febrile convulsions FEBRILE CONVULSIONS J. GORDON MILLICHAP M.D., M.R.C.P. Professor of Pediatrics and Neurology, Northioestern University Medical School AND Head, Division of Neurology and Seizure Clinic, Childrens Memorial Hospital, Chicago The Macmillan Company new york Collier-Macmillan Limited London © Copyright, J. Gordon Millichap, 1968 All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photo- copying, recording or by any information storage and retrieval system, without permission in writing from the Publisher. First Printing Library of Congress catalog card number; 67-24785 THE MACMILLAN COMPANY, NEW YORK COLLIER-MACMILLAN CANADA, LTD., TORONTO, ONTARIO PRINTED IN THE UNITED STATES OF AMERICA To LOUIS S. GOODMAN whose generosity and wise counsel have guided my introduction to basic scientific research, and in memory of WILLIAM GORDON LENNOX whose dedication and literary contributions sparked my interest in epilepsy FOREWORD The sudden occurrence of a convulsion in a child is one of the most terrifying things that young parents face. Even the question of life and death intrudes on their thoughts. After the first seizure abates, their concern continues, for a new series of questions arises: Why did this happen? Is this “epilepsy”? Is he going to be mentally retarded? The physician must also face a series of prac- tical questions: Why did this happen? Is it going to happen again? What further investigations must I suggest? Shall I start him on a regime of anticonvulsant drugs? Dr. Millichap has drawn together herein the results of his own extensive investigations on febrile convulsions in children and a detailed survey of the world literature on this subject. His broad background of clinical training in England and the United States, as well as his intimate experience with convulsive disorders in the pediatric department of the Albert Einstein College of Medicine, the Mayo Clinic, and now the Children’s Memorial Hospital of Chicago, puts him in an ideal position to write this monograph, which answers the above and many other questions. FOREWORD The field of convulsive disorders has suffered for a long time from the adoption of unfounded assumptions and from the failure to correlate the laboratory with the clinic. The result has been too little attention paid to febrile seizures by the busy practitioner, on the one hand, and over-enthusiastic resort to expensive and dangerous diagnostic studies and therapeutic trials, on the other. The magnitude of the problem is not appreciated by the average physician until he is faced with its repeated occurrence in prac- tice. Because of the diversity of opinion regarding the diagnostic approach and the therapeutic regime, a more complete review of the subject has been needed for many years. I am very happy that Dr. Millichap has done this, and I trust that his book will not only be of value to all physicians who care for children but also serve as a base for new investigative studies into this com- mon, and frequently serious, problem. Robert B. Lawson, M.D. Chairman, Department of Pediatrics, Northwestern University Medical School; Chief of Staff, Children’s Memorial Hospital, Chicago PREFACE In this treatise on febrile convulsions, the most common and yet perhaps the most controversial neurologic disorder of infancy and childhood, the authors personal experience and investiga- tions are described and the world literature on febrile convulsions and related disorders is reviewed. The general clinical manifesta- tions, laboratory findings, prognosis, and treatment are discussed, as are clinical and basic research data obtained from patients and experimental animals. The monograph is written for (1) the gen- eral practitioner and pediatrician, whose experience with febrile convulsions is frequent and chiefly clinical and therapeutic; (2) the pediatric neurologist, whose interest lies in clinical and laboratory research; (3) the consultant neurologist, whose atten- tion is directed to the uncommon patient with complications; (4) the basic scientist, who is concerned with biochemical and pharmacologic aspects of the mechanism and development of potential new therapies; (5) the electroencephalographer, whose advice regarding the significance of transient or persistent abnor- malities is important in prognosis and therapy; (6) the specialist PREFACE in infectious diseases and epidemiology, whose investigations of viral causes of febrile illnesses in the young child would help to elucidate the causes of febrile convulsions; and (7) the parents of infants and children with febrile convulsions, who are en- trusted with the responsibility for prompt and adequate treatment and preventive measures, and whose apprehension regarding the immediate and long-term prognosis is frequently exaggerated by lack of knowledge and understanding. It is estimated that close to 3 per cent of the world population under 5 years of age and one-half million children in the United States are affected. The mechanism of febrile convulsions is in- completely understood, the management is unsatisfactory and controversial, and opinions concerning prognosis are diverse, varying with the diagnostic criteria employed and the degree of specialization of the physician. Reports of well-controlled clinical studies and original data are limited, and the publication of vi- gnettes, describing personal experiences, and apparently authori- tative conclusions without confirmatory evidence have confused our knowledge of the significance and optimum treatment of febrile convulsions. The discussion and recommendations are based on the authors opinions and on statistics and data culled from original articles published in various journals from most parts of the world. The relative lack of reports of original clinical studies from medical centers in England, Canada, South America, Soviet Russia, and Japan is remarkable, when compared to the frequency of publica- tions in the United States, India, Germany, France, Italy, and Denmark. The interests of a small number of dedicated investi- gators, including Peterman, Bridge, W. G. Lennox, M. A. Lennox, and Livingston, explain the apparent high incidence of febrile convulsions in the United States. The bibliography includes 270 references to both original and review articles published in English and foreign languages, and is meant to be complete; if significant publications have been omitted inadvertently or the data and conclusions misconstrued, the author accepts full responsibility and offers his apologies to those investigators involved. My own investigations have in- cluded clinical, electroencephalographic, and therapeutic aspects PREFACE of febrile convulsions, in which I was ably assisted by Drs, L. M. Aledort, J. A. Madsen, and B. I. Kramer; and basic pharmaco- logical studies in laboratory animals, performed with the faithful and expert technical help of Michael R. Zales, now Dr. Zales, and Patricio Hernandez and Lawrence A. Halpern, now pharmacolo- gists, all of whom were at one time associated with the Depart- ments of Pediatrics and Pharmacology, Albert Einstein College of Medicine. I am indebted to Dr. Henry L. Barnett, chairman of the Department of Pediatrics, and to Dr. Alfred Gilman, chair- man of the Department of Pharmacology, of the Albert Einstein College of Medicine, for providing the exceptionally adequate laboratory and clinical facilities and the opportunity to conduct this research while I was a member of the staff as an assistant and, later, an associate professor in their departments. My re- search was initially made possible by the generous support of Burroughs, Wellcome & Co. (U.S.A.), Inc., Tuckahoe, New York, and it is a pleasure to acknowledge the interest and assistance obtained from Mr. W. Creasy, president of that company, and his colleagues, Dr. W. Colvin and the late Dr. E. J. deßeer. Sub- sequently, grants (B-1627, NB 05161-01) were obtained from the National Institute of Neurological Diseases and Blindness, United States Public Health Service, Bethesda, Maryland, for continuation of this research; more recently, the Epilepsy Foun- dation, Washington, D.C., has contributed to the support of these investigations. My interest in febrile convulsions stemmed from a period of study with the late Dr. William Gordon Lennox in Boston, supported by a British Medical Research Council Travel- ing Fellowship; and the pharmacologic and biochemical investi- gations were stimulated by previous training and experience with Dr. Louis S. Goodman, chairman of the Department of Pharma- cology, University of Utah, and his colleagues Dr. Dixon M. Woodbury and Dr. Horace W. Davenport, the latter now chair- man of the Department of Physiology, University of Michigan. The progress of the research was facilitated by the interest and encouragement of Dr. Alfred Gilman in New York City; my understanding of the neurology and neuropathology of seizures was enlightened and broadened by Drs. Philip R. Dodge and Ray- mond D. Adams in Boston; and the fruition of this monograph PREFACE has resulted from the opportunities afforded in my present posi- tion by Dr. Robert B. Lawson, chief of staff, and the Board of Directors of Children’s Memorial Hospital, and the support of the Brain Research Foundation, in Chicago. Several members of my staff and other friends have contributed in some way to the completion of this work. In particular, it is a pleasure to men- tion the following; Mrs. Lynda deVoursney, whose assistance with the bibliography and assemblage of articles for review was invaluable; Dr. Neville R. Belton and Mr. Gary Pitchford, who provided numerous statistical analyses of data; and Drs. Mari- anne Larsen, Eda Hackl, and Cesar Porres, and Rosemary Whittle, for their translations of publications from Danish, Swedish, German, Polish, Czechoslovakian, Italian, Belgian, Por- tuguese, and Spanish. The devotion and tireless efforts of my secretary, Mrs. Connie Sylak, deserve special acknowledgment. Finally, the completion of this monograph has depended on the cooperation and helpful advice of Miss Joan Carolyn Zulch, editor, of The Macmillan Company, and the gentle but persua- sive encouragement and assistance of my wife, Mary. The subject of febrile convulsions involves many scientific dis- ciplines, including pediatrics, neurology, epidemiology and pub- lic health, pharmacology, and biochemistry. The frequency of the problem during infancy and childhood and the potential for de- velopment of recurrent seizures in adult life should demand the attention of investigators at both clinical and basic levels. The persistence of the stigma and social ostracism associated with the terms “epilepsy,” “seizures,” and “fits” is related inversely to the advancement of medical knowledge and education of the public. This monograph is presented as a comprehensive and critical analysis of our understanding of the significance, etiology, prog- nosis, and management of febrile convulsions. In addition to the factual information provided, the book is intended as a guide to future research projects and as a stimulus to the better delinea- tion and consequent acceptance oLthe problem of epilepsy and related disorders in our society. March 7, 1967 Chicago J. G. M. CONTENTS CHAPTER 1 Definition and Statistics 1 DEFINITION AND DIAGNOSTIC CRITERIA 2 INCIDENCE 5 SEX INCIDENCE 12 GEOGRAPHIC DISTRIBUTION 12 RACIAL FACTORS 15 SUMMARY 16 SUMMARY 16 CHAPTER 2 CHAPTER 2 Clinical Evaluation and Manifestations 17 NEUROLOGIC EVALUATION 17 AGE AT FIRST FEBRILE CONVULSION 23 ASSOCIATED EXTRACRANIAL INFECTIONS 25 PATTERNS AND DURATION OF SEIZURES 28 ASSOCIATED EPISODIC SYMPTOMS 32 BIRTH HISTORY AND NEUROLOGIC ABNORMALITIES 32 CONTENTS PSYCHOMOTOR DEVELOPMENT 34 SUMMARY 36 SUMMARY 36 CHAPTER 3 Electroencephalography and Other Laboratory Tests 37 THE ELECTROENCEPHALOGRAM 37 CEREBROSPINAL FLUID CONSTITUENTS 53 BLOOD CHEMISTRY ANALYSES 58 CHAPTER 4 CHAPTER 4 Etiologic Factors, Mechanism, and Seizure Threshold 59 FEVER AND HEIGHT OF BODY TEMPERATURE 60 SIGNIFICANCE OF INFECTIONS 68 AGE AND CEREBRAL MATURATION 74 INFLUENCE OF SEX AND MALE PREPONDERANCE 77 GENETIC FACTOR 77 ACQUIRED BRAIN LESIONS 82 MISCELLANEOUS FACTORS 83 SUMMARY 86 SUMMARY 86 CHAPTER 5 Prognosis and Sequelae 89 RECURRENCE OF FEBRILE SEIZURES 90 OCCURRENCE OF SPONTANEOUS SEIZURES 93 FACTORS PREDICTIVE OF SPONTANEOUS SEIZURE OCCURRENCE 101 INCIDENCE AND SIGNIFICANCE OF HEMIPARESIS 106 SUMMARY 110 SUMMARY 110 CHAPTER 6 Treatment of Febrile Convulsions 111 TREATMENT OF THE INITIAL CONVULSION 111 PROPHYLACTIC TREATMENT AND LONG-TERM MANAGEMENT 115 CONTENTS REVIEW AND ANALYSIS OF RECOMMENDED THERAPIES 132 SUMMARY 135 CHAPTER 7 Experimental Febrile Convulsions, Artificial Fever, and Hyperpyrexia 137 STUDIES IN ANIMALS 137 CLINICAL INVESTIGATIONS 172 HYPERPYREXIA, CONVULSIONS, AND BRAIN PATHOLOGY 180 Bibliography 185 SUMMARY AND COMMENT 183 Index 205 CHAPTER 1 DEFINITION AND STATISTICS The febrile convulsion is the most common neurologic disorder of infants and young children. It is estimated that close to 3 per cent of the population under 5 years of age and one-half million children in the United States are affected. “Fever cramps,” “stomach spasms,” and “teething convulsions” are lay terms used in referring to the febrile convulsion; although dramatic and disturbing to parents at the time of their occurrence, the episodes are often dismissed, once the child has recovered, as a reaction to fever of little consequence, comparable to the chill or rigor in a febrile adult. The significance of febrile convulsions in infants and young children is controversial. Some investigators have considered them benign and the prognosis sanguine, whereas others have em- phasized the gravity of febrile seizures and have regarded them as evidence of an organic cerebral lesion. The reported incidence of spontaneous and recurrent nonfebrile seizures in these patients has varied widely and the relation of febrile convulsions to the epilepsies is disputed, some authors regarding them as a specific 1 2 FEBRILE CONVULSIONS disorder distinct from the epilepsies, others using the terms “mild” or “pure” form of epilepsy. The diversity of opinions among authorities may be explained principally by differences in the degree of specialization of the investigator and clinic, and the definitions and criteria employed in selection of cases. DEFINITION AND DIAGNOSTIC CRITERIA The term “febrile convulsion” is used to describe a convulsion that occurs with fever and infections other than those primarily involving the brain. Apart from the exclusion of patients with meningitis and encephalitis, a seizure preceded by a significant degree of fever was the only criterion for the diagnosis of a febrile convulsion in the author’s group of cases. The classifica- tion of febrile convulsions into subgroups on the basis of the severity of the seizure, age at onset, electroencephalographic ab- normalities, and family history is of practical value in the assess- ment of prognosis and in terms of therapy. In our understanding of the mechanisms of febrile seizures, however, this arbitrary division of cases is confusing. Gelegenheitskrampfe or “casual convulsions” was the term coined by Hochsinger (1905) for convulsions that herald acute infection in infancy and childhood. Husler (1921) regarded the Gelegenheitskrampfe as an unusual reaction to infection in some children with a constitutional predisposition, comparable to shivering with rigors or chills in adults, and unrelated to “genu- ine” epilepsy. Husler compared the incidence of convulsions at the onset of infectious fevers in nonepileptic children with that in children with known epilepsy. Among 1,026 patients with scar- let fever and 1,908 with diphtheria, the illness was heralded by convulsions in seven and 33, respectively, whereas children with epilepsy in this study had no convulsions at the time of these infectious fevers. However, only one child with Gelegenheits- krampfe was followed to puberty and the subsequent history of the remaining 39 patients was not mentioned (Faerber, 1929). Faerber (1929) was one of the first authors to use the term “febrile convulsion.” He questioned the evidence and basis for a DEFINITION AND STATISTICS 3 distinction between “casual” or febrile convulsions and epilepsy, since few authors had studied their patients for a sufficiently long period after the initial convulsion with fever. Among 10 children with febrile convulsions in Faerber’s study, two de- veloped petit mal and grand mal and seven had a family history of epilepsy. It followed that the febrile convulsion could not be regarded as an entity distinct from epilepsy, although Faerber was prepared to acknowledge that a group, and perhaps the greater proportion, of children with febrile convulsions would develop normally without other manifestations of “genuine” epi- lepsy. He concluded that the significance of the initial convulsion with fever in the individual child could be assessed only after long-term observation. As an indication of uncertainty in his own definition of the febrile convulsion, Faerber continued to use the term Gelegenheitskrampfe for those seizures uncomplicated by “epileptic” convulsions. Several investigators have classified the febrile convulsion as a form of epilepsy, although data have been obtained mainly from retrospective studies of selected patients examined in epilepsy and neurology clinics. Lederer (1934) reported 30 patients with epilepsy whose first convulsion had occurred during a febrile illness in infancy or childhood. Bridge (1949) stated “there is no good reason for considering febrile convulsions as a clinical entity distinct from epilepsy. In reality, both belong in a single group, best described with the name of convulsive disorders. The differences are not of a fundamental nature but only of type and degree.” Peterman (1950) defined a febrile convulsion as “a major sei- zure precipitated by a nonspecific fever of variable degree in a person with a potential convulsive disorder.” The frequent asso- ciation with an organic cerebral lesion in this series of cases led Peterman to the conclusion that the febrile convulsion was of grave prognostic significance. W. G. Lennox (1953), Bjerglund and Brandt (1954), and Breg and Yannet (1962) used the terms “mild” or a “pure” form of epilepsy in describing the nature of the' febrile convulsion, and Lennox and Lennox (1960) referred to “hyperthermic-precipitation epilepsy.” Breg and Yannet re- garded the febrile convulsion as “an activated seizure state which 4 FEBRILE CONVULSIONS differs from our ordinary concept of epilepsy only in the fact that in the former, the trigger mechanism is recognized and in the latter, the trigger mechanism is as yet unknown.” Livingston’s opinion (1954) differed from that of his colleague Bridge (1949) and reverted to the older concept of a specific entity distinct from the epilepsies. He introduced the term “simple” febrile convulsion and restricted the definition to patients under 6 years of age with short, generalized seizures and a normal electroencephalogram. Febrile seizures that were prolonged, focal in type, or associated with electroencephalographic abnormalities were classified as “epileptic seizures” in which fever acted as a precipitating factor. A similar definition and selection of patients were favored by Friderichsen and Melchior (1954) and by Cary (1956), who omitted from their studies all patients with a history of birth trauma, cerebral dysgenesis, and other factors of known etiologic importance in epilepsy. Prichard and McGreal (1958) avoided the controversial word “epilepsy” and divided patients with febrile convulsions into “simple” and “atypical” groups. Simple febrile convulsions were described as benign, and the subsequent occurrence of spontaneous afebrile seizures in this group was less than 2 per cent; a specific disease entity was implied by reference to a genetic factor but was unsupported by data. The criteria emphasized for a diagnosis of,atypical febrile convulsions were mainly quantitative and related to the duration of the seizure, the degree of pyrexia, and the frequency of seizure recurrence; a higher incidence of spontaneous seizures and the occasional occurrence of permanent neurologic lesions could be expected in this atypical group. The most restrictive diagnostic criteria were those suggested by Cavazzuti and Lavagna (1961), who eliminated from the definition of “simple” febrile convulsions all patients who developed afebrile seizures at any time. The controversy concerning the correct classification of the febrile seizure stems largely from the varying degrees of selection of patient material studied by different investigators, undue em- phasis on the rare report of an unusually high genetic predispo- sition to febrile seizures, and the continuing reluctance of some authors to abandon the traditional concept of the term epilepsy as a disease per se, distinct from convulsive disorders sympto- DEFINITION AND STATISTICS 5 matic of known etiologic factors. Until more convincing evidence is provided in support of a “genuine or true epilepsy” and a “sim- ple febrile convulsion” as disease entities with distinct genetic or acquired factors in etiology, it is more appropriate to regard all epilepsies as manifestations of a lowered threshold to seizures, caused by inherited and/or acquired disorders of brain function of known and unknown etiologies, and to classify the febrile convulsion from a scientific standpoint as a form of epilepsy. An etiologic classification of seizures or the epilepsies, which includes the febrile convulsion, is shown in Table 1-1. The thresh- old to febrile seizures may be lowered by functional and re- versible disorders affecting the brain primarily or secondarily to systemic disease, or by acquired and structural lesions of the brain. A diagnosis of epilepsy should be determined by the facility and frequency with which the seizure mechanism is triggered. The present-day misunderstanding and stigma associated with the term epilepsy precludes its use in infants and children with an occasional febrile convulsion uncomplicated by recurrent spon- taneous seizures. From the lay and legal standpoints, seizures occurring only in response to known, temporary precipitating factors such as fever and intoxications should not be labeled epileptic, since in most cases the prognosis is good and the incidence of spontaneous seizures is low. A diagnosis of epilepsy in a child with febrile convulsions should be deferred and applied only if nonfebrile seizures develop and become recurrent. The prognosis of the individual case of febrile convulsions may be determined by various factors of which the more reliable are the duration of the seizure and the electroencephalographic find- ings. The arbitrary grouping of febrile convulsions as “simple” or “atypical” is justifiable in terms of prognosis, but to ignore the relation between febrile and nonfebrile convulsions and to regard the so-called simple febrile convulsion as a distinct disease entity is unwarranted on the basis of present evidence. INCIDENCE The statistics of febrile convulsions were derived from approxi- mately 10,000 cases reported from various countries in 51 publi- Table 1—1. Etiologic Classification of Seizures* A. WITHOUT STRUCTURAL CEREBRAL LESIONS 1. Cryptogenic 2. Symptomatic (a) Fever and extracranial infection (“febrile convulsion”) Acute tonsillitis, otitis media, pneumonia (b) Intracranial infection Meningitis, encephalitis (c) Electrolyte imbalance Water intoxication, hyponatremic and hypernatremic states (d) Metabolic disorder (e) Toxic encephalopathy Hypocalcemia, hypoglycemia, pyridoxine deficiency, phenylpyruvic oligo- phrenia, porphyria i. Heavy metals Lead, thallium, arsenic ii. Drugs Camphor, thujone, pentylenetetrazol, strychnine, aminophylline, phenothia- zines, antihistamines (f) Hypertensive encephalopathy Acute glomerulonephritis ( g) Acute anoxia Anesthesia, congenital heart disease (h) Reflex sensory stimuli Light, touch, music, noise, odors (i) Miscellaneous Emotional disturbance, allergy 6 Table 1-1. Etiologic Classification of Seizures (Continued)* B. . WITH STRUCTURAL CEREBRAL LESIONS 1. Posttraumatic Laceration, contusion 2. Posthemorrhagic Injury, hemorrhagic disease, ruptured aneurysm 3. Postanoxic Asphyxia neonatorum 4. Postischemic Thrombosis, embolism, vasospasm 5. Postinfective Meningitis, encephalitis, encephalopathy, immunization, cerebral abscess, congenital syphilis 6. Posttoxic Kernicterus, lead and thallium poisoning 7. Congenital Aplasia, porencephaly, hydrocephalus, Sturge-Weber disease, arteriovenous malformation, tuberous sclerosis 8. Degenerative Lipidoses, leukoencephalopathies 9. Parasitic Toxoplasmosis, cysticercosis 10. Neoplastic Tumor * From Millichap, J. G.: Pediat Clin N Amer 7: 583 (Aug.), 1960 (iii); and Physiological Pharmacology, Vol. II, Aca- demic Press, New York, 1965, pp. 97-173. 7 8 FEBRILE CONVULSIONS cations between 1924 and 1965 (Table 1-2). An early reference to the incidence of febrile convulsions was that of Patrick and Levy (1924) in Chicago. These authors found that patients with febrile convulsions accounted for 42 per cent of children less than 6 years of age with seizure disorders. Thirty authors have estimated the incidence in relation to other types of seizures and 13 reports include the percentage of patients with febrile con- vulsions among all children attending hospitals and clinics. A total of 5,679 case reports of febrile convulsions represented 30 per cent of 19,334 children examined because of various seizure disorders, and 3,740 case reports amounted to 2 per cent of a total census of 188,690 children attending hospitals or clinics. The frequency of a diagnosis of febrile convulsions varied from a low of 1 per cent (Pascual and McGovern, 1952) to as high as 9 per cent (Shanks, 1949); and an incidence of 11 per cent (Friderichsen and Melchior, 1954) was recorded among a group of children with febrile illnesses. The variations in incidence observed in these studies might be explained by differences in the maximum age of the patients examined and the type of hos- pital or clinic where the investigation was performed, but a close correlation with these variables was not established. In two rela- tively early reports from Boston by different authors, the inci- dence in infants less than 2 years of age was 5 per cent (Brown, 1935) and in children less than 5 years old it was 2 per cent (Thom, 1942). However, the most extreme variations in inci- dence occurred in two reports which concerned children of all ages (Shanks, 1949; Pascual and McGovern, 1952). That the inci- dence of the febrile convulsion has not changed significantly in the past 25 years is evident from the figure of 3 per cent reported by Costeff in 1965. Apparently, the advent and general use of antibiotics, particularly penicillin, has had no demonstrable effect on the problem of febrile convulsions. The incidence of febrile convulsions in children with seizures ranged from 4 per cent to 61 per cent; the low figure was reported in a group of 128 children of all ages (Peterman, 1950), and the high incidence occurred in 634 patients examined at an age period between 1% and 3 years, when susceptibility to febrile convul- sions is enhanced (Bamberger and Matthes, 1959). Febrile con- Table 1-2, Statistics of Febrile Convulsions in i Children AUTHORS YEAR GEOGRAPHIC LOCATION NO. PATIENTS PER CENT OF CHILDREN IN HOSPITAL OR CLINIC AND INCIDENCE PER CENT OF CHILDREN WITH SEIZURES MALE SEX FEMALE Patrick and Levy 1924 Chicago, U.S. 13 42 ( < 6 years) * Lederer, E. 1934 Budapest, Hungary 30 10 Peterman 1934 Milwaukee, U.S. 100 23 Brown 1935 Boston, U.S. 120 5.3 (< 2 years) 30 ( < 2 years) Bergemann 1936 Leipzig, Germany 20 13 Herlitz 1941 Stockholm, Sweden 776 432 344 Coelho 1942 Bombay, India 81 38 Thom 1942 Boston, U.S. 44 2.2 (< 5 years) 34 Peterman 1946 Milwaukee, U.S. 834 1.9 33 Lennox, M. A. 1947 Boston and 153 41 101 52 New Haven, U.S. 52 24 28 Livingston, Bridge, Kajdi 1947 Baltimore, U.S. 94 12 Neyroud 1947 Geneva, Switzerland 70 39 31 Chaptal et al. 1948 Montpellier, France 21 5 Kaplan et al. 1948 Paris, France 15 9 6 10 4 6 Bridge 1949 Baltimore, U.S. 350 2.4 35 Gallant and Livingston 1949 Baltimore, U.S. 18 12 6 * Age group studied. 9 Table 1- -2. Statistics of Febrile Convulsions in Children (Continued) AUTHORS YEAR GEOGRAPHIC LOCATION NO. PATIENTS PER CENT OF CHILDREN IN HOSPITAL OR CLINIC AND INCIDENCE PER CENT OF CHILDREN WITH SEIZURES MALE SEX FEMALE Lennox, M. A. 1949 Boston and 240 149 91 New Haven, U.S. 77 36 41 Shanks 1949 Glasgow, Scotland 487 8.5 Ekholm, Niemineva 1950 Helsinki, Finland 91 41 (< 3 years) 57 34 Peterman 1950 Milwaukee, U.S. 128 4.3 72 56 Kao 1951 Nanking, China 515 46 Pine et al. 1951 Durham, N.C., U.S. 21 12 ( < 5 years) Calzetti and Borghi 1952 Bologna, Italy 100 2 50 63 37 Pascual and McGovern 1952 Washington, D.C., U.S. 455 0.9 252 203 Giraud et al. 1953 France 73 28 Lennox, W. G. 1953 Boston, U.S. 407 26 Tibrewalla 1953 Bombay, India 149 52 Bjerglund, Brandt 1954 Copenhagen, Denmark 129 77 52 Chandy et al. 1954 Vellore, India 18 30 Friderichsen, Melchior 1954 Copenhagen, Denmark 405 4.2 (< 7 years) 224 181 11 (of febrile patients) Pardelli, Ardito 1954 Pisa, Italy 10 6 4 Cary 1956 Sydney, Australia 100 1.3 54 46 Moller 1956 Gothenburg, Sweden 448 1.5 10 Table I- -2. Statistics of Febrile Convulsions in Children (Continued) AUTHORS year GEOGRAPHIC LOCATION NO. PATIENTS PER CENT OF CHILDREN IN HOSPITAL OR CLINIC AND INCIDENCE PER CENT OF CHILDREN WITH SEIZURES MALE SEX FEMALE Coelho 1957 Bombay, India 120 84 36 Hrbek 1957 Prague, Czecho- Slovakia 274 2.3 45 150 124 Phadke 1957 Poona, India 28 12 Broberger 1958 Stockholm, Sweden 30 17 13 Laplane et al. 1958 Paris, France 116 43 Lerique-Koechlin et al. 1958 Paris, France 1005 45 Radermecker 1958 Antwerp, Belgium 50 29 21 Bamberger, Matthes 1959 Basel, Switzerland 634 61 (li—3 years) 367 268 Turinese 1959 Padua, Italy 167 34 (< 8 years) 102 65 Vyas 1959 Bhavnagar, India 41 9.2 Millichap et al. 1960 New York, U.S. 110 68 42 Cavazzuti, Trovarelli 1961 Modena, Italy 230 148 82 MacDougall 1962 Nairobi, Kenya 30 23 7 Gandhi 1963 Jamnazar, India 37 12 Horstmann, Schinnerling 1963 Freiburg, Germany 108 1-2 60 48 Frantzen et al. 1964 Gentofte, Denmark 220 121 99 Lennox-Buchthal 1964 Copenhagen, Denmark 100 62 38 Costeff 1965 Beer-Sheva, Israel 15 3 24 Totals 9969 2% of 188,690 30% of 19,334 2842 2061 11 12 FEBRILE CONVULSIONS vulsions occur most commonly between 6 months and 3 years of age, and the highest incidence would be expected in reports concerning children within this age range. W. G. Lennox (1953) estimated that 5 per cent of all children under 5 years of age have one or more seizures and one half of these have convulsions with fever. The inconstancy of this correlation indicates that factors in addition to age must contribute to the number of patients with febrile convulsions reported by various authorities. A review of the data presented in Table 1-2 shows that the febrile convulsion accounts for 30 per cent of all childhood sei- zure disorders and 2 per cent of all childhood illnesses. The population statistic of the United States in 1960 for children under 5 years of age was approximately 20 million, and the number with febrile convulsions in this age period may be esti- mated as one-half million. SEX INCIDENCE In each of 29 series of cases reported in the literature, boys were affected more commonly than girls, and in a total of 4,903 patients with febrile convulsions the sex ratio was 1.4 to 1 (Table 1-2). Studies by M. A. Lennox (1947) conducted in Boston and New Haven showed that boys were affected twice as frequently as girls, but in a more recent report by the same author (Lennox- Buchthal, 1964) from Copenhagen, Denmark, the ratio of boys to girls was 1,6 to 1. Our own study (Millichap, Madsen, and Aledort, 1960 iii) included 68 boys and 42 girls, approximately the same ratio as that observed among the total number of patients reported in the literature. The possible significance of this predilection for males is discussed in relation to the mech- anism of the febrile convulsion (Chapter 4). GEOGRAPHIC DISTRIBUTION The statistics of febrile convulsions according to the geographic distribution shown in Figure 1-1 are based on approximately 12,000 case reports culled from 71 publications of studies in 19 Figure 1-1. Geographic distribution of 11,777 patients with febrile convulsions from repons published in 19 countries during the period 1924-1965. 13 14 FEBRILE CONVULSIONS different countries. The numbers of publications from each coun- try are listed in parentheses in order of decreasing frequency as follows: United States (21), India (8), Germany (7), France (6), Italy (5), Denmark (4), Sweden (4), Switzerland (3), Israel (2), Belgium (1), Australia (1), China (1), Czechoslovakia (1), Finland (1), Hungary (1), Kenya (1), Mexico (1), Poland (1), and Scotland (1). The largest number of patients, 3,532, was reported from the United States. Studies in France and Sweden included 1,450 and 1,645 case reports, respectively; in Denmark and Switzerland they numbered 854 and 809; in Germany, 785; Italy, 533; China, 515; India, 491; Scotland, 487; Czechoslovakia, 274; Australia, 100; Finland, 91; Israel, 51; Belgium, 50; Poland, 35; Hungary, 30; Kenya, 30; and Mexico, 15. A variety of factors, including population statistics, incidence of certain types of infection, and genetic influences might explain the relatively large number of reports from the United States, but the most likely reason for the apparent high incidence of febrile convulsions in this country is the dedication of a small number of interested pediatricians, neurologists, and other investigators. Peterman (1946), Bridge (1949), M. A. Lennox (1947, 1949), W. G. Lennox (1953), and Livingston (1954) reported the ma- jority of the cases, and these authors specialized in epilepsy at centers in Baltimore, Boston, Milwaukee, and New Haven, where large numbers of children with seizure disorders were investi- gated and treated. Single reports of 1,005 patients from a center in Paris (Lerique-Koechlin et ah, 1958) and 776 patients from Stockholm (Herlitz, 1941) account for the relatively high inci- dence in France and Sweden. One report of 634 patients from Basel (Bamberger and Matthes, 1959) includes more than 75 per cent of the cases from Switzerland, and a report of 405 patients from Copenhagen (Friderichsen and Melchior, 1954) represents approximately 50 per cent of those from Denmark. An unusual and perhaps epidemic proportion of convulsions caused by gastroenteritis was apparent in reports from Sweden (Faxen, 1935), Scotland (Shanks, 1949), Germany (Tille, 1950), China (Kao, 1951), Alabama (Donald et ah, 1956), and Israel (Fischler, 1962); but the increased incidence of certain infections DEFINITION AND STATISTICS 15 was an infrequent explanation for the occurrence and publication of febrile convulsions in the majority of reports (Table 2-1) The relative lack of reports of original clinical data concerning febrile convulsions in England, Canada, South America, Soviet Russia, and Japan is remarkable and unexplained, Ounsted (1952) from Oxford, England, and Pupo (1962) in South America have made important contributions to our understanding of the influ- ence of genetic or acquired factors in the etiology of both febrile and afebrile seizures, but detailed statistics regarding the patients with febrile convulsions were not included in these special studies. That the problem of the child with convulsions and fever is acknowledged in Canada and Japan is evident from clinical re- views published by Prichard and McGreal (1958) and Prichard (1961), and from reports concerning the mechanism of febrile seizures in experimental animals investigated by Ikeda (1962) and Kashiwase (1962). Studies of the epidemiology of febrile convulsions in these and other countries might elucidate further the cause and facilitate the prevention of this common childhood illness. RACIAL. FACTORS The wide geographic distribution of case reports shown in Figure 1-1 indicates that febrile convulsions may affect children of all races, but studies regarding the possible importance of racial factors in etiology are limited. In our own investigation of 110 patients seen in New York City (Millichap, Madsen, and Aledort, 1960), the incidence in white, Negro, and Spanish-American races was 53, 26, and 21 per cent, respectively. A similar racial distribution was observed in the general pediatric population at the hospital center where this study was performed, and the propensity to febrile seizures was not significantly greater in white patients compared with those of other races. The paucity of reports from some well-populated areas of the world might indicate a peculiar resistance to febrile convulsions among certain races, but a relative lack of concern of physicians and a denial of the problem by the laity in these countries are alternative explanations. The persistence of the 16 FEBRILE CONVULSIONS stigma and social ostracism associated with the terms epilepsy, seizures, and fits is related inversely to the advancement of med- ical knowledge and education of the public, and greater atten- tion to the need for original scientific research is an essential precursor to a better delineation and acceptance of the problem of epilepsy and related disorders in our society. SUMMARY The febrile convulsion is defined as a convulsion with fever and infection not primarily involving the brain. It is the com- monest neurologic disorder of childhood and affects an estimated one-half million children in the United States. Two per cent of all children attending hospitals and clinics and 30 per cent of children with seizures have febrile convulsions. Boys are affected more frequently than girls, and the disorder is reported in all races and from 19 countries. Original clinical studies in England, Canada, South America, Soviet Russia, and Japan are limited or unreported. CHAPTER 2 CLINICAL EVALUATION AND MANIFESTATIONS The need for a complete neurologic evaluation in a child with a febrile convulsive disorder is determined by the frequency and severity of the convulsions and the occurrence of other symptoms of neurologic dysfunction. An infant or young child who recovers rapidly after a single, uncomplicated febrile convulsion of short duration may not require a detailed neurologic examination, but the febrile seizure disorder that is recurrent, prolonged, or asso- ciated with one or more spontaneous convulsions demands special consideration. neurologic evaluation The examination is best performed by a neurologist trained and experienced in diseases of infants and young children. Several of the common febrile illnesses may present with convulsions in the age period between 6 months and 5 years, and the diagnosis, antibiotic therapy, and management of associated electrolyte 17 18 FEBRILE CONVULSIONS disorders require the skills of both pediatrician and neurologist. The evaluation should include the following: Present, Past, and Family History (a) Description of the pattern, severity, and duration of the convulsion, height of the body temperature immediately before or soon after the onset of the convulsion, frequency of febrile convulsions, and occurrence of nonfebrile spontaneous seizures. (b) History of pregnancy, birth, and early development, in- cluding evidence of cerebral anoxia or head injury. (c) Family history of seizures or related neurologic disorders. Neurologic and General Physical Examinations The neurologic examination of the infant and child is described fully in textbooks (Dodge, P, R., and Farmer, T. W., 1964; Dejong, R. N., 1967), a symposium (Paine, R. S., and Perlstein, M. A., 1960), and a monograph (Paine, R. S., and Oppe, T. E., 1966). Examination During the Acute Episode. The differential di- agnosis at the time of the acute seizure with fever will include bacterial meningitis, encephalitis, toxic encephalopathy, subdural effusion and hematoma, and cerebral abscess. Signs of increased intracranial pressure, meningeal irritation, and focal cerebral pathology should be investigated further by examination of the cerebrospinal fluid and subdural puncture; a neurosurgical opin- ion and radiographic contrast studies are advised when a space- occupying intracranial lesion is suspected. The indications for neurosurgical diagnostic procedures in infants and young children differ from those in adult patients, and the focal seizure pattern alone is not a reliable criterion of a localized or expanding lesion. Provided that bacterial infection is excluded or treated ade- quately, a conservative approach is preferred and repeated neu- rologic examinations may be needed in some cases before a definite diagnosis is established. In children with febrile seizures that respond rapidly to anti- pyretic and anticonvulsant therapy and are associated with ob- vious signs of systemic infection, examination of the cerebrospinal fluid at the time of the acute episode is rarely indicated. When CLINICAL EVALUATION AND MANIFESTATIONS 19 consciousness is recovered, the patient is observed further for signs of postictal paresis which, if persistent, may negate a diag- nosis of febrile convulsion and indicate a focal encephalitis or cerebrovascular accident. Postictal electroencephalographic ab- normalities may persist for up to seven days following a febrile convulsion (Livingston, 1954), and the value of the record in assessment of diagnosis and prognosis is more accurate if de- ferred until recovery from these reversible and transient changes is complete. Examination after Recovery from Acute Episode. The extent of the evaluation will vary with the nature of complications. In addition to the clinical examination, an electroencephalogram, x-ray of skull, determinations of blood glucose and calcium, and urinalysis are advised in all patients with a recurrence of febrile seizures or one or more spontaneous seizures. The need for examination of the cerebrospinal fluid and radiographic contrast procedures will depend on the clinical evidence for central ner- vous system disease of a progressive type. Signs of psychomotor retardation may require a more complete and specialized study if behavior disorders are troublesome and placement in a normal school program seems inexpedient. A CASE HISTORY A 3-year-old boy was examined and treated by the Neu- rology Service at Children’s Memorial Hospital because of a convulsion during an acute febrile illness. The con- vulsion was generalized and clonic in pattern, the body temperature was 104°F, and an exudative pharyngitis was the apparent cause of the fever. The immediate administration of phenobarbital sodium, 120 mg (30 to 60 mg per year of age), by subcutaneous injection, sponging with tepid water, and rubbing the skin with alcohol controlled the convulsive movements and re- duced the body temperature. Neurologic examination in the postictal state revealed normal fundi, pupils of equal size which responded to light, symmetry of facial move- ments, withdrawal of all extremities following stimulation, generalized hypotonia and hyporeflexia, bilateral Babinski 20 FEBRILE CONVULSIONS responses, and absence of nuchal rigidity. The general physical examination was negative except for signs of pharyngitis, and a spinal puncture and examination of the cerebrospinal fluid were considered unnecessary. Clinical recovery from the effects of the convulsion was rapid, and cultures of pharyngeal and nasal secretions were negative for hemolytic streptococci and other pathogenic bacteria. Treatment included prophylactic phenobarbital, 15 mg orally at intervals of 6 hours (5 mg/kg daily), and acetyls alicylic acid, 300 mg orally at intervals of 8 hours (65 mg/kg daily), until the fever subsided. Antibiotics were withheld pending the reports of bacterial cultures because the fever responded promptly to antipyretic therapy and the clinical condition of the patient was improved. In the past history, the birth was prolonged and oxygen was administered during the first day. The neonatal period was otherwise uneventful and the milestones of develop- ment were reached at the normal age. A first febrile con- vulsion occurred when the infant was 5 months old, and approximately 10 recurrences were reported in 2% years, despite regular daily administration of phenobarbital and diphenylhydantoin (Dilantin) prescribed by a local physi- cian. The febrile illnesses had been diagnosed as infections of the upper respiratory tract, and penicillin had been given routinely because laboratory facilities for throat cultures were not readily available. Convulsions had com- plicated only those infections with high fevers, usually 39.4° C (103°F) and above, the pattern of the seizures was mainly generalized and clonic but sometimes asymmetrical, and the duration of the convulsive movements was short, rarely greater than 5 or 10 minutes. Recovery from each convulsion had been prompt, but brief episodes of staring and inactivity in recent months were suggestive of absence seizures. A history of seizures in the family was not elicited. Further investigations included a normal x-ray of the skull, a serum calcium of 5.1 mEq/L, fasting blood sugar of 70 mg/100 ml, and normal urinalysis. An electroenceph- CLINICAL EVALUATION AND MANIFESTATIONS 21 alogram performed one day following the convulsion and before the fever had subsided completely was abnormal, showing diffuse slowing and runs of high-voltage, irregular, and asymmetrical slow waves with rudimentary spikes. The electroencephalogram was repeated on the following day when the child had recovered; apart from an asymmetry of amplitude in the occipital leads of questionable sig- nificance the record was normal (Fig. 2-1). The subsequent long-term management included pheno- barbital in average anticonvulsant doses of 15 mg twice daily and the gradual withdrawal of diphenylhydantoin therapy. Regular daily treatment with phenobarbital was advised because febrile seizures had recurred frequently and the probability of spontaneous seizures was increased (see Chapter 5). Diphenylhydantoin is generally ineffec- tive or contraindicated for febrile and absence seizures (Millichap, 1964), and its continued administration was unwarranted. The parent was instructed to increase the regular dose of phenobarbital to 30 mg at the earliest signs of a febrile illness and to continue treatment with 15 mg doses at intervals of 6 hours until the fever subsided. Sponging with tepid water was recommended and acetyl- salicylic acid was permitted in amounts not to exceed 900 mg each 24 hours. Neurologic reevaluation after a period of observation of not longer than 1 year was ad- vised. The prognosis was guarded because of the history of equivocal spontaneous seizures, frequent recurrence of febrile seizures, and possible anoxia at birth. Factors in the case history indicating a relatively good prognosis include the short duration of the febrile convulsions and the return of the electroencephalogram and neurologic examination to normal after recovery from the acute illness. The suscep- tibility to febrile seizures is expected to decrease sharply after 3 years of age, and the observed response to treat- ment and reevaluation in 1 year would allow a more ac- curate assessment of the eventual outcome. Figure 2-1. Electroencephalograms of a 3-year-old boy with a history of recurrent febrile convulsions. (A) An initial recording one day after a convulsion is abnormal, showing diffuse slowing and runs of irregular and asymmetrical high-voltage slow waves with rudimentary spikes. (B) A recording one day later shows improvement, with only an asymmetry of amplitude in the occipital leads. (A) One Day Postictal (j 3 yrs (B) Two Days Postictal 22 CLINICAL EVALUATION AND MANIFESTATIONS 23 age at first febrile CONVULSION In the author’s study of 110 patients (Millichap, Madsen, and Aledort, 1960), the first febrile seizure occurred most commonly between 6 months and 3 years of age. An onset before 6 months was observed in only two patients and after 5 years in six; the latest age of onset was 8 years (Fig. 2-2). The age of the patient was recorded in several publications and a total of 7,000 case reports. These data were tabulated and the numbers of patients with onset of seizures during each age period are shown in the histogram (Fig. 2-3). The age distribu- tion among this large group of patients is almost identical with that observed in the relatively small sample studied by the author. The highest incidence was between the first and second birth date, and this period accounted for one third of the patients. Approximately 60 per cent of patients with febrile seizures have had their first attack by the second birth date, approximately 80 per cent have been affected by the third birth date, and 95 per cent by the fifth birth date (Fig. 2-4). The predilection of the febrile seizure for the young may pos- No. Pafienfs Age in Years Figure 2-2. Age at first febrile convulsion in the authors series °f 110 patients. (From Millichap et al.: Neurology (Minneap) 10: 643,1960 in.) 24 FEBRILE CONVULSIONS Per Cent No. Patients Age at First Febrile Seizure Figure 2-3. Age at first febrile convulsion of 7000 patients re- ported in the literature. sibly be explained by the increased exposure and susceptibility to upper respiratory tract infection at this age. Furthermore, the relatively low incidence in the first 6 months of life might be due to the infrequency of tonsillitis and pharyngitis at this time. However, the almost complete absence of seizures of this type in children more than 10 years of age would militate against this theory. Laboratory experiments in animals (Chapters 4 and 7) suggest that the high threshold to seizures in the newborn is related to a low concentration of carbonic anhydrase and a reciprocal high level of carbon dioxide in the brain (Millichap, Balter, and Hernandez, 1958), whereas resistance to febrile seizures in older animals may be explained by changes in the balance of water and electrolytes with maturation and the possible development of enzyme systems that inhibit the seizure discharge. In addition, the increased susceptibility of young compared to adult animals has been correlated with developmental changes in the ratio of intra- cellular to extracellular cations in the brain (Millichap, 1960 i). CLINICAL EVALUATION AND MANIFESTATIONS 25 Per Cent No. Patients with History of Febrile Seizures Age in Years Figure 2-4. Incidence of a history of febrile convulsions in relation to age in 7000 patients. ASSOCIATED EXTRACRANIAL INFECTIONS In the authors study, 142 febrile episodes were recorded in HO patients. The causes of the fever and the frequency of their occurrence are shown in Table 2-1. In 71 per cent of the author’s series, the febrile seizure was associated with acute tonsillitis, pharyngitis, or otitis media. The prevalence of these infections in young children between the ages of 1 and 3 years may account for this high incidence. Among 6,790 patients with febrile seizures reported in 33 publications between 1929 and 1964, the causes of fever were noted in a total of 7,036 febrile episodes (Table 2-1). Tonsillitis or pharyngitis and otitis media were diagnosed in 4,371 Table 2-1. Causes of Fever in Children with Febrile Convulsions AUTHORS YEAR GEOGRAPHIC NO. LOCATION PATIENTS NO. FEBRILE EPISODES TONSILLITIS OR PHARYNGITIS OTITIS MEDIA BRONCHITIS OR PNEUMONIA GASTROENTERITIS SHIGELLOSIS SALMONELLOSIS EPIDEMIC FEVER VAR. MEASLES ROSEOLA INFANTUM PERTUSSIS VACCINATION IMMUNIZATIONS < s < < SEPTICEMIA OR SEPSIS PYELITIS MISCELLANEOUS | UNKNOWN Faerber 1929 Berlin, Germany 10 10 4 2 2 i 1 Brown 1935 Boston, U.S. 120 120 31 4 47 3 25 5 1 Faxen 1935 Gothembourg, Sweden 238 238 125 16 46 2 21 4 18 4 2 Herlitz 1941 Stockholm and Uppsala, Sweden 776 776 618 22 45 31 20 7 11 22 Neyroud 1947 Geneva, Switzerland 70 70 14 4 9 13 1 i 1 6 21 Kaplan et al. 1948 Paris, France 25 23 4 3 2 2 3 i 5 3 Shanks 1949 Glasgow, Scotland 487 487 310 104 16 46 11 Ekholm, Niemineva 1950 Helsinki, Finland 91 91 64 1 6 10 4 1 4 1 Tille 1950 Berlin, Germany 96 141 56 5 56 5 4 7 8 Kao 1951 Nanking, China 515 512 83 3 134 75 71 30 30 28 30 24 4 Calzetti, Borghi 1952 Bologna, Italy 100 90 49 16 7 4 24 Pascual, McGovern 1952 Washington, D.C. 455 557 407° 60 90 Friderichsen, Melchior 1954 Copenhagen, Denmark 405 465 290 53 46 44 12 20 Livingston 1954 Baltimore, U.S. 201 201 115 32 19 3 10 8 14 Pardelli, Ardito 1954 Pisa, Italy 10 10 5 1 4 Turnbull 1955 Mexico 15 15 . 10 3 1 1 Cary 1956 Sydney, Australia 100 100 40 13 15 5 10 5 12 * Includes some cases of otitis media and bronchitis. 26 Table 2-1. Causes of Fever in Children with Febrile Convulsions (Continued) • AUTHORS YEAR GEOGRAPHIC LOCATION NO. PATIENTS NO. FEBRILE EPISODES TONSILLITIS OR PHARYNGITIS OTITIS MEDIA BRONCHITIS OR PNEUMONIA GASTROENTERITIS SHIGELLOSIS SALMONELLOSIS EPIDEMIC FEVER VAR. MEASLES ROSEOLA INFANTUM PERTUSSIS VACCINATION IMMUNIZATIONS MALARIA SEPTICEMIA OR SEPSIS PYELITIS MISCELLANEOUS UNKNOWN Donald et al. 1956 Birmingham, U.S. 41 41 12 29 Moller 1956 Gothenburg, Sweden 448 448 358 8 16 7 34 25 Broberger 1958 Stockholm, Sweden 153 154 66 4 22 11 i 30 3 3 3 11 Laplane et al. 1958 Paris, France 39 39 34 1 i i i 1 Lerique-Koechlin et al. 1958 Paris, France 1005 1017 590 160 267 Radermecker 1958 Antwerp, Belgium 50 46 17 4 2 23 Bamberger, Matthes 1959 Basel, Switzerland 634 634 403 22 40 46 7 31 9 3 26 i 16 4 20 Vyas 1959 Bhavnagar, India 41 41 2 8 11 20 Millichap et al. 1960 New York 110 142 78 24 7 4 6 10 2 1 10 Fischler 1962 Zrifin, Israel 36 46 29 17 MacDougall 1962 Nairobi, Kenya 30 30 10 4 2 2 2 1 8 1 Mehta 1962 Jaipur, India 17 17 10 6 1 Dobrzynska 1963 Gdansk, Poland 35 35 28 4 2 1 Laplane, Salbreux 1963 Paris, France 210 210 185 25 Ehrengut, Ehrengut 1964 Hamburg, Germany 7 10 2 4 1 5 1 Frantzen et al. 1964 Gentofte, Denmark 220 220 173 15 8 18 6 Totals 7036 4162 209 481 509 119 177 88 98 60 49 7 49 55 123 403 440 Incidence (%) 59 2.9 6.9 7.2 1.7 2.5 1.2 1.4 0.9 0.7 0.1 0.7 0.8 1.7 5.7 6.2 27 28 FEBRILE CONVULSIONS or 62 per cent of these children, a slightly lower incidence than in the author’s small series. Gastroenteritis occurred with relatively greater frequency in eight of the published reports, and those from Belgium, Switzer- land, Scotland, Finland, Germany, China, Denmark, and Israel contributed particularly to the high incidence of 7.2 per cent. Shanks (1949) in Glasgow, Scotland, and Kao (1951) in Nanking, China, reported 487 and 512 patients with febrile episodes, and the frequency of occurrence of gastroenteritis in these two large series was 21 per cent and 28 per cent, respectively. One study by Tille (1950) of Berlin, Germany, included 96 patients with 141 febrile episodes with convulsions of which 56 (40 per cent) were caused by gastroenteritis. Apart from the study by Donald and co-workers (1956) in Birmingham, Alabama, in which all of 41 patients had gastroenteritis, shigellosis, or salmonellosis, febrile convulsions in the United States were associated very infrequently with gastroenteritis. A low incidence of this infection was noted also in studies conducted in France. Other causes of fever, listed in order of decreasing frequency, included bronchitis and pneumonia in 481 (6.8 per cent) and measles, roseola infantum, and pertussis in 423 (6 per cent) pa- tients. The inclusion of these epidemic fevers presupposes that the convulsions occurred as a result of the associated fever and not as a symptom of encephalitis or encephalopathy. A similar problem in classification of the seizures arises in regard to re- ports of 49 patients with malaria, the same number following smallpox vaccination, and 55 patients with septicemia or other sepsis. In 843 patients the causes of fever were unknown or were classed as miscellaneous. The significance of infection in the etiol- ogy of febrile seizures is discussed in Chapter 4. PATTERNS AND DURATION OF SEIZURES In the author’s study a clonic type of seizure pattern was re- ported in 84 per cent of 96 patients and was associated with a tonic component in 57 per cent. The pattern was principally tonic in only 7 per cent and was flaccid or akinetic in 5 per cent. It CLINICAL EVALUATION AND MANIFESTATIONS 29 was not documented in the remainder. A focal seizure pattern was reported in 15 per cent of cases, and the incidence in the total group was not significantly different from that observed in patients with a history of spontaneous as well as febrile seizures. The duration of the febrile seizure was usually less than 5 min- utes. Among 79 patients in whom this information could be eval- uated satisfactorily, the longest seizure was less than 5 minutes in 43 per cent, 5 to 10 minutes in 22 per cent, 10 to 20 minutes in 19 per cent, and more than 20 minutes in 16 per cent. The clinical patterns and grades of febrile seizures were re- corded in nine publications by seven authors between 1947 and 1964. The total number of patients was 764; a publication by M. A. Lennox in 1949 included some data previously reported in 1947. The number and percentage of patients with seizure pat- terns and grades recorded as generalized, focal, mainly clonic, mainly tonic, akinetic, or flaccid and mild or severe are shown in Table 2-2. The mild seizure was defined as generalized and brief, the severe seizure was focal or of more than 1 hour in duration. Eighty-six per cent of patients had generalized febrile seizures and 11 per cent had focal patterns. The convulsions were mainly clonic in type, and tonic and akinetic seizures were infrequent. An additional reference to the preponderance of clonic seizure patterns concerned 228 patients, but the exact incidence was not quoted (Lerique-Koechlin et al., 1958). M. A. Lennox (1949) and Fois and Malandrini (1957) divided their patients into two groups according to the presence or absence of abnormal neuro- logic signs or electroencephalographic disturbances. In one series of 229 patients (Lennox, 1949) whose first convulsion was at- tributed to fever, 53 were classed as epileptic. In this group, 33 (61 per cent) patients had severe convulsions, more than one half had grand mal seizures, and one third had symptoms indicating a focal onset of the convulsions. The severity of the convulsion was by far the most important criterion of a poor prognosis. Duration of febrile seizures was determined in seven studies between 1924 and 1964 (Table 2-3), The results in the total group of 1,229 patients are in agreement with the author’s find- ings in 96 patients. The duration of the convulsion was less than 5 minutes in 39 per cent, less than 20 minutes in 76 per cent, less Table 2-2. Clinical Patterns of Febrile Seizures AUTHORS year NO. PATIENTS SEIZURE PATTERNS AND GRADES GENER- ALIZED FOCAL MAINLY CLONIC MAINLY TONIC AKINETIC OR FLACCID MILD SEVERE Lennox, M. A. 1947 116 75 41 Lennox, M. A. 1949 176 106 70 Lennox, M. A. 1949 53 20 33 Fois, Malandrini 1957 82 56 26 Broberger 1958 30 27 3 Laplane et al. 1958 39 25 3 4 4 Millichap et al. 1960 96 82 14 83 7 5 Dobrzynska 1963 35 28 4 2 1 Gandhi 1963 37 37 21 16 Lennox-Buchthal 1964 100 91 8 Totals 764 290 32 104 29 10 257 170 Percentages 86 11 80 14 6 60 40 30 Table 2-3. Duration of Febrile Convulsions AUTHORS year NO. PATIENTS DURATION OF CONVULSIONS < 5 MIN. NO. PER CENT < 20 MIN. NO. PER CENT < 60 MIN. NO. PER CENT > 60 MIN. NO. PER CENT Patrick, Levy 1924 13 12 92 12 92 12 92 1 8 Herlitz 1941 732 285 39 732 100 0 Pascual, McGovern 1952 221 85 38 199 90 22 10 Broberger 1958 30 22 73 30 100 0 Millichap et al. 1960 96 34 35 66 84 96 100 0 Gandhi 1963 37 15 40 31 84 36 97 1 3 Lennox-Buchthal 1964 100 44 44 78 78 100 100 0 Totals 1229 475 39 209 76 1205 98 24 2 31 32 FEBRILE CONVULSIONS than 60 minutes in 98 per cent, and more than 60 minutes in only 2 per cent. The patients were mainly unselected and had not been grouped as simple or atypical according to rigid arbitrary criteria. The data were obtained principally from observations of parents but despite the limitations in accuracy the findings in the seven groups of patients show few exceptional variations from the mean. Very prolonged seizures that might be complicated by brain damage occurred in only 2 per cent of the total group of patients. ASSOCIATED EPISODIC SYMPTOMS A review of the literature revealed only three studies of ad- ditional and prominent symptomatology among patients with febrile convulsions. Recurrent episodic symptoms included be- havior disorders, and abdominal pain and vomiting. In the au- thor’s study recurrent attacks of behavior disorder were reported in 35 per cent of patients and were manifested by aggressive out- bursts, temper tantrums, and hyperactivity. Recurrent and epi- sodic abdominal pain or vomiting without evidence of visceral pathology occurred in 21 per cent and showed some, though not a significant, predilection for patients with nonfebrile as well as febrile seizures. Neyroud (1947) in a study of 70 patients re- ported nervousness in 18, a character disorder in 14, enuresis in 10, and speech disorders in 6; and Friderichsen and Melchior (1954) found behavior disorders in 12 (4 per cent) of 282 patients. BIRTH HISTORY AND NEUROLOGIC ABNORMALITIES In the author’s series of 95 patients evidence suggestive of possible prenatal injury, birth injury, and head injury in infancy or early childhood was obtained in 21 per cent, 20 per cent, and 17 per cent of patients, respectively. The incidence of birth trauma or anoxia determined in 19 series of patients and 17 pub- lications between 1933 and 1963 is shown in Table 2-4. Evidence of possible brain injury caused by trauma or anoxia was present CLINICAL EVALUATION AND MANIFESTATIONS 33 Table 2-4. Incidence of Birth Injury or Anoxia in Children with Febrile Convulsions AUTHORS year TOTAL PATIENTS BRAIN INJURED NO. PER CENT Freund 1933 1 1 Mendez 1945 5 2 Lennox, M. A. 1947 130 27 21 Lennox, M. A. 1947 47 16 34 Lennox, M. A. 1949 187 51 27 Lennox, M. A. 1949 66 17 26 Peterman 1950 97 27 28 Peterman 1952 275 55 20 Pardelli, Ardito 1954 10 1 10 Turnbull 1955 15 1 7 Hrbek 1957 215 48 23 Palesi 1957 101 3 3 Laplane et al. 1958 39 5 12 Lerique-Koechlin 1958 981 198 20 Bamberger, Matthes 1959 634 43 7 Millichap et al. 1960 95 19 20 Cavazzuti, Trovarelli 1961 230 21 9 Pupo 1962 191 23 12 Horstmann, Schinnerling 1963 108 17 16 Totals 3,427 575 17 in 575 (17 per cent) of a total of 3,427 patients, an incidence similar to that observed in the author’s series of 95 patients. The number of patients in each study varied from one to as high as 981; the mean was 202. The numbers of patients with a history of birth complications were one to 198, the mean being 34. The in- cidence of birth injury or anoxia varied from 3 per cent to 34 per cent, with a mean of 17 per cent. The evidence for brain injury at birth is often presumptive and equivocal, but the agreement among figures quoted in these 19 series of patients was remark- able. In reports that analyzed complicated cases compared to those having febrile seizures alone (Lennox, M. A., 1949; Milli- chap et al., 1960 hi), the incidence of birth injury was not sig- nificantly changed by this arbitrary division of patients. Trauma 34 FEBRILE CONVULSIONS or anoxia at birth is a frequent finding in the history of the febrile convulsive disorder and is implicated with equal frequency in patients with “simple” febrile seizures and with “atypical” febrile seizures. Abnormal Neurologic Signs Details of neurologic examinations were recorded infrequently in the literature. The incidence of neurologic deficits in five series of patients, including that of the author and his assistants, is shown in Table 2-5. Abnormalities in the neurologic examinations were Table 2-5. Incidence of Abnormal Neurologic Signs in Children with Febrile Seizures AUTHORS YEAR NO. PATIENTS NEUROLOGICALLY IMPAIRED NO. PER CENT Herlitz 1941 732 3 0.4 Friderichsen, Melchior 1954 282 3 1.0 Stevens 1957 73 23 32 Millichap et al. 1960 96 19 20 Lennox-Buchthal 1964 100 7 7.0 Totals 1,283 55 4.3 found in 20 per cent of the author’s series. These included hyper- kinesia in 6, retarded motor development in 3, mild hypotonia in 3, transient hemiparesis in 2, retrolental fibroplasia in 2, Klippel- Feil syndrome and cleft palate in 1, atresia of the auditory canal in 1, and strabismus in 1. Measurements of height and weight were available in 54 patients; the height in 20 per cent and the weight in 13 per cent were above the 90th percentile, and only four per cent had values below the 10th percentile. The incidence of signs of structural cerebral lesions was small in this group of 96 patients, and the average frequency of neurologic deficits among a total of 1283 patients was only 4.3 per cent. The signifi- cance of these findings in the etiology of the febrile seizures was undetermined. CLINICAL EVALUATION AND MANIFESTATIONS 35 PSYCHOMOTOR DEVELOPMENT Psychomotor retardation was noted in nine (9 per cent) of 96 patients in the author’s series, and the incidence was identical among a total of 2,764 patients recorded in 15 publications be- tween 1935 and 1964, the mean being 9 per cent with a range from 1 per cent to as high as 44 per cent of patients (Table 2-6). Table 2-6. Incidence of Psychomotor Retardation in Children with Febrile Seizures AUTHORS year NO. PATIENTS RETARDED PATIENTS NO. PER CENT Faxen 1935 238 10 4 Herlitz 1941 732 14 2 Mendez 1945 5 1 20 Lennox, M. A. 1947 136 18 13 Lennox, M. A. 1947 45 7 15 Lennox, M. A. 1949 162 35 22 Lennox, M. A. 1949 68 15 22 Friderichsen, Melchior 1954 282 5 2 Hrbek 1957 215 12 5 Palesi 1957 101 10 10 Broberger 1958 30 1 3 Lerique-Koechlin et al. 1958 228 48 21 Turinese 1959 167 73 44 Millichap et al. 1960 96 9 9 Cavazzuti, Trovarelli 1961 230 2 1 Horstmann, Schinnerling 1963 108 7 6 Keith 1964 74 5 7 Totals 2,917 269 9 Compared to patients with infantile spasms, in whom the inci- dence of psychomotor retardation is 84 per cent (Gibbs et al., 1954; Millichap et al., 1962), the prognosis regarding mental de- velopment in patients with febrile seizures is good and the occur- rence of neurologic deficits and structural cerebral lesions is relatively infrequent. 36 FEBRILE CONVULSIONS SUMMARY Febrile convulsions occur most commonly between 6 months and 3 years of age, and 95 per cent of patients have their first convulsion by the fifth birth date. The fever is caused by acute infection of the upper respiratory tract in 60 to 70 per cent of patients, and bronchitis and pneumonia, gastroenteritis, and epi- demic fevers are associated infrequently. The pattern of the con- vulsion is generalized in 86 per cent and focal in 11 per cent, with clonic movements predominating in 80 per cent of patients. The duration is less than 20 minutes in 76 per cent and more than 1 hour in only 2 per cent. Episodic behavior disorder is the most frequently associated symptom; the incidence of birth trauma or anoxia is 17 per cent; abnormal neurologic signs are elicited in 4 per cent, and psychomotor retardation occurs in 9 per cent of patients. CHAPTER 3 ELECTROENCEPHALOGRAPHY AND OTHER LABORATORY TESTS THE ELECTROENCEPHALOGRAM Authors Findings Of a total of 110 unselected patients in the study (Millichap et al., 1960 hi), 76 had at least one electroencephalogram and some patients had two or three recordings. The total number of records obtained was 89. A Grass 8-channel machine was em- ployed and bipolar and monopolar recordings were performed. A sleep recording was obtained in 52 per cent of patients and hyperventilation was possible in 20 per cent of patients. The electroencephalogram was normal in 52 patients (68 per cent). Table 3-1 shows the abnormalities observed and the frequency of their occurrence; the findings in patients with febrile seizures alone and in those with nonfebrile spontaneous seizures in addi- tion are compared. Borderline abnormalities in six patients (8 per cent) consisted of minor degrees of slow or fast activity and amplitude asymmetries (Fig. 2-lb). Electroencephalographic seizure discharges consisting of paroxysmal spikes, sharp waves, 37 FEBRILE CONVULSIONS 38 Table 3-1. Electroencephalographic Abnormalities in Children with Febrile Seizures Uncomplicated and Complicated by Spontaneous Nonfebrile Seizures FEBRILE SEIZURES ELECTROENCEPHALOGRAPHIC ALONE ABNORMALITIES ( 58 ) * FEBRILE AND NONFEBRILE SEIZURES (18) TOTAL PATIENTS EVALUATED (76) Synchronous and symmetrical 4 Focal 3 Total seizure discharges 7 (12% ) f 7 4 11 (61% ) f 11 7 18 (24%) Data derived from Millichap et ah: Neurology (Minneap) 10; 643, 1960 iii. 0 Number of patients. f Difference significant: x2 — 15.3, P < 0.001. spike and wave, and high-voltage slow waves were observed in 18 (24 per cent) patients (Fig. 2-la and Fig. 3-1); they were significantly more frequent in patients with nonfebrile seizures (61 per cent) than in those with febrile seizures alone (12 per cent). The paroxysmal discharges were of the centrencephalic type and symmetrical and synchronous in 11 patients (Fig. 3-2), and focal in seven (Fig. 3-3). The incidence of focal abnormali- ties in patients with nonfebrile seizures was greater than in pa- tients with febrile seizures alone but the difference was not significant. Age was a significant factor in relation to the incidence of abnormalities in the electroencephalogram. Of 18 patients with seizure discharges, 13 (72 per cent) were 5 to 10 years of age when the record was obtained. Of 58 patients with normal or borderline records, 49 (85 per cent) were less than 5 years old and only nine (15 per cent) were in the 5-to-10-year age group (Fig. 3-4). The patients with abnormal records were 3 to 10 years old (average, 7 years) and those with normal records were 1 to 7 years old (average, 3 years). Review of the Literature Of 36 publications in which the electroencephalogram was dis- cussed in relation to febrile convulsions 23 included details of original investigations. Table 3-2 shows the incidence of seizure ELECTROENCEPHALOGRAPHY AND OTHER TESTS 39 Figure 3-1. Electroencephalogram of a 6-year-old hoy with a history of minor motor seizures of cryptogenic type in early child- hood and two subsequent febrile convulsions at 4 years of age. An electroencephalogram obtained shortly after recovery from the febrile illness had shown slow waves with asymmetry in the occipital leads but no paroxysmal discharges. The child had received regular treatment with phenobarbital and had been free of seizures for 2 years. The present recording, obtained at follow-up examination, is abnormal, with frequent symmetrical and synchronous slow spike and wave discharges. The case report illustrates the late emergence of paroxysmal electroencephalo- graphic abnormalities in a child with febrile and previous afebrile seizures. 40 FEBRILE CONVULSIONS Figure 3-2. Electroencephalogram of a 7-year-old girl showing centrencephalic-type spike-and-tcave paroxysmal discharges. The child had febrile seizures at 9 months and 7\ years of age and one nonfehrile seizure at 7 years. (From Millichap et al.: Neurology (Minneap) 10: 643,1960 in.) discharges in electroencephalograms obtained in 2281 patients with febrile convulsions and reported during the period 1947 to 1964. The incidence varied from 2 per cent ip the publication by Bamberger and Matthes (1959) to 86 per cent in that of Peterman (1950). The average incidence calculated from these data was 25 per cent, and was almost identical with that observed in the author’s series of unselected patients (24 per cent of 76). Laplane and his co-workers in France have published results of three studies of the electroencephalogram in febrile convulsions. The first paper in 1947, one of the earliest of such investigations, was concerned with only eight patients, of whom three had spe- cific abnormalities and five had nonspecific changes in the records. In 1958, the same chief author with different co-workers reported on 39 patients with electroencephalograms. All patients were less than 4 years old and 72 per cent were followed for more than 6 months. Five patients had seizure discharges in the electroen- ELECTROENCEPHALOGRAPHY AND OTHER TESTS 41 Figure 3-3. Electroencephalogram of a 3-year-old boy showing spike-and-slow-wave focal discharges with phase reversal in left occipital area. The child was moderately retarded and had retrolental fibroplasia. He had a history of three febrile seizures, of which only one teas focal, and two subsequent nonfebrile seizures, both of which were generalized in pattern. (From Milli- chap et al.: Neurology (Minneap) 10: 643,1960 in.) cephalograms, and two of these developed recurrent nonfebrile seizures. Twenty-six of the 39 patients had entirely normal records when examined one month after the febrile convulsion. In a fur- ther study in 1963, Laplane and Salbreux examined 210 patients within 24 to 36 hours of the febrile convulsion and 177 (84 per cent) had abnormal records at this time: paroxysmal slow waves were present in 63 records, slow spikes in 32, rapid spikes in 17, spike and wave in 61, and 3 cps spike and wave discharges in four patients. The patients were observed for periods up to 6 years after the convulsion, and follow-up ex uninations showed 126 (60 per cent) with persistent abnormalities in the records. Of these 126 patients, 34 (27 per cent) had developed recurrent spontaneous nonfebrile seizures and 66 (52 per cent) had re- peated febrile convulsions which were complicated by afebrile 42 FEBRILE CONVULSIONS Normal EEG Abnormal EEG Patients with EEG Records Age in Years Figure 3-4. Age at the time of electroencephalo graphic record- ing in 76 patients with febrile seizures. Abnormal records showed recurrent paroxysmal seizure discharges and were found more fre- quently in the older children. (From Millichap et al.: Neurology (Minneap) 10: 643,1960 Hi.) seizures in 23 cases. Of the 84 patients with normal records, only one developed epilepsy and 23 (27 per cent) had recurrent febrile convulsions. A total of 35 (17 per cent) patients in the study had developed epilepsy at the time of the follow-up evaluation. One of the more extensive investigations of the electroncephalo- gram in febrile convulsions was conducted by M. A. Lennox (1947 and 1949). In the first study of 153 children with convulsions which occurred only with fever, 55 per cent had normal electro- encephalograms and 10 per cent had paroxysmal seizure dis- charges. Eleven per cent had extremely slow records; this abnor- mality was thought to reflect brain damage produced by the effects of prolonged fever and severe convulsions on the immature brain. The slowing was confined to the occipital leads and oc- curred particularly in children with a history of many febrile convulsions. The clinical findings in patients with paroxysmal electroencephalographic abnormalities were compared with those ELECTROENCEPHALOGRAPHY AND OTHER TESTS 43 Table 3—2, Incidence of Electroencephalographic Evidence of Seizures in Children with Febrile Convulsions AUTHORS year PATIENTS WITH ELECTROENCEPHALOGRAMS TOTAL NO. WITH SEIZURE NO. DISCHARGES PER CENT Laplane et al. 1947 8 3 37 Lennox, M. A. 1947 153 15 10 Neyroud 1947 70 2 3 Lennox, M. A. 1949 240 16 7 Lennox, M. A. 1949 77 26 33 Peterman 1950 88 76 86 Peterman 1952 180 133 74 Melin 1954 72 18 25 Pardelli, Ardito 1954 10 1 10 Fois, Malandrini 1957 82 14 17 Palesi 1957 101 7 7 Stevens 1957 50 19 38 Stevens 1957 23 9 39 Laplane et al. 1958 39 5 12 Radermecker 1958 50 17 34 Bamberger, Matthes 1959 305 13 4 Bamberger, Matthes 1959 109 2 2 Isch-Treussard, Rohmer 1959 50 10 20 Millichap et al. 1960 76 18 24 Cavazzuti, Lavagna 1961 80 19 24 Horstmann, Schinnerling 1963 108 28 26 Laplane, Salbreux 1963 210 126 60 Lennox-Buchthal 1964 100 9 9 Totals 2281 586 25 See text for report of 1,005 patients, Lerique-Koechlin et al., 1958. in patients with normal recordings and the data are summarized in Table 3-3. Statistical analyses of these data showed significant differences between the numbers of patients with severe convul- sions and the incidence of mental retardation in the paroxysmal and normal groups. Apparent differences in age at onset, fre- quency of convulsions, histories of abnormal birth, and epilepsy in the family were not significant. The majority of children with paroxysmal electroencephalographic records were over 5 years of Table 3-3. Correlation of Electroencephalographic Findings and Clinical Data in Children with Febrile Convulsions CLINICAL DATA ELECTROENCEPHALOGRAMS NORMAL ( % ) PAROXYSMAL ( % ) a(85)* B (169)* A(15)* B(16)* Age at examination: 1- 2 years 65 59 7 6 3- 4 years 15 15 33 31 5-10 years 16 19 60 63 Age at onset: < 1 year 18 22 f 43 38t 1-2 years 74 66 29 38 3-4 years 7 8 28 19 No. of convulsions: 1-4 74 47 5 or more 26 f 53f Severity of convulsions: Mild 74 (of 66)t 69 (of122)| 40 (of10)| 45 (of 11)t Severe 26 31§ 60 55§ 44 Table 3-3. Correlation of Electroencephalographic Findings and Clinical Data in Children with Febrile Convulsions (Continued) ELECTROENCEPHALOGRAMS CLINICAL NORMAL ( % ) PAROXYSMAL ( % ) DATA A (85)° B (169)° A (15)° B (16)° Height of fever: to 103° 13 11 17 17 103° or more 87 89 83 83 Duration of fever: 1-4 days 00 O0 o Hh 00 86 (of 66)t 67 (of 3)| 75 (of4)| 5 or more 17 14 33 25 Mental retardation 8 15§ (of115)| 15 38§ of 13)| Abnormal birth 19 25f (of 136)t 17 50f (of 12)* Epilepsy in family 12 19f (of134)| 23 25f (of 8)* Data derived from Lennox, M. A.: (A) Res Publ Ass Nerv Merit Dis 26: 342, 1947; (B) Amer J Dis Child 78: 868, 1949. 0 Number of patients, f Differences not significant. | Number of patients when different from *. § Significant differences between normal and paroxysmal groups. 45 46 FEBRILE CONVULSIONS age and this observation of a higher incidence of abnormalities in older children was confirmed by the author’s study (Millichap et ah, 1960 iii). As a result of these findings, Lennox recommended repeated electroencephalograms subsequent to a febrile convul- sion in order to determine prognosis and the necessity for therapy with anticonvulsant drugs. M. A. Lennox (1947, 1949) also compared the electroencepha- lographic findings in groups of children with febrile convulsions and approximately the same numbers of children with grand mal epilepsy; the findings are summarized in Table 3-4. Normal elec- troencephalograms were found in more than half the cases, and the total percentages of abnormalities in the two groups corre- Table 3-4. Electroencephalographic Findings Compared in Children with Febrile Convulsions and Idiopathic Grand Mal ELECTROENCEPHALO- GRAPHIC FINDINGS PATIENTS WITH CONVULSIONS FEBRILE (%) GRAND MAL (%) a(152)* b (240) A(170) B (295) Normal records 55 71 53 66 Paroxysmal discharges lot 7| 25 f 20 \ Focal abnormalities 2 5 3 7 Slow activity 7 5 8 4 Fast activity 5 5 4 2 Extreme slowing 10 7 5 1 Occipital slowing § 11 2 Data derived from Lennox, M. A.: (A) Res Publ Ass New Merit Dis 26:342, 1947; (B) Amer J Dis Child 78:868, 1949. * Number of patients. f and | Significant differences between febrile and grand mal groups. § Later considered normal. spend surprisingly closely at each age level. However, the inci- dence of the various types of abnormality was not equal in the febrile paroxysmal and grand mal groups; paroxysmal discharges were significantly less frequent and generalized slowing and oc- cipital slowing were more evident in patients with febrile con- vulsions. ELECTROENCEPHALOGRAPHY AND OTHER TESTS 47 Margaret Lennox-Buchthal (1964) analyzed the electroen- cephalograms of 100 patients with febrile convulsions from a group of 220 studied previously by Frantzen and collaborators (1964). The electroencephalograms had been obtained on the third, fourth, or fifth day of admission, 10 days to 2 weeks after admission, and at 3, 6, and 12 months and every year at the follow-up visits. The patients were divided into four groups as follows: Group 1, those with pronounced slow wave activity in the electroencephalogram (28); Group 2, moderate slow wave activity (28); Group 3, normal records (35); Group 4, a mini- mum of one paroxysmal record (9). The incidence of 9 per cent of patients with paroxysmal electrographic abnormalities in this study performed in Denmark was the same as that obtained by the author in her previous studies in the United States (Lennox, M. A., 1947 and 1949). The electroencephalograms showing slow frequencies at the third to fifth day after a febrile convulsion had a trace of the same abnormality in some after 10 days to 2 weeks, but the record was normal after an interval of 3 months. Com- pared to the patients with normal or borderline initial electroen- cephalograms, those with pronounced slow frequencies had a greater incidence of recurrence of febrile seizures and paroxysmal abnormalities in subsequent records, but the differences in inci- dence were not significant. The possible relation of the postictal slow wave abnormality to seizure recurrence and later complica- tions could not be proven statistically. Neyroud (1947) of Geneva, Switzerland, found that only 3 per cent of his patients had electroencephalograms compatible with a diagnosis of epilepsy, but an additional nine (13 per cent) had records suspicious of epilepsy. A family history of seizures of various types in 97 per cent of this series of patients was an un- usually high incidence, but only 10 per cent had a family history of recurrent nonfebrile seizures. Peterman (1950) of Milwaukee considered the electroencepha- logram the most valuable aid in diagnosis, and 86 per cent of patients in his study had dysrhythmias thought to be indicative of cerebral organic lesions. Abnormal electroencephalograms were found also in the siblings of six of eight children with febrile seizures and in the parents of three of 10 patients examined. In a 48 FEBRILE CONVULSIONS subsequent study Peterman (1952) found dysrhythmias in the records of 74 per cent of patients, and abnormalities were de- scribed as typical of petit mal or grand mal epilepsy, scattered spikes indicative of previous brain injury, and slow waves com- patible with chronic encephalitis. A history of an abnormal birth was frequent and was found in 20 per cent of this series of cases; 33 per cent of patients had recurring spontaneous convulsions. Peterman concluded that an organic basis for febrile convulsions was present in 52 per cent of cases; idiopathic epilepsy was diag- nosed in 66, cerebral birth injury with a residual lesion in 35, encephalitis in 17, brain hemorrhage residual in eight, cerebral dysgenesis in five, cerebral anoxia in two, and miscellaneous causes in 21. The remainder had occasional febrile convulsions of undetermined etiology. Fois and Malandrini (1957) of Siena, Italy, grouped febrile convulsions as “simple” in type or of an “epileptic” nature, mainly in accordance with electroencephalographic findings. The elec- troencephalograms were normal in all of 56 patients included in the “simple” group, and showed spike and wave discharges in eight and focal abnormalities in six of 26 patients in the “epi- leptic” group. Stevens (1957) used the terms “simple” and “com- plicated” in his classification of febrile seizures, and 31 per cent of 73 cases were typed as “complicated” because of a history of previous neurologic abnormalities. Electroencephalographic find- ings were not a criterion for this selection and generalized parox- ysmal and focal abnormalities in the records of these patients occurred with equal frequency in the two groups (Table 3-2). In a review of publications by Lerique-Koechlin (1958) and Lerique-Koechlin and associates (1958) from Paris, France, con- cerning a total of 1,005 patients with electrencephalographic re- cordings, the largest series reported in the literature, the exact number with seizure discharges in the records could not be de- termined with certainty. The results are omitted from Table 3-2 and are summarized as follows: 582 of 1,005 electroencephalo- grams obtained soon after an initial febrile convulsion were re- ported as normal, 313 records showed slow waves, 110 had spike discharges, and 141 were abnormal because of focal features. The electroencephalograms with seizure discharges would include ELECTROENCEPHALOGRAPHY AND OTHER TESTS 49 the group of 110 with spikes and some of those with slow wave and focal abnormalities; an incidence greater than 11 per cent could be expected. At follow-up examinations after an interval of more than one year, the electroencephalogram was repeated in 228 of the original 1,005 patients and seizure discharges were found in the records of at least 13 per cent of this group. An addi- tional 43 per cent (98 patients) had slow wave abnormalities, some of which may have been paroxysmal and compatible with seizure discharges. The electroencephalographic findings were correlated with the clinical manifestations of the febrile convul- sions and the incidence of subsequent occurrence of spontaneous seizures. The results summarized in Table 3-5 show that in the total group of 228 patients re-examined, improvement and dete- rioration of electroencephalograms had occurred in 25 per cent and 27 per cent of records, respectively. Patients were divided into those with simple febrile convulsions, some with complicat- ing nonlocalizing neurologic abnormalities, and a large group with focal febrile seizures including some with hemiplegia. The unusually high incidence of focal seizures in this study suggests that the patients who attended for re-evaluation were principally those with complicating factors. Spontaneous seizures had oc- curred in 32 per cent of children with focal febrile convulsions and in only 1.3 per cent of those with simple febrile convulsions; the incidence of hemiplegia in patients with focal seizures may have contributed to this significant disparity in prognosis. As might be expected, patients with complicated febrile convulsions and abnormal neurologic signs, and particularly those with hemi- plegia, showed a higher incidence of electroencephalographic abnormalities and recurrent nonfebrile seizures than those with simple febrile convulsions. Radermecker (1958) of Antwerp, Belgium, found a higher in- cidence of abnormal electroencephalograms among patients ex- amined after 4 years of age than in infants and young children, and the influence of age on the electroencephalogram in febrile convulsions reported in this study was confirmed by the present author (Millichap et al., 1960 iii). Diffusely abnormal or focal “epileptic” tracings were reported in 17 per cent of 23 patients examined before 3 years of age and in 50 per cent of an equal Table 3-5. Correlation of Clinical and Electroencephalographic Findings and Prognosis in Febrile Convulsions EEG REPEATED INITIAL EEG AFTER > 1 YEAR SPIKE SEIZURE SLOW WAVE LESS MORE NUMBER INCIDENCE DISCHARGES’' ABNORMALITIES ABNORMAL ABNORMAL FEBRILE OF OF CONVULSIONS PATIENTS “epilepsy” GENERALIZED FOCAL GENERALIZED FOCAL “Simple” 76 1.3% 4% 2.6% 34% 8% 35% 9% “Complicated” 47 16% 6% 2% 34% 6% 24% 11% Focal, with or without hemiplegia 105 32% 8% 13% 10% 34% 23% 50% Total 228 41 (18%) 14 (6%) 17 (7%) 53 (23%) 45(20%) 56 ( 25%) 63 (27%) Data derived from Lerique-Koechlin et al.: Rev Neurol 99: 11, 1958. # Incidence in group with focal convulsions is significantly greater than in “simple” group (x2 = 6.1, P< 0.02). 50 ELECTROENCEPHALOGRAPHY AND OTHER TESTS 51 number in whom records were obtained between 4 and 15 years of age. Seizure discharges occurred in 34 per cent of the total series. The incidence of complications was relatively high com- pared to that in the majority of reports; frequently recurring nonfebrile seizures were recorded in 60 per cent and neurologic or mental sequelae in 34 per cent of patients at the time of follow-up examination. Bamberger and Matthes (1959) of Basel, Switzerland, exam- ined 634 patients with febrile convulsions between 1930 and 1955. Out of 314 patients who were re-evaluated, 159 were followed for 10 to 25 years and 155 patients were followed for 3 to 10 years. Electroencephalograms were obtained for 414 patients and these were divided into two groups according to the time interval after the occurrence of the febrile convulsion; 305 patients were ex- amined between 1 and 7 days and 109 patients had recordings between the seventh and thirtieth day after the febrile convulsion. The results summarized in Table 3-6 show that abnormalities Table 3-6. Comparison of Electroencephalographic Abnormalities in 414 Patients Examined Within 7 Days or Between 7 and 30 Days after a Febrile Convulsion TIME OF RECORDING PATIENTS ELECTROENCEPHALOGRAPHIC ABNORMALITIES AFTER CONVULSION WITH eeg’s PARYOXYSMAL SEIZURE DISCHARGES FOCAL DIFFUSE 1-7 days 305 4%* (13) 8% (22) 27% (83) 7—30 days 109 2 %• (2) 7% (7) 10% (11) Data derived from Bamberger, Ph., and Matthes, A.; Anfdlle im Kindesalter, S. Karger, Basel/New York, 1959. * Difference in incidence not significant (x2 0.75, P < 0.05). were found more frequently in the early recordings than in the late, but a 3 per cent incidence of seizure discharges observed for the total group was much lower than the average estimated from all published reports (Table 3-2). Cavazzuti and Lavagna (1961) of Modena, Italy, studied 250 patients of whom 80 were followed for 1 to 6 years. Initial elec- 52 FEBRILE CONVULSIONS troencephalograms were normal in 128 (57 per cent) and showed discharges significant of epilepsy in 29 (12 per cent); an addi- tional 51 (20 per cent) records with spikes and hypersynchronism were suggestive of epilepsy or “pre-epilepsy.” At the follow-up examination of 80 patients, of whom 43 per cent had normal initial electroencephalograms, 24 per cent had seizure discharges in the repeat records and 8 per cent had recurrent nonfebrile seizures. The electroencephalogram had become normal in 40 per cent of those whose initial records were abnormal. Horstmann and Schinnerling (1963) of Freiburg, Germany, obtained electroencephalograms of 108 children with febrile con- vulsions. Seventy-seven were uncomplicated cases and 31 had fe- brile seizures which later became associated with latent or overt clinical manifestations of epilepsy. Seizure discharges or abnor- malities “suspicious” of epilepsy were found in the electroen- cephalograms of 26 per cent of the total group of patients and in 27 per cent of those with febrile seizures alone. At the time of follow-up examination, 83 per cent of the patients were more than 5 years old and 25 per cent were older than 10 years. Nine- teen (18 per cent) patients developed a chronic convulsive dis- order; the interval between the last febrile seizure and the onset of recurrent spontaneous seizures varied from a few months to 6 years, and in most cases was less than 1 year. The authors rec- ommended that electroencephalograms be obtained on all chil- dren with a febrile convulsion and that those with abnormal tracings be observed and treated with anticonvulsant drugs when possible. Persistence of electroencephalographic abnormalities was considered an important criterion in estimation of prognosis and second only to evidence of preceding brain damage. Summary The incidence of electroencephalographic seizure discharges in children with febrile convulsions varies from 2 per cent to as high as 86 per cent according to criteria employed in diagnosis and selection of cases. The average incidence estimated from all published reports is 25 per cent, which agrees with the figure observed in a small unselected group of patients. Pronounced or moderate slow wave frequencies occur in approximately 50 per cent of initial records obtained shortly after the convulsion, but ELECTROENCEPHALOGRAPHY AND OTHER TESTS 53 the abnormality is transient and rarely persists longer than 10 days. A correlation of the postictal slow wave abnormality with recurrence of seizures and other complications is suggested but not proven statistically. Persistent and significant abnormalities in the electroencephalogram are found more commonly in chil- dren 5 years of age and older than in younger age groups, and the incidence of paroxysmal discharges in records of patients who develop spontaneous seizures is five times that observed in chil- dren with febrile seizures alone. Paroxysmal tracings are a more frequent finding in patients with prolonged or focal febrile con- vulsions than in patients with short seizures. The incidence of paroxysmal records is lower in patients with febrile convulsions than in children of the same age with afebrile grand mal epilepsy. The value of the electroencephalogram in assessment of prognosis is established; repeated recordings at intervals are advised in all patients with a recurrence of the febrile convulsion or other com- plications so that delayed emergence of paroxysmal abnormalities and a potential susceptibility to spontaneous seizures may be excluded. CEREBROSPINAL. FLUID CONSTITUENTS Authors Study The cerebrospinal fluid was essentially normal in 86 patients examined. The concentration of sugar was greater than 80 mg per 100 ml in 24 patients and 100 mg per 100 ml or higher in 11, but the significance of these elevations was not determined; abnor- mally low concentrations were not reported. Patients whose febrile seizures were associated with infectious fevers such as measles, chicken pox, mumps, and roseola infantum were examined closely for signs of meningoencephalitis and spinal fluid abnormalities; no such signs were observed. Review of the Literature Cerebrospinal fluid findings in children with febrile convul- sions were reported in 18 publications between 1934 and 1964; these are summarized in Table 3-7. Spinal taps were performed in more than 500 of 705 children. Exanthem subitum was the Table 3- -7, Cerebrospinal Fluid Findings in Children with Febrile Convulsions AUTHORS YEAR PATIENTS WITH FEBRILE CONVULSIONS CHIEF CAUSES OF FEVER PATIENTS NO. EXAMINED WITH SPINAL NO. ABNORMAL FLUID EXAMINATIONS ABNORMALITIES Wallfield 1934 1 exanthem subitum 1 0 Habel, Lucchesi 1938 41 pertussis 22 6 pleocytosis,* increased sugar Rosenblum 1945 1 exanthem subitum 1 0 Holliday 1950 6 exanthemata, various 5 1 pleocytosis* Calzetti, Borghi 1952 100 respiratory infections 100 0 Giraud et al. 1953 74 U.R.I. 74 0 Chandy et al. 1954 18 U.R.L, malaria, etc. 18 3 increased cells,* protein or sugar Chandy et al. 1954 [18] U.R.I., malaria, etc. 7 6 increased amino acids Pardelli, Ardito 1954 10 U.R.L, gastroenteritis 2 0 Windorfer 1954 35 exanthem subitum 28 3 pleocytosis* (8-12 cells) Donald et al. 1956 29 shigellosis 17 0 Donald et al. 1956 12 other gastroenteritides 3 0 54 Table 3-7. Cerebrospinal Fluid Findings in Children with Febrile Convulsions (Continued) PATIENTS WITH SPINAL FLUID EXAMINATIONS PATIENTS WITH FEBRILE CHIEF CAUSES NO. NO. AUTHORS YEAR CONVULSIONS OF FEVER EXAMINEE I ABNORMAL ABNORMALITIES Moller 1956 34 exanthem subitum 29 1 pleocytosis* (9 cells) Phadke 1957 28 various 28 28 increased sugar Broberger 1958 30 exanthem subitum 17 1 increased protein (76 mg % ) Kowlessar, Forbes ; 1958 29 shigellosis 0 Millichap et al. 1960 110 tonsillitis, otitis 86 24 increased sugar (> 80 mg % ) Millichap et al. 1960 [110] tonsillitis, otitis [86] 11 increased sugar (100 mg % or more) Fischler 1962 17 shigellosis 0 MacDougall 1962 30 U.R.I., malaria, etc. 30 30 low protein Lennox-Buchthal 1964 100 various 63 15 pleocytosis* Totals 705 531 + 118(22%) * Indicative of encephalitis, and “febrile convulsion” doubtful. 55 56 FEBRILE CONVULSIONS cause of the fever in 101 patients and of 76 tested, 71 had normal spinal fluid, four had a slight increase in lymphocytes, and one showed an elevated protein content of 76 mg per 100 ml. Pleocyto- sis in 15 of 63 patients was reported by Lennox-Buchthal (1964) and was associated with an increased incidence of slow wave activity in initial electroencephalograms, though the correlation was not significant. An increase in cells in the spinal fluid was noted in six of 22 patients with convulsions and fever associated with pertussis (Habel and Lucchesi, 1938); the mortality in this group of patients was 66 per cent and a diagnosis of encephalitis or encephalopathy was more likely than febrile convulsion. The protein content of the cerebrospinal fluid was increased in one patient with exanthem subitum. Low values were reported in a series of 30 patients with various infections examined by MacDougall (1962); the total spinal fluid protein was below 5 mg/100 ml in 16 patients and sto 10 mg/100 ml in 14 children. A spinal tap repeated in 15 children 2 to 10 days after the febrile convulsion showed a mean protein level of 11.4 mg/100 ml and a range of 7-15 mg/100 ml. The low level of protein in the spinal fluid at the time of the febrile convulsion could not be explained by poor nutritional status and was not correlated with low levels of total protein or albumin in the serum. The patients with upper respiratory tract infections and malaria, who presented the most typical picture of benign febrile convulsions, tended to have the lowest concentrations of protein in the spinal fluid. An elevation in the sugar content of the spinal fluid was re- ported in two publications in addition to the author’s observation. Chandy and collaborators (1954) examined the fluid by paper chromatography in seven patients and found a generalized in- crease in all the amino acids in six of these specimens. Fever per se did not result in changes in the cerebrospinal fluid amino acids and in patients with seizures due to other causes the amino acid concentrations were decreased. Summary The cerebrospinal fluid in the majority of children with febrile convulsions is normal, but a slight increase in cells, a decrease in protein content, an increase in the sugar content and an increase in the amino acids may be observed in limited numbers of cases. Table 3-8. Blood Chemistries in Children with Febrile Convulsions AUTHORS YEAR PATIENTS WITH FEBRILE CHIEF CAUSES CONVULSIONS OF FEVER PATIENTS NO. EXAMINED WITH BLOOD CHEMISTRY ANALYSES* NO. ABNORMAL ABNORMALITIES Habel, Lucchesi 1938 41 pertussis “few” 0 Holliday 1950 6 exanthemata, various 2 0 Donald et al. 1956 29 shigellosis 29 0 Donald et al. 1956 12 other gastroenteritides 12 0 Kowlessar, Forbes 1958 29 shigellosis 8 0 Millichap et al. 1960 110 various 17 4 serum Na 130 mEq/L or less Fischler 1962 17 shigellosis 17 1 low Ca Ehrengut, Ehrengut 1964 6 various 6 4 decreased immunoglobu- lins Totals 250 91 + 9(10%) * Including Na, Ca, P, sugar, and nonprotein nitrogen. 57 58 FEBRILE CONVULSIONS BLOOD CHEMISTRY ANALYSES Authors Studies The level of serum sodium was 130 mEq/L or lower in four (24 per cent) of 17 patients examined within 1 or 2 hours of a febrile seizure and 131 to 138 mEq/L in the remainder of this group. Concentrations of calcium, phosphorus, nonprotein nitro- gen, and sugar in the blood were within normal limits in five patients. Review of the Literature Six studies of febrile convulsions included values of blood chemistries and a total of more than 86 patients was examined. Apart from a low serum calcium in one patient reported by Fischler (1962) and the author’s observations of hyponatremia, the blood chemistries were normal in these patients (Table 3-8). Ehrengut and Ehrengut (1954) performed serum electro- phoresis and found a lack of beta-2A and beta-2M and a reduc- tion of gamma globulin in six patients with febrile convulsions. One patient had developed fever and a convulsion after polio- myelitis vaccination (Type I), and two patients had febrile sei- zures within 3 days of smallpox vaccination. In one of these patients antibodies against vaccinia could not be found at the 12th day after vaccination and the levels of beta-2A and gamma globulin were diminished. A lack of one or sometimes two im- munoglobulins was found in each of these patients and weakness of defense mechanisms by antibody formation against infections was considered a possible cause of the febrile convulsions. CHAPTER 4 ETIOLOGICAL FACTORS, MECHANISM, AND SEIZURE THRESHOLD Fever is one of several factors known to alter neuronal metab- olism and to produce a temporary increase in susceptibility to seizures. Certain drugs, toxins, electrolyte disorders, and acute anoxia are other recognized causes of isolated convulsions. The occurrence of a seizure in the individual patient will depend on the degree of hyperthermia or severity of the noxious stimulus and the facility of spread of the seizure discharge. The threshold to seizures varies with age and cerebral maturation, and the predisposition of infants and young children to certain types of epilepsy and electroencephalographic abnormalities may be cor- related with an immaturity or deficiency of those systems which normally inhibit seizure activity (Millichap, 1965). The threshold to febrile seizures is lowered by unknown or “cryptogenic” factors and by demonstrable structural lesions in the brain. The nature of the underlying lesion in cryptogenic seizures is presumed to be biochemical, and an acquired or genetically determined enzyme defect has been postulated. In children with a temporary defect or delayed maturity of enzyme systems the febrile seizure disorder will be “simple” or “benign,” 59 60 FEBRILE CONVULSIONS whereas in those with more persistent biochemical or structural cerebral lesions the febrile convulsions will be complicated by spontaneous seizures and other neurologic abnormalities. FEVER AND HEIGHT OF BODY TEMPERATURE Authors Studies Among 110 children with febrile seizures, the mean of 149 recordings of body temperature was 40.0° C (104.0°F) and the range was 38.5 to 41.4° C (101.3 to 106.5°F). A body temperature of 38.3° C (101.0°F) or above taken rectally immediately before or at the onset of the seizure was regarded as a significant degree of fever in the diagnosis of febrile convulsions. Of 110 patients examined in a 2-year period, 54 per cent had more than one sei- zure with fever. At each follow-up appointment records were obtained of the body temperature during febrile episodes asso- ciated with seizures and of fevers greater than 38.3° C (101.0°F) that were unassociated with seizures. The mean body tempera- ture at which convulsions occurred and the highest temperature unaccompanied by seizures were compared in individual patients and in the group of 51 patients in which evaluation was possible (Table 4-1), The mean of 101 recordings with seizures was 40.0° C (104.0°F); the range was 38.5 to 41.4° C (101.3 to 106.5°F) and the standard error was 0.07° C (0.13°F). In the same pa- tients, the mean of 51 recordings of the highest temperature un- associated with seizures was 39.6° C (103.3°F); the range was 38.5 Table 4-1. Threshold Convulsive Temperature in 51 Children with History of Febrile Convulsions FEBRILE EPISODES IN 51 CHILDREN NO. TEMPERATURE RECORDINGS MEAN BODY TEMPERATURE All fevers with convulsions 101 104.0 ± 0.13°F* Highest fevers without convulsions 51 103.3 ± 0.14°F* From Millichap, J. G.: Pediatrics 23: 76, 1959. * Significantly different (P < 0.001). ETIOLOGY AND MECHANISM 61 to 40.6° C (101.3 to 105.0°F) and the standard error was 0.07° C (0.14°F). The differences between these values was highly sig- nificant (P < 0.001). In individual patients and in the group as a whole, the height of the body temperature appeared to be an important factor in the occurrence of the febrile seizure. Seizures occurred when the degree of fever reached or surpassed the threshold convulsive temperature for each individual patient (Millichap, 1959). Review of the Literature Details of body temperature recordings at the time of febrile convulsions were noted in 11 publications. Table 4-2 shows the range of body temperature at which the maximum incidence of convulsions was noted in approximately 2,000 febrile episodes. In the majority of these series of cases, the mean temperature at the time of the convulsion was in the range 39.0 to 39.9° C (102.2 to 103.8°F). In the author’s series and in that of Keith (1964), the mean temperature was 40° C (104.0°F). A mean convulsive tem- perature below 39.0° C (102.2°F) was not recorded in any of the 11 reports. Friderichsen and Melchior (1954) studied the incidence of febrile convulsions in young children with fevers below 39.0° C and above 39.0° C (102.2°F). Table 4-3 shows that the incidence was related directly to the height of the body temperature and was significantly higher in the group of patients with fevers of 39.0° C (102.2°F) and above. Kowlessar and Forbes (1958) made similar observations in children of 6 years of age and younger with Shigella enteritis; the incidence of convulsions in- creased in direct relation to the height of the body temperature (Table 4-4). Pascual and McGovern (1952) found that the dura- tion of fever before the occurrence of a convulsion was most commonly from 1 hour to 2 days, and fever usually persisted 1 to 2 days after the convulsion. The occurrence of the convulsion was not related to the duration of the febrile illness. These authors also observed that a group of Negro children were less susceptible to febrile seizures than white children, and the incidence of oc- currence of convulsions was related to the height of the body temperature (Table 4-5). It is of interest that Peterman (1946), Table 4-2. Height of Body Temperature of Children in Relation to Occurrence of Febrile Convulsions AUTHORS YEAR NO. PATIENTS OR TEMPERATURE RECORDINGS RANGE OF BODY TEMPERATURE AT MAXIMUM INCIDENCE OF CONVULSIONS < 39°C 39.0-39.39.9°C > 39.9°C (< 102.2°F) (102.2-103.8°F) (> 103.8°F) Herlitz 1941 732 + Lennox, M. A. 1947 153 [103°F or higher in 82% ] Neyroud 1947 70 + Calzetti, Borghi 1952 100 + Pascual, McGovern 1952 341 + Hrbek 1957 215 + Broberger 1958 30 + Laplane et al. 1958 39 + Bamberger, Matthes 1959 72 + Millichap et al. 1960 149 + Keith 1964 26 + 62 ETIOLOGY AND MECHANISM 63 Table 4-3. Height of Body Temperature and Incidence of Febrile Convulsions BODY TEMPERATURE NO. YOUNG CHILDREN WITH FEVER FEBRILE NO. CONVULSIONS PER CENT 37.7-38.9°C 914 58 6.3* 39 °C and above 593 113 19* Totals 1507 17 11 From Friderichsen and Melchior: Acta Paediat Suppl 43: 307, 1954. * Significant difference in incidence (x2 = 6.6, P < 0.01). on the basis of clinical observations, postulated that most children with febrile convulsions have a convulsive threshold or a tem- perature level beyond which the seizure is precipitated, and that infections or fevers cause convulsions only in children with a potential convulsive state. The results of subsequent clinical in- vestigations outlined above confirm Peterman’s viewpoint and indicate that the height of the body temperature is an important determinant of the occurrence of febrile convulsions which may be used as a measure of the convulsive threshold in individual patients. Table 4-4. Height of Body Temperature and Incidence of Febrile Convulsions in Children up to 6 Years of Age with Shigellosis RANGE OF BODY TEMPERATURE PATIENTS WITH SHIGELLOSIS CONVULSIONS NO. PER CENT 98.6-100.4°F (37-38°C) 6 0 100.6-102.2°F (38.1-39°C) 18 2 11 102.4-104°F (39.1-40°C) 30 9 30 104.2°F+ (40.1°C+) 43 18 42 From Kowlessar, M., and Forbes, G. B.: New Eng J Med 258: 520, 1958. 64 FEBRILE CONVULSIONS Table 4-5. Influence of Racial Factor on Threshold Convulsive Temperature in Children with Febrile Convulsions RACIAL GROUP NO. PATIENTS BQDY TEMPERATURE < 103°F WITH FEBRILE CONVULSIONS NO. PER CENT White 205 106 51* Negro 130 41 31* From Pascual, R., and McGovern, J. P.: Clin Proc Child Hasp {Wash) 8: 92, 1952. * Difference significant (x2 — 12.3, P < 0.01). The Rapidity of Rise of Rody Temperature This factor has been invoked as an important precipitant of the febrile convulsion in several review publications, but there is no real evidence in support of this frequently repeated hypoth- esis. Wegman (1939), as a result of his experimental study of kittens with hyperthermia, was one of the first authors to conclude that rate of rise of temperature meant more than height of body temperature in animals which convulsed with fever. A review and statistical analysis of these data, however, indicate that the height of the temperature was no less significant than the rate of rise (Millichap, 1959; and Chapter 7). Authors of clinical studies have quoted the experiments and conclusions of Wegman in discussions of the mechanism of febrile convulsions but have offered no significant and confirmatory clin- ical data. Friderichsen and Melchior (1954), having proved the importance of the height of the body temperature by analyses of patients’ records, referred to a subjective impression that a sudden rise in temperature may also be contributory. Lennox- Buchthal (1964), in a study of 100 patients, observed a slow rate of temperature elevation in relation to convulsions in 65 and a rapid rate in only 35 patients. That the rapid rise of body tem- perature is unimportant in the precipitation of febrile convulsions is evident from calculations based on data obtained in children with artificially induced fever (Baird and Garfunkel, 1956 [analy- sis of data by Millichap]). Table 4-6 shows that the rate of ETIOLOGY AND MECHANISM 65 Table 4-6. Rate of Rise of fiody Temperature in Children with or without Convulsive Response to Induced Hyperthermia CHILDREN MAXIMUM BODY RATE OF MEAN RATE PREDISPOSED TEMPERATURE TIME TEMPERATURE OF RISE TO CONVULSIONS (°f) (hrs) RISE (°F/Hr) (°f/hr) Convulsive 104 24 2.2 response 103 44 1.0 1.7 to hyper- 104 3 1.8 thermia No convulsion 103 2 2.3 with hyper- 105 3 2.2 thermia 104 24 2.2 2.0 105 34 1.9 103 24 1.4 From data obtained by Baird, H. W., and Garfunkel, J. M.: J. Pediat 48: 28, 1956. temperature rise in the patients who convulsed in response to hyperthermia was not significantly different and tended to be slower than that in children who did not convulse. These results of clinical studies have been confirmed in experiments with various small animals; in mice, rats, guinea pigs, and kittens the occurrence of fever-induced convulsions was determined by the height of the body temperature and was unrelated to the rate of rise (Millichap, 1959). More convincing clinical data will be required to justify the prevalent concept that rapid temperature elevation is essential or even contributory to the precipitation of febrile convulsions in children. Mechanism of Convulsions in Response to Fever Hypotheses regarding the mechanism of convulsant effects of fever have been suggested as a result of observations in animals with hyperthermia induced artificially, but evidence based on clinical studies is limited and confined mainly to chemical analyses of the blood and electroencephalographic findings during fever. Baird and Garfunkel (1956) obtained unique information from a study of the electroencephalogram in 10 children with hyper- thermia induced by typhoid vaccine. Seven patients had a history 66 FEBRILE CONVULSIONS of three or more convulsions with fever, but all had normal elec- troencephalograms immediately before the study. High-voltage slow waves or spike and wave formations occurred in the majority of patients when the body temperature had reached levels ranging from 37.3 to 40.0° C (99.2 to 104.0°F). Three convulsions occurred at temperatures of 39.5 to 40.0° C (103.0 to 104.0°F), two in patients with a history of febrile seizures, and one in a boy whose electroencephalogram had been abnormal on a previous occasion. The records of all patients returned to normal within 3 days and usually within 24 hours after the febrile episode, the abnormalities being less persistent than those reported in patients with fever caused by infections. Determinations of the plasma C02 content and pH were normal in one patient examined during the occurrence of electroencephalographic abnormalities, and the authors concluded that hyperthermia per se was the only sig- nificant factor in the induction of the cerebral dysrhythmia. That fever associated with infection and in the absence of convulsions is sufficient to induce abnormalities in the electro- encephalogram has been shown by Livingston (1954), who ob- tained tracings of 10 children during upper respiratory infections. Abnormal slow wave activity occurred in the records of nine patients and persisted in four patients for at least 3 days after fever had subsided. The high-voltage slowing in electroenceph- alograms during hyperthermia resembles the changes induced by hyperventilation, and respiratory alkalosis as a consequence of the hyperpnea that complicates fever is a possible factor in etiology. Changes in cerebral metabolism, oxygen consumption, and blood flow as a result of hyperthermia are additional con- tributory mechanisms to be considered (Meyer and Handa, 1967). Both respiratory alkalosis and acidosis have been reported in subjects with artificially induced fever, and the relation of changes in acid-base balance to the seizure threshold is inconstant in animal experiments (Millichap, 1965). Hemodilution is one of the earliest responses of the body fluids to heat stress, and the increase in plasma and blood volume is associated with little change in composition of the blood (Bass and Henschel, 1956). With continued heat stress the red cells and circulating protein ETIOLOGY AND MECHANISM 67 are increased, and with sweating and dehydration the blood volume is reduced. Changes of this magnitude occur in heat stroke and severe hyperthermia, but their significance in relation to the mechanism of febrile convulsions in children is doubtful. How- ever, the possible importance of water and electrolyte changes in the brain as a result of fever is suggested by the occurrence of hyponatremia in a significant proportion of patients in one clinical study (Millichap et al., 1960 iii). In some patients the cause of the fever is unrelated to obvious infection and a neurogenic basis for both the fever and convul- sion may be entertained. W. G. Lennox and M. A. Lennox (1960) have referred to “thermal epilepsy” and have reported case his- tories of patients with periodic bouts of fever and seizures, un- accompanied by infection, which responded to treatment with anticonvulsant drugs. The authors postulated a seizure focus in- volving part of the hypothalamus concerned with the regulation of temperature. Neurogenic hyperthermia related to structural and functional lesions of the brain has been described by several authors (Ribadeau-Dumas and Fouet, 1925; Friedman, 1931; Erickson, 1939; Mendez, 1945; Lennox, W. G., 1953; Ylppo and Ylppo, 1956; and Kanof and Volk, 1961). The distinction between fever of noninfectious central nervous system origin and that due to infection may be confirmed by serial determinations of the plasma fibrinogen. Kanof and Volk (1961) estimated plasma fibrinogen and erythrocyte sedimentation rate in 23 children with diseases involving the nervous system, including Tay-Sachs lipid- osis, Schilder’s disease, hydrocephalus, cerebral trauma, tumor, Sydenham’s chorea, and familial dysautonomia. During each febrile episode, the children were examined thoroughly to rule out infection, and in no instance was a high plasma fibrinogen level maintained in the absence of infection. A normal fibrinogen concentration seemed to indicate a central origin for an elevated temperature, regardless of change in the sedimentation rate. This determination may prove of value in further studies of the eti- ology and mechanism of febrile seizures, but in the majority of cases signs of infection are obvious and a neurogenic cause for the fever is not plausible. 68 FEBRILE CONVULSIONS SIGNIFICANCE OF INFECTIONS Febrile seizures occur most commonly in association with acute infections of the upper respiratory tract. In approximately 7,000 patients reported in 33 publications between 1929 and 1964, ton- sillitis or pharyngitis and otitis media accounted for 62 per cent of the febrile episodes, gastroenteritis occurred in 7.2 per cent, bronchitis or pneumonia in 6.9 per cent, and various epidemic fevers in 6.0 per cent (Table 2-1), Miscellaneous and unknown causes were listed in 11.9 per cent of cases. The incidence and severity of these infections are known to vary at different ages and may be determined by changes in immunity and the oppor- tunities for exposure to bacteria and certain viruses. The preva- lence of acute tonsillitis and pharyngitis during childhood may explain in part why febrile seizures occur frequently between the ages of 6 months and 5 years. Tonsillitis and Pharyngitis Epidemiologic information is limited concerning the etiologic organisms in children with upper respiratory tract infections, particularly those complicated by febrile convulsions. In the past, acute throat infections were attributed to bacteria and mainly the members of group A hemolytic Streptococcus. More recent studies have indicated that nonbacterial agents, pre- sumably viruses, are responsible for the majority of pharyngeal infections. However, except for group A Coxsackie viruses in herpangina, adenovirus type 3 in pharyngoconjunctival fever, and adenovirus types 1, 2, 3, and 5 in some cases of pharyngotonsillitis, the organism has not been identified, and the evidence that viruses may be responsible for exudative lesions of the pharynx and tonsils is indirect (Parrot and Nelson, 1963). Viruses were recovered only occasionally from newborn infants in a hospital nursery in a study of routine cultures (Eichenwald and Kotsevalov, 1960) but were found frequently in cultures from infants older than 6 months (Moscovici et ah, 1959). Adenoviruses and enteroviruses were isolated with increasing frequency during the first 4 months of life at the same time that maternal antibodies have been noted to decrease in concentration, and the incidence ETIOLOGY AND MECHANISM 69 of these organisms in the young infant was five to ten times greater than that observed in older children (Moffett and Cramb- lett, 1962). In a study of the prevalence of Coxsackie viruses among children with various types of illnesses (Cramblett et ah, 1964), the incidence at different ages varied from 2to 5 per cent. The maximum recovery of virus was obtained in children 1 to 3 years of age, when febrile seizures are most common, and an associated undifferentiated febrile illness was complicated by convulsions in six of 25 patients affected. Twenty-four of a total of 120 patients of all ages who excreted Coxsackie viruses had no illness or an unrelated disease, and 54 per cent of infants with positive cultures for enterovirus and adenovirus were asympto- matic. In view of the apparent frequency of isolation of these viruses in children with acute respiratory illnesses, it is tempting to ascribe an etiologic role to these agents in febrile convulsions. However, the isolation of a virus is not proof of a causal relation- ship to infection and the data must be interpreted with caution. We need similar epidemiologic studies with particular attention to patients with febrile convulsions. Epidemic Infectious Fevers Epidemic diseases are a relatively infrequent cause of febrile convulsions. Broberger (1958) stated that among 2,325 patients with scarlet fever, measles, and diphtheria in the Hospital for Epidemic Diseases in Stockholm, only 0.9 per cent had convul- sions. Roseola infantum (exanthem subitum) is the infectious fever most commonly associated with febrile convulsions. Of 3,168 patients reported in 13 publications, roseola infantum was the cause of the convulsion in 4 per cent, and the incidence varied from 0.6 to 7.6 per cent (Table 4-7). M oiler (1956), who found that convulsions were a relatively frequent complication of roseola, invoked an encephalitic process and a direct involvement of the brain in this disease. Evidence in support of roseola en- cephalitis is lacking, however, and in Moller’s own series of pa- tients only one of 29 with spinal fluid examinations showed a mild pleocytosis (Table 3-7). The average incidence of convulsions among 581 patients with roseola infantum reported in 11 publi- cations was 22 per cent (Table 4-8). Compared to the common 70 FEBRILE CONVULSIONS Table 4-7. Incidence of Roseola Infantum as Cause of Febrile Convulsions AUTHORS year NO. PATIENTS OR FEBRILE CONVULSIONS ROSEOLA NO. INFANTUM PER CENT Wallfield 1934 1 1 Brown 1935 120 4 3.3 Herlitz 1941 776 5 0.6 Rosenblum 1945 1 1 Posson 1949 3 3 Friderichsen, Melchior 1954 465 <12 2.6 Livingston 1954 201 8 4.0 Moller 1956 448 34 7.6 Bamatter et al. 1958 4 4 Broberger 1958 153 30 20 Bamberger, Matthes 1959 634 9 1.4 Millichap et al. 1960 142 2 1.4 Frantzen et al. 1964 220 15 6.9 Totals 3,168 128 4.0 Table 4-8. Incidence of Convulsions in Patients with Roseola Infantum AUTHORS year NO. PATIENTS WITH ROSEOLA INFANTUM CONVULSIONS NO. PER CENT Wallfield 1934 1 1 Barenberg, Greenspan 1939 54 1 2 Greenthal 1941 100 6 6 Valquist 1942 35 8 23 Clemens 1945 80 5 6 Rosenblum 1945 1 1 Posson 1949 3 3 Windorfer 1954 117 35 30 Moller 1956 125 34 27 Bamatter et al. 1958 4 4 Broberger 1958 61 30 50 Totals 581 128 22 ETIOLOGY AND MECHANISM 71 occurrence of convulsions in the general pediatric population roseola infantum is a relatively rare illness, and the proposed importance of this infection as a specific cause of the febrile convulsion seems improbable. Broberger (1958) suggested a parencephalitic or toxic allergic process as an explanation for the convulsion with roseola, and this author in addition to Green- thal (1941) and Clemens (1945) argued against a specific in- volvement of the brain by the infection. Ten per cent of patients were affected at less than 6 months, an unusual age period for the occurrence of febrile convulsions. The remarkably high body temperature that accompanies roseola infantum may be sufficient to explain the frequent complication of convulsions, and further and more convincing data are required to support a possible encephalitic etiology. Shigella Dysentery The importance of shigellosis in the etiology of febrile seizures has been indicated in some reports, and neurotoxin formed by Shigella dysenteriae has also been implicated as a possible con- vulsive agent. Table 4-9 shows the incidence of convulsions in children with shigellosis and in patients with Shigella-negative diarrheas. Among 1,292 patients in nine publications regarding convulsions and shigellosis, the incidence varied from “occasional” and 4 per cent to as high as 45 per cent. Among 2,241 patients in two studies of Shigella-negative diarrheas the incidence of con- vulsions was only 1.7 per cent. Forbes (1954) and Kowlessar and Forbes (1958) found that febrile convulsions were quite uncom- mon in the young infant with shigellosis and occurred more fre- quently in children between the ages of 1 to 5 years. In the discussion concerning the cause of the convulsions, Forbes stressed the importance of the age of the patient at the time of the infec- tion, the height of the body temperature, and a history of epilepsy in the family; he considered that febrile convulsions and the epilepsies were related conditions. The incidence of convulsions associated with shigellosis was independent of the species of Shigella that included Flexner and Sonne bacillary dysenteries unaccompanied by neurotoxin formation. A specific cerebral in- volvement by the neurotropic toxin produced by Shigella shiga Table 4-9. Incidence of Convulsions in Children with Shigellosis and Other Gastroenteritides AUTHORS YEAR PATIENTS WITH SHIGELLOSIS PATIENTS WITH SHIGELLA-NEGATIVE DIARRHEAS NO. PATIENTS NO. AND PER CENT WITH CONVULSIONS NO. PATIENTS NO. AND PER CENT WITH CONVULSIONS Felsen et al. 1934 100 4 (4) Dennison, de Holl 1935 35 “occasional” Cooke 1940 2 2 Hardy, Watt 1945 473 49 (ID Forbes 1954 115* 26 (23) Donald et al. 1956 64 29 (45) 100 12 (12) Thompson et al. 1957 68 15 (22) Kowlessar, Forbes 1958 82* 27 (33) Fischler 1962 353 17 (4.8) 2141 26 (1.2) Totals 1292 169 (13) 2241 38 (1.7) * Only patients up to 5 years of age included. 72 ETIOLOGY AND MECHANISM 73 was emphasized by Donald and co-authors (1956) as the ex- planation for convulsions in an unusually high percentage of their patients with shigellosis. In a report by Fischler (1962) convulsions had occurred as a complication of Shigella infections of all main types, and examination of the spinal fluid had provided no evidence in support of a diagnosis of encephalitis or toxic encephalopathy. The higher incidence of febrile seizures with shigellosis in comparison with Shigella-negative diarrheas may be related to differences in severity of these infections and the incidence of complications of water and electrolyte imbalance. Pertussis A diagnosis of febrile convulsion may be questioned in some cases of pertussis because of the possibility of occurrence of an encephalopathy. However, in the majority of patients reported in the literature, the cerebrospinal fluid findings were normal and any associated neurologic abnormalities were usually transient so that the diagnosis seemed justified. The incidence of pertussis among infections causing febrile convulsions in 7,000 cases was 0.9 per cent. The disease was mentioned in only five of 33 reports but may have been included with infections listed as miscella- neous or various epidemic diseases. In one study of a group of 516 patients with pertussis, convulsions occurred in 8 per cent; patients under 2 years of age and those with bronchopneumonia and severe paroxysms of coughing were particularly susceptible (Habel and Lucchesi, 1938). Immunizations Immunizations against pertussis and smallpox were included among causes of febrile convulsions in occasional reports and the incidence in the total series was 0.8 per cent (Table 2-3). In a group of 776 patients with febrile seizures reported by Herlitz (1941), a reaction to smallpox vaccination was the cause in 0.9 per cent; among a smaller group of 141 patients (Tille, 1950) the incidence was 3 per cent. Postvaccinial encephalitis apparently was not the explanation for the seizures in these cases. The various infections that are complicated by febrile convul- sions in early childhood may play a role in the etiology of the 74 FEBRILE CONVULSIONS convulsion by one or more of several possible mechanisms: (1) a result of the fever per se; (2) an effect of toxic products of bacteria directly on the brain; (3) an abnormal immune state and allergic response to the infection; and (4) an unrecognized and mild viral encephalitis or transient toxic encephalopathy. The importance of the febrile response to infection and the height of the body temperature has been proved by experiments in animals and clinical observations, and the evidence is statistically significant. The alternative or additional mechanisms enumerated above are conjectural and reliable data in support of these hy- potheses are not yet available. Whereas children frequently have convulsions in response to infection and fever, adult patients with epilepsy may have a reduction in the incidence of seizures during acute febrile illnesses (Guthrie, 1930). Changes in the diet and degree of hydration during these fevers may have accounted for the apparent modification of the seizure threshold by infection in the adult. AGE AND CEREBRAL MATURATION The influence of age is the most obvious and well-documented factor concerned with the threshold to febrile convulsions, but the exact mechanism is undetermined. Febrile convulsions occur most commonly between 6 months and 5 years of age (Chapter 2), and in the author’s material onset before 3 months and re- currence after 9 years were not observed. In Figure 2-3, which shows the age at the time of the first febrile seizure in 7,000 patients, the highest incidence was between 1 and 2 years and this period accounted for one third of the cases. The susceptibility of this age group of patients, discussed in Chapter 2, appears to be related to the stages of anatomical, physiological, and bio- chemical maturation of the brain. Anatomical Maturation In a full-term infant at birth the average weight of the brain is 335 grams and all major lobes are clearly distinguishable. The cerebral cortex and white matter are poorly demarcated and myelination of the hemispheres is restricted to a few fibers of the primary afferent projection systems. The neurons have relatively ETIOLOGY AND MECHANISM 75 large nuclei in proportion to the size of the cytoplasm and small numbers of poorly formed Nissl bodies and dendritic processes. The immaturity of the cerebellum is recognized by the presence of a thick external granular layer. By 2 years of age when the susceptibility to febrile seizures has reached a peak, the average weight of the brain is about 1,064 grams and all long projection and association tracts are well myelinated. The size of the nucleus in proportion to the cytoplasm of the nerve cell is similar to the mature state, Nissl bodies and dendritic processes are more prominent, and the cerebellar cortex is completely differentiated (Conel, 1939; Dekaban, 1959). Electroencephalographic Maturation At birth the electroencephalogram of a wakeful infant consists of low-voltage irregular and asynchronous waves of 3 to 5 cps and during drowsiness and sleep the frequencies are increased in voltage. Between 6 months and 2 years of age the waking record shows a progressive increase in frequency and amplitude of the basic rhythms, particularly prominent in the occipital areas, and drowsiness is manifested by hypersynchrony or high-voltage slow waves (Kellaway and Fox, 1952). Paroxysms of slow activity during the drowsy state become maximal in the preschool years, are noted less frequently during later childhood, and finally dis- appear in early adult life. An exaggeration of hypersynchrony has been reported in children subject to isolated convulsions, often of the febrile type (Samson-Dollfus et ah, 1964), and the appear- ance of high-amplitude slow waves in response to hyperventila- tion is more frequent and of greater magnitude in children less than 5 years old than at subsequent age periods (Gibbs and Gibbs, 1950). Biochemical Maturation Information concerning the chemical constituents of the brain during development has been obtained principally from studies in animals, and investigations of the brains of infants and young children are limited. Tingey (1956) studied brain lipids in relation to myelination in material obtained at autopsies from newborns, young infants, children, and adults. The brain of the neonate contained considerable amounts of cholesterol and cerebroside 76 FEBRILE CONVULSIONS but negligible quantities of sphingomyelin. The cholesterol and cerebroside had reached maximum adult levels in the white mat- ter between 9 months and 5 years and in the cerebral cortex by 2 months of age. Sphingomyelin concentrations were one third to one half the adult values at age 9 months and provided the best indication of the degree of myelination of the brain. Cumings and colleagues (1958) found that sphingomyelin and cerebroside in- creased in amount in the white matter at a rate expected for components of myelin, whereas hexosamine and neuraminic acid increased at different rates. The concentration of hexosamine was highest from late fetal life to 7 months of age while neuraminic acid reached peak levels between 3 months and 7 years. The water content of the human brain showed a maximum level of 91.0 gm/100 gm fresh tissue at birth, and the contents at 7 months and 2, 3, and 5 years were 81.4, 75.7, 74.9, and 71.2 gm/100 gm, respectively. The activity of the enzyme carbonic anhydrase is lower in the newborn than in the adult (Ashby and Schuster, 1950; Millichap, 1957), and relatively high levels of cerebral oxygen consumption and blood flow in early childhood show a fall toward adult values during the time of puberty (Kennedy, 1956; Kety, 1955). In studies of the biochemical maturation of the brain of ani- mals, changes in the activity of carbonic anhydrase and the con- centrations of water and electrolytes have been correlated with developmental modifications in the threshold to experimental febrile seizures (Millichap, 1958 ii and 1960 i; Millichap et al., 1958). Ontogenetic studies of myelin formation and the histo- logical appearances of the cerebral cortex and white matter show only partial correlations with susceptibility to febrile seizures, since evidence of continuing maturation of the brain after 5 years of age is associated with a sharp decrease in seizure susceptibility. The development of inhibitory seizure mechanisms at this age period may be postulated as an explanation for this apparent disparity in the functional and anatomical maturation of the brain. Further observations of the enzyme chemistry and electro- lyte concentrations in the human brain at different ages and stages of development might provide a more definitive explanation for the occurrence of febrile seizures within a narrow age period. ETIOLOGY AND MECHANISM 77 INFLUENCE OF SEX AND MALE PREPONDERANCE Boys are affected more commonly than girls in a ratio of 1.4 to 1 (Chapter 1). The reason for this predilection of the febrile convulsion for males is not clear, W. G. Lennox (1953) has sug- gested that males are more liable to congenital cerebral defects and birth injuries that would tend to lower the threshold to con- vulsions, and he cites a similar increased susceptibility to pen- tylenetetrazol seizures in the male adult compared to the female. In a study of shigellosis in children, Kowlessar and Forbes (1958) reported febrile convulsions in 39 per cent of 49 males and in 21 per cent of 48 females. The groups were not comparable, however, and the male patients had a greater susceptibility to convulsions based on the personal and family histories of epilepsy and the severity of the febrile response to infection, A significant difference in incidence related to sex was reported by Pascual and McGovern (1952) whose series of 455 children with febrile seizures consisted of 252 males and 203 females (\2 = 5.3, P < 0.05). Attempts to demonstrate significant differences in sus- ceptibility to experimental febrile seizures in male compared to female animals might lead to valuable correlations of the seizure threshold and the chemistry of the brain and other organs of the body. GENETIC FACTOR A genetically determined susceptibility to febrile seizures is stressed in several review articles as an important factor in the etiology and mechanism of the febrile convulsion. Schmidt and Ward (1955) have stated that “perhaps no single factor can be so strongly implicated as an hereditary predisposition,” Mainly on the basis of this postulated genetic factor, febrile convulsions have been divided into two groups: (1) simple febrile seizures, and (2) atypical febrile seizures, sometimes described as “epi- lepsy precipitated by fever” (Livingston, 1954). Prichard and McGreal (1958) quote the probable mode of inheritance of simple febrile seizures as recessive, but data in support of this 78 FEBRILE CONVULSIONS statement are not provided. It is admitted that the nature of this genetic factor is unknown, but some authors have postulated an inherited defect in physiological mechanisms concerned with the stability of the reticular activating system. As in other epilepsies, an inherited predisposition related to some enzyme or alternative biochemical defect may be important in etiology in some cases, but there is no evidence for an unusually high genetic determination in febrile seizures. This hypothesis, often repeated in clinical reviews without the support of addi- tional data, is based on one report in which the incidence of a family history of febrile convulsions among 201 patients was 58 per cent (Livingston, 1954). That this high figure is exceptional may be seen from the data in Table 4-10, which shows the familial incidence of febrile convulsions in 2,109 patients reported be- tween 1948 and 1963 in 12 different publications. The incidence varied from a low figure of 2 per cent to that of 58 per cent and the mean incidence was 17 per cent. Table 4-10. Familial Incidence of Febrile Convulsions AUTHORS year NO. PATIENTS WITH FEBRILE CONVULSIONS FAMILY HISTORY OF FEBRILE CONVULSIONS NO. PER CENT Kaplan et al. 1948 25 4 16 Peterman 1950 128 29 22 Friderichsen, Melchior 1954 405 18 4.5 Livingston 1954 201 116 58 Pardelli, Ardito 1954 10 2 20 Coelho 1958 130 6 4.6 Hrbek 1957 215 30 14 Laplane et al. 1958 39 9 23 Bamberger, Matthes 1959 523 96 18 Millichap et al. 1960 95 29 30 Cavazzuti, Trovarelli 1961 230 4 2.0 Horstmann, Schinnerling 1963 108 23 21 Totals 2109 366 17 ETIOLOGY AND MECHANISM 79 In an unselected group of 95 patients examined by the author (Millichap et al., 1960) the incidence of a familial history of febrile convulsions was 30 per cent. In this series of patients evidence for an inherited predisposition to febrile seizures was equally strong in those who suffered from nonfebrile seizures in addition to febrile seizures. Further, a tendency to nonfebrile seizures was inherited no less frequently than the tendency to febrile seizures. Our own figures are in approximate agreement with those determined from 24 series of patients reported in the literature between 1929 and 1963 and shown in Table 4-11. There was a total of 5,576 patients with febrile convulsions, and data regarding a family history of “seizures of all types” and of “epi- lepsies” were obtained in 3,771 and 3,798 patients, respectively. The incidence of a family history of seizures of all types was 32 per cent and the familial incidence of seizures regarded as “epi- lepsies” was 15 per cent. In most studies the diagnosis of epilepsy was restricted to those patients having frequently recurring non- febrile seizures. These analyses of case reports in the literature suggest that a predisposition to febrile seizures may be inherited but that this genetic factor is not of greater significance in the febrile seizure than in other epilepsies. The genetically determined abnormalities that may result in a low threshold to febrile seizures in some cases is associated with an equally prevalent genetic predisposition to nonfebrile seizures. It is possible that in febrile states the ten- dency to seizures is enhanced and is manifested at an earlier age. The hereditary factor in febrile seizures was studied in a group of 93 children with shigellosis, of whom nine (9.3 per cent) had a past history of convulsions and seven (7.2 per cent) had a positive family history for convulsions (Kowlessar and Forbes, 1958). Of 15 patients with a personal or family history of con- vulsions, 10 (66 per cent) had febrile convulsions with shigellosis, whereas in 78 patients with negative histories, 19 (24 per cent) had convulsions. The difference in incidence of convulsions with shigellosis in these two groups was very significant (x2 = 8.59, P = 0.003). The importance of a hereditary factor is illustrated by this study, but the evidence is insufficient to postulate a specific Table 4-11. Familial Incidence of Seizures of All Types and Recurrent Nonfebrile Seizures (“Epilepsies”) Among Children with Febrile Convulsions FAMILIAL INCIDENCE SEIZURES NO. PATIENTS “epilepsies” ALL TYPES WITH FEBRILE AUTHORS YEAR CONVULSIONS NO. PER CENT NO. PER CENT Faerber 1929 10 8 80 7 70 Herlitz 1941 424 39 9.2 3 0.7 Lennox, M. A. 1947 153 66 44 21 14 Lennox, M. A. 1947 49 22 43 12 25 Livingston et al. 1947 94 44 46 17 18 Neyroud 1947 70 68 97 7 10 Kaplan et al. 1948 25 5 20 Zellweger 1948 105 15 14 15 14 Lennox, M. A. 1949 185 76 41 14 8 Lennox, M. A. 1949 77 25 38 9 14 Peterman 1950 128 91 71 Peterman 1952 302 145 48 145 48 Lennox, W. G. 1953 407 212 53 214 53 Friderichsen, Melchior 1954 405 10 2.5 10 2.5 80 Table 4-11. Familial Incidence of Seizures of All Types and Recurrent Nonfebrile Seizures (“Epilepsies”) Among Children with Febrile Convulsions (Continued) authors year NO. PATIENTS WITH FEBRILE CONVULSIONS NO. FAMILIAL SEIZURES ALL TYPES PER CENT INCIDENCE “epilepsies” NO, PER CENT Pach© 1954 281 20 7 Brusa et al. 1955 26 4 1.5 4 1.5 Cary 1956 100 37 37 Hrbek 1957 215 89 41 7 3.3 Laplane et al. 1958 39 1 2.5 Bamberger, Matthes 1959 523 144 27 25 4.8 Millichap et al. 1960 95 27 28 22 23 Cavazzuti, Trovarelli 1961 230 14 6.0 14 6.0 Gandhi 1963 37 2 5.0 Horstmann, Schinnerling 1963 108 36 33 7 6.5 Totals 1199 32 574 15 (of 3,771) (of 3,798) 81 82 FEBRILE CONVULSIONS disease entity; the authors concluded that a multiplicity of factors must determine the occurrence of febrile convulsions. That the factor of inheritance among patients with febrile seizures cannot be distinguished from that of other convulsive disorders is evident from a study by Ounsted (1952), who found many different types of seizures within single families. For example, in the family of a young child with a simple febrile seizure, a sibling had grand mal epilepsy, the father had petit mal, his brother had suffered from grand mal epilepsy all his life until he died at age 59 during a convulsion, and the paternal grandmother had died in status epilepticus at the age of 70. Of 333 propositi with whole family trees or half trees, 130 (39 per cent) had a kinsman with a clear history of a convulsive disorder. Among these patients and their relatives, epilepsy and febrile convulsions could not be segregated from one another. Of 100 patients with febrile convulsions, 41 per cent had a positive family history of seizures of various types, and of 106 children whose first convulsion was afebrile, a positive family history was elicited in 46 per cent. In our understanding of the mechanism of seizures a distinction between febrile sei- zures and the epilepsies is unjustified on the basis of genetic evidence obtained in this and other studies. ACQUIRED BRAIN LESIONS Febrile convulsions may occur in patients with a history of cerebral trauma or evidence of structural brain pathology, and definitions which exclude these cases are arbitrary and without scientific basis. Peterman emphasized the acquired etiologies and considered a febrile convulsion as a symptom of an organic cerebral lesion precipitated by an elevated body temperature; he found a history of birth injury in 20 per cent of 275 patients and in 28 per cent of a group of 97 patients with febrile seizures (Peterman, 1952 and 1950). Table 2-4 shows the incidence of birth injury or anoxia in 3,427 patients with febrile convulsions reported in 19 series of patients between 1933 and 1963. The incidence was 17 per cent in the total group, ranging from 3 per cent to 34 per cent. In the author’s series of patients a history of birth trauma was obtained in 20 per cent and the incidence in ETIOLOGY AND MECHANISM 83 children who subsequently had nonfebrile seizures was twice that of the group with febrile seizures alone, but the difference was not significant. Records of abnormal neurologic signs were found in only five publications in the literature; the incidence was 4.3 per cent in 1,283 patients (Table 2-5). The incidence varied from 0.4 per cent among 732 patients reported by Herlitz (1941) to as high as 32 per cent of the 73 patients of Stevens (1957). The tendency of some authors to exclude from a diagnosis of febrile convulsions patients with obvious neurologic lesions would modify and ex- plain the wide variation in these data; Livingston (1954) grouped these patients among those with epilepsy and apart from his cases of febrile seizures, and Friderichsen and Melchior (1954) did not include patients whose febrile convulsive disorder was com- plicated by severe trauma at birth, congenital cerebral anomalies, and other known etiologic factors for seizures. In unselected series of patients, however, evidence suggests that the threshold to febrile seizures may be lowered by an inherited and functional defect of an unknown nature or by acquired and structural lesions of the brain. In patients with cryptogenic seizures the prognosis is good; in those with acquired brain pathology the seizures are expected to be more severe and protracted. MISCELLANEOUS FACTORS Other factors which may increase the susceptibility to febrile convulsions in young children include the following: electrolyte imbalance, hypersensitivity reactions to bacteria, drugs, and tox- ins, and economic and racial factors; even alterations in the weather have been postulated. Water and Electrolyte Imbalance The threshold to febrile seizures may be lowered in animals by the experimental induction of hyponatremia, and low levels of serum sodium have been observed in close relation to the oc- currence of febrile seizures in children (Millichap, 1960 i). Table 3-8 shows a low incidence of abnormal blood chemistries in children with febrile convulsions. Fischler (1962) found a low 84 FEBRILE CONVULSIONS level of serum calcium in one of 17 patients with shigellosis, but in all other reports in the literature the blood calcium was normal. Herlitz (1941) mentioned spasmophilia as a possible contributing cause in some patients whose rectal temperature was not markedly elevated. When rickets was a common disease, it is conceivable that latent tetany could have become manifested under the stress of the acute infection, but today there is no evidence in support of hypocalcemia in the etiology of febrile seizures. Fever may also precipitate convulsions and encephalopathy in patients with latent lead intoxication, but the distinction should not be difficult. Allergies and Immune Reactions The reaction of the child to infection is influenced by the age and the stage of development of the immune state. The newborn infant has a level of serum gamma globulins and protective anti- bodies derived in utero from the mother that may be sufficient to prevent or modify certain viral infections for the first 6 months of life. Protection from various bacterial infections by means of maternal antibodies is less adequate and persists for only 1 or 2 months. The synthesis of the child’s own gamma globulins begins normally at about the fourth week of life and maximal levels are attained at 1 to 3 years of age. A continuing succession of anti- genic stimuli from infections and active immunizations determines the pattern of immune antibodies within the gamma globulins, and the development of specific hypersensitivity to infections may vary in intensity according to age and inherited allergic ten- dencies (Jane way, 1963). A decrease in immunoglobulins in the plasma and a lack of antibodies in response to vaccination have been observed in one study (Ehrengut and Ehrengut, 1964). A relatively high inci- dence of allergic symptoms was reported in the author’s series of patients and in their relatives. Immune reactions to the toxic products of infecting organisms may occur or the patient may be allergic to penicillin, aspirin, and other agents used in the treat- ment of the febrile illness (Millichap et al., 1960 iii; Millichap, 1965). In a study by Hrbek (1957), amidopyrin administered as the antipyretic was thought to be a factor in the precipitation of the febrile convulsions. ETIOLOGY AND MECHANISM 85 Allergies occurred in 20 per cent and a family history of aller- gies was obtained in 38 per cent of patients examined by the author, whereas a positive family history in only 6 per cent was reported by Lennox-Buchthal (1964). Stevens (1966), in a pub- lication concerning the possible association of allergy and epi- lepsy, advised further investigation of this hypothesis, and Dees (1953) has emphasized the significance of occipital dysrhythmia in the electroencephalograms of children with allergy complicated by convulsions. The possible role of histamine in the initiation of the seizure discharge has been examined in animals with varying degrees of sensitivity to this agent (Kohn and Millichap, 1958); the results of both clinical and experimental studies indicate that further work concerning allergy and epilepsy is warranted. Economic Factors and Race W. G. Lennox (1953) studied the incidence of febrile seizures among office patients compared to clinic patients. He argued that a relatively high rate might be expected among clinic pa- tients, who presumably are exposed to more infectious diseases or else receive less prompt and effective treatment. Among the clinic patients with convulsions in the first decade, febrile sei- zures accounted for 29 per cent and in the office group the proportion was 25 per cent; the difference was not significant. Pascual and McGovern (1952) considered the influence of race on the convulsive threshold in infants and children with febrile seizures, finding the white infant more susceptible than the Negro child to the convulsive effects of a fever and body temperature below 103°F (39.5° C) (Table 4-5). Meteorohiologic Factor Tille (1950) studied 96 patients with febrile convulsions in relation to their occurrence with the passage of weather fronts. In the Berlin Charity Children’s Hospital over a period of 15 years, 81 of the 96 patients had febrile convulsions at time of cold fronts and the author concluded that this seizure should be classified as a “meteorotropic illness.” Petersen (1934), in an extensive treatise concerning the weather and disease, found that the in- cidence of epilepsy among draftees and the mortality of epileptics 86 FEBRILE CONVULSIONS in the United States was highest in the northeastern part of the continent and was related to meteorobiologic factors. Boldrey and Millichap (1967) have studied the influence of mild to mod- erate changes in barometric pressure on the susceptibility and severity of experimental seizures in animals; no relationship be- tween seizure susceptibility and barometric pressure was deter- mined, and it was assumed that meteorologic factors other than atmospheric pressure might be important in the precipitation of febrile seizures. Syncope Of 15 children with febrile convulsions whose electroenceph- alograms and electrocardiograms were normal, ocular compres- sion resulted in a marked cardioinhibitory effect in 11 patients, slow hypersynchronous waves in the electroencephalogram in nine patients, and syncope in seven patients. In eight children with seizures complicated by abnormal electroencephalograms, ocular compression had no effect. On the basis of this study Gas- taut and Gastaut (1957) considered that syncope is a possible mechanism in the etiology of febrile seizures. SUMMARY Fever, infection, and the age of the child are the principal determinants in the etiology of the febrile convulsion. The height of the body temperature at the onset of the convulsion is a measure of the threshold to febrile seizures, and the average value is 39.0 to 40.0° C (102.2 to 104.0°F). A previously proposed re- lationship between rate of rise of temperature and the precipita- tion of febrile seizures has not been confirmed in subsequent clinical and laboratory studies. The role of infections in etiology appears to be nonspecific and related chiefly to the febrile re- sponse; tonsillitis and pharyngitis of probable nonbacterial origin occur most commonly. The predilection for infants and young children between 6 months and 5 years of age shows a partial correlation with the anatomical, physiological, and biochemical immaturity of the brain. A preponderance in males and an in- creased susceptibility of white compared to Negro children are 87 ETIOLOGY AND MECHANISM unexplained. A family history of febrile seizures occurs in an average of 17 per cent of patients and a family history of “epi- lepsy” in 15 per cent. A distinction between febrile seizures and epilepsy is unjustified on the basis of genetic evidence. Inherited factors which may lower the threshold to febrile seizures also appear to increase susceptibility to spontaneous seizures. Birth injury or anoxia and structural cerebral pathology may be factors in etiology and should not negate the diagnosis of febrile con- vulsion. Among miscellaneous mechanisms are electrolyte im- balance, lack of immunoglobulins, and hypersensitivity reactions to bacteria, toxins, and drugs. CHAPTER 5 PROGNOSIS AND SEQUELAE The prognosis in a child with a febrile convulsive disorder may be determined from the clinical history of the duration and severity of the seizures and from observations of neurologic and electroencephalographic signs of seizure activity. If the definition and diagnostic criteria are restricted, as in the group termed simple febrile convulsions, the prognosis is excellent, whereas in series which include patients with prolonged seizures and electro- encephalographic abnormalities, the outlook is less good. In the author’s study of unselected patients, which included both simple and complicated cases, the febrile seizures recurred in 54 per cent, but only 12 per cent had more than four seizures (Fig. 5-1). The chance of seizure recurrence during a single febrile episode was about 1 in 3. Spontaneous nonfebrile seizures occurred in 17 per cent of patients; they were recurrent in 12 per cent and frequent in only 4 per cent. Abnormalities in the electroencephalogram and febrile seizures of long duration were the principal criteria indicative of a poor prognosis and subsequent occurrence of spontaneous sei- 89 90 FEBRILE CONVULSIONS Febrile Nonfebri le Patients with Seizures Febrile Seizures per Patient Figure 5-1. Incidence of spontaneous nonfebrile seizures in relation to the frequency of occurrence of febrile seizures in 91 patients. zures. The suggestion that febrile seizures may predispose to spontaneous seizures was not confirmed, and this complication appeared dependent upon genetic or acquired etiologic factors, common to both types of seizures (Millichap et al., 1960 hi). RECURRENCE OF FEBRILE SEIZURES Table 5-1 shows the frequency of febrile seizures and the age at the latest recurrence among 4,107 patients reported in 25 dif- ferent publications between 1947 and 1964. Recurrence of febrile seizures was recorded in 44 per cent of patients and more than four seizures occurred in 18 per cent. The age at the latest re- currence varied from 6 to 15 years. Patients with evidence of Table 5-1. Frequency of Febrile Seizures and Age at Latest Recurrence AUTHORS YEAR NO. PATIENTS FREQUENCY OF FEBRILE SEIZURES ONE ONLY RECURRENT MORE NO. PER CENT NO. PER CENT NO. THAN 4 PER CENT AGE AT LATEST RECURRENCE ( YRS ) Lennox, M. A. 1947 153 38 25 115 75 46 31 Lennox, M. A. 1947 52 18 35 34 65 Lennox, M. A. 1949 240 198* 82 43 18 Lennox, M. A. 1949 77 47* 61 30 39 Ekholm, Niemineva 1950 82 34 41 48 59 9 11 Peterman 1950 128 92 72 36 28 10 Calzetti, Borghi 1952 100 93 93 7 7 Pascual, McGovern 1952 455 330 72 125 28 Peterman 1952 302 86 28 216 72 Lennox, W. G. 1953 407 167 41 240 59 132 32 9 Bjerglund, Brandt 1954 129 57 44 72 56 12 Pardelli, Ardito 1954 10 5 50 5 50 Coelho 1957 120 14 12 106 88 6 Hrbek 1957 215 129 61 86 39 10 Palesi 1957 101 75 75 26 25 12 Radermecker 1958 50 27 54 23 46 7 14 15 91 Table 5-1. Frequency of Febrile Seizures and Age at Latest Recurrence (Continued) FREQUENCY OF FEBRILE SEIZURES AGE AT NO. ONE ONLY RECURRENT MORE THAN 4 LATEST RECURRENCE AUTHORS YEAR PATIENTS NO. PER CENT NO. PER CENT NO. PER CENT ( YRS ) Bamberger, Matthes 1959 314 164 52 150 48 66f 21 9 Vyas 1959 41 17 40 24 60 12 Millichap, et al. 1960 110 52 47 58 53 13 12 9 Cavazzuti, Lavagna 1961 250 151 60 99 40 231 9 Cavazzuti, Trovarelli 1961 230 139 60 91 40 14 6 >8 Dobrzynska 1963 35 9 26 26 74 7 Horstmann, Schinnerling 1963 108 77* 71 31 29 Laplane, Salbreux 1963 192 116 60 76 40 18 9 Frantzen et al. 1964 206 168 78 38 22 6 Totals 4107 2303 56 1805 44 328 18 * Includes some cases with one recurrence. f More than 5 seizures. | 4 or more seizures. 92 PROGNOSIS AND SEQUELAE 93 brain injury were less susceptible to frequent recurrences than those with simple febrile seizures; 5.3 per cent of patients with acquired cerebral lesions compared to 11.2 per cent of crypto- genic cases had 10 or more febrile seizures (Lennox, W. G., 1953). The incidence of recurrence of febrile seizures during a single febrile episode was reported infrequently. Laboratory studies have shown that a postictal elevation in the seizure threshold occurs in animals subjected to hyperthermia (Millichap, un- published observations) and a relatively higher body tempera- ture would be required to induce a second seizure. OCCURRENCE OF SPONTANEOUS SEIZURES Table 5-2 shows the incidence of spontaneous nonfebrile sei- zures among 5,576 children with febrile convulsions recorded in 35 publications between 1929 and 1964 from various parts of the world. Of 2,343 patients followed by 14 different investigators for varying periods of time up to a maximum of 29 years, 29 per cent had at least one spontaneous seizure; of 4,459 patients ob- served by 31 authors, 20 per cent had frequent spontaneous sei- zures that were classified as “epilepsy.” The lowest incidence of epilepsy reported was 2.6 per cent and the highest was 100 per cent; the wide variations were explained by the differences in diagnostic criteria and selection of patients. Those with short fe- brile seizures and normal electroencephalograms were grouped in prospective studies and rarely developed spontaneous seizures (Livingston, 1954), whereas patients with complicated febrile seizures had often presented at a later age with epilepsy and were diagnosed in retrospect (Lennox, M. A., 1947). Of 2,753 patients in 15 published series with various degrees of suscepti- bility to spontaneous seizures, 27 per cent had electrographic evidence of seizures, and of 1,904 patients with data available 17 per cent had records of brain dnjury at birth. The average incidence of a positive family history of epilepsy in these patients was 15 per cent and the familial incidence of seizures of all types was 32 per cent (Table 5-2). The average incidence of spontaneous epilepsies among the patients culled from the total literature is higher than the inci- Table 5-2. Incidence of Nonfebrile Seizures Among Children with Febrile Convulsions AUTHORS YEAR GEOGRAPHIC LOCATION FEBRILE CONVULSIONS (NO. PATIENTS PATIENTS WITH SEIZURES NONFEBRILE SEIZURES at : ) NO. LEAST ONE PER CENT FREQUENT NO. PER CENT Faerber 1929 Berlin, Germany 10 2 20 Faxen 1935 Gothembourg, Belgium 238 12 5.0 Bergemann 1936 Leipzig, Germany 20 3 15 Herlitz 1941 Stockholm, Sweden 424 11 2.6 Lennox, M. A. 1947 New Haven and 153 Lennox, M. A. 1947 Boston, U.S. 52 52 100 Livingston et al. 1947 Baltimore, U.S. 94 76 81 Neyroud 1947 Geneva, Switzerland 70 2 2.9 Kaplan et al. 1948 Paris, France 25 10 40 9 36 Zellweger 1948 Zurich, Switzerland 105 21 20 15 14 Lennox, M. A. 1949 New Haven and 240 20 8 Lennox, M. A. 1949 Boston, U.S. 77 77 100 Ekholm, Niemineva 1950 Helsinki, Finland 66 5 7.6 4 6.0 Peterman 1950 Milwaukee, U.S. 128 51 40 40 31 Peterman 1952 Milwaukee, U.S. 302 100 33 Lennox, W. G. 1953 Boston, U.S. 407 313 77 Friderichsen, Melchior 1954 Copenhagen, Denmark 282 62 22 11 3.9 Livingston 1954 Baltimore, U.S. 201 6 3.0 Livingston 1954 Baltimore, U.S. 297 276 93 Melin 1954 Germany 263 33 12 17 6.5 Pache 1954 Munich, Germany 281 70 25 Brusa et al. 1955 Milan, Italy 26 8 3.0 Cary 1956 Sydney, Australia 100 6 6 Hrbek 1957 Prague, Czechoslovakia 215 19 8.7 Laplane et al. 1958 Paris, France 39 2 5 Lerique-Koechlin et al. 1958 Paris, France 228 41 18 Radermecker 1958 Antwerp, Belgium 50 30 60 Bamberger, Matthes 1959 Basel, Switzerland 314 34 11 30 10 Millichap et al. 1960 New York, U.S. 95 18 19 4 4.0 Cavazzuti, Lavagna 1961 Modena, Italy 80 6 8.0 Cavazzuti, Trovarelli 1961 Modena, Italy 230 7 10 (70)° Dobrzynska 1963 Gdansk, Poland 35 7 20 Gandhi 1963 Jamnazar, India 37 7 19 Horstmann, Schinnerling 1963 Freiburg, Germany 108 31 29 19 18 Laplane, Salbreux 1963 Paris, France 210 35 17 Keith 1964 Rochester, U.S. 26 2 8.0 Keith 1964 Rochester, U.S. 48 6 12 Totals 5,576 689 29 921 20 (2343)" (4459)° “ Total patients evaluated when different from number in left-hand column. 94 FAMILIAL INCIDENCE SEIZURES “EPILEPSIES” ALL TYPES PATIENTS WITH ELECTROGRAPHIC SEIZURE DISCHARGES PATIENTS WITH HISTORY OF BRAIN INJURY AT BIRTH FOLLOW-UP PERIOD ( YRS ) NO. PER CENT NO. PER CENT NO. PER CENT NO. PER CENT 8 80 7 70 “several” 1-12 3-21 39 9.2 3 0.7 3-12 66 44 21 14 15 10 29 22 (130)” — 22 43 12 25 (49)“ 16 34 (47)” — 44 46 17 18 0-14 68 97 7 10 2 3 1-4 5 20 — 15 14 15 14 — 76 41(185)” 14 8(185)” 16 7 51 27 (187)” 0-3 25 38 9 14 26 33 17 26 0-3 7-29 91 71 76 86 (88)“ 27 28 (97) “ 0-25 145 48 145 48 133 74 (180)” 55 20 (275)” 0-27 212 53 214 53 — 10 2.5 (405)” 10 2.5 (405)“ 1-15 10 10 18 25 (72)“ > 15 20 7 1-20 4 1.5 4 1.5 1-12 37 37 10 89 41 7 3.3 48 23 2 1 2.5 5 12 5 12 21 (72%) 251 25 (1005)” 1-10 17 34 — 144 27 (523)” 25 4.8 (523)” 13 4(305)” 43 7(634)“ 3-25 27 28 22 23 18 24 (76)” 19 20 1-2 19 24 1-6 14 6.0 14 6.0 1-6 3-5 2 5.0 0-2 36 33 7 6.5 28 26 17 16 0-12 126 60 0-6 8-12 8-12 1199 32 574 15 763 27 327 17 (3771)” (3798)* (2753)“ (1904)“ 95 96 FEBRILE CONVULSIONS dence usually quoted in clinical reviews of febrile seizures. For example, W. G. Lennox (1953) stressed the limitations of retro- spective studies in which a high proportion of adult patients with chronic epilepsy are included, and he estimated that less than 5 per cent of children whose initial seizure is fever-induced will have spontaneous seizures subsequently. Friderichsen and Mel- chior (1954), who studied a selected series of patients with simple febrile convulsions, reported “epilepsy” in 3 per cent but admitted to the occurrence of “nonfebrile convulsions” in 20 per cent. Among unselected patients in the author’s report, spontaneous nonfebrile seizures occurred in 17 per cent, were recurrent in 12 per cent, and were numerous in 4 per cent (Millichap et al., 1960 iii). The follow-up period was not longer than 2 years, and a greater proportion of children with frequent seizures might be expected in extended studies. In compiling the data in Table 5-2 patients with frequently recurrent nonfebrile seizures were tabulated separately from those without recorded evidence of recurrence in order to accommodate for both conservative and more liberal definitions of the term epilepsy. In some reports which failed to indicate the frequency of the spontaneous seizures, as many as 77 per cent of patients were affected (Lennox, W. G., 1953); the exclusion of all these data from the group with frequent seizures has undoubtedly compromised the calculated average incidence of “epilepsy” in patients with febrile seizures, and a figure greater than 20 per cent may prove more accurate. The incidence of spontaneous seizures expected in a selected group of patients with simple febrile seizures would be lower than 20 per cent and the expected incidence in complicated cases would exceed the average estimate. Time of Onset of Spontaneous Seizures Reports of the incidence of occurrence of spontaneous seizures will show variable results dependent in part on the prospective compared to retrospective analysis. Among individual series of case reports in the literature the period of observation was gen- erally expressed as a range, so that some patients in the group had been followed for up to 29 years whereas others had been ex- amined only shortly before completion of the study (Table 5-2). PROGNOSIS AND SEQUELAE 97 Despite some obvious advantages of long-term follow-up, in practice the problems incurred in maintaining close and frequent contact with patients are often insurmountable and investigators are compelled to rely on data obtained at irregular intervals and in retrospect. Data collected at frequent and personal interviews with parents are more dependable, and the reports of short-term well-supervised studies may be more reliable. Information concerning the time of development of spontaneous seizures in relation to the age at the first febrile seizure should provide an index of the optimum length of a prospective study. That extension of a period of observation beyond 10 years or even 4 years is of little advantage compared to a shorter time is evi- dent from the results obtained by W. G. Lennox (1953) in a retrospective study of 313 patients who had histories of both types of seizures. Approximately two thirds of this group of patients had spontaneous seizures within 1 year of the onset of febrile seizures, 77 per cent were affected within 4 years, and 82 per cent by 10 years. Only 7.6 per cent had a delay in occurrence of spontaneous seizures beyond a 10-year period of follow-up (Table 5-3). Of Table 5-3. Time of Appearance of Spontaneous Nonfebrile Seizures in Relation to Onset of Febrile Seizures in Patients with Both Types of Seizures TIME OF APPEARANCE OF NONFEBRILE SEIZURES PATIENTS WITH FEBRILE SEIZURES NO. PER CENT Prior to febrile seizures 20 6.4 Close to febrile seizures 21 6.7 Years after febrile seizures A-i 147 46.9 1-4 54 17.3 5-9 47 15.1 10-14 14 4.5 15-35 10 3.1 Totals 313 100 From Lennox, W. G.: Pediatrics 11; : 341, 1953. 98 FEBRILE CONVULSIONS 18 patients with both types of seizures in the author’s series, 61 per cent had developed the nonfebrile seizures before 5 years of age (Fig. 5-2). Patterns of Spontaneous Seizures The type of seizure pattern is dependent on the cerebral locali- zation of origin, the severity of the stimulus, and facility of spread of the seizure discharge in the brain. It will be modified by the age and degree of maturation of the brain (Millichap, 1965). The principal clinical patterns include grand mal, petit mal, psychomotor, autonomic, and focal seizures. Generalized major seizures and petit mal, sometimes termed centrencephalic seizures, have their origin in subcortical structures of the central integrating systems of the higher brain stem and diencephalon (Penfield and Jasper, 1954). Psychomotor attacks, characterized by automatic, stereotyped movements and semipurposeful but bizarre behavior, are frequently associated with a structural ab- normality in the temporal lobe, and paroxysmal recurrent epi- sodes of abdominal pain and cyclic or periodic vomiting, as mani- festations of an autonomic epilepsy, may originate from areas of neuronal dysfunction in the hypothalamus and the temporal Patients with Seizures Febrl le i Nonfebri le Age at Onset (Years) Figure 5-2. Age at onset of seizures of 18 children with a history of both febrile and nonfebrile seizures. (From Millichap et ah: Neurology (Minneap) 10: 643,1960 in.) PROGNOSIS AND SEQUELAE 99 lobes. Focal or partial seizure patterns are dependent on the functional representation of the area of motor or sensory cortex affected by the epileptic discharge. A structural lesion is a fre- quent finding, but in infants and children a localized pathological defect in the cortex may not be demonstrable, and the focal sei- zure is sometimes related to a disturbance of metabolism. In the immature brain the development and myelinization of projection and association fibers may occur irregularly, and the propagation of a seizure discharge from deep subcortical to cortical areas may be manifested clinically in asymmetrical or focal patterns. Thus, a focal febrile seizure in an infant or young child may be more indicative of an immature state of anatomical and physiological systems than of structural lesions in the cerebral cortex, and the significance in prognosis is usually more sanguine than in adults with focal motor seizures (Fig. 5-3), A comparison of patterns of spontaneous seizures in patients with a history of febrile seizures and in patients with mixed seizure types has shown a significantly higher incidence of psy- chomotor patterns among the group with febrile seizures (Table 5-4; Lennox, W. G., 1953). Whereas grand mal and petit mal occurred with approximate equal frequency in the two groups, psychomotor seizures were diagnosed in 17 per cent of the febrile seizure group and in 5.9 per cent of the mixed seizure group. In Table 5-4. Patterns of Nonfebrile Seizures in Febrile Seizure Patients and in Patients of All Ages with Mixed Seizure Types NONFEBRXLE SEIZURE PATTERN patients with seizures FEBRILE GROUP (per cent) MIXED GROUP (per cent) Grand mal 66.2 80.0 Petit mal 11.5 9.8 Psychomotor 17.1* 5.9* Total patients 272 1,900 From Lennox, W. G.: Pediatrics 11: 341, 1953. * Difference significant. Figure 5-3. (A) At 8 months of age: Electroencephalogram showing asymmetry, with irregular high-voltage slow waves over the right hemisphere and low-voltage activity on the left. The same asymmetry occurred in a bipolar record. (B) At 18 months of age: Symmetrical electroencephalogram during sleep. The pa- tient had a history of two febrile seizures, focal and on the left side, at 8 months and no recurrence at 18 months. The mothers pregnancy was complicated by toxemia, but birth and develop- ment were normal. The case history demonstrates the lack of per- sistence of electroencephalographic abnormalities and an appar- ent good prognosis despite focal febrile seizures in infancy. PROGNOSIS AND SEQUELAE 101 the author’s study, which was prospective and restricted to chil- dren, typical psychomotor seizures occurred only in one o£ 110 patients, but recurrent and paroxysmal disorders of behavior were recorded in 35 per cent. Furthermore, a history of periodic attacks of abdominal pain and vomiting without evidence of visceral pathology was suggestive of autonomic seizures in 21 per cent (Millichap, et ah, 1955 i and 1960 iii). Of patients with seizure discharges in the electroencephalogram, 65 per cent showed a centrencephalic type of electrographic pattern (Fig. 3-2), and an involvement of subcortical structures in febrile seizures, as sug- gested by Schmidt and Ward (1955), appears likely. FACTORS PREDICTIVE OF SPONTANEOUS SEIZURE OCCURRENCE Prolonged febrile seizures and abnormalities in the electro- encephalogram were the most significant criteria of a poor prog- osis in the author’s patients. Duration of the Febrile Seizure This information was available in 79 patients of the author’s series; the longest seizure in each patient was less than 5 minutes in 43 per cent, 5 to 10 minutes in 22 per cent, 10 to 20 minutes in 19 per cent, and more than 20 minutes in 16 per cent. In 13 pa- tients who had at least one febrile seizure that lasted more than 20 minutes, spontaneous seizures occurred subsequently in 38 per cent and the electroencephalogram was abnormal in 36 per cent. The incidence of nonfebrile, spontaneous seizures and of seizure discharges in the electroencephalogram was significantly greater in patients with prolonged febrile seizures than in those with short seizures of less than 20 minutes (Fig. 5-4 and Table 5-5). Spontaneous seizures were recurrent in 30 per cent of pa- tients with prolonged febrile convulsions and in only 5 per cent of patients with short convulsions. E lectroencephalographic Abnormalities The significance of electrographic seizure discharges in prog- nosis is discussed in Chapter 3. The incidence of paroxysmal tracings in patients who developed spontaneous seizures was five 102 FEBRILE CONVULSIONS Febrile Nonfebri le Patients with Seizures Duration of Febrile Seizures (min) Figure 5-4. Incidence of spontaneous nonfebrile seizures com- pared in patients with febrile seizures of short and prolonged duration. times that observed in children with uncomplicated febrile seizures. In a clinical and electroencephalographic study by M. A. Len- nox (1949) which included 77 patients with both febrile and nonfebrile seizures, the ages ranged from less than 1 year to 45 years, and one half the patients were 5 years of age or younger at the time of evaluation. The interval between the last febrile and first nonfebrile convulsion varied from a few months to 25 years and in the majority of cases was less than 1 year. The patterns of nonfebrile seizures were of all types: grand mal oc- curred in more than one half the patients, a focal onset was re- corded in one third, petit mal was diagnosed in 9 per cent, and psychomotor seizures in 8 per cent. The electroencephalogram showed paroxysmal discharges in 33 per cent of patients who had developed nonfebrile seizures, and focal abnormalities in 17 per cent. During a 3-year follow-up period, the appearance of the electroencephalogram recorded within a week of the febrile con- PROGNOSIS AND SEQUELAE 103 Table 5—5. Relation of Length of Febrile Convulsion to Incidence of Nonfebrile Seizures and Electroencephalographic Abnormalities CLINICAL AND DURATION < 20 MIN DURATION > 20 MIN EEC FINDINGS NO. PER CENT NO. PER CENT Nonfebrile seizures 5 8 5 38 Recurrent seizures 3 5 4 30 Electroencephalograpbic abnormalities 5/48 10 4/11 36 From Millicbap et al.: Neurology (Minneap) 10: 643, 1960 iii. vulsive episode was correlated with the subsequent occurrence of nonfebrile seizures. Of patients with normal electroencephalo- grams, only 5 per cent developed convulsions without fever whereas those with various abnormalities in the records showed a significantly greater proportion with spontaneous seizures. Fifty per cent with abnormally fast frequencies, 33 per cent with ex- tremely slow or focal patterns, 25 per cent with paroxysmal dis- charges, and 17 per cent with a moderately slow abnormality had nonfebrile in addition to febrile seizures. Repeated electroencephalographic recordings in patients with a recurrence of febrile seizures are of established value in pre- diction of the likelihood of occurrence of spontaneous seizures and the choice of therapy to be advised. Recurrence of Febrile Seizures The incidence of spontaneous seizures was directly related to the frequency of recurrence of febrile seizures in 91 of a total of 110 patients who had fewer than four recurrences during a 6-month to 2-year period of observation (Fig. 5-1; Millichap et ah, 1960 iii). In patients with one, two, three, and four febrile sei- zures, the likelihood of occurrence of spontaneous seizures is 6 per cent, 20 per cent, 30 per cent and 50 per cent, respectively. The failure of spontaneous seizures to develop in three patients who had 20 and 35 recurrences with fever was probably related to the effects of administration of anticonvulsant drugs. 104 FEBRILE CONVULSIONS A tendency for the frequent recurrence of febrile seizures to be associated with complications was also noted by M. A. Lennox (1949) in a comparison of clinical and electroencephalographic findings in 240 patients with febrile convulsions of the benign type and 77 patients with “epilepsy” and a past history of febrile seizures. The incidence of patients with recurrent febrile con- vulsions in the complicated group of cases was twice that in the benign group (Table 5-6). Additional unfavorable and significant prognostic features in the study by Lennox included a history of prolonged or focal convulsions and the occurrence of febrile seizures in females Table 5-6. Comparison of Clinical Findings in Children with “Benign” Febrile Convulsions and Febrile Convulsions Complicated by Later Development of Recurrent Nonfebrile Seizures FEBRILE AND FEBRILE SEIZURES NONFEBRILE TOTAL PATIENTS “benign type” SEIZURES EVALUATED NO. NO. NO. CLINICAL DATA PXS. PER CENT PTS. PER CENT PTS. PER CENT Number of Patients 240 77 317 Males 149 62* 36 47* 185 58 Onset 1-4 years 172 73 48 64 220 70 3 or more febrile convulsions 43 18 29 38 72 22 Severe febrile convulsions 70 40* 33 62* 103 45 Mental retardation 35 22 15 22 50 22 Abnormal birth 51 27 17 26 68 27 Family history of convulsions 90 45 25 38 115 43 “Epilepsy” 14 7f 9 131 23 9 Data derived from Lennox, M. A.: J Pediat 35:427, 1949. * Significant differences between benign and complicated group. f Differences not significant. PROGNOSIS AND SEQUELAE 105 (Table 5-6). The possible importance in prognosis of focal sei- zure patterns deserves further consideration. Focal Febrile Seizures Among 322 case reports in six publications, the average inci- dence of focal febrile seizure patterns was 11 per cent. In the author’s series of 96 patients, focal seizures occurred in 14 per cent and the incidence in patients with simple uncomplicated fe- brile seizures was not significantly different from that in patients with both febrile and nonfebrile seizures. Focal electroencephalo- graphic abnormalities were of greater prognostic significance than focal clinical seizure patterns, and were found more frequently in patients who developed spontaneous seizures. The febrile con- vulsions classified as severe by M. A. Lennox (1949) were either prolonged or focal and the incidence was significantly greater in patients with complicated febrile seizures than in the group with benign seizures (Table 5-6). The numbers of patients in each group with focal febrile seizures were not listed separately, however, and the exact significance of this pattern could not be determined. Several authors of clinical reviews refer to a poor prognosis associated with focal seizures, and Livingston (1954) excludes such cases from his definition of simple febrile convulsions. Con- vincing and significant data regarding a correlation between the focal febrile and the occurrence of spontaneous seizures are limited, and the duration of the seizure and electroencephalo- graphic abnormalities are of greater prognostic import. Miscellaneous Factors Without Significance A history of abnormal birth and a family history of epilepsy are sometimes invoked as possible criteria of a poor prognosis, and the febrile seizures in patients with these findings are often re- garded as atypical. In the author’s series of patients and in the study by M. A. Lennox the incidence of these complications in patients with simple uncomplicated febrile seizures was not sig- nificantly different from that observed in patients with nonfebrile seizures (Millichap et ah, 1960 iii; and Table 5-6). 106 FEBRILE CONVULSIONS INCIDENCE AND SIGNIFICANCE OF HEMIFARESIS The possible occurrence of a permanent brain injury and hemiparesis as complications and sequelae of febrile convulsions is stressed by some authors (Lennox, M. A., 1949; Lennox, W. G., 1953; Schmidt and Ward, 1955; Fowler, 1957), but the data in support of these opinions are limited and often not strictly appli- cable. The evidence is derived from case histories that are fre- quently atypical (Lennox, M. A., 1949; Fowler, 1957) and from observations of electroencephalographic abnormalities immedi- ately following the febrile convulsion, which are usually transient (Lennox, M. A., 1947 and 1949). Furthermore, studies of effects of epilepsies in children (Zimmerman, 1938) and of hyperthermia in kittens (Lennox, M. A., et ah, 1954) are invoked in support of the hazardous nature of febrile convulsions. The brain pathology observed in children with epilepsy resulted from chronic, re- fractory convulsions with prolonged anoxia, and the neurologic lesions induced in kittens were secondary to severe degrees of artificial hyperthermia, attended by adverse environmental condi- tions and water and electrolyte imbalance. The significance of such studies in relation to children with acute and occasional febrile convulsions is debatable. Admittedly, febrile convulsions may sometimes be complicated by structural brain pathology and neurologic deficit, but these cases are relatively rare. They may be explained by a convulsion of unusual severity and duration and a fever of exceptional height; in some instances the diagnosis of febrile convulsion is equivocal and the history is suggestive of viral encephalitis or a toxic en- cephalopathy. Some examples of atypical case histories and uncommon com- plications are those presented by M. Fowler (1957) as evidence of brain damage after febrile convulsions. Summaries of the clinical manifestations of illness in the five patients studied are as follows: case 1, a boy with coma and convulsions complicating measles; case 2, high fever, unusually prolonged convulsion, per- sistent coma, and abnormal cerebrospinal fluid constituents with 14 cells and protein content of 60 mg/100 ml; case 3, convulsion, PROGNOSIS AND SEQUELAE 107 1 hour in duration and followed by persistent coma, a skin rash possibly indicative of an exanthematous infectious disease, and urine containing many granular casts; case 4, an initial convulsion of 24 hours’ duration associated with a temperature of 105°F (40.6° C) and bloody diarrhea, and followed by a persistent coma and the subsequent occurrence of typical measles; case 5, a 10- month-old female infant with scurvy and salmonella infection at the time of the febrile convulsion, which was followed by coma and spasticity. Results of serum electrolyte determinations were not recorded in the case histories of patients with gastroenteritis, and electrolyte imbalance may have contributed to the symptom- atology in these patients. The neuropathologic findings at autopsy included neuronal necrosis of variable distribution and severity in the cerebral cor- tex, basal ganglia, and cerebellum. The changes in the neurons were thought to result from systemic anoxia caused by the con- vulsions, but the author conceded that a specific mechanism act- ing locally in the brain was a possible alternative explanation for the signs of brain damage. The case histories and pathologic reports appear compatible with a diagnosis of acute toxic en- cephalopathy (Lyon et ah, 1961) in which convulsions are symp- tomatic of a disease entity of unknown mechanism which involves the brain directly. In the use of the term febrile convulsion to describe his cases, the author may intend a more liberal definition than that commonly accepted. However, the inclusion of such atypical reports among cases in which the diagnostic criteria have been applied more strictly might lead to confusion in our under- standing not only of prognosis but also of the mechanism of febrile convulsions. The incidence of hemiparesis associated with more typical manifestations of febrile convulsions was mentioned in four pub- lications of large series of patients, and additional information could be obtained only from isolated case reports and small series which included a total of 13 cases (Table 5-7). Exanthem subitum, measles, and immunizations against diphtheria, per- tussis, and tetanus were the causes of fever in eight (62 per cent) of the case reports and a toxic or allergic encephalopathy is a possible alternative diagnosis in these patients. 108 FEBRILE CONVULSIONS Table 5-7. Incidence of Hemiparesis as Complication of Febrile Convulsions AUTHORS YEAR NO. PATIENTS WITH FEBRILE CONVULSIONS PATIENTS WITH HEMIPARESIS NO. PER CENT NATURE OF FEBRILE ILLNESS Davis 1939 2 2* — tonsillitis, pyelitis Herlitz 1941 722 26 3.6 — Rosenblum 1945 1 1* — exanthem subitum Lennox, M. A. 1949 240 7t 2.9 — Posson 1949 3 3 — exanthem subitum Schwartz- man 1949 1 1* — tonsillitis, pneumonia Grace 1950 2 2 — DPT immunization Holliday 1950 5 1 — measles Calzetti, Borghi 1952 100 7* 7 — Pardelli, Ardito 1954 10 1 10 tonsillitis Fowler 1957 5 1 — pneumonia, otitis media Bamatter et al. 1958 4 1 — exanthem subitum Millichap et al. 1960 95 2* 2.1 — Totals 1190 55 4.6 w Transient Todd’s paresis only, f Transient in 5 of the 7 patients. In one large series of 722 patients with febrile convulsions, Herlitz (1941) reported 3.6 per cent with hemiparesis, but the nature of the illness in these cases could not be determined from the data available. The hemiparesis in 16 (3.7 per cent) of 435 patients evaluated in three different publications was mainly of the transient Todd’s variety. A persistent hemiparesis was re- PROGNOSIS AND SEQUELAE 109 ported in only two (0.2 per cent) patients, and these were in- cluded in the series of M. A. Lennox, (1949). In one of these patients the fever reached an unusually high peak of 41.7° C (107°F) and persisted for 4 days, and a hemispheric brain lesion may have preceded the occurrence of the focal febrile convulsion as a result of the difficult and prolonged birth history. Clinical manifestations of mild congenital hemipareses may not be evident or easily recognized during infancy, and structural lesions of the brain resulting from developmental defects or birth injuries may underlie the apparent onset of the abnormal neuro- logic signs at the time of a febrile convulsion. The average inci- dence of birth injury or anoxia among patients with febrile con- vulsions is 17 per cent (Chapter 4), and this complication occurs with sufficient, frequency to explain or contribute to the severity and sequelae of those occasional febrile convulsions associated with a permanent neurologic deficit. Acute hemiplegia of infancy, the Marie-Striimpell polioenceph- alitis of earlier literature, is caused by cerebral arterial or venous thromboses, arterial emboli, focal encephalitis, and unknown fac- tors. The paralysis is frequently the earliest manifestation, focal seizures are common but not uniformly associated, and fever is mild to moderate and usually occurs late. The previous history is rarely of significance and second attacks are almost unknown. Vascular anomalies and subintimal arterial plaques have been documented but vascular alterations induced by seizures, a cause postulated by some authorities (Schmidt and Ward, 1955), have not yet been proved. The clinical manifestations of this syndrome and the definitions of febrile convulsions cannot be equated and a causal relationship should not be implied. In the majority of published cases of febrile convulsions, the occurrence of hemiparesis is not recorded, and among the larger series of case reports listed in Table 5-7, only 0.2 per cent had a history of permanent hemiparesis. 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 electro- encephalogram (Chapters 3 and 4), The true incidence of brain 110 FEBRILE CONVULSIONS 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. SUMMARY Febrile seizures are likely to recur in 44 per cent of patients, but only 18 per cent will have more than four seizures. During a single febrile illness, the chance of recurrent seizures is about one in three if the fever is not controlled. Occurrence of spontaneous nonfebrile seizures will vary in incidence according to differences in diagnostic criteria and selection of patients. In groups that include all types of cases, nonfebrile seizures may be expected to occur in 29 per cent, and these may recur frequently in 20 per cent. In selected patients with simple febrile seizures the ex- pected incidence of spontaneous seizures will be lower than 20 per cent, and in complicated cases the incidence will exceed the average estimate. If spontaneous seizures have not developed within 1 year of the onset of febrile seizures, the chance of sub- sequent occurrence is one in three; after 4 years the risk is one in four. Prolonged febrile seizures, an abnormal electroencephalo- gram, and frequent recurrence of febrile seizures are associated with a significant increase in incidence of spontaneous seizures. After one febrile seizure, the risk is 6 per cent, and after four seizures 50 per cent of patients may develop nonfebrile seizures. Focal electroencephalographic abnormalities are of greater prog- nostic significance than focal febrile seizure patterns. Girls are less susceptible to febrile seizures but have a greater incidence of complications. Febrile seizures generally do not predispose to spontaneous seizures, and genetic or acquired factors common to both types of seizures may be important in etiology. Permanent brain damage and hemiparesis as a direct result of the febrile convulsion is a rare complication, occurring in less than two of 1,000 patients. CHAPTER 6 TREATMENT OF FEBRILE CONVULSIONS TREATMENT OP THE INITIAL CONVULSION The management of the acute febrile seizure may be discussed in four parts in order of urgency and importance: Control of the Convulsion An anticonvulsant should be administered by the parenteral route, either subcutaneously or intravenously according to the severity and pattern of the convulsion. When the seizure is moderately severe and mainly clonic in pattern, the subcutaneous route of administration is indicated, whereas tonic and protracted seizures attended by asphyxia demand more urgent control, and intravenous injection is advisable. The choice of drug is deter- mined largely by the experience of the individual physician, but a potent anticonvulsant with additional antipyretic properties would be preferred. The initial dose should be optimal and the maximal tolerated amount for the age, weight, or surface area of the child. A relatively large, initial single dose is superior in 111 112 FEBRILE CONVULSIONS efficacy to small divided doses given at intervals. Anticonvulsants and their optimum doses are shown in Table 6-1. Phenobarbital sodium has both anticonvulsant and antipyretic properties (Millichap, 1960 ii), and the risk of laryngeal spasm and apnea following intravenous administration is less than that which attends the use of short-acting barbiturates. Phenobarbital is also superior to diphenylhydantoin (Dilantin Sodium), which even exacerbates the febrile clonic seizure and fails to control the fever (Millichap et al., 1960 ii). If the convulsive movements continue, half the initial dose of phenobarbital is given after an interval of 20 minutes, and if further medication is required to control the seizure, amobarbital sodium (Amytal), a short-acting barbiturate, is preferred. When the patient regains consciousness, doses of anticonvulsant should be continued orally until the temperature is normal and recovery from the infection is complete. In febrile states, larger doses of anticonvulsant are required than those usually necessary to con- trol nonfebrile seizures; phenobarbital, 5 mg/kg daily, in three or four equal parts, is usually required to prevent recurrence of the Table 6-1. Anticonvulsant Drugs of Value in Treatment of Febrile Convulsions and Suggested Dosages AVERAGE OPTIMUM DOSE RANGE DRUG PREPARATION (cm) PARENTERAL ( mg/kg ) ORAL (mg/kg daily) Phenobarbital sodium (Luminal) Amps. 0.125, 0.3 Tabs. 0.016, 0.032, 0.065, 0.1 3-5 (maximum total, 250 mg) 3-6 Amobarbital sodium (Amytal) Amps. 0.065, 0.125, 0.25 3-5 •— Secobarbital sodium (Seconal) Supp. 0.03, 0.06, 0.125, 0.2 3-5 (rectal) — Pentobarbital sodium (Nembutal) Supp. 0.03, 0.06, 0.12, 0.2 3-5 (rectal) — Primidone (Mysoline) Tabs. 0.05, 0.25 — 5-20 TREATMENT 113 seizure in a patient with fever. If primidone (Mysoline) is em- ployed, an alternative choice of orally administered anticonvul- sant, an initial dose of 5 mg/kg daily in three or four divided amounts should be increased according to tolerance up to an optimum amount of 20 mg/kg per day. Reduction of Body Temperature In the treatment of hyperthermia, medical opinion is divided almost equally between those who favor conductive cooling by immersion of the patient in an iced bath, and those who prefer evaporative cooling by wet sheets and fanning with cool dry air. In both methods vigorous massage is requisite to maintain the cutaneous circulation (Minard and Copman, 1963). Ferris and co-workers (1938) recommend ice baths for patients with tem- peratures greater than 41.7° C (107°F), but evaporative cooling is considered adequate in the treatment of milder degrees of fever. The height of the fever is the most significant determinant of a febrile convulsion, and treatment should be aimed toward pre- vention of a rise in temperature to the threshold convulsive level. Barbiturates and chlorpromazine (Thorazine) are the most effec- tive antipyretic drugs during the pyrogenic stage of a fever, whereas acetylsalicylic acid (aspirin) acts only at the height of a fever or during its defervescence (Millichap et ah, 1960 ii). Phenobarbital inhibits heat production, whereas aspirin, the ap- plication of tepid water, and rubbing of the skin with alcohol facilitate heat loss. Water at an approximate temperature of 37° C (98°F) should be used to sponge the anterior surface of the body including the axillae and groins, and total immersion in cold water, which induces constriction of superficial vessels, is advised only if an ice bath is employed. Enemas of cold water, advocated in the past, are now considered inadvisable because of the dangers of retention and water intoxication. Treatment of Acute Infection Nonbacterial pharyngitis and tonsillitis, presumably viral in origin, are more common than streptococcal and other bacterial infections of the upper respiratory tract during infancy and early childhood, and the routine administration of antibiotics at the 114 FEBRILE CONVULSIONS onset of the febrile convulsion is not generally advisable. A diag- nosis of acute bacterial meningitis must be considered when fever is persistent and unexplained and the seizure is refractory to initial doses of anticonvulsants; the absence of nuchal rigidity in infants is not sufficient evidence for the deferral of a spinal tap. A throat culture is obtained for confirmation of a bacterial pharyn- gitis whenever practicable, and the administration of unneces- sarily large doses of penicillin should be avoided. Penicillin is a convulsant when applied locally to the cerebral cortex; it causes seizure activity in the electroencephalogram when injected intra- spinally, and in large intravenous and intramuscular doses in infants it may enhance the tendency to febrile seizures. In addi- tion to these toxic effects on the brain, hypersensitivity reactions to penicillin occur in some patients and a seizure is precipitated shortly after the injection (Millichap, 1965). Management of Electrolyte Disturbances A determination of the level of serum sodium is advised if febrile seizures are prolonged and resistant to anticonvulsant therapy. Hyponatremia has been observed in association with febrile convulsions, resulting from rapid administration of hypo- tonic parenteral solutions, cold-water enemas, and excessive quantities of oral fluids. The treatment of water intoxication and hyponatremia consists of intravenous hypertonic saline (2.5 per cent), 5-10 ml/kg body weight, administered over a period of 1 or 2 hours. Water is withheld, and osmotically active substances, such as 15 per cent mannitol or 6 per cent urea, are given if the hyponatremia is complicated by signs and symptoms of increased intracranial pressure (Forbes, 1964). Nursing Care of Child During Seizure Most convulsions are brief, self-limited, and relatively harmless, unless complicated by a fall. All hard, injurious objects and soft pillows are removed from possible contact with the patient, and food and other foreign matter are extracted from the mouth. Aspiration of saliva or vomitus is prevented by elevating the foot of the crib or bed, and impending suffocation and asphyxia are relieved by loosening clothing around the neck and chest and TREATMENT 115 raising the tongue to a forward position with the operator’s finger placed under the jaw. Oxygen is administered if the skin is cyanotic and when a tonic seizure is prolonged. Restraint of convulsive movements during the seizure and stimulation in the subsequent sleeping phase are inadvisable. Reassurance and avoidance of excitement and apprehension are necessary on awakening and recovery. PROPHYLACTIC TREATMENT AND LONG-TERM MANAGEMENT Although authorities are generally in agreement on the method of control of the initial febrile seizure, the prophylactic treatment and long-term management are controversial. Some recommend that phenobarbital or diphenylhydantoin be given continuously to all patients after the first febrile seizure and until 3 to 6 years of age. However, most authors are opposed to continuous ther- apy without discrimination and prescribe phenobarbital or other anticonvulsants only at the time of subsequent febrile episodes. Continuous therapy with anticonvulsants is recommended and reserved for patients with certain complicating and defined factors. Proponents of Continuous Therapy Authors in favor of universal employment of continuous anti- convulsant therapy and those opposed are listed in Table 6-2. The various rationales presented in support of continuous therapy are as follows: (1) the frequency of recurrence of febrile seizures is high and might be controlled by anticonvulsant therapy, (2) sub- sequent occurrence of prolonged seizures or status epilepticus is unpredictable, (3) febrile seizures may result in irreversible brain damage, (4) reliable criteria for separation of benign types of febrile convulsions from those less benign are lacking, and (5) the parent cannot be expected to anticipate a fever with infections of subtle onset and to administer anticonvulsant and antipyretic therapy sufficiently early for successful prevention of the seizure. The majority of these arguments cannot be reconciled, however, with statistics and other evidence available in the Table 6-2, Recommendations For and Against Anticonvulsant Drug Treatment of Febrile Convulsions by Various Authors PATIENTS WITH RECURRENCE OF FEBRILE SEIZURES CONTINU- OUS DRUG QUALIFYING REPORT INTER- THERAPY DURATION CRITERIA OF DRUG TOTAL CON- MITTENT CON- ADVISED OF THERAPY FOR TRIAL PATIENTS TINUOUS OR NO PERIOD OF TROLLED FOR ALL RECOM- REGULAR AUTHORS YEAR DATA IN STUDY THERAPY THERAPY OBSERVATION STUDY PATIENTS MENDED TREATMENT Livingston et al. 1947 + 94 52% 49% 2-14 yrs — — Gallant, Livingston 1949 — to age Lennox, M. A. 1949 — + 3 yrs none Peterman 1950 + 128 84% 1-25 yrs — — Peterman 1952 — — Bjerglund, Brandt 1954 to age 1-3 Jennings 1954 — + 3 yrs seizures Livingston 1954 — — Schmidt, Ward 1955 — + none Fois, 1-3 Malandrini 1957 — + mths none Prichard, McGreal 1958 — — to age Schmidt 1958 — + 4 yrs + 116 Table 6-2. Recommendations For and Against Anticonvulsant Drug Treatment of Febrile Convulsions by Various Authors (Continued) PATIENTS WITH RECURRENCE OF CONTINU- FERRILE SEIZURES OUS DRUG QUALIFYING REPORT INTER- THERAPY DURATION CRITERIA OF DRUG TOTAL CON- MITTENT CON- ADVISED OF THERAPY FOR TRIAL PATIENTS TINUOUS OR NO PERIOD OF TROLLED FOR ALL RECOM- REGULAR AUTHORS year DATA IN STUDY THERAPY THERAPY OBSERVATION STUDY PATIENTS MENDED TREATMENT > 1 Chao et al. 1958 — + 1 yr seizure Bamberger, Matthes 1959 — — Millichap et al. 1960 + 47 57 % 53% 4-2 yrs + — Prichard 1961 — — Doose 1962 — — Dobrzynska 1963 - - Horstmann, Schinnerling 1963 - — Keith 1963 — - to age Carter, S. 1964 — + 6 yrs none Frantzen et al. 1964 + 206 21% 23% 1 to >2 yrs + — Millichap 1964 — — Costeff 1965 — — Hammill, to age Carter, S. 1966 — + 6 yrs none Totals 475 57% 33% 117 118 FEBRILE CONVULSIONS literature. Original data obtained from controlled clinical and therapeutic studies show that the choice of continuous therapy is unwarranted and inadvisable except in selected patients and according to certain well-defined criteria. Arguments Against Universal Use of Continuous Therapy The regular and continuous administration of anticonvulsants after the first febrile seizure in all patients is considered irrational for the following reasons: 1. The incidence of febrile convulsions in the population under 5 years of age in the United States has been estimated at one-half million, and the supervision of long continued anticonvulsant therapy in all cases for 1 to 5 years would be impractical. Ob- jective evidence based on well-designed studies should be made available in order to justify this recommendation. 2. Febrile seizures recur in 44 per cent of children and an incidence of more than four seizures is recorded in only 18 per cent of patients (Table 5-1). The majority of these statistics were obtained in communities where long-term continuous therapy with anticonvulsants would not be readily available. Thus, if prescribed in all patients, treatment would be superfluous in 56 per cent. 3. Prolongation of the febrile seizure for more than 60 minutes occurred in only 2 per cent of 1,229 patients, and the duration of the seizure was less than 20 minutes in 76 per cent. Status epilepticus as a complication of a typical febrile seizure is rare, and small doses of anticonvulsants given regularly would offer little protection. 4. The incidence of permanent hemiparesis following febrile convulsions has been estimated at less than 0.2 per cent, and in reports of brain damage following febrile convulsions the diag- nosis is frequently equivocal or the possibility of preexisting neurologic lesions caused by birth injury, anoxia, or other factors has not been excluded. In the reference by M. A. Lennox (1949), often quoted by those who emphasize the potential hazards of febrile convulsions, hemiparesis was recorded in 2.9 per cent of 240 patients but only two had a permanent disorder. Furthermore, the birth history was complicated and may have contributed to TREATMENT 119 the signs of brain damage in at least one of these patients. The statistics culled from a majority of reports do not substantiate the malignant nature of the febrile convulsion attested to by a numerical minority (Chapter 5). 5. Patients with a strong potential for spontaneous seizures have clinical manifestations which distinguish them from those less likely to develop complications. The division of cases into benign or simple and complicated or atypical is arbitrary in terms of the mechanism of febrile convulsions but useful in management and prognosis. The incidence of occurrence of spon- taneous seizures in patients with febrile seizures is related prin- cipally to two criteria: (1) the duration of the febrile seizures, and (2) electroencephalographic abnormalities, persistent and indicative of susceptibility to seizures. Patients with short febrile convulsions and a normal electroencephalogram have a good prognosis irrespective of therapy, whereas those with prolonged convulsions and epileptiform discharges in the electroencephalo- gram may benefit from regular administration of anticonvulsant drugs. Treatment may be prescribed according to these and other criteria and in relation to the needs of the individual patient; and the impracticalities and potential dangers of long-term, inade- quately supervised, anticonvulsant therapy in many thousands of patients with febrile convulsions may be avoided. 6. That subsequent infections may be subtle in onset and the febrile response sudden is admitted, but a fast-acting barbiturate administered promptly in a rectal suppository and the use of optimum methods of cooling should result in improved control of seizures. This immediate and interval form of treatment is no less effective in the prevention of febrile seizures than the con- tinuous anticonvulsant regimen (Millichap et ah, 1960 i), and the risk of drug side-effects is reduced. More active and less toxic medications than those presently available, having both antipy- retic and anticonvulsant properties, are required, however. A physician’s prescription for continuous anticonvulsant drug treatment is not proof of a parent’s compliance and strict adher- ence to this recommendation. The reluctance of some parents to accept this form of therapy for an illness generally regarded by the laity as benign is not an uncommon experience among physi- 120 FEBRILE CONVULSIONS dans in pediatric practice, and reports of apparent benefit derived from therapy may be misleading unless patients are examined and interrogated frequently. The publication of vignettes describing personal experiences and apparently authoritative conclusions without confirmatory evidence may confuse our understanding of the optimum management of the febrile convulsive disorder. A clinical trial of anticonvulsant medications which satisfied the demands and criticisms of all statisticians would be difficult to design, but recommendations based on prospective studies which incorporate some controls and frequent observations might rea- sonably be accepted in preference to reports unsupported by original data. Controlled Comparisons of Continuous and Intermittent Methods of Therapy Two prospective studies of the relative efficacy of regular and intermittent or no anticonvulsant treatment are reported (Milli- chap et al., 1960 i; Frantzen et al., 1964). In the author’s study, which included all patients with febrile convulsions who attended an emergency service at a general hospital during a 2-year period, one of two methods of treatment was advised as follows: some patients received phenobarbital (mean daily dose 3.Bmg/kg) only at the time of each febrile episode and parents were advised to administer the drugs orally at the earliest sign of infection; alternate patients were given regular daily doses of phenobarbital (mean dose 3.lmg/kg) in addition to that received at the time of each subsequent febrile episode. The effect of diphenylhy- dantoin administered regularly (mean daily dose 10 mg/kg) was observed in a small group of patients which included some who had previously received phenobarbital. All patients were treated with aspirin, tepid water baths, and antibiotics when necessary. The number of recurrences of febrile seizures was recorded and patients were grouped according to the method of therapy and degree of response. Clinical and electroencephalographic findings were compared so that, apart from the treatment under study, any systematic difference in the factors which could influence the incidence of seizures might be recognized. The results are shown in Table 6-3. Febrile seizures recurred in 53 per cent of 19 pa- TREATMENT 121 Table 6-3. Recurrence of Febrile Convulsions in 40 Children Treated Intermittently or Regularly with Phenobarbital PRETRIAL NO. CONVULSIONS INTERMITTENT THERAPY REGULAR DRUG THERAPY NO. PATIENTS PATIENTS WITH RECURRENCES NO. PATIENTS PATIENTS WITH RECURRENCES 1-3 14 6 12 4 4-10 5 4 9 5 Total 19 10(53%)* 21 9(43%)* Modified from Millichap et al.: J Pediat 56: 364, 1960 i. * Difference not significant. tients treated intermittently and in 43 per cent of 21 patients treated continuously with phenobarbital. All seven patients given diphenylhydantoin by continuous administration had recurrences. The number of recurrences in patients not benefited by pheno- barbital or diphenylhydantoin ranged from one to seven, with an average of two seizures per patient. Sedation or ataxia occurred in two patients treated continuously with phenobarbital and in two who were treated with diphenylhydantoin; toxic effects were not observed in patients who received phenobarbital intermit- tently. The period of observation ranged from 6 months to 2 years. The two groups of patients who received phenobarbital were com- parable with regard to frequency of complicating factors in the personal and family histories and the incidence of epileptiform discharges in the electroencephalograms. Abnormal clinical find- ings in the patients treated with diphenylhydantoin were about equal in frequency to those in the groups treated with pheno- barbital. Of the total number of patients in the study, 22 (53 per cent) had a recurrence of febrile seizures. The incidence of recurrence was greater among patients with a previous history of four or more febrile seizures (64 per cent) than in those with a history of one to three seizures (38 per cent), but the difference was not significant. Apart from the incidence of focal seizures, the frequency of abnormal clinical and electroencephalographic findings was not significantly different in patients whose seizures 122 FEBRILE CONVULSIONS were controlled and in those who had recurrences. Contrary to reports of a poor prognosis attending focal febrile seizures, the majority of patients with this type of seizure had no recurrence (Table 6-4). Table 6-4. Clinical Findings in Patients with and without Recurrences of Febrile Convulsions Following Intermittent or Regular Therapy with Anticonvulsant Drugs CLINICAL FINDINGS PATIENTS WITH FEBRILE CONVULSIONS RECURRENT NONRECURRENT No. patients 22 19 Mean convulsive temperature 39.8'°C 39.8°C Focal febrile convulsions 2(9%)* 7(37%)* Nonfebrile seizures 3 (14%) 5(26%) Abnormal electroencephalogram 5(23%) 6(32%) Modified from Millichap et al.: J Pediat 56: 364, 1960 i. * Difference significant (P = 0.03). In this comparative study of therapies, the prophylactic value of continuous administration of phenobarbital was not superior to that of intermittent therapy alone; control of febrile seizures was poor and of questionable degree in the patients of both groups. Of 110 unselected patients with febrile seizures examined by the author at the same hospital, the majority had not been treated systematically with anticonvulsant drugs in the past and a recurrence of febrile seizures was recorded in only 54 per cent. Thus, the incidence of recurrence of the seizures observed in these patients was not greater than in those of the present series whose treatment was continuous or intermittent but well con- trolled. The doses of anticonvulsants used in this study were average, and at the times of subsequent illnesses patients in the group treated regularly with phenobarbital received twice the average initial dose and within the maximal range generally tolerated by infants and children. The results in children are con- firmed by laboratory studies in animals, which have demonstrated the need for large toxic doses of phenobarbital and the ineffective- ness of diphenylhydantoin in the prevention of febrile seizures. TREATMENT 123 Frantzen and associates (1964) evaluated the effect of long- term, prophylactic, antiepileptic treatment in febrile convulsions in a controlled study of 220 children under the age of 7 years, all with a first convulsion associated with an elevation of tempera- ture greater than 37.5° C (99.5°F). The children were grouped and treated as follows: 100 patients, born on odd dates, received phenytoin (Dilantin) 5 mg/kg daily and 106 patients, born on even dates, received no regular therapy. The period of follow-up was at least 12 months and in some cases more than 2 years. Febrile episodes occurred in 87 and 88 of the patients in each group during the periods of observation. Febrile convulsions re- curred in 18 (21 per cent) patients treated with diphenylhy- dantoin and in 20 (23 per cent) patients who received no regular therapy (Table 6-5). The results in this large group of patients Table 6-5. Recurrence of Febrile Convulsions in 100 Patients Treated Regularly with Diphenylhydantoin and 106 who Received no Prophylactic Therapy PATIENTS PATIENTS PROPHYLACTIC TOTAL DURATION WITH WITH ANTICONVULSANT PATIENTS OF RECURRENT RECURRENT THERAPY OBSERVED OBSERVATION FEVERS CONVULSIONS Diphenylhydantoin 5 mg/kg daily 100 1-2 yrs 87 18 (21%) No therapy 106 1-2 yrs 88 20 (23%) From Frantzen et al.: Ugeskr Laeg 126: 207, 1964. confirm those of our own study of a smaller number of patients; treatment with regular and continuous doses of anticonvulsants has no advantages when compared with intermittent forms of therapy, given only at the times of subsequent febrile episodes. The ineffectiveness of continuous anticonvulsant therapy in the prophylaxis of febrile convulsions has been reported by other authors (Table 6-2). In 1947, Livingston, Bridge, and Kajdi found a recurrence of febrile seizures in 15 (49 per cent) of 31 patients who had received irregular or no treatment and in 33 (52 per cent) of 63 patients who were treated regularly with anticonvul- 124 FEBRILE CONVULSIONS sant drugs. The type and dose of anticonvulsant were not speci- fied in this report. In subsequent publications, Livingston (1954, 1964) distinguished between the young child with simple febrile seizures and the patient over 5 years old with prolonged or focal seizures associated with abnormalities in the electroencephalo- gram. In the benign group of cases, daily administration of anti- convulsants was not advised, but for the complicated group, referred to as epileptic seizures precipitated by fever, Livingston advocated regular therapy for a period of at least 4 years. In Peterman’s study of 128 patients (1950), which included a rela- tively large number with complicating factors, the results of con- tinuous therapy were very discouraging, and febrile seizures recurred in 84 per cent of patients treated. Indications for Continuous Administration of Phenobarbital The indications for regular daily administration of pheno- barbital in patients with a history of febrile convulsions are as follows: 1. The occurrence of one or more spontaneous nonfebrile sei- zures. 2. A seizure associated with a minimal degree of fever of questionable significance. 3. A febrile convulsion of longer duration than 20 minutes and associated with (1) seizure discharges in an electroencephalo- gram obtained not earlier than 5 to 7 days after recovery from the convulsion, or (2) persistent abnormal neurologic signs and evidence of a structural cerebral lesion. Criteria and indications which are less well defined include a recurrence of febrile seizures at intervals of less than 3 months, focal seizure patterns, an onset after 5 years of age and febrile seizures in female patients. The above criteria are listed as guides to therapy, but recom- mendations will vary in relation to individual circumstances. Choice of Anticonvulsant Drug for Treatment of Febrile Convulsions The ideal compound for the treatment of febrile convulsions should elevate the threshold and prevent the seizure in doses which are relatively nontoxic. Since febrile seizures are likely to TREATMENT 125 occur only at the time of an acute febrile illness, the action of the drug should be rapid and allow intermittent administration, so that the necessity for long-term treatment with increased risk of side-effects may be avoided. A compound with anticonvulsant and antipyretic properties would be most desirable, and of bar- biturates with these properties phenobarbital is the most potent. Of drugs presently available, phenobarbital is recommended as the agent of choice, but restlessness and, in large doses, sedative side-effects detract from its effectiveness. In laboratory investigations of potential new agents for the prevention of febrile seizures, phetharbital (Fyrictal), mepro- bamate, (Miltown, Equanil), and trimethadione (Tridione) had higher protective or therapeutic indices against seizures induced by hyperthermia (Fig. 6-1), but, apart from phetharbital, the N-Phenylbarbi tal (B.W. No. 401) PROTECTIVE INDEX Trimethadione Phenobarbital Meprobamate Figure 6-1. Structural formulas and therapeutic indices of phe- tharbital (N-phenylbarbital, Fyrictal), meprobamate, trimetha- dione, and phenobarbital. The therapeutic index is a measure of the margin of safety and potential usefulness of a drug; it is ob- tained by dividing the dose which causes minimal signs of toxicity in 50 per cent of animals (TDSO) by that which protects 50 per cent of animals from the experimental febrile seizure (EDSO). 126 FEBRILE CONVULSIONS potency of these drugs was significantly less than that of pheno- barbital (Millichap et al., 1960 ii). Trimethadione may be contra- indicated because of the risk of leukopenia that attends its use, but the possible value of meprobamate in the treatment of febrile seizures warrants clinical investigation. Phetharbital, a nonofficial new drug, was developed as a potential specific therapy by means of a laboratory assay designed for screening compounds of various chemical structure (Millichap, 1960 ii). Of 60 new chemicals tested in animals, phetharbital was the most potent anticonvul- sant; its action was rapid in onset and toxicity was relatively low. The antipyretic activity of phetharbital, determined by the per- centage retardation of rate of rise of body temperature, was more regular and constant than that of phenobarbital, and nontoxic doses were effective (Fig. 6-2). As a result of these investigations and assay procedures in laboratory animals, phetharbital was recommended for clinical trial in patients with febrile and non- febrile seizures (Millichap, 1960 ii; Millichap et al., 1960 ii). -N-Phenyl barbital (B. W. No. 401) Rate of Rise in Body Temp. % Retardation Phenobarbital Meprobamate Trlmethadione Dose (mg/kg) Figure 6-2. Antipyretic effect of phetharhital (N-phenylharbital, Pyrictal) compared with conventional anticonvulsants in animals. X phetharhital, • phenoharbital, □ meprobamate, O trimetha- dione. (From Millichap et ah: Neurology (Minneap) 10: 575, 1960 ii.) TREATMENT 127 Clinical Evaluations of Phetharhital (Pyrictal), The antipy- retic activity of phetharhital was examined by the author and co-workers in 71 infants and children aged 3 months to 8 years admitted to hospital with fever greater than 38.3° C (101°F) caused principally by acute upper respiratory tract infection and otitis media. The average single dose of phetharhital administered orally was 23 mg/kg; the range was 14 to 30 mg/kg. The body temperature was recorded rectally before, and 1 hour after, ad- ministration of the drug and a control group of febrile patients was observed for 1 hour before administration of antipyretic therapy. A reduction of fever occurred in a significantly greater proportion of children treated with phetharhital than that ob- served in the control group of patients. Small doses of 2 to 5 mg/kg were without effect, whereas doses of 15 to 30 mg/kg were asso- ciated with a fall in temperature. In 20 institutionalized patients with major, minor, or combined seizures, previously refractory to conventional drugs, a reduction in the incidence of major seizures of 75 per cent or more was ob- tained in 53 per cent of cases and a similar reduction in incidence of minor seizures was obtained in 90 per cent of patients. Myo- clonic, opisthotonic, and absence seizures were controlled more effectively than the major tonic or tonic-clonic pattern and a com- bination of diphenylhydantoin and phetharhital was necessary for satisfactory treatment of major epilepsies. The combined antipyretic and anticonvulsant properties of phetharhital were examined in 14 infants and children with a previous history of febrile seizures. The drug was administered orally in doses of 20 mg/kg at the onset of subsequent infections, and smaller doses were given at intervals of 8 hours until the temperature was less than 38.3° C (101°F). Febrile seizures did not recur in the 14 patients treated intermittently with phetharbi- tal, whereas the incidence of recurrence in those who had received phenobarbital intermittently or continuously on previous occa- sions was 48 per cent. Between one and three febrile episodes occurred in each child during the trial of phetharhital. Side-effects were not observed with single doses of 14 to 30 mg/kg admin- istered at the time of fever. In patients with epilepsy the drug was given as a supplement to other medications, and a critical 128 FEBRILE CONVULSIONS evaluation of minor side-effects was precluded by the low men- tality of the patients treated; symptoms possibly related to phe- tharbital included restlessness in one patient, anorexia in one, and increased drowsiness in four. The apparent correlation of results in laboratory experiments and in patients should encourage the continued use of these methods in evaluation of new medica- tions, and further trials of phetharbital in patients with febrile seizures seemed warranted. A clinical trial of phetharbital was conducted by C. H. Carter (1962) in 75 institutionalized epileptic patients whose seizures had not been controlled completely by previous medication. Most of the patients had both grand mal and minor epilepsies. Records of the numbers of seizures were available on all patients for at least 12 months prior to therapy and phetharbital was admin- istered for a 12-month period. The dose of phetharbital, 50 to 100 mg every 6 to 8 hours initially, was increased to as much as 500 mg every 6 hours in some patients, and all previous antiepi- leptic drugs were maintained at the same dosage. A total of 260 grand mal seizures per month occurred in the 75 patients during the pretrial control period and only 24 seizures per month were observed during the period of the phetharbital trial. Fifty-eight patients were completely free from seizures and 16 had their seizures partially controlled. Of 49 patients with spike-and-wave discharges in the electroencephalogram, pretrial records showed an average of 12 spike-and-wave bursts per hour of record, and after therapy with phetharbital for 6 months the average number of spike-and-wave complexes per hour was reduced to four. The anticonvulsant activity of phetharbital appeared to be confirmed in these patients with previously refractory nonfebrile seizures. Optimum Duration and Withdrawal of Anticonvulsant Therapy In patients with febrile seizures who require regular anti- convulsant therapy, the optimum time to discontinue treatment is difficult to determine. Factors of importance include: (1) oc- currence and incidence of spontaneous febrile seizures, (2) the findings on neurologic examination and in the electroencephalo- graphic record, and (3) the age of the patient. Regular anti- convulsant treatment should be continued in patients with a his- TREATMENT 129 Tory of frequent nonfebrile seizures and persistent neurologic or electroencephalographic evidence of a structural cerebral lesion. Withdrawal of medication may be advised 2 years after the last seizure if the neurologic examination and electroencephalogram are normal. Phenobarbital and other antipyretic measures at the time of subsequent febrile episodes should be employed up to 5 years of age, although their value in the prevention of febrile seizures may be limited. Continuous anticonvulsant therapy is intended mainly for the prevention of nonfebrile spontaneous seizures; it is recommended in patients with a history of com- plications and spontaneous seizures or neurologic and electro- encephalographic evidence of a low threshold to seizures. Medications should be withdrawn whenever possible before the child reaches school age. In older children, the emotional disturbance and social ostracism which attend recurrence of a seizure must be considered and weighed against the minor in- conveniences of prolongation of therapy. When withdrawal of anticonvulsants has been successfully accomplished, the patient should be advised to take phenobarbital at times when factors known to increase susceptibility to seizures may be experienced. In addition to fever, the excessive exposure to heat, sunlight, and fatigue, certain immunization procedures, the administration of antihistamines, phenothiazines, and penicillin, and emotional dis- turbances are known to precipitate seizures in susceptible pa- tients. The threshold to seizures is lowest and the withdrawal of therapy most hazardous at 7 to 10 years of age and at puberty. Two methods of withdrawal of medications may be offered ac- cording to individual circumstances: (1) The patient may be ad- mitted to the hospital and the effects of sudden withdrawal of medication observed under close and expert supervision. In pa- tients with a low threshold to seizures, sudden withdrawal of anticonvulsants would result in a recurrence of seizures and par- enteral doses of phenobarbital could be used to obtain rapid control in these circumstances. The ability of the patient to with- stand this mode of withdrawal without seizure recurrence is in- dicative of a high or normal threshold, and the continuation of anticonvulsant drugs would be unnecessary. (2) Alternatively, and when the parents are reliable and live close enough to a 130 FEBRILE CONVULSIONS hospital or physician, the gradual withdrawal of medications may be preferred. The dose of phenobarbital is reduced by one half for 1 month and to one quarter of the original amount for a fur- ther month; complete withdrawal is advised if seizures have not recurred. In the general management of the patient, some limita- tions of physical activities are mandatory for the safety of the patient, but unnecessary restrictions should be avoided. Parents should be informed concerning the nature of seizures and their prognosis; the tendency for a child with seizures to be rejected is minimized if the fears and anxieties of parents are dispelled. Indications and Methods for Intermittent Prophylactic Therapy The principal aim in therapy is the prevention of a rise in body temperature above the threshold level at which seizures have occurred previously. Specific forms of treatment include (1) anti- pyretics to inhibit heat production, (2) anticonvulsants to raise the threshold convulsive temperature, and (3) antibiotics to combat infection when indicated. Aspirin, phenobarbital, and diphenylhydantoin, the therapeutic agents most commonly used, have been found inadequate in controlled studies; nevertheless, antipyretic measures and some form of anticonvulsant treatment are indicated and the methods of their administration should be outlined in detail to parents for optimum results. The indications for intermittent therapy are (1) a history of previous febrile seizures, shorter in duration than 20 minutes, in- frequent in occurrence, and uncomplicated by spontaneous sei- zures and other neurologic or electroencephalographic abnor- malities, and (2) early signs and symptoms of infection or a body temperature of 37.8° C (100°F) or higher. Choice of Anticonvulsant Drug. Among compounds readily available, phenobarbital is the agent of choice. An initial dose of 30 mg per year of age up to a maximum of 120 mg should be given orally at the earliest signs of infection, and a dose of 5 mg/kg daily in four equal parts at intervals of 6 hours should be con- tinued until the fever has subsided. Antipyretic Therapy. Barbiturates cause a slight fall in normal body temperature owing to lessened activity and depression of TREATMENT 131 the central-temperature-regulatory mechanism, and the effect is proportional to the dose (Goodman and Gilman, 1955). Short- acting barbiturates, pentobarbital and secobarbital, should be more effective as antipyretics than the long-acting analogues, phenobarbital and mephobarbital (Mebaral); in uncontrolled studies a suppository of pentobarbital has prevented the recur- rence of febrile seizures in a small number of patients previously treated unsuccessfully with phenobarbital. In anesthetic doses all barbiturates employed clinically are capable of inhibiting convul- sions, but phenobarbital has a selective anticonvulsant action shared only by mephobarbital, metharbital, and phetharbital (Goodman and Gilman, 1955; Millichap, 1965). Thus, the ap- parent prevention of a febrile seizure by pentobarbital may be related more to an antipyretic effect than anticonvulsant activity, and a single rectal suppository of pentobarbital should be sup- plemented by intermittent oral doses of phenobarbital until the fever and risk of convulsions have subsided. Suppositories that melt at body temperature are more effective than those which depend on solution for their action, but uniform results cannot be expected because of constipation or diarrhea which frequently complicate a febrile illness. Aspirin is the most frequently used antipyretic drug, but over- dosage results in poisoning, especially when fever is associated with dehydration. Heat production is not inhibited, but heat dissi- pation is augmented by increased peripheral blood flow and sweat- ing. The initial stimulatory effect of toxic doses of salicylates and the hyperventilation with respiratory alkalosis which occur with therapeutic doses may increase susceptibility to seizures. Aceta- minophen (Tempra, Tylenol) is said to be less toxic than salicy- lates but more experience is needed to confirm this claim. In laboratory studies of antipyretic agents, aspirin, and acetophe- netidin (Phenacetin) failed to retard the rate of temperature rise induced by radiotherm diathermy in animals, and aspirin in doses of 600 mg/kg lowered the threshold convulsive temperature and exacerbated the febrile seizure. In summary, antipyretics such as aspirin may be given in small doses to facilitate heat loss by pro- motion of sweating and to relieve discomfort attending the illness, 132 FEBRILE CONVULSIONS but large and toxic doses should be avoided. Tepid baths, rubbing the skin with alcohol, and administration of barbiturates are more reliable measures for the control of a rising temperature. REVIEW AND ANALYSIS OF RECOMMENDED THERAPIES Apart from one paper by Turnbull (1955) concerning tonsil- lectomy in patients with febrile seizures, a proposed method of prophylaxis of doubtful value, contributions to the literature deal- ing with treatment consider principally the indications and results of administration of anticonvulsant drugs. Of 25 publications dealing with treatment of febrile convulsions, only four present original data in support of opinions and recommendations (Table 6-2). The results of these studies, of which two were controlled and prospective in design, showed that continuous treatment with anticonvulsant drugs, either phenobarbital or diphenylhy- dantoin, failed to prevent recurrence of febrile seizures and was not superior in efficacy to the intermittent form of therapy given at times of subsequent infections (Livingston et al., 1947; Peter- man, 1950; Millichap et al., 1960 i; Frantzen et al., 1964). A total of 475 patients was included in these trials, and periods of ob- servation varied from 6 months to 25 years. Recurrence of febrile seizures was reported in an average of 57 per cent of those treated continuously and in 33 per cent of patients who received anti- convulsants only at intervals or not at all. The failure and inadvisability of continuous anticonvulsant therapy for the prevention of uncomplicated febrile seizures was mentioned in 17 different publications, and the universal pre- scription of this treatment was recommended in eight publications from six separate centers (Table 6-2). Those authors who advise continuous therapy for all patients do not provide supporting data with details of specific drug trials, and some admit that the recommendation represents an unconfirmed opinion based on personal experience of isolated cases. The high proportion of patients with complications referred to a neurology specialty clinic may have prejudiced these authors against a more conservative treatment schedule based on clinical TREATMENT 133 and electroencephalographic findings in each individual patient. The duration of therapy advised varies from a minimum of 1 month (Fois and Malandrini, 1957) to a maximum of up to 6 years of age (Hammill and Carter, S., 1966). M. A. Lennox (1949) advises regular administration of diphenylhydantoin to all chil- dren with febrile seizures up to 3 years of age, and Chao and associates (1958) continue therapy for not less than 1 year in patients with a single recurrence of a febrile convulsion. The importance of the electroencephalogram in determining the need for continuous anticonvulsant therapy is stressed by many authors who favor a discriminatory approach to treatment (Gallant and Livingston, 1949; Dobrzynska, 1963; and Horst- mann and Schinnerling, 1963). Some list a family history of spontaneous seizures or an abnormal electroencephalogram as an indication for continuous therapy (Bjerglund and Brandt, 1954); others defer treatment until the third recurrence of a febrile con- vulsion and include an unusually prolonged or asymmetrical sei- zure as sufficient indication (Prichard and McGreal, 1958); and the majority consider the occurrence of one or more spontaneous seizures as an absolute criterion for regular administration of anticonvulsant drugs. Costeff (1965) dissents from this recom- mendation and reserves this form of treatment for patients more than 4 years old with seizures which are focal, prolonged, more than twice recurrent, or associated with other evidence of cerebral dysfunction, regardless of the body temperature at the time of their occurrence. Costeff studied the natural history of febrile and nonfebrile seizures in children who attended for repeated health examina- tions at a clinic in Beer-Sheva, Israel, between 1962 and 1963. Of 500 children examined between 3 and 5 years of age, 63 (12.6 per cent) had suffered at least one seizure; in 15 (3.0 per cent) patients the seizure occurred with fever. Seizures had been brought to previous medical attention in one half the total group of patients and in two thirds of the group with febrile convulsions. Only one of the 63 patients had received prophylactic anticonvul- sant medication, and the frequency of recurrence was the same in children with febrile or nonfebrile seizures. Costeff concludes that the risk of epilepsy in adult life following a single seizure, either 134 FEBRILE CONVULSIONS febrile or nonfebrile, in early childhood is less than 5 per cent. He stresses an excellent prognosis for seizures in children under 5 years of age, irrespective of an association with or without fever, and advises against the universal use of regular anticonvulsant therapy. The impracticality of treatment initiated after a first or even a second febrile or nonfebrile seizure is considered in regard to the incidence of seizures in childhood; according to CostefFs observations, the number of patients with seizures is so large as to make treatment an almost normal phenomenon. The incidence of convulsive disorders in young children re- ported by Costeff has been questioned because of the possible inclusion of patients with breath-holding attacks and without true seizures (Huttenlocher, 1966). If nonconvulsive seizures are excluded, the adjusted incidence of 6.6 per cent is comparable to that reported by Thom (1942) and Miller and associates (1960). These authors, in their surveys of young children in Boston, Massachusetts, and in Newcastle, England, did not distinguish between febrile and nonfebrile seizures, but the prevalence of convulsive disorders at this age period is many times greater than the incidence in the adult population, which is estimated at 0.2 to 0.5 per cent. Similar epidemiological studies in adults might reveal a true incidence of epilepsy of much greater magnitude than that apparent from the number of patients who attend for medical treatment. Nevertheless, on the basis of statistics avail- able, the number of adults with seizures is considerably smaller than the incidence in children; of the population under 5 years of age in the United States, approximately one and one-half million have seizures and one third of this number have febrile seizures. If the likelihood of these children developing epilepsy in later life is less than 5 per cent, the need for prolonged anti- convulsant therapy after the first seizure, an almost unanimous recommendation among authors in regard to nonfebrile seizures, seems questionable. In contrast to the published opinions of neurologists and other specialists, that most or all seizures in childhood warrant pro- phylactic anticonvulsant therapy, in practice it seems that anti- convulsants are rarely prescribed by physicians or administered by parents for long periods to young children (Miller et al., 1960; TREATMENT 135 Costeff, 1965). The general practitioner and pediatrician, who obviously favor a more individual approach to therapy, may rightfully question the evidence in support of universal use of anticonvulsant drugs, particularly in children with uncomplicated febrile seizures. Controlled prospective studies have demonstrated that febrile seizures which are prolonged and associated with electroencephalographic abnormalities or persistent neurologic defects have a poor prognosis and warrant continuous anticon- vulsant therapy. Short, infrequent febrile seizures without electro- encephalographic evidence of epileptic activity do not require regular prophylactic treatment and, in children under 5 years of age, the arguments against a similar recommendation for un- complicated nonfebrile seizures are limited and unsupported by significant data. An increased incidence of recurrent nonfebrile seizures may be expected in patients with prolonged or frequent febrile seizures and electroencephalographic abnormalities, symmetrical or focal in type; the significance of the focal clinical seizure pattern and other complications suggestive of a poor prognosis has not been proved statistically. From a therapeutic standpoint, each child with febrile convulsions should be considered individually; con- tinuous prophylactic anticonvulsant medication should be re- stricted to patients with clinical and electroencephalographic evidence significant of a potential recurrent seizure disorder. SUMMARY Treatment of the initial febrile convulsion includes pheno- barbital to control the seizure, tepid baths to facilitate heat loss, and antibiotics to combat bacterial infection. Excessive doses and needless treatment with penicillin should be avoided and electro- lyte imbalance corrected. Prophylactic anticonvulsant treatment is of two types—intermittent and continuous. The principal aim in therapy is the prevention of a rise in body temperature above the threshold convulsive level for the individual patient. Inter- mittent therapy consists of pentobarbital given rectally at the onset of subsequent infection with fever and phenobarbital orally until the fever subsides. Continuous daily administration of phe- 136 FEBRILE CONVULSIONS nobarbital is indicated in patients whose initial febrile convulsion is prolonged and associated with seizure discharges in the electro- encephalogram or persistent abnormal neurologic signs indicative of a structural cerebral lesion. The occurrence of one or more spontaneous nonfebrile seizures is considered an absolute reason for regular therapy by most authorities, but, in children under 5 years of age, the recommendation may be modified by individual circumstances. Indications for continuous anticonvulsant treat- ment that are less well defined include frequent recurrence of febrile-seizures, focal seizure patterns, onset after 5 years of age, and febrile seizures in female patients. In the absence of com- plications, medications are withdrawn before the child reaches school age whenever possible. CHAPTER 7 EXPERIMENTAL FEBRILE CONVULSIONS, ARTIFICIAL FEVER, AND HYPERPYREXIA STUDIES IN ANIMALS A satisfactory model of the febrile convulsion in experimental animals is difficult to design because the definitive etiology of the clinical seizure is undetermined. The importance of fever, par- ticularly the height of the fever, is reasonably well documented in a few controlled studies in patients and the restriction of the fe- brile seizure to a limited age period is undisputed. The significance of infection in the etiology of febrile seizures is incompletely defined; present evidence suggests that the role is nonspecific and related chiefly to the associated febrile reaction. The influ- ence of toxins has not been investigated fully, but a neurotoxin associated with some Shigella infections has occasionally been cited as a cause. Other factors which may be contributory include water and electrolyte imbalance, immune reactions to the toxic products of infecting organisms, allergy to medications, previous brain injury, and some genetically determined enzyme defect. In the majority of experimental studies in animals febrile convulsions have been induced by artificial hyperthermia, and radiant heat, 137 138 EXPERIMENTAL FEBRILE CONVULSIONS radiotherm diathermy, or the intravenous injection of pyrogenic substances has been employed. Methods of Experimental Induction of Hyperthermia Fever induced by a hot environment or by radiotherm dia- thermy has been found more satisfactory and easier to control than that which results from injections of pyrogens. The experi- mental production of hypothalamic lesions, another known cause of fever, has not been employed in the investigation of febrile convulsions. Radiant Heat. A radiant heat chamber was used by Wegman (1939) in his investigation of factors that might influence con- vulsions caused by hyperthermia. The apparatus consisted of an elliptical cylinder of copper with a heating coil along the entire length of the upper focus. At the other focus of the ellipse was a removable slab of wood perforated to allow the animal’s legs to be inserted and secured with leather thongs. The temperature was taken by a clinical thermometer inserted in the rectum and maintained in position by means of a wire spring. The electric input was 110-volt direct current, and a resistance of 20 ohms in series with the coil permitted modifications of the rapidity of rise of temperature. Two additional methods of induction of fever were tried by this investigator but were abandoned in favor of the radiant heat chamber. One apparatus consisted of an ordinary bacteriologic incubator and the second was a high-frequency generator with an electromagnetic field established between two large electrode plates. The major difficulties encountered with these methods were the failure to permit variation and control of the rapidity of temperature rise and the generation of current with sparking when the animal came in contact with the moisture of urine, feces, saliva, and excessive lachrymation. The present author attempted to use a radiant-heat chamber similar to that employed by Wegman, but the method was con- sidered unsatisfactory for several reasons: (1) the necessity to tether and restrain the animal precluded an evaluation of the patterns of febrile convulsions, produced modifications of the seizure patterns, introduced the probability of changes in the re- sponses to fever by increased muscular activity associated with ARTIFICIAL FEVER AND HYPERPYREXIA 139 struggling, and prolonged each experiment so that large numbers of animals with observations susceptible to statistical analysis could not be employed; (2) extreme heat of the external environ- ment could cause complications and reactions in the animal dif- ferent from those which attend endogenous fevers; (3) the method of monitoring the temperature of the animal inside the chamber proved difficult; and (4) the maximum rise of temperature at- tained in larger animal species was insufficient to induce a con- vulsion, so that the relative importance of the height of the body temperature and the rapidity of rise of temperature could not be determined. Microwave Diathermy. In the author’s experiments, the micro- wave diathermy generator proved the most satisfactory method for the controlled induction of hyperthermia in animals (Milli- chap, 1959). The power output, indicated on a milliammeter calibrated in per cent, is controlled by a variable autotransformer, and the selected dosage of radio frequency energy is applied by means of a coaxial power cable and director (Fig. 7-1). The body temperature of the animal is recorded rapidly and simply with a telethermometer and thermistor hypodermic probe inserted into the deep subcutaneous tissue of the back. The temperature re- mains constant for approximately 30 seconds after the animal is removed from the testing chamber, and errors in reading are avoided since the time constant of the thermistor probe is 1 to 2 seconds. The advantages of microwave diathermy compared to radiant heat are as follows: (1) the tissues of the animal are heated without change in the environmental temperature; (2) the ani- mal is unrestrained and is exposed to the heating stimulus in a glass chamber, so that observation of the behavior and seizure patterns is facilitated; (3) the radiofrequency energy is suffi- ciently powerful to induce maximal tolerated temperatures in small animal species and a body temperature of sufficient height in kittens to induce convulsions in all animals tested; and (4) the rate of rise of temperature may be varied in proportion to the intensity of the microwave current and compared in animals of different species, ages, and weights and in relation to other vari- able factors. In order to attain fever of sufficient degree to induce 140 EXPERIMENTAL FEBRILE CONVULSIONS Figure 7-1. Microwave diathermy generator for the induction of hyperthermia and febrile convulsions in experimental animals. {From Millichap: Pediatrics 23:76, 1959.) convulsions in larger animal species such as the monkey, a dia- thermy generator with a maximum power output greater than that employed in the author’s experiments would be necessary; other- wise, the rate of heat loss would exceed the heat gain, and a body temperature of sufficient height to induce a seizure would not be attained. A similar problem was apparently encountered in ex- periments in monkeys when the induction of hyperthermia was attempted with an electric heating pad. Electric Heating Pad. This method, employed by Schmidt and co-workers (1956), produced an elevation of temperature to 40.6° C (105.0°F) in young monkeys, and a body temperature of 42.2° C (108°F) was attained in two animals. The heating pad was applied to the restrained animal wrapped in woolen blankets, and the temperature elevation was insufficient to activate an epi- leptogenic cortical focus. ARTIFICIAL FEVER AND HYPERPYREXIA 141 Hot Water Baths. Fever has been induced in dogs by immer- sion to the shoulders in water at a temperature of 50° C (122°F). Richards and Egdahl (1956) used this method in a study of the effects of acute hyperthermia on adrenal cortical function in anesthetized dogs. A rapid increase in rectal temperature to 42° C (107.6°F) in 15 to 20 minutes was accompanied by a two- to sevenfold increase in adrenal corticoid output and a 10 to 20 per cent increase in adrenal venous blood flow. When a rectal temperature of 44 to 45° C (111 to 113°F) was reached, circula- tory failure occurred and a concomitant decrease in adrenal corticoid output and venous blood flow ensued. The necessity for anesthesia would preclude the use of this method in investigations of febrile convulsions in animals. However, hot water applied to the abdomen by means of a bag has been found to induce fever and changes in the electrical activity of the cortex of curarized adult cats and kittens, but the height of the temperature attained in these experiments was not noted in the published abstract (Kashiwase, 1962). Insulated Heating Cabinet. Swinyard and Toman (1948) in- duced hyperthermia in rats by placing the animals, restrained in circular wire-mesh holders, in an insulated heating cabinet at 55° C (131°F). Rectal temperatures were recorded with a mer- cury thermometer immediately before and after experimental seizures, and the maximal body temperature attained by this method was 43° C (109.4°F). The experiments were designed to investigate the effects of alterations in body temperature on the properties of convulsive seizures induced by electroshock, pen- tylenetetrazol, and picrotoxin. Steam Heat. Krakau and Nyman (1953), in their study of the effects of fever on the electroencephalogram in man, used a steam chamber which enclosed the subject apart from his head. The temperature inside the chamber is maintained by a thermostat and the body temperature is recorded continuously by a rectal thermocouple. Injections of Pyrogens. Pyrogens are metabolic products of microorganisms which are active even in submicrogram quanti- ties; they are relatively heat stable and are not affected by steril- ization. Endogenous pyrogens are derived from granulocytes of 142 EXPERIMENTAL FEBRILE CONVULSIONS acute inflammatory exudates and from leukocytes damaged by endotoxins of gram-negative organisms and antibodies (Bennet and Beeson, 1953). In patients with “Salvarsan fever,” “milk fever,” and artificial hyperthermia induced in the past by injec- tion of other foreign substances, the elevation in body tempera- ture was caused by pyrogenic contaminants of bacterial origin. In addition to bacteria, certain viruses, molds, and a yeast have been found to produce pyrogens. Highly purified pyrogens of a polysaccharide nature have been separated from bacterial sources, and increasing values of reducing sugar and decreasing amounts of nitrogen in the preparations serve as criteria for evaluating comparative purity. The analysis of some purified pyrogens was reported as follows: reducing sugars after hydrolysis, including aldohexose, glucosamine, and methylpentose, 65 to 70 per cent as glucose; lipids, 16 per cent; phosphorus, 1.1 per cent; and nitrogen, 2.2 per cent. The minimum pyrogenic dose of this sub- stance was 0.005 p,g/kg body weight. Bacterial pyrogens are re- ported to elicit temperature elevations in men, dogs, and rabbits but not in mice, rats, guinea pigs, or chicks; in some studies man was more sensitive and in others less sensitive when compared to rabbits. The response to pyrogens is modified by the sex and the resting normal temperature; male animals and animals with a relatively low resting temperature have the highest temperature elevations following the injection. Tolerance occurs following repeated injections of pyrogens in the rabbit (Berger et ah, 1956). Ginger, Windle, and Johnson (1952) have compiled an an- notated bibliography of a specially prepared bacterial pyrogen, Piromen, and its use in the treatment of various diseases. This preparation may be used for the induction of hyperthermia in animals. However, the present author found that the response to Piromen in small animal species was inconstant and the tempera- ture elevation was insufficient to induce convulsions. Baird and Garfunkel (1956), in an investigation of the effects of fever on the electroencephalogram in children, induced hyperthermia by the intravenous administration of 1.0 ml of a typhoid vaccine containing 100 million organisms per ml; a maximum rise in temperature to 40.0 and 40.9° C (104.0 and 105.6°F) was obtained in approximately 3 hours after the injection. ARTIFICIAL FEVER AND HYPERPYREXIA 143 Patterns of Experimental Febrile Seizures The author has reported five stages of behavior and seizure activity in response to artificial hyperthermia induced by micro- wave diathermy in small animal species; stage 1, intermittent exploratory behavior, occasional running, and rubbing of the face with forepaws; stage 2, head tremors and occasional myoclonic jerks of the limbs; stage 3, generalized clonus with loss of posture, preceded by a sudden opisthotonic spasm; stage 4, tonic flexion followed by extension of forelimbs; stage 5, mild preagonal tonic extensor spasms of hindlimbs, flexion and, rarely, extension of forelimbs, and subsequent mild clonus of distal parts of limbs. The clonic, stage 3 seizure pattern is accompanied by salivation and urination and the tonic, stage 5 seizure is complicated by a nasal discharge of frothy and, sometimes, hemorrhagic fluid and cyanosis, immediately before death. Influence of Age on Febrile Seizure Patterns. Changes in the patterns of nonfebrile seizures in relation to age have been demonstrated in young rats with electroshock-induced seizures (Millichap, 1957), and similar changes are observed in fever- induced seizures in rats between 1 and 35 days old. In rats under 12 days old, clonus is milder and asymmetrical and tonic spasms less complete and fewer than in older animals; after 30 days of age, the clonic stage of the seizure is not preceded by opisthotonic spasm and loss of posture; and in adult rats, seizures in response to fever are absent or consist only of mild tonic extension of the forelimbs and clonus immediately before death. In all experi- mental febrile seizures, the tonic extensor component of fore- and hindlimbs occurs much less frequently and in milder degree than that of maximal electroshock seizures. In mice and rats between 12 and 28 days old, the generalized clonic, stage 3 seizure is most constant and well defined; and this pattern offers the clearest end-point in timing the onset of the febrile seizure in experimental studies. Species Differences in Febrile Seizure Patterns. In the guinea pig, the tonic component of the febrile seizure is absent in both young and adult animals. Clonus is symmetrical in newborn guinea pigs, and asymmetrical clonus occurs in animals between 144 EXPERIMENTAL FEBRILE CONVULSIONS 20 and 38 days old. The absence or inconstancy of tonic exten- sion of hindlimbs is observed also in maximal seizures induced by electroshock in the guinea pig (Millichap, 1957). In kittens 7 to 14 days old, the febrile seizure pattern is mainly clonic and asymmetrical and tonic spasm is not observed (Millichap, 1959). Convulsions with hyperthermia have been induced in alligators, and susceptibility varied in different species (Almeida et al., 1950). Experimental Modification of Febrile Seizure Patterns. Fac- tors known to alter experimental febrile seizure patterns include drugs, electrolyte imbalance, and section of the spinal cord. Steinschneider and co-workers (1964), using the method of fever induction introduced by the author (Millichap, 1959), studied the effects of transverse section of the spinal cord on the pattern of the febrile seizure in 21-day-old weanling rats. Whereas the neurologically intact animal responded to hyperthermia with a generalized clonic-tonic but mainly clonic convulsion affecting all limbs equally, section of the spinal cord at the midthoracic level resulted in differences in the seizure patterns in the upper and lower limbs but not an abolition of seizure activity in the hindlimbs, despite the resultant flaccid paraparesis. The febrile seizures were entirely tonic in pattern in the forelimbs and clonic in the hindlimbs; the temperature at the onset of convulsions was not recorded. No single hypothesis was offered to explain the apparent change in seizure patterns following cord transection, but a diminution of the inhibitory influence normally generated by higher centers upon the midbrain was invoked as a likely explanation for the tonic seizure induced by hyperthermia. The investigation of febrile seizures in animals with cord transection is interesting and of value in the elucidation of the physiology of seizures in general. However, in the interpretation of the differences in patterns observed in fore- and hindlimbs, it must be remembered that the pattern and severity of the febrile convulsion are dependent on the height of the temperature, and the occurrence of tonic flexor and extensor spasms of the forelimbs is inconstant in the intact rat. In animals with spinal cord tran- section, the finding of sustained tonus in the forelimbs may indi- cate exacerbation of the seizure activity above the site of the cord lesion, and a purely clonic pattern in the hindlimbs may represent a mitigation of the response below the lesion caused by ARTIFICIAL FEVER AND HYPERPYREXIA 145 spinal shock. In support of this explanation for the reported changes in the febrile seizure patterns in these animals, clinical experience has shown that the duration of spinal shock following injury to the cord is variable (Merritt, 1963), and an exaggera- tion of deep-tendon reflexes may be noted in the upper limbs of infants with injuries to the lower cervical and upper thoracic spinal cord sustained at birth. Ruch and Watts (1934, 1935) have alluded to reciprocal changes in reflex activity of forelimbs in- duced by postbrachial “cold-block” of the spinal cord; this feed- back or Schiff-Sherrington phenomenon is mentioned as a possible explanation for the apparent changes in the patterns of the febrile seizure in animals with cord transection (Steinschneider et al., 1964). The role of spinal reflex mechanisms as determinants of patterns of seizures has been elucidated by Esplin and Laffan (1957); these investigators demonstrated that the hindlimb tonic ex- tensor component of the maximal electroshock seizure is abolished by section of the lumbosacral dorsal roots, and the flexor-extensor sequence of seizure patterns represents the response of the spinal cord to an intense supraspinal barrage. In an unanesthetized cat with spinal section above C.l, massive stimulation of the spinal cord through an electrode inserted longitudinally from C.l to C.4 resulted in a sequence of motor events identical to those observed in an intact cat during a maximal electroshock seizure; the seizure consisted of an initial latent phase, a brief tonic flexion, and an abrupt change to an extensor pattern. If the frequency of stimu- lation was increased from very low to high values, the motor re- sponses began as clonic movements, later became tonic-flexor, and finally extensor in type. Extensor-tonic seizures, flexor-tonic seizures, and pure clonic seizures appeared to represent the responses of the spinal cord to supraspinal discharges of diminish- ing intensities. The seizure patterns observed in these experiments in cats were similar to those induced by various electroshock stimuli in intact rats (Millichap, 1957). Threshold to Experimental Febrile Seizures A threshold to febrile seizures dependent on the height of the body temperature has been demonstrated in animals of various species. In mice subjected to different intensities of heat energy, EXPERIMENTAL FEBRILE CONVULSIONS 146 between 30 and 100 per cent of maximum power output, the mean body temperature at the onset of the clonic, stage 3 seizure was constant at 43.2° C (109.7°F) while the duration of the heating stimulus varied from 70 to 393 seconds (Fig. 7-2), In mice sub- jected to weaker intensities of current, the highest temperature recorded was 40° C (104.0°F) and convulsions did not occur; it was necessary to raise the body temperature to 43.2 ± O.5°C (109.7°F) in order to induce a generalized clonic seizure. The threshold convulsive temperature was independent of the in- tensity and duration of the heating stimulus and the rate of rise of body temperature; the temperature of the animal at the onset of clonus was constant at 43.2° C although the rate of rise of tem- perature varied from 0.8 to 4.4° C per minute. Similarly, in two groups of kittens which convulsed in response to stimuli of weak and strong intensities, the rapidity of temperature rise differed significantly, whereas the convulsive temperature was the same in both groups. An accurate parameter for measurement of the r _ . . / \ Convulsive Time of Convulsion (sec) Temp (°F) Intensity of Heating Current {%) Figure 7-2. Convulsive temperature in mice in relation to dura- tion and intensity of heating current. (From Millichav: Pediatrics 23:76,1959.) ARTIFICIAL FEVER AND HYPERPYREXIA 147 threshold to febrile seizures in animals was provided by this experimental method. Influence of Age on Febrile Seizure Threshold. The different body temperatures at the onset of clonic convulsions in rats and guinea pigs from birth to 21 months of age are shown in Fig- ure 7-3. Rats were most susceptible to fever-induced seizures between 3 and 8 days of age, and at 1 day and above 8 days of age higher convulsive temperatures were observed. A rapid rise in threshold occurred at 10 days and again at 33 and 35 days of age. That the threshold convulsive temperature is related to the stage of maturation is shown by subsequent experiments in guinea pigs, an animal more mature than the rat at birth; in the newborn guinea pig the threshold convulsive temperature was 43.5° C (110.3°F) and was equal to that of the 20-day-old rat. Range of Temperature Elevation and Febrile Seizure Thresh- old. Modifications of the resting body temperature had no effect on the threshold convulsive temperature. In mice with tempera- tures lowered by means of chlorpromazine or by exposure to low ambient temperatures, the threshold convulsive temperature was Guinea Pigs Rats Convulsive Temperature (°F) Age in Days Figure 7-3. Convulsive threshold temperature of young animals in relation to age. (From Millichap: Pediatrics 23: 76, 1959.) 148 EXPERIMENTAL FEBRILE CONVULSIONS not different from that of controls with normal resting tempera- tures (Fig. 7-4). Animals subjected to fever with convulsions subsequently became hypothermic; reapplication of the micro- wave diathermy induced a second seizure and the threshold con- vulsive temperature was higher but not altered significantly in this experiment. The duration of the heat stimulus and the range of temperature change did not influence the febrile seizure threshold. Water and Electrolyte Metabolism and Febrile Seizure Thresh- old. The effects of changes in the balance of water and electro- lytes on brain excitability have been demonstrated in children with grand mal seizures (McQuarrie, 1929) and petit mal (Milli- chap and Jones, 1964) and also in animals with seizures induced by electroshock (Swinyard, 1949; Millichap et al., 1958). Yannet Basal and Threshold Convulsive Temperature (°F) Control Fever Chlor- Low Very Low Group Stlm, promazine -Ambient- Repeat. Treated Temp, Temp. Figure 7-4. Convulsive threshold temperature in mice with lowered resting bodti temperatures. (From Millichap: Pediatrics 23: 76,1959.) ARTIFICIAL FEVER AND HYPERPYREXIA 149 and Darrow (1938) examined the distribution of electrolytes in the brains of cats subjected to hyperthermia; they found a shift of water unaccompanied by cations from the cell to the extra- cellular fluid of the brain, and the alteration in the distribution of water could not be correlated with the occurrence of- febrile seizures. In the authors studies (Millichap, 1960 i) the applica- tion of the fever stimulus for determination of the seizure thresh- old was short and nonlethal, and the concentrations and distribu- tion of electrolytes were measured in control animals not sub- jected to hyperthermia. The influence of the fever and convulsion per se on the brain chemistry was not a complication in these experiments, and changes in the seizure threshold with age and other factors could be correlated with biochemical maturation or imbalance. The correlations obtained in developmental studies in young rats were confirmed by observations in adult animals in which the seizure threshold was lowered experimentally by the intraperitoneal injection of 5.5 per cent glucose solution and con- sequent production of hypoelectrolytemia (Figs. 7-5, 7-6, 7-7, 7-8, and 7-9). The threshold to febrile seizures in rats was related inversely to total water content, volume of intracellular water, and concen- tration of extracellular potassium in the brain. The threshold was related directly to concentration of extracellular sodium and ratio of extracellular to intracellular sodium, concentration of total potassium and intracellular potassium, and ratio of intracellular to extracellular potassium in the brain; it was related directly to the concentration of sodium in the plasma. Similar correlations were observed in mice in which the seizure threshold was modi- fied by dehydration or by the administration of acetazolamide and diphenylhydantoin (Figs. 7-5 and 7-10). Wafer Balance and the Febrile Seizure Threshold. Swinyard (1949) reported a significant increase in brain-cell volume in rats when analyses of plasma and brain tissue were performed within the first 4 hours after the intraperitoneal injection of 5.5 per cent glucose; the change observed in water balance was correlated with a maximal reduction in the threshold of electroshock sei- zures. With few exceptions, factors which influence the threshold to electroshock seizures also modified the febrile seizure threshold 150 EXPERIMENTAL FEBRILE CONVULSIONS Control Water Depriv. Isos molar Glucose Isotonic Saline Febrile Seizure Threshold Temp. (°F) Body Weight % Change Figure 7-5. Elevation of the febrile seizure threshold in mice with dehydration, and reduction of threshold following brain cell overhydration induced by the intraperitoneal injection of 5.5 per cent glucose. Injection of isotonic sodium chloride and expansion of the extracellular fluid space of the brain failed to modify the febrile seizure threshold. (Millichap, 1960 i); the heightened susceptibility to fever-induced seizures in 1-month-old rats compared to adult rats is related to the cellular hydration of the brain and is independent of the vol- ume of extracellular water (Fig. 7-6). Electroencephalographic studies have shown that slowing of cortical potentials occurs with hydration and similar changes are observed with hyperthermia (Lennox et ah, 1954; Baird and Garfunkel, 1956); these electrical changes in the brain are generally interpreted as a sign of dimin- ished cortical excitability. The increased sensitivity to convulsions with overhydration and hyperthermia may be explained by the release of the brain stem reticular system from cortical inhibition ARTIFICIAL FEVER AND HYPERPYREXIA 151 Young Adult Adult (I. P. Glucose) Febrile Seizure Threshold Extracellular Water Intracel lular Water Solids % Brain Tissue Figure 7-6. Febrile seizure thresholds in young (I month old) and adult rats in relation to the volume and distribution of water in the brain. Correlations in young animals are compared with those in adult rats with seizure threshold lowered by the intra- peritoneal injection of 5.5 per cent glucose solution. (From Milli- chap: Neurology (Minneap) 10: 312,1960 i.) and the consequent occurrence of seizures of subcortical origin (Millichap, 1960 i). Sodium Balance and Febrile Seizure Threshold. The concen- tration of extracellular sodium and distribution of sodium in the brain are closely correlated with the threshold to febrile seizures in healthy young and adult rats and also in mice with dehydration or hypoelectrolytemia (Figs. 7-5 and 7-7). These observations in animals are consistent with clinical findings and the occurrence of hyponatremia in relations to febrile seizures in children (Milli- chap et ah, 1960 iii). In one case report a recurrence of febrile seizures was related to the administration of a tap-water enema 152 EXPERIMENTAL FEBRILE CONVULSIONS Young Adult Adult (I. P. Glucose) Febrile Seizure Threshold Temp. F Extracellular Sodium mEq/kg/Brain Water r.h. [Na]c Figure 7-7. Febrile seizure thresholds in young and adult rats in relation to concentration and distribution of sodium in the brain. (From Millichap: Neurology (Minneap) 10: 312, 1960 i.) and resultant overhydration (Crawford and Dodge, 1959). Nyhan and Cooke (1956) have noted the association of hyponatremia and seizures in children with acute infection of the central nervous system; the low serum sodium concentrations appeared to be due to an acute expansion of the extracellular fluid volume. A similar expansion of the plasma volume may also occur as the result of fever per se in patients with acute infections that do not involve the nervous system directly (Soule et ah, 1928). The anticonvulsant drugs, acetazolamide and diphenylhydan- toin, which are known to modify the balance of water and electro- lytes in the brain (Woodbury et al., 1958; Millichap, 1965), have opposite effects on the febrile seizure threshold. Acetazolamide, which tends to stabilize the nerve-cell membrane and decreases ARTIFICIAL FEVER AND HYPERPYREXIA 153 the rate of influx of sodium into brain cells, causes an elevation of the febrile seizure threshold (Fig. 7-10), whereas diphenyl - hydantoin, which increases the rate of transport of sodium and enhances permeability of brain cell membranes, exacerbates the febrile seizure in animals. The ineffectiveness of diphenylhy- dantoin in the prevention of febrile seizures has also been re- ported in children (Millichap et al., 1960 i; Frantzen et al., 1964). Potassium Balance and Febrile Seizure Threshold. The con- centration and distribution of potassium in the brain are closely correlated with the threshold to febrile seizures in animals (Figs. 7-8 and 7-9). Acetazolamide increases cellular potassium and the Young Adult- Adult (I. P. Glucose) Febrile Se i zure Threshold Temp. °F Potassium Sodium Chloride mEq/kg/Brain Tissue Figure 7-8. Febrile seizure thresholds in young and adult rats in relation to concentration of total potassium in the brain. (From Millichap: Neurology (Minneap) 10: 312,1960 i.) 154 EXPERIMENTAL FEBRILE CONVULSIONS Young Adult Adult (I. P. Glucose) Febrile Seizure Threshold Temp.0 F Intracellular Potassium mEq/kg/Brain Water Ratio [KJe Figure 7-9. Febrile seizure thresholds in young and adult rats in relation to concentration and distribution of potassium in the brain. (From Millichap: Neurology (Minneap) 10: 312, 1960 i.) ratio of cellular to extracellular potassium in the brain (Wood- bury et ah, 1958; Millichap et ah, 1955 ii) and raises the threshold convulsive temperature, whereas diphenylhydantoin has no effect on the concentration of cellular potassium (Woodbury et ah, 1958) and fails to raise the threshold to febrile seizures in animals. Histology of Developing Brain and Febrile Seizure Threshold. Sugita (1917) has described the postnatal changes in the histol- ogy of the brain of the rat and these have been correlated with changes estimated in the volume of the cellular and extracellular spaces in the brain during development. An increase in the size of neurons and the development of axons, dendrites, and glial cells in the brain of 1-month-old and adult rats compared to ARTIFICIAL FEVER AND HYPERPYREXIA 155 newborn rats are associated with a relatively greater volume of intracellular water in the mature brain. Myelination and the selective affinity of lipids for electrolytes may explain the changes with maturation in the balance of water and electrolytes which correlate in part with a reduction in seizure susceptibility with increasing age. The disparity between the average convulsive threshold temperatures observed in children and in animals may possibly be explained by differences in degree of histologic and chemical maturation of the brain at various age levels and by genetic and/or acquired factors which contribute to a low febrile seizure threshold in some children. Effects of Drugs on Febrile Seizure Threshold. Various classes of drugs have been tested for their ability to (1) modify the threshold convulsive temperature, (2) abolish the febrile seizure, and (3) retard the rate of rise in body temperature during in- duction of hyperthermia in animals (Millichap et ah, 1960 ii). The results of these studies are shown diagrammatically in Fig- ures 7-10 and 7-11. The febrile seizure threshold is elevated by the anticonvulsants phetharbital (Pyrictal), phenobarbital (Lu- minal), metharbital (Gemonil), meprobamate (Miltown), tri- methadione (Tridione), and acetazolamide (Diamox) and by atropine; it is lowered by diphenylhydantoin (Dilantin), diphen- hydramine (Benadryl), and by acetylsalicylic acid (aspirin) in large doses. The febrile seizure is prevented by anticonvulsants, except acetazolamide and diphenylhydantoin (Table 7-1 and Fig. 7-12), and by atropine; it is exacerbated by antihistaminics, re- serpine (Serpasil), penicillin, oxytetracycline (Terramycin), and cortisone; and is mitigated by acetazolamide and desoxycorti- costerone acetate (DOGA). Steinschneider et ah (1964) have also reported inhibition of the tonic seizure pattern but persistence of clonus in rats with spinal cord transection after treatment with acetazolamide. The antipyretic effect of some anticonvulsant drugs (phetharbital, phenobarbital, meprobamate, and trimetha- dione) and tranquilizing agents was more potent than that of acetyls alicylic acid, which failed to retard the rate of temperature rise. Phetharbital was antipyretic in nontoxic doses, in contrast to the effects of phenobarbital and other anticonvulsants (Fig. 6-2). The advantages of phetharbital in the prevention of febrile 156 EXPERIMENTAL FEBRILE CONVULSIONS 4) L_ D "o 8-p E 2^ « cn 0) c > D IS > c o U Pheno- barbil-al Trimet-ha- dione Acefazol- amide Diphenyl- hydant’oin Acelyl- salicylic Acid Dose (mg/kg) 20 500 60 50 500 Figure 7-10. Elevation of the febrile seizure threshold in mice treated with phenoharhital, trimethadione, and acetazolamide. The degree of reduction of the seizure threshold in animals given diphenylhydantoin was not significant, hut the severity of the febrile seizure was exacerbated. Aspirin caused a significant re- duction in seizure threshold. Drug Threshold Clonus T° Rise Antipyretics ~ I - — Antibiotics — t — Antihistaminics ♦ f t Vasodi lators — — ♦ Tranqui lizers - - ♦ ♦ Anticholinergics ♦ ♦ ♦ Anticonvulsants * ♦ ♦ Figure 7-11. Effects of drugs on experimental febrile seizures and the rate of temperature elevation in mice subjected to hyperthermia. ARTIFICIAL FEVER AND HYPERPYREXIA 157 Table 7—l. Relative Efficacy of Various Anti-epileptic Compounds Against Experimental Febrile Convulsions in Mice DRUG TIME OF MAXIMAL EFFECT (hrs) ANTICONVULSANT EFFECTIVE dose5o ( mg/kg ) TOXIC DOSEgo ( mg/kg ) PROTECTIVE INDEX* Phetharbital (Pyrictal) u 150 190 1.3 Trimethadione (Tridione) 14 1380 1420 1.0 Meprobamate (Miltown) 14 310 275 0.9 Phenobarbital (Luminal) 3 103 61 0.6 Metharbital (Gemonil) U 150 74 0.5 Acetazolamide (Diamox) 3 2000 0 Diphenylhydantoin (Dilantin) 3 — 77 0 From Millichap et al.: Neurology (Minneap) 10: 575, 1960 ii. * Protective or therapeutic index is the ratio of the toxic doseso and effec- tive doseso. seizures have been confirmed in preliminary clinical trials in chil- dren (Millichap, 1960 ii). Effect of the Febrile Convulsion on Seizure Threshold. In a child with a continuing infection which is heralded by a febrile convulsion, recurrence of the seizure is infrequent despite the persistence of fever. In an attempt to confirm this apparent re- fractory state which follows an initial febrile convulsion, the threshold to maximal electroshock seizures was measured in groups of mice before and after induction of a febrile seizure and in unheated control animals. The initial convulsant currents (CCSO) in test and control animals were approximately equal but the elevation in electroshock seizure threshold determined at about 1 hour after the induction of a febrile convulsion was sig- nificantly greater than the elevation noted in control animals 158 EXPERIMENTAL FEBRILE CONVULSIONS Control Dipheny Ihydantoin Phenobarbi tal Me probamate N-Phenyl barbital (B.W. No. 401) Figure 7-12. Movements of mice during artificial hyperthermia recorded by means of a cage and a strain-gauge transducer. In control animals, initial exploratory behavior is followed by a short quiescent period which immediately precedes the clonic con- vulsion, indicated by the arrow. Diphenylhydantoin suppresses the exploratory behavior but does not prevent the convulsion, whereas phenobarbital and meprobamate are anticonvulsant in sedative and muscle relaxant dosages. N-phenylbarbital (Pyrictal, B.W. No. 401) is anticonvulsant in nonsedative doses. ARTIFICIAL FEVER AND HYPERPYREXIA 159 examined after a similar time interval (Fig. 7-13). The threshold had returned to normal levels when re-examined at 22 hours after the febrile convulsion. Repeated febrile convulsions induced in groups of mice over a period of seven days caused a persistent elevation in the febrile seizure threshold, and the threshold to maximal electroshock seizures tested at seven days and after the induction of four febrile convulsions was significantly increased when compared in control unheated animals (Fig. 7-14), It seems probable that the infrequency of recurrence of febrile seizures despite continuing infection may be explained by a postictal rise in threshold, and the degree of fever needed for the ► Fever convulsion induced at 1 -1-j hrs. ) Controls unheated Maximal Electroshock Seizure Threshold (CC 50 in mi 11 lamps) % Increase In Seizure Threshold Time In Hours Figure 7-13. Threshold to maximal electroshock seizures in mice before and after induction of febrile convulsions. Each point represents the current required to produce nonfebrile seizures in 50 per cent of a group of animals. 160 EXPERIMENTAL FEBRILE CONVULSIONS Febrile Seizure Threshold Temp. (F°) Test Control Electroshock Seizure Threshold % Change Time in Days Figure 7-14. Effect of recurrent seizures with hyperthermia on the febrile and electroshock seizure thresholds in mice. Each point represents the mean value in at least 10 animals tested. induction of a second seizure is greater than that causing the initial febrile convulsion. Rate of Temperature Rise and Febrile Seizures: No Significant Relationship The rapidity of rise in body temperature is often invoked as the important determinant of a convulsion with fever, and the experiments by Wegman (1939) are usually quoted in support of this theory. In these studies kittens and cats were subjected to hyperthermia by means of an incubator, a high-frequency generator, and a radiant heat chamber. Of 82 animals tested, only 32 had convulsions. A convulsive response was obtained in 26 (48 per cent) kittens, 2 (17 per cent) adult cats, and 4 (25 per cent) animals of intermediate size. An analysis of these data shows that the difference in the incidence of convulsions in kittens and adult cats is not significant (x2 = 2.79, P > 0.05). ARTIFICIAL FEVER AND HYPERPYREXIA 161 The total number of kittens tested was divided into two groups according to the duration of application of the heating stimulus; of 42 animals treated for less than 100 minutes, 25 (59 per cent) had convulsions, and of 12 heated for a longer period, 1 (8 per cent) convulsed. At the end of the experiment the majority of kittens heated for a short period recovered, whereas all animals subjected to the heating stimulus for long periods were either dead or dying. The significance of results in animals subjected to irre- versible brain damage is questionable, and the failure to induce convulsions in all animals tested precludes an evaluation of the relative etiologic importance of rate of rise and height of body temperature. Despite his observation that the occurrence of con- vulsions was more frequent in animals with the higher tempera- tures, Wegman concluded that “rapidity of temperature increase is the determining factor in the production of convulsions.” How- ever, the author’s calculations based on the data of Wegman have shown that the kittens with convulsions had a mean body temperature of 43.4° C (110.2°F) and kittens which failed to con- vulse had a mean body temperature of only 42.2° C (107.9°F). The difference between these values is highly significant (P < 0.001), and the occurrence of febrile convulsions is related to the height of the body temperature. The methods of induction of hyperthermia available for these experiments were not sufficiently adequate to allow an elevation of temperature to the threshold convulsive levels of all test animals, and the maintenance of subthreshold degrees of fever in kittens for prolonged periods resulted in irreversible pathology and death. Notwithstanding these apparent limitations of technique and critical analyses of his conclusions, Wegman was the pioneer of the laboratory method of investigation of febrile convulsions; he demonstrated that fever without an accompanying infection was sufficient to induce convulsions in the immature animal. In the author’s studies in which the microwave diathermy generator was employed, the threshold and susceptibility to fe- brile seizures in mice, rats, guinea pigs, and kittens were inde- pendent of the rate of temperature rise and the severity of the pyrogenic stimulus (Millichap, 1959). The rate of rise of temper- ature varied from 2.0 to B.O°F per minute, while the temperature 162 EXPERIMENTAL FEBRILE CONVULSIONS at the onset of the convulsion remained constant at 109.7°F (Fig. 7-15). In mice treated with chlorpromazine, the rate of rise of temperature is retarded, but the threshold convulsive temperature is the same as in control untreated animals (Fig. 7-16). The rate of rise of body temperature in rats between 3 and 35 days old is related directly to the threshold convulsive temperature and inversely to the susceptibility to febrile seizures; the rate of rise is slower in adult animals but is not corre- lated with the weight of the animal nor with growth of fur (Figs. 7-17 and 7-18). In two groups of kittens which convulsed at mean body temperatures of 45.7° C (114°F) and 45° C (113°F), the mean rate of rise of temperature in response to different heating currents was O.5°C and I.5°C per minute, respectively. The rapidity of temperature rise differed significantly (P < 0.01) in the two groups, whereas the convulsive temperature was un- altered (P < 0.2). In these experiments febrile convulsions were induced in all animals tested, and the importance of the height of the body temperature in the induction of the seizure was established (Millichap, 1959). Rate of Temp. Convulsive Rise (°F per min) Temp.( F) Intensity of Heating Current (%) Figure 7-15. Body temperature at the onset of febrile convul- sions in mice with various rates of temperature elevation. (Milli- chap: Pediatrics 23:76, 1959.) ARTIFICIAL FEVER AND HYPERPYREXIA 163 Convulsive Temp. (°F) Rate of Temperature Rise (°F per min) Control Group Chlor- promozine Treated Low Very Low -Ambienf— Temp. Temp. Figure 7-16. Rates of temperature elevation and convulsive temperatures in mice treated with chlor promazine and in controls with normal or low resting body temperatures. (From Millichap: Pediatrics 23:76, 1959.) Effects of Hyperthermia and Febrile Convulsions on the Electroencephalogram and Behavior The experimental method introduced by Wegman (1939) was used subsequently by M. A. Lennox and colleagues (1954) in order to test the hypothesis that brain injury may occur as a result of the febrile convulsion and may predispose to the sub- sequent development of spontaneous seizures. These authors studied 32 kittens 2 to 16 weeks of age, before, during, and after 49 bouts of artificially induced fever. Electroencephalograms were recorded by means of scalp electrodes, and the brains of nine animals were examined histologically at autopsy. Convul- sions occurred on only nine occasions and in 18 per cent of the episodes of hyperthermia; five animals died during the convulsive 164 EXPERIMENTAL FEBRILE CONVULSIONS Rate of Temp. Rise (°F per min) Threshold Convulsive Temp. (°F) Age in Days Figure 7-17. Rate of rise of body temperature in relation to threshold convulsive temperature in rats between 3 and 35 days of age. (From Millichap: Pediatrics 23:76, 1959.) episode. Duration of hyperthermia and height of the fever were the most important factors that determined the occurrence of convulsions, and all animals in which the rectal temperature rose above 42.8° C (109.0°F) in less than 30 minutes had convulsions; the majority occurred in kittens 5 to 8 weeks old. The incidence of convulsions was higher and mortality was lower when sub- cutaneous fluids were administered. The electroencephalogram showed slow wave activity that varied in degree in direct relation to the height and duration of fever and inversely with the age. The administration of sub- cutaneous fluids appeared to prevent the electroencephalographic changes in some cases, whereas the inhalation of 5 per cent carbon dioxide in oxygen had no obvious effect. A combination of convulsions and fever resulted in prolongation of the slow wave activity to as long as 24 hours. The brains of three kittens which showed the most marked electroencephalographic slowing ARTIFICIAL FEVER AND HYPERPYREXIA 165 Rate of Temp. Rise (°F per min) Rats Guinea Pigs Weight- in Grams Figure 7-18. Rate of rise of temperature in relation to body weight during artificial hyperthermia of constant intensity in rats and guinea pigs. had the most striking changes on microscopic examination at autopsy. Cellular changes consisted of shrinkage, hyperchromasia, and loss of cells with microglial reaction. The cerebellum was involved most consistently and more prominently than the cere- brum, whereas the basal ganglia, pons, and brain stem were spared. Animals which died acutely had no cellular changes in the brain, but the slowing in the electroencephalogram was marked. In these studies as in those of Wegman, the relative importance of the height of the temperature and the rapidity of rise of temperature could not be evaluated satisfactorily since convulsions occurred in only nine of 32 kittens subjected to a total of 49 febrile episodes. Furthermore, a high proportion of the animals died, and the mortality and severe electroencephalo- graphic changes appeared to be related to dehydration. In experiments with microwave diathermy as the method of induction of hyperthermia, the majority of animals recovered after the fever-induced convulsion, and the lack of permanent damage to the brain was demonstrated by the finding of normal levels of enzyme activity in the brain of test animals compared 166 EXPERIMENTAL FEBRILE CONVULSIONS with control animals not subjected to hyperthermia or febrile convulsions (Millichap, 1960 i). Werbolf and Havlena (1963), using the same experimental method as that described by Milli- chap (1959), have demonstrated that the febrile convulsion has no deleterious effect on the behavior of young rats. On the con- trary, the infantile febrile convulsion appeared to act similarly to other types of stimulation administered to the very young animal and had a facilitating effect on some aspects of behavior. Thirty female albino rats and their litters were assigned to one of three treatment groups: experimental, with febrile convulsions induced at 3 days of age when the threshold is lowest; control animals which were handled; and reference animals which were not handled. After the distribution of the animals at birth, none of the groups received any further handling until weaning at 30 days of age. Each pup was then weighed and evaluated for either learning ability on a Lashley 111 water-maze, activity-wheel per- formance, or susceptibility to audiogenic seizures. Animals which had been subjected to the febrile convulsion were heaviest and were most resistant to audiogenic seizure stimuli; no differences were found in maze-learning ability or activity levels. The results suggest that the febrile convulsion is not a noxious experience which results in cerebral damage, but the findings are in agree- ment with other evidence that stimulation early in life facilitates a variety of subsequent behavioral performances. Clinical re- ports that emphasize the occurrence of brain damage as a com- plication of febrile convulsions are not supported by evidence obtained in experimental animals in which febrile convulsions, induced by microwave diathermy, are short in duration and un- accompanied by dehydration and other complicating factors. Etiology of Reversible Electroencephalographic Changes with Febrile Convulsions The electroencephalographic changes demonstrated by M. A. Lennox and coworkers (1954) in animals with artificially induced hyperthermia are similar to those induced by hyperventilation and may be related to the hyperpnea which complicates fever. Fluid loss and acidosis have been invoked as contributing causes, but the mechanism of the changes in the electroencephalogram ARTIFICIAL FEVER AND HYPERPYREXIA 167 are not definitely determined. Hemodilution is one of the earliest responses of the body fluids to heat stress, and both respiratory alkalosis and an acidosis have been reported during hyper- thermia. Bass and Henschel (1956) found that the increase in plasma and blood volume is associated with little change in the composition of the blood. With continued heat stress the red cells and circulating protein are increased, and with sweating and dehydration the blood volume is reduced. Yannet and Darrow (1938) demonstrated a shift of water unaccompanied by cations from the intracellular to the extracellular fluid in the brain of cats with hyperthermia, but no correlation was observed between these changes and the occurrence of seizures; the animals died as a result of the fever and the significance of these experiments is questioned. Rodbard and colleagues (1951) reported no changes in plasma or blood volume of rabbits and chickens during hyper- thermia induced by incandescent lamps, but again the data are difficult to interpret since the animals were obviously dehydrated and the effects of lethal hyperpyrexia cannot be compared with the typical febrile convulsion. Ikeda (1962) performed biochem- ical studies on the arterial blood of rabbits in which febrile convulsions were induced by the rapid elevation of the environ- mental temperature. The carbon dioxide content of the blood was decreased in mature and young animals and the oxygen content and oxygen saturation of the blood were markedly decreased in weanling but not in older animals. No change was noted in the serum calcium level, but hyperglycemia occurred, particularly in the mature animal. The water content of the blood was slightly elevated in all groups. The author concluded that hypoxemia was important in the pathogenesis of febrile convulsions, especially when the rectal temperature was elevated rapidly. The height of the body temperature in these experimental animals was not recorded in the abstract of the communication. Effects of Fever on Cortical Epileptogenic Lesions Schmidt, Ward, and Wolfe (1956) investigated the effects of hyperthermia on experimental epileptogenic foci produced by in- jection of alumina cream. Hyperthermia was induced by the application of an electric heating pad to the restrained animal 168 EXPERIMENTAL FEBRILE CONVULSIONS wrapped in woolen blankets. Electroencephalographic recordings were made before and after the rectal temperature had risen to 40.6° C (105°F) or above. In normal monkeys, hyperthermia caused little alteration in the electroencephalogram unless the degree of fever was extreme; in animals with epileptogenic lesions there was no evidence of activation of the epileptic focus. The authors concluded that febrile convulsions in the human infant are not precipitated by temperature elevation alone unless the seizures are of subcortical origin. Body temperatures equiva- lent to those usually encountered in infants with febrile seizures were produced in the monkeys in these experiments, but clinical seizures occurred in only eight of 18 animals. It seems probable that the threshold to febrile seizures in the monkey is higher than in the child, and the degree of fever required to produce con- vulsions may be similar to that observed in the mouse, rat, guinea pig, and kitten (Millichap, 1959). The method of induc- tion of hyperthermia in the experiments on monkeys was not sufficiently adequate to raise the temperature to a convulsive level in a short period of time; two deaths occurred with a rise in rectal temperature above 42.2° C (108°F) and prolonged hy- perthermia at subconvulsive threshold levels would be likely to cause dehydration and to result in a fatal outcome. It would be of interest to repeat these experiments using a method of in- duction of fever to attain convulsive levels without irreversible pathological changes. It is possible that a cortical focus may be more resistant to hyperthermia than a subcortical epileptogenic lesion, but the induction of clinical or electroencephalographic seizures or rapidly lethal temperatures without dehydration in both experimental models would be necessary in order to test this hypothesis. Kashiwase (1962) studied the effects of fever on the electro- corticograms of 25 curarized adult cats and 12 curarized kittens or immature cats. A dissociation between the electrical activity of the neocortex and the limbic system was generally observed; hippocampal arousal waves were more prominent and continu- ous, whereas the neocortical spindle bursts and slow waves were increased. Spontaneous seizure discharges were observed in kit- tens during fever and the majority originated in the neocortex, ARTIFICIAL FEVER AND HYPERPYREXIA 169 particularly in the anterior sigmoid gyrus; these discharges were propagated to the caudate nucleus and thalamus and finally to the whole brain. The production of localized seizure discharges by electrical stimuli of threshold intensity in kittens showed that the susceptibility to seizure activity in the neocortex was greater than in the amygdala and hippocampus. The author concluded that the convulsive effects of hyperthermia were correlated with phylogenetic and ontogenetic factors and were most prominent in the neocortex of the brain of kittens and immature cats. In contrast to the report by Schmidt et al. (1956), the results obtained by Kashiwase do not favor a subcortical origin for the febrile seizure discharge. Teschan and Gellhorn (1950) also demonstrated an increased sensitivity of convulsive cortical neurons in response to heat ap- plied to the cerebral cortex of cats by direct radiation from a 250-watt incandescent infrared lamp. Potentials were recorded from various parts of the exposed cortex and local convulsive activity was induced by topical application of picrotoxin. Con- trol experiments without application of heat showed little or no change in picrotoxin-induced convulsive activity. The tempera- ture of the cortex was measured with thermocouples, and the rate of heating to 50° C (112°F) was either 12 or 48 minutes. The convulsive potentials disappeared at a temperature of 45.5° C and were more sensitive to heat than normal electrical potentials which persisted up to 50° C. The disappearance of convulsive activity was preceded by a period of progressively diminishing spike discharges, and the effects of hyperthermia were com- parable to the selective action of anoxia and asphyxia on con- vulsive potentials (Gellhorn and Heymans, 1948). Anoxia may be a major factor in the action of hyperthermia on convulsive neurons, but the changes induced by the temperatures applied in this experiment were irreversible. When more moderate degrees of heat to 44° C (111.2°F) were applied repeatedly and for short periods, convulsive discharges were increased in fre- quency, and the effects were compared to the increased frequency of alpha potentials which occurs during an elevation in tempera- ture (Hoagland et al., 1939). Thus, the action of fever on epi- leptogenic cortical lesions may vary in relation to the duration 170 EXPERIMENTAL FEBRILE CONVULSIONS and degree of hyperthermia, and the more moderate tempera- tures with activation of convulsive neurons are more comparable to the conditions in clinical seizures than the higher degrees of heating employed in these experiments. Effects of Fever on Electroshock Seizure Susceptibility Swinyard and Toman (1948) studied the effects of an increase in body temperature on the properties of experimental seizures in rats. The study was intended to elucidate the paradoxical occurrence of convulsions in response to fever in children in con- trast to a reported remission of seizures during febrile states in some patients. Minimal electroshock seizure thresholds were com- pared at normal and elevated temperatures, with a period of at least 12 hours between tests in each animal; changes in the pat- tern and duration of maximal seizures were also determined. The minimal electroshock seizure threshold was increased with an elevation of body temperature and decreased during hypo- thermia. Similar effects of subconvulsive elevations of body tem- perature have been noted in mice with minimal electroshock seizures (Brown, W. C., 1953). The total duration of the maximal seizure varied inversely with the body temperature and the re- covery phase was hastened by fever; above 42° C (107.6°F) the seizure consisted of a tonic flexion without extension and with superimposed fine clonic movements. The elevation in seizure threshold during hyperthermia in animals may confirm the ob- served remission of seizures in some patients with epilepsy dur- ing febrile illnesses, and may suggest that factors other than fever are responsible for febrile convulsions in children. However, the effects of age, of importance in relation to febrile convulsions in young and immature subjects, were not determined in these experiments. Effect of Convulsions on Body Temperature Elevations of body temperature are observed in patients with prolonged seizures, and increased muscular activity is sometimes invoked as the explanation for the fever. However, Hoyt and Rosyold (1951), in studies of the effects of electroconvulsive shock on the body temperature of the rat, were unable to relate ARTIFICIAL FEVER AND HYPERPYREXIA 171 the resultant fever to changes in activity. Twenty adult rats were divided equally into control and experimental groups, and con- vulsions were induced in the experimental animals daily by a 50 milliampere current for 0.2 second applied by clip-electrodes attached to the ears. Rectal temperatures were taken in two animals from each group on five occasions between 9 a.m. and 9 p.m. each day. The series of electroshocks with convulsions re- sulted in a decreased variability of the body temperature; a high temperature persisted in shocked animals, while the temperature of control animals was reduced during periods of inactivity. Modi- fication of the temperature-regulating mechanism in the hypo- thalamus was considered the most likely explanation for the change in temperature following convulsions in these animals, and a similar mechanism may occur in children with convulsions. Effects of Behavior and Arousal on Brain Temperature The relation of brain temperature to the behavior and arousal of animals has been studied by Hull and co-workers (1965). The electroencephalogram and brain temperatures were monitored simultaneously during waking and sleeping states and during an associative conditioning situation in which peripheral sensory stimuli and intracranial stimulation were used to signal the presentation of food to a hungry animal. Experiments were made in eight cats with brain recording and stimulation devices per- manently implanted. A decrease in temperature of the brain occurred in all animals with the onset of electroencephalographic synchrony associated with sleep or drowsiness; and a relatively sharp decrease in temperature occurred concomitant with the onset of 8 to 12 cps spindles in the electroencephalogram. An in- crease in temperature of the brain occurred when the animal awakened, and the rate and duration of the temperature increase paralleled the apparent degree of wakefulness attained. Con- ditioning with a light and electrical stimulation of the lateral geniculate body as signals for subsequent presentation of food reinforcement caused increases in temperature of the brain. The ingestion of cool milk at a temperature of 15 to 25 °C caused a sharp and relatively marked decrease in the brain temperature which persisted for 1 to 3 minutes; during this time stimuli which 172 EXPERIMENTAL FEBRILE CONVULSIONS ordinarily evoked an increase in temperature were ineffective. The authors concluded that a change in brain temperature is a useful indicator in studies correlating brain function with be- havior and may be caused by modifications of blood flow and blood temperature and by heat produced from the metabolism of the brain itself. These experiments are of interest in relation to the mechanism and management of febrile convulsion, and similar studies in animals with experimental seizures should be of value. CLINICAL INVESTIGATIONS Studies of experimental artificial hyperthermia in children with febrile convulsions are limited because of ethical considerations and possible hazards involved in such procedures. Effects of Artificial Hyperthermia on the Electroencephalogram Baird and Garfunkel (1956), in a carefully controlled investi- gation of 12 children, seven of whom had a history of febrile convulsions, recorded electroencephalograms during hyperther- mia induced by the intravenous injection of typhoid vaccine. Rectal temperatures were recorded every 30 minutes and aspirin (60 mg per year of age) was given orally or rectally when the temperature reached 40° C (104°F) or when definite electro- encephalographic abnormalities appeared. Sodium phenobarbital (5 mg/kg of body weight) was administered intramuscularly if convulsions occurred or seemed imminent. Maximum temperatures ranging from 38.9 to 40.9° C (102° to 105.6°F) were recorded between 2 and 34 hours after the in- jection of typhoid vaccine; complaints or symptoms of stiffness of neck and back, headache, malaise and restlessness, and chills, nausea, and vomiting were common. At the higher temperatures, the children became quiet but seldom slept. Electroencephalo- graphic abnormalities, observed in all patients, consisted of high- voltage slow waves, increased delta activity, and spike-and-wave formations; the records became normal within 3 days and usually within 24 hours. Three patients had febrile convulsions during ARTIFICIAL FEVER AND HYPERPYREXIA 173 hyperthermia; two of these had a history of febrile seizures and in one a previous electroencephalogram was abnormal. The authors considered that hyperthermia was the only factor involved in the induction of electroencephalographic abnormali- ties in their patients. They cautioned that an abnormal electro- encephalogram with a febrile illness should not be regarded as indicative of disease of the central nervous system, since high- voltage slow waves may persist for at least 3 days after a fever unassociated with convulsions. They add that their observations tend to minimize the significance of electroencephalographic abnormalities following febrile convulsions and to negate inter- pretations of these changes as evidence of irreversible pathology in the brain and a precursor of spontaneous seizure susceptibility. That the common form of febrile convulsions, with its associated reversible electroencephalographic changes, should produce ir- reversible brain damage is unlikely, and may not be inferred on the basis of studies in kittens with degrees of hyperthermia of lethal severity. Baird and Garfunkel conclude that artificially induced fever, as employed in their experiments, is worthless as an activating procedure in the distinction of children with con- vulsive disorders from those with a normal seizure threshold. They recommend that all children with high fevers, irrespective of a history of febrile seizures, should receive treatment with antipyretics and sedatives as well as any necessary antibiotics and fluid therapy. In view of the marked electroencephalographic changes produced in these children by artificial hyperthermia, the authors considered that further experiments of this type were inadvisable. Other experiments concerning the effect of artificial fever on the electroencephalogram have been performed in adults, of which some were normal volunteers and others were suffering from diseases such as general paresis. Hoagland (1936) found an increase in the frequency of the electroencephalogram with an elevation of temperature, more marked in patients with advanced general paresis than in normal human subjects; for every degree centigrade the increase in frequency was 1.0 and 0.45 cps, re- spectively. Krakau and Nyman (1953) used frequency analysis 174 EXPERIMENTAL FEBRILE CONVULSIONS of the electroencephalogram in a study of the effect of artificial fever on the alpha activity in man. No consistent change in the frequency of the alpha activity was noted but the amplitude of the alpha peak was lowered and the shape of the alpha distribu- tion curve was changed. Pathophysiology of Artificial Hyperthermia Lenggenhager (1952) studied the pathogenesis of febrile con- vulsions in children by observing the effects of hyperthermia on the pH of venous blood, the oxygen consumption, and the res- piratory volume. Exogenous fever was produced by immersion in hot baths and endogenous fever by injections of a pyrogenic substance, Pyrifer. An elevation in temperature to 39° C (102.2°F) caused an increase in the pH of venous blood, more marked during exogenous fever than with endogenous fever. Oxygen consumption was increased by 78 per cent following immersion in a hot bath, and a smaller increase of 40 per cent occurred after injection of Pyrifer. The respiratory volume was increased by 300 per cent and by 45 per cent during exogenous and endogenous fevers, respectively. The greater oxygen consumption during ex- ogenous fever than during endogenous fever is explained by the disparity in the degree of hyperthermia in various parts of the body produced by these two methods; with endogenous fever heat is lost from the skin by evaporation of perspiration and the periphery of the body is cooler than deeper tissues, whereas in exogenous fever all parts of the body are almost equally hyper- thermic. The author concludes that febrile convulsions are physio- logical and distinct from epilepsy and may be explained by the relatively small lung volume compared to body surface in infancy; the increased oxygen consumption during hyperthermia results in a relatively greater respiratory volume in the child than in the adult and is associated with hyperventilation alkalosis and con- sequent convulsions with loss of consciousness. A decrease in the ionization of serum calcium during alkalosis and the direct effects of heat on nerve cell function may contribute to the etiology of febrile convulsions in children. The adult is protected from the harmful effects of hyperventilation alkalosis during fever because of his relatively large lung surface and his ability to maintain ARTIFICIAL FEVER AND HYPERPYREXIA 175 sufficient oxygen consumption without excessive increases in respiratory volume. However, with exogenous fever and heat stroke the demand for oxygen is increased in all cells of the body, and the increase in respiratory volume required to offset the anoxia to tissues may contribute to convulsions even in the adult under these conditions. Prolonged and continuous anticonvulsant therapy was considered inappropriate in children with febrile convulsions since, in the author’s opinion, every child suffering from an equal degree of hyperthermia would develop the same symptoms; an increased susceptibility to febrile seizures in some children due to genetic or acquired factors was not considered in this publication. Oxygen Saturation and pH of Blood During Hyperthermia. Yamakita (1921) has reviewed the early literature regarding the effects of fever on the pH of the blood and the consequent changes in the per cent saturation of hemoglobin with oxygen. Barcroft (1914) found that warm blood binds oxygen less than cooler blood and that the dissociation curve of blood is an ac- curate index of its hydrogen ion concentration. An increase in the partial pressure of carbon dioxide or addition of lactic and other acids to blood causes reduction in the oxygen combining power, whereas alkalinity accelerates the rate of oxidation of blood. Yamakita investigated the pH of blood in patients with pulmonary tuberculosis, typhoid fever, and influenzal pneumonia; he also studied the effects of fever without infection induced in rabbits by heat puncture, a mechanical stimulus to the corpus striatum, and by injections of peptone, tuberculin, and typhoid toxin. A decrease in the oxygen saturation of the blood was observed in patients with fever greater than 38.5° C (101.3°F) but not during nonfebrile periods. The associated acidosis, caused by the hyper- thermia and toxic breakdown of body tissues, resulted in severe dyspnea in some patients. That toxins may contribute more than fever to the breakdown of body tissues and consequent acidosis was suggested by differences in- degree of effect on pH and oxygen saturation of blood observed in animals with heat puncture and in those with typhoid toxin injection. On the basis of these findings in patients and in animals with endogenous fever and pre- vious reports of acidosis as a result of exogenous artificial hyper- 176 EXPERIMENTAL FEBRILE CONVULSIONS thermia (Haldane, 1905), Yamakita advocates the administration of sodium bicarbonate in patients with febrile illnesses in order to lessen the acidosis and improve the oxygen saturation of the blood. The effect of hyperthermia on the pH of blood appears to vary according to the length of exposure and the degree of fever. Daly and Knudson (1932), in a study of the acid-base equilibrium and phosphorus metabolism in dogs heated by radiothermy or hot water, found either alkalosis, no change in pH, or acidosis, de- pending upon the length of exposure and maximum increase in temperature. They concluded that when the loss of C02 is greater than the production of C02 and lactic acid, alkalosis results; if the loss of C02 just compensates for the production of systemic acidosis, no change in pH is found; and if the loss of C02 is less than the systemic acidosis caused by the formation of C02 and lactic acid, acidosis is produced. Both acid-soluble phosphorus and inorganic phosphorus were decreased in whole blood and plasma during hyperthermia and a slight increase in phosphorus in the blood cells was only an apparent change, since the cells shrink as the result of water loss during heating. White (1925) studied the bicarbonate reserve and the dissocia- tion curve of oxyhemoglobin in 10 patients with various febrile ill- nesses. A shift to the right in the dissociation curve of hemoglobin meant that oxygen is more readily available to the tissues during fever. The bicarbonate reserve was above normal in five patients, normal in one, and low in the remainder. The effect of the ele- vated temperature on the dissociation curve of oxyhemoglobin was not uniform but was, on the average, greater than that ex- pected from the alkalosis and results of experiments on normal blood in vitro. Apparently, factors in addition to fever, such as alterations in the concentration of various electrolytes, accounted for the changes observed in the bicarbonate reserve and in the oxygen combining power of the blood. Bischoff, Long, and Hill (1931) reported an increase in plasma pH and a fall in total C02 content of whole blood during hyper- thermia induced by short radio waves in one healthy volunteer and two patients with general paresis. A shift of bases to blood ARTIFICIAL FEVER AND HYPERPYREXIA 177 proteins and an increased oxygenation of the hemoglobin of venous blood occurred with the change of pH to the more alkaline reaction, and examination of constituents of urine and sweat re- vealed no evidence of a compensatory acidosis. With the excep- tion of a fall in blood volume, no difference was noted between effects produced by short radio waves and hyperthermia induced by diathermy, warm air, or hot water baths. The hemoconcentra- tion noted during fever following exposure to condenser plates in circuit with a short-wave radio transmitter could result from the greater degree of perspiration associated with this method. The elimination of nitrogen in perspiration accounted for a decreased rate of urinary excretion of nitrogen, and the heightened meta- bolic rate with conversion of inorganic to organic phosphorus in the blood resulted in a fall in urinary excretion of phosphorus (Bischoff, Maxwell, and Hill, 1931). Gordon, Darling, and Shea (1949) found the influence of hyper- thermia on the oxygen and carbon dioxide equilibria in the blood to result from diverse factors, representing the sum of physico- chemical and physiological changes. Samples of arterial blood were obtained after elevation of the body temperature to 40° C (104°F) by a hot humid-air cabinet, and on different days at times of normal temperatures, from eight male adults of whom two were normal volunteers and the remainder had neurosyphilis or other diseases. Sedatives, given to allay restlessness and dis- comfort, caused a relative increase in carbon dioxide tension in the blood and were omitted in some experiments. The oxygen dissociation curve during elevated temperatures showed a shift to the right, and at a given oxygen saturation the increment of p02 with a rise in temperature from 37° C (98.6°F) to 40.6° C (105.1°F) was about 13 per cent. Unsedated patients responded with an uncompensated hyperventilation alkalosis, reflected by an elevated blood pH, decreased arterial pC02, and minor changes in alkali reserve, usually in a positive direction. In se- dated patients, the blood pH was normal or slightly reduced, with the pC02 correspondingly increased. Whereas the acid-base equi- librium was altered by hyperthermia, the p02 (mm Hg) and oxy- gen saturation of the blood were unaffected in patients with or 178 EXPERIMENTAL FEBRILE CONVULSIONS without sedation. The possible occurrence of anoxic anoxia, sometimes invoked as an etiologic factor in convulsions with hyperthermia, was not confirmed in these experiments. Blood Volume and Electrolyte Changes During Hyperthermia. Soule, Buckman, and Darrow (1927), using Congo red dye as an indicator, found plasma volumes in six children between 25 and 30 per cent higher during pneumonia and typhoid fever than at times of convalescence from the febrile illnesses. Hemoconcen- tration during hyperthermia, reported by Simon (1936) and Keutmann, Bassett, and Warren (1939), appears to result from dehydration and loss of fluids from the skin and lungs, and is more marked with some heat cabinets than other methods of fever induction. Brown and Clark (1939) emphasize the value of measurements of the blood specific gravity, in order that dangers of dehydration or overhydration during artificial hyperthermia may be avoided. Osborne (1941) found no significant changes in the specific gravity of the blood during fever of 40° C (104°F) maintained for 4 hours in 10 patients with arthritis; fluids con- taining sodium chloride were administered by mouth in amounts judged appropriate on the basis of the physician’s experience. Bass and Henschel (1956), in a review of the literature concern- ing responses of body fluid compartments to heat and cold, con- clude that hemodilution is one of the earliest effects of heat stress, reflecting a shift of interstitial fluid to the vascular bed in response to peripheral vasodilatation. With continued heat stress the body mobilizes red cells and circulating proteins, subserving the ho- meostasis of an expanded blood volume. In the absence of de- hydration with sweating and loss of fluids via the lungs, acute hyperthermia elicits responses which, under controlled conditions, would result in increased plasma and blood volumes, with little change in composition. Keutmann and colleagues (1939) studied electrolyte and fluid loss through sweating during artificial fever therapy sustained for periods up to 48 hours in two patients and 4 hours in two additional patients. Determinations of hematocrit, hemoglobin, and serum proteins pointed to moderate dehydration in three patients. The amount of sodium and chloride lost in perspiration represented from 7 to 19 per cent of the amount estimated in the ARTIFICIAL FEVER AND HYPERPYREXIA 179 extracellular water at the beginning of the treatment; differences in the losses through the skin were dependent largely on variations in ability of the individual to sweat. Appreciable losses of potas- sium and nitrogen were encountered during prolonged fever but not with fever of short duration. Sodium concentration in the serum decreased more than the sum of the decreases in bicar- bonate and chloride, and a primary base deficit was suggested. The authors stressed the need for adequate preparation of the patient beforehand and replacement therapy, particularly sodium chloride and water, in order to prevent the development of symp- toms of dehydration during hyperthermia. Paul and Kemp (1940) found no changes in the potassium levels and no significant altera- tions of sodium, calcium, and sugar concentrations in the blood during hyperpyrexia induced in 15 patients by means of a heat cabinet or malaria; hematocrit determinations showed that the degree of hydration of the blood had been maintained within narrow limits by the intravenous administration of appropriate fluids. When fluid losses have not been corrected, however, as in in 10 patients with fever of 39.4° C (103°F) maintained for only 3 to 4 hours (Simon, 1936), hemoconcentration occurred and blood chemistry determinations revealed elevations of sugar, nonprotein nitrogen, and creatine and a reduction of chloride. The average increase in blood sugar during fever was more than 20 per cent compared to initial levels. In children with febrile con- vulsions, elevations of spinal fluid sugar concentration have been reported, but abnormal levels of blood sugar have not been ob- served (Chapter 3). Dehydration, hypotension, hypoxia, and acidosis have been reported in patients who develop hyperthermia during anesthesia (Saidman et ah, 1964), and in a study of 215 patients who under- went surgery at Childrens Memorial Hospital, Chicago, a body temperature above 37.8° C (100°F) was recorded in 50 per cent (Bigler and McQuiston, 1951). Convulsions during anesthesia have occurred in response to diethyl ether in patients with hyper- thermia and in experimental animals. Respiratory acidosis or administration of carbon dioxide during ether anesthesia has been invoked by some authors (Cassels et al., 1940) as an im- portant factor in the etiology of these convulsions. The inhala- 180 EXPERIMENTAL FEBRILE CONVULSIONS tion of carbon dioxide may cause an increase or decrease in brain excitability depending on the concentration (Woodbury and Karler, 1960), and a rapid withdrawal from anesthetic concentra- tions of carbon dioxide will induce convulsions in mice and rats. However, acute changes in acid-base equilibrium of the blood have only minimal or no effects on the threshold and patterns of experimental seizures (Millichap, 1965), and factors other than respiratory acidosis must be considered in the etiology of ether convulsions. Hyperthermia, a constant finding in patients and animals with ether convulsions, may result in hemoconcentration and sludging of blood in cerebral capillaries. Microscopic exam- ination of the brain and lung tissue of animals subjected to hyperthermia and ether anesthesia has revealed a high incidence of fat embolization and neurologic deficit (Owens, 1958), whereas ether administered to animals with normal body temperatures did not promote development of fat emboli. Heparin prevented the electroencephalographic evidence of seizures in 90 per cent of animals, and the frequency of occurrence of fat emboli and neurologic deficit was correspondingly reduced. It is apparent that ether is contraindicated in the treatment of febrile convul- sions and that the administration of heparin may be considered in patients with prolonged seizures and evidence of impending neurologic deficits. HYPERPYREXIA, CONVULSIONS, AND BRAIN PATHOLOGY Febrile convulsions are generally associated with moderate elevations in temperature, and brain damage is a rare complica- tion. In patients with hyperpyrexia, however, pathological lesions in the brain are common, although some patients have recovered without sequelae despite elevations of temperature up to 43.3° C (110°F). Hyperpyrexia and Brain Lesions The disorders which result from excessive heat include heat stroke, heat exhaustion, and heat cramps. Heat stroke is associ- ated with hyperpyrexia, whereas in heat exhaustion and cramps ARTIFICIAL FEVER AND HYPERPYREXIA 181 the body temperature is normal or only slightly changed. The hyperpyrexia in heat stroke is complicated by coma, delirium, convulsions, and circulatory collapse; damage to the central ner- vous system and the thermoregulatory center results in impair- ment of sweating and further exacerbation of the fever (Malamud et al., 1946). Paradoxically, fever induced artificially has been found to inhibit the formation of scar tissue in the spinal cord of animals and to promote the regeneration of neurons and axons (Windle, 1952); artificial fever has also been advocated for the treatment of various neurologic disorders (Bailey et al., 1952). Hartman (1937) described the pathological changes in various organs following hyperpyrexia in animals. Gross changes consisted of engorgement and congestion of blood vessels, degeneration and hemorrhage of the adrenals, hemorrhages in the brain, marked edema and congestion of the lungs, contraction and bloodlessness of the intestine, and parenchymatous degeneration of the liver and kidneys. Microscopic examination revealed acute passive congestion of all organs and tissues and cellular degenera- tion and hemorrhages of varying degrees in the adrenals, liver, brain, lungs, and kidneys. The changes were typical of anoxia and similar to those produced by prolonged asphyxia and carbon monoxide poisoning. Malamud, Haymaker, and Custer (1946), in a clinicopathologic study of 125 fatal cases of heat stroke, found that changes in the central nervous system were most con- spicuous and consisted of (1) progressive degeneration of neu- rons and replacement by glia, especially in the cerebellum, cere- bral cortex, and basal ganglia, and sparing the hypothalamus and brain stem; and (2) congestion, edema, and hemorrhages, most commonly in the region of the third ventricle, the aqueduct, and the fourth ventricle, all of which were inconstant and regarded as terminal. In the opinion of the authors, the cellular changes were caused by excessive heat and the hemorrhages by shock. More recent reports have invoked fibrinolysis and afibrinogenemia as the cause of hemorrhages in fatal heat stroke (Shibolet et al., 1962). In experiments in dogs with temperatures maintained for at least 1 hour above 41.7° C (107°F), no disturbance in motor func- tion or behavior was observed, provided that intravenous saline 182 EXPERIMENTAL FEBRILE CONVULSIONS to replenish fluid and salt deficits and oxygen to combat anoxemia were employed; acute parenchymal lesions of the brain at the time of the hyperthermia were apparently reversible in animals which survived (Dobin et ah, 1949). These authors cite many instances of reported recovery from hyperpyrexia in patients, but a cerebellar syndrome following protracted hyperthermia and heat stroke is not an infrequent finding (Cruickshank, 1951; Freedman and Schental, 1953; Gottschalk and Thomas, 1966). Indeed, in the pathology of heat stroke, changes in the cerebellum are more striking, more consistent, and more rapid in develop- ment than in any other part of the brain (Malamud et ah, 1946). If brain damage is a frequent complication of febrile convulsions, as suggested by M. A. Lennox (1949), Schmidt and Ward (1955), Fowler (1957), and others, cerebellar ataxia might be expected as a common sequela. A review of the literature has shown that neurologic deficits attributable to febrile convulsions are rare and consist chiefly of hemiparesis (Chapter 5). Neurogenic Hyperthermia. Fever as a complication of various brain lesions has been documented in children (Akerren, 1946; Ylppo and Ylppo, 1956; O. T. Bailey, 1959) and should be con- sidered as a possible explanation for convulsions with fever of unknown origin, particularly if the child has signs of neurologic impairment. Friedman (1931) reported fever as a consequence of a brain tumor in one adult, and Erickson (1939) referred to neurogenic hyperthermia following head injury or operation in the region of the third ventricle or posterior fossa. In children with thrombosis of the superior saggital sinus, elevations in tempera- ture to 41.1° C (106°F) and even 42.8° C (109°F) were attributed to interference with the central control of temperature regulation, but necroses and hemorrhages in the cerebral hemispheres were so extensive that a single lesion could not be designated as the one responsible for the fever (O. T. Bailey, 1959), Convulsions and Brain Lesions Febrile convulsions are usually short in duration and clonic in pattern, and brain pathology as a result of prolonged asphyxia and tonic spasm is an uncommon complication. ARTIFICIAL FEVER AND HYPERPYREXIA 183 Zimmerman (1938) described the histopathological findings in seven children who died during or after severe convulsions or respiratory failure, occurring in the course of a serious infectious illness, and in nine children who died as a result of either severe convulsions or asphyxia, unassociated with infectious illness. The lesions were similar to those described by Spielmeyer (1930) and Scholz (1951) in patients with epilepsy, and consisted of extensive necrobiotic changes of the ganglion cells in the cerebral cortex, especially the cornu ammonis of the temporal lobes, the basal ganglia, and the cerebellar cortex. Severe convulsions and pro- longed periods of respiratory difficulty or asphyxia dominated the clinical pictures, and although seven of the children had an underlying infectious illness that suggested the possibility of encephalitis as the cause of the symptoms, none had inflammatory cerebral lesions at autopsy. It was concluded that cerebral anox- emia, resulting from the convulsions, the respiratory difficulty, or the asphyxia, was responsible for the pathologic condition in the nervous system. Meyer, Beck, and Shepherd (1955) reported the pathologic findings in a child aged 9 years who died 2 years fol- lowing the onset of Still’s disease complicated later by status epilepticus and bacterial infections with toxemia. Brain lesions were diffuse but most severe in the first temporal convolution, adjacent Sylvian areas, Ammon’s horn, the hippocampal gyrus, medial orbital, and anterior cingular regions. The pathologic find- ing was attributed to status epilepticus and was considered sec- ondary to circulatory and anoxic disturbances. The case history was certainly not characteristic of a febrile convulsive disorder and, as in several of the examples of brain pathology with child- hood fevers cited in the literature, the patient had a previous history of retarded motor and speech development suggestive of antecedent neurologic deficits. SUMMARY AND COMMENT Experimental febrile convulsions in animals are induced most satisfactorily by use of microwave diathermy, and the body temperature at the onset of generalized clonus is a measure of the 184 EXPERIMENTAL FEBRILE CONVULSIONS convulsive threshold. Factors found to modify the threshold con- vulsive temperature include age, biochemical and anatomical maturation of the brain, electrolyte imbalance, and drugs. Short exposures to moderate and even excessive degrees of fever are generally not complicated by brain damage, but prolonged hyper- pyrexia, especially when associated with dehydration and anox- emia, may result in permanent pathology. Investigations involv- ing artificial hyperthermia in children are limited to observations of electroencephalographic abnormalities and are inadvisable on the basis of potential hazards and ethical considerations. Future clinical research should be directed toward elucidation of (1) the epidemiology of febrile convulsions in all parts of the world; (2) the exact role of viral and bacterial infections in eti- ology; (3) the nature of genetic and presumed enzyme defects that may increase susceptibility to febrile seizures in some patients and their families; (4) factors related to anatomic, physiologic, and biochemical immaturity in causation of a lowered febrile seizure threshold; (5) the significance of toxins, lack of antibodies, and immune reactions to infections and drugs; and (6) the value of new anticonvulsant and antipyretic medications in the pre- vention of febrile convulsions. BIBLIOGRAPHY Akerren, Y.: Les symptomes cerebraux et les lesions cerebrales post-mortelles dans des etats morbides malins, aigus, hyper- febriles, se declarant pendant 1’enfance, Acta Med Scand suppl 170: 705-732, 1946. Almeida, M. 0., Cavalcanti, T., and Dias, M. V.: Sobre o ataque de convulsoes generalizadas produzido pelo reaquecimento, Mem Inst Cruz 48: 351-355, 1950, Altschule, M. D.: Mechanisms of fever, Med Sci 7: 694-699 (May 25), 1960 (review article). Ashby, W., and Schuster, E. M.: Carbonic anhydrase in brain of newborn in relation to functional maturity, J Biol Chem 184: 109-116 (May), 1950. Atkins, E.: Pathogenesis of fever, Physiol Rev 40: 580-646 (July), 1960. Atkins, E.: Fever and its pathogenesis, Med Sci 11:57-74 (Jan. 10), 1962. Bailey, A. A., Rooke, E. D., and Rodin, E. A.: Investigation of a bacterial pyrogen as a therapeutic agent in neurologic dis- orders, Proc Mayo Clin 27: 340-343 (Aug. 13), 1952. 185 186 BIBLIOGRAPHY Bailey, O. T.: Results of long survival after thrombosis of the superior sagittal sinus, Neurology (Minneap) 9: 741-746 (Nov.), 1959. Baird, H. W., 111, and Garfunkel, J. M.: Electroencephalo- graphic changes in children with artificially induced hyper- thermia, / Pediat 48: 28-33 (Jan.), 1956. Bamatter, F., Gautier, A., and Jeanneret, O.; Une etiologie trop peu connue des convulsions dans la premiere enfance: la fievre de 3 jours avec exantheme subit, Praxis 47: 1093-1096 (Nov. 13), 1958. Bamberger, Ph., and Matthes, A.: Anfdlle im Kindesalter, S. Karger, Basel, Switzerland, 1959, pp. 365-432, Barcroft, J.: The Respiratory Function of the Blood, Cambridge U. P., Cambridge, England, 1914. Barenberg, L. H., and Greenspan, L.: Exanthema subitum (ro- seola infantum), Am J Dis Child 58: 983-993 (Nov.), 1939. Bass, D. E., and Henschel, A.; Responses of body fluid com- partments to heat and cold, Physiol Rev 36: 128-144 (Jan.), 1956. Bennett, I. L., Jr., and Beeson, P. M.: Studies on pathogenesis of fever; effect of injection of extracts and suspensions of un- infected rabbit tissues upon body temperature of normal rab- bits, / Exp Med 98: 447-492 (Nov.), 1953. Bennett, I. L., Jr., and Cluff, L. E.; Bacterial pyrogens, Pharma- col Rev 9: 427,1957. Bergemann, E.: Über das spatere Schicksal von Kindern mit Krampfanfallen, Mschr Kinderheilk 65: 116-140 (March), 1936. Berger, A., Elenbogen, G. D., and Ginger, L. G.: Pyrogens, Advances Chem 16: 168-197, 1956. Bertoye, A.; Convulsions et maladies infectieuses, Rev Prat 13: 763-768 (Mar. 1), 1963. Bigler, J. A., and McQuiston, W. O.: Body temperatures during anesthesia in infants and children, JAMA 146: 551-556 (June 9), 1951. Binet, L., Laudat, M., and Auclair, J.; Abaissement de la reserve alcaline et mouvement du chlore dans de sang au cours de Thyperthermie provoquee par les ondes courtes, C R Acad Sci (Paris) 199: 422-444 (Aug. 6), 1934. BIBLIOGRAPHY 187 Bischoff, F., Long, M. L., and Hill, E.: Studies in hyperthermia. 11. The acid-base equilibrium in hyperthermia induced by short radio waves, J Biol Chem 90: 321-329 (Jan.), 1931. Bischoff, F., Maxwell, L. C., and Hill, E.: Studies in hyper- thermia. 111. The phosphorus equilibrium, / Biol Chem 90: 331-339 (Jan.), 1931. Bjerglund, R., and Brandt, S.: Elektroencefalografiske fund hos born med “febrile kramper,” Ugesk Laeg 116: 1423-1428 (Oct. 7), 1954. Boldrey, E. E., and Millichap, J. G.; Barometric pressure and seizures, Proc Soc Exp Biol Med 123: 968-970 (Dec.), 1966, Breg, W. R., and Yannet, H.: The child in a convulsion, Pediat Clin N Am 9: 101-112 (Feb.), 1962. Bridge, E. M.: Epilepsy and Convulsive Disorders in Children, McGraw-Hill, New York, 1949, p. 607. Broberger, O.; Exanthema subitum och feberkramper, Nord Med 59: 523-525 (April 10), 1958. Brown, H. R., Jr., and Clark, W. F.: Plasma specific gravity and control of fluid administration in artificial fever, Proc Soc Exp Biol Med 40: 490-494 (March), 1939. Brown, J. E., Jr.: Convulsions in infancy. A statistical study of 400 cases, Ohio Med J 31: 423-430 (June), 1935. Brown, W. C.; Properties and alterations of electrically-induced seizures in mice, Epilepsia (Amst) 2: 127-137 (Nov.), 1953. Brusa, P., Bizzi, G., and Breschi, F.; La prognosi a distanza delle convulsioni infantili, Minerva Pediat 7: 25-32 (Jan. 13), 1955. Calzetti, G., and Borghi, G.: Le convulsioni con iperpiressia (contribute clinico-statistico), Clin Pediat 34: 473-492 (July), 1952. Carter, C. H.: Phetharbital in epilepsy, Pediat Digest 4: 853- 858 (Oct), 1962. Carter, S.; Diagnosis and treatment; management of the child who has had one convulsion, Pediatrics (New York) 33: 431- 434 (March), 1964 (editorial review article). Cary, W.; Febrile convulsions in infancy and childhood, Med J Aust 43: 254-256 (Aug. 18), 1956. Cassels, W. H., Becker, T. J., and Seevers, M. H.; Convulsions 188 BIBLIOGRAPHY during anesthesia. An experimental analysis of the role of hy- perthermia and respiratory acidosis, Anesthesiology 1: 56-68 (July), 1940. Castiglione, J. B.: The child who has had one convulsion, Pedi- atrics (New York) 34: 890 (Dec.), 1964 (letter to the Editor). Cavazzuti, G. 8., et ah: Rev Neurol (Napoli) 27: 332-349, 1957; Clin Pediat 43: 608-621, 1961; Arch Franc Pediat 18: 389-391, 1961; G Psichiat Neuropat 89: 499-534, 1961. Chandy, J., Singh, 8., and Isaiah, P.; Fits in children. A review of 60 consecutive cases, Indian J Child Health 3: 263-270 (June), 1954. Chao, D. H-C., Druckman, R., and Kellaway, P.: Convulsive Disorders of Children, W. B. Saunders, Philadelphia, 1958. Chaptal, J., Passovant, P., Jean, R., Campo-Bechard, C., Barjon, P., and Ribstein, M.; Convulsions chez le nourisson et chez I’enfant. Etude statistique, clinique, electroencephalographique et therapeutique des 424 cas observes dupuis janvier 1948, Pedi- atrie 8: 629-637,1953. Chopra, D. R.: Significance of febrile convulsions and their management (a short review with some observations), Burma Med J 9: 126-230 (July), 1961. Clemens, H. H.: Exanthem subitum, J Pediat 26: 66-77, 1945. Coelho, G.: PyrexHl convulsions in children, Med, Bull 10: 290, 1942; Indian ] Child Health 6: 244-248 (May), 1957. Conel, J. L.; The Postnatal Development of the Human Cere- bral Cortex, Harvard U. P., Cambridge, Mass., 1939. Cooke, R. T.: Fulminating sonne dysentery; report of 2 cases, Brit Med J 2: 151-152 (Aug. 3), 1940. Costeff, H.: Convulsions in childhood. Their natural history and indications for treatment, New Eng J Med 273: 1410-1413 (Dec. 23), 1965. Cramblett, H. G., Moffet, H. L., Black, J. P., Schulenberger, H., Smith, A., and Colonna, C. T.: Coxsackie virus infections: clinical and laboratory studies, J Pediat 64: 406-414 (March), 1964. Crawford, J. D., and Dodge, P. R.: Complications of fluid ther- apy in patients with neurologic diseases, with special emphasis on water intoxication and hypertonic dehydration, Pediat Clin N Amer 6: 257-279 (Feb.), 1959. BIBLIOGRAPHY 189 Cruickshank, E. K.: Hyperpyrexia with cerebellar damage in acute rheumatism, Lancet 1: 1052-1053 (May), 1951. Comings, J. N., Goodwin, H., Woodward, E. M., and Curzon, G.: Lipids in the brains of infants and children, J Neurochem 2: 289-294,1958. Daly, C. A., and Knudson, A.: Acid-base equilibrium and phos- phorus metabolism in hyperthermia, J Biol Chem 97:57-58 (July), 1932. Darrow, D. C., and Yannet, H.: Changes in the distribution of body water accompanying increase and decrease in extracellu- lar electrolyte, J Clin Invest 14: 266-275 (March), 1935. Davis, E.: Temporary hemiplegia following symptomatic con- vulsions, Arch Dis Child 14: 87-88 (March), 1939. Dees, S.: Electroencephalography in allergic epilepsy, Southern Med J 46: 618-620 (June), 1953. Dejong, R. N.: The Neurologic Examination, Hoeber, New York, 1967. Dekaban, A.: Neurology of Infancy, Williams & Wilkins, Balti- more, 1959. Dennison, G. A., and deHoll, G.: Bacillary dysentery in infants and children; clinical and bacteriologic study of 35 cases, J Infect Dis 56: 124-141, 1935. Dobin, N. 8., Neymann, C. A., and Osborne, S. L.: Pathologic changes in the central nervous system resulting from experi- mentally produced hyperpyrexia, J Neuropath Exp Neurol 8: 295-304 (July), 1949. Dobrzynska, L.: Drgawki goraczkowe, Pediat Pol 38: 1501- 1506,1963. Dobrzynska, L.: Drgawki dzieciece a padaczka u dzieci, Neurol Neurochir Psychiat Pol 13: 797-805 (Nov.-Dee.), 1963. Dodge, P. R.: Neurologic history and examination, in Farmer, T. W. (ed.): Pediatric Neurology, Hoeber, New York, 1964. Donald, W. D., Winkler, C. H., Jr., and Bargeron, L. M., Jr.: The occurrence of convulsions in children with shigella gastro- enteritis, I Pediat 48: 323-327 (March), 1956. Doose, H., Volzke, E., Petersen, C. E., und Herzberger, E.: Fie- berkrampfe und Epilepsie. 11. Elektroencephalographische Verlaufsuntersuchungen bei sogenannten Fieber- oder Infekt- krampfen, Arch Psychiat Nervenkr 208: 413-432, 1966. 190 BIBLIOGRAPHY Ehrengut, W., and Ehrengut, J.: Fieberkrampfe bei Antikor- permangel-zustanden, Deutsch Med Wschr 89: 166-170 (Jan), 1964. Eichenwald, H. F., and Kotsevalov, O.: Immunologic response of premature and full-term infants to infection with certain viruses, Pediatrics (New York) 25: 829-839 (May), 1960. Ekholm, E., and Niemineva, K.: On convulsions in early child- hood and their prognosis, an investigation with follow-up exam- inations of patients treated for convulsions at the Childrens Clinic of Helsinki University, Acta Paediat 39: 481-501, 1950. Erickson, T. G.: Neurogenic hyperthermia (a clinical syndrome and its treatment), Brain 62: 172-190 (June), 1939. Esplin, D. W., and Laffan, R. J.: Determinants of flexor and extensor components of maximal seizures in cats, Arch Int Pharmacodyn 113: 189,1957. Faerber, E.: Klinische Beobachtungen iiber die Initialkrampfe im Kindesalter, Jahrb Kinderheilk 124; 148-158, 1929. Faxen, N.: Le pronostic des convulsions de I’enfance, Rev Franc Pediat 11: 665-688, 1935. Felsen, J., Rundlett, E. V., Sullivan, J., and Gorenberg, H.: Atypical flexner dysentery; preliminary report of Jersey City epidemic, JAMA 103: 1055-1058 (Oct. 6), 1934. Ferris, E. 8., Jr., Blankenhorn, M. A., Robinson, H. W., and Cullen, G. E.; Heat stroke: clinical and chemical observations on 44 cases, J Clin Invest 17: 249-262 (May), 1938, Fischler, E.: Convulsions as a complication of shigellosis in children, Helv Paediat Acta 17: 389-394, 1962. Fois, A., and Malandrini, F.: Le convulsioni con febbre, Riv Clin Pediat 59: 481-488 (June), 1957. Forbes, G. B.: The febrile convulsion, Missouri Med 51: 723- 726 (Sept.), 1954. Fowler, M.: Brain damage after febrile convulsions, Arch Dis Child 32: 67-76 (April), 1957. Frantzen, E., Nygaard, A., and Wulff, H.: Febrile kramper hos bprn. Prognostiske studier og forspg pa vurdering af effekten af profylaktisk antiepileptisk langtids-behandling en forelpbig meddelelse, Ugeskr Laeg 126:207-210 (Feb.), 1964. BIBLIOGRAPHY 191 Freedman, D. A., and Schental, J. E.; A parenchymatous cere- bellar syndrome following protracted high body temperature, Neurology (Minneap) 3:513-516 (July), 1953. Freund, W.; Monatsschr f Kindern 57: 195-204, 1933. Friderichsen, C., and Melchior, J.: Febrile convulsions in chil- dren, Acta Paediat Suppl 43: 307-317 (Oct), 1954. Friedman, E. D.: Two cases of cerebral fever, Med Clin N Amer 14: 1419-1431 (May), 1931. Fruthaler, G. J., and Tilden, T.: Management of hyperpyrexia in children, Postgrad Med 35: 643-646 (June), 1964. Gallant, L. J., and Livingston, S.: An electroencephalographic and clinical study of children with febrile convulsions, Trans Amer Neurol Ass 74: 201-205, 1949. Gandhi, V. K.: Convulsions in children, Indian J Child Health 12: 502-513 (Aug.), 1963. Gastaut, H., and Gastaut, Y.: Syncopes et convulsions. A propos de la nature syncopale de certaines spasmes du sanglot et de certaines convulsions essentielles hyperthermiques ou a froid, Rev Neurol (Paris) 96; 158-163 (Feb.), 1957. Gellhorn, E., and Heymans, C.: Differential action of anoxia, asphyxia and carbon dioxide on normal and convulsive poten- tials, /Neurophysiol 11: 261-273 (May), 1948. Gibbs, E. L., Fleming, M. M., and Gibbs, F. A.; Diagnosis and prognosis of hypsarhythmia and infantile spasms, Pediatrics 13: 66-73 (Jan.), 1954. Gibbs, F. A., and Gibbs, E. L.; Atlas of Electroencephalography, 2nd ed., Vol. I. Methodology and Controls, Addison-Wesley, Cambridge, Mass., 1950, 327. Ginger, L. G., Windle, W. F., and Johnson, I. E.: Bacterial Pyrogens. An Annotated Bibliography, Baxter Laboratories, Morton Grove, 111., 1952. Giraud, P., Bernard, R., and Vincent, P.: Bilan actuel des con- vulsions infantiles observees a la clinique medicale infantile au cours des dix dernieres annees, Pediatrie 8: 625-627, 1953. Goodman, L. S., and Gilman, A.: The Pharmacologic Basis of Therapeutics: A Textbook of Pharmacology, Toxicology, and Therapeutics for Physicians and Medical Students, 2nd ed., Macmillan, New York, 1955, 1831. 192 BIBLIOGRAPHY Gordon, E. E., Darling, R. C., and Shea, E.: Effects of physical hyperthermia upon blood gas equilibria in man, J Appl Physiol 1:496-511 (Jan.), 1949. Gottschalk, B. G., and Thomas, J. E.: Heat stroke, Proc Mayo Clin 41: 470-482 (July), 1966. Grace, H. B.: Convulsions and hemiplegia in pertussis prophy- laxis, Canad Med Ass J 63: 129-131 (Aug.), 1950. Greenthal, R. M.: Roseola infantum (exanthem subitum), Wis- consin Med J 40: 25-27 (Jan,), 1941. Guthrie, R. H.: Influence of intercurrent febrile disorders on pre- existing epilepsy, Arch Neurol Psychiat, 24: 753-758 (Oct), 1930. Habel, K., and Lucchesi, P. F.: Convulsions complicating per- tussis: a clinical study, Amer J Dis Child 56: 275-286 (Aug.), 1938. Haldane, J. B. S.; The influence of high air temperatures, / Hyg (Camb) 5: 494, 1905 (quoted by Yamakita, M., 1921). Hammill, J. F., and Carter, S.: Febrile convulsions, New Eng J Med 274: 563-565 (Mar. 10), 1966 (current concepts, review article). Hardy, A. V., and Watt, J.: Studies of acute diarrheal diseases. XIV. Clinical observations, Public Health Rep 60: 521-531 (May 11), 1945. Harrell, G. T., Jr., and Aikawa, J. K: Alterations in the permea- bility of membranes during infection, JAMA 147; 232-238 (Sept. 15), 1951. Hartman, F. W.: Lesions of the brain following fever therapy, etiology and pathogenesis, JAMA 109: 2116-2121 (Dec. 25), 1937. Herlitz, G.; Studien iiber die sog. initialen Fieberkrampfe bei Kindern, Acta Paediat (Suppl 1) 29: 1-142, 1941. Hoagland, H.: Pacemakers of human brain waves in normals and in general paretics, Amer J Physiol 116: 604-615 (Aug.), 1936. Hoagland, H., Rubin, M. A., and Cameron, D. E.; Brain wave frequencies and cellular metabolism. Effects of dinitrophenol, J Neurophysiol 2: 170-172 (March), 1939. BIBLIOGRAPHY 193 Hoagland, R. J., and Bishop, R. H., Jr.: A physiologic treatment of heat stroke, Amer J Med Sci 241: 415-422 (April), 1961. Hochsinger: Deutsch Klinik 7: 500, 1905 (quoted by Faerber, E.: 1929). Holliday, P. 8., Jr.: Pre-eruptive neurological complications of the common contagious diseases—rubella, rubeola, roseola, and varicella, / Pediat 36; 185-198 (Feb.), 1950. Horstmann, W., and Schinnerling, W.: Zur Prognose de sog. Fieberkrampfe, Mschr Kinderheilk 111: 52-57 (Jan.-June), 1963. Hoyt, R., and Rosvold, H. E.: Effect of electroconvulsive shock on body temperature of the rat, Proc Soc Exp Biol Med 78; 582-583 (Nov.), 1951. Hrbek, A.: Fieberkrampfe im Kindesalter, Ann Paediat {Basel) 188: 162-182 (March), 1957. Hull, C. D., Buchwald, N. A., Dubrovsky, 8., and Garcia, J.: Brain temperature and arousal, Exp Neurol 12: 238-246 (July), 1965. Husler, J.: Zur systematik u klinik epileptiformer Krampfkrank- heiten im Kindesalter, Ergebn Inn Med Kinderheilk 19: 624, 1921 (quoted by Faerber, E., 1929 and Lennox, M. A., 1949). Huttenlocher, P. R.; Seizures in childhood, New Eng J Med 273: 857-858 (April 14), 1966 (letter to the Editor). Ikeda, T,: Experimental study on the pathogenetic mechanism of febrile convulsions. Part 11. Biochemical study on the blood in experimental febrile convulsions, Acta Paediat Jap 66: 207- 221 (April), 1962. Isch-Treussard, C., and Rohmer, F.; L’electroencephalogramme dans les convulsions de I’enfant de moins de 3 ans: interet diagnostique et prognostique a-propos de 250 cas suivis, Arch Franc Pediat 16: 822-825, 1959. Jacobowsky, B.: Sugar content of blood in fever, Acta Soc Med Upsal 28:215-236 (Feb.), 1923 (illus.); abstract, JAMA 80: 1038 (April 7), 1923. Janeway, C. A.: Infection, immunity and allergy in relation to pediatrics, in Nelson, W, E. (ed.): Textbook of Pediatrics, W. B. Saunders, Philadelphia, 1963, pp. 381-384. 194 BIBLIOGRAPHY Jennings, C. G.: Febrile convulsions in childhood. Review of their significance, J Mich Med Soc 53: 272-274 (March), 1954. Kanof, A., and Volk, B. W.: Plasma fibrinogen in the diagnosis of neurogenic hyperthermia, J Pediat 59: 674-683 (Nov.), 1961. Kao, Y. E.: Convulsions in childhood. A statistical study of 1125 cases with special reference to its occurrence in various types of diseases, Chin Med J 69: 400-406 (Sept.-Oct.), 1951. Kaplan, M., Mozziconacci, P., Lerique, and Picard: L’electro- encephalogramme dans les convulsions d’hyperpyrexie de I’enfant, Sent Hop Paris 24; 3240-3244 (Dec. 30), 1948. Karlsson, B.: Om feberkrampens langtidsprognos, Svensk La- kartidn 54; 601-609 (March 1), 1957. Kashiwase, Y.: An electrophysiological study of effects of febrile state on seizure discharge in relation to clinical problems of the febrile convulsion, Brain Nerve (Tokyo) 14: 698-715, 1962. Keith, H. M.: Convulsive Disorders in Children, Little, Brown, Boston, 1963, p. 44. Keith, H. M.: Convulsions in children under three years of age: a study of prognosis, Proc Mayo Clin 39: 895-907 (Dec.), 1964. Kellaway, P., and Fox, B. J.: Electroencephalographic diagnosis of cerebral pathology in infants during sleep; rationale, tech- nique, and characteristics of normal sleep in infants, J Pediat 41:262-287 (Sept.), 1952. Kennedy, C.: Cerebral metabolic rate in children, in Korey, S. R., and Nurnberger, J. I. (eds.): Neurochemistry: Progress in Neurobiology, Vol. I, Hoeber, New York, 1956, p. 244. Kety, S. S.; Changes in cerebral circulation and oxygen con- sumption which accompany maturation and aging, in Waelsch, H. (ed.): Biochemistry of the Developing Nervous System, Academic Press, New York, 1955, pp. 208-217. Keutmann, E. H., Bassett, S. H., and Warren, S. L.: Electrolyte balances during artificial fever with special reference to loss through the skin, J Clin Invest 18: 239-250 (March), 1939. Kohn, R., and Millichap, J. G.: Properties of seizures induced by histamine, Proc Soc Exp Biol Med 99: 623-628, 1958. Kowlessar, M., and Forbes, G. B.: The febrile convulsion in shigellosis, New Eng J Med 258: 520-526 (March 13), 1958. BIBLIOGRAPHY 195 Krakau, C. E. T., and Nyman, G. E.: On effect of artificial fever on the alpha activity in man, Acta Fhysiol Scand 29: 281-292, 1953. Laplane, R., Fischgold, H., and Brisac, C.: Le probleme de I’epilepsie debutante ou atypique de I’enfant du point du vue electroencephalographique, Arch Franc Pediat 4: 375-377, 1947. Laplane, R., Humbert, R., Laget, P., Salbreux, R., and Debray, P.: Suites immediates et lointaines des convulsions febriles avant quatre ans, Rev Neurol 99: 26-38 (July), 1958. Laplane, R., and Salbreux, R.: Les convulsions hyperpyretiques Rev Prat 13: 753-761 (March 1), 1963. Lederer, E.; fiber Bedeutung und folgen der Krampfe im Sauglings und Kindesalter, Arch Kinderheilk 102: 1-15, 1934. Lenggenhager, K.: Zur Genese der Fieberkrampfe im Kinde- salter, Schweiz Med Wschr 82: 390-392 (April 12), 1952. Lennox, M. A.: Febrile convulsions in childhood, Proc Ass Res Nerv Meat Dis (Chapter XXIV) 26: 342-365, 1947. Lennox, M. A.; Febrile convulsions in childhood: their relation- ship to adult epilepsy, J Pediat 35: 427-435 (Oct.), 1949. Lennox, M. A.: Febrile convulsions in childhood; a clinical and electroencephalographic study, Amer J Dis Child 78; 868-882 (Dec.), 1949. Lennox, M. A., Sibley, W. A., and Zimmerman, H. M.: Fever and febrile convulsions in kittens. A clinical, electroencepha- lographic, and histopathological study, J Pediat 45: 179-190 (Aug.), 1954. Lennox, W. G.: Significance of febrile convulsions, Pediatrics {NewYork) 11: 341-357 (April), 1953. Lennox, W. G., and Lennox, M. A.: Epilepsy and Related Dis- orders, Vol. I, Little, Brown, Boston, 1960. Lennox-Buchthal, M.: Febrile kramper hos born. EEG-fund og kliniske korrelationer en forelobig meddelelse, Ugeskr Laeg 126: 203-206 (Feb. 13), 1964. Leonard, C. A., and Harrison, J. W. E.; Effect of hyperthermic shock on “free and bound” potassium levels, J Am Pharm Ass 45: 116-120 (Feb.), 1956. 196 BIBLIOGRAPHY Lerique-Koechlin, A.: Colloque sur Telectroencephalogramme dans les convulsions de la premiere enfance, Rev Neurol {Paris) 99: 1-10 (July), 1958. Livingston, S.; The Diagnosis and Treatment of Convulsive Dis- orders in Children, Charles C. Thomas, Springfield, 111., 1954. Livingston, S.: The child who has had one convulsion, Pediatrics (New York) 33: 1001-1002 (June), 1964 (letter to the Editor). Livingston, S., Bridge, E. M., and Kajdi, L.: Febrile convul- sions: a clinical study with special reference to heredity and prognosis, J Pediat 31: 509-512 (Nov), 1947. Lormans, J.: Hyperthermische stuipen, Belg T Geneesk 18: 708- 710 (Aug 1), 1962. Lyon, G., Dodge, P. R., and Adams, R. D.: The acute encepha- lopathies of obscure origin in infants and children, Brain 84: 680-708,1961. MacDougall, L. G.: Low cerebrospinal fluid protein in African children with febrile convulsions, Arch Dis Child 37: 309-313 (June), 1962. McQuarrie, I.; Epilepsy in children: relationship of water bal- ance to the occurrence of seizures, Am J Dis Child 38: 451-467 (Sept), 1929. Malamud, N., Haymaker, W., and Custer, R. P.: Heat stroke. A clinico-pathologic study, Mil Surgeon 99: 397-449, 1946. Medlinsky, H. L.: The child who has had one convulsion, Pedi- atrics (New York) 34: 438, 1964 (letter to the Editor). Mehta, J. B.: Convulsions in children, Indian J Child Health 10: 485-488 (Oct), 1961. Melin, K.-A.: Msehr Kinderheilk 102; 62-63,1954. Mendez, M.: Epilepsia y regulation termica, Rev Neurpsiquiat 8: 50-56 (March), 1945. Merritt, H. H.: A Textbook of Neurology, 3rd ed., Lea & Feb- iger, Philadelphia, 1963. Meyer, A., Beck, E., and Shepherd, M.: Unusually severe lesions in the brain following status epilepticus, J Neurol Neurosurg Psychiat 18: 24-33 (Feb), 1955. Meyer, J. S., and Handa, J.: Cerebral blood flow and metabo- lism during experimental hyperthermia (fever), Minnesota Med 50: 37-44 (Jan.), 1967. BIBLIOGRAPHY 197 Miller, F. J. W., Court, S. D. M., Walton, W. S., and Knox, E. G.; Growing up in Newcastle-upon-Tyne: A Continuing Study of Health and Illness in Young Children Within Their Families, Published for the Nuffield Foundation, Oxford U. P., New York, 1960, pp. 164-173. Millichap, J. G.: Methods of evaluation of new anticonvulsant compounds, Neurology (Minneap) 6; 484-490 (July), 1956. Millichap, J. G.: Development of seizure patterns in newborn animals. Significance of brain carbonic anhydrase, Proc Soc Exp Biol Med 96; 125-129,1957. Millichap, J. G.: Seizure patterns in young animals. Significance of brain carbonic anhydrase. 11. Proc Soc Exp Biol Med 97: 606-611,1958 (i). Millichap, J. G.: The febrile seizure threshold in relation to maturation of brain water, electrolyte, and acid-base balance of newborn animals (abst), Amer J Dis Child 96: 492-493, 1958 (ii). Millichap, J. G.; Studies in febrile seizures. I. Height of body temperature as a measure of the febrile seizure threshold, Pedi- atrics (New York) 23: 76-85 (Jan.), 1959. Millichap, J. G.: Studies in febrile seizures. 11. Febrile seizures and the balance of water and electrolytes, Neurology (Min- neap) 10; 312-322 (March), 1960(i). Millichap, J. G.: A potential new therapy for febrile seizures, preliminary report on development of N-phenylbarbitone, Brit Med] I: 1111-1112 (April 9), 1960(h). Millichap, J. G.: Diagnosis and management of convulsive dis- orders in children with special emphasis on seizures amenable to specific therapies, Pediat Clin N Amer 7: 583-603 (Aug.), 1960 (hi). Millichap, J. G.: Treatment of convulsive disorders, including febrile seizures in children, in Dejong, R. N. (ed.): Modern Treatment, Hoeber, New York, 1: 1087-1103 (Sept.), 1964. Millichap, J. G.: Anticonvulsant drugs, in Root, W. S., and Hof- mann, F. G. (Eds,): Physiological Pharmacology, Vol. 11, Aca- demic Press, New York, 1965, pp. 97-173. Millichap, J. G.; Treatment of febrile convulsions, New Eng J Med 274: 1329 (June), 1966 (letter to the Editor). 198 BIBLIOGRAPHY Millichap, J. G.: The effects of fever on the central nervous sys- tem, Lecture presented at The Johns Hopkins Hospital, and sponsored by the Joseph P. Kennedy, Jr., Memorial Foundation, Baltimore, in press. Millichap, J. G., Lombroso, C. T., and Lennox, W. G.: Cyclic vomiting as a form of epilepsy in childhood, Pediatrics (New York) 15: 705-714 (June), 1955(i). Millichap, J. G., Woodbury, D. M., and Goodman, L. S.: Mech- anism of the anticonvulsant action of acetazolamide, a carbonic anhydrase inhibitor, J Pharmacol Exp Ther 115: 251-258 (Nov.), 1955(h). Millichap, J. G., Balter, M., and Hernandez, P.: Development of susceptibility to seizures in young animals. 111. Brain water, electrolyte, and acid-base metabolism, Proc Soc Exp Biol Med 99:6-11 (Oct), 1958. Millichap, J. G., Aledort, L. M., and Madsen, J. A.: A critical evaluation of therapy of febrile seizures, J Pediat 56: 364-368 (March), 1960(i). Millichap, J. G., Hernandez, P., Zales, M. R., Halpern, L. A., and Kramer, B. I.: Studies in febrile seizures. IV. Evaluation of drug effects and development of potential new therapy (Py- rictal), Neurology (Minneap) 10:575-583 (June), 1960(h). Millichap, J. G., Madsen, J. A., and Aledort, L. M.: Studies in febrile seizures. V. A clinical and electroencephalographic study in unselected patients, Neurology (Minneap) 10:643-653 (July), 1960 (hi). Millichap, J. G., Bickford, R. G., Klass, D. W., and Backus, R. E.: Infantile spasms, hypsarhythmia, and mental retardation. A study of etiologic factors in 61 patients, Epilepsia (Amst) 3: 188-197 (June), 1962. Millichap, J. G., and Jones, J. D.: Acid-base, electrolyte, and amino-acid metabolism in children with petit mal. Etiologic significance and modification by anticonvulsant drugs and the ketogenic diet, Epilepsia {Amst) 5: 239-255,1964. Minard, D., and Copman, L.: Elevation of body temperature in disease, in Hardy, J. D. (ed.): Temperature, Its Measurement and Control in Science and Industry, Vol. 3, Part 3, Biology and BIBLIOGRAPHY 199 Medicine, Reinhold, New York, and Chapman & Hall, London 1963, pp. 253-273. Moffet, H. L., and Cramblett, H. G.: Viral isolations and ill- nesses in young infants attending a well-baby clinic, New Eng J Med 267: 1213-1218 (Dec. 13), 1962. Moller, K. L.: Exanthema subitum and febrile convulsions, Acta paediat4s: 534-540 (Sept.), 1956. Moscovici, C., Ginevri, A., and Kempe, C. H.: Distribution of poliomyelitis and ECHO viruses in children’s institution, / Dis Child 98: 139-143,1959. Neyroud, M.; Etude clinique et electroencephalographique des convulsions hyperpyretiques, Praxis 36: 459-465 (Nov. 6), 1947. Nyhan, W. L., and Cooke, R. E.: Symptomatic hyponatremia in acute infections of the central nervous system, Pediatrics 18: 604-613 (Oct.), 1956. Osborne, S. L.: Hyperpyrexia and the specific gravity of blood, Arch Phys Ther 22: 407-409 (July), 1941. Ounsted, C.: The factor of inheritance in convulsive disorders in childhood (abridged), Proc Roy Soc Med 45: 865-868 (Dec.), 1952. Owens, G.: Further studies on ether convulsions; fat emboliza- tion and its association with neurologic deficits, Neurology (Minneap) 8: 827-831 (Nov.), 1958. Pache, H. D.: Krampfe im Kindesalter, Mschr Kinderheilk 102: 42-49,1954. Paine, R. S.: Neurologic examination of infants and children, in Perlstein, M. A. (ed.): Symposium on Neuropediatrics, Pediat Clin N Amer7: 471-510 (Aug.), 1960. Paine, R. S., and Oppe, T. E.: Neurological examination of children, in Clinics in Developmental Medicine, Vol. 20/21, Spastics Society/Heinemann, London, 1966. Palesi, S.: Prognosi a distanza delle convulsioni febbrili. Studio clinico ed elettroencefalografico su 100 casi, Clin Pediat (Bo- logna) 39: 685-694 (Sept.), 1957. Pardelli, L., and Ardito, R.: Attuali possibilita diagnostiche e prognostiche nelle convulsioni febbrili dell’infanzia. Contribute 200 BIBLIOGRAPHY clinico ed elettroencefalografico, Riv Clin Pediat 54: 233-244 (Oct), 1954. Parrott, R. H., and Nelson, W. E.: Acute pharyngitis, bacterial and nonbacterial (viral), in Nelson, W. E. (ed.): Textbook of Pediatrics, W. B. Saunders, Philadelphia, 1963, pp. 750-753. Pascual, R., and McGovern, J. P.: Febrile convulsions in child- hood. A statistical analysis of the clinical features in 455 cases, Clin Proc Child Hosp (Wash) 8: 92-98 (May), 1952. Patrick, H. T., and Levy, D. M.: Early convulsions in epileptics and in others, JAMA 82: 375-381 (Feb. 2), 1924. Paul, W. D., and Kemp, C. R.: Variations in concentration of certain electrolytes of blood serum during induced hyperpy- rexia, Proc Soc Exp Biol Med 45 : 427-431 (Oct.), 1940. Penfield, W., and Jasper, H.: Epilepsy and the Functional Anat- omy of the Human Brain, Little, Brown, Boston, 1954. Peterman, M. G.: Convulsions in childhood. Study of 419 cases, Trans Sect Pediat Amer Med Assn, 54-65,1932. Peterman, M. G.: Convulsions in childhood, JAMA 102: 1729- 1732 (May 26), 1934. Peterman, M. G.: Convulsions in childhood. Twenty year study of 2,500 cases, Amer J Dis Child 72; 399-410 (Oct.), 1946. Peterman, M. G.: Febrile convulsions in children, JAMA 143: 728-730 (June 24), 1950. Peterman, M. G.: Febrile convulsions, J Pediat 41; 536-540 (Nov.), 1952. Petersen, W. F.: The Patient and the Weather, Edwards Broth- ers, Inc., Ann Arbor, Mich., 1934-1938, Vols. I-IV. Phadke, M. V.: Convulsions in children, Indian J Child Health 6: 665-671 (Sept), 1957. Pine, 1., Reynolds, D. H., and O’Rear, H. B.: Convulsive dis- orders in children: a study of the Duke Hospital Pediatric Con- vulsive Clinic, North Carolina Med J 12; 129-138 (April), 1951. Posson, D. D.: Exanthem subitum (roseola infantum) compli- cated by prolonged convulsions and hemiplegia, J Pediat 35: 235-236 (Aug.), 1949. BIBLIOGRAPHY 201 Prichard, J. S.: Treatment of febrile convulsions, Appl Ther 3: 101-103 (Feb.), 1961 (review article). Prichard, J. S., and McGreal, D. A.: Febrile convulsions, Med Clin N Amer 42: 379-387 (March), 1958 (review article). Pupo, P. P.; Convulsoes na infancia; Possiveis fat ores pre- disponentes. Fatores geneticos e fatores adquiridos por ocasiao do parto, Rev Paul Med 61: 199-213 (Sept.), 1962. Radermecker, J.: Les convulsions hyperthermiques chez I’en- fant, Acta Neurol Belg 58: 50-64 (Jan.), 1958. Ribadeau-Dumas, L., and Fouet, A.: Les troubles de la regula- tion thermique par lesion du systeme nerveux central chez le nourrisson, Rev Franc Pediat 1: 3-15 (June), 1925; abstract, JAMA 85: 780 (Sept. 5), 1925. Richards, J. 8., and Egdahl, R. H.: Effect of acute hyperthermia on adrenal 17-hydroxycorticosteroid secretion in dogs, Am J Physiol 186; 435-439 (Sept.), 1956. Rodbard, S., Saiki, H., Malin, A., and Young, C.: Significance of changes in plasma and extracellular volumes in induced hyperthermia and hypothermia, Amer J Physiol 167: 485-498 (Nov.), 1951. Rosenblum, J.; Roseola infantum (exanthem subitum) compli- cated by hemiplegia, Amer J Dis Child 69; 234-236 (April), 1945. Ruch, T. C.: Evidence of non-segmental character of spinal re- flexes from analysis of the cephalad effects of spinal transection (Schiff-Sherrington phenomenon), Amer J Physiol 114: 457-467 (Jan.), 1935. Ruch, T. C., and Watts, J .W.: Reciprocal changes in reflex ac- tivity of forelimbs induced by post-brachial “cold-block” of the spinal cord, Amer J Physiol 110: 362-375 (Dec.), 1934. Saidman, L. J., Havard, E. S., and Eger, E. 1., Ill: Hyperthermia during anesthesia, JAMA 190: 1029-1032 (Dec, 21), 1964. Sampson, H. A., and Yuen, L.: The use of ACTH in heat sick- ness, New York J Med 54: 3420 (Dec, 15), 1954, Samson-Dollfus, D., Forthomme, J., and Capron, E.; EEG of the human infant during sleep and wakefulness during the first year of life. Normal patterns and their maturational changes; 202 BIBLIOGRAPHY abnormal patterns and their prognostic significance, in Kell- away, P., and Petersen, I. (eds.): Neurological and Electroen- cephalographs Correlative Studies in Infancy, Grune and Stratton, New York, 1964, pp. 208-229. Schmidt, R. P.: Sequelae of Febrile Convulsions, in Kellaway, P., and Conn, H. F. (eds.): The Convulsive Disorders, Med Clin N Amer 42: 389-397 (March), 1958. Schmidt, R. P., and Ward, A. A., Jr.: Febrile convulsions, Epi- lepsia (Amst) 4:41-47 (Nov.), 1955 (review article). Schmidt, R. P., Ward, A. A., Jr., and Wolfe, W. J.: Failure of hyperthermia to activate experimental epileptogenic foci in monkey, / Pediat 48: 180-186 (Feb.), 1956. Scholz, W.; Die Krampfschadigungen des Gehirns, Springer, Berlin, 1951. Schwartzman, J.: Transient hemiplegia associated with febrile convulsions, Arch Pediat 66: 489-491 (Nov.), 1949. Shanks, R. A.: The nature of infantile convulsions, Amer J Dis Child IS: 763-774 (Nov.), 1949. Shibolet, S., Fisher, S., Gilat, T., Bank, H., and Heller, H.; Fi- brinolysis and hemorrhages in fatal heat stroke, New Eng J Med 266: 169-173 (Jan. 25), 1962. Simon, J. F.: Effects of hyperpyrexia on the human blood count, blood chemistry and urine, J Lab Clin Med 21: 400-404 (Jan.), 1936. Soule, H. C., Buckman, T. E., and Harrow, D.C.: Blood volume in fever, J Clin Invest 5: 229-242 (Feb.), 1928. Spielmeyer, W.; Anatomic substratum of convulsive state, Arch Neurol Psychiat 23: 869-875 (May), 1930. Steinschneider, A., Ginsberg, T., George, E. D., and Lipton, E. L.: Febrile convulsions: the effects of spinal cord transection and acetazolamide, Neurology (Minneap) 14: 362-368 (April), 1964. Stevens, A.: Donnees electroencephalographiques dans les con- vulsions febriles essentielles du jeune enfant, Rev Med Liege 12: 675-681 (Dec. 1), 1957. Stevens, H.: Allergy and epilepsy, Epilepsia (Amst) 6: 205-216 (Sept.), 1966. BIBLIOGRAPHY 203 Sugita, N.: Comparative studies on the growth of the cerebral cortex, / Comp Neurol 28: 495,1917. Swinyard, E. A.; Effect of extracellular electrolyte depletion on brain electrolyte pattern and electroshock seizure threshold, AmerJ Physiol 156: 163-169 (Feb.), 1949. Swinyard, E. A., and Toman, J. E. P.; Effects of alterations in body temperature on properties of convulsive seizures in rats, AmerJ Physiol 154: 207-210 (Aug.), 1948. Teschan, P., and Gellhorn, E.: Temperature and convulsive activity, Arch Int Pharmacodyn 84: 57-67 (Nov. 1), 1950. Thom, D. A.: Convulsions of early life and their relation to the chronic convulsive disorders and mental defect, Amer J Psy- chiat 98: 574-580 (Jan.), 1942. Thompson, S. G., and Panos, T. C.: Shigellosis: pediatric as- pects, with special reference to central nervous system mani- festations, Texas Med J 53: 320-323 (May), 1957. Tibrewalla, N. S.; Convulsions in children, Indian J Child Health 2; 421,1953. Tille, D.: Initiale Fieberkrampfe und Wetter, Kinderdrztl Praxis 18:227-229 (May-June), 1950. Tin gey, A. H.: Human brain lipids at various ages in relation to myelination, J Ment Sci 102: 851-855 (Oct.), 1956. Turinese, A.: Sulle convulsioni febbrili dell’infanzia. Studio clinico ed elettroencefalografico, Riv Pat Nero Ment 80: 807- 828,1959. Turnbull, J. M.: Convulsiones febriles en el niho, Gac Med Mex 85: 541-550 (July-Aug.-Sept,), 1955. Valquist, B.: Hopade fall av “tredagarsfeber” hos barn, Svensk Lakartidn 39: 1273, 1942 (quoted by Moiler, K. L., 1956). Vyas, K. J.: Convulsions in infants and children, J Indian Med Ass 33: 272-274 (Oct. 1), 1959. Wallfield, M. J.: Exanthem subitum with encephalitic onset, J Pediat 5; 800-801 (Dec.), 1934. Wegman, M. E.: Factors influencing the relations of convulsions and hyperthermia, J Pediat 14: 190-202 (Feb.), 1939. Werboff, J., and Havlena, J.: Febrile convulsions in infant rats, and later behavior, Science 142: 684-685 (Nov.), 1963. 204 BIBLIOGRAPHY White, A. C.: The bicarbonate reserve and the dissociation curve of oxyhemoglobin in febrile conditions, / Exp Med 41: 315-326 (March), 1925. Windle, W. F.; Regeneration of the spinal cord, / Parapleg 2: 3-5,1952. Windorfer, A.: Das dreitagefieber-exanthem der kleinen Kinder Exanthema Subitum, Deutsch Med Wschr 79: 1201-1204 (Aug. 13), 1954. Wood, W. 8., Jr.: The pathogenesis of fever, Triangle 5: 101- 106 (July), 1961. Woodbury, D. M., and Karler, R.: The role of carbon dioxide in the nervous system, Anesthesiology 21: 686, 1960. Woodbury, D. M., Koch, A., and Vernadakis, A.: Relation be- tween excitability and metabolism in brain as elucidated by anticonvulsant drugs, Neurology (Minneap) (Suppl. 1) 8: 112- 116 (April), 1958. Yamakita, M.: Changes in the dissociation curve of the blood in experimental fever and feverish diseases, Tohoku J Exp Med 2: 290-323 (Sept. 10), 1921. Yannet, H., and Darrow, D. C.: Effect of hyperthermia on the distribution of water and electrolytes in brain, muscle and liver, / Clin Invest 17: 87-94 (Jan.), 1938. Ylppo, L., and Ylppo, A.: Chronic cerebral fever in cerebral palsy. Temperature athetosis, Acta Paediat 45: 483-488 (July), 1956. Zellweger, H.: Betrachtungen zum Problem der Kinderkrampfe, Deutsch Med Wschr 78: 1253-1256 (Sept. 11), 1953. Zimmerman, H. M.: The histopathology of convulsive disorders in children, J Pediat 13: 859-890 (Dec.), 1938. INDEX Abdominal pain, 32 Acetaminophen, 131 etiological significance of, 84, 85 family history of, 85 incidence of, 85 Amobarbital sodium, 112, 112 t. Anatomical maturation seizure threshold and, 74, 75 Anesthetic ether-convulsions hyperthermia and, 179, 180 Anoxia, 32, 33 t., 34 Acetazolamide experimental seizure response to, 152-58 water and electrolyte metabolism and,152—53 Acetophenetidin, 131 Acetylsalicylic acid, 20, 113, 155, 156 f. overdosage, 131 Antibiotics, in experimental animals, 155 Acid-base balance artificial hyperthermia and, 176 seizure threshold and, 66 Antibodies, maternal, 68 Anticonvulsant drugs doses of, 111-13, 112 t. in experimental animals, 152-58 Acquired brain lesions, 82, 83 Acute infection, treatment of, 113, 114 Anticonvulsant treatment, 111-13, 1121. choice of drug, 111-13, 124, 130 continuous, 115, 116-171., 118- 24 arguments against, 116-171., 118-20 duration of, 116-17 t., 133 failure of, 132 impracticality of, 134 indications for, 124, 132-34 ineffectiveness of, 123, 124 vs. intermittent, 116-17 t., 120- 23, 121-23 t. vs. none, 123, 123 t. optimum duration of, 128-30 proponents of, 115, 116-17 t. recurrence of seizures with, 116-17 t. Adenoviruses, 68, 69 Age cerebral maturation and, 74-76 at onset, 23-25 febrile vs. nonfebrile seizures, 98 /., 102 prognosis and, 104 t. seizure patterns and, in animals, 143 seizure recurrence, latest, 91-92 t. seizure threshold and, 74-76 in animals, 147 Akerren, Y., 182 Alcohol rubs, temperature control, 113 Aledort, L. M., 12, 15, 23 Allergies EEG abnormalities and, 85 205 206 INDEX universal use of, 116-17 #., 118- 20 controlled studies, 116-17 *., 120- 23, 121-23 ts. duration of trial, 116-17 *. reports of data, 116—17 *. discriminatory, 133 individual approach, 135 intermittent, 116-17 *., 120-23, 121-23 ts. vs. continuous, 116-17 #., 120- 23, 121-23 ts. indications for, 130 methods of, 130 recurrence of seizures with, 116-17 *. optimum duration of, 128—30 potential new agents, 125 practical aspects of, 134 properties of ideal compound, 124 regular; see also Anticonvulsant treatment, continuous withdrawal, criteria and meth- ods of, 128-30 EEG activation by, 65-66, 172- 74 electrolyte changes in, 178, 179 endogenous and exogenous fever, 174, 175 hemoglobin dissociation curve in, 176, 177 induction methods of, 65, 138-42, 140 174, 175, 177 pathophysiology, 174-80 phosphorus metabolism in, 176, 177 rate of temperature rise in chil- dren, 65 *. regeneration of nerve tissue in, 181 respiratory acidosis and alkalosis in, 66 respiratory volume and, 174, 175 spinal cord scar tissue, effects on, 181 treatment of neurologic disorders, 181 Antihistaminics experimental seizure response to, 155 seizure threshold and, 129 Ashby, W., 76 Aspirin; see Acetylsalicylic acid Associated episodic symptoms, 32 Atropine, experimental seizure re- sponse to, 155 Antipyretic treatment, 113, 125-27, 130-32 acetaminophen, 131 acetophenetidin, 131 acetylsalicylic acid, 131 alcohol rubs, 132 barbiturates, 130-31 phetharbital; see Phetharbital Atypical febrile seizures, 4, 5, 34, 77 Autonomic seizure incidence, 101 Bailey, A. A., 181 Bailey, O. T., 182 Baird, H. W., 64, 65*., 142, 150, 172, 173 Balter, M., 24 Bamatter, F., 70 *., 180 *. Bamberger, P., 8, 11 *., 14, 27 *., 33 *., 40, 43 *., 51, 51 *., 62 *., 70 *., 78 *., 81 *., 92 *., 94 *., 117 *. Barcroft, J., 175 Antipyretics, 113, 125-27, 130-32, 162-63 experimental seizure response to, 155, 156 f. Ardito, R., 10 *., 26 *., 33 *., 43 *., 54*., 78*., 91 *., 108*. Artificial hyperthermia acid-base changes in, 176 blood and gases in, 174-80 oxygen saturation in, 175-78 pH in, 175-78 specific gravity in, 178-79 volume in, 178, 179 convulsive response in children, 66 Barenberg, L. H., 70 *. Barometric pressure and seizures, 86 Bass, D. E., 66, 167, 178 Bassett, S. H., 178 Beck, E., 183 Behavior arousal, and brain temperature, 171-72 INDEX 207 disorders, 32, 101 learning, and experimental sei- zures, 163, 166 rate of rise in children, 64, 65, 65 t. in experimental seizures, 160- 62, 162-65 fs. recordings of, 62 t. reduction of, in treatment, 113, 125-27, 130-32 Benadryl (diphenhydramine), 155 Benign febrile convulsions; see Sim- ple febrile convulsions Bergemann, E., 9 t., 94 t. Berger, A., 142 Bigler, J. A., 179 Body weight and rate of tempera- ture rise, 165 f. Biochemical maturation and seizure threshold, 75 Boldrey, E. E., 86 Borghi, G., 10 #., 26 t., 54 t., 62 t., 911., 108 t. Birth anoxia, 32, 33 t., 34, 48, 82, 83 history, 32, 33 t., 34, 48, 82, 83 injury, 32, 33 t., 34, 48, 82, 83 Brain damage complication, 106-10, 108 t. epilepsy and, 93, 106, 183 nonfebrile seizure occurrence and, 93 significance in etiology, 82-83 Bischoff, F., 176, 177 Bjerglund, R., 3, 10 t., 91t., 116 t., 133 Blood chemistry abnormalities artificial hyperthermia and, 174-80 endogenous vs. exogenous fever effects, 174, 175 gases, 174-80 specific gravity, 178, 179 volume and electrolytes, 178, 179 immunoglobulin decrease, 57 t., 58 serum calcium decrease, 57t., 58 serum sodium decrease, 57t., 58 significance in etiology, 83-84 nonprotein nitrogen, 57 t., 58 sugar, 57 t., 58 volume and hyperthermia, 66, 178, 179 Brain lesions convulsions and, 183 hyperpyrexia and, 183 Brain temperature behavior, arousal and, 171-72 measurement in animals, 169 Brandt, S., 3, 10 t., 911., 116 t., 133 Breg, W. R., 3 Bridge, E. M., 4, 9 t., 14, 123 Broberger, 0., 11t., 27 t., 30 t., 31t., 35 t., 551., 62 t, 69, 70 t, 71 Bronchitis, 28, 68 Brown, H. R., Jr., 178 Brown, J. E., 8, 91., 26 t., 70 t. Brown, W. C., 170 Brusa, P., 811., 94 t. Buckman, T. E., 178 Calzetti, G., 10 t., 26 t., 54 t., 62 t., 911., 108 t. Carbon dioxide and brain excitabil- ity, 180 Carbonic anhydrase in brain age and, 76 seizure threshold and, 76 Body fluids and hyperthermia, 178, 179 Body temperature convulsions and effects on, 170, 171 convulsive threshold and, 61, 63, 63 t., 64 t. height of, 60, 61, 63, 62-64 ts. etiological significance in, 60, 60 t., 61, 62-64 ts., 64-65 maximum, 65 t. range of, 62-63 ts. Carter, C. H., 128 Carter, S., 117 t., 133 Cary, W., 4, 10 t., 26 t., 811., 94 t. Cassels, W. H., 179 Casual convulsions, 2, 3 Causes; see Etiological factors Cavazzuti, G. 8., 4, 111., 33 t., 351., 208 INDEX 43 t, 51, 78 t., 811., 92 t, 94 t. Chandy, J., 10 t, 54 t., 56 Chao, D. H.-C., 117 t., 133 Chaptal, J., 9 t. Chicken pox, 53 Cerebellar ataxia and hyperthermia, 182 Cerebral anatomical maturation and seizure threshold, 74, 75 Chlorpromazine, 113 experimental seizure response to, 162, 163/. Clark, W. F., 178 Cerebral anoxia, incidence of, 48 Cerebral biochemical maturation and seizure threshold, 75 Clemens, H. H., 70 *., 71 Clinical evaluation, 17-22 Cerebral birth injury, incidence of, 48 Coelho, G., 91., 11 t., 78 t., 911. Cold water enema, hazards of, 113 Cerebral blood flow age and, 76 hyperthermia and, 66 Cerebral dysgenesis, incidence of, 48 Complicated febrile convulsions vs. benign, 1041. clinical findings with, 104 t. EEC abnormalities and, 49, 50 t. epilepsy and, 50 t. neurologic abnormalities and, 48, 49 nonfebrile seizures and, 49 Cerebral dysrhythmia and hyper- thermia, 66 Cerebral hemorrhage, incidence of, 48 Cerebral oxygen consumption age and, 76 hyperthermia and, 66 Conel, J. L., 75 Continuous anticonvulsant treat- ment; see also Anticonvulsant treatment, continuous inadvisability of, 132 universal use of, arguments against, 118-20 Cerebral trauma, incidence of, 82, 83 Cerebral venous thrombosis and hy- perthermia, 182 Cerebrospinal fluid abnormalities, 53, 54-55 t., 56 amino acid increase, 54 t. cell increase, 54-55 t. pleocytosis, 54-55 t. EEC showing and, 56 protein decrease, 54-55 t. malaria and, 56 U.R.I. and, 56 protein increase, 54-55 t. sugar increase, 54-55 t. Convulsions; see also Seizure duration of, 28-32, 311., 89, 101, 103 t. histopathology and, 183 Convulsive threshold, 60-64, 66, 85; see also Seizure threshold height of body temperature and, 60-64 temperature, 60-64; see also Ex- perimental febrile convulsions Cerebrospinal fluid examination, in- dications for, 19 Cooke, R. E., 152 Cooke, R. T., 72 t. Cooling techniques, 113, 130-32 Copman, L., 113 Cerebrospinal fluid findings, 53-57 exanthem subitum, 53, 54-551., 56 exanthemata, var. and, 54 t. gastroenteritides, 54-55 t. malaria, 54-551. pertussis, 54 t., 73 respiratory infections, 54 t., 56 roseola infantum, 69 shigella dysentery, 73 shigellosis, 54-55 t., 56 Cortical epileptogenic lesions, exper- imental activation by fever, 167-70 Cortisone, experimental seizure re- sponse to, 155 Costeff, H., 8, 111., 117 t., 133, 134, 135 Countries, incidence; see Geographic location Coxsackie viruses in etiology, 68, 69 INDEX 209 Cramblett, H. G., 69 Crawford, J. D., 152 Cruickshank, E. K., 182 Cumings, J. N., 76 Custer, R. P., 181 Cyclical vomiting, 98, 101 Diphtheria, 69 Dobin, N. 8., 182 Dobrzynska, L., 27*., 30 *., 92 *., 94*., 117 *., 133 Dodge, P. R., 18, 152 Donald, W. D., 14, 27*., 28, 54*., 57 *., 72 *., 73 Doose, H., 117 *. Darling, R. C., 177 Darrow, D. C., 149, 167, 178 Davis, E., 108 *. Dees, S., 85 Definition, 1-5 activated seizure state, 3 atypical febrile seizures, 4, 5, 34 hyperthermic-precipitation epi- lepsy, 3 lay terms, 1 mild or pure epilepsy, 3 relation to epilepsies, 1, 2, 3 simple febrile seizures, 4, 5, 34 deHoll, G., 72 *. Dejong, R. N., 18 Dekaban, A., 75 Dennison, C. A., 72 *. Desoxycorticosterone, experimental seizure response to, 155 Diagnosis criteria for, 2-3 differential bacterial meningitis, 18, 114 cerebral abscess, 18 C.S.F. examination in, 18 cerebrovascular accident, 19 encephalitis, 18 focal encephalitis, 19 neurosurgical tests in, 18 postical paresis in, 19 subdural effusion, 18 subdural hematoma, 18 toxic encephalopathy, 18 Diphenylhydantoin sodium (Dilan- tin) electrolyte metabolism and, 152, 153 exacerbating effect of, 112, 20, 21 experimental seizure response to, 152-58, 156/., 157*., 158/. regular daily administration, 120— 23, 123 #., 133 Diphenhydramine, experimental sei- zure response to, 155 Economic status in etiology, 83-85 Education, parental, 130 Egdahl, R. H., 141 Ehrengut, J. and W., 27 *., 57 *., 84 Eichenwald, H. F., 68 Ekholm, E., 10 *., 26 *., 91 *., 94 *. Electric heating pad, 140; see also Hyperthermia Electrocorticogram and hyperther- mia, 167-70 Electroencephalogram, 37-53, 65- 66 age and, 42 /., 44 *., 75 artificial hyperthermia and, 65, 66 clinical correlations, 42, 43, 44- 45 *., 49 diagnostic aid, 47 follow-up examination, 52 indications for, 52, 53 maturation of, 75 seizure threshold and, 75 normal records, 44^45*., 46 *., 48, 52 incidence of epilepsy and, 42 incidence of febrile seizure re- currence, 42 postical abnormalities, 19 prognostic value, 53, 103 spontaneous seizure occurrence and, 49, 103 *. Electroencephalographic abnormali- ties age and, 38, 44*., 49, 53 allergies and, 85 amplitude asymmetries, occipital, 21 artificial hyperthermia and, 66, 172 birth abnormalities and, 45 *. blood pH, CO2, and, 66 centrencephalic spike-and-wave, 40 /. 210 INDEX cerebral organic lesions and, 47 clinical correlations, 42, 43, 44- 45 t., 49 convulsions and, duration of, 101, 103 t. frequency of, 44 t. severity of, 44 t. diffuse dysrhythmia, 49 “epileptic” discharges, 47, 52 family history of epilepsy and, 45 t. fast activity, 46 t. fever and duration of, 45 t. height of, 45 t. infection with, 66 focal, 38 t., 46 t., 48, 49 grand mal and, 46 t., 48 vs. febrile convulsions, 46 t. height of body temperature and, 66 incidence, 40, 46 t. febrile vs. afebrile grand mal, 53 simple vs. complicated febrile seizures, 38 t. indications for treatment, 52 late emergence of, 39 f. mental retardation and, 45 t. parents with, 47 paroxysmal discharges focal seizures and, 53, 105 incidence, 44-451., 46 t.} 51-53, 94-95 t., 102, 103 t. seizure duration and, 53, 103 spontaneous seizure occurrence and, 53, 102, 103 103 persistent, 41 petit mal and, 48 phase reversal, 41 /. phetharbital effect on, 128 plasma CO2 content and, 66 postical slow wave seizure recurrence and, 47, 53 subsequent paroxysmal records, 47 time to disappearance of, 47, 52-53 prognosis and, 89, 94-951., 101- 103 febrile seizure recurrence, 41 nonfebrile seizure occurrence, 41 respiratory alkalosis, fever, and, 66 seizure discharges brain injury at birth and, 94- 95 t. centrencephalic, 40 f., 101 family history of epilepsy and, 94-95t. of seizures, all types, 94-95 t. focal, 38 t., 46 t., 48, 49 follow-up examination and, 49, 52, 94-95 t. incidence of, 381., 43 f., 44- 45 t., 46 t., 51-53, 94-95 t., 102, 103 t. nonfebrile seizure occurrence and, 94-951., 102, 1031., 103 siblings with, 47 slow wave, 46 t., 47, 48, 52, 53, 66 encephalitis and, 48 slowing, extreme, 42, 46 t. brain damage and, 42 frequent convulsions and, 42 occipital, 42, 46 t. prolonged fever and, 42 severe convulsions and, 42 spike discharges, 48 brain injury and, 48 spikes and hypersynchronism, 52 spike-and-wave, 48 synchronous, symmetrical, 38 t. temperature elevation and, 66 typhoid vaccine and, 66 time interval after convulsion, 22 f., 41, 47, 511. upper respiratory infection and, 66 Electroencephalogram, animals, brain temperature and, 171-72 convulsions, fever, and, 163, 166- 69 hyperthermia-induced abnormali- ties, 163, 166-69 blood chemistry and, 166, 167 etiology and mechanism, 166-69 Electrolyte disturbances, manage- ment of, 114 Electrolyte imbalance, shigella dys- entery and, 73 INDEX 211 Electrolyte metabolism artificial hyperthermia and, 178, 179 threshold to febrile seizures in ani- mals and,l4B-54 Electroshock seizures, hyperthermia, in animals, 170 Emotional disturbance and seizure threshold, 129 Encephalitis, incidence of, 48 Enteroviruses, 68 Enuresis, 32 Epidemic fevers, 26-271., 68, 69; see also Infections Epidemiological studies, 15, 68, 69 Epilepsies, see also “Epilepsy” febrile convulsions, relation to, 1-5, 71, 90 cerebral anoxia, 48 cerebral dysgenesis, 48 cerebral hemorrhage, 48 drug reactions, amidopyrin, 84 economic status, 83-85 electrolyte imbalance, 73, 83, 84 epidemic infectious fevers, 69-73 fever, 59-67, 86 genetic, 77-82 height of body temperature, 60- 65, 60 62-64 ts., 71 hereditary, 77-82 histamine, experimental, 85 hypersensitivity reactions, 83-85 immune reactions, 84-85 immunizations, pertussis, small- pox, 73 immunoglobulin lack, 57 t., 58, 84 infections, 68-73 latent lead intoxication, 84 meteorobiologic, 83-85 organic, 48 pertussis, 73 race, 64 t., 83-85 rapidity of temperature rise, 64- 65, 65 t. roseola infantum, 69-71, 70 t. encephalitic, 71 toxic allergic, 71 shigella dysentery, 71-73, 72 t. spasmophilia, 84 syncope, 86 temperature elevation height of, 60-65, 71 rate of, 64—65, 65 t. tonsillitis, pharyngitis, 68-69 water, electrolyte, imbalance, 73, 83, 84 “Epilepsy”; see also Nonfebrile and Spontaneous seizures cryptogenic, 48 definition, I—s, 96 genuine or true, 2-5 relation to febrile convulsions, 1-5, 71, 90 diagnostic criteria and incidence, 93, 96 familial incidence, 80-81 t. fever and, 74, 77 idiopathic, 48 “Epileptic” febrile seizures, 48; see also Atypical febrile seizures EEG findings in, 48 Epileptogenic lesions, hyperthermia, in animals, 167-70 Episodic symptoms, 32 Equanil; see meprobamate Erickson, T. C., 67, 182 Erythrocyte sedimentation rate in neurogenic hyperthermia, 67 Esplin, D. W., 145 Ether convulsions, hyperthermia and, 179, 180 treatment hazards of, 180 Etiological mechanisms allergic response to infection, 74 bacterial toxins, 74 encephalitis, unrecognized, 74 encephalopathy, toxic, 74 fever per se, 74 immune state abnormality, 74 Etiological factors, 59-87, 137 acquired brain lesions, 48, 82-83 age, 71 allergies, 84-85 barometric pressure changes, 86 birth injury, 48, 82-83 Exanthem subitum, 26-27 t., 28, 53, 54-551., 56, 69; see also Roseola infantum hemiparesis and, 107, 108 t. Exanthemata, 54 t., 57 t. 212 INDEX Experimental animals, 64-65, 137- 80 Faxen, N., 14, 26 t., 94 t. Experimental febrile convulsions, 137-80 behavior and, 163, 166 blood chemistry abnormalities and, 166, 167 brain histology and, 164-65 cortical epileptogenic lesions and, 167-70 electroencephalogram and, 163, 166-69, 172-73 neuropathology of, 164-65 seizure patterns, 143-45 age and, 143 cord transection and, 144, 145 species differences, 143, 144 seizure threshold, 145-60 age and, 147 body weight and, 165 f. brain histology and, 154, 155 drugs and, 152-57, 162, 163 electrolyte metabolism and, 148-54 myelination and, 155 previous convulsions and, 157, 159, 159-60 fs. temperature elevation range of, 147, 148 rate of, 160-62, 162-65, 162- 65 fs. water and electrolytes, 148-54 Febrile episodes, 25, 26-271., 54- 551., 571.; see also Fever; Infections Febrile illness, epileptic seizures and, 74, 170 Felsen, J., 72 t. Ferris, E. 8., Jr., 113 Fever; see also Artificial hyperther- mia; Body temperature artificial, 64-66, 138-42, 172-81 causes, 25, 26-27 *., 28, 54-55 *., 56, 57 *. bronchitis, pneumonia, 26-27 *. epidemic fever, var., 26-27 *., 54-55*., 56, 57*., 69-73 exanthem subitum, 26-27 *., 28, 53, 54-55 *., 56, 69-71 exanthemata; see also Epidemic fevers immunizations, 26-27 *., 73 malaria, 26-27 *., 54-55 *., 56 measles, 26-27 *., 28, 53 otitis media, 25, 26—27 *., 55 *. pertussis, 26-27 *., 28, 54 *., 56, 57 *., 73 poliomyelitis vaccination, 58 pyelitis, 26-27 *. roseola infantum, 26-27 *., 28, 53, 54-55 *., 56, 69-71 salmonellosis, 26-27 *., 28 septicemia or sepsis, 26-27 *., 28 shigellosis, 26-27 *., 28, 54- 55 *., 57 *.,71, 72*., 73 smallpox vaccination, 26-27 *., 28, 58, 73 tonsillitis, pharyngitis, 20, 25, 26-27*., 54-55*., 56, 68-69 unknown, 26-27 f. etiological significance, 59-67, 86 neurogenic, 67 pre-convulsive, duration of, 61 Experimental fever, 64-66, 138-42, 172-81; see also Artificial hyperthermia Experimental hyperthermia, see Arti- ficial hyperthermia Extracranial infections, 25-28; see also Infections Faerber, E., 2, 3, 26 *., 80 *., 94 *. Familial incidence, 47, 78-82 epilepsies, 47, 80-81 *., 94-95 *. prognosis and, 94-95 *. recurrent nonfebrile seizures, 47 seizures, all types, 47, 80-81 #., 94-95 *. Fischler, E., 14, 27*., 55*., 57*., 72*., 73, 83 Focal EEG abnormalities, 50 *. prognosis and, 100 102, 105 Focal febrile convulsions, 49, 50 *. cerebral immaturity and, 99 EEG abnormalities and, 50 t. epilepsy and, 50 *. Family history, 18, 45 t., 94-95 t. Farmer, T. W., 18 Fatigue and seizure threshold, 129 INDEX 213 hemiparesis and, 49, 50 f. incidence, 102 prognosis and, 105 significance in infants, 99 spontaneous seizures and, 49, 105 England, 15, 16 Newcastle, 134 Oxford, 15 Finland, 14, 28 Helsinki, 10 #., 26 #., 94 f. France, 10 #., 14, 28, 40 Montpellier, 9 f. Paris, 9f., Ilf., 14, 26-27#., 40, 48, 94 f. Germany, 14, 28, 94 f. Berlin, 26 f., 85, 94 f. Freiburg, 11 f., 52, 94 f. Hamburg, 27 f. Leipzig, 9 f., 94 f. Munich, 94 f. Hungary, 14 Budapest, 9 f. India, 14 Bhavnagar, 11 f., 27 f. Bombay, 9 f., 10 #., Ilf. Jaipur, 27 f. Jamnazar, 11 f., 94 f. Poona, Ilf. Vellore, 10 f. Israel, 14, 28 Beer-Sheva, 11 f., 133 Zrifin, 27 f. Italy, 14 Bologna, 10 f., 26 f. Milan, 94 f. Modena, 11 f., 51, 94 f. Padua, Ilf. Pisa, 10 f., 26 f. Siena, 48 Japan, 15, 16 Kenya, 14 Nairobi, 11 f., 27 f. Mexico, 14, 26 f. Poland, 14 Gdansk, 27 f., 94 f. Scotland, 14, 28 Glasgow. 10 f., 26 f., 28 South America, 15, 16 Soviet Russia, 15, 16 Sweden, 14 Gothenburg, 10 f., 26—27 f. Stockholm, 9 f., Ilf., 14, 26- 27 f., 69, 94 f. Uppsala, 26 f. Switzerland, 14 Basel, 11 f., 14, 27#., 51, 94 f. Fois, A., 29, 30 f., 43#., 48, 116 f., 133 Follow-up, duration, 93, 94-95 #., 96 Forbes, G. 8., 55 #., 57 #., 61, 63 #., 71,72 f., 77, 114 Fouet, A., 67 Fowler, M., 106, 108 #., 182 Fox, B. J., 75 Frantzen, E., 11#., 27 f., 47, 70#., 92 f., 117 f., 120, 123, 123 #., 132, 153 Freedman, D. A., 182 Freund, W., 33 f. Friderichsen, C., 4, 8, 10 f., 14, 26 #., 32, 34 #., 35 #., 61, 63 f., 64, 70 #., 78 #., 80 #., 83, 94 f., 96 Friedman, E. D., 67, 182 Gallant, L. J., 9 #., 116 f., 133 Gandhi, V. K., 11 f., 30 f., 31 #., 81 f., 94 f. Garfunkel, J. M., 64, 65 f., 65, 142, 150, 172, 173 Gastaut, H. and Y., 86 Gastroenteritis, 14, 28, 54 f., 68 Gelegenheitskrampfe, 2, 3 Gellhom, E., 169 Gemonil (metharbital), 155, 157 f. General management, 130 Genetic factor, 77-82 Geographic location, 9-11 f., 12-16, 13 f. Australia, 14 Sydney, 10 f., 26 #., 94 #. Belgium, 14 Antwerp, 11 f., 27 f., 49, 94 f. Canada, 15, 16 China, 14, 28 Nanking, 10 f., 26 f., 28 Czechoslovakia, 14 Prague, 11 #., 94 #. Denmark, 14, 28, 47 Copenhagen, 10-11 f., 12, 14, 26 f., 94 f. Gentofte, 11 #., 27 f. 214 INDEX Geneva, 9 #., 26 #., 47, 94 #. Zurich, 94 #. United States, 9-11#., 13/., 14, 28, 47, 134 Baltimore, 9 #., 26 #., 94 #. Birmingham, Ala., 14, 27 #., 28 Boston, 8, 9-10 #., 12, 14, 26 #., 94 #., 134 Chicago, 8, 9 #., 19, 179 Durham, N.C., 10 #. Milwaukee, 9 #., 10 #., 14, 47, 94 f. New Haven, 9 #., 12, 14, 94 #. New York, 11 #., 15, 27 #., 94 #. Rochester, Minn., 94 f. Washington, D.C., 10 #., 26 #. Hemiparesis acute infantile type, 109 DPT immunizations and, 107, 108 t. exanthem subitum and, 107, 108 #. febrile convulsions and, 106-10 incidence and significance, 106— 10, 108 #. Marie-Striimpell polioencephalitis and, 109 Hemoglobin dissociation and hyper- thermia, 176, 177 Henchel, A., 66, 167, 178 Hereditary predisposition, 77-82 febrile vs. nonfebrile seizures, 79- 82, 80-81#. Gibbs, E. L., 35, 75 Herlitz, G., 9 #., 14, 26 #., 31#., 34 #., 35 #., 621., 70 t., 73, 80 t., 83, 84, 94 #., 108 t. Gibbs, F. A., 75 Gilman, A., 130 Ginger, L. G., 142 Giraud, P., 10 #., 54 t. Goodman, L. S., 130 Gordon, E. E., 177 Gottschalk, B. G., 182 Grace, H. 8., 108 t. Greenspan, L., 70 t. Greenthal, R. M., 70 t., 71 Guthrie, R. H, 74 Hernandez, P., 24 Heymans, C., 169 Hill, E., 176, 177 Histology, developing brain, and sei- zure threshold, 154, 155 Hoagland, H., 169, 173 Hochsinger, 2 Holliday, P. 8., Jr., 54 #., 57 #., 108 #. Horstmann, W., 11 #., 33 #., 35 #., 43#., 52, 78#., 81#., 92#., 94 #., 117 #., 133 Habel, K., 54 t., 56, 57 t., 73 Haldane, J. B. S., 176 Hot water baths, 141; see also Hy- perthermia Hammill, J. F., 117 #., 133 Handa, J., 66 Hoyt, R., 170 Hrbek, A., Ilf., 33 f., 35 f., 62 f., 78#., 81#., 84, 91#., 94#. Hardy, A. V., 72 #. Hartman, F. W., 181 Havlena, J., 166 Hull, C. D., 171 Husler, J., 2 Haymaker, W., 181 Head injury, 32, 33 #., 34 Heat cramps, 180 Huttenlocher, P. R., 134 Hyperactive behavior, 32 Hyperpyrexia, 180-83 cerebellar syndrome in, 182 Heat exhaustion, 180 Heat puncture, 175 Heat stroke, 180-81 cerebellar syndrome in, 182 fibrinolysis, afribrinogenemia, in, 181 Hyperpyrexia and brain lesions, ISO- -83 Hypersensitivity reactions, etiologi- cal significance of, 83-85 Heating cabinet, 141; see also Hy- perthermia Hyperthermia anesthesia and, 179 artificial, 64-66, 138-42, 172-81 blood volume, 66 brain lesions and, 182 cerebellar syndrome, 182 Height of body temperature, 60-65, 60 #., 62-64 ts.; see also Body temperature INDEX 215 cerebral blood flow and, 66 cerebral oxygen consumption and, 66 cerebral venous thromboses and, 182 convulsive response to, 65 #. EEC abnormalities, cerebral me- tabolism, and, 66 experimental, 64-66, 138-42, 172-81 adrenal cortical function, 141 behavior and, 163, 166 blood chemistry and, 166-67 clinical significance, 106 cortical epileptogenic lesions, 167-70 EEG and, 163, 166-69 electroshock seizures and, 170 head injury and, 182 induction methods, 138-42 electric heating pad, 140 heat puncture, 175 hot water baths, 141 injections of pyrogens, 141, 142 insulated heating cabinet, 141 microwave diathermy, 139, 140, 140 /. radiant heat, 138, 139 radio short waves, 177 steam heat, 141 typhoid vaccine, 65, 142 plasma proteins and, 67 red blood cells and, 66 abnormal neurologic signs, 34, 34 #. age and, 23-25 all childhood illnesses, 9-11#., 12 all seizures, 9-11#. birth complications, 32, 33, 33 #., 34 children in hospital or clinic, 9- 11 #. clinic vs. office patients, 85 convulsive disorders, young chil- dren, 133-34 economic factors and, 85 geographic, 9-11#., 12, 13/., 14- 16 maximum, 62 #. nonfebrile seizures adults, 134 young children, 133-34 racial factor, 15—16, 85 recurrence of febrile seizures, 90- 93, 91-92 #. sex factor, 9—ll #., 12, 77 in young children, 133-34 Infections adenoviruses, 68, 69 bronchitis, 28 chicken pox, 53 Coxsackie viruses, 68, 69 diphtheria, 69 enteroviruses, 68 epidemic fever, van, 26-27 #., 57#., 69 exanthemata, 54 #., 57 f. extracranial, 25-28 gastroenteritis, 14, 28, 54 #., 57 #., 68 herpangina, 68 malaria, 26-27 #., 28, 54-55 #., 56 measles, 26-27 #., 28, 53, 69 miscellaneous, 26-27 #., 28, 68 mumps, 53 otitis media, 25, 55 #., 68 pertussis, 26-27 #., 28, 54 #., 56, 57 #., 73 pharyngitis, 25, 54-55 #., 56, 68 pneumonia, 28, 68 pyelitis, 26-27 #. roseola infantum, 28, 53, 54-55 #., 56 salmonellosis, 26-27 #., 28 Hypocalcemia, 57 #., 58 Hypogammaglobulinemia, 57 #., 58 Hyponatremia, 57 #., 58, 67, 114 etiology and, 83 increased intracranial pressure and, 114 treatment of, 114 Ikeda, T., 15, 167 Immune reactions, etiology and, 84- 85 Immunizations convulsions and, 73 seizure threshold and, 129 Immunoglobulins, 57 #., 58 etiology and, 84 Incidence, 5, 8, 9-11 #., 12-16, 72 #.; see also Statistics 216 INDEX scarlet fever, 69 sepsis, 26-27 t., 28 septicemia, 26-27 t., 28 shigellosis, 28, 54-55 £., 57 t., 61, 63 t. statistics, geographic location and, 26-27 t., 28 streptococcal, 68 tonsillitis, 20, 25, 26-271., 54- 55 £., 56, 68 U.R.1., 20, 25, 26-27 t., 54-551, 56, 68 viral, 68, 69 Laboratory investigations animals in, 137-80 blood sugar, 20 EEG, 20, 37 serum calcium, 20 urinalysis, 20 x-rays of skull, 20 Laffan, R. J., 145 Laplane, R., Ilf., 27 t., 30 t., 33 t., 40, 41, 43 t., 62 t, 78 t., 811., 92 t., 94 t. Lavagna, E., 4, 43 t., 51, 94 t. Lead intoxication, etiology and, 84 Lederer, E., 3, 9 t. Inheritance, 77-82 epilepsies and, 82 Lenggenhager, K., 174 Intermittent treatment, 116-171., 120-23, 121-23 ts.; see also Anticonvulsant treatment indications, 130 methods, 130 Lennox, M. A., 3, 9 t., 10 L, 111., 12, 14, 29, 30 t, 311., 33 t., 33, 34 £., 35 t., 42, 43 t., 44- 45 t., 46, 46 t., 47, 551., 56, 62 t., 64, 67, 80 t., 85, 911., 93, 94 t., 102, 104, 104 t., 105, 106, 1081., 1161., 118, 133, 150, 163, 166, 182 Isch-Treussard, C., 43 t. Janeway, C. A., 84 Jasper, H., 98 Lennox-Buchthal, M. A.; see Len- nox, M. A. Jennings, C. G., 116 t. Johnson, I. E., 142 Lennox, W. G., 3, 10 t., 12, 14, 67, 77, 80 t., 85, 911., 93, 94 L, 96, 97, 97 t, 99 t, 106 Jones, J. D., 148 Lerique-Koechlin, A., 111., 14, 27 t., 29, 33 t., 351., 43 n., 48, 50 t., 94 t. Kajdi, L., 9 t., 123 Kanof, A., 67 Kao, Y. E., 10 14, 26 28 Levy, D. M., 8, 91., 311. Livingston, S., 4, 9 t., 14, 19, 26 t., 66, 70 t., 77, 78, 78 t., 80 L, 83, 93, 94 L, 105, 116 t., 123, 124, 132, 133 Kaplan, M., 9 t., 26 t., 78 t., 80 t., 94 t. Karler, R., 180 Kashiwase, Y., 15, 141, 168, 169 Keith, H. M., 35#., 61, 62#., 94#. 117#. Long, M. L., 176 Long-term treatment, 115-32; see also Anticonvulsant treatment Kellaway, P., 75 Kemp, C. R., 179 Kennedy, C., 76 Kety, S. S., 76 Lucchesi, P. F., 541., 56, 57 t., 73 Lyon, G., 107 MacDougall, L. G., 111., 27 f., 55 t., 56 Keutmann, E. H., 178 Knudson, A., 176 Kohn, R., 85 Kotsevalov, 0., 68 Kowlessar, M., 551., 57 t., 61, 63 71, 72 t., 77 Krakau, C. E. T., 141, 173 McGovern, J. P., 8, 101., 26 t., 311., 61, 62 t., 64 t, 77, 85, 911. McGreal, D. A., 4, 15, 77, 1161., 133 McQuarrie, L., 148 McQuiston, W. 0., 179 INDEX 217 Madsen, J. A., 12, 15, 23 Malamud, N., 181, 182 42 /., 43 #., 46, 49, 53, 55 #., 57#., 58, 59, 60, 601., 61, 62 #., 64, 65, 66, 67, 70 #., 74, 76, 78 #., 79, 81#., 82, 83, 84, 85, 86, 90, 92 #., 93, 94 #., 96, 98, 98/., 101, 103#., 103, 105, 108t., 112, 113, 114, 117#., 119, 120, 121#., 122, 122#., 126, 126/., 127, 131, 132, 138, 139, 140/., 143, 144, 145, 146/., 147/., 148, 148 /., 150, 151 /., 151, 152 /., 152, 153, 153/., 154/., 154, 155, 157#., 157, 161, 162, 162 /., 163 /., 164 /., 166, 168, 180 Malandrini, F., 29, 301., 431., 48, 116 t., 133 Malaria, 26—27 #., 28, 54-55 #., 56 Matthes, A., 8, 11 #., 14, 27 #., 33#., 40, 43 #., 51, 51 #., 62 #., 78 #., 81#., 92#., 94#., 117#. Maxwell, L. C., 177 Measles, 28, 53, 69, 107 Mebaral (mephobarbital), 131 Mechanism, 59-87 body temperature and, height of, 60-65, 60#., 62- 64 ts., 160-162 rate of rise of, 64-65, 65 #., 160-62, 162-65 fs. electrolyte imbalance, 73, 83, 84 epilepsies, relation to, 82 fever and, 59-67, 86 hypersensitivity reactions, 83-85 immunoglobulin lack, 58, 84 syncope, 86 Miltown (meprobamate), 125, 126, 126 /., 155, 157 t. Minard, D., 113 Moffet, H. L., 69 Moller, K. L., 10 t., 27 t., 55 t., 69, 70 t. Moscovici, C., 68 Mumps, 53 Mehta, J. 8., 27 t. Melchior, J., 4, 8, 10 #., 14, 26 #., 32, 34#., 35#., 61, 63#., 64, 70 #., 78 #., 80 #., 83, 94 #., 96 Melin, K.-A., 43 #., 94 f. Mendez, M., 33 #., 35 #., 67 Myelination, and threshold, 155 Mysoline (primidone), 112 t., 113 Natural history, 133 Nelson, W. E., 68 Meningitis, 114 Meningoencephalitis, 53 Mental development, 35, 35 t., 36 Mental sequelae, 51 Nembutal (pentobarbital), 112#., 131 Neurogenic hyperthermia, 67, 182; see also Hyperthermia erythrocyte sedimentation, 67 plasma fibrinogen, 67 Neurologic abnormalities, 32, 34 t, 35, 83 Mephobarbital, 131 Meprobamate, 125, 126, 126 /., 155, 157 #. Merritt, H. H., 145 Meteorobiologic factor in etiology, 83-85 Neurologic examination, 17-22, 34, 34 t. Metharbital, 155, 157 #. Meyer, J. S., 66, 183 Microwave diathermy, 139, 140, 140 /., 161, 165; see also Hy- perthermia Neurologic sequelae, 51 Neuropathologic findings, 107 experimental seizures, 164, 165 secondary to convulsions, 183 Neyroud, M., 9 #., 26#., 32, 43#., 47, 62#., 80 #., 94 #. Niemineva, K., 10#., 26#., 91#., 94 #. Nonfebrile seizures, 47, 48, 51-52, 89, 90 /., 93, 94-95 #., 96, 97, 97 #., 99, 99 #., 101-3, 102 /., Milk fever, 142; see also Pyrogens Miller, F. J. W., 134 Millichap, J. G., 7 n., 111., 12, 15, 21, 23, 24, 27 #., 28, 29, 30#., 31#., 32, 33 #., 33, 34, 34 #., 351., 35, 37, 38 t., 40 /., 41 /., 218 INDEX 103 #., 133, 134; see also Spontaneous seizures Nursing care during seizure, 114-15 Nyhan, W. L., 152 Nyman, G. E„ 141, 173 Pharyngoconjunctival fever, 68 Phenacetin (acetophenetidin), 131 Phenobarbital sodium, 19, 21, 112, 112#., 113, 125, 126, 129; see also Anticonvulsant treat- ment antipyretic properties, 112, 113 continuous treatment, 120-22, 124 dosage, 112, 112 #., 113 intermittent treatment, 120-22, 129 regular daily treatment, 120-22, 124 Oppe, T. E., 18 Osborne, S. L., 178 Otitis media, 25, 26-27 #., 55 #., 68 Ounsted, C., 15, 82 Owens, G., 180 Oxygen saturation and pH of blood, 175-78; see also Artificial hyperthermia Phenothiazines and seizure thresh- old, 129 Phenylbarbital, 125, 126 f.; see also Phetharbital Phenytoin, 123; see also Diphenyl- hydantoin sodium Phetharbital anticonvulsant properties clinical evaluations, 127-28 laboratory evaluations, 125 /., 126 /., 155, 156, 157 #., 158 f. antipyretic properties clinical evaluation, 127 laboratory evaluation, 126 f., 127 EEG spike-and-wave, response to, 128 Oxytetracycline, experimental sei- zure response to, 155 Pache, H. D., 81#., 94 t. Paine, R. S., 18 Palesi, S., 33 t., 35 t., 43 t., 911. Pardelli, L., 10 t., 26 t., 33 t., 43 #., 54 #., 78 #., 91 #., 108 #. Parrott, R. H., 68 Pascual, R., 8, 10 #., 26 #., 31 #., 61, 62 #., 64 #., 77, 85, 91 #. Patients history of, 18 neurologically impaired, 34 #. retarded, 35, 35 t. Patrick, H. T., 8, 9 #., 31 #. Paul, W. D., 179 Phosphorus metabolism, and artifi- cial hyperthermia, 176, 177 Penfield, W., 98 Penicillin convulsant effects, 114, 129, 155 Pentobarbital sodium, 112 #., 131 Periodic abdominal pain, 101 Periodic vomiting, 101 Perlstein, M. A., 18 Pertussis encephalopathy, 73 incidence and, 26-27 #., 28, 54 #., 56, 57 #., 73 Peterman, M. G., 3, 8, 9 #., 10 #., 14, 33 #., 40, 43 #., 47, 48, 61, 63, 78 #., 801., 82, 91#., 94#., 116 #., 132 Petersen, W. F., 85 Physical activities, restrictions, 130 Picrotoxin convulsions and hyper- thermia, 169 Pine, L, 10 #. Piromen, 142; see also Pyrogens Plasma fibrinogen, 67 Plasma proteins, 67 Pneumonia, 28, 68 Poliomyelitis vaccination, 58 Population statistics, U.S., 134 Posson, D. D., 70 #., 108 t. Potassium metabolism, seizure threshold and, 148-54 Prenatal injury, 32 Prichard, J. S., 4, 15, 77, 116-17 #., 133 Phadke, M. V., 11 #., 55 t. Primidone, 112 #., 113 Prognosis, 89-110, 135 Pharyngitis, 25, 54-55 #., 56, 68 INDEX 219 acquired brain pathology and, 83, 93 brain injury complication, 106, 107, 1081., 109, 110 brain pathology and, 83 epilepsy in adulthood, 133-34 febrile seizure recurrence, 90, 91- 92 t., 93 follow-up period and, 94-951., 97 t. hemiparesis complication, 106-10, 108 t. simple cases, 83 vs. complicated cases, 89, 93 spontaneous seizure occurrence, 93, 94-95 t., 96-98 Todd’s paresis, 108 Negro and, 15, 61, 64 t. Spanish-American and, 15 white, 64 t. Radermecker, J., 111., 271., 43 t., 49, 91 f., 94 t. Radiant heat, 138, 139; see also Hy- perthermia Recurrence, febrile seizures, 103 age at latest, 91-92 t. anticonvulsant treatment and, continuous, 116-171., 118-24, 121-23 ts. intermittent, 116-17 t., 118-28, 121-23 ts. none, 123, 123 t. duration of febrile seizure and, 101, 103 t. EEG abnormalities and, 122 t. focal seizure pattern and, 122, 122 t. nonfebrile seizure occurrence and, 122 t. previous incidence and, 121, 1211. prognosis and, 104, 104 t. single febrile illness and, 93 spontaneous seizure occurrence and, 103, 104 Red cells and hyperthermia, 67 Refractory seizures, differential diag- nosis and, 114 Prognostic factors, 135 age at onset, 104 t. birth abnormalities, 104 t., 105 duration of febrile seizure, 89, 101 EEG abnormalities, 101-3 family history convulsions, 104, 104 t. epilepsy, 104 t., 105 focal febrile seizures, 105 mental retardation, 104 t. recurrence of febrile seizures, 103- 4, 1041. severity of febrile seizures, 104 t. sex, 104 t. Reserpine, experimental seizures and, 155 Prophylactic therapy, 115-32; see also Anticonvulsant treat- ment Respiratory acidosis, 66 Respiratory alkalosis, 66 Psychological testing, 19 Psychomotor retardation, 35, 351., 36 Respiratory infections, upper, 20, 54-551., 56, 68 Respiratory volume, 174, 175 Ribadeau-Dumas, L., 67 Psychomotor seizures, 98, 99 t., 101, 102 Richards, J. 8., 141 Rodbard, S., 167 Rohmer, F., 43 t. Pupo, P. P., 15, 33 t. Pyelitis, 26-27 t. Pyrictal, 125-28, 155-58; see also phetharbital Pyrifer, 174; see also Pyrogens Pyrogens, 141, 142; see also Hyper- thermia Rosenblum, J., 54 t., 70 t., 108 t. Roseola infantum, 26-27 t., 28, 53, 54-551., 56 convulsions and incidence, 69, 70 t., 71 encephalitis and, 71 hemiparesis and, 107, 108 t. spinal fluid in, 69 toxic allergic process and, 71 Race etiology and, 83-85 incidence and, 15, 64 t. 220 INDEX Rosvold, H. E., 170 Ruch, T. C., 145 causes, 6-7 ts. etiological classification cryptogenic, 5, 6 t. symptomatic, 5, 6 t., 7 t. with structural cerebral lesions, 5, It. without structural lesions, 5, 6 t. familial incidence, 47, 80-81 t. nonfebrile classification 5, 6-7 t. family history, 47 precipitants of, 129 Saidman, L. J., 179 Salbreux, R., 27 t., 41, 431., 92 t., 94 t. Salicylates, toxic effects, 131 Salmonellosis, 26-27 #., 28 brain pathology and, 107 Salvarsan fever, 142 Samson-Dollfus, D., 75 Scarlet fever, 69 Schental, J. E., 182 Sepsis, 26-27 28 Septicemia, 26-27 t., 28 Schinnerling, W., lit., 331., 35#., 431., 52, 78#., 81#., 92#., 94 #., 117 #., 133 Sequelae; see Prognosis; Prognostic factors Schmidt, R. P., 77, 100, 106, 109, Serpasil (reserpine), 155 Serum; see also Blood chemistry ab- normalities calcium, 57 t., 58 globulins, 57 t., 58 phosphorus, 57 t., 58 sodium, 57 t., 58 Sex incidence, 9-111., 12, 77 male preponderance, 77 ratio, 12, 77 116#., 140, 167, 169, 182 Scholz, W., 183 Schuster, E. M., 76 Schwartzman, J., 108 t. Secobarbital sodium (Seconal), 112 t. Seizure grades, 28-32, 30-31 ts. Seizure pattern age and, 98 cerebral maturation and, 98 clinical, 28-32, 98-99, 102 akinetic or flaccid, 28-30 autonomic, 98 clonic, 28-30 focal, 28-30, 98-99, 102 generalized, 28-30 grand mal, 98-99, 99 t., 102 petit mal, 98, 99 t., 102 psychomotor, 98, 99 t., 102 tonic, 28-30 experimental, 143-45 species differences, 143, 144 localization of lesion and, 98-99 prognosis and, 105 Shanks, R. A., 8, 10 #., 14, 26 #., 28 Shea, E., 177 Shepherd, M., 183 Shibolet, S., 181 Shigella dysentery, 26-27 t., 28, 54- 55 f., 57 t., 61, 63 t. etiological significance of, 71, 72 t., 73 age and, 71 familial epilepsy and, 71, 79 hereditary predisposition and, 79 neurotoxin formation, 71 water and electrolyte imbal- ance, 73 spinal fluid findings in, 73 Shigella-negative diarrheas, 71, 72 t., 73 Shigellosis; see Shigella dysentery Short wave radio, hyperthermia, 177 Simon, J. F., 178, 179 Simple febrile convulsions, 4, 5, 34, 48, 49, 50 t., 77, 1041. Seizure susceptibility; see Convul- sive threshold, Seizure thresh- old Seizure threshold, 60-64, 66, 85, 145-60; see also. Experimen- tal febrile convulsions Seizures; see also Spontaneous sei- zures INDEX 221 EEC findings, 48, 50 t. incidence of “epilepsy,” 50 t. of spontaneous seizures, 49 Smallpox vaccination, 58 Sodium metabolism, seizure thresh- old and, 148-54 Swinyard, E. A., 141, 148, 149, 170 Syncope, 86 Temper tantrums, 32 Temperature, 60-65; see also Body temperature control, 113, 130-32 Temporal lobe seizures, 98, 991., 101, 102 Tempra (acetaminophen), 131 Terrainycin (oxytetracycline), 155 Teschan, P., 169 Thermal epilepsy, 67 Thom, D. A., 8, 9 t., 134 Thomas, J. E., 182 Thompson, S. G., 72 t. Thorazine; see Chlorpromazine Soule, H. C„ 152, 178 Spasmophilia, 84 Speech disorder, 32 Spielmeyer, W., 183 Spinal cord transection, 144, 145 Spontaneous seizures adulthood, risk of, 133-34 family history, 47 incidence, 51, 52, 89, 90 /., 93, 94-95 t., 96 birth injury and, 93, 94-95 t. duration of febrile seizure and, 101, 102/., 1031. EEC abnormalities and, 93, 94- 95 t. familial epilepsy and, 94-951. frequency of febrile seizures and, 103 geographic location and, 94- 95 t. prospective vs. retrospective studies and, 93, 96 simple vs. complicated cases, 96 time after onset of febrile sei- zures, 96, 97, 97 t. patterns of febrile vs. mixed seizure cases, 99, 99 t. recurring, 48 time of onset, 96, 97, 97 t., 102 Threshold, 59-86; see also Convul- sive threshold; Seizure thresh- old age and, 74, 129 antihistamines and, 129 biochemical maturation and, 75- 76 cerebral anatomical maturation and, 74-75 convulsive temperature, 61, 63, 631., 64 t. treatment of, 113 EEC maturation and, 75 experimental; see Experimental febrile convulsions fatigue and, 129 penicillin and, 129 phenothiazines and, 129 race and, 85 sex and, 77 sunlight and, 129 Statistics; see Incidence Status epilepticus, histopathological findings, 183 Steam heat, 141 Steinschneider, A., 144, 145, 155 Stevens, A., 34 t., 43 #., 83 Tibrewalla, N. S., 10 t. Tille, D., 14, 26 t., 28, 73, 85 Tingey, A. H., 75 Todd’s hemiparesis, 19, 108 t., 108 Toman, J. E. P., 141, 170 Stevens, H., 85 Sugita, N., 154 Sunlight, seizure threshold and, 129 Susceptibility; see also Seizure threshold genetic factor and, 77-82 sex and, 77 Tonsillectomy, 132 Tonsillitis, 20, 25, 26-27 t., 54-55 t., 56, 68 Toxic encephalopathy, 107 Tranquilizing agents, seizure thresh- old and, 155 Trauma, 32, 33 t., 34 222 INDEX Treatment analysis of, 132 antibiotics, 20, 113, 114, 130 anticonvulsant, 111-13, 1121., 130 antipyretic, 19, 113, 130 control of convulsion. 111 electrolyte disturbance, 114 hazards cold water enema, 113 ether, 180 infection and throat culture, 114 long-term, 115 nursing care, 114-15 oxygen, 115 prophylactic, 115-32 review of literature, 132-35 specific forms, 130 temperature control, 19, 113 tonsillectomy, 132 Valquist, 8., 70 t. Varicella (chicken pox), 53 Volk, B. W., 67 Vomiting, 32 Vyas, K. J., 11£., 27 £., 92 t. Wallfield, M. J., 54 t., 70 t. Ward, A. A., Jr., 77, 100, 106, 109, 116 £., 167, 182 Warren, S. L., 178 Water and electrolyte imbalance etiological significance, 83-84 shigella dysentery and, 73 seizure threshold in animals and, 148-54 Watt, J., 72 t. Watts, J. W., 145 Wegman, M. E., 64, 138, 160, 161, 163 Werboff, J., 166 White, A. C., 176 Windle, W. F., 142, 181 Windorfer, A., 54 t., 70 t. Wolfe, W. J., 167 Woodbury, D. M., 152, 154, 180 Trimethadione (Tridione), experi- mental seizure response to, 125, 126, 126 f., 155, 156 157 t. Trovarelli, A., 111., 33 t., 35 t., 78 811., 92 t., 94 t. Turinese, A., 111., 35 t. Turnbull, J. M., 26 t., 33 t., 132 Tylenol (acetaminophen), 131 Typhoid vaccine; see also Hyper- thermia convulsions, 65, 66, 142 EEG changes, 65, 66 Yamakita, M., 175, 176 Yannet, H., 3, 148, 167 Ylppo, A. and L., 67, 182 Upper respiratory infections, 20, 54-55 t., 56, 68 Zellweger, H., 80 t., 94 t. Zimmerman, H. M., 106, 183