10.1192/bjp.152.2.246Access the most recent version at DOI: 1988, 152:246-252.BJP 

S J Logsdail and B K ToonePost-ictal psychoses. A clinical and phenomenological description.

Referenceshttp://bjp.rcpsych.org/content/152/2/246#BIBLThis article cites 0 articles, 0 of which you can access for free at:

permissionsReprints/

permissions@rcpsych.ac.ukto To obtain reprints or permission to reproduce material from this paper, please write

to this article atYou can respond http://bjp.rcpsych.org/cgi/eletter-submit/152/2/246

from Downloaded

The Royal College of PsychiatristsPublished by on April 5, 2012http://bjp.rcpsych.org/

http://bjp.rcpsych.org/site/subscriptions/ go to: The British Journal of PsychiatryTo subscribe to

British Journal of Psychiatry (1988), 152, 246—252

The association between epilepsy and psychosis hasgenerated considerable debate and interest since thenineteenth century. Acute states of excitation relatedto seizures were first described by Esquirol in 1838and later taken up by Jackson (1875). Modernresearch using scientific measurement (Slater et a!,1963) and methods of observation of knownreliability (Perez & Trimble, 1980; Toone eta!, 1982)has concentrated on the chronic, or inter-ictalpsychoses; observations on post-ictal psychotic stateshave been largely ignored. Some studies include aproportion of such patients in a larger epilepticpsychiatric cohort (Clark & Lesko, 1939; Levin,1952; Dongier, 1959; Betts, 1974), but the salientfeatures that would permit precise diagnosticclassification are not always available or data isarranged in such a way that it is not possible to drawconclusionsabout thisparticulargroup. Basicinformation remains unknown (e.g. type of epilepsy,phenomenology, time course and outcome), and itis not possible to examine the possible links betweenpost-ictal and inter-ictal epileptic psychoses. Thepresent study aims to clarify the position by describing14 cases.

Method

Admissions to the Bethlem Royal and Maudsley Hospitalswere screened by computer search for the years 1967—1985;eight cases were thus identified. The department ofPsychological Medicine at King's College Hospital wasscreened for 1980—1985;four cases were identified. Onepatient attended the National Hospitals for NervousDiseasesand one the David LewisCentre for Epilepsy. Eachpatient had been in the care of a consultant psychiatristwith a special interest in epilepsy and had been studied asan in-patient at the time of psychosis. An EEG was availablein all cases and most patients had undergone multiple

recordings. Case-notes and EEG recordings were carefullyscrutinised; further information from other hospitals wasobtained wherever possible. The inclusion criteria forepilepsy were those of Gunn & Fenton (1969), and thisdiagnosis was unequivocal in each case.

A diagnosis of post-ictal psychosis was accepted if thefollowing four criteria were fulfilled.

The first was that the episode of confusion or psychosismanifested immediately upon a seizure or emergedwithin a week of the return of apparently normal mentalfunction.

The second criterion was that the psychosis had aminimum length of 24 hours and a maximum length of threemonths.

The third criterion was that the mental state wascharacterised by one of the following:

(a) cloudingof consciousness,disorientation,or delirium(b) delusions, hallucinations, in clear consciousness(c) a mixture of (a) and (b).The fourth criterion was that there was no evidence of

the following extraneous factors which might havecontributed to the abnormal mental state:

(a) anticonvulsant toxicity —¿�based on anticonvulsantlevels where possible and also physical examinationfor evidence of cerebellar dysfunction in each case

(b) a previous history of inter-ictal psychosis(c) EEG evidence of minor status(d) recent history of head injury, or alcohol or drug

intoxication.

Clinical evaluation

Information included the following: onset age of epilepsy;type of seizure; anticonvulsant drug treatment; onset ageof psychosis; lucid interval; mode of presentation; familyhistory of psychiatric illness; premorbid personalityand social situation; physical examination; psychiatricphenomenology; laboratory examination, including anticonvulsant levels; CT scan; serial EEG examinations;psychometry and long-term follow-up.

246

Post-Ictal PsychosesA Clinical and Phenomenological Description

S. J. LOGSDAILand B. K. TOONE

Post-ictalpsychoseshavesofar receivedlittle attention. The clinicaldetailsof 14 cases,diagnosedaccordingto newly formulatedcriteria,were examined.Psychoseswere usuallyprecipitated by a run of seizures and occurred after a lucid interval. The seizures werepartial complex with secondary generalisation in 11 cases. Catego analysis of the PresentState Examination confirmed pleomorphic phenomenology. Follow-up details wereavailable in all cases, for up to eight years. Psychoses tended to recur. Similarities withchronic epileptic psychosis are discussed, and a possible organic aetiology for post-ictalpsychosis is proposed.

POST-ICTAL PSYCHOSES 247

When possible the Present State Examination (PSE)(Wing et a!, 1974), a structured psychiatric interview ofestablished high reliability, was employed. Both authorshad been trained in the use of the PSE. The Syndrome CheckList, a schedule derived from the PSE and adapted forretrospective evaluation of case records, was employed whenPSE information gained at the time of the psychosis wasnot available. Symptoms thus recorded were fed into theCatego computer program, which classified them in ahierarchical manner.

Resufts

The results are summarisedin Table I.

General characteristics

There were nine males and five females, average age 35 yearsat the time of the study. Five had one psychoticepisode,six had 2—4episodes, and three had multiple post-ictalpsychoses.

Familyhistorieswereunremarkableexceptfor onepatientwho had a father who wassaid to drink alcohol to excess,and another patient who had a cousin who was reputed tosuffer seizures.

Five patients were married, two lived with parents, onelived in a convent and the remaining six lived in hostels.Five patients were employed. Six were given an extradiagnosis of personality disorder by their doctors accordingto ICD—9criteria (World Health Organization, 1978).

General characteristicsof epilepsy

The age of onset of epilepsyranged from a few monthsto 41 years (mean 16.7 years) and the onset of first postictal psychosis was 17—56years (mean 32.2 years). The gapbetween onset of epilepsy and onset of psychosis was 3-33years (mean 15.5 years). Epilepsy classification wasundertaken according to the criteria of Gastaut (1970).Three of the 14 patients had primary generalised epilepsy(POE) causing grand ma! seizures, and the remaining 11had focal epilepsy causing partial complex seizures:generalisation of the seizures occurred intermittently in all11. Two patients were also thought to suffer hystericalseizures, although these seizures were not temporally relatedto psychosis.

Seizure control, despite careful treatment, was variable.Five patients had fits once monthly or lessoften; one patienthad them more than monthly but less than weekly; sevenhad seizuresat least weekly.One patient varied in seizurefrequency between these extremes, probably because of

intermittent non-compliance in taking medication.Four patients were taking a single anticonvulsant

drug; the remainder were on polytherapy. Phenytoin,carbamazepine, sodium valproate and phenobarbitonewere the most commonly prescribed, in varying combinations and doses. Blood levels taken at the time ofpsychosis were within the therapeutic range in 12 assays,and below in eight.

Seizures prior to the onsetof psychosis

In 12of the patients there was a clear history of an increasein major seizure frequency prior to the onset of psychosis,usually as a cluster of two or three. In one case the patientsuffered one unusually prolonged fit from which it tookhim longer than usual to recover, and in the last case noreliable history could be obtained.

Lucid Interval

Elevenof the 14patientsmade a recoveryfrom the seizuresand immediatepost-ictalconfusion to the point wherethey(and a witness where available) said they were back tonormal. One furtherpatient developed clear consciousnessprior to psychosis, but had not made a full recovery fromthe seizures, as her Todd's paralysis was incompletelyresolved. It was after this interval of apparent normalityof mental state that the onset of psychosis occurred. Thislucid interval was 1—2days in eight of the 12 patients; therange for the 12 was 1—6days, mean 2.5 days. Theremaining two patients were difficult to categorise: one wasconfused for 24 hours after recovering from epileptic statusbefore becoming psychotic, the second was havingnumerous fits daily for a month before psychosis occurred.

Mode of presentation

Three patients were brought to a casualty department byfamily members, three were admitted after OP referral, fourmorewereadmittedaftera domiciliaryvisitby a psychiatrist(two requests were from hostel wardens) and four were seenas in-patient consultations to other hospital departments.In this latter group of four, one had been under the careof physicians because of epileptic status, another wasadmitted to a labour ward suffering the delusion that shewas in labour, and the last two had been admitted toneurological wards primarily because of mental statedisturbance, but in the care of consultants who had beenmanaging their epilepsy previously. Each admission wasconsidered urgent, as were the ward consultations, althoughonly one patient caused a significant public disturbanceprior to coming into hospital.

Features of psychosis

All patients were given a diagnosis of post-ictal psychosisby theirmedicalattendants.Of the nine who wereconfusedat the onset of the psychosis, seven had experienced theintervening lucid interval of clear consciousness afterrecovery from seizures and before the onset of psychoticphenomena. This confusion persisted throughout thepsychosis in five. Three further patients had no evidenceof confusion during their psychosis, and the remaining twocould not co-operate with testing to help clarify theirconscious level. Only one of the patients had primarydelusions and thought disorder. Nine patients had amarkedly abnormal mood: elevated in three, depressed infour, and both (at different times) in two more. Three

248 LOGSDAIL AND TOONE

.@

@ .@@@ .@0@@

@ .0 i@ .0'- .0 0.@.g:.@@@ ‘¿�#‘@@v@ u u

@ •¿�

‘¿�->@‘-@D ‘¿�-@‘-@@ @-2@-.@

@u.h2 HH@ .@2@ @i@@EU 2C.) C.)

.9 @‘¿�‘@J@@ ,,@

@ -@@@ .@.@9@

H

@@ b@@@@ .@

.0 @. ,@. .@ .@ .@ .0

0. N

.@@0) U U@

@ .

@@ n@@@ .0 0@@

0 0 .@U'-0@ -@—0 0 @0 0 o

i@ ‘¿�‘aii :.ii :.@i@0. 3@0.@0. 0.@ 6@0.@@

.0'- . •¿� .0.@ @.h,.0@ .0@@@@ n.@r@O@@ cX@. a,0.@X0.@@ c@.

0 0 N 0 —¿� N C') —¿� —¿� N- - -E@ A A A

0@ f) N N N @0 N—¿� N m N C')

U

+ + + + + )@ + + ++ + + + + + + + + + I + + +

.@@ .@.@@

.@@ @-a@@@@@@@@@ 60 60 @ne••@.50@ IL.N @)ON@).Nt1ONI@ON l&,N ION@O0@U@OOOUIZON15NLL.N @o0@UlL,N@

—¿� N N r') 0 0@ 0 C')—¿� —¿� —¿� N —¿� C') .0

POST-ICTALPSYCHOSES 249

patients presented in stupor: one was catatonic; anotherturned out to be depressed; no definite diagnosis was madein the third case. Paranoid delusions were reported in sixpatients; these were poorly systematised. Six patientsreported visual hallucinations, six had auditory hallucinations, and two somatic hallucinations. The orthodoxdescription of post-ictal psychosis (Lishman, 1978) hasfour main elements —¿�confusion, visual and auditoryhallucinations, and paranoid ideation; all of these featureswere present in only 5 of the 14 cases.

PSE data

An analysis was undertaken of the phenomenology of onepsychotic episode in each patient. The best documentedepisode was taken in those cases where there was more thanone. In each of these cases the abnormalities of mental statedid not differ markedly from one episode to another, andapart from the confusion, did not change notably duringeach episode. Eighteen Catego subclasses were found: fiveschizophrenia (including one catatonic schizophrenia), fourparanoid psychosis, three affective psychosis, two mania,one hypomania, one psychotic depression, and twoborderline psychosis. The final Catego classes wereschizophrenic psychosis, four; manic and mixed affectivepsychosis, three; paranoid psychosis, four; and otherpsychosis, three. The presence of confusion did not leadto a markedly differing phenomenology: the four PSEclasses were found within the group of five patientsconfused throughout psychosis as well as within theremaining group of seven patients who were known to beconfused initially or not at all.

Neurological examination

One patient had a longstanding left hemiparesis. Duringthe psychoticphase, one had an unsteadygait, but no otherabnormalities on examination. Another developed a rightsided Todd's paralysis and dysphasia: the dysphasiaresolved over four days, and the weakness had almost goneby day 8, when the psychosis started.

Investigations

CT scan abnormalities were detected by a neuroradiologistin 5 of 12 patients: one had a right occipital pole tumour,one had an area of lowattenuationsuperficiallyin the rightparietal lobe, two had dilated lateral ventricles, and onehad widened fissuresand sulci.

EEG examination of the patients when not psychoticshowed five patients with slow wave abnormalities only.There wereintermittent spikesand/or sharp wavesin theother nine: right unilateral location in two, left unilateralin one and bilateral in six. Repeat EEG examination duringpsychosis was undertaken in ten patients. There was nochange in one, reduced abnormalities in another, andincreased spikes and/or sharp waves in six. Slow waves wereincreased in six. There was no particular association betweenEEG abnormalities and clinical picture, apart from anincrease in slow waves in those who were confused at the

time of assessment.

Psychometry was undertaken in 11 patients; in three,assessment was during the psychotic state. IQ ranged from80 to 120 in the ten patients in which this was measured;the mean IQ of the group was 92. There were localisingabnormalities (visual memory, visuospatial ability, arithmetic,and naming ability) in seven cases and these correlated withan EEG focus or localised CT scan abnormality in three.

B12 and folate assays were performed in eight patients.Serum B12 was in the normal range for each. One patienthad a red cell folate estimation, which was in the normalrange. Seven others had estimations of serum folate: thiswas marginally reduced in three -2.0, 2.8, 2.8 @tg/l(normalrange 3—15gig/I).

Course and outcome

Six patients had a serious behaviour disturbance while inpatients, including one who made a suicide attempt andtwo who werephysicallyaggressive.Two patientswereheldunder the Mental Health Act. The length of psychosisvariedfrom one to 90 days, mean 14.3 days, although eight ofthe patients had recovered within one week.

Eight were treated with major tranquillisers: trifluoperazine, chlorpromazine, or haloperidol. One patient wasgiven lithium. ECT was not prescribed.

Six patients had further generalised seizures whilepsychotic, and in five these were followed by a worseningof psychotic symptoms. Two of these patients also hadpartial complex seizures without secondary generalisation,and in one this led to an exacerbation of the psychosis.

Follow-up details were available in all of the patients,for a period of three months to eight years after the firstpost-ictal psychotic episode. Four patients died: one froman astrocytoma, two from complications of epilepticseizures, and one was found drowned in a river (the causewas unknown). Two patients developed a chronic psychosis,in each eight years after the first post-ictal episode, anda further patient started to experience fleeting inter-ictalauditory hallucinations six months following the onlyepisode of post-ictal psychosis she had experienced.

ifiustrative case histories

Case 1 (third case in Table 1)

MB, a 33-year-old farm worker, first developed complexpartial seizures with intermittent secondary generalisationat the age of 12.At 25 the patient experienced the first postictal psychotic episode, the symptoms manifesting themselves 48 hours after a major seizure. At the same time,the seizure frequency increased and the aura became morecomplex. Investigations yielded evidence of a right occipitalpole astrocytoma.

During the remaining ten years of the patient's life,post-ictal psychotic episodes occurred at ever-decreasing

intervals, while the duration of the episodes progressivelyincreased from 24 hours at the onset of the disorder to apoint when, shortly before death, chronic psychosis ensued.During this period the lucid interval diminished from 24hours to 6 hours. At the beginning, psychosis appeared

250 LOOSDAIL AND TOONE

only after major seizures; towards the end it followed uponpartial seizures.

A full mental state examination was carried out at a pointwherechronic psychosis was about to supervene. Primarydelusions, auditory hallucinations, thought disorder andideas of reference were noted. The patient's originaldelusional idea, that he had been in communication withthe angelsviaa specialpentheyhad givenhim, waspresent.He now believed that he had been responsible for WorldWar III and had destroyed the world, and that the toothfairies were using his brain to smuggle drugs.

He was treated with major tranquillisers;these reducedthe intensity of the psychotic phenomena but did notcompletely dispel them. Radiotherapy was given a fewmonths prior to his death, but this had no marked effect onhis psychotic phenomena, which had become continuous.

Case 2 (seventh case in Table 1)

PB is a housewife who suffered blank spells since childhood.Duringher teenageyears, thesebecameassociatedwithanunpleasant abdominal feeling and a sensation of fear. Theattacks occurred approximately weekly and led to tonicclonic seizures, 2-3/annum. Despite careful treatment thesepartial complex seizures with intermittent secondarygeneralisation increased in frequency, and her first psychosisoccurred after a run of generalised seizures.

Over the ensuing six years, seizures have occurred withgradually increasingfrequency: she experiencesmajor attacksapproximately monthly and minor episodes weekly. Psychoseshavebeencorrespondinglymorefrequent,nowapproximatelymonthly, and are occasionally precipitated by a run ofpartial complex seizures without secondary generalisation.

Post-ictal psychotic features conformed to a consistentpattern. One day after the seizure she would developelevated mood, become overtalkative and suffer persecutoryideas if prevented from undertaking irrational, grandioseprojects. Visual hallucinations were present from thebeginning; auditory hallucinations developed as the episodesbecame more frequent. The content of the psychosis varies;on one occasionshe wanted to fly to Scotland on a setteeto visit a friend, despiteacknowledgingthat she had dieda year previously.On another occasion,whenpreoccupiedwith religious ideas, she said she could purify her sister,that the neighbours were evil and that her husband andbrotherwereagainsther. The prescriptionof lithium(levelsof 0.5—0.9mmol/1) reduced the affectiveelement of thepsychosis. Major tranquilhisers have been given in an effortto avoid admission to hospital. As a result of herintermittently disturbed mental state and despite thefact that she returns to normal between episodes, heremployment has been terminated and her marriage is nowunder considerable strain.

Sheiscurrentlyundergoingevaluationfor a neurosurgicalprocedure in an effort to help control her seizures and theirsequelae.

DiscussionPost-ictal psychosis is a relatively rare condition: acomputerised search of the medical records of the

Maudsley Hospital showed a referral ratio of 12:1chronic or inter-ictal psychosis to post-ictal psychosis.As the condition is transient and presents acutely,the number may be unrepresentative, as some areprobably admitted to their local psychiatric hospital,or perhaps not admitted to hospital at all. In viewof the small size of the group reported here, and thepossibility that teaching hospital referrals may beunrepresentative, any conclusions made are tentative.

Three features of the methodology merit furtheramplification. The diagnostic operational criteriarequired that the onset of psychosis should occurwithin a week of return of normal function after aseizure, and should endure for at least 24 hours butfor no longer than three months. Little is known ofthe time course of the post-ictal psychoses. Thecriterion suggested, therefore, is to a degree arbitrary,and may subsequently need to be modified. However,all of our cases fulfilled this criterion, and we didnotencountera furthercasethatsatisfiedtheotheroperational criteria but fell outside this time-scale.A second point concerns the exclusion of patientswho had previously suffered an inter-ictal psychosis.This criterion was used to prevent possible diagnosticuncertainty in the post-ictal psychotic group, although

no patients had to be excluded on this basis. Thethird point concerned the use of the PSE. This wasnot designed for use in organically determinedabnormal mental states, and the Catego classificationpresented here must be viewed with some caution(Wing, 1983). Nevertheless, the PSE and its derivativethe Syndrome Check List are useful instruments forsystematic collection of clinical data, and for thisreason alone seem appropriate for a study of thiskind.

The principal findings are as follows. The postictal psychoses are defined by their temporalrelationship to seizure activity. Psychosis usuallyoccurred after an exacerbation in seizure frequencyor intensity and appeared after a lucid interval.The episodes were characterised by pleomorphicpsychotic phenomena occurring either in clear or inclouded consciousness and dominated by markedand varied mood changes. Spontaneous resolutionwas usual, but with a tendency to recur.

We were unable to confirm certain previouslyreported observations. Post-ictal psychotic states aresaid usually to occur against a background ofclouded consciousness (Lishman, 1978), but only fiveof 14 cases reported here provided evidence ofconfusion sustained throughout the episode, whilethree more showed no confusional features at anytime. Perhaps repeated testing of patients previouslyreported would have identified a proportion whoresumed clear consciousness while still psychotic, as

POST-ICTAL PSYCHOSES 251

occurred in four of this series. The presence orabsence of confusion was confirmed during mentalstate examination by one of the authors in five ofthe cases. The medical notes confirmed that thissymptom was repeatedly sought in each of theremaining nine, and verbatim responses to questionsof orientation were recorded. Organically determinedstates may occur in the absence of confusionaccording to Cutting (1980), who identified 74patients with psychosis and cerebral dysfunction:38@Vowere in clear consciousness at the time ofassessment.

Previously reported work (Dongier, 1959) hasobserved an association between post-ictal psychosisand generalised epilepsy. Although the majority ofpatients in this series experienced generalised seizures,in most cases this was due to secondary generalisationfrom complex partial seizures. It has also beensuggested that post-ictal psychotic episodes mayremit spontaneously following the occurrence offurther seizure activity, and therefore ECT could bebeneficial in refractory states (Toone, 1981). In thisstudy, further seizure activity was associated with anexacerbation of psychotic symptoms in five out ofsix subjects.

Other clinical observations seem worthy ofcomment. Stuporose states are said to occur postictally: three patients presented in such a way. Onewas mute and catatonic, with waxy flexibility;intensive nursing care, including tube feeding, wasnecessary. Another, also admitted mute and noteating or drinking, recovered in three weeks. Hevoiced depressive ideas when he started talking andhad a depressed affect. The third case of stupor lastedonly a matter of days. Diagnostic phenomenologywas never elicited.

All commonly used anticonvulsant drugs had beengiven to the patient group: no particular drug couldbe inculpated. Three of the seven patients had reducedserum folate levels. Reynolds (1967) postulated thatfolate deficiency might contribute to the developmentof a psychotic state in epileptic patients. Unfortunately, red cell folate assay, a more stableestimate of folate availability, was not carried out.However, the patient with the lowest level, 2.0 g@g/l,had previously recorded a level of 1.1 gtg/l when notpsychotic (normal range 3—15gig/l). The role offolate deficiency in the generation of post-ictalpsychoses therefore remains unclear.

The distinction between inter-ictal and postictal psychosis is not usually problematical. Brief,spontaneously remitting psychotic episodes occurringin epileptic patients have been described by Landolt(1958), but these are exceedingly rare, and the

relationship of psychotic episodes to preceding

seizure activity distinguishes such episodes from postictal psychoses. The course of the chronic, inter-ictalpsychosis is probably not greatly influenced byseizure activity, and spontaneous remission isunusual once the disorder is well established. In onlytwo patients was the onset of psychosis morethan 48 hours after the return of normal cerebralfunctioning following a transient, marked increasein seizure frequency. One had a lucid interval of sixdays (during which time he was reported to haveincreasingly odd behaviour), but his usual fitfrequency was one seizure every month. Moreover,he suffered a seizure while still psychotic, markedlyworsening his mental state, which ultimately recoveredafter three months. The other patient had a lucidinterval of three days. Her usual fit frequency wasone seizure every three months.

Post-ictal psychosis should be borne in mind asan unusual clinical entity that may occasionally bemisdiagnosed and treated inappropriately. It is alsoof theoretical importance as it provides the possibilityof an organic model for psychoses in general and forthe inter-ictal psychoses in particular. A closerfamiliarity with its mode of presentation, phenomenology and clinical course may thus provide clues toaetiology of a far commoner group of disorders. Itis admittedly not an entirely adequate model.First-rank Schniederian experiences are relativelyuncommon, and many of the episodes tend towardsphenomenological polymorphism rather than presentwith discrete, easily recogmsable schizophrenic ormanic—depressivesymptoms. Even so, mental statephenomena occur that are identical to those whichare found in, and are diagnostic of, the functionalpsychoses, and they occur frequently in clear consciousness. The post-ictal and inter-ictal psychosesalso have points in common: the lack of familyhistory of psychosis, the over-representation ofpartial complex seizures, the similar age of onset ofboth epilepsy and psychosis, and the observationthat, in a minority of cases, post-ictal psychoses willprogress to chronic inter-ictal states.

The observation that psychosis may occur morecommonly in patients with epilepsy, especiallytemporal lobe epilepsy, and that it usually develops10—15years after epilepsy onset (Slater et a!, 1963),has led some authorities to propose that a disturbancein limbic function may play a crucial role in theevolution of the functional psychoses in general(Bogerts et a!, 1985), and particularly in the interictal psychoses of epilepsy (Torrey & Peterson, 1974).A similar explanatory hypothesis may be advanced toaccount for the post-ictal psychotic syndrome. Theincrease in temporal lobe spike/sharp wave activityis certainly consistent with an active process rather

252 LOGSDAIL AND TOONE

than neuronal exhaustion. An increase in slow waveactivity, noted in the same number of patients, wasassociated with confusion, and this pattern may havemore in common with the usually more brief periodsof confusion occurring commonly after majorgeneralised seizures.

Such a hypothesis raises as many questions as itanswers. It is not at all clear why some form of asyet unspecified limbic malfunction should in oneinstance manifest itself as a chronic inter-ictalpsychosis and in another as a post-ictal psychosis,nor how the latter may in some cases evolve towardsthe chronic condition. A lucid interval preceding theonset of psychosis was observed in 12of the 14cases,but its significance remains uncertain. Little is knownof the effect of spontaneously occurring seizureactivity in man on cerebral monoamine systems, butexperimental seizures induced in animals may resultin increased post-synaptic dopamine sensitivity(Modigh, 1975). It has not yet been possible toreplicate these results in man. Whatever the mechanism,the presence of a delay in the onset of post-ictalpsychosis does suggest the possibility of an indirecteffect involving one or more neurotransmitter systems.

Acknowledgements

We wish to thank colleagues who have contributed patients to thisseries, and in particular to Dr Ron, Dr Fenwickand ProfessorLishman,whoalsoadvisedon themanuscript.MrsGuiseprovidedinvaluable secretarial assistance.

References

Barrs, 1. A. (1974)A follow-upof a cohort of Englishpatientswith epilepsy admitted to psychiatric care in an English city. InEpilepsy:Proceedingsof theHansBergerCentenarySymposium(eds P. Harris & C. Mawdsley).Edinburgh: Churchill Livingstone.

BOGERTS, B., Msaaiz, E. & SCHONFELDT-BAUSCH, R. (1985) Basal

ganglia and limbic system pathology in schizophrenia. Archivesof GeneralPsychiatry,42, 784—791.

Ci@a@,R. A. & LESKO,J. M. (1939)Psychosesassociatedwithepilepsy. American Journal of Psychiatry, 96, 595-607.

CUTrING, J. (1980) Physical illness and psychosis. British JournalofPsychiatry,136,109—119.

DoNoLER,S. (1959)StatiSticalstudy of clinical and electroencephalographic manifestations of 536 psychotic episodes occurring in516 epileptics between clinical seizures. Epilepsia, 1, 117—142.

EsQulitol,E. (1838)MaladiesMentales.Paris: J.-B. Balliere.GASTAUT,H. (1970) Clinical and electroencephalographical classifi

cation of epileptic seizures. Epilepsia, 11, 102—113.Gu@, J. & Fr.r.rroN,G. W. (1969) Epilepsy in prisons: a diagnostic

survey.BritishMedicalJournal, iv, 326—328.JAcKsoN, J. H. (1875) On temporary mental disorders after

epileptic paroxysms. West Riding Lunatic Asylum MedicalReports,5,105—129.

LANDOLT, H. (1958) Serial electroencephalographic investigations

during psychotic episodes in epileptic patients and duringschizophrenic attacks. In Lectures in Epilepsy (ed. A. M. Lorentzde Haas). Amsterdam:Elsevier.

LEvIN, S. (1952) Epileptic clouded states. Journal of Nervous andMenial Disease,116, 214—225.

LISHMAN, W. A. (1978) Organic Psychiatry. Oxford: Blackwell

Scientific Publications.MODIGH, K. (1975) Electroconvulsive shock and postsynaptic

catacholamine effects; increased psychomotor stimulant actionof apomorphine and clonidine in reserpine pretreated mice byECS.Journalof NeuralTransmission,36, 19-32.

PEREZ, M. M. & TRIMBLE, M. R. (1980) Epileptic psychosis:diagnostic comparison with process schizophrenia. BritishJournalofPsychiatry,137,245-249.

REYNOLDS, E. H. (1967) Schizophrenia-like psychoses of epilepsy

anddisturbancesoffolateandB12metabolism.BritishJournalof Psychiatry, 113, 911—919.

SLATER, E., BEARD, A. W. & CLITHEROE, E. (1963) Theschizophrenia-likepsychosesof epilepsy.British Journal ofPsychiatry, 109, 95—105.

TOONE, B. K. (1981)Epilepsy and Psychiatry (eds E. H. Reynolds

& M. R. Trimble).London: ChurchillLivingstone.—¿�,GARRALDA,E.&RoN,M.A.(1982)Thepsychosesofepilepsyand the functionalpsychoses:a clinicaland phenomenologicalcomparison.BritishJournal of Psychiatry,141,256-261.

TORREY,E. F. & PETERSON,M. R. (1974) Schizophrenia and thelimbicsystem.TheLancet, ii, 942-946.

WING, J. K. (1983) Use and misuse of the PSE. British Journal ofPsychiatry,143,111-117.

—¿�, CoopER, J. E. & SARTORIUS, N. (1974) The Measurement and

Classjficaiion of Psychiatric Symptoms. London: CambridgeUniversityPress.

WoRLD HEALm ORGANiZATION(1978) Mental Disorders: Glossaryand Guide to their Classification in Accordance with the NinthRevision of the International Classification of Diseases. Geneva:WHO.

* S. J. Logsdail, MRCP, MRCPsych, Research Senior Registrar, Department of Psychiatry, The National

Hospital for Nervous Diseases; B. K. Toone, MRCP,MPhil,MRCPsych,Consultant Psychiatrist, Kings CollegeHospital, London

5Correspondence: Queen Square, London WCJN 3BG