Pediatric Annals

FEBRILE CONVULSIONS: DIAGNOSIS, TREATMENT, AND PROGNOSIS

Samuel Livingston, MD; Lydia L Pauli, MD; Irving Pruce; Irving I Kramer, MD

Abstract

In 1973, Pampiglione1 directed attention to the confusion surrounding the designation of "febrile convulsions," stating that "it covers a wide range of circumstances and it is not uniformly defined." We agree completely that there is no universally accepted definition of "febrile convulsions" and have observed that the interpretation of this term varies from physician to physician and from one clinic to another.

Without amplification or elaboration, the expression "febrile convulsions" is essentially meaningless, because it refers to all seizures associated with an elevation of temperature without regard to such factors as the duration and character (focal or generalized) of the attack, the electroencephalographic findings, and the developmental and neurologic status of the patient before the initial seizure. This inconsistency in nomenclature and classification is, in our opinion, undoubtedly responsible for some of the contradictory and perplexing reports that have appeared in the literature relative to the diagnosis, treatment, and prognosis of seizures associated with fever in infants and young children.

We have had a special interest in this disorder since 1936 and have written continually on this subject since our original article in 1945.2 During the course of our early investigations, it became apparent that childhood "febrile convulsions" can, in most instances, be classified into two specific types, and in 1954 we3 introduced the terms "simple febrile convulsions" and "epileptic seizures precipitated by fever" to designate the two groups. Table 1 presents the characteristic features of these two types of "febrile seizures."

Our studies were continued, and in 1958 we4 published the diagnostic, therapeutic, and prognostic data relative to 622 patients who were seen by us at the time of or shortly after their first "febrile seizure." All of these patients had been followed up by us for intervals ranging from 15 to 22 years. Two hundred fifty-six of these patients were classified as having had a simple febrile convulsive disorder and the other 366 as suffering with epilepsy. It should be noted that none of the 622 children manifested evidence of a cerebral infection or intoxication in association with their initial seizure, nor did any of them present a history or evidence of having had a cerebral disturbance of any type before their initial "febrile convulsion."

This article includes a resume of our present classification of "febrile convulsions" based on the results of long-term documented investigations, as well as a discussion of some of the current controversies regarding the diagnosis, treatment, and prognosis of so-called febrile convulsions. Throughout, the term "simple febrile convulsions" will be designated as SFC, and FC will be used to denote seizures associated with an elevation of temperature.

Table

1. PampigUone, G. Children with "febrile convulsions." Lancet 2 (1973), 1035.

2. Livingston, S., and Kajdi, L. Importance of heredity in the prognosis of febrile convulsions. Am. J. Dis Child. 69 (1945), 324.

3. Livingston, S . The Diagnosis and Treatment of Convulsive Disorders in Children. Springfield, IU.; Charles C Thomas, 1954.

4. Livingston, S. Convulsive disorders in infants and children. In Levine, S. Z. (ed.). Advances in Pediatrics, Volume 10. Chicago: Year Book Publishers, 1958.

5. Eisner, V., Pauli, L. L., and Livingston, S. Hereditary aspects of epilepsy. Bull Johns Hopkins Hasp. 105 (1959), 245-271.

6. Miller, F. M. W., et al. Crowing Up in Newcastle Upon Tyne. London: Oxford University Press, 1960.

7. Gomez, J. G., Arriniegas, E., and Torres, J. Prevalence of epilepsy in Bogotá, Colombia. Neurology 28 (1978), 90-94.

8. Nelson, K. B., and Ellenberg, J. H. Prognosis in children with febrile seizures. Pediatrics 61 (1978), 720-727.

9. Friderichsen, C, and Melchior, J. Febrile convulsions in children, their frequency and prognosis. Acta…

In 1973, Pampiglione1 directed attention to the confusion surrounding the designation of "febrile convulsions," stating that "it covers a wide range of circumstances and it is not uniformly defined." We agree completely that there is no universally accepted definition of "febrile convulsions" and have observed that the interpretation of this term varies from physician to physician and from one clinic to another.

Without amplification or elaboration, the expression "febrile convulsions" is essentially meaningless, because it refers to all seizures associated with an elevation of temperature without regard to such factors as the duration and character (focal or generalized) of the attack, the electroencephalographic findings, and the developmental and neurologic status of the patient before the initial seizure. This inconsistency in nomenclature and classification is, in our opinion, undoubtedly responsible for some of the contradictory and perplexing reports that have appeared in the literature relative to the diagnosis, treatment, and prognosis of seizures associated with fever in infants and young children.

We have had a special interest in this disorder since 1936 and have written continually on this subject since our original article in 1945.2 During the course of our early investigations, it became apparent that childhood "febrile convulsions" can, in most instances, be classified into two specific types, and in 1954 we3 introduced the terms "simple febrile convulsions" and "epileptic seizures precipitated by fever" to designate the two groups. Table 1 presents the characteristic features of these two types of "febrile seizures."

Our studies were continued, and in 1958 we4 published the diagnostic, therapeutic, and prognostic data relative to 622 patients who were seen by us at the time of or shortly after their first "febrile seizure." All of these patients had been followed up by us for intervals ranging from 15 to 22 years. Two hundred fifty-six of these patients were classified as having had a simple febrile convulsive disorder and the other 366 as suffering with epilepsy. It should be noted that none of the 622 children manifested evidence of a cerebral infection or intoxication in association with their initial seizure, nor did any of them present a history or evidence of having had a cerebral disturbance of any type before their initial "febrile convulsion."

This article includes a resume of our present classification of "febrile convulsions" based on the results of long-term documented investigations, as well as a discussion of some of the current controversies regarding the diagnosis, treatment, and prognosis of so-called febrile convulsions. Throughout, the term "simple febrile convulsions" will be designated as SFC, and FC will be used to denote seizures associated with an elevation of temperature.

Table

TABLE 1FACTORS THAT HELP TO DIFFERENTIATE SIMPLE FEBRILE CONVULSIONS FROM EPILEPTIC SEIZURES PRECIPITATED BY FEVER

TABLE 1

FACTORS THAT HELP TO DIFFERENTIATE SIMPLE FEBRILE CONVULSIONS FROM EPILEPTIC SEIZURES PRECIPITATED BY FEVER

SIMPLE FEBRILE CONVULSIONS

It should be noted that some authors designate an SFC as a "benign," "pure," "typical," or "uncomplicated" febrile seizure.

We make a diagnosis of an SFC disorder in children who present the following:

Brief, generalized seizures, seldom lasting longer than five or so minutes, that occur soon after the onset of fever; no clinical or laboratory evidence of a cerebral infection or intoxication; and a normal electroencephalogram after the patient has been afebrile for at least 10 days.

Simple febrile convulsions usually appear initially between nine and 18 months of age, seldom occur in infants under six months, and rarely if ever commence after six years of age. These seizures are almost always associated with an extracranial infection, such as tonsillitis, pharyngitis, or otitis media.

An SFC usually occurs soon after the onset of the fever, generally within two to six hours. We have never observed an SFC to occur as the initial convulsive reaction to an elevation of temperature later than 24 hours after the onset of an acute febrile illness. It is most unlikely that such a seizure would be a manifestation of an SFC disorder; a patient who experiences a seizure at that time should be intensively investigated for an intracranial infection or epilepsy.

Most children will experience one simple febrile convulsion during a febrile episode, but some may have several such seizures. It is extremely rare, however, for a simple febrile convulsion to be repeated during the same illness if 24 hours have elapsed since the onset of fever. For example, it would not be unusual for a child who had a short generalized seizure soon after the onset of fever associated with a pharyngitis to experience a similar convulsion two to four hours later. If, however, the fever should persist and the seizure be repeated about 24 hours later, the chances are that it is not an SFC but a symptom of a more serious illness, such as meningitis, or a manifestation of an epileptic disorder.

Electroencephalographic examinations performed during the first week or so after the onset of fever revealed normal findings in approximately 67 per cent of our 256 patients with SFC; in the remaining 33 per cent, the electroencephalogram obtained during this period of time revealed abnormal slowing, which disappeared in all instances within 10 days after defervescence. This slowing was most commonly bilaterally synchronous.

We performed electroencephalographic examinations in 10 young children during a4 febrile episode associated with an extracranial infection. There were no signs of cerebral infection at the time of examination; convulsions were not associated with the illness, nor did any of the children have a history of seizures of any type. The electroencephalograms of nine of these patients revealed bilaterally synchronous, abnormally slow high-voltage waves similar to those exhibited in our patients with SFC for at least one week or so after defervescence.3 We consider such occipital slowing not to be an inherent electrical abnormality but, rather, a transient episode of electrical dysfunction due to the fever per se.

In view of the results of the electroencephalographic examinations in our patients with an SFC disorder and the electrographic findings in the 10 patients with fever unassociated with convulsions, we recommend that electroencephalograms of patients with FC be obtained at a time when they have been afebrile for at least 10 days.

EPILEPTIC SEIZURES PRECIPITATED BY FEVER

An "epileptic seizure precipitated by fever" is classified by some physicians as a "complex," "complicated," "nonbenign," or "severe" febrile seizure.

A diagnosis of "epileptic seizures precipitated by fever" is strongly indicated by one or more of the following findings:

Prolonged convulsions, focal seizures of any duration, seizures in a child six or more years of age, or electroencephalographic abnormalities, such as those observed in epileptic patients whose seizures are unassociated with fever.

This type of epilepsy may commence at any age, and the duration of the seizure varies from patient to patient. The attack may take place any time during a febrile episode and may recur on a daily to yearly basis.

In approximately 50 per cent of our 366 patients with "epileptic seizures precipitated by fever," an electroencephalogram obtained after the patient had been afebrile for at least 10 days revealed seizure discharges (spikes or spike and wave forms) such as are seen in persons with overt epilepsy. The remainder of the patients presented normal electrographic findings. We emphasize that the incidence of normal electroencephalograms in patients with definite clinical evidence of grand-mal epilepsy is exceedingly high, particularly in young children.

WILL "FEBRILE CONVULSIONS" MAKE THE CHILD PRONE TO EPILEPSY?

Physicians and parents both want to know if the occurrence of simple febrile convulsions in a child will render him more prone to the subsequent development of epilepsy. Our investigations do not support such a conclusion. We have found consistently that the incidence of afebrile seizures (e.g., epilepsy) subsequent to a simple febrile convulsive disorder is about the same as the incidence of afebrile seizures in the general population. On the other hand, the vast majority of children who fulfilled our criteria for a diagnosis of "epileptic seizures precipitated by fever" subsequently developed afebrile seizures.

In 1958, we4 reported that seven (2.9 per cent) of 256 children developed epilepsy following an SFC disorder. Frank epileptic seizures appeared in these seven patients at intervals varying from three to 11 years after the initial SFC but in almost all instances before the age of five years. The incidence of subsequent epilepsy found in our study of children with SFC approximates the prevalence rates in the general population reported by us in 1959,5 by Miller et al. in I960,6 and by Gomez et al. in 1978.7

Recent confirmation of the paucity of epilepsy following simple febrile convulsions was given by the massive investigation by Nelson and Ellenberg,8 in which the authors presented data based on the findings of a study by the National Institute of Neurological and Communicative Disorders and Stroke Collaborative Perinatal Project. In this study, 1,706 children who had had febrile convulsions were followed in an effort to determine the consequences of FC and their relationship to certain "risk factors."

Nelson and Ellenberg8 found 971 children who were neurologically and developmentally normal before their first seizure, who had no family history of afebrile seizures, and whose initial FC was "simple." They reported that 0.9 per cent of the 971 children subsequently developed "epilepsy" and 1.9 per cent developed "afebrile seizures."

These authors8 classified "recurrent asymptomatic afebrile seizures" and "epileptic" seizures as two separate, distinct entities. We do not believe there is scientific justification for such a nosologic differentiation. We regard asymptomatic afebrile seizures as unquestionable epileptic attacks. Consequently, our interpretation of their data is that 2.8 per cent of the children who had had a simple febrile convulsive disorder subsequently developed epilepsy, an incidence that is concordant with our findings.

Therefore, on the basis of the results of our study,4 the data reported by Nelson and Ellenberg,8 and the findings of other investigators,9,10 we remain convinced that there is no justification for fear that a child will have a significantly elevated risk of developing epilepsy if he has had a simple febrile convulsion.

RISK FACTORS

Over a period of many years, many investigators have endeavored to identify features that could be reliably employed as indicators or predictors of epilepsy in children who have experienced FC. The most prominent of these risk factors are the clinical character of the seizure, the duration of the attack, electroencephalographic findings, the neurologic and developmental status of the patient before the initial seizure, the rate of seizure recurrence, and the family genetic history.

In 1958, we4 reported that recurrent afebrile seizures developed in 355 (97 per cent) of 366 patients whose initial seizure was prolonged or focal or occurred after six years of age or whose electroencephalogram revealed specific abnormalities.

Nelson and Ellenberg8 presented prognostic data relative to 636 patients who were neurologically or developmentally "abnormal or suspect" before their initial seizure or whose first seizure was complex (more than 15 minutes' duration or focal or more than one seizure in 24 hours) or who had an immediate family history of afebrile seizures. They reported that 11.6 per cent of the 636 patients subsequently developed epilepsy and that 15.8 per cent developed "recurrent asymptomatic afebrile seizures." As mentioned above, we do not believe there is any difference between "epilepsy" and "recurrent asymptomatic afebrile seizures," and, therefore, our evaluation of their data is that 27.4 per cent of the 636 children with one or more risk factors developed epilepsy.

Character and duration of the seizure. There is essential agreement that if the character of the initial FC is other than generalized, the subsequent risk of development of epilepsy is increased.

Most investigators concur that the duration of an SFC is brief; however, there is a significant divergence of opinion among authors as to how long an FC must persist to qualify for a designation of "atypical," "complex," "complicated," "prolonged," or "severe." An aggravating factor is that the physician rarely has an opportunity to witness an FC and must usually depend on the parents' estimation of the duration of the attack, which is seldom precise. In many instances, the reliability of the parents' approximation is especially suspect because of the intensity of their emotional involvement.

In our experience, the vast majority of patients with an SFC disorder had a seizure that persisted for five minutes or less; in a small number, the convulsion continued for about 10 minutes or so, and on rare occasion the seizure lasted for approximately 15 or so minutes.

"Complex," "severe," and "complicated" febrile seizures. Nelson and Ellenberg8 classified an initial "febrile seizure" as "complex" if its duration exceeded 15 minutes, and Wolf et al.11 defined an FC persisting for longer than 10 minutes as "severe." On the other hand, Wallace12 designated as "complicated" a seizure associated with fever that continued for more than one-half hour, in agreement with Lennox-Buchthal,13 who stated that "severe" FCs are usually defined as "lasting longer than 30 minutes."

Thus, it is obvious that a rigid, inflexible limit of duration cannot be assigned to every FC that would infallibly separate "severe" attacks from SFC. We regard an initial FC persisting for up to 10 or so minutes that is unassociated with other characteristics of "epileptic seizures precipitated by fever" as an SFC, and we re-evaluate the situation at periodic intervals. We consider a diagnosis of "epilepsy precipitated by fever" in patients whose initial FC lasts for approximately 15 minutes but who present no additional findings suggestive of this type of FC disorder. Needless to say, such patients are kept under close surveillance, with frequent re-evaluations.

Electroencephalographic findings. Because electroencephalographic examinations "were not routinely available" at the participating facilities, Nelson and Ellenberg8 reported no electrographic findings in their study group. This constitutes a major deficit in their investigation, in our opinion, and consequently affects their results - particularly those regarding prognosis.

When an electroencephalogram has been obtained at least 10 days after defervescence in a patient who has had an FC and specific abnormalities are present (spikes or spikeand-wave forms), there is almost unanimous agreement that this is an important prognostic factor for the likelihood of subsequent clinical afebrile seizures (e.g., epilepsy). Wallace14 investigated the records of 112 children eight to 10 years after their first FC in an attempt to identify factors that might influence the outcome of such seizures. She found that spikes or spike-wave complexes in the EEG occurring within three years of the initial attack correlated significantly with subsequent afebrile seizures.

Neurologic and developmental status. Nelson and Ellenberg8 have reported that children who are "other than normal" in their neurologic or developmental status before their first FC are more likely to develop epilepsy than children who are "normal" before their first "febrile seizure." This is not surprising. It is a finding consistent with the observations of many other authors, notably Lennox-Buchthal13 and Wallace.14 We did not include children who were developmentally or neuroiogically abnormal before their initial FC in our investigations, because it is obvious that they are more prone to developing epilepsy than "normal" children, regardless of whether or not they experience "febrile seizures.

Rate of seizure recurrence. Nelson and Ellenberg 15 examined the recurrence rate of febrile convulsions to see if repetition of such episodes could be considered a prognostic factor for the development of subsequent epilepsy. They found that children who were "normal" prior to their first fever-associated seizure "showed no increasing tendency to acquire epilepsy as the number of febrile seizures increased." Children who were abnormal before the first seizure and who experienced at least three seizures "demonstrated an increasing rate of epilepsy" when compared with children who had had only one or two seizures. In their study of 1,706 children, 67 per cent experienced only a single FC.8

Of the 256 children with SFC in our 1958 investigation,4 123 (48 per cent) had one convulsion, 121 (47.3 per cent) experienced one to four seizures yearly, and 12 (4.7 per cent) suffered six or more convulsions each year. There was no recurrence of "febrile seizures" in any of these patients after six years of age. Of the seven (2.9 per cent) patients who subsequently developed afebrile seizures, six had previously experienced four or more brief FCs per year. On the basis of these findings, we do not believe that a rate of recurrence of SFC of less than four per year is a significant factor in the development of later epilepsy.

Genetic aspects. Nelson and Ellenberg8 reported a positive immediate family history of afebrile seizures in 14 per cent of their patients with FC and considered this finding to be an important prognostic factor in the development of epilepsy. Although we reported a family history of epilepsy in 1 8 per cent of our patients with FC in 1947,16 our subsequent investigations of the hereditary aspects of FC4,17 did not confirm our initial observations, and we currently do not regard a positive family history of afebrile seizures in patients with FC as a significant predictor of epilepsy subsequent to an FC disorder.

On the other hand, our investigations did indicate that in children who experience SFC, the familial incidence of similar seizures is exceedingly high. In 1954, we3 reported a family history of SFC in 58 per cent of our study group of 201 children who were classified as having an SFC disorder. Millichap10 found a familial incidence of FC of 30 per cent in 95 children who themselves experienced "febrile seizures." Certainly, these figures (58 percent and 30 per cent) are considerably higher than most investigators have reported relative to the familial incidence of epilepsy in patients who suffer with overt epilepsy (afebrile seizures). We therefore believe that the exceptionally elevated familial incidence of SFC can be utilized as an additional supportive criterion in categorizing the type of FC and also in prognosticating the outcome in young children whose initial seizure is associated with an elevation of temperature.

PROGNOSIS

The prognosis is excellent for children with simple febrile convulsions. These seizures tend to recur for several years during early childhood, with most children who have had them experiencing from one to three convulsions each year. In our series there was no recurrence of such attacks after the child reached the age of six. All of our SFC patients have been followed for 15 years or more (many for as long as 22 years) since their first seizure, and only 2.9 per cent of them developed afebrile seizures (epilepsy).

On the basis of our findings and those of others - particularly Nelson and Ellenberg8 - we believe the pediatrician should advise parents whose child has experienced his first SFC that there is a good prognosis. Of course he cannot be absolutely dogmatic and tell the parents that there is no chance the child will develop epilepsy, for, as we indicated above, the odds are about the same as they are in the general population: a little less than three out of every 100 such children will subsequently have an afebrile seizure.

We recommend that the child with SFC be observed periodically. If he has additional seizures that are not typical of SFC, if subsequent electroencephalographic studies reveal electrical abnormalities diagnostic of epilepsy, or if the "febrile seizures" recur after the age of six, the physician should then diagnose the condition as an epileptic disorder and manage the patient accordingly.

USE OF ANTICONVULSANT MEDICATIONS

Should children be treated with anticonvulsant medication after their first " febrile seizure?" We do not believe the question can be answered meaningfully until the initial "febrile seizure" is classified as either an SFC or "epilepsy precipitated by fever." The reason is that the treatment of an SFC disorder should be completely different from that of a child with "epileptic seizures triggered by fever."

Proponents of continuous phenobarbital treatment. Some physicians who previously did not treat children with an SFC disorder are now prescribing daily anticonvulsant medications to such patients on the basis of the studies by Faero et al.18 and Wolf and co-workers.11 In addition, other recommendations - devoid of personal experiences - have subsequently appeared in the literature, based on the results of these two investigations and suggesting that children with SFC be treated continuously with phenobarbital.

In our opinion these recommendations have been based on unsubstantial, unconvincing findings and unconventional therapeutic procedures. Therefore we would like to examine briefly these two studies and their findings as they relate to SFC.

1. The study of Faero et al.18 These investigators treated a group of 60 children who were "hospitalized for febrile convulsions" with daily dosages of phenobarbital. At the conclusion of a six-month follow-up they reported that 4 per cent of the children with constant serum concentrations of 16 µg/ml. or above experienced a recurrence of "febrile seizures," whereas 21 per cent of the children whose phenobarbital blood level was consistently 15 µ./ml. or less had a recurrence of FC. This study has been criticized,19,20 primarily because it consisted of a diverse group of children and also because of the absence of prior neurologic and developmental data. Regarding the first point, at least 25 per cent of the children in the study had experienced "epileptic seizures precipitated by fever" (prolonged and/or focal), and so the study was hardly one of a homogeneous series of children with SFC. Regarding the second point, data were lacking regarding the neurologic and developmental status of the children before the initial FC and at the time of the initial FC.

As pediatricians will realize, a follow-up period of six months in evaluating the results of therapy in children with FC is practically worthless, since many children with an SFC disorder experience only one or two such attacks annually, and others experience only an attack or two in their entire lives. Also, Faero et al. say that "of the 48 children without recurrence of any kind, 18 (37.5 per cent) did not have a subsequent episode of fever during the six-month follow-up period." Obviously, these findings cannot be summarily applied to children with SFC. When Heckmatt and coworkers21 duplicated the study by Faero and associates in a group of 165 children, they reported that phenobarbital was ineffective in preventing the recurrence of FC, despite "prophylactic drug serum concentrations."

To maintain a phenobarbital serum concentration of 16 µg./ml. or above requires a daily dosage that will yield a blood level sufficiently high so that even a 30 per cent decrease will not result in a serum concentration of less than 16 µg./ml. The reason is that young children excrete the drug at a rate of about 25 to 30 per cent every 24 hours. It should also be borne in mind that there may be considerable variation in the day-to-day or week-to-week blood-level determination values in a young child receiving a constant daily dosage of phenobarbital, principally due to factors uncontrollable by the prescriber.

2. The study by Wolf et al.11 These investigators studied 355 children seen at the time of their initial FC. In their treatment, the standard method of prescribing phenobarbital was replaced by a new method of continuous plus intermittent therapy. That is, phenobarbital was administered continuously at the rate of 3-4 mg./kg. daily. At the onset of fever an additional loading dose of 30 mg. per year (up to a maximum of 120 mg.) was given, and the daily phenobarbital dosage was increased to 5 mg./kg. for the duration of the fever. This complicated therapeutic regimen has been referred to in the literature as "continuous phenobarbital therapy,"22 and a reader not familiar with the unique methodology employed by Wolf et al. might easily be led to believe he was following their recommendations for preventing recurrence of FC simply by instituting a program of regular phenobarbital treatment on a daily basis.

In our opinion, the therapeutic program proposed by Wolf and his co-workers would be difficult to implement, principally because of the fact that a child's seizure is frequently the first indication the parents have that he is sick. We noticed this sequence of events in approximately half of our patients with SFC. Wolf et al. also were confronted with this situation, since they reported that "parents were unaware of the fever prior to the seizure in about 30 per cent of the cases." But they did not state how, or if, the problem was resolved. The program would also be difficult to implement in the parents were of modest intellectual status, we believe, since they might experience problems in executing the complicated phenobarbital dosage schedule recommended by these authors.

This study, too, consisted of a heterogeneous group of patients with FC and was composed of subjects whose initial "febrile seizure" was akinetic, focal, or prolonged, in addition to children with SFC. The segment of their study population that received only intermittent phenobarbital therapy included patients whose initial "febrile seizure" was focal in 5 per cent, akinetic in 4 per cent, and from 10 to more than 60 minutes' duration in 12 per cent. The group that received no anticonvulsant therapy included children whose first seizure associated with fever was focal in 2 per cent, akinetic in 7 per cent, and from 10 to more than 60 minutes in duration in 10 per cent. Therefore, we believe that data derived from such a conglomerate may be confusing and misleading and cannot be imposed on children with SFC.

How effective was this unusual phenobarbital regimen in preventing seizure recurrences in children with SFC and in those with "epilepsy triggered by fever"? Wolf presented an analysis in a subsequent article,23 discussing statistics that were either absent from or obscure in the original publication. This report showed that the average rate of SFC recurrence during the follow-up period of six to 36 months in the groups receiving either intermittent or no phenobarbital therapy (20.6 per cent) was twice that of the children who received continuous-plus-intermittent treatment (10.2 per cent). But we do not believe that these data constitute a compelling reason to treat children with an SFC disorder with their extraordinary phenobarbital program, for reasons that are outlined below.

Opponents of continuous phenobarbital treatment.

1. Livingston et al. In 1947, we16 published the results of our investigation of FC in 94 children, of whom 63 had received daily anticonvulsant drug therapy and 31 had received irregular or no treatment. "Febrile seizures" recurred in 33 (52 per cent) of the 63 patients treated with daily anticonvulsant medication and in 15 (49 per cent) of the 31 who had received irregular or no therapy. These findings indicated to us that the recurrence rate of FC was not influenced by daily anticonvulsant drug therapy.

Our 1958 study4 of the effect of phenobarbital therapy on the rate of recurrence of FC was made with a large number of children who were divided into two groups. Each of the children was alternately placed in one of the two groups. Children in group 1 were treated with the following doses of phenobarbital: children under two years of age, 48 mg. daily; two to four years of age, 64 mg. daily; four and five years of age, 96 mg. daily. Children in the other group received no medication.

Barbiturate-blood-level examinations were not performed in these patients, but we believe that the dosage in the majority of children was sufficient to have produced a "prophylactic" serum concentration. (Our subsequent experience with phenobarbitalblood-level determinations in hundreds of children six months to six years of age revealed that the administration of this drug in the dosages specified yielded "therapeutic" drug-serum levels in the preponderance of such cases.) We explained to the parents in detail and at great length the absolute necessity and importance of administering the medication exactly as prescribed, and, to our knowledge, noncompliance was not a significant factor in our study group.

We found that children with SFC who received phenobarbital on a continuous basis had a seizure-recurrence rate that was essentially the same as in those with SFC who did not receive any therapy. On the other hand, the recurrence rate of seizures in children with "epileptic seizures precipitated by fever" was much less in the patients receiving daily phenobarbital dosages than in those who did not receive medication.

The children taking part in our 1972 investigation17 were located in essentially the same manner as those in our 1958 study. The staffs of the general outpatient clinic and emergency department at the Johns Hopkins Hospital Department of Pediatrics were instructed to refer to our clinic the first 200 children seen with a first "febrile seizure" that was unassociated with cerebral disturbance of any type either before or at the time of the initial attack. These patients received a complete seizure work-up and were placed consecutively into one of the following four groups: 1. Those who received daily doses of phenobarbital consistent with our 1958 dosage schedule; 2. those who were given phenobarbital daily with instructions to be given aspirin at the first sign of fever or infection; 3. those who were given a placebo daily; and 4. those who received a daily placebo with instructions to take aspirin at the initial sign of fever or infection. Placebos were prescribed in this study as an incentive for the parents to return their children for follow-up visits.

Unfortunately, this investigation, in which the patients were to have been followed up for at least 15 years, was never concluded because some of the records were lost when our offices were transferred from the Harriet Lane Home to the Childrens Medical and Surgical Center of the Johns Hopkins Hospital. However, statistical projection of the available data relative to the efficacy of phenobarbital in preventing "febrile seizure" recurrences yielded findings compatible with the results of our 1958 study - i.e., the rate of seizure recurrence in children with SFC who received daily phenobarbital did not differ significantly from the incidence of additional attacks in those with SFC who took a placebo every day, whereas the seizure-recurrence rate in patients with "epilepsy triggered by fever" who received continuous phenobarbital therapy was significantly decreased as compared with those with the same form of FC disorder who received the placebo regimen.

From the results of our three investigations, we concluded that the administration of daily anticonvulsive therapy to children with an SFC disorder was of no avail but that chronic treatment with phenobarbital significantly reduced the recurrence rate of attacks in patients with "epileptic seizures precipitated by fever."

Our conclusions, therefore, are:

Children with SFC: no daily prophylactic anticonvulsant therapy.

Children with "epilepsy precipitated by fever": daily antiepileptic medication, just as with patients whose seizures are not associated with elevated temperature.

Children whose parents reject or cannot accept a management program excluding daily anticonvulsants or who are unduly anxious during implementation of such a program: administration of phenobarbital and aspirin at the first sign of fever or infection. Although of little or no therapeutic value to the patient, this plan gives parents "something to do" and may alleviate their anxiety.

2. Heckmatt et al.21 Faero and associates18 had reported that daily phenobarbital therapy reduced the recurrence rate of FC. In an endeavor to corroborate their results, Heckmatt et al. investigated 161 children without known neurologic disorder who had experienced their initial "febrile seizure" between six months and three years of age. Phenobarbital was prescribed to 88 children in a daily dosage of 4-5 mg./kg., and the drug plasma level was monitored at intervals of one to three months; the drug dosage was adjusted, if necessary, to maintain a phenobarbital blood level between 15 and 30 µg./ml. The control group consisted of 73 subjects who received no anticonvulsant treatment.

At the conclusion of the study, 10 of the 88 children treated with phenobarbital experienced a recurrence of FC, as against 14 of 73 children in the control group. Forty-nine of the 88 children in the treatment group took their medication regularly, and four of them had additional FC, "a proportion not significantly different from that in the controls." All four patients presented mean phénobarbital plasma concentrations in excess of 16 /UgJmI. during the period of investigation, and in three cases the blood level was 16 /igJml. or above within eight hours of the repeat convulsion. The authors concluded that the results of their study did not confirm the findings reported by Faero and co-workers and that daily phénobarbital therapy "does not seem to prevent febrile convulsions," despite the maintenance of "prophylactic" drug plasma levels.

Intermittent therapy. It is apparent that some, if not many, physicians do not appreciate the fact that intermittently administered phénobarbital is of practically no value in preventing the occurrence of an FC - as indicated by the survey of Asnes et al.,24 who found that 54 per cent of the responding pediatricians advised the parents "to administer phenobarbital only when the child has a febrile illness." It should be noted, however, that a "number" of these physicians commented that they employed this therapy "primarily for parental reassurance."

The ineffectiveness of intermittent phénobarbital therapy in the prevention of a recurrence of FC is primarily due to the following two factors: 1. in many instances (at least 30 per cent), the convulsive episode is the first indication to the parents that their child is ill, and 2. even if the conventional oral dosage of phénobarbital is doubled, a "therapeutic" serum concentration will not be achieved for two days. In addition, it is difficult, if not impossible, to accurately and completely evaluate the efficacy of intermittent treatment, because not all children with FC experience a seizure every time they have an elevation of temperature.

Negative findings relative to the value of phénobarbital administered daily or intermittently in the prevention of FC recurrences have been published by Millichap et al.25 These authors reported that "febrile seizures" recurred in 53 per cent of patients who received intermittent phénobarbital therapy and in 43 per cent of those treated continuously with this drug, and Millichap10 stated that "control of febrile seizures was poor and of questionable degree in the patients of both groups." The results of the investigation by Wolf and his co-workers11 also demonstrated that intermittently administered phénobarbital is not effective in preventing FC recurrences.

Phenytoin therapy. A number of investigations have confirmed the ineffectiveness of phenytoin (Dilantin) in preventing or reducing the recurrence rate of FC,25"28 negating undocumented recommendations to the contrary by Gold and Carter,29 among others. Especially noteworthy is the study of the efficacy of phenytoin in preventing FC recurrences by Melchior et al.,28 in which the FC recurrence rate of 29 adequately treated children was compared with that of 218 who did not receive any medication. The phenytoin dosage was adjusted, when necessary, to maintain a serum concentration of 10-20 /Lig./ml. The authors reported that the rate of FC recurrence in subjects under three years of age was the same in patients who received daily phenytoin as in the untreated group. "Decisive proof" that phenytoin does not prevent FC in children under three years of age was found in four patients who experienced repeat seizures when the phenytoin serum concentration was, respectively, 17, 18, 21, and 27 /ig JmI.

It was gratifying to note that none of the pediatricians who participated in the study by Asnes et al.24 employed phenytoin either continuously or intermittently in their management of patients with SFC. These authors also reported that only 15 per cent of the respondents prescribed phénobarbital on a daily basis to children with an SFC disorder.

Treatment in relation to "risk factors." Predicated on their data regarding the significance of various "risk factors" in the development of epilepsy following an FC disorder, Nelson and Ellenberg8 concluded that children with FC of the type we classify as SFC "would not seem an appropriate target population for chronic treatment aimed at the prevention of afebrile seizures." We have continually made exactly the same recommendation since our 1954 publication,3 and Millichap10 has expressed a similar opinion. On the other hand, Rabe30 and Gellis31 advise that children be placed on daily anticonvulsant-drug therapy after their second SFC, and Pollack22 says that "continuous phénobarbital administration during the preschool years is indicated for most children who have a simple febrile convulsion." However, none of these authors present personal documentation of their own findings in support of their recommendations.

Nelson and Ellenberg8 suggest that children with "febrile seizures" who possess two of the three possible "risk factors" should be considered for chronic treatment. The risk factors they list are complicated initial seizure, family history of afebrile seizures, and a pre-existing neurologic abnormality. As noted above, on the basis of our 19543 and 19584 data and our subsequent experience, we continue to recommend that children whose FC disorder fulfills our criteria for a diagnosis of "epilepsy precipitated by fever" be treated with daily anticonvulsant therapy as intensively as patients whose epileptic seizures are not associated with pyrexia. Gellis31 and Rabe30 also advise that anticonvulsant-drug treatment be administered to children after their first FC if the convulsion is other than an SFC. It is emphasized that Nelson and Ellenberg8 found that 5 per cent of their patients with only one risk factor subsequently developed "afebrile seizures" or "epilepsy."

In the opinion of Chao et al.,32 all children with an FC disorder should be treated daily with phénobarbital, because of the "difficulty in distinguishing the benign [simple] from the nonbenign [complex, complicated, severe] febrile convulsions and in identifying the complications that may result from a seizure." This is a unique and isolated position that is inconsistent with the findings of most other authors, and we find it difficult to justify in the light of the investigations cited above.

RECURRENCE OF SFC

Simple febrile convulsions by and large are a benign and self-limited disorder. They recur in from 30 to 50 per cent of the children who have an initial seizure. While it is certainly desirable to prevent the recurrence of one, the point of prime importance is not whether treatment will prevent a subsequent SFC but whether or not it will result in a reduced likelihood of epilepsy.

We have already noted that there is no statistical evidence to support such a claim. We re-emphasize the large series reported by Nelson and EUenberg,8 who found that the only significant risk factors identified with development of epilepsy following FC were a complicated initial seizure, a family history of afebrile seizures, and pre-existing neurologic abnormalities. Children who were neurologically and developmentally normal before their first FC "showed no increasing tendency to acquire epilepsy as the number of febrile seizures increased."15

Some physicians treat children with SFC with daily anticonvulsant medication because of their "hope" that the prevention or a diminution in the recurrence rate of such "febrile seizures" will result in a decreased incidence of subsequent epilepsy. We stress that there is currently no evidence to support such attitudes and that no investigation has demonstrated that a reduction in the recurrence of SFC has resulted in a diminution of the prevalence of subsequent afebrile seizures (epilepsy). It should be noted that, to our knowledge, all studies that consisted of or included the administration of anticonvulsant drugs to children with FC were designed to determine the efficacy of medication in reducing the recurrence rate of "febrile seizures" and not to assess the value of such agents in diminishing the incidence of epilepsy following an FC disorder.

HAZARDS OF CONTINUOUS PHENOBARBITAL TREATMENT

Continuous phénobarbital therapy is not without hazard or encumbrance, and attention is directed to the difficulties encountered by Wolf and Forsythe.33 These authors reported that 46 (42 per cent) of 109 children treated with daily phénobarbital manifested a behavioral disorder. Of these 46 children, 38 were hyperactive and eight exhibited disobethence, increased irritability, insomnia, persistent lethargy, or tantrums. These investigators were compelled to discontinue the phénobarbital regimen in 22 of the 38 children with hyperactivity and in three of the eight with other behavioral aberrations. Of the 22 subjects in whom phénobarbital was discontinued because of hyperactivity, the behavioral abnormality did not disappear in six.

Despite their vigorous efforts, Wolf and co-workers11 found "considerable parental resistance to the continuous phénobarbital regimen," and noted that compliance was a major problem. The only way the physician could be assured of maintaining adequate serum concentrations in the continuously treated child, they said, was by ordering frequent blood-level determinations and by frequently encouraging the parents. Heckmatt et al.21 also encountered the problem of parental resistance, even in parents whose children had experienced prolonged, severe "febrile seizures."

The maintenance of "therapeutic" phénobarbital serum levels in infants and young children, as suggested by Wolf and his coworkers, may be a formidable task. Heckmatt et al.21 emphasized the difficulty of doing so for even relatively brief periods of time. The objections to performing blood-level determinations continually in infants and young children because of the adverse physical and psychological effects upon the child are patent, not to mention the penalties that would have to be borne by the parents in anxiety, money, and time. Brett34 aptly stresses that there are strong arguments against the frequent withdrawal of blood in young children, "if only because it can turn what should be a pleasurable visit to the hospital or clinic into a dreaded ordeal." He adds that the practice could lead to an increase in defaulters, "since parents need to be convinced that a painful procedure is essential if they are to tolerate it for their children."

The frequent withdrawal of blood may explain to a large extent the high rate of attrition in the studies conducted by Wolf et al.11 and Heckmatt and associates.21 Gellis31 has also called attention to the problems associated with frequent blood-level determinations in infants and young children.

What about the effect of continuous phénobarbital treatment on the child's performance at school? The effects on academic achievement have not been precisely established. But there are indications that the mental performance of persons receiving this drug in doses appropriate for seizure control is not optimal.35

CONCLUSION

The prevalency rate of epilepsy subsequent to SFC essentially parallels that of the general population. Since there appears to be little relationship between the recurrence rate of SFC and the development of afebrile seizures (epilepsy), we believe administration of continuous phénobarbital therapy is untenable in children with an SFC disorder, because of the hazards, difficulties, and consequences associated with such therapy. We believe continuous phénobarbital therapy is justified in the case of "epileptic seizures precipitated by fever."

Questions Pediatricians Ask About Febrile Convulsions

Do "febrile convulsions" result in mental retardation or adversely affect the child's intellectual ability?

From our discussions with thousands of physicians over the past four decades, we have learned that a major concern relative to "febrile seizures" is that such attacks, particularly if recurrent, often result in intellectual impairment or mental retardation. This notion has prompted some clinicians to institute daily anticonvulsant drug therapy in all children after their first FC, even if it is an SFC.

We reviewed the literature concerning the relationship of FC to subsequent brain damage and found that many of the reports indieating such an association included significant numbers of children who were developmentally or neurologically abnormal before the initial convulsion, or whose seizures were prolonged or accompanied by an acute cerebral disorder such as encephalitis or meningitis. We are in agreement with Millichap's statement:10

The concept that the febrile convulsion per se may precipitate pathological and permanent alterations in the immature brain is based on equivocal evidence obtained from isolated and atypical cases and indirectly from observations of the effects of fever and convulsions on the electroencephalogram. The true incidence of brain injury and permanent hemiparesis resulting from the effects of febrile convulsions is probably less than 0.2 per cent, and the majority of reported cases lack detailed data regarding the previous history and clinical manifestations of the illness.

Ellenberg and Nelson36 recently conducted a comprehensive, controlled study of the intellectual performance of 862 children - 431 whose initial seizure was associated with fever, and an equal number of siblings with no history of a seizure disorder. They concluded that febrile seizures - initial or recurrent - are not likely to cost the child a measurable decrement in intelligence or early academic achievement.

As far as we know this is the first investigation in the United States that examined through standardized testing procedures the subsequent intellectual performance of children who were neurologically and developmentally normal before an initial FC that was unassociated with any cerebral infection or neurologic illness. The results of their investigation are in accord with our previous clinical observations that SFCs are not associated with subsequent brain damage.17,37 These findings should certainly allay to a great extent the anxiety concerning FC-induced intellectual deficit. It was especially reassuring to note that the intellectual performance of children with FC was impaired only when there had been a neurologic or developmental problem prior to the initial seizure.

Although we did not routinely employ standardized instruments to examine intellectual status, we have, by means of questionnaire or personal interview or both, followed during their entire scholastic careers a large group of patients who experienced SFC during early childhood and did not observe an elevated incidence of academic deficiency. In fact, none of our patients who were developmentally, intellectually, and neurologically normal before their first SFC exhibited a deterioration in mentality as a result of such convulsive episodes, whether single or recurrent.

Are "febrile seizures" a causative factor in psychomotor (temporal-lobe) epilepsy?

There has been some anxiety relative to the development of psychomotor (temporal-lobe) epilepsy in a child with an FC disorder. We believe this anxiety is due principally to the reports by Falconer and his associates38"40 in which they described the clinical and pathologic findings and results of unilateral anterior lobectomy performed in 200 patients with psychomotor seizures.

Mesial-temporal (Amnion's horn) sclerosis was found in approximately half of the patients, and was attributed in 35 per cent of these cases to a prolonged "febrile seizure," serial "febrile seizures," or febrile status epilepticus. The fact that a large percentage of these patients benefited from excision of the anterior temporal lobe indicates that mesialtemporal sclerosis was responsible for the psychomotor attacks. Falconer41 studied 30 patients nine to 15 years of age who underwent anterior temporal lobectomy for psychomotor epilepsy and identified mesialtemporal sclerosis in 20 of these children. A history of a prolonged "febrile convulsion" was obtained in 15 of the patients who exhibited this lesion.

Ounsted et al. 42 investigated the genetic aspects of temporal-lobe epilepsy in 100 children, 32 of whom had experienced a prolonged "febrile seizure" but presented no other findings suggestive of cerebral insult. "Febrile convulsions" had also occurred in 30 per cent of the siblings of these 32 patients, but none oí them had psychomotor epilepsy, nor had any of them sustained an FC of long duration. These authors concluded that children predisposed to "febrile seizures" early in life and who experience an FC longer than 30 minutes in duration may develop psychomotor epilepsy because "the seizure itseli damages the brain. . . . The brothers and sisters, having milder febrile convulsions, do not suffer any secondary brain damage and make full recoveries."

Gastaut and associates43 had previously observed that the majority of patients whose "febrile seizures" included a iocal component subsequently experienced psychomotor attacks, and Wallace14 recently reported that the development of psychomotor epilepsy is significantly related to a prolonged or repeated FC with unilateral features.

Thus, it is apparent that psychomotor epilepsy following an FC is essentially restricted to children whose "febrile seizures" are prolonged or focal or both. The etiologic findings previously cited are consistent with our clinical and electroencephalographic data that children with SFC are quite unlikely candidates for psychomotor epilepsy; in fact, none of our patients with SFC subsequently experienced psychomotor seizures. On the other hand, Chao and her co-workers32 stated without documentation or supportive data that psychomotor epilepsy is more likely to develop in children whose FCs are "benign" than in those with complex or severe "febrile seizures." We are perplexed by this expression, since, to our knowledge, it is discordant with the reports and experience of all other investigators.

It should also be noted that Nelson and Ellenberg8 reported that their large group of children with FC did not develop psychomotor attacks "proportionately more often than other forms of epilepsy."

Is an attack of febrile status eptlepticus likely to occur after a child has had an initial simple febrile convulsion?

No, we have found no evidence to support such a statement. It is true that Pollack said in one report that "the possibility that a child who has a simple febrile convulsion will subsequently develop febrile status epilepticus is a 'major concern' ".22 And this "possibility"is not infrequently cited in support of a recommendation to treat the child with an SFC with anticonvulsant medication on a daily basis. As far as we know, the statement is based entirely on the retrospective study done by Aicardi and Chevrie44 - a study we believe to be totally irrelevant insofar as any relation between SFC and status epilepticus is concerned.

Aicardi and Chevrie reported on 239 children who had been hospitalized for an episode of status epilepticus. They noted that 67 of their patients - 28 per cent - had an elevated temperature due to "a common upper respiratory infection or one of the childhood exanthemas." We emphasize, because it has been overlooked or misinterpreted, that the fever in their patients occurred in association with the attack of status epilepticus and not with an antecedent convulsive episode. In fact, there is no statement in the article that any of the 239 children had a history of a "febrile seizure" prior to the occurrence of the attack of febrile status epilepticus.

We have never had a patient who developed febrile status epilepticus if his initial "febrile seizure" was an SFC. And we are quite sure that - Pollack's statement to the contrary - there is no need for any concern about such an eventuality. Nor have we seen any data in the literature to support such a conclusion. For example, in their study of 1,706 children with FC, Nelson and Ellenberg8 do not even mention febrile status epilepticus as a consequence of seizures associated with fever.

Should a lumbar puncture be performed in the child following an initial "febrile convulsion"?

We don't think a lumbar puncture should be performed routinely, and it is our policy not to perform one when there is a simple febrile convulsion unless there are extenuating circumstances of the kind mentioned below. A lumbar puncture is not a completely innocuous procedure. In addition, it is costly, particularly when repeated, and can be traumatic emotionally to both the patient and the parents.

When one reviews the literature regarding spinal taps in the child with FC, the yield is ambiguous and confusing statements, a great many differences of opinion, and a number of specific recommendations made without convincing supporting data. For example, Krugman and Ward45 feel that "it is best to err on the side of doing an unnecessary spinal tap rather than to miss a diagnosis of meningitis," since early, accurate diagnosis followed by appropriate therapy has a profound effect on the child's prognosis. But, in the next paragraph, these same authors caution that "lumbar punctures should not be done indiscriminately and in the absence of a reasonable suspicion of meningitis."

Some physicians recommend that a spinal tap be performed on all children following their first FC regardless of the age of the patient46, whereas others advise that this examination be carried out in children whose initial FC occurs before the age of 15 months,47 18 months,48 or 24 months.40

Rutter and Smales48 reviewed the histories of 328 children who were hospitalized after their first FC; lumbar puncture had been performed in 96 per cent. Four cases of meningitis were diagnosed exclusively by this examination, all in infants under 1 8 months of age. These authors note that signs of meningitis are usually obvious in patients over the age of 18 months but are equivocal or absent in younger children, and they therefore recommend "routine lumbar puncture in children under the age of 18 months presenting with a first febrile convulsion."

In a subsequent publication,50 however, these authors point out that **?? an experienced pediatrician kept a careful and constant eye on each child, a routine lumbar puncture would not be necessary. ..."

Ratcliffe and Wolf 46 reviewed the records of 325 children with meningitis hospitalized during a 10-year period and reported that 43 (13 per cent) oí this group presented with a seizure associated with fever. Of these 43 patients, 15 (35 per cent) lacked signs of meningeal irritation, 25 (58 per cent) manifested signs of meningitis, and in three (7 per cent) the absence or presence of clinical indications "could not be determined." Eleven of the 15 children without signs of meningitis were under three years of age, and three of these 15 patients had previously experienced FC. Of the children for whom such information was available, the duration oí the seizure was reported as less than five minutes in 83 per cent and was described as generalized in 94 per cent, and consequently the authors concluded that "most of the febrile convulsions with meningitis might have been classified as 'simple' febrile convulsions."

On the basis of the findings of their retrospective survey, Ratcliffe and Wolf46 recommended that lumbar puncture be performed on all children with an initial FC and that it be considered in children under three years of age with recurrent FC. They suggest that it might be omitted in the older child with recurrent FC if "(1) there is no clinical suspicion of meningitis, (2) one finds an alternative cause of fever, and (3) rapid improvement occurs under observation in the hospital."

In a commentary entitled "To Tap or Not to Tap," Menkes51 stated that "a lumbar puncture has become nearly routine in the evaluation of a child who has just experienced his first febrile convulsion." This statement allows the reader no alternative except to assume that this procedure is "mandatory" in the initial evaluation of every child who experiences an FC, regardless of the clinical manifestations of the seizure and the data derived from the medical history. Kafka52 agreed that Menkes's statement is applicable to children seen in a teaching hospital or an outpatient department of a university clinic but challenged the implication that it was appropriate for pediatricians in private practice. Menkes53 responded that his recommendation was directed not to physicians with extensive experience in pediatrics but to "those physicians who have not yet acquired the 'feel' for a seriously ill child." Lewak54 noted that Menkes's statement relative to the obtention of a lumbar puncture pertained to a child's initial FC, and he cogently inquired, "Couldn't that cause the non-thinking resident not to tap a child with meningitis simply because the child was having his second convulsion associated with fever?"

Asnes et al.24 reported that 41 per cent of the pediatricians in private practice who participated in their survey routinely employed a lumbar puncture in their diagnostic evaluation of children with an initial FC. During the past 35 years we have participated in the scientific assemblies at innumerable national and regional medical meetings and have had the opportunity to interrogate by means of personai interview or questionnaire thousands of pediatricians in regard to their policy of performing a spinal tap in children following their first FC. We found that approximately 75 per cent of those we questioned did not routinely order a lumbar puncture in such situations but utilized an individualistic approach in their selection of diagnostic procedures, based on the patient's medical history and the clinical findings.

From our discussions with many hundreds of pediatricians during the past several years, we have learned that the litigatory atmosphere currently surrounding the practice of medicine in the United States has had a great impact on the performance of some diagnostic procedures, particularly in the case of such potentially catastrophic diseases as meningitis. As a result many admittedly unnecessary examinations are ordered by some physicians as "protection" against the possibility of subsequent legal action by a patient. We are convinced that the threat of a lawsuit, rather than an authentic suspicion of meningitis, has induced many physicians to perform a lumbar puncture in a child with an FC and may account in great measure for the large number of pediatricians who routinely employ this examination in their diagnostic evaluation of children following their initial FC.24

Our experience contrasts sharply with the opinion of Ratcliffe and Wolf46 that "most of the febrile convulsions with meningitis" in their retrospective review "might have been classified as 'simple* febrile convulsions." We do not recall having seen a seizure that occurred soon after an elevation of temperature and fulfilled our criteria for an SFC to be associated with meningitis. Our studies of children with meningitis have revealed that the presenting FC episode, if there is one, generally does not occur proximate to the onset of fever but does so some 12 to 18 or even 24 or so hours after the patient has become pyrexic. In addition, the initial FC associated with meningitis in our experience is not typically brief and generalized.

We cannot refrain from wondering how a who, in accordance with some pubrecommendations, routinely performs a puncture in children with their first will diagnose a child's second, third, or "febrile seizure." Certainly, normal of a lumbar puncture in a child with initial FC might lull a physician into a false of security with regard to subsequent FCs. We have seen patients whose initial seizure" was unassociated with a infection but in whom a subsequent was one of the clinical manifestations of meningitis. If one accepts the premise that a lumbar puncture is indicated after a child's initial FC, it follows that this examination should be performed after all subsequent FCs, and we do not believe that such diagnostic methodology is justified or warranted. We agree with Lewak's advice that "the diagnostic process (for meningitis or any other condition) should be based on a careful and thoughtful analysis of the patient" and not on the implementation of rigid "rules" or instructions.54

On the basis of comprehensive and extensive experience, we are committed to the concept that each child with an FC should be evaluated on an individual basis and that diagnostic procedures should be ordered in accord with the physician's clinical findings and judgment. It is our general current policy not to perform a lumbar puncture in a child with a "febrile seizure" of the type that we classify as an SFC unless the patient is under six months of age or the initial attack occurred after the age of six years or the child may not be available for re-examination during the three-to-four-day period following the convulsive episode.

BIBLIOGRAPHY

1. PampigUone, G. Children with "febrile convulsions." Lancet 2 (1973), 1035.

2. Livingston, S., and Kajdi, L. Importance of heredity in the prognosis of febrile convulsions. Am. J. Dis Child. 69 (1945), 324.

3. Livingston, S . The Diagnosis and Treatment of Convulsive Disorders in Children. Springfield, IU.; Charles C Thomas, 1954.

4. Livingston, S. Convulsive disorders in infants and children. In Levine, S. Z. (ed.). Advances in Pediatrics, Volume 10. Chicago: Year Book Publishers, 1958.

5. Eisner, V., Pauli, L. L., and Livingston, S. Hereditary aspects of epilepsy. Bull Johns Hopkins Hasp. 105 (1959), 245-271.

6. Miller, F. M. W., et al. Crowing Up in Newcastle Upon Tyne. London: Oxford University Press, 1960.

7. Gomez, J. G., Arriniegas, E., and Torres, J. Prevalence of epilepsy in Bogotá, Colombia. Neurology 28 (1978), 90-94.

8. Nelson, K. B., and Ellenberg, J. H. Prognosis in children with febrile seizures. Pediatrics 61 (1978), 720-727.

9. Friderichsen, C, and Melchior, J. Febrile convulsions in children, their frequency and prognosis. Acta Paediatr. (Suppl) 100 (1954), 307-317.

10. MUlichap, ]. G. Febrile Convulsions. New York: Macmillan, 1968.

11. VVolf, S. M., et al., The value of phénobarbital in the child who has had a single febrile seizure: a controlled prospective study. Pediatrics 59 (1977), 378-385.

12. Wallace, S. J. Factors predisposing to a complicated initial febrile convulsion. Ardi. Dis. Child. 50 (1975), 943-947.

13. Lennox-Buchthal, M. A. Febrile Convulsions: a Reappraisal. Amsterdam: Elsevier Publishing Company, 1973.

14. Wallace, S. J. Spontaneous fits after convulsions with fever. Arch. Dis. Child. 52 (1977), 192-196.

15. Nelson, K. B., and Ellenberg, ). H. Predictors of epilepsy in children who have experienced febrile seizures. N. Engl. J. Med. 295 (1976), 1029-1033.

16. Livingston, S., Bridge, E. M. and Kajdi, L. Febrile convulsions: a clinical study with special reference to heredity and prognosis. /. Pediatr. 31 (1947), 509-512.

17. Livingston, S.: Comprehensive Management of Epilepsy in Infancy, Childhood and Adolescence. Springfield, ??.; Charles C Thomas, 1972.

18. Faero, O., et al. Successful prophylaxis of febrile convulsions with phénobarbital. Epilepsia 13 (1972), 279-285.

19. Livingston, S., and Pauli, L. L. Treatment of "febrile seizures."/. Pediatr. 89 (1976), 164-165.

20. Ouellette, E. M. The child who convulses with fever. Pediatr. Clin. North Am. Il (1974) 467-481,

21. Heckmatt, J. Z., et al. Failure of phenobarbitone to prevent febrile convulsions. Br. Med. J. 1 (1976), 559-561.

22. Pollack, M. A. Continuous phénobarbital treatment after a 'simple febrile convulsion'. Am. /. Dis. Child. 132 (1978), 87-89.

23. Wolf, S. M.: Effectiveness of daily phénobarbital in the prevention of febrile seizure recurrencies in "simple" febrile convulsions and "epilepsy triggered by fever". Epilepsia 18 (1977), 95-99.

24. Asnes, R. S., et al. The first febrile seizure: A study of current pediatric practice. J. Pediatr. 87 (1975), 485-488.

25. MUlichap, J. G., Aledort, L. M., and Madsen, J. A. A critical evaluation of therapy of febrile seizures. /. Pediatr. 56 (1960), 364-368.

26. Frentzen, E., Nygaard, ?., and Wulff, H.: Febrile kramper hos bom. Prognostíke studier og forsog pa vurdering af effeckten af profylaktisk antiepUeptisk langtidsbehandling en forelobig meddelelseUgeskr. Laeger 126 (1964), 207-211.

27. Frentzen, E., Lennox-Buchthal, M., and Nygaard, A.: Longitudinal EEG and clinical study of children with febrile convulsions. Electroeticephalogr. Clin. Neurophyswrf. 24 (1968), 197-212.

28. Melchior, J. C, Buch thai. F., and Lennox-Buchthal, M. The ineffectiveness of diphenylhydantoin in preventing febrile convulsions in the age of greatest risk, under three years. Epilepsia 12 (1971), 55-62.

29. Gold, A. P., and Carter, S. Pediatric neurology. In Shirkey, H. C (ed.). Pediatric Therapy. St. Louis: The C V. Mosby Company, 1966.

30. Rabe, E. F. Comment on predictors of epUepsy in children who have experienced febrile seizures. In Gellis, S.S. (ed.) The Year Book of Pediatrics. Chicago: Year Book Medical Publishers, 1978, p. 307.

31. GeIUs, S. S. Editorial comment on continuous phénobarbital treatment after a "simple febrile convulsion." Am. J. Dis. Child. 132 (1978) 89.

32. Chao, D., Carter, S., and Gold, A. P.: Paroxysmal disorders. In Rudolph, A. M., Barnett, H. L., and Einhorn, A. H. (eds.). Pediatrics. New York: Appleton-Century-Crofts, 1977.

33. Wolf, S. M., and Forsythe, A. Behavior disturbance, phénobarbital, and febrile seizures. Pediatrics 61 (1978), 728-731.

34. Brett, E. Implications for measuring anticonvulsant blood levels in epilepsy. Deo. Med. Child. Neurol. 19 (1977), 245-251.

35. Hurt, S. J., Jackson, P. M., Belsham, A., and Higgins, G.: Perceptual-motor behaviour in relation to blood phenobarbitone level. A preliminary report. Deo. Med. Chad. Neurol. 10 (1968), 626632.

36. EUenberg, J. H., and Nelson, K. B. Febrile seizures and later inteUectual performance. Arch. Neurol. 35 (1978), 17-21.

37. Livingston, S. Seizure disorders. In Gellis, S. S., and Kagan, B. M. (eds.). Current Pediatric Therapy. Philadelphia: W. B. Saunders Company, 1966.

38. Falconer, M. A., Serafetinides, E. A., and CorseUis, ). A. N. Etiology and pathogenesis of temporal lobe epilepsy. Arch. Neurol. 10 (1964), 233-248.

39. Falconer, M. A., and Taylor, D. C. Surgical treatment of drug-resistant temporal lobe epilepsy due to mesial temporal sclerosis: etiology and significance. Arch. Neurol. 18 (1968), 353-361.

40. Falconer, M. A. Genetic and related aetiological factors in temporal lobe epilepsy. A review. Epilepsia 12 (1971), 13-31.

41. Falconer, M. A. Mesial temporal (Amnion's horn) sclerosis as a common cause of epUepsy: aetiology, treatment, and prevention. Lancet 2 (1974), 767-770.

42. Ounsted, C, Lindsay, }., and Norman, R. Biological Factors in Temporal Lobe Epilepsy. London; William Heineman Medical Books. 1966.

43. Gastaut, H., et al. H. H. E. syndrome: hemiconvulsions, hemiplegia, epUepsy. Epilepsia 1 (1960), 418-447.

44. Aicardi, J., and Chevrie, J. J. Convulsive status epilepticus in infants and chUdren. A study of 239 cases. Epilepsia U (1970), 187197.

45. Krugman, S., and Ward, R. Infectious Diseases of Children. St. Louis: The C. V. Mosby Company, 1968.

46. Ratcliffe, J. G, and Wolf, S. M. Febrile convulsions caused by meningitis in young chUdren. Ann. Neurol. 1 (1977), 285-286.

47. Solomon, G. E., and Plum, F. Clinical Management of Seizures: a Guide for the Physician. Philadelphia: W. B. Saunders Company, 1976.

48. Rutter, N., and Smales, O. R. C. Role of routine investigations in children presenting with their first febrile convulsion. Arch. Dis. Child. 52 (1977), 188-191.

49. Menkes, J. H. Textbook of Child Neurology. Philadelphia: Lea & Febiger, 1974.

50. Rutter, N., and Smales, O. R. C. Lumbar puncture in children with convulsions. Lancet 2 (1977), 190-191.

51. Menkes, J. H. To tap or not to tap. Pediatrics 51 (1973), 560-561.

52. Kafka, H. L. To tap, or not to tap, in the febrile convulsion. Pediatrics 52 (1973), 619.

53. Menkes, J. H. To Up, or not to tap, in the febrile convulsion. Pediatrics 52 (1973), 619-620.

54. Lewak, N. To tap or not to tap in convulsion (free wUl vs damnation). Pediatrics S3 (1974), 766-767.

TABLE 1

FACTORS THAT HELP TO DIFFERENTIATE SIMPLE FEBRILE CONVULSIONS FROM EPILEPTIC SEIZURES PRECIPITATED BY FEVER

10.3928/0090-4481-19790301-04

Sign up to receive

Journal E-contents