychotic episodes. Neurology 47:1199, 1996.
Kanemoto, K., J. Kawasaki, E. Mori. A close relation between violence and epilepsy. Epilepsia 40:107, 1999.
Ketter, T., B. A. Malow, R. Flamini, et al. Anticonvulsant withdrawal-emergent psychopathology. Neurlogy 44:55, 1994.
King, D. W., B. B. Gallagher, A. J. Murvin, et al. Pseudoseizures: diagnostic evaluation. Neurology 32:18, 1982.
King, M. A., M. R. Newton, G. D. Jackson, et al. Epileptology of the first seizure presentation: a clinical, electroencephalographic and magnetic resonance imaging study of 300 consecutive patients. Lancet 352:1007, 1998.
Kluver, H., P. C. Bucy. Preliminary analysis of functions the temporal lobes in monkeys. Arch Neurol Psychiatry 42:979, 1939.
Kolorsky, A., K. Freund, J. Machek, O. Polak. Male sexual deviations. Arch Gen Psychiatry 17:735, 1967.
Koo, B. Vigabatrin in the treatment of infantile spasms. Pediatr Neurol 20:106, 1999.
Krishnamoorthy, E. S., M. R. Trimble. Forced normalization: clinical and therapeutic relevance. Epilepsia 40 (Suppl 10):57, 1999.
Kwan, P., M. J. Brodie. Epilepsy after the first drug fails: substitution or add-on? Seizure 9:464, 2000.
Laskowitz, D. T., M. R. Sperling, J. A. French, et al. The syndrome of frontal lobe epilepsy: characteristics and surgical management.Neurology 45:780, 1995.
Lennox, W. G. The heredity of epilepsy as told by relatives and twins. JAMA 146:529, 1951.
Lennox-Buchtal, M. Febrile convulsions: A reappraisal. Electroenceph. Clin Neurophysiol Suppl 32:1, 1973.
Lesser, R. P., H. Leuders, D. D. Dinner. Evidence for epilepsy is rare in patients with psychogenic seizures. Neurology 33:502, 1983.
Lewis, D. O., J. H. Pincus, S. S. Shanok, G. H. Glaser. Psychomotor epilepsy and violence in a group of incarcerated adolescent boys. Am J Psychiatry 139:882, 1982.
Li, L. M., D. R. Fish, S. M. Sisodiya, et al. High resolution magnetic resonance imaging in adults with partial or secondary generalized epilepsy attending a tertiary referral unit. J Neurol Neurosurg Psychiatry 59:384, 1995.
Lilienfeld, A. M., B. Psamanick. Association of maternal and fetal factors with the development of epilepsy. JAMA 155:719, 1954.
Lin, Y. Y., M. S., Su, C. H., Yiu. Relationship between mesial temporal seizure focus and elevated prolactin in temporal lobe epilepsy.Neuorlogy 49:528, 1997.
Liporace, J., W. Tatum IV, C. L. Morris III, et al. Clinical utility of sleep deprived versus somputer assisted ambulatory 16-channel EEG in epilepsy patients: a multi center study. Epilepsy Res 32:357, 1998.
MacLean, P. D., D. W. Ploog. Cerebral representation of penile erection. J Neurophysiol 25:29, 1962.
MaClean, P. The limbic system and its hippocampal formation. J Neurosurg 11:29, 1954.
Mahaffey, K. R., J. L. Annest, J. Robert, et al. National estimates of blood lead levels:United States 1976-1980: Association with selected demographic and socioeconomic factors. N. Engl J Med 307:573, 1982.
P.54
Malamud, N. Psychiatric disorder with intracranial tumors of limbic system. Arch Neurol 17: 113, 1967.
Malkowicz, D. E., A. Legido, R. A., Jackel, et al. Prolactin secretion after repetitive seizures. Neuorlogy 45:448, 1995.
Marcotte, D. Use of topiramat, a new anti-epileptic as a mood stabilizer. J Affect Disord 50:245, 1998.
Margerison, J. H., J. Corsellis. Epilepsy and the temporal lobes. Brain 89:499, 1966.
Mark, V. H., F. R. Ervin. Violence and the Brain. Harper and Row, New York, 1970.
Marshall, A. G., E. O. Hutchinson, J. Honisett. Heredity in common diseases: a retrospective survey of twins in a hospital population. Br Med 71:1, 1962.
Martin, R., R. Kuzniecky, S. Ho, et al. Cognitive effects of topiramate, gabapentin and lamotrigine in healthy young adults. Neurology52:321, 1999.
Mathers, C. Group therapy in the management of epilepsy. Br J Med Psychol 65:279, 1992.
Matthews, W., Barabas, G: Suicide and epilepsy. Psychosomatics 22:515-524, 1981.
Matsuura, M. Epileptic psychosis and anticonvulsant drug treatment. J Neurol Neurosurg Psychiatry 67:231, 1999.
Mattson, R. H., K. L. Pratt, J. R. Calverley. Electroencephalograms of epileptics following sleep deprivation. Arch Neurol 13:310, 1965.
Mattson, R. H., J. A. Cramer, J. F. Collins. A comparison of valproate with carbamazapine for the treatment of partial complex seizures and secondarily generalized tonicclonic seizures in adults. W Engl J Med 327:765, 1992.
Maxmen, J., N. Ward, Psychotropic Drugs Fast Facts, 2nd ed., Norton, NY, 1995.
McConnell, H., P. J. Snyder, J. D. Duffy, et al. Neuropsychiatric side effects related to treatment with felbamate. J Neuropsychiatry Clin Neurosci 8:341, 1996.
McConnel, H., Snyder, P. (eds). Psychiatric Comorbidity in Epilepsy, American Psychiatric Press, Washington, DC, 1998.
McElligott, R. F. An analysis of the Epilepsy Center at Veteran's Admiistration Hospital, West Haven, Connecticut, Master's Thesis, Yale University School of Medicine, 1974.
McKenna, P., J. Kane, K. Parrish, et al. Psychotic syndromes in epilepsy. Am J Psychiatry 142:895, 1985.
McNamara, M. E., B. S. Fogel. Anticonvulsant-responsive panic attacks with temporal lobe EEG abnormalities. J Neuropsychiatry Clin Neurosci 2: 193, 1990.
Meldrum, B. Physiological changes during prolonged seizures and epileptic brain damage. Neuropadiatrie 9:203, 1978.
Mendez, M. E, R. C. Doss, J. L. Taylor. Interictal violence in epilepsy. Relationship to behavioral science variables. J Nerv Ment Dis181:566, 1993.
Mendez, M. F., J. L. Taylor, R. C. Doss, et al. Depression in secondary epilepsy: relation to lesion lateratlity. J Neurol Neurosurg Psychiatry 57:232, 1994.
Mesulam, M. Principles of Behavioral and Cognitive Neurology, 2nd Ed. Oxford University Press, New York, 2000, p. 55.
Metrakos, K., J. D. Metrakos. Genetics of convulsive disorders: II Genetic and encephalographic studies in centrencephalic epilepsy.Neurology 11:474, 1961.
Mignone, R. J., E. F. Donnelly, D. Sadowsky. Psychological and neurological comparisons of psychomotor and nonpsychomotor epileptic patients. Epilepsia 11:345, 1970.
Multicenter Study (no author) Neurology 45:224, 1995.
Murialdo G., C. A. Galimberti, S. Fonzi, et al. Sex hormones and pituitary function in male epileptics. Epilepsia 36:360, 1995.
P.55
Nelson, K. B., J. H. Ellenberg. Predictors of epilepsy in children who have experienced febrile seizures. N Engl J Med 295:1029, 1976.
Neugebauer, R., M. M. Weissman, R. Ouellette, et al. Comorbidity of panic disorder and seizures. J Anxiety Disorders 7:21, 1993.
Ney, G. C., C. Zimerman, N. Schaul, Psychogenic status epilepticus induced by a provocative technique. Neurology 46:546, 1996.
Ounstead, C., J. Lindsay, and R. Norman. Biologic Factors in Temporal Lobe Epilepsy. William Heinemann Medical Books, London, 1966.
Ounsted C. Aggression and epilepsy rage in children with temporal lobe epilepsy. J Psychosom Res 13:237, 1969.
Pakalnis, A., M. E. Drake, K. John, et al. Forced normalization. Arch Neurol 44:289, 1987.
Papez, J. W. A proposed mechanism of emotion. Arch Neurol Psychiatry 38:725, 1937.
Perrine, K., B. P. Hermann, K. J. Meador, et al. The relationship of neuropsychological functioning to quality of life in epilepsy. Arch Neurol 52:997, 1995.
Petersen, I., O. Erg-Olofsson, U. Selder. Paroxysmal activity in EEG of normal children. In: Cinical Electroencephalography of Children, P. Kellaway, I. Petersen eds. Grune & Stratton, New York, 1968, p. 167.
Pirkle, J. L., R. B. Kaufman, D. J. Brody, et al. Exposure of the US population to lead, 1991-1994. Enviorn Health Perspect 106:745, 1998.
Polkey, C., Epilepsy surgery. Acta Neurol Scand Suppl. 152:183, 1994.
Pond, D. A. Psychiatric aspects of epilepsy. J Ind Med Prof 3:1441, 1957.
Popkin, M and G. Tucker, Mental disorders due to a general medical condition and substance induced disorders. Eds widiger, T, A. Frances, h. Pincus et al. In DSM-IV Sourcebook, American Psychiatric Press, Washington DC, 1994, p. 243.
Porrino, L. and P. Goldman-Rakic, Brainstem innervation of prefrontal and anterior cingulate cortex in the rhesus monkey revealed by retrograde transport of HRP. J Comp Neurol 10:63, 1982.
Post, R. M., T. A. Ketter, P. J. Pazzaglia, et al. Rational polypharmacy in the bipolar affective disorders. Epilepsy Res Suppl 11:153, 1996.
Post, R. M., S. R. Weiss, D. M. Chuang. Mechanisms of action anticonvulsants in affective disorders: comparisons with lithium. J Clin Psychopharmacol 12:23S, 1992.
Prichard, P. B., III, B. B. Wannamaker, J. Sagel, et al. Endocrin function following complex partial seizures. Ann Neurol 14:27, 1983.
Ramani, V., F. F. Quesney, D. Olson, et al. Diagnosis of hysterical seizures in epileptic patients. Am J Psychiatry 137:705, 1980.
Rapin, I. and D. Allen. Syndromes in developmental dysphasia and adult aphasia. Res Pub Assoc Res Nerv Ment Dis 66:57, 1988.
Reed, D. M., F. J. Staley. The Epidemiology of Prematurity. Urban and Schwarzenberg, Baltimore, 1977.
Regan, K., G. Banks, and R. Beran. Therapeutic recreation programs for children with epilepsy. Seizure 2:195, 1993.
Reeves, A. G., F. Plum. Hyperphagia, rage and demential accompanying a ventromedial hypothalamic neoplasm. Arch Neurol 20:616, 1969.
Reynolds, E. H. Anticonvulsant drugs, folic acid metabolism, fit frequency and psychiatric illness. Psychiatr Neurol Neurochirg 74:167, 1971.
Reynolds, E. H., R. D. Travers. Serum anticonvulsant concentrations in epileptic patients with mental symptoms. Brit J Psychiat 124:440, 1974.
Riley, T. L., W. L. Brannon. Recognition of pseudoseizures. J Fam Pract 10:213, 1980.
P.56
Robertson, M. Mood disorders associated with epilepsy, in McConnel, H., Snyder, P., (eds) Psychiatric Comorbidity in Epilepsy, Washington, DC, American Psychiatric Press, 1998, pp. 133-168.
Rodin, E. A. The Prognosis of Patients with Epilepsy. C C Thomas, Springfield, IL, 1968.
Rodin, E. A. Psychomotor epilepsy and aggressive behavior. Arch Gen Psychiatry 28:210, 1973.
Rodin, E. A., M. Katz, K. Lennox. Differences between patients with temporal lobe seizures and those with other forms of epileptic attacks. Epilepsia 17:313, 1976.
Rossen, R. Electroencephalographic studies in idiopathic epilepsy, idiopathic syncope, and related disorders in a U.S. Naval Hospital. Am J Psychiatry. 107:391, 1974.
Roy, A. Hysterical seizures. Arch Neurol 36:447, 1979.
Sachdev, P. Schizophrenia-like psychoses and epilepsy: The status of the association. Am J Psychiatry 155:325, 1998.
Saint-Hilaire, J. M., M. Gilbert, G. Bouner. Aggression as an epileptic manifestation: two cases with depth electrode study. Epilepsia21:184, 1980.
Schiff, H. B., T. D. Sabin, A. Geller, et al. Lithium in aggressive behavior. Am J Psychiat 139:1346, 1982.
Schiottz-Christensen, E., P. Bruhn. Intelligence, behavior and scholastic achievement subsequent to febrile convulsions: An analysis of discordant twin pairs. Dev Med Child Neurol 15:565, 1973.
Schmitz, E. B., M. M. Robertson, M. R.Trimble. Depression and schizophrenia in epileps: social and biological risk factors. Epilepsy Res35:59, 1999.
Schorner, W., H. J. Meencke R. Felix. Temporal lobe epilepsy: comparison of CT and MR imaging. Am J Roent 149:1231, 1987.
Scolnik, D., I. Nulman, J. Rovet, et al. Neurodevelopment of children exposed in utero to phenytoin and carbamazepine monotherapy.JAMA 271:767, 1994.
Septien, L., P. Gras, M. Giroud, et al. Depression and temporal lobe epilepsy. The possible role of laterality the epileptic foci and gender.Neurophysiol Clin 23:327, 1993.
Serafetinides, E. A. Aggressiveness in temporal lobe epileptics and its relation to cerebral dysfunction and environmental factors. Epilepsia 6:33, 1965.
Shamansky, S. L., G. H. Glaser. Socioeconomic characteristics of childhood seizure disorders in the New Haven area: An epidemiologic study. Epilepsia 20:457, 1979.
Shaywitz, S. E., B. A. Shaywitz, K. R. Pugh et al. Functional disorganization for reading in the brain in dyslexia. Proc Natl Acad Sci USA95:2636, 1998.
Silveira, D., E. Souza, J. Carvalho, et al. Interictal Hyposexuality inmale patients with epilepsy. Arq Neuropsiquistr 59:23, 2001.
Slater, E., A. W. Beard, E. Glithero. The schizophrenia-like psychosis of epilepsy. Br J Psychiatry 109:95, 1963.
Small, J. G., V. Milstein, J. R. Stevens. Are complex partial epileptics different: Arch Neurol 7:187, 1962.
Smith, J., Episodic rage in Limbic Epilepsy and the Dyscontrol Syndrome. M. Grigis and L. Kiloh, eds. Elsevier/North, Holland, Biomedical Press, 1980, p. 255.
Smith, D. B., R. H. Mattson, J. A. Cramer, et al. Results of a nationwide Veteran's Administration Cooperative Study comparing the efficacy and toxicity of carbamazapine, phenobarbitol, phenytoin and primidone. Epilepsia 28 (Suppl 3):550, 1987.
Sommerville, E. and J. Bruni. Tonic status epilepticus presenting as confusionsal state. Ann Neurol 13:549, 1983.
P.57
Southall, D. P., M. C. Plunkett, M. W. Banks, et al. Covert video recordings of life threatening child abuse: lessons for child protection.Pediatrics 100:735, 1997.
Spanaki, M. V., I. G. Zubal, J. MacMullen, et al. Periictal SPECT localization verified by simultaneous intracranial EEG. Epilepsia 40:267, 1999.
Spencer, S. S. Depth electrode encephalography in selection of refractory epileptics for surgery. Ann Neurol 9:207, 1981.
Spencer, S. S., D. D. Spencer, P. D. Williamson, R. H. Mattson. The localizing value of depth electroencephalography in 32 refractory epileptic patients. Ann Neurol 12:248, 1982.
Spencer, S. S. Sexual automatisms in complex partial seizures. Neurology 33:527, 1983.
Spencer, S. S., D. D. Spencer, G. H. Glaser, et al. More intense focal seizure types after callosal section. The role of inhibition. Ann Neurol 16:686, 1984.
Sperling, M. R., M. J. O'Connor, A. J. Saykin, et al. Temporal lobectomy for refractory epilepsy. JAMA 276:479, 1996.
Spitz, M. C. Panic disorder in seizure patients: a diagnostic pitfall. Epilepsia 32:33, 1991.
Srole, I., et al. Mental Health in the Metropolis. McGraw-Hill, New York, 1962.
Staden, U., E. Isaacs, S. G. Boyd, et al. Language disfunction in children with Rolandic Epilepsy. Neuropediatrics 29:242, 1998.
Standage, K. F. The etiology of hysterical seizures. Can Psychiatr Assoc 20:67, 1975.
Standage, K. G., G. W. Fenton. Psychiatric symptom profiles of patients with epilepsy. Psychol Med 5:152, 1975.
Stark-Adamec, C., R. Adamec, J. Graham et al. Complexities in the complex partial seizures personality controversy. Journal of the University Ottawa 10:231, 1985.
Stevens, J., Psychiatric implications of psychomotor epilepsy. Arch Gen Psychiat 14:461, 1966.
Stevens, J., V. Mark, F. Erwin, et al. Deep temporal stimulation in man. Arch Neurol 21:157, 1969.
Stevens J. R., Interictal clinical manifestations of complex partial seizures. Radiology 117:113, 1975.
Stevens, J. R, Hermann, B. P. Temporal lobe epilepsy, psychopathology, and violence: the state of the evidence. Neurology 31:1127, 1981.
Strauss, E., J. Wada, A. Moll. Depression in male and female subjects with complex partial seizures. Arch Neurol 49:391, 1992.
Suter, C. Medical-legal aspects of EEG. Presdntation course at American EEG Society, 34th Annual Meeting, Boston, 1980.
Taylor, D. C. Sexual behavior and temporal lobe epilepsy. Arch Neurol 21:510, 1969.
Taylor, D. C. Factors influencing the occurrence of schizophrenia-like psychosis in patients wit htemporal lobe epilepsy. Psychol Med5:249, 1975.
Teicher, M. H., C. A. Glod, J. Surrey. Early childhood abuse and limbic system ratings in adult psychiatric outpatients. J Neuropsychiatry Clin Neurosci 5:301, 1993.
Theodore, W. H. R. J. Porter. Removal of sedative-hypnotic antiepileptic drugs from the regimen of patients with intractable epilepsy.Ann Neurol 13:320, 1983.
Thomas P. B. Zifkin, O. Migneco, et al. Nonconvulsive status epilepticus of frontal origin. Neurology 52:1174, 1999.
Thierry, A., Y. Gioanni, E. Degenetais, et al. Hippocampo-prefrontal cortex pathway. Hippocampus 10:411, 2000.
P.58
Tranel, D. Acquired sociopathy, the development of sociopathic behavior following focal brain damage. In: Progress in Experimental Personality and Psychopathology Research, D.C., Fowles, P. Sutker, S. H. Goodman, eds. New York Springer, 1994.
Trimble, M. Non-monoamine oxidase inhibitor antidepressant and epilepsy: A review. Epilepsia 19:241, 1978.
Trimble, MR. Anticonvulsant drugs and cognitive function: a review of the literature. Epilepsia 28 (suppl):S37-S45, 1987.
Trimble, M., Cull, CA. Antiepileptic drugs, cognitive function, and behavior in children. Clev Clin J Med 56 Suppl: S140, 1989.
Trimble M. Epilepsy and behavior. Epilepsy Res 10: 71, 1991.
Trojaborg, W. Changes in spike foci children. In: Clinical Electroencephalography in Children, P. Kellaway, ed. Grune and Stratton, New York, 1968.
Troupin, A. S., L. M. Ojemann, L. M. Halpern, et al. Carbamazepine (Tegretol): A double blind comparison and phenytoin (Dilantin).Neurology 26:342, 1976.
Tucker, G. J. Seizure disorders presenting with psychiatric symptomatology. Psychiatr Clin North Am 21:624, 1998.
Tucker, G. J., T. Detre, M. Harrow, H. Glaser. Behavior and symptoms of psychiatric patients ands the electroencephalogram. Arch Gen Psychiat 12:278, 1965.
Van Elst, L., F. Woerman, L. Lemieux, et al. Affective aggression in patients with temporal lobe epilepsy: a quantiative study of the amygdala. Brain 123:234, 2000.
Varney, N. R., M. J. Garvey, B. L. Cook, et al. Identification of treatment-resistant depressives who respond favorably to carbamazepine.Ann Clin Psychiatry 5:117, 1993.
Velasco, F., M. Velasco, A. L. et al. Electrical stimulation of the centromedian thalamic nuclei in control of seizures: long term studies.Epilepsia 36:63, 1995.
Victaroff, J. L, F. Benson, S. T. Grafton, et al. Depression in complex partial seizures. Electroencephalography and cerebral metabolic correlates. Arch Neurol 51:155, 1994.
Walden, J., B. Hesslinger, D. VonKalker, et al. Addition of lamotrigine to valproate may enhance efficacy in the treatment of bipolar disorder. Pharmacopsychiatry 29:193, 1996.
Watson, C, C. R. Jack, F. Cendes. Volumetric MRI. Arch Neurol 54: 521, 1997.
Waxman, S. G., N. Geschwind The interictal behavior syndrome of temporal lobe epilepsy. Arch Gen Psychiatry 32:1580, 1975.
Weilberg, J. B., S. Schacter, Worth, et al. EEG abnormalities in patients with atypical panic attacks. J Clin Psychiatry 56:358, 1995.
Wiebe S., W. T. Blume, J. P. Girvin, et al. A randomized, controlled trial of surgery for temporal lobe epilepsy. N Engl J Med 345:311, 2001.
Wilder, G. J., E. R. Ramsey, J. Murphy, et al. Comparison of valproic acid and phenytoin in newly diagnosed tonic-clonic seizures.Neurology 12:1474, 1983.
Williams, D. Neural factors related to habitual aggression. Brain 92:50, 1969.
Williamson, P. D., S. S. Spencer. Clinical and EEG features of complex partial seizures of extratemporal origin. Epilepsia 27 (Suppl 2): 46, 1986.
Williamson, P. D., D. D. Spencer, S. S. et al. Complex partial seizures of frontal lobe origin. Ann Neurol 18:497, 1985a.
Williamson, P. D., D. Spencer, S. S. et al. Complex partial status epilepticus: A depth electrode study. Ann Neurol 18:647, 1985b.
Williamson, P. D., V. M. Thadani, T. M. Darcey, et al. Occipital lobe epilepsy: clinical characteristics, seizure spread patterns and results of surgery. Ann Neurol 31:3, 1992.
Wilson, S. J., P. Kincade, M. M. Saling, et al. Patient readmission and support utilization following anterior temporal lobectomy. Seizure8:20, 1999.
Wirrell, E. C. Benign epilepsy of childhood with centrotemporal spikes. Epilepsia 39 (Suppl 4):S32–41 1998.
P.59
Woermann F., L. Van Elst, M. Koepp, et al. Reduction of frontal neocortical grey matter associated with affective aggression in patients with temporal lobe epilepsy: an objective voxel by analysis of automatically segmented MRI. J Neurol Neurosurg Psychiatry 68: 162, 2000.
Wyllie, E., J. P. Glazer, S. Benbadis, et al. Psychiatric features of children and adolescents with pseudo seizures. Arch Pediatr Adolesc Med 153:244, 1999.
Zivin, L., C. Aimone Marsan. Incidence and prognostic significance of epileptiform activity in the EEG of non-epileptic subjects. Brain91:751, 1968.