474

THE MINOR EPILEPSIES OF CHILDHOODBy NEIL GORDON, M.D.EDIN., M.R.C.P., M.R.C.P.E.

Senior Registrar to the Departmeni of Neurology, St. Mary's Hospital, London W.2.

IntroductionThere is no doubt that children suffering from

epilepsy tend to manifest its minor forms withgreater frequency than adults. However, beforeconsidering this problem further, it is essential todefine the terms to be used, as a certain amountof confusion has arisen as the result of cliniciansand --electroencephalographists sometimes usingthe same terms to refer to attacks of a differentnature. No classification of the epilepsies is en-tirely satisfactory but it seems reasonable to referto all seizures, other than generalized convulsionsor grand mal, as minor epileptic fits. Petit mal,when the term is unqualified, should only be usedwhen referring to those attacks which appear tobe of diencephalic origin and are associated withoutbursts of three cycle per second spike and waveactivity in the electroencephalogram. Symonds(I955) prefers to describe them as minimalseizures or central epilepsy, as he maintains thatpetit mal has lost its purely clinical meaning andis a definition in terms of the electroencephalo-gram. This group may be further subdivided, anumber of the patients suffering from attacks ofthis kind being grouped under the term pykno-lepsy. This matter will be discussed in detaillater, but suffice it to say at the moment that thedistinction proposed is based not only on a numberof clinical features but on certain electroence-phalographic findings as well.

Seizures defined as petit mal include the familiarbrief lapse of consciousness and also the myoclonicattack consisting of involuntary jerks of themuscles. It is almost certain that there is a thirdtype in which the patient falls to the groundwithout loss of consciousness. Lennox (1945) hasreferred to this as an akinetic attack, but Gastaut(I954) does not consider that it is a separateentity and claims it is due to a sudden involvementof the lower limbs by a myoclonic jerk. It isgenerally assumed that these fits, included underthe term petit mal, are the result of some func-tional disturbance and are not caused by anydemonstrable alteration of structure. However,attacks which are clinically indistinguishable cer-

tainly can occur after an acquired lesion, and, forthis reason alone, it seems justifiable to use theterm petit mal in the manner suggested, com-bining both clinical and electroencephalographiccriteria in the definition.

It is then possible to refer to such attacks,when they are acquired, as symptomatic petit mal.In this group, as will be discussed later, thehistory and examination will show evidence thatthe patient is suffering from the results of organicbrain damage in addition to epilepsy. Thisdamage is most commonly of a diffuse nature,such as that due to a lipoidosis or a subacuteencephalitis of the Dawson and van Bogaert type.

Focal epileptic attacks do occasionally occur inchildren, and these, as in adults, most frequentlyarise from the temporal lobes. However, owingto the restricted capacity of a child to describe theodd features that often characterize this type ofseizure (Gordon, 1957), it may not be possible onclinical grounds to be certain if they really are ofcortical origin. The impairment of consciousnessmay be very brief and, if this is unassociated withany remembered symptoms, these ' pseudo-lapses' of temporal lobe epilepsy may easily beconfused with petit mal. Recent work (Gastaut,I954) has suggested that the term focal epilepsyfor this type of attack is inappropriate, implyingthat the fits arise in a restricted area of the cerebralcortex. It may be better to refer to them aspartial epilepsy as there is considerable evidencethat they do not develop in cortical areas or sub-cortical areas alone but in the sectors consistingof the two closely interconnected in both direc-tions.

Petit MalEpileptic attacks of this kind undoubtedly occur

more frequently than any others during childhoodand they most often take the form of an inter-ruption of consciousness, myoclonic or akineticseizures being rarer. If the term petit mal is usedin the restricted sense suggested, it is assumed thatthese fits do not arise from any demonstrablelesion and in the majority of patients suffering

copyright. on A

ugust 1, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.34.395.474 on 1 S

eptember 1958. D

ownloaded from

GORDON: The Minor Epilepsies of Childhood

with petit mal there is certainly nothing in thehistory or an examination to suggest the presenceof any structural defect. There is, however, con-siderable evidence that the disturbance of functionunderlying this type of epilepsy arises within thediencephalon (Jasper and Droogleever-Fortuyn,1947), and, whatever its nature, it does renderthese patients unduly liable to seizures and un-doubtedly in many cases it is transmitted as anhereditary factor. The fits, however frequentthey may have been, tend to cease as the childgrows older, although occasionally in adult lifethe patient may continue to suffer from grand mal,the minor seizures stopping altogether or onlyoccurring with increasing rarity.

Although considerable stress has been laid onthe electroencephalographic findings in the diag-nosis of petit mal, the clinical features of theattacks are often typical enough for this to be madewith confidence without further investigationsbeing needed. However, seizures of a differentkind can cause what appears to be only an inter-ruption of consciousness so that the finding ofthree-per-second spike and wave activity in theelectroencephalogram will undoubtedly assist inconfirming this diagnosis.As with any other investigation, the electro-

encephalographic findings cannot be taken as theonly criteria of diagnosis and must be combinedwith the clinical features of the attacks, and onlywhen both of these are typical can the diagnosis beaccepted without further question. This com-bination of clinical and electroencephalographicevidence should dispel many of the criticisms ofthis definition of petit mal and, in any case inwhich there are unusual features, the diagnosisshould not be made at least until further investiga-tions and perhaps the passage of time have estab-lished the nature of the fits beyond all reasonabledoubt.Unusual types of seizures have been recorded

under the heading of petit mal or centrencephalicepilepsy, and, to account for these, it has beensuggested that there may be an asymmetrical andrestricted spread by continuity of the epilepticdischarge in the central areas of the brain, followedby projection of the discharge to a restricted areaof the cortex in one hemisphere. This wouldthen lead to an attack mimicking a focal corticalseizure (Howell, 1955).

If the outbursts of spike and wave activity inthe electroencephalogram are in any way atypical,they cannot be accepted as pathognomonic of petitmal. This applies both to the frequency of theabnormal waves and to their symmetry. As willbe discussed later, spike and wave activity oc-curing at a frequency of less than three cyclesper second has an entirely different significance;

also, if these discharges showed a consistent asym-metry, it would have to be considered whetherthey could be secondary to a focal epileptogeniclesion in some part of the cerebral cortex. Asym-metrical discharges of this type may turn out tobe of no particular significance, but they mustalways be viewed with suspicion as the spike andwave complex is undoubtedly a characteristicresponse of the infantile cerebrum to epileptogeniclesions of varying situation (Calderon and Paal,'957).There is one type of focal epilepsy that is par-

ticularly liable to cause difficulty in diagnosis, evenin the older age groups. If the attack starts on themedial surface of the cerebral hemisphere, the onlyelectroencephalographic abnormality may be gen-eralized outbursts of spike and wave activity(Tiikel and Jasper, 1954). However, the com-plexes are usually asymmetrical, irregular in formand of a frequency varying between two and two-and-a-half cycles per second. In explanation ofthese generalized abnormalities arising from focalcortical lesions it seems reasonable to suppose thatthe epileptic discharge, originating in the para-sagittal areas, fires into a centrally-placed region inthe diencephalon, from whence it is relayed back tothe cortex on both sides (Penfield and Jasper,I947). Such fits are rare, but should be con-sidered when the spike and wave discharges areatypical and are associated with clinical mani-festations unlike those of petit mal. These mayinclude periods of confusion and automatic be-haviour, adversive tonic-clonic convulsions, sensa-tions referred to the head and throat, arrest ofspeech and forced thinking.An example of such a diagnostic problem has

been given by Bates, Cobb and Williams (I956).A child of twelve had been suffering for some yearsfrom grand mal and outbursts of aggressive be-haviour. All investigations, which included air-encephalography and cerebral arteriography, werenormal except from the electroencephalogram.This showed frequent bilateral sharp and slowwaves complexes in the frontal regions, but therewere slight and consistent differences on the twosides. As the child was quite unmanageable to theextent of having to be admitted to an adult mentalhospital, it was decided on the electroencephalo-graphic evidence alone to turn a frontal flap. Alarge angiomatous malformation with evidence ofprevious haemorrhage and thrombosis was found,and after this had been removed there was suffi-cient improvement for the patient to return to anordinary school.

It has also been maintained that confirming thediagnosis of petit mal by finding typical spike andwave discharges in the electroencephalogram maybe misleading, because they sometimes occur in

September 1958 475copyright.

on August 1, 2020 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.34.395.474 on 1 Septem

ber 1958. Dow

nloaded from

POSTGRADUATE MEDICAL JOURNAL

the records of patients suffering only from grandmal (Jasper and Kershaw, 1941). This does notseem to be a justifiable criticism, as the same couldbe said of the occasional, well-marked focal abnor-malities in the electroencephalograms of patientswith only grand mal; the spread of the dischargebeing so rapid that there is never any clinicalevidence of the focal origin of the attacks. Ineither case it has been the electroencephalogramwhich has provided evidence as to where theepilepsy has originated from, grand mal being theend result of any such disturbance.

Petit mal attacks may occur with a frequency ofseveral hundred a day, but, just as in other types ofepilepsy, recovery. may not be complete betweeneach individual seizure and then the patient canbe said to be in petit mal status. This manifestsitself clinically by periods of confusion and auto-matic behaviour and, if it is not realized that thechild is an epileptic, a diagnosis of a behaviourdisorder may be made. ' Status' of this kind is,undoubtedly, commoner in symptomatic petit malthan in the ' ideopathic ' variety.

PyknolepsyAnyone who has treated a number of children

with petit mal will have been struck by the factthat the response to treatment may vary widely.For instance, one such child, when started onTroxidone, will cease to have attacks, whileanother with a somewhat similar history will notrespond at all. When reviewing the electro-encephalograms of these children it has beennoticed that although all of them might show out-bursts of spike and wave activity only some wouldmanifest runs of occipital delta activity (Elston,Gordon and Cobb, 1956). This rhythm wasusually at a rate of three cycles per second, wasoften of high voltage, was strikingly blocked byeye opening and was increased in continuity andamplitude by overbreathing (Cobb, 1954). Whenthe histories of the group of children showingthese occipital delta rhythms in their encephalo-grams were analysed and compared with those ofthe remaining children with petit mal, certaindifferences were discovered. Their fits started atan earlier age, they had fewer major attacks butmore frequent minor ones, and the incidence ofepilepsy in their families was higher. However,the finding which differentiated the two groupsmore than any other was the response to treatment.This can only be assessed over a reasonably longperiod, as petit mal attacks frequently stop andstart for no definite reason and, if this fact is nottaken into account, unjustifiable credit may begiven to a particular therapy. The patients show-ing occipital delta activity were the ones who, withfew exceptions, had never shown conclusive evi-

dence that treatment had affected the frequencyof their fits to any marked degree. Some of themseemed to have obtained temporary benefit, but itwas commonplace for these patients to have beengiven several different types of drugs in an effortto control their seizures. In contrast to this thecontrol group of patients with petit mal usuallypresented no particular therapeutic problem,especially when they were given troxidone or oneof the other oxozalidine-dione derivatives. There-fore, although no adequate explanation can begiven for the presence of these delta waves in thepost-central regions, the finding of them may beof prognostic significance and suggest that thechild's fits may respond poorly to treatment.The natural history of the illness in the occipital

delta group was closely akin to that of pyknolepsyas it was originally defined by Friedmann (I906),,Sauer (I9I2) and Adie (1924). This syndromewas described as occurring in childhood andadolescence and consisting of very frequent minorattacks, even as many as a hundred or more a day,only lasting for a few seconds at a time. Theseseizures started suddenly between the ages of fourand twelve, were of uniform severity, might lastfor a few weeks or for several years, and were un-influenced by anti-convulsant remedies. They didnot impede normal mental and physical develop-ment and ultimately ceased spontaneously, neverto return. However, Adie (1924) considered thatif the patient developed grand mal the prognosiswas altered.

It has been suggested that the term pyknolepsyshould be dropped (Williams, 1950), and that thereis no justification for separating a special groupfrom the general run of children suffering frompetit mal. However, as Tizard (I954) has main-tained, the so-called ideopathic epilepsies of child-hood may well be separate disease entities withsimilar symptoms and he would oppose anytendency to group them together in case it shouldslacken the search for the underlying pathology.The occurrence of occipital delta rhythms in

the electroencephalograms of some patients withpetit mal suggests that the disturbance of functionunderlying this type of epilepsy may indeed notbe uniform in its nature. This is supported bythe slight variations in the clinical behaviour ofthe occipital delta group as exemplified by thedescriptions of pyknolepsy, particularly as far asthe response to treatment is concerned. Theseelectroencephalographic findings.may be helpfulwhen planning treatment as, although it willalways be justifiable to try the effect of variousdrugs and combination of drugs, a failure tocontrol the fits after a reasonable period of observa-tion may lead to a decision to abandon them en-tirely. This will avoid submitting these patients

476 September 1958copyright.

on August 1, 2020 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.34.395.474 on 1 Septem

ber 1958. Dow

nloaded from

GORDON: The Minor Epilepsies of Childhood

to unnecessary risks associated with such drugs astroxidone, apart from the fact that the generalsense of well-being is often improved when treat-ment is stopped.

Symptomatic Petit MalIt has been stated that the diagnosis of ' petit

mal' assumes the absence of any discoverablestructural lesion. However, it seems possible thatan injury to the diencephalon early in life mightcause a similar predisposition to minor seizure ofthis type, which was not constitutional. Jasperand Kershaw (I94I) in a series of patients showingthree-per-second spike and wave activity in theirelectroencephalograms reported that, in 2I percent. of them, there was a previous history ofcerebral trauma either at birth or soon afterwards,and, in 9 per cent., a history of various types ofinfection which might have involved the centralnervous system. From this evidence they deducedthat this type of epilepsy could be acquired.Gastaut (I954) has also stressed that brief inter-ruptions of consciousness may be caused by alesion of the diencephalon resulting from ' en-cephalitis.' He maintains that these ' lapses' fol-lowing such an illness are as common as sympto-matic bilateral myoclonus, although the latter hasreceived so much more attention in the literature.However, the myoclonus is usually associatedwith definite evidence of structural damage, whichmay include the cerebellum as well as the cerebrum(Unerricht, I89I; Hunt, 192I).

It is impossible to prove for certain that thesepatients are suffering from symptomatic epilepsy,as physical examination and investigations willshow no definite evidence of cerebral damage. Onthe other hand some patients whose minor fitsconsist of brief interruptions of consciousness,indistinguishable from those of petit mal as alreadydefined, will reveal definite evidence of structuraldisease of the central nervous system. Theirattacks will usually be accompanied by electro-encephalographic changes consisting of generalizedoutbursts of spikes or slow waves or the combina-tion of the two in ragged complexes, at varyingfrequencies. The underlying disease is most oftenof a diffuse nature, such as that due to a cerebrallipoidosis or a subacute encephalitis of the Dawsonand van Bogaert type. For instance, a patientwho is ultimately diagnosed, perhaps only at post-mortem examination, as an example of the firstcondition may present with frequent minor epi-leptic fits, soon to be followed by increasing mentaland physical deterioration. Sometimes the elec-troencephalogram is of considerable diagnosticassistance in patients suspected of having cerebrallipoidosis or ani inclusion-body encephalitis. Inthe former disease the encephalogram may show

random generalized sharp transients on a back-ground of slow waves (Cobb, Martin andPampiglione, 1952), and in the latter the charac-teristic abnormality is the appearance of regularepisodes of high-voltage slow waves at intervalsof about eight seconds (Cobb and Hill, I950).

Gibbs, Fleming and Gibbs (I954) have describedthe encephalographic findings in a group ofchildren suffering from what they term ' infantilespasms': attacks consisting of sudden jerking ofthe head, rolling of the eyes, upward flinging ofthe arms, and quivering of the entire body. Theseattacks might occur at extremely frequent intervalsand it might be difficult to tell if consciousness wasimpaired or not. About a third of the childrenhad grand mal as well. The electroencephalo-grams showed high-voltage random slow wavesand spikes in all cortical areas and the term' hypsarhythmia' was used to describe thesechanges. The clinical attacks were accompaniedby outbursts of fast waves and high-voltage spikes.These children were frequently mentally retardedor showed other evidence of cerebral damage. Theetiology was not known in over half of the cases,but encephalitis or birth injury were two of thecommonest causes. Hess and Neuhaus (I952)have described similar electroencephalographicfindings accompanying ' Blitz, Nick und Salaam-krampfen.'

Temporal Lobe EpilepsyMost varieties of focal -or partial epilepsy are

rare during childhood, as they usually arise fromacquired lesions, more likely to be present thelonger the patient has lived, and motor and sensoryfits of this type do not usually present any par-ticular diagnostic problems. However, seizuresarising from within the temporal lobes are thecommonest type of focal epilepsy at any age and,although their incidence increases markedly inadult life (Lennox, 1951) they are still a fairlyfrequent cause of minor epilepsy in childhood.An adult often has difficulty enough in des-

cribing the clinical features of these attacks and achild may be able to add little information to thatgiven by witnesses. All that may be apparent toan observer is an interruption of consciousness,perhaps associated with abnormal behaviour ofsome kind, so that it is not surprising that adiagnosis of ' petit mal ' is made.

If these patients are closely questioned, sus-picion may at least be aroused that there is moreto the minor fits than just an interruption ofconsciousness. In fact, temporal lobe epilepsycan be exceedingly varied in the way it manifestsitself, although, however complex these seizuresmay be, there are many features in the clinicalpicture sufficiently characteristic to suggest the

September 1958 477copyright.

on August 1, 2020 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.34.395.474 on 1 Septem

ber 1958. Dow

nloaded from

POSTGRADUATE MEDICAL JOURNAI,

correct diagnosis. The fits may consist of symp-toms that would be expected to result from stimu-lation of this part of the cerebral cortex, and thesecan roughly be divided into four classes: motor,sensory, visceral, and psychic. The motor typemay be exceedingly brief and consist of only ageneralized tonic contraction of the muscles, incontrast to the clonic contractions sometimes asso-ciated with petit mal. The sensory type is charac-terized by a distortion of a normal sensation of onekind or another. For instance, the patient maycomplain of a strange smell or taste which isusually unpleasant, or visual or auditory hallucina-tions may occur. It will naturally be difficult toelicit a history of such attacks from a young child,but he may well have told his mother about them,who, until questioned by the doctor, may not haverealized their significance. The commonest mani-festation of an attack starting in the visceral brain,a term used to include parts of the insula, frontaland temporal lobes, is the ' epigastric aura.'Finally, the psychic group may cause greaterdiagnostic difficulties than any of the others.Different emotional feelings may occur, the mostcommon of these being fear; an intense feeling offamiliarity, the so-called dej"a vu phenomenon mayconstitute the fit; or, conversely, it may consist ofa feeling of strangeness. When this affects theenvironment, it may seem to the patient that he isasleep and dreaming. This is by no means acomplete list of such manifestations and they donot always occur in isolation, but may be combinedin various ways.Any part of the cerebral cortex occupied or

'exhausted' by an epileptic discharge cannot beused for normal integrative functions, and thisaccounts for other features of temporal lobeepilepsy. For instance, if parts of the temporallobe of the dominant hemisphere are inactivated inthis way, it is likely that a disturbance of speechwill result and it is not uncommon for transientattacks of dysphasia to occur in this type of focalepilepsy. Although the temporal lobes subservemany complex functions they do not seem to beessential for the more stereotyped ones, whichprobably explains the frequent association ofautomatism with temporal lobe attacks. Duringsuch episodes one is observing the integratedaction of the brain excluding that part of thetemporal lobe involved in the epileptic disturbance.The patient may perform a few confused move-ments which are not even purposive, or may carryout co-ordinated movements, though in such away that even the inexpert eye can recognize thatsomething is wrong. Smacking and licking of thelips is a common example of such attacks. How-ever, occasionally the conduct of the patient duringthese periods may appear relatively normal. He

may talk, walk around and even carry out skilledactions. On recovery, the patient will have norecollection of what has happened, and this is nodoubt related to the fact that parts of the temporallobe appear to be essential to the laying down ofmemory patterns.

Patients with this type of epilepsy are particu-larly liable to disorders of emotion and behaviour.This may present a considerably greater problemthan the occurrence of occasional fits and may bethe reason for seeking medical advice. This asso-ciation is so common that, if an epileptic childpresents with abnormalities of behaviour, thepossibility that the fits are arising in the temporallobe should be seriously considered.

Although the diagnosis of temporal lobe epilepsyin children may be particularly difficult, help canoften be obtained from the electroencephalogram.The specific seizure discharge is the occurrence ofwell-marked focal abnormalities in the temporalregions. These may consist of irregular slowwaves or of random spikes and sharp waves.However, the record taken when the patient isawake is frequently normal or it may only show anon-specific dysrhythmia. It is then of value totake a record while the patient is going to sleep orduring a period of deep sleep (Gibbs and Gibbs,1947). Spikes and other focal abnormalities mayappear, most frequently as the patient becomesdrowsy, but sometimes only when he is asleep.Once the diagnosis of temporal lobe epilepsy has

been made, the implications are the same as forany other type of focal epilespy. Further inves-tigations may well be indicated, as even in child-hood these attacks may be the first symptom of acerebral tumour. However, as in any other typeof epilepsy, there will be many cases in which nodefinite underlying pathology can be demonstratedalthough the etiology may be suggested from otherevidence. For instance, a history of cerebraltrauma is not uncommon, the hippocampal gyrusbeing particularly liable to damage in closed headinjuries. This may be due to generalized cerebraloedema causing herniation of the hippocampusthrough the tentorial opening. A lesion in thisarea may then result from interference with theblood supply. A similar mechanism possiblyoperates in birth injuries and it has been suggestedthat the lesion may ripen into an epileptogenicfocus in later life (Earle and Penfield, I953).The treatment of temporal lobe epilepsy presents

particular difficulties and these fits will not usuallyrespond to the same drugs that are likely to controlpetit mal. Therefore, quite apart from its differentdiagnostic and prognostic implications, it is im-portant to separate this variety of epilepsy fromthe other minor seizures of childhood. Wheninvestigations have helped to exclude any underly-

478 September 1958copyright.

on August 1, 2020 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.34.395.474 on 1 Septem

ber 1958. Dow

nloaded from

GORDON: The Minor Epilepsies of Childhood

ing condition likely to be amenable to surgery, adecision will have to be taken on medical treat-ment. It is always worth while trying the effect ofbarbiturates, either alone or combined withphenytoin sodium. Then, if the fits are stillpoorly controlled, other drugs, such as methoinand phenacetylurea, may have to be given,although they carry a risk of toxic complications.Finally, if it has to be admitted that this treatmenthas failed and there is a well-marked electro-encephalographic focus, the question of its surgicalexcision should be considered. In such a case,other investigations may well have been negativeand even at operation no definite macroscopicalevidence of a lesion may be found. However,microscopical examination of the material removedsometimes shows evidence of focal damage, suchas atrophy or a small vascular anomaly.

SummaryWhen a diagnosis of epilepsy is made in child-

hood particular attention should be paid to thecharacteristics and pattern of any minorattacks that may have occurred, although it willnot always be possible to classify these into a par-ticular group which has an especial significanceof its own. It will be found that the majority ofthe minor fits will consist of a brief lapse ofconsciousness or of myoclonic jerks. The term' petit mal ' should only be used to refer to fitsof this kind, which appear to have their origin indeep, mid-line structures and are associated withthree-cycle-per-second spike and wave activity inthe electroencephalogram. There is a subdivisionof this group, which, it is suggested, is a definiteclinical entity of its own and to which the termpyknolepsy should be restricted. The minor fitsof pyknolepsy have no specific characteristicsexcept, perhaps, their tendency to occur at veryfrequent intervals. However, they are particularlydifficult to control with anti-convulsant drugs ofany kind, and it is common to find that patientssuffering from pyknolepsy have been tried onnumerous drugs and have been taken to see manydifferent doctors by their distressed parents.Although the immediate prognosis is poor, theultimate prognosis is good and the attacks fre-quently tend to stop at puberty or soon afterwards.The reassurance that can, therefore, be given tothe parents is often of great comfort. Theclinical diagnosis of attacks of this kind may onlyl)e made in retrospect, but the associated electro-encephalographic changes can be of considerablehielp. In addition to the paroxysmal outbursts ofspike and wave complexes, runs of occipital deltaactivity may be found in the records of thesepatients. It is suggested that the slight variationsin the natural history of the fits and the electro-

encephalographic abnormalities warrant separat-ing, under the term pyknolepsy, a number ofpatients suffering from brief lapses of conscious-ness, as it is possible that the underlying disordermay in some way differ from that causing otherattacks of a similar kind.

Brief lapses of consciousness and myoclonicjerks may be the initial manifestation of an organicdisease of the brain, and these attacks may bereferred to as symptomatic petit mal. The patho-logical process is usually found to be diffuse anddegenerative, such as a cerebral lipoidosis or asubacute encephalitis. The course of the illnessand the results of special investigations will oftenindicate the correct diagnosis. The electroen-cephalogram may show not only paroxysmal dis-charges, but a continuous background of diffuseabnormalities. The paroxysmal outbursts will nothave the regular pattern associated with petit mal,but will consist of spikes or slow waves, eitheralone or very irregularly combined into raggedcomplexes at a frequency which is usually belowthree cycles per second. These changes may besufficiently specific to support a particular diag-nosis.There is a group of children whose minor

attacks consist not only of impaired consciousnessbut also of myoclonic jerks or spasms. Their elec-troencephalographic findings have been groupedunder the term hypsarhythmia, the abnormalspikes and slow waves being of particularly highvoltage. These children show, in most cases,other evidence of damage to the central nervoussystem, especially mental defect, and there maybe a previous history of such conditions as en-cephalitis or birth injury.

Focal epileptic fits are relatively uncommonduring childhood, and even when they do occurmay not have the same significance as in adult life.In very young children the apparent focus oforigin of the seizures may change and no definiteevidence of a corresponding lesion may ever befound. Another point of confusion is that inchildren an epileptogenic lesion in the cortex maycause a spike and wave discharge. However, thisis usually well localized except in the case oflesions on the medial surface of the cerebralhemispheres, when this type of discharge may begeneralized.Although not as common as in adults, fits

arising in the temporal lobes may sometimes causediagnostic difficulties, particularly as a child willfind it hard to describe the varied manifestationsof this type of epilepsy. The attacks may consistof features that would be expected to occur fromstimulation of this part of the cortex, such asolfactory, visual and auditory hallucinations or,because the cortex cannot be used for normal

Se-ptennber I1958 479copyright.

on August 1, 2020 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.34.395.474 on 1 Septem

ber 1958. Dow

nloaded from

480 POSTGRADUATE MEDICAL JOURNAL September 1958

functions while it is occupied by an epilepticdischarge, there may be evidence of the limitedintegration of the rest of the brain in the form ofperiods of ' automatism.'The diagnosis from petit mal is of special im-

portance as the presence of a progressive destruc-tive lesion in the temporal lobe will have to beexcluded, and if there is no indication for surgicaltreatment, the type of drug likely to control theseizures will be entirely different. Troxidone isone of the most successful drugs in the treatmentof petit mal, while focal fits arising in the temporallobes may respond best to one of the hydantoinateseries.Even if it is not always possible to make an exact

diagnosis of the variety of epilepsy from which aparticular child is suffering, a detailed study ofthese seizures will frequently produce a great dealof information on the various disorders that mayunderlie them. For instance, it seems probablethat petit mal may be the result of more than onekind of disturbance of cerebral function, thesimilarity of the fits only being a reflection of thelimited way in which a child's brain can react tosuch stimuli.

BIBLIOGRAPHYADIE, W. J. (1924), Brain, 47, 96.

BATES, J. A. V., COBB, W., and WILLIAMS, D. J. (I956),Electroenceph. cin. Neurophysiol., 8, i6i.

CALDERON, A., and PAAL, G. (I957), Ibid., 9,350.COBB, W. (igs4), Proc. roy. Soc. Med., 47, 846.COBB, W., and HILL, D. (ig5o), Brain, 73, 392.COBB, W., MARTIN, F., and PAMPIGLIONE, G. (1952), Ibid.,

75, 343-EARLE, K. M., and PENFIELD, W. (I9S3), Arch. Neur. Psychiat,

(Chicago), 69, 27.ELSTON, C., GORDON, N. S., and COBB, W. (I956), Electroen-

ceph. din. Neurophysiol., 68, I6i.FRIEDMANN, M. (I906), Dtsch. Z. Nervenheilk., 30, 462.GASTAUT, H. (I954), 'The Epilepsies,' Charles C. Thomas,

Springfield, Illinois, U.S.A.GIBBS, E. L., and GIBBS, F. A. (1947), Proc. Ass. Res. Nerv.

Ment. Dis., 26, 366.GIBBS, E. L., FLEMING, M. M., and GIBBS, F. A. (1954),

Pediatrics, 13, 66.GORDON, N. S. (I9S7), Postgrad. med. J., 33, 228.HESS, R. J., and NEUHAUS, T. (I952), Arch. Psychiat. Nervenkr.,

I89, 37.

HOVELL, D. A. (I955), Brain, 78, 199.HUNT, R. (1921), Ibid., 44, 490.JASPER, H., and DROOGLEEVER-FORTUYN, J. (I947), Proc.

Ass. Res. Nerv. Ment. Dis., 26, 272.JASPER, H., and KERSHMAN, J. (I94, Arch. Neur. Psychiat.

(Chicago), 45, 903.LENNOX, W. G. (I945), J. Amer. med. Ass., 129, I069.LENNOX, W. G. (Igsi), Neurology, 1, 357.PENFIELD, W., and JASPER, H. H. (I947), Proc. Ass. Res. Nerv.

Ment. Dis., 26, 252.SAUR (1912), Mschr. Physiat. Neurol., 40, 27.SYMONDS, C. P. (I955), Brit. med. Y., i, 1235.TIZARD, J. P. M. (I954), Proc. roy. Soc. Med., 47, 845.TUKEL, K., and JASPER, H. H. (I955), Electroenceph. clin.

Neurophysio., 4, 48I.UNVERRICHT, H. (I891), 'Die Myoclonie,' Leipzig.WILLIAMS, D. (1950), Brit. med. Y., i, 685.

CARDAC DIS ASEPrice 3s. lld. post free

INTRODUCTION DRUG TREATMENT OF HYPERTENSIONWalter Somerville, M.D., M.R.C.P. E. G. McQueen, M.B., M.R.C.P., and F. H.

ANGIOGRAPHY Smirk, M.D., F.R.C.P.J. Norman Pattinson, M.B., B.Chir., D.M.R.D., TREATMENT OF BACTERIAL

F.F.R.ENDOCARDITISBEDSIDE DIAGNOSIS OF CONGENITAL Ian G. W. Hill, C.B.E., T.D., M.B., F.R.C.P.E.,HEART DISEASE MRCP,FRSEWaltcr Somerville, M.D., M.R.C.P. M.R.C.P., F.R.S.E.

SURGICAL TREATMENT OF CONGENITAL THE MANAGEMENT OF COR PULMONALEHEART DISEASE J. F. Goodwin M.D., M.R.C.P.W. P. Cleland, M.R.C.P., F.R.C.S. J

PREGNANCY AND RHEUMATIC HEART THE CARDIAC RISK IN ANAESTHESIADISEASE AND SURGERYSamuel Gram, M.D., F.R.C.P. Graham W. Hayward, M.D., F.R.C.P.

Published byTHE FELLOWSHIP OF POSTGRADUATE MEDICINE

60, Portland Place, London, W.A

copyright. on A

ugust 1, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.34.395.474 on 1 S

eptember 1958. D

ownloaded from