A diagnosis of epilepsy can be made in most patients on the basis of the clinical history and the exclusion by physical, neurologic, and laboratory examinations of other disorders known to be associated with seizures - e.g., intracranial neoplasm, hypoglycemia, and hypocalcemia.
In our center, the initial work-up on a patient with seizures consists of a detailed medical history, general physical and neurologic examinations, complete blood-cell counts (including a platelet count or at least an estimation of the number of platelets), blood chemistries consisting of fasting-blood-sugar and calcium determinations, and kidney- and liver-function tests. (These laboratory examinations serve as a means of detecting disorders other than epilepsy and also as a baseline to be utilized in the follow-up of patients who are treated with anticonvulsant medications.)
Also included in the initial work-up is a urinalysis, roentgenograms of the skull, an electroencephalographic study, a lumbar puncture when indicated, and specific cerebral diagnostic procedures when indicated.
Medical history. The importance of a comprehensive medical history cannot be overemphasized, and the physician is urged to devote adequate time and effort to this part of the overall evaluative procedure. Special attention should be directed to the child's age at onset of the seizures, the occurrence of an aura, and the ictal and postictal features of the attack. The physician should ask detailed questions about any events or other factors that may have been precipitating factors immediately preceding the seizure episode.
Age at onset. In some cases, the age at onset of the symptoms may provide the physician with an initial or preliminary diagnostic clue. For example, the occurrence of hysterical convulsions in a two-year-old child would be most unusual, and convulsions associated with a brain tumor are more likely to be observed in an adult than in a very young child.
Aura. An aura, or prodrome, is a disturbance that sometimes precedes an epileptic seizure (usually a grand- mal convulsion) and is an integral part of the attack. It is usually very short in duration, generally lasting a minute or so. The nature and characteristics of the aura vary from patient to patient but remain constant in individual patients.
Auras may be broadly classified as:
1. Sensory. These auras may appear in the following forms: auditory, olfactory, somatosensory, or visual. The somatosensory type is the most frequently encountered aura in children.
2. Psychic. Some patients exhibit such disturbances as increased irritability or restlessness, fear, anxieties, mental confusion or dullness, and various types of bizarre behavior before the occurrence of a seizure. Although these auras are generally brief, usually persisting for several minutes, we have observed some epileptic children who displayed personality changes and behavioral aberrations for two or three days before the onset of a major seizure.
3. Motor. Some patients present muscular twitching or spasms of isolated muscles or muscle groups or a series of generalized myoclonic jerks before the onset of a major motor seizure. The myoclonic phenomena, especially when generalized, most commonly begin when the patient awakens and persist for varying periods of time. They usually terminate in a grand-mal seizure.
We have found the presence or absence of an aura to be very useful at times in differentiating epileptic seizures from functional disorders, such as syncopal attacks.
Ictal characteristics of the epileptic seizures most frequently observed in children are presented in Table 1.
Postictal features are encountered most commonly in patients with major motor epilepsy. At the end of a grand-mal seizure, usually of prolonged duration, the patient often passes into a deep sleep or coma and, on recovery, manifests a variety of postconvulsive or postictal features. Generalized weakness, nausea, vomiting, severe headache, muscle soreness, general fatigue, marked restlessness, increased irritability, and various types of abnormal behavior are common. Impaired speech, mental confusion, and transient muscular paralysis are also frequently observed. In some patients, postconvulsive disturbances may persist for several days or longer. Postictal symptoms differ from one patient to another but usually remain the same in a given individual.
Precipitating factors. In some patients, epileptic seizures are associated with certain specific circumstances, events, situations, or other precipitating factors (Table 2). Diagnostically, recognition of the précipitants of seizures in a patient provides valuable information, since in many instances the attacks can then be reproduced under conditions of optimal observation so that details relative to the provocation and onset of the seizure can be studied.
There are a number of other précipitants of epileptic seizures that, while uncommon, are significant and of interest. These include:
CLINICAL AND EEG FEATURES OF EPILEPTIC SEIZURES
Flickering light. Epileptic seizures have been reported to occur when patients are suddenly exposed to bright light and in some instances after patients have been subjected to light for a period of time. Most investigators, however, agree that photic seizures occur most frequently in association with a flickering light. This observation has been confirmed in the electroencephalographic laboratory with the use of the stroboscope. Among the sources of intermittent light that have been designated as précipitants of epileptic seizures are, in addition to that produced by the stroboscope, television sets,1 rhythmically interrupted sunlight (such as that seen shining through the branches of trees or reflected from undulating waves of water2), and sunlight interrupted by the patient himself (by blinking his eyes while standing exposed to the sun or waving his hand before his eyes - so-called self-induced epilepsy3).
Patterns. Keith and his co-workers4 have described a patient in whom electroencephalographic changes and clinical epileptic seizures were precipitated by the visual stimulation produced by looking at fine mesh patterns.
Music. Critchley5 has reported several patients who had epileptic seizures after listening to certain types of music. This type of seizure disorder has been classified as musicogenic epilepsy.
Reading. Bickford et al·6 have reported clinical and electroencephalographic evidence of epileptic seizures precipitated by reading - a syndrome designated as reading epilepsy.
Sensations and tactile stimulation. We have seen many patients in whom clinical seizures and concurrent electroencephalographic epileptic discharges were precipitated by sudden noise or other startling sensations and by tactile stimuli. Similar instances of sensory precipitation epilepsy (sometimes referred to as reflex epilepsy) have been reported by others.7,8
PRECIPITANTS OF EPILEPTIC SEIZURES
The resumes of the following case histories may help illustrate the diagnostic relevance of precipitating factors.
Case History No. 1.
A nine-year old girl had experienced "passing-out spells" for a period of several years. The episodes occurred in the morning, just as she left her house for school.
Her parents* first reaction was that she was "faking" and simply did not want to go to school. The family physician considered a diagnosis of school phobia or malingering. The child was referred to our clinic for evaluation. Detailed interrogation of the patient and her parents revealed that she experienced these "passing-out spells" only on sunny days and never on cloudy or sunless occasions. Diagnostic investigations, including sudden exposure to bright light during a routine EEG examination, established that these "passing-out" episodes were epileptic seizures triggered by sudden exposure to direct sunlight when the child stepped out of her house in the mornings.9
Case History No. 2.
A seven-year-old boy experienced a seizure on several occasions while watching television. The parents thought that the attacks were "caused" by the excitement and tension engendered by the television programs. The family physician was consulted. The results of routine physical, neurologic, and laboratory examinations were within normal limits. An EEG study, which did not include intermittent photic stimulation, revealed normal findings. The physician considered that the seizures were probably functional in origin but referred the patient to us for further investigation. Exhaustive questioning elicited the fact that the television set was defective and flickered frequently. In addition, it was established that the boy experienced a seizure only when the set was malfunctioning - i.e., when the normal picture was replaced or accompanied by flickering lines and/or light. We subsequently confirmed that this child was sensitive to photic stimulation and postulated that his seizures were precipitated by the flickering light of the television set.1
Case History No. 3.
A five-year-old girl with a seizure disorder of recent onset was referred to us for evaluation. The results of the physical, neurologic, and laboratory examinations were within normal limits. During the obtention of the clinical history, the patient's mother remarked, "I don't know how to explain it, but it seems like my daughter only has a seizure when Tm wearing this skirt, which I bought about a month or two ago." The mother's patterned skirt consisted of small, colorful, finely lined boxes on a solid background. We then considered the possibility that this patient was sensitive to visual stimulation produced by the pattern of her mother's garment.
Later in the day, the mother walked into the EEG laboratory, where her daughter was seated on a chair having the electrodes affixed to her head by the EEG technician, and the child experienced a clinical seizure after staring at the patterned skirt for a short period. During the EEG examination, the child was instructed to gaze intently at her mother's skirt, which was appropriately displayed by the EEG technician. After briefly staring at the garment, the patient had a clinical seizure with concomitant EEG abnormalities. Exhibition of a solid-colored skirt during a subsequent portion of the recording induced neither a clinical attack nor EEG changes. Clinical epileptic seizures with concurrent EEG abnormalities were provoked on subsequent occasions by exposing the child to her mother's finely patterned skirt.
General Diagnostic Tools
Skull radiography. We obtain x-ray films of the skull in every patient, with the exception of children with a simple febrile convulsive disorder or with "pure" petit-mal epilepsy (e.g., three-per-second spike-and-wave dysrhythmia).
Roentgenograms of the head do not reveal abnormalities in the majority of cases that will be of use in determining the cause of the seizure disorder. But on rare occasions they do, and in such instances they are invaluable. Over the past years we have had at least 10 patients in which the origin of the seizure disorder was found because of such x-rays. In each case the manifestations were not detectable on clinical examination. In one patient there was a localized thinning of the frontal area of the skull due to an intracranial neoplasm. Four had intracranial calcifications from brain tumors, and two had intracranial calcifications from tuberous sclerosis. In the other three the causative factors were, respectively, Sturge-Weber syndrome (Figure 1), toxoplasmosis (Figure 2), and cysticercosis (Figure 3).
Skull radiography is a simple procedure. Since it is impossible to predict when such examinations will yield significant or unsuspected findings, we believe at least three views of the skull should be taken routinely - Towne, Caldwell, and lateral views.
Figure 1. Intracranial calcifications. These were discovered by routine skull roentgenography of a 17-year-old patient with convulsions. The boy had the Sturge-Weber syndrome but without the characteristic facial hemangioma (A). In lateral x-ray (B), calcifications in the occipital area typical of the Sturge-Weber syndrome can be seen clearly on the actual x-ray, although they may not be quite obvious in this reproduction. A poste roanterior view (not shown) indicated calcifications were limited to the right side.
Figure 2. Toxoplasmosis. X-ray of the skull showed several areas of cerebral calcification. The patient presented the tetrad of findings of infantile toxoplasmosis - onset of symptoms at birth or shortly thereafter; bilateral focal chorioretinituis, usually affecting both maculae; convulsions; and evidence of cerebral calcification. Convulsions began when the child was two, and the diagnosis was confirmed by a positive neutralizing antibody test.
Figure 3. Cystlcercosls. Note small calcified nodules in the brain of this 11 -year-old boy. Generalized major motor convulsions had started when he was eight. The parasitic tapeworm (Cysticercus) was confirmed by biopsy of a nodule at the time of craniotomy.
Electroencephalography. In our center, a routine electroencephalographic examination consists of recordings obtained during (1) the awake state, (2) natural or artificially induced sleep, (3) hyperventilation, and (4) intermittent photic stimulation. Special techniques may be of considerable value in selected cases, such as the use of nasopharyngeal or sphenoidal electrodes in patients suspected of having psychomotor epilepsy but whose surface electroencephalographic examinations revealed normal states.
Misunderstandings concerning the practical value of the electroencephalogram in epilepsy are still widespread. Some physicians have the impression that the electroencephalogram invariably supplies definitive and relevant information in all patients suspected of having an epileptic disorder. It is true that, in certain forms of epilepsy, the electroencephalographic study reveals characteristic abnormalities that are diagnostically significant, such as the spike-wave discharge of petit-mal epilepsy and the hypsarrhythmic pattern of infantile myoclonic epilepsy. However, in other types of seizure disorders - particularly grand-mal epilepsy - the interictal electroencephalogram may show equivocal findings or may be normal. On the basis of serial electroencephalographic recordings performed over many years on thousands of patients, we now know that the incidence of normal electroencephalographic findings in patients with definite clinical evidence of grand-mal epilepsy is exceedingly high, especially in young children.
Electroencephalography should be viewed as a valuable adjunct to, but not a substitute for, other methods of diagnosis, since it has definite limitations, particularly in major motor epilepsy. For example, in a five-yearold child who has recurrent grand- mal convulsions and in whom appropriate investigative procedures have failed to elicit a cause, a diagnosis of a convulsive disorder of undetermined origin (epilepsy) should be made, regardless of whether the electroencephalographic examination yields normal or abnormal findings.
We feel strongly that a diagnosis of epilepsy should not be made on the basis of electroencephalographic findings interpreted as "equivocal, " "borderline," "minimally abnormal," or "slightly irregular." The policy of predicating a diagnosis of epilepsy solely on dubious electrographic irregularities in patients whose clinical symptoms are only suggestive of epilepsy warrants severe deprecation, primarily because of the serious social and economic penalties that almost automatically accompany the diagnosis.
The results obtained from an electroencephalographic study can in some instances provide the "key" to the diagnosis. For example, (1) in a subject who has had an attack that was not completely observed or was poorly described by the historian; (2) in patients who present such symptoms as recurrent behavioral aberrations or bizarre motor movements that the physician is unable to clinically classify as either psychomotor epilepsy or an emotional disorder; (3) in patients with such disturbances as recurrent attacks of abdominal pain, dizziness, fainting spells, or headache in whom the physician is unable to cVmicaWy differentiate autonomic epilepsy from a functional disturbance.
Electroencephalography has proved to be exceedingly valuable in classifying the various types of epileptic seizures, particularly those of brief duration (Table 3). Ability to distinguish between the different forms of epileptic seizures is important because drugs that are capable of controlling one type of seizure may be ineffective against other forms and may, in fact, increase their frequency.
In general, the electroencephalogram has little to offer relative to prognostic implications.10"12 Information from an electroencephalographic examination should rarely be used as a complete indicator of a patient's progress10,13 except in petit mal, in which there is almost invariably a direct relationship between the disappearance of the classic electrical discharge and the control of clinical spells,14 and in myoclonic epilepsy of infancy, in which the absence of the typical hypsarrhythmic pattern concurrently with the cessation of attacks has been noted in some cases.15'16
Over the past four decades, we have performed serial electroencephalographic examinations on many thousands of epileptic patients, and, aside from those with petit-mal spells and some with infantile myoclonic seizures, we have not observed a direct relationship between electroencephalographic findings and clinical status in a large number of cases. In many instances, the electroencephalogram became more abnormal even though the patient experienced fewer or no seizures. In others, the electroencephalogram continued to reveal abnormalities despite the fact that the patient had remained free of seizures for a period of years. Similar experiences have recently been reported by Hopkins and Scambler.10 It should also be noted that, in many children, the electroencephalogram tends to revert to normal during adolescence, regardless of their clinical progress. We emphasize that the results of electroencephalographic examinations are not prognostic for the time after termination of drug therapy and that the presence of specific electroencephalographic abnormalities usually need not interdict discontinuance of an anticonvulsant- medication-withdrawal program.11,17,18
Lumbar puncture. We perform a spinal tap only in patients who present neurologic signs or symptoms pointing to intracranial disease other than the seizure itself. There has been a relative paucity of significant information obtained from lumbar punctures on thousands of our patients, and the examination is not a completely innocuous procedure. In addition, it is often psychologically traumatic to the child and/or the parents.
Specific Diagnostic Procedures
Signs or symptoms of an intracranial neoplasm, cerebral vascular anomaly, or other specific types of cerebral pathology are indications for such cerebral examinations as arteriography, isotopie brain scanning, and pneumoencephalography. Since the indications for these conditions are well known to pediatricians, they will not be discussed further here.
Echoencephalography and computerized axial tomography are two other noninvasive diagnostic tools we have found useful in certain patients.
EEG FEATURES THAT ASSIST IN DIFFERENTIATING BRIEF EPILEPTIC SEIZURES
Echoencephalography employs ultrasonic echo to measure the position of the midline intracranial structures. The primary indication for this examination is suspicion of an expanding intracranial lesion. Echoencephalography has also proved useful in the diagnosis of subdural hematoma, cerebral abscess, and cerebral and cerebellar hemorrhage.
Computerized axial tomography (CAT scan, EMI scan) is a more recent noninvasive procedure that utilizes electronic and computer technology to measure the transmission of x-ray photons through tissue. The cranium is scanned in successive layers by a narrow beam of x-rays in such a manner that the transmission of x-ray photons across a particular "slice" can be measured. By means of a computer and a suitable algorithm a detailed differential picture of the internal structure and tissue of the brain can be constructed.
Use of the CAT scan has revolutionized the evaluation of patients with seizure disorders and now has, by and large, replaced the older methods of cerebral examinations mentioned above. Its great advantage is the fact that it can detect subtle differences in tissue density, allowing the inner parts of the brain to be demonstrated. Not only mass lesions can be studied, but also early changes in the ventricular system and the cerebral cortex. The scanning procedure is accomplished without danger or discomfort to the patient.
In the following case history the CAT scan demonstrated intracranial pathology (lowgrade astrocytoma) after the standard cerebral examinations gave normal findings. Case History No. 4.
A left-handed, adult man began to experience episodes of feeling "light-headed" and "dizzy" in 1970. The attacks, he said, were similar to the auras he now experiences. In July, 1973, he had his first "major attack." Those who had observed the attack described it this way: "His left hand curled up at the wrist in claw-like fashion; he stared vacantly into the air. His mouth twitched. There was jerking of the head and twisting of the head to the right. He became completely rigid and remained so for about five minutes. Then he relaxed but remained unconscious for another 20 minutes."
A week after this "major attack" a neurologic examination and an EEG were performed, with neither indicating any possible clue to the cause of the attack. The patient was started on phenytoin and phenobarbital but continued to experience psychomotor attacks with increasing frequency, with and without automatisms. These were interspersed with grand-mal attacks, which were usually related to the patient's discontinuing his medication. The seizure disorder was accompanied by a change in libido and by destructive behavior, which worsened as time progressed.
From July, 1973, to June, 1975, the patient was seen and evaluated by a number of competent neurologists and psychiatrists without success. All examinations, including radioactive brain scanning, arteriography, and pneumoencephalography were consistently negative. More than a dozen EEGs were taken. All exhibited normal findings with the exception of one, which revealed slowing in the left temporal area. Anticonvulsant therapy was administered, but the results were unrewarding. Among the drugs used were various dosages and combinations of acetazolamide, diazepam, mephobarbital, phenobarbital, phenytoin, and primidone.
The patient came to our center in July, 1975. The physical, neurologic, and EEG examinations were normal; however, in view of the relatively recent onset of the seizure disorder in a young adult, the increase in the frequency of the psychomotor seizures despite intensive anticonvulsant drug therapy and the severe behavioral aberrations, he was hospitalized for appropriate cerebral studies. The first of such investigations, the CAT scan, demonstrated an abnormal area in the right posterior temporal region. Subsequent cerebral angiogram, isotopie brain scan, and pneumoencephalogram were essentially nonrevealing.
Right parietotemporal craniotomy disclosed flattened pale cortex in the posterior portion of the temporal lobe. Histologic study confirmed a grade-II astrocytoma.
CONFIRMING THE DIAGNOSIS BY RESPONSE TO MEDICATIONS
In the past it was customary to use the patient's favorable reponse to antiepileptic medications as the primary criterion for confirming a diagnosis of epilepsy in patients with normal EEGs whose recurrent spells were not typical of classic epilepsy. We no longer do this routinely. It is now known that a number of factors can cause a favorable response in such cases that have little or no bearing on the efficacy of the medication itself. For example, there can be a "placebo effect" in some patients who will be benefited regardless of the medication taken; others have a favorable psychotherapeutic reaction because of the periodic conferences with the physician.
The diagnosis of epilepsy is a serious thing, and it will have a profound effect on the patient's subsequent life style. It should not be made for any patient unless the physician is able to elicit specific evidence of an epileptic disorder. Despite recent major advancements in the repeal or modification of restrictive legislation and despite the gradual elimination in recent years of some of the unwarranted stigmas and misconceptions that have been associated with epilepsy since Biblical times, there still are people who regard epilepsy as unacceptable and epileptics as second-class citizens. These people may form a relatively small portion of the total population, but they are unyielding. And the results of such a viewpoint can be found in many unexpected places.
For example, a few years ago we conducted a survey in our center and found out about a hazard of being an epileptic that has received essentially no publication in the literature: the diagnosis of epilepsy is a great deterrent to marriage, particularly if the epileptic is a girl.19 By intensive interrogation of our single adult female patients we discovered they almost universally shared the opinion that epilepsy is a formidable obstacle to marriage. Fully 60 per cent of our women patients have remained unmarried.
This is unfortunate, and it is particularly regrettable for those patients with nonconvulsive symptoms who were originally referred to our clinic as "definite epileptics" and in whom we have been unable to substantiate an unequivocal diagnosis of epilepsy. These patients, however, have already been stigmatized with the diagnosis, and many of them have been subjected to other discriminatory attitudes by ignorant and misinformed people.
So the physician who is tempted to diagnose a seizure condition as "epilepsy" before all the facts are in should bear in mind that, for many patients, almost insuperable socioeconomic obstacles and hardships will become inextricably associated with the diagnosis "epilepsy."
The diagnosis of epilepsy also should not be summarily assigned to a brain-damaged person who exhibits atypical or nonconvulsive symptomatology unless definite evidence of epilepsy can be demonstrated. We have seen many brain-injured patients with recurrent, paroxysmal emotional or psychiatric disorders that had been previously diagnosed as epilepsy solely on the assumption that the brain damage had "induced" epileptic manifestations. We are also cognizant of two brain-damaged children with grand-mal convulsions who were diagnosed as having epilepsy on the basis of previously sustained permanent cerebral injury. Investigation of these two patients revealed that the seizures were due to an intracranial neoplasm in one child and to ketotic hypoglycemia in the other. It is urged that all brain-damaged subjects with a seizure disorder be investigated as intensively as patients with no history of cerebral damage in an endeavor to elicit a cause for the symptoms.
Should one assign a diagnosis of epilepsy to a person who has experienced a seizure of undetermined cause and in whom the EEG examination gives normal findings? This is a controversial question. It is our present policy to make a tentative diagnosis of epilepsy in such instances and to continue with this diagnosis unless the passage of time proves that the seizure was a manifestation of some other disorder. It is imperative that such patients be observed very closely for a prolonged period of time for signs or symptoms of a specific cause for the seizure disorder, especially an intracranial neoplasm. An intracranial neoplasm should always be a pertinent etiologic consideration in the older child and the adult, since convulsions appearing initially at these stages of life are more likely to be a manifestation of a brain tumor than those occurring during early childhood. However, the initial onset of epilepsy during adulthood is not uncommon,20 and an analysis made of a large group of patients in our clinic indicated that there are many individuals who do not develop epilepsy until after the age of 20.*
DIFFERENTIATING EPILEPSY FROM EMOTIONAL OR PSYCHIATRIC DISORDERS
A diagnosis of epilepsy can be made in most patients with relative ease by using the diagnostic tools described above. However, the establishment of a precise diagnosis in patients with certain symptoms is frequently difficult, since recurrent disturbances, such as loss or impairment of consciousness, convulsions, bizarre motor and/or psychic performances, and syncopal attacks may be manifestations of a psychiatric disorder as well as epilepsy. Symptoms of epilepsy that also can be manifestations of psychiatric or behavior disorders are indicated in Table 4. A discussion of each is given below.
Recurrent attacks of staring and stiffness of the body. Still noted as early as 190921 that masturbation "may easily be mistaken for some epileptic manifestation" and reported that he had seen at least two cases in consultation in which the erroneous diagnosis was due to such a cause. "The mistake is the more natural when the mother mentions, as I have noted in several oí my cases, that the child 'fixes her eyes' or 'stares' during the paroxysm, and also that she 'quivers' or 'waves her hands in a sort of tremor,' and that after the attack the child is sleepy or actually goes to sleep," he said.
SYMPTOMS THAT MAY BE MANIFESTATIONS OF A PSYCHIATRIC OR AN EPILEPTIC DISORDER
A few years ago we reported three such cases,9 and subsequently eight other children have been referred to our clinic with a diagnosis of epilepsy because of recurrent attacks of "staring and stiffness of the body." All were girls. Of the 13 patients, the "spells" commenced between 15 months and two years of age in 10 of them. Physical, neurologic, and electroencephalographic findings were normal initially and remained so on subsequent examinations. Prolonged observation and psychiatric studies - in collaboration with Dr. Leo Kanner in some of the cases - confirmed the diagnosis oí masturbation rather than epilepsy in each of these patients. The following is a resume of a typical case history.
Case History No. 5
A seven-year-old girl was referred to us with a diagnosis of epilepsy. The history revealed that she had started to have two-tothree-minute episodes of rocking movements of the body on a daily basis beginning when she was three years and four months of age. She would go to an obscure corner of the room, or to another room, and lie on her abdomen. She would then stiffen her body, rock back and forth, stare blankly and begin to sweat profusely. Then she would resume her normal play activities, or seem drowsy and fall asleep.
Two months after the condition was first noticed she was hospitalized for an investigation of the condition. Urinalysis, blood chemistries and counts, and cerebrospinal fluid all were within normal limits, and skull roentgenography and brain scan revealed no abnormalities. An EEG recording was read as "irregular spike activity in the right occipital area associated with slow-wave complexes." The hospital-discharge diagnosis was a seizure disorder of undetermined origin, and phenobarbital was prescribed. The episodes diminished in frequency and then disappeared but recurred two months later. Dilantin and Zarontin were subsequently added to the therapeutic regimen, but the spells recurred at their previous frequency. A second EEG gave normal findings.
The child was referred to our center for evaluation. Physical and neurologic examinations were within normal limits, except for marked hypertrichosis on the back, extremities, and shoulders. Routine blood chemistries and counts, urinalysis, skull and wrist x-rays, and EEG were all normal.
The mother described the patient's spells as follows: "During the rocking movements, her hands are kept tightly between her thighs. She lies on her stomach, stiffens her legs, and rocks. After a few minutes, her eyes stare and she sweats. Sometimes she is drowsy after the spell. When we see her during a spell, she continues to rock. We are able to communicate with her but she prefers not to be disturbed by us."
On intensive questioning, the parents admitted that they had never observed the onset of a spell, as the child always went to another room. One exception occurred recently in a motel. When the patient initiated the act and realized that her parents were present, she covered herself so that they could not observe her.22
Obviously, the initial diagnosis of epilepsy in this case was made on the basis of a history of recurrent attacks of staring and stiffness of the body and the abnormal EEG. However, the clinical data derived from our interrogation of the parents, the description of the spells, and the activities and condition of the child both before and after the episodes were generally incompatible with a diagnosis of epilepsy but highly indicative of masturbation.
The following features were important in our differential diagnosis: (1) the occurrence of epileptic seizures is generally abrupt, spontaneous, and involuntary, whereas the performance of masturbation is volitional; (2) rocking movements are rarely, if ever, observed during epileptic attacks but are characteristic of masturbation in early childhood, especially in girls; (3) the epileptic patient usually has no choice regarding his or her location at the time of occurrence of a seizure, whereas the child desiring to masturbate will almost invariably seek seclusion; (4) epileptic seizures are generally associated with loss of consciousness, whereas consciousness is maintained throughout masturbation.
We emphasize that rocking movements in a young child who presents "attacks of staring and stiffness of the body" warrant consideration of a diagnosis of masturbation. Still21 described several cases in which rocking of the body was a prominent feature of childhood masturbation, one of which we cite in part: "... a girl aged five years, was brought with the history that she would several times a day, when she thought herself not observed by her mother, rock herself backwards and forwards as she sat on her chair, at the same time turning red in the face, and perspiring, and then turning white as if exhausted. Inquiry showed that masturbation occurred at night, also . . ." Recurrent syncopal attacks. Fainting is a very common complaint and there are few people who have not experienced this sort of disturbance at least once in their lives. Most syncopal attacks are without major implications, but if such episodes are prolonged or occur often, precise diagnosis is frequently difficult. When such disorders as cerebral and cardiac disturbances, orthostatic hypotension, carotid-sinus sensitivity, etc., have been ruled out on the basis of physical, neurologic, and laboratory examinations, the diagnosis is almost always a functional disturbance or epilepsy.
Hundreds of patients who suffered recurrent episodes of syncope were referred to our center with a diagnosis of epilepsy based on "minimal," "borderline," or "nonspecific" electroencephalographic "irregularities" and/ or a favorable response to antiepileptic medication. Intensive investigation and prolonged follow-up study of such patients, however, revealed that the symptomatology in the preponderance of cases was functional and not epileptic in origin.
In patients with normal interictal electroencephalograms, the differentiation between functional syncopal attacks and epilepsy often presents a diagnostic dilemma, and in many instances a definite diagnosis can be made only after the passage of time. However, the occurrence in persons with fainting spells of intersymptom electroencephalographic abnormalities of the type encountered in patients with overt epilepsy constitutes evidence in favor of an epileptic disorder. Obviously, the presence of seizure discharges in an electroencephalogram recorded during an attack of syncope is conclusive evidence of epilepsy. Unfortunately, however, one rarely has the opportunity to obtain an electroencephalogram at such a time.
Functional syncopal attacks often simulate atonic epileptic seizures. The following are some features that may assist in differentiating these two disorders:
1. Functional syncopal episodes are almost always preceded by an exciting or disturbing situation, and the subject is usually aware of the waning of consciousness, whereas in epilepsy the attacks generally occur spontaneously and abruptly.
2. Functional syncopal incidents are usually variable in duration, severity, and clinical manifestations, whereas epileptic seizures are generally stereotyped in an individual patient.
3. Biting of the cheeks or tongue and injury are generally unassociated with attacks of syncope; the reverse is true in many patients with epilepsy.
4. Fine trembling movements of the extremities occur, in some patients with functional syncopal attacks, whereas the repetitive motions associated with epileptic seizures are usually intense, pronounced clonic movements.
5. During a functional syncopal episode, the patient is generally pale and the skin is cool and clammy; the epileptic patient is apt to be flushed and cyanotic during a seizure.
6. Attacks of syncope are usually associated with shallow respirations and a weak pulse; the epileptic seizure is frequently accompanied by prolonged apnea and stertorous breathing.
7. Pronounced postictal phenomena are generally absent in episodes of functional syncope, whereas postconvulsive symptomatology is characteristic of the epileptic attack.
8. The electroencephalogram during and immediately following recovery from a functional syncopal attack is normal, whereas the electroencephalogram recorded during an epileptic seizure shows seizure discharges, and the electroencephalogram obtained during the postictal phase of an epileptic seizure reveals abnormalities consisting of marked generalized slowing.
9. Functional syncopal episodes are not preceded by a specific stereotyped aura, whereas many patients with epilepsy experience such a prodrome.
The "hyperventilation syndrome," which is thought to be functional in origin, may simulate an epileptic disorder in some of its clinical manifestations. This disturbance is occasionally observed in the teenager, but it occurs most commonly in older people, particularly nervous, anxious women who have other functional disturbances related to tension.
Nightmares, night terrors, and somnambulism. The differentiation of disturbances of sleep (e.g., nightmares, night terrors, and somnambulism) of epileptic origin from those of psychiatric origin is frequently very difficult on clinical grounds. Investigations now being carried out at several sleep-research centers23 may be expected to yield significant findings in this respect in the near future.
The occurrence of electroencephalographic abnormalities concomitant with clinical symptoms during either natural or artificially induced sleep is definite proof of an epileptic disorder. On the other hand, the presence of abnormalities in an electroencephalogram recorded during the awake state or during sleep unassociated with clinical symptoms does not warrant a positive diagnosis of epilepsy. In patients with normal electroencephalograms, the eventuation of overt epileptic symptomatology, such as grand-mal seizures, diurnally or nocturnally, suggests an epileptic basis for the sleep aberrations, whereas the presence of marked emotional disturbances, such as fears and anxieties, suggests a functional basis.
Recurrent attacks of abdominal pain and headache. These are among the most common complaints with which the pediatrician is confronted. Since these symptoms may be manifestations of many disorders, it is essential that all known causes be thoroughly explored before a specific diagnosis is assigned. In some instances a definite neurotic basis can be established, such as in the case of children with school phobia or other anxieties and frustrations. In many patients, however, a definite cause for the symptoms cannot be elicited, and consequently they are usually designated as functional in origin.
In some patients, gastrointestinal disturbances (particularly abdominal pain), headache, or other symptoms suggestive of autonomic dysfunction occur at the onset (aura) or during the postictal phase of a frank epileptic seizure, especially a major motor attack. When such symptoms occur at these times, they are unquestionably of epileptic origin and constitute an integral part of the complete epileptic seizure. Therefore, when such disturbances recur and are not associated with overt epileptic symptoms, the possibility that they may be the only manifestations of an epileptic disorder (autonomic epilepsy) should be considered.
Headache and gastrointestinal disturbances, especially abdominal pain, may occur independently or in combination as clinical manifestations of autonomic epilepsy and are the most common expressions of this disorder. Autonomic epilepsy is discussed in detail in the following article.
A definite diagnosis of epilepsy can be made if an electroencephalographic recording reveals seizure discharges concomitant with the clinical disturbance in question; however, the obtention of an electroencephalogram at the time of the occurrence of clinical symptoms, which are usually brief, is exceedingly rare. Therefore, other criteria are necessary for a diagnosis in the vast majority of cases. Here are some suggestions that may prove helpful in making the definite diagnosis:
1. When the EEG reveals abnormalities without concurrent symptomatology: A diagnosis of epilepsy can be made when there is (a) impairment of the sensorium that occurred at the time of the symptom, and (b) the disturbance is followed by drowsiness, sleep, stupor, or other postictal phenomena similar to those observed in patients with overt epileptic seizures.
2. When the EEG is normal with paroxysmal symptoms.- A diagnosis of epilepsy can be considered if the paroxysmal symptoms are associated with impairment of the sensorium or are followed by postictal manifestations similar to those encountered in patients with frank epileptic seizures.
In patients whose symptomatology fulfills the clinical criteria for a diagnosis of autonomic epilepsy, a favorable response to anticonvulsant medication is regarded as an additional confirmatory finding. However, in subjects whose symptoms did not satisfy one or more of the diagnostic criteria, a beneficial reaction to antiepileptic drug therapy is considered only a finding that favors a diagnosis of epilepsy.
Migraine. The diagnosis of migraine should be considered in children who experience paroxysmal headaches associated with gastrointestinal disturbances. The classic unilateral headache is not encountered frequently in children. Migrainous headaches in children are almost always associated with gastrointestinal disorders, such as nausea, vomiting, and abdominal pain. In fact, these disturbances are frequently the predominant symptoms of a migrainous disorder in the young child and in many instances precede the onset of the headache. Unilateral headache, visual aberrations, and a history of migraine in the family are additional factors that confirm the diagnosis. Cerebral-electric abnormalities have been reported in the electroencephalograms of some children with migraine. In our experience, the electroencephalogram does not present irregularities diagnostic of migraine.
The possibility of a relationship between epilepsy and migraine has been investigated by some physicians, with Lennox24 postulating that the syndrome classified as migraine is actually a form of epilepsy. On the other hand, Friedman and Merritt25 did not find significant evidence of a relationship between epilepsy and migraine. In our opinion, the headaches associated with epilepsy and those with migraine seem to be symptoms of two distinct types of cerebral dysfunction. The following features may be helpful in distinguishing between headaches associated with epilepsy and headaches associated with migraine:
1. Onset. In epilepsy the onset of the headache is abrupt, and it is of maximum severity at the very beginning. Onset is almost invariably accompanied by impairment of consciousness. It is followed by drowsiness, sleep, or stupor.
In migraine, the onset is gradual, and the pain gradually increases in intensity until maximum severity is reached. Migrainous headaches usually have not been associated with impairment of consciousness in our experience, although they may be followed by drowsiness or sleep. Some other physicians, however, have reported patients who suffered impairment or loss of consciousness during a migraine attack.
2. Duration. The epileptic headache is usually brief, lasting perhaps five minutes. The migraine headache generally is protracted, lasting for several hours or longer.
3. Association with GI disturbances. Epileptic headaches are rarely associated with gastrointestinal disturbances, whereas such disturbances are frequently reported by patients suffering from migraine.
4. Family history. The family history is not a significant diagnostic factor in children with epilepsy, whereas most children with migraine have a family history of migrainous disorders.
5. Electroencephalogram. The interseizure EEG reveals specific abnormalities in many patients with epilepsy, whereas in children with migraine the EEG is usually normal or displays nonspecific electrical irregularities.
Narcolepsy. This disorder is characterized by paroxysmal and recurrent diurnal attacks of irrepressible sleep. It is frequently associated with one or more of the following auxiliary symptoms: cataplexy, sleep paralysis, and hypnagogic hallucinations.
Cataplectic attacks consist of sudden loss of muscular tone and weakness. Such episodes may simulate atonic epileptic seizures in the brevity and general appearance of the spells. The cataplectic attack differs from the epileptic seizure in the following respects: (1) Cataplexy is almost always precipitated by some emotional disturbance or sudden stimulation, particularly laughter; the atonic epileptic seizure generally occurs spontaneously in the absence of an obvious precipitating factor. (2) Consciousness is maintained throughout the cataplectic attack, whereas there is at least a momentary impairment of consciousness with the epileptic seizure.
Most investigators have reported normal electroencephalographic findings in patients with narcolepsy. In the majority of narcoleptic patients who manifest cataplexy, electroencephalograms obtained during sleep attacks and/or nocturnal sleep almost invariably begin with rapid-eye-movement waves, but only 10 per cent of narcoleptic patients without cataplexy exhibit REM waves at the onset of their seizures.23
Convulsions preceded by emotional upsets. The possibility of a hysterical convulsive disorder should be seriously considered in persons whose seizures are always immediately preceded by an acute emotional disturbance. In most cases, the differentiation between a hysterical convulsion and a true epileptic seizure is not difficult on clinical grounds.
As a rule, the onset of the hysterical convulsion is less abrupt than that of true epilepsy, and in most instances bodily injury from a hysterical seizure does not occur. Postconvulsive sleep or stupor, which is commonly observed following an overt epileptic convulsion, is generally not seen after a hysterical convulsion; when it does occur, the patient can be easily aroused.
Hysterical convulsions are not usually associated with biting of the tongue, and generally there is neither fecal nor urinary incontinence.
The interseizure electroencephalogram in many patients with frank epilepsy reveals specific electrical aberrations, such as spikes, random spike and wave forms, and abnormally slow waves; in patinets with genuine hysteria, it is normal. An electroencephalographic examination performed on a hysterical patient during or immediately subsequent to an active convulsion reveals normal findings, whereas an electrographic tracing recorded during a frank epileptic seizure exhibits continuous spiking and abnormally slow high-voltage waves during the postconvulsive phase.
Convulsions following holding of the breath. Breath-holding spells of the convulsive type usually appear initially during the first two years of life but rarely before six months of age. The attacks generally recur infrequently at the onset of the disorder, but in most cases the rate of recurrence gradually increases as the child advances in age. They tend to disappear spontaneously after four years of age and rarely recur after the age of six years.
Breath-holding attacks of the convulsive form can usually be readily differentiated from epilepsy on clinical grounds, since the following sequence of symptoms is encountered in no other condition:26 (1) a precipitating factor such as a slight injury or some other minor provocative incident that arouses an acute emotional response, (2) crying, usually of short duration, (3) holding of the breath in the expiratory phase of respiration, (4) cyanosis, (5) generalized rigidity (opisthotonos), with impairment of consciousness, and (6) frank clonic convulsive movements.
The following are some features that assist in distinguishing breath-holding convulsions from epileptic seizures:
1. Breath-holding attacks are almost always preceded by an obvious precipitating factor, whereas in most instances, epileptic seizures occur spontaneously.
2. Patients with breath-holding episodes cry before holding of the breath, whereas young children with epilepsy rarely cry before an attack. Breath-holding spells rarely occur after prolonged intervals of crying but characteristically follow two or three short, violent cries.
3. Cyanosis precedes the convulsive movements of breath-holding spells, whereas in epilepsy, cyanosis, if present, usually appears after the onset of the convulsion.
4. Opisthotonos is almost invariably associated with breath-holding spells, whereas it rarely occurs with epileptic seizures. (See illustration at the beginning of this article.)
5. The electroencephalogram is normal in children with breath-holding attacks, but it is abnormal in many patients with epilepsy.
Lombroso and Lerman27 reported a high incidence of breath-holding spells in which cyanosis was absent or did not represent a conspicuous aspect of the symptoms; however, we have not encountered children with severe breath- holding spells who did not manifest cyanosis in association with their attacks. Our observations are in accord with those of Ibrahim,28 who stated that his patients with convulsive breath-holding episodes turn bluish-red during the period of respiratory arrest and pale after the resumption of breathing. Peiper29 has reported concordant findings.
Reflex anoxic seizures secondary to provoked cardioinhibition in infants and young children frequently are incorrectly diagnosed as epileptic and treated with anticonvulsant drugs. These attacks may take the form of breath-holding spells with vagotonia^ breath-holding spells of the pallid type, pallid infantile syncope, reflex anoxic cerebral seizures with white syncope, or white breathholding attacks. Stephenson describes a typical case:30
An unsteady toddler on his own trips and falls. His mother hears the bump but no succeeding cry and hurries to him. She finds her child lying deathly still with eyes fixed upwards, lips dusky. As she lifts him, he abruptly stiffens into rigid extension with jaw clenched and hands fisted, gives a few jerks, and after what seems an age (but in fact in less than half a minute) relaxes limply with an absent far-away look. Then he opens his eyes, at once recognizes his mother, cries a little, and drifts off to sleep, his face distinctly pale.
There is substantial evidence to support the hypothesis that "white" reflex anoxic seizures result from vagal-mediated cardiac arrest, and the differentiation between these attacks and epileptic seizures can generally be accomplished by ocular compression accompanied by electroencephalographic and electrocardiographic monitoring. Ocular compression in children with such "vagal attacks" usually results in asystole of abnormal (at least four seconds') duration.30
In 1943, we2e proposed vagotonia as a possible cause (mechanism) of breath-holding attacks in some children. In three of our patients, breath-holding spells occurred in association with a bradycardia and an asystole produced by pressure on the eyeballs. In one of these patients, mild pressure on the eyeballs arrested the heartbeat for about 20 seconds. This boy's cheeks became flushed, and he appeared unconscious for approximately onehalf minute after the heart resumed beating.
It should be noted that many of our patients with breath- holding attacks manifested such behavioral aberrations as enuresis, head banging, and temper tantrums. In some instances, these disturbances appeared during the early years of life concurrent with the breathholding spells; however, in many cases, behavioral abnormalities did not become apparent until after the breath-holding episodes had disappeared. Kanner31 studied a group of children with breath- holding spells and observed similar behavioral aberrations. Gellis32 is impressed with "the number of children with this condition who, as they grow older, abandoned the breath-holding portion of the syndrome but exhibited simple attacks of syncope when subjected to pain, fright or anger." Lombroso and Lerman27 emphasized the frequency of syncopal attacks occurring in "direct continuation of" or subsequent to breath-holding spells. We have also encountered syncopal attacks in many of our patients who had previously experienced breathholding episodes.
1. Livingston, S. Comments on a study of light-induced epilepsy in children. Am. J. Dis. Child. 83 (1952), 409.
2. Marshall, C, Walker, A. E., and Livingston, S. Photogenic epilepsy: parameters of activation. Arch. Neurol. Psychiatry 69 (1953), 760-765.
3. Robertson, E. G. Photogenic epilepsy: self-precipitated attacks. Brain 77 (1954), 232-251.
4. Keith, H. M., et al. A study of light-induced epilepsy in children. Am. J. Dis. Child. 83 (1952), 408-409.
5. Critchley, M. Musicogenic epUepsy. Brain 60 (1937), 13-27.
6. Bickford, R. G., et al. Reading epilepsy: clinical and electroencephalographic studies of a new syndrome. Trans. Am. Neurol. Assoc. 81 (1956), 100-102.
7. Cohen, N. H., McAuliffe, M., and Aird, R. B. "Startle" epilepsy treated with chlordiazepoxide (Librium). Dis. Nerv. Syst. 22 (1961), 20-27.
8. Calderon-Gonzalez, R., Hopkins, I., and McLean, W. T. Tap seizures. A form of sensory precipitation epilepsy. J.A.M.A. 298 (1966), 521-523.
9. Livingston, S. The Diagnosis and Treatment of Convulsive Disorders in Children. Springfield, Dl.: Charles C Thomas, 1954.
10. Hopkins, A., and Scambler. G. How doctors deal with epilepsy. Lancet 1 (1976), 183-186.
11. Gastaut, H., and Broughton, R. Epileptic Seizures: Clinical and Electrographic Features, Diagnosis and Treatment. Springfield, Ill.: Charles C Thomas, 1972.
12. Livingston, S. Convulsive disorders in infants and children. In Levine, S. TL. (ed.). Advances in Pediatrics. Volume X. Chicago: The Year Book Publishers, 1958.
13. Livingston, S., and Berman, W.: Classification and diagnosis of epileptic seizures. Pediatr. Ann. 2:8 (1973), 23-40.
14. Livingston, S., et al. Petit mal epilepsy: results of a prolonged follow-up study of 117 patients. f.A.M.A. 194 (1965), 227-232.
15. Pauli, L., et al. Minor motor epilepsy: treatment with corticotropin (ACTH) and steroid therapy. JAMA. 174 (1960), 1408-1412. 16. Harper, J. R.: True myoclonic epilepsy in childhood. Arch. Dis. Child. 43 (1968), 28-35.
17. Holowach, J. Thurston, D. L-, and OT-eary, J. Prognosis in childhood epilepsy: foilow-up study of 148 cases in which therapy had been suspended after prolonged anticonvulsant control. N. Engl. J. Med. 286 (1972), 169-174.
18. Livingston, S., Pauli, L. L., and Pruce, I. Control of antiepileptic drug treatment. Dev. Med. Child. Neurol. 18 (1976), 820-822.
19. Livingston, S. The opinionated reader. Med. Opinion 4 (1975), 63-64.
20. Livingston, S. Etiologic factors Ln adult convulsions: an analysis of 689 patients whose attacks began after twenty years of age. N. Engt. f. Med. 254 (1956), 1211-1216.
21. Still, C. F. Common Disorders and Diseases of Childhood. London: Oxford University Press, 1909.
22. Livingston, S., Berman, VV., and Pauli, L. L.: Masturbation simulating epilepsy. Clin. Pediatr. 14 (1975), 232-234.
23. Kales, A., and Kales, J. D. Sleep disorders: recent findings in the diagnosis and treatment of disturbed sleep. N. Engl. J. Med. 290 (1974), 487-499.
24. Lennox, W. G. Epilepsy and Related Disorders. Boston: Little, Brown and Company, 1960.
25. Friedman, A. P., and Merritt, H. H. Headache: Diagnosis and Treatment. Philadelphia: F. A. Davis, 1959.
26. Bridge, E. M., Livingston, S., and Tietze, C. Breath-holding spells: their relationship to syncope, convulsions, and other phenomena. /. Pediatr. 23 (1943), 539-561.
27. Lombroso, C. T., and Lerman, P.: Breathholding spells (cyanotic and pallid infantile syncope). Pediatrics 39 (1967), 563-581.
28. Ibrahim, J. Über respiratorische Affektkrampfe im frühen Kinedsalter (das sogenannte "Wegbleiben" der Kinder). Z. Gesamte Neurot. Psychiat. 5 (1911), 388-413.
29. Peiper, A. Cerebral Function in infancy and Childhood. New York Consultants Bureau, 1963.
30. Stephenson, J.B.P.: Reflex anoxic seizures ("white breathholding"): nonepÜeptic vagal attacks. Arch. Dis. Child. 53 (1978), 193-200.
31. Kanner, L. Child Psychiatry, Third Edition Springfield, Dl.: Charles C Thomas, 1962.
32. GeIUs, S. S. (ed.). The Year Book of Pediatrics. Chicago: The Year Book Publishers. 1955-1956, p. 368.
CLINICAL AND EEG FEATURES OF EPILEPTIC SEIZURES
PRECIPITANTS OF EPILEPTIC SEIZURES
EEG FEATURES THAT ASSIST IN DIFFERENTIATING BRIEF EPILEPTIC SEIZURES
SYMPTOMS THAT MAY BE MANIFESTATIONS OF A PSYCHIATRIC OR AN EPILEPTIC DISORDER