E

R

DC

B

DQ

RA

0d

pilepsy Research (2009) 85, 172—186

journa l homepage: www.e lsev ier .com/ locate /ep i lepsyres

EVIEW

elusions, illusions and hallucinations in epilepsy: 2.omplex phenomena and psychosis

rent Elliott, Eileen Joyce, Simon Shorvon ∗

epartment of Clinical and Experimental Epilepsy, National Hospital for Neurology and Neurosurgery,ueen Square, London WC1N 3BG, United Kingdom

eceived 31 October 2008; received in revised form 8 March 2009; accepted 15 March 2009vailable online 12 May 2009

KEYWORDSEpilepsy;Psychosis;Postictal psychosis;Interictal psychosis;Schizophrenia

Summary In this second paper the clinical features and electrophysiological underpinningsof more complex psychotic states associated with epilepsy are reviewed. (a) Complex par-tial status epilepticus, in particular of temporal lobe origin, may result in mental statesremarkably similar to those seen in the primary psychoses. This non-convulsive state is associ-ated with prolonged epileptic discharges on intracranial stereoelectroencephalography (SEEG)in hippocampal and other mesial temporal structures, sometimes without abnormalities onthe scalp EEG. Where hallucinatory or psychotic symptomatology does occurs, it can beconsidered an examples of an ictal psychosis. The phenomenology and electrophysiologicalfeatures of this condition are reviewed. (b) Postictal psychosis is noted for its similarity toschizophrenia-like/paranoid and affective psychoses and there is convincing SEEG evidencethat, for some cases at least, the psychosis is not in fact postictal but rather an ictal psy-chosis due to ongoing limbic seizure activity and a form of non-convulsive status epilepticus.It has been suggested that postictal psychosis should be divided into two sub-groups: theclassical ‘nuclear’ postictal type and an atypical periictal type. (c) Interictal hallucinosis inepilepsy has been poorly studied, but is probably commoner than appreciated. To what extentit represents subclinical epileptic discharges (i.e. auras) is not known. It may interestinglyalso be associated with abnormal affective states in epilepsy. (d) The interictal psychosis ofepilepsy is often indistinguishable from primary schizophrenia. It occurs more commonly intemporal lobe (limbic) epilepsy, in those with frequent seizures and only in patients with along history of epilepsy (usually over 10 years). There is convincing SEEG evidence of fre-

quent, semi-continuous and sometimes continuous epileptic activity in limbic structures at thetime of psychotic and hallucinatory ideation and behaviour, suggesting that in some cases atleast, the epileptic activity is the cause of the symptoms. Whether the psychosis is directly‘driven’ by subclinical electrographic activity or is indirectly a consequence of function changeinduced by such activity is not clear. An intriguing question also arises as to whether sim-ilar electrophysiological changes could underpin psychosis in patients without epilepsy butevidence on this point is sparse. The effects of temporal lobe surgery on the psychoses of

∗ Corresponding author. Tel.: +44 0207 837 3611.E-mail address: s.shorvon@ion.ucl.ac.uk (S. Shorvon).

920-1211/$ — see front matter © 2009 Elsevier B.V. All rights reserved.oi:10.1016/j.eplepsyres.2009.03.017

Delusions, illusions and hallucinations in epilepsy 173

epilepsy are described and these might throw light on the mechanisms of epileptic psychosis.

The principles of pharmacological therapy of epileptic hallucinosis and psychosis are outlined.© 2009 Elsevier B.V. All rights reserved.

Contents

1. Complex hallucinatory/psychotic states................................................................................ 1732. Ictal psychosis: complex partial status epilepticus ..................................................................... 1733. Postictal psychosis ..................................................................................................... 1754. Interictal psychosis in people with epilepsy ............................................................................ 177

4.1. Definition ....................................................................................................... 1774.2. Prevalence ...................................................................................................... 1774.3. Relationship to epilepsy subtype ................................................................................ 1774.4. Interictal hallucinosis in patients with epilepsy.................................................................. 1784.5. Psychopathology in interictal psychosis: comparison with primary psychoses .................................... 1784.6. SEEG studies .................................................................................................... 1794.7. Neuropathology ................................................................................................. 1804.8. Structural magnetic resonance imaging ......................................................................... 1804.9. Functional brain imaging ........................................................................................ 1804.10. Neurogenetics.................................................................................................. 180

5. Comparison with electrophysiological, neuropathological, structural and functional imaging findings in schizophrenia . 1815.1. EEG studies ..................................................................................................... 1815.2. Neuropathology ................................................................................................. 1815.3. Structural and functional brain imaging ......................................................................... 181

6. The effect of temporal lobe surgery on the psychoses of epilepsy...................................................... 1817. Therapy of hallucinations and psychosis in epilepsy .................................................................... 1828. Discussion.............................................................................................................. 183

References ............................................................................................................. 184

1. Complex hallucinatory/psychotic states

It is clear that ‘experiential’ phenomena and many of theelementary hallucinatory phenomena described in paper 1differ markedly from the complex psychotic symptoms seenin the functional psychoses or in the psychoses of epilepsy. Itis also clear however that some complex psychotic states canoccur as a consequence of an epileptic discharge. The bestevidence for this can be found in stereoelectroencephalog-raphy (SEEG) recordings during episodes of complex partialstatus epilepticus and during certain cases of apparent ‘pos-tictal’ and possibly ‘interictal’ psychosis, the results of thesestudies are summarised in Table 1. Any review of this litera-ture relies on a limited number of individual case reports andpublication bias may be a significant problem. However, theliterature in this area comprises case reports only, and herewe have attempted to choose illustrative and representativecases.

The psychoses of epilepsy are usually categorised accord-ing to their temporal relationship to seizures or to clustersof seizures. One useful approach is to divide these statesinto periictal (preictal, ictal, postictal) and interictal cate-gories. Below we consider the ictal, postictal and interictalpsychosis of epilepsy.

of the early literature was confined to tonic-clonic statusepilepticus (Shorvon, 1994). The term ‘complex partial sta-tus epilepticus’ was coined by Gastaut in 1956 and was animportant step in appreciating that a range of phenom-ena, including hallucinations, could be caused by prolongedseizure activity (Gastaut et al., 1956; Shorvon, 1994). Thisculminated in the Xth Marseilles Colloquium of 1962 inwhich various forms of convulsive and non-convulsive statusepilepticus was formally defined (Gastaut et al., 1967) andsince then a number of subclassfications of complex partialand non-convulsive status epilepticus have been proposed(Shorvon, 2007).

Symptomatology is pleomorphic but major featuresinclude confusion and inaccessibility. Psychiatric featuresinclude delusions, illusions, visual and auditory halluci-nations (which may have a religious content), ideas ofreference, paranoia and thought disorder. There is often acharacteristic perseverative obsessive insistence on oppo-sitions (such as black/white, good/bad, right/left). Thehallucinatory experiences often have religious content. Theattacks typically last hours, but can be much more prolonged(occasionally continuing for months or years) and the symp-toms frequently fluctuate or ‘cycle’ (Shorvon, 1994). Thetypical attack is characterised by several hours of fluctu-ating restless fearful and agitated behaviour with delusions

2. Ictal psychosis: complex partial statusepilepticus

The term status epilepticus (état de mal épileptique) firstappeared in French in Calmeil’s thesis of 1824, althoughmost

a

cpm

nd hallucinations.An interesting comparison between absence status (32

ases), psychomotor status (complex partial status of tem-oral lobe; 9 cases) and frontal polar status (19 cases) wasade by Rohr-Le Floch et al. (1988). Symptoms which were

174 B. Elliott et al.

Table 1 Summary of EEG data correlating ‘limbic’ epileptiform activity with complex psychotic states.

Study EEG Electrophysiology Psychopathology

Wieser et al. (1985) SEEG Temporal lobe status epilepticus. Stickiness, aggressivity, dysphoriaand depression

Trimble (1991) Sphenoidal EEG Frequent sharp waves with phasereversals on the right.

Paranoid schizophreniform psychosis

Takeda et al. (2001) SEEG Left amygdala discharges becomingalmost continuous.

Internal dialogue with the voices ofher parents, restless, anxious andfearful

Kanemoto (1997) SEEG Clear cut epileptiform discharge inleft amygdalo-hippocampal region.

Capgras syndrome

Kristensen andSindrup (1978)

Spehnoidal EEG Temporal medio-basal spike foci.More frequent and more likely

in no

Paranoid/hallucinatory interictalpsychosis

mpifios

wrstIfsts(

aOttpcirabehw

Fb

bilateral thancontrols.

uch more frequent in psychomotor status were: ideas ofersecution (30%), anxiety/fear (44%), negativism (22%),rritability or aggression (56%). This was in contrast to therontal polar cases and absence cases who typically exhib-ted indifference, perplexity, mutism or poverty or slownessf speech. The frontal polar cases also often smiled oreemed hilarious during the episodes and confabulated.

The first SEEG case was described by Wieser (1980) inhich the clinical and SEEG features were meticulously cor-

elated, and it is worth describing this case in detail. Ithould also be noted that the scalp EEG was ‘uncharac-eristic’ although at times desynchronisation was noted.n contrast the SEEG showed impressive changes—–withrequent focal seizures in the right hippocampal region

ubsequently replaced by continuous discharges. Spread tohe lateral neocortex was associated with a dream-liketate and to Herschl’s gyrus with musical hallucinationsFig. 1).

csbr

igure 1 Discharge starting in the gyrus of Heschl and remainingy a hallucination of music (Wieser, 1980).

n-psychotic

During electrographic status epilepticus, the patient wasnxious, tense and experienced hallucinations of music.thers have reported similarly detailed studies in whichhe SEEG showed various patterns including high-frequencyonic spike discharges, clonic or clonic-tonic dischargeatterns. As Weiser has pointed out, the EEG in areaslose to the discharging focus may show attenuation (‘crit-cal aplattisement’) and the surface EEG may show aegional or even generalised attenuation with disappear-nce of interictal spikes described as ‘forced normalization’y Landolt (Wieser, 1998). It is clear that prolongedpileptiform EEG discharges (characteristic of status) inippocampal and amygdaloid regions can be associatedith behavioural abnormalities without clear cut scalp EEG

hanges. Wieser described two other cases of temporal lobetatus epilepticus in which the patients exhibited a markedehavioural syndrome with stickiness, aggressivity, dyspho-ia and depression (Weiser et al., l985).

there for several hours during which time it was accompanied

Delusions, illusions and hallucinations in epilepsy

Figure 2 A recording from a patient in whom the symp-

aa

atmoottwaacpsctfvptiramif

eac

wcmfiass

ittiHm

atwdpwiethe EEG showed spike activities in the (predominantly) left

tomatology of non-convulsive status epilepticus resembledschizophrenia. The top four leads are from the right and threeand four represent sphenoidal recordings (Trimble, 1991).

Psychotic symptomatology is common in complex par-tial status, and occurs in more or less florid forms. Trimble(1991) provides a case of non-convulsive status producinga state resembling schizophrenia in a 22-year-old man withcomplex partial and generalised tonic-clonic seizures sincethe age of 3.5 years. He complained of the sudden onsetfeeling that rays were being passed through his body to ster-ilise him as well as voices in the second and third personscriticising him. His behaviour varied from torpid and hardlyresponsive to periods of restlessness. He was admitted toa psychiatric hospital and diagnosed with schizophrenia. AnEEG, which included sphenoidal electrodes was carried outand is shown below (Fig. 2). The main features includedfrequent sharp waves on the right side with phase rever-sals in the right sphenoidal leads which occurred during aperiod of florid paranoid psychosis. The mental state andEEG both demonstrated a quick response to intravenousdiazepam.

3. Postictal psychosis

The concept of postictal psychosis can be dated back tothe work of Jackson (1875) on ‘temporary mental disor-ders after epileptic paroxysms.’ It did not however becomethe focus of attention until much later (Logsdail and Toone,1988; Kanner et al., 1996), despite being the most frequentpsychotic condition observed in clinical practice, account-ing for as much as 25% of cases of psychosis in one earlyepilepsy series (Dongier, 1959). Despite its frequency therecontinues to be no universally agreed definition, althoughthat used by Kanemoto et al. (2001), including any psychoticepisode occurring within seven days of the last generalisedtonic-clonic seizure or cluster of complex partial seizures,is useful.

Common findings among the different case series weresummarised by Kanner and Barry (2001), and include: (1)delay between the onset of psychiatric symptoms and thetime of last seizure; (2) relatively short duration of psy-chosis; (3) affect-laden symptomatology; (4) the clusteringof symptoms into delusional and affective-like psychosis; (5)

an increase in the frequency of secondary generalised tonic-clonic seizures preceding the onset of postictal psychosis;(6) the onset of postictal psychosis after a long duration ofepilepsy (for a mean period of more than 10 years); and (7)

m

cs

175

prompt response to low dose antipsychotics or benzodi-zepines.

The phenomenology of postictal psychosis has beenddressed by a number of studies. One early review washat of Levin (1952) who investigated 52 cases noting thatost episodes occurred in patients between 30 and 40 years

f age, hallucinations (mainly auditory) occurred in aboutne-third and delusions (mainly persecutory) in one quar-er. In two-thirds the psychotic episode was preceded bywo or more seizures and developed, in three quarters,ithin 24 h of the last seizure. In a later series by Logsdailnd Toone (1988) the lucid interval lasted between onend six days (mean 2.5) in 78%, with episodes following alear increase in generalised seizure frequency in 86%. Theiratients fulfilled Present State Examination (PSE) criteria forchizophrenic psychosis in four, manic/mixed affective psy-hoses in three, and paranoid psychoses in four. They foundhe psychopathology to be identical to that found in theunctional psychoses. Mood was markedly abnormal (ele-ated, depressed or both) in three quarters, one half hadaranoid delusions and both auditory and visual hallucina-ions were common. Barczak et al. (1988) and Byrne (1988)n their series also note the prevalence of hypomania andeligiosity following temporal lobe seizures. Kanemoto etl. (1996) suggest that patients with postictal psychosis areore likely to experience grandiose and religious delusions

n the presence of elevated mood and a feeling of mysticusion of the body with the universe.

It is possible however that at least some cases of appar-nt ‘postictal psychosis’ are in fact due to ongoing seizurectivity and should therefore be categorised as cases of non-onvulsive status epilepticus.

Takeda et al. (2001) provide data on a 25-year-old womanith intractable temporal lobe epilepsy. Two days after aluster of 18 seizures she was elated, engaged in an internalonologue and could hear the voices of her parents. The

ollowing day she was mute and stuporose before graduallymproving over the next two days. She then became restless,nxious and fearful for a further 24 h before the psychiatricymptoms disappeared. A schematic representation of hereizures and their relation to the psychosis is given in Fig. 3.

It is interesting to note that in this example dischargesn the amygdala are not elevated for the entire duration ofhe psychotic episode but that the psychosis can persist inheir absence, a point also highlighted by Kanemoto (1997)n their study of a postictal Capgras syndrome (see below).er EEG before the seizure cluster showed spike activityainly in the left medial temporal lobe (Fig. 4).After the cluster, seizure discharges restricted to the left

mygdala appeared and increased in frequency and dura-ion. During this period she felt estranged from the outsideorld and had auditory hallucinations. The left amygdalaischarges became almost continuous when she was stu-orose (Fig. 5). Rhythmic spike and slow wave dischargesere observed towards the end of the psychotic episode

n the left amygdala and coincided with feelings of anxi-ty, restlessness and fear. When the psychosis had improved

edial temporal lobe, as before the cluster.A second ‘positive’ study by Kanemoto (1997), reports the

ase of a 24-year-old woman with complex partial seizuresince the age of four, who developed a periictal Capgras syn-

176 B. Elliott et al.

Figure 3 Schematic representation of seizures occurring during SEEG recording, psychotic symptoms, seizure discharges in theleft amygdala (duration of seizure discharges in every four hours in percentages), and the course of medication. CPS, complex partialseizure; GTC, secondarily generalised tonic-clonic seizure; CBZ, carbamazepine; PB, phenobarbital; NZP, nitrazepam (Takeda et al.,2001).

drome (Capgras and Reboul-Lachaux, 1923) following boutsof ictal fear. Capgras syndrome is the delusion that a per-son close to the patient has been replaced by one or moreimpostors. For two years prior to a neurosurgical assess-ment for intractable seizures she experienced episodes ofictal fear, i.e., complex partial seizures which occasionallyled to brief psychotic episodes in which she felt that closefriends or relatives had become entirely different persons.This could last a matter of hours or weeks. During the 12-dayexamination which included intracranial EEG, one episode of

Figure 4 Intracranial EEG recording prior to the psychoticepisode. Note interictal discharges in the medial structure ofthe left temporal lobe (Takeda et al., 2001).

postictal Capgras syndrome occurred in association with 15isolated ictal fears and 2 complex partial seizures. The psy-chotic episode, in which she believed her best friends whowere just visiting had assumed totally different personali-ties, lasted several hours. She was orientated in time, placeand person and there was no clouding of consciousness. All17 seizures showed clear cut epileptiform discharge in theleft amygdalo-hippocampal region (Fig. 6).

Of interest is the observation that she described her ictalfear as the feeling that someone was standing behind her.Similar findings have been reported by McLean (1952) and

Figure 5 Intracranial EEG recording during the psychoticepisode with frequent seizure discharges restricted to the leftamygdale (Takeda et al., 2001).

Delusions, illusions and hallucinations in epilepsy 177

Figure 6 Depth EEG of ictal fear in the course of Capgras syndrome. L1—15 depth electrodes inserted along the longitudinalaxis of the left hippocampus (1, most anterior; 15 most posterior). Subdural electrodes placed over anterior third of left middle

cleuS1—

4

4

Atopsnd

4

RwscwriratEl(wb1

4

temporal gyrus: 1 most anterior; 4 most posterior. L1—3: left nugyrus; R1—3: right nucleus amygdala; R4—6: right hippocampus;diamond) (Kanemoto, 1997).

Williams (1959), with Gloor et al. (1982) describing thissensation as ‘the feeling of somebody being nearby’. Moreimportantly perhaps, Kanemoto noted that the episodes ofictal fear ended precisely with the disappearance of the cor-responding ictal EEG discharge. When discharges occurredin rapid succession some abnormal mental state persistedbeyond the scope of the ictal EEG discharge, but it was onlywhen the interval between successive ictal fears was lessthan one hour that an overt psychotic reaction occurred.

It is also clear that some episodes of postictal psychosisappear at least, to be unrelated to ongoing epilepsy. Wewere able to identify two SEEG studies which have beenreported as ‘negative’ (So et al., 1990; Mathern et al.,1995). However, given the issues of sampling bias and ‘crit-ical applatisement’, it remains possible that even in thesecases, focal seizure activity was present. Nevertheless, nei-ther of these two psychotic states can be attributed, on theevidence provided, to anything but an indirect after effectof the seizure clusters.

An attempt by Kanemoto’s group has recently been madeto clarify this issue (Oshima et al., 2006). In their prospec-tive study of 108 patients with temporal lobe epilepsy, sixexperienced postictal psychotic episodes. Three of thesepatients demonstrated a lucid interval between the seizurecluster and the psychotic episode, relatively infrequentpsychotic episodes (once only, twice and yearly) and ini-tial brief manic states occurred in two. The other threepatients lacked a lucid interval and in this group psy-chotic episodes occurred more frequently (>1 per month),additional seizures occurred in the midst of the psychoticepisodes and initial manic states were not seen. Theyconclude by suggesting that postictal psychosis should besubdivided into two types: first, a nuclear type representing

the established clinical picture as described by Kanner (seeearlier) and occurring as an indirect after effect of seizureactivity, and second, an atypical periictal type, occurring asa direct manifestation of limbic epileptic discharges, thatwe may speculate could be due to complex partial status.

Tatd

s amygdala; L4—8: left hippocampus; L9—10: parahippocampal4 anterior left middle temporal gyrus. Onset of ictal fear (solid

. Interictal psychosis in people with epilepsy

.1. Definition

s Sachdev (1998) has pointed out in his review on epilep-ic psychoses most studies have lacked a precise definitionf either ‘psychosis’ or ‘epilepsy’. Definitions for interictalsychosis need to exclude periictal psychosis, allow for theubcategories of drug induced and the possibility of ‘alter-ative’ psychosis, and be present in the absence of seizureischarges.

.2. Prevalence

eported prevalence rates for psychosis in epilepsy varyidely. Dongier (1959), in what was one of the earliest

tudies, reviewed 516 epileptic patients with 536 psy-hotic episodes and found rates of 40.2% and 44% for thoseith ‘centrencephalic’ epilepsy and ‘psychomotor’ epilepsy,

espectively. These seem rather high compared with laternvestigations, with for example Pond (1974), reportingates of 7% and Mendez et al. (1993) of 9.25%. Sengoku etl. (1983) initially found the rate of psychosis in epilepsyo be much lower (4.3% and 6% in Idiopathic Generalisedpilepsy (IGE) and TLE groups, respectively) although in aater review (Sengoku et al., 1997) more intermediate rates19.4% and 15.2% in the IGE and TLE groups, respectively)ere reported and which correspond well with recent worky Filho et al. (2008) who reported psychotic disorders in5.8% of 85 patients with TLE+ mesial temporal sclerosis.

.3. Relationship to epilepsy subtype

hroughout the 20th century, there have been numerousttempts made to assign different psychiatric features tohe different types of epilepsy. Kraeplin (1913), for instance,istinguished between patients with what he called reac-

178 B. Elliott et al.

Table 2 Occurrence of hallucinations, illusions and other psychosensory symptoms. From Silberman et al. (1994).

Symptom type Ictal occurrence (%) Interictal occurrence (%)

Hallucinations (visual, auditory, olfactory, gustatory, vestibular) 71 43Sensory illusions and distortions (visual, auditory, olfactory,

gustatory, tactile, proprioceptive, autoscopic, apperceptive)33 43

Cognitive illusions (déjà vu, jamais vu, illusions of significance) 24 76Paroxysmal affect (fear, sadness, rage, euphoria, sexual) 14 52Time distortion 5 0

tiaamtwI

(apeRd(tabwobcbibwlcrffihifca

4e

Ipbtt

ppssis(nd(

cotsctsnfitdis(stscmn

4c

Iwt‘tn‘

Thought and speech disturbanceMotor automatisms (simple and complex involuntary acts)

ive epilepsy, whom he described as restless and occasionallympulsive with alternating periods of stupor and excitementnd those with genuine epilepsy who were described as dullnd circumlocutory with hallucinations and delusions. In aore recent study, Sengoku et al. (1997) found hallucina-

ions, delusions and irritability to be more common in TLE,hilst perplexity and impairment of attention prevailed in

GE.Several studies have not found significant differences

e.g., Dongier, 1959) although classification was not inccord with modern schema. More recent attempts to relatesychopathology to epilepsy localisation have also beenquivocal. Adachi et al. (2000) used the Brief Psychiatricating Scale to evaluate psychiatric symptoms in patientsivided into the following groups: (1) frontal lobe epilepsyFLE) with interictal psychosis (n = 8); (2) FLE with postic-al psychosis (n = 3), (3) TLE with interictal psychosis (n = 29)nd (4) TLE with postictal psychosis (n = 8). Whilst the num-ers were small the only significant finding was that patientsith FLE and interictal psychosis scored more highly on twof four ‘hebephrenic’ features (emotional withdrawal andlunted affect: NB this contrasts with the psychiatric con-ept of hebephrenia in which affect is preserved). Paranoidehaviour showed only a trend towards increased frequencyn the TLE groups. Finally, in one of the largest population-ased studies to date, Qin et al. (2005) found that peopleith a history of epilepsy were two to three times more

ikely to develop schizophrenia or schizophrenia-like psy-hosis than people who had never had epilepsy. Significantisk factors included a later age of onset of epilepsy and aamily history of psychosis, but again they did not find a dif-erence according to the type of epilepsy. On the face oft these results are rather surprising given the quite widelyeld clinical view that interictal psychosis is more commonn patients with TLE. It is difficult to draw firm conclusionsrom these studies except to reiterate the fact that psy-hosis can occur with all types of epilepsy, both generalisednd partial.

.4. Interictal hallucinosis in patients withpilepsy

nterictal hallucinosis in the absence of a seizure is one sus-ects, on the basis of clinical experience, rather common,ut the literature on this subject is small. One excep-ion is that of Silberman et al. (1994), who investigatedhe relationship between aura phenomena and interictal

adoif

100 2486 5

sychopathology (as measured by SADS-L/DSM-IIIR) in 21atients with a variety of seizure types. Psychosensoryymptoms were identified using the Silberman-Post Psycho-ensory Phenomena Scale (Silberman et al., 1985) whichncludes categories covering hallucinations, sensory illu-ions, depersonalisation—derealisation, cognitive illusionse.g., déjà vu, jamais vu, illusions of reference or sig-ificance, mystical experiences), paroxysmal affect, timeistortion, thought disturbance and motor automatismsTable 2).

As a group 90% had generalised seizures and 72% hadomplex partial seizures, with or without seizures ofther types. The authors whilst accepting that the symp-oms could be due to undiagnosed auras (simple partialeizures) chose to retain the designation ‘interictal as aonservative designation that makes the fewest assump-ions’. Clearly this is a hypothetical stance and theseymptoms could well be aura phenomena and thereforeot ‘interictal’ in the strictest sense at all. The mainnding was that of a strong correlation between ‘interic-al’ hallucinations, sensory illusions (macropsia, microspia,epersonalisation/derealisation, and autoscopy), cognitivellusions (déjà vu, jamais vu, illusions of reference orignificance, mystical experiences) and psychopathologyprimarily depression and anxiety), as well as more timepent in psychiatric treatment. There was no such rela-ionship between ictal symptoms and psychopathology. Theyuggest that limbic ‘kindling’ by chronic seizure dischargesould alter the individual’s affective response to environ-ental stimuli. The physiological basis of the symptoms was

ot explored but deserves further study.

.5. Psychopathology in interictal psychosis:omparison with primary psychoses

n the 1950s Hill and Pond observed that whilst patientsith psychosis in epilepsy could experience positive symp-

oms (delusions and hallucinations), they tended to remainwarm and appropriate’ and the course of illness lackedhe usual deterioration of personality seen in schizophre-ia (Trimble, 1991). Slater et al. (1963) also noted theseschizophrenia-like’ positive symptoms but agreed that the

ffect tended to remain preserved. He added that ‘theelusions and hallucinations of patients with the psychosisf epilepsy are empathiseable’ (i.e., the patient remainsn our world), although this point is arguable. Other keyeatures are the absence of negative symptoms and bet-

179

Fc

be

aptpaK

datb1tmi1rthttbcrTchanges in gene and protein expression, synaptic changes,neuronal death, neurogenesis, etc. and so it would be rea-sonable to postulate that these secondary changes mightresult in pathological circuitry causing psychosis. A detailed

Delusions, illusions and hallucinations in epilepsy

ter pre-morbid function (Perez et al., 1985; Slater et al.,1963). Crowe and Kuttner (1991) found that patients withTLE experienced more positive symptoms, particularly hal-lucinations and bizarre behaviour, with relatively fewernegative symptoms than patients with chronic schizophre-nia. In their series the TLE group also manifested a highersubjective level of depression. Interictal affective disorders,in particular the classical manic or bipolar presentations, areconsidered rare. Notable exceptions include Flor-Henry’s(1969) series in which an affective psychosis occurred in40% and the prevalence rate of 70% for depressive psychosisreported by Fenton (1978).

By way of contrast, Mendez et al. (1993) compared 62patients with epilepsy and psychosis with the same num-ber of patients with schizophrenia and concluded that ‘theschizophrenia of epilepsy appears to conform to the usualschizophrenic categories’. Perez et al. (1985) using thePresent State Exam (PSE) found that 46% of the epilepsygroup had nuclear schizophrenia symptoms almost indistin-guishable from those seen in patients with schizophrenia.Matsuura and Trimble (2000) highlight Akimoto, Hachiyaand Hosokawa’s observation of positive symptoms, in atotal of 55 patients, identical to those seen in schizophre-nia, although negative symptoms were less common. Tosummarise, it would seem that the key differences lie ina preservation of affect, fewer negative symptoms andarguably greater insight, whilst the greatest similarities canbe seen in positive symptomatology, i.e., that of thoughtdisorder, delusions and hallucinations.

4.6. SEEG studies

Clearly, the psychopathology of interictal psychosis is closein nature to that seen in the functional psychoses. The pos-sibility that the interictal psychosis of epilepsy may in partbe ‘driven’ by ongoing ictal discharges in the limbic systemis intriguing, and there is a body of suggestive data fromdepth EEG studies largely in patients with treatment resis-tant temporal lobe epilepsy. This has provided an interestingopportunity to study at first hand the anatomical substratefor these mental phenomena.

The recordings by Wieser (1983) and Wieser et al. (1985)and others which demonstrate continuous seizure activity(amounting to non-convulsive status epilepticus) during psy-chotic episodes are described in paper 1. However, thereis also evidence that psychosis can develop at times offrequent, but not continuous, interictal epileptiform dis-charges. Kristensen and Sindrup (1978) reviewed sphenoidalelectrode recordings in 96 patients with partial seizureswho developed a paranoid/hallucinatory psychosis after amedian of 18 years, and compared these with an epilep-tic control group. They found that the psychotic patientshad a significantly larger number of temporal medio-basalindependent spike foci (Fig. 7) than controls.

Patients with psychosis also had significantly more fre-quent bilateral than unilateral medio-basal spike foci than

controls (Fig. 8) with significant slow wave admixture sug-gesting more extensive and severe epileptogenic lesions inthe psychotic patients. They found no correlation betweenpsychosis and unilateral EEG foci in either temporal lobe.Interestingly, Hughes (1985) found that the incidence of

Fc

igure 7 Multiple independent spike foci in patient with psy-hosis (Kristensen and Sindrup, 1978).

ilateral as opposed to unilateral foci in temporal lobepilepsy increased with age at a rate of almost 1% per year.

Whilst association does not prove causation, it doesppear that patients with complex partial seizures are morerone to the schizophrenia like psychosis particularly inhose with bilateral mesio-basal spike foci and that thesychosis is associated with frequent discharges (Kristensennd Sindrup, 1978; Reynolds and Trimble, 1981; Sindrup andristensen, 1979).

Any mechanisms by which epilepsy might underlie theevelopment of interictal psychosis are unclear and wouldnyway be highly speculative. It is worth briefly noting thathere are two predominant theories. The first is the possi-ility that seizures ‘kindle’ psychosis (Smith and Darlington,996). This rests on the idea that the spread of seizureso the ventral tegmental area by secondary epileptogenesisay result in a potentiation of dopaminergic transmission

n the limbic system and neocortex (Stevens and Livermore,978; Stevens, 1991). However there are only limited dataelating to epileptiform activity in the VTA, current ampli-udes needed to induce kindling in the VTA are relativelyigh, and kindling is not normally associated with struc-ural brain damage. Smith and Darlington (1996) concludehat overall ‘the evidence in favour of a causal relationshipetween kindling and the development of psychosis is notompelling’. The second is that the epilepsy sets up a neu-onal network which itself results in psychotic symptoms.here is no doubt that epileptic seizures result in massive

igure 8 Independent bilateral medio-basal spike foci in psy-hotic patient (Kristensen and Sindrup, 1978).

1

dantassmotiil

4

Tceffisrh(i(apceaaiaia

4

Miinnuawicmataesisai2

pTcso(tbewoiosu

4

Fphaftfladreuadema

4

Te(ea5ootwcpmiiwhh

80

iscussion of this area is beyond the scope of this paperlthough an excellent review in relation to epilepsy (butot psychosis) is provided by Morimoto et al. (2004). A fur-her interesting possibility is that these changes converge,s in some animal models of human psychosis, in dopamineupersensitivity and an increase in the high affinity states oftriatal dopamine D2 receptors (D2HIGH), a mechanism whichay represent a final common pathway in the development

f psychosis (Seeman et al., 2006). One immediate key ques-ion in our view is to determine the extent to which thenterictal psychosis is dependent on ongoing epileptic activ-ty and the extent to which this is a phenomenon limited toimbic structures.

.7. Neuropathology

he neuropathological series on patients with interictal psy-hoses are in general rather old and usually based on anxtremely selected sub-group of patients with TLE, oftenrom institutions. This seriously limits the validity of thendings, however, the key points include: (1) ‘Alien tis-ue’ (small tumours, hamartomas, focal dysplasia) in theesected temporal lobe appear to be associated with aigher risk of psychosis than mesial temporal sclerosisTaylor, 1975) a finding not confirmed in structural MRI stud-es; (2) lesions that (a) originate in the fetus or perinatally,b) affect neurons in the medial temporal lobe, and (c) given early age of first fit, are significantly associated with are- or post-operative diagnosis of schizophrenia-like psy-hosis in patients with medically intractable TLE (Robertst al., 1990); (3) gangliogliomas and Ammon’s horn sclerosisre disproportionately associated with psychosis (Roberts etl., 1990; Suckling et al., 2000), and (4) severe neuron lossn CA1 is more common in patients without psychosis whilst,n increased density of calbindin-immunoreactive neuronsn CA4 is found in psychotic patients with TLE (Suckling etl., 2000).

.8. Structural magnetic resonance imaging

agnetic resonance imaging has certain advantages in thenvestigation of the interictal psychosis of epilepsy, hav-ng advanced to the stage that it approaches ‘in vivoeuropathology’. It avoids the inevitable selection bias ofeuropathology. Again, the key findings include: (1) ventric-lar enlargement and reduced temporal lobe, frontoparietalnd superior temporal gyrus grey-matter volumes in patientsith TLE both with and without psychosis of epilepsy and

n patients with schizophrenia when compared with healthyontrols (Marsh et al., 2001). In this study the abnor-alities were greatest in the psychotic/epileptic group,

lthough a later study by Rüsch et al. (2004), did not supporthis; (2) a 16—18% enlargement of the amygdala bilater-lly in patients with epileptic psychosis (Tebartz van Elstt al., 2002), in contrast to the reduction in amygdala sizeeen in schizophrenia, although the validity of this finding

s unclear; (3) magnetisation transfer and diffusion ten-or imaging findings of subtle frontotemporal white matterbnormalities possibly contributing to the cognitive deficitsdentified in patients with interictal psychosis (Flugel et al.,006a,b); (4) temporal lobe dysplasia is more frequent in

ttwaa

B. Elliott et al.

atients with postictal psychosis (Breillmann et al., 2000).he problem with all such studies is the lack of pathologi-al confirmation and in schizophrenia, for instance, imagingtudies have shown abnormalities in many regions, manyf which have no known genuine pathological significancesee below). Similarly, it is now known that epilepsy dueo mesial temporal sclerosis is associated with widespreadut subtle changes in volume of many other temporal andxtratemporal structures, and which may not be visibleithout quantitative analysis (Moran et al., 2001). Mostf the studies test no serious hypothesis, and the find-ngs could simply reflect factors correlated with featuresf epilepsy more than psychosis, and for all these rea-ons, this corpus of work is in our view of very limitedtility.

.9. Functional brain imaging

unctional brain imaging has also been used to study thesychosis of epilepsy. The lack of clear pathophysiologicalypotheses and the confounding effects of the underlyingetiology and the epilepsy itself, each of which can causeunctional changes, however, limit the usefulness of thisechnique also. Numbers are often small and the results con-ict; however, the main findings are: (1) left temporal lobend amygdala hyperperfusion on SPECT in patients with TLEuring a psychotic episode (Jibiki et al., 1993); (2) left supe-ior temporal hypoperfusion in patients with psychosis ofpilepsy (Mellers et al., 1998; Marshall et al., 1993); (3) fail-re to find an association between the psychosis of epilepsynd laterality (Fong et al., 2000); and (4) no significantifferences between patients with mesial temporal lobepilepsy with and without psychosis, after Bonferroni adjust-ent, in rCBF patterns using interictal SPECT (Guarnieri et

l., 2005).

.10. Neurogenetics

he genetic study of epilepsy has also yielded some inter-sting findings. One such study is that of Winawer et al.2000) linking the syndrome of autosomal dominant partialpilepsy with auditory features (ADPEAF) in families to anrea of chromosome 10q. Auditory auras were present in5% of these individuals who had complex partial and sec-ndary generalised seizures. Another example is the studyf Magnusson et al. (2003) of autosomal dominant noc-urnal frontal lobe epilepsy (ADNFLE) which is associatedith mutations in two genes coding for the nicotinic acetyl-holine receptor (CHRNA4 and CHRNB2). The frequency ofsychiatric disorders in two families with different CHRNA4utations (776ins3 and Ser248Phe) was studied. In the fam-

ly with the 776ins3 mutation at least four of the tenndividuals diagnosed with epilepsy had been in contactith psychiatric services, one had schizophrenia, anotherad experienced at least two severe psychotic episodes andad been taking antipsychotic medications for years. The

hird family member had been hospitalised at least threeimes for psychiatric problems whilst the fourth needed helpith activities of daily living due to incapacitating apathy,lthough she lacked a formal psychiatric diagnosis. Theseuthors suggested examining patients with the schizophre-

amgopngCllcicD

olid

6p

Wieptipcttt(icr

ttc1i1d1(AascoIb

Delusions, illusions and hallucinations in epilepsy

nia like psychosis of epilepsy for this mutation althoughthese findings have not yet been followed up.

5. Comparison with electrophysiological,neuropathological, structural and functionalimaging findings in schizophrenia

5.1. EEG studies

Investigators have reported spike activity in the septal aswell as other regions in patients with schizophrenia undergo-ing assessment for neurosurgery (Kendrick and Gibbs, 1957;Crandall et al., 1969; Hanley et al., 1972). Sixty-two ofKendrick and Gibbs (1957) series had schizophrenia. Dur-ing the process of assessment for neurosurgical treatmentof intractable psychiatric symptoms, they noted that nearlyhalf of these patients demonstrated a clear spike focus inthe anterior temporal and/or frontal region with the com-monest site of discharge being the mesial orbital cortex ofthe frontal lobe. The medial temporal lobe was almost asfrequently involved and spread from medial temporal toorbitofrontal regions was seen commonly in both groups.

5.2. Neuropathology

A detailed consideration of this complex topic is outside thescope of this review article. Controversy surrounds almostevery point, but based on a resurgence of post-mortemstudies throughout the 1980s the view has evolved thatrather than having a distinct neuropathological signature,schizophrenia consists of quantitative alterations in variousnormal parameters of neural microcircuitry. For the inter-ested reader, two excellent reviews are provided by Harrison(1999) and Harrison and Weinberger (2005).

5.3. Structural and functional brain imaging

A systematic review of volumetric MRI studies in schizophre-nia by Lawrie and Abukmeil (1998) suggested that inschizophrenia there was a median 40% increase in lateraland third ventricle size which is accompanied by a 3% lossof brain tissue. There is no consistent correlation betweenthe two. Larger tissue reductions are seen in the temporallobe as a whole (∼8%) and in medial temporal structures.In unmedicated and first episode patients caudate volumesmay be reduced (Shihabuddin et al., 1998) and there is alsoincreasing evidence for the pathophysiological involvementof the cerebellum (Andreasen et al., 1996).

The left temporal lobe has attracted the most attentionin structural imaging studies relating to auditory halluci-nations in schizophrenia. The superior temporal gyrus hasprovided the main focus. In a review by Stephane et al.(2001), 10 studies examined the STG. Five found someassociation with hallucinations and in three there was a neg-ative correlation between AH severity and STG volume. This

inverse correlation between STG volume and AH severity wasfirst noted by Barta et al. (1990) and subsequently replicatedby Rajarethinam et al. (2000); mirroring findings in epilepsy(Moran et al., 2001) and possibly related to genetic as wellas acquired factors.

rn1dp

181

Diencephalic structures are also involved. Shapleske etl. (2002) using voxel based morphometry in seventy-twoen with schizophrenia found localised areas of reduced

rey-matter affecting the medial temporal lobes, insula,rbitofrontal cortex (including anterior cingulate) and therecuneus and lingual gyri. Contrasting hallucinator andon-hallucinator groups revealed a single region of reducedrey-matter tissue proportion affecting the left insula.respo-Facorro et al. (2000) also used MRI to explore insu-

ar cortex morphology and found a significant reduction ineft insula grey-matter volume in patients compared withontrols. Involvement of the insular cortex has been clearlymplicated in the SEEG studies outlined above as well as inognitive neuropsychological models of psychosis (Frith andolan, 1997).

Functional imaging studies in general support the findingsutlined above in that they suggest a role for the temporalobe in general and superior temporal gyrus in particularn hallucinations associated with schizophrenia. Again, aetailed review is provided by Allen et al. (2008).

. The effect of temporal lobe surgery on thesychoses of epilepsy

hen temporal lobectomy was first introduced in the 1950s,t had been hoped that it would alleviate the psychosis ofpilepsy as well as seizures, indeed, psychosis was seen as aositive indication. However, it was rapidly recognised thatemporal lobectomy could also make psychosis worse. Theres now a general acceptance that the presence of interictalsychosis in particular carries a significant risk of recrudes-ent post-operative psychosis and it is generally consideredo be a contra-indication to surgery. It is usual in our practiceo counsel patients of a 5—10% risk, although the litera-ure provides quite varying figures. The underlying aetiologyhippocampal sclerosis, alien tissue, etc.) is not thought tonfluence prognosis in this regard (in spite of the pathologi-al literature reviewed above), but most are agreed that theisk of psychosis is greater if seizures continue after surgery.

A detailed review of the literature on this topic is outsidehe scope of this article, but we here emphasise key themeshat emerge out of this extensive literature: (1) Patientsan be cured of their psychosis following surgery (Falconer,958; Taylor, 1972), although length of follow up may be anssue. (2) The psychosis can get worse (Simmel and Counts,958), and often does. (3) Postictal psychosis can arisee novo following temporal lobectomy (Manchanda et al.,993). (4) De novo interictal psychosis is also well recognisedTaylor, 1972; Simmel and Counts, 1958; Shaw et al., 2004).review by Matsuura (1997) puts the incidence at about 4%,

nd in two-thirds of these patients complete seizure remis-ion was not achieved. In two of Shaw’s series there was alose temporal link between the recurrence of seizures post-peratively and the first appearance of psychotic symptoms.nterestingly, one of the significant risk factors includedilateral pre-operative EEG abnormalities. (5) Right sided

esections appear to be associated with an increased vul-erability to developing psychosis post-operatively (Stevens,990; Manchanda et al., 1993; Matsuura, 1997). This isespite the over-representation of left sided pathology atre-operative assessment highlighted by Trimble (1992). The

1

dc

7e

IahbitCaoar

ioc(ca

cti

cyAaom(apao0ibaffrnmtpiw

82

ata is not entirely consistent and it is clear also that psy-hosis can follow both left and right sided operations.

. Therapy of hallucinations and psychosis inpilepsy

ctal hallucinations are best treated by controlling the ictus,nd thus by antiepileptic drugs. In cases where postictalallucinosis or psychosis is thought to be due to ongoing lim-ic status epilepticus, acute therapy with benzodiazepiness usually advised, sometimes with additional antipsychoticherapy (see below) if the psychosis is particularly florid.lobazam is a common antiepileptic drug choice given orallyt 20—30 mg/day over several days. The postictal psychosisf epilepsy clearly can respond to ECT (Farkas, 2002),lthough it is hardly ever necessary to resort to this in cur-ent practice.

A detailed exposition of the treatment of hallucinationsn the interictal psychosis of epilepsy is beyond the scopef this article, and a recent review of the use of antipsy-hotic drugs in epilepsy has been provided by Koch-Stoecker2002). Usually antipsychotic drugs in low doses are suffi-ient to control the overt signs and there is little role for

ntiepileptic drugs in this regard.

Several points are worth emphasis though. Treatment isomplicated by the fact that all antipsychotic drugs havehe propensity to cause paroxysmal EEG abnormalities andnduce seizures (Itil and Soldatos, 1980). Generalised tonic-

eatsi

Table 3 Frequently used antipsychotic drugs in patients with epi

Antipsychotic Startingdose/timing

Usualtreatment dose

Max dose P

Risperidone 2 mg 4—6 mg 16 mg(doses > 8 mgare rarely used)

5�

Olanzapine 5 mg 5—15 mg 20 mg mD5

Quetiapine 25—100 mg(dependingon age)

400—600 mg 750—800 mg D5H

Amisulpiride 200 mg 400—800 mg 1200 mg Dp

Clozapine 12.5 mg 300—700 mg(adjustedaccording toplasma leveland response)

900 mg D�

m3

B. Elliott et al.

lonic seizures were attributed to chlorpromazine barely aear after its introduction in 1952 (Anton-Stephens, 1953).classic study in baboons demonstrated that haloperidol

t doses of 0.6—1.2 mg/kg produced an increased incidencef spontaneous EEG spikes and waves and a great enhance-ent of paroxysmal EEG activity during photic stimulation

Meldrum et al., 1975). Almost all available antipsychoticsre mildly epileptogenic, with seizure incidence rates (inatients who previously had not had seizures) ranging frompproximately 0.1% to approximately 1.5% (the incidencef the first unprovoked seizure in the general population is.07—0.09) (Pisani et al., 2002). EEG changes seem to occurn about 7% of patients treated with antipsychotic drugs,ut in most cases are of little clinical consequence (Benkertnd Hippius, 1998). In one recent review seizure incidenceor the different atypical antipsychotics was given as 0.3%or risperidone and 0.9% for olanzapine and quetiapine. Theisk with clozapine increased in a dose dependant man-er from 1.0% with low doses (<300 mg/day), to 2.7% withoderate doses and 4.4% with high dose (600—900 mg/day)

reatment (Alldredge, 1999). Paccia and Devinski (1994) in aost marketing study of more than 5000 patients on clozap-ne showed a lower rate (1.3%) with no dose dependency, andhen Langosch and Trimble (2002) treated six patients with

pilepsy and severe psychosis with clozapine, none showedn increase and in three there was a substantial reduc-ion in seizure frequency. The fear of eliciting additionaleizures has led to what we consider exaggerated cautionn antipsychotic drug use, and in patients co-medicated

lepsy (doses, receptor binding and other effects).

harmacology Adverse/other effects

-HT2A, D2,1, �2, 5HT7

EPSE at higher doses, hyperprolactinaemia,hyper triglyceridaemia, insulin resistance,HONK. Less weight gain than olanzapineand clozapine. Higher doses may berequired due to enzyme induction bycarbamazepine/Phenytoin (CYP 3A4).

ACh, 5-HT2A,1-4, �1/2, H1,HT-3/6, 5HT-C

Weight gain, sedation (less than clozapine).Poses definite cardiometabolic,dyslipidaemia and diabetes risks. EPSE’s notcommon. Enzyme induction possible withcarbamazepine (CYP 1A2).

2,-HT2A/1A/2C,1, �1, �2, M1

Virtually no EPSE. No hyperprolactinaemia.Sedation. Intermediate to high risk ofdyslipidaemia/insulin resistance.

3. Possible D2artial agonist

Low EPS. May cause dose dependant QTcprolongation. Can increase prolactin.

1-4,5HT2A/C,1/2, H,1ACh, 5HT

/6/7.

Weight gain, sedation, hypersalivation,constipation, bowel obstruction,agranulocytosis, myocarditis, seizures.Sudden hyperosmolar syndrome/diabeticketoacidosis. High cardiometabolic risk.Reduces suicide risk Cannot be combinedwith carbamazepine or lamotrigine becauseof risk of agranulocytosis, Needs slowintroduction with EEG monitoring.Specialist monitoring registration.

twoftpwa1ttshidtpso(1tcuiodnl(

aincisFte(saaa

tntphdicse

Delusions, illusions and hallucinations in epilepsy

with antiepileptic drugs (as is almost always the case inpatients with epilepsy and psychosis) the risk of an increasein seizures is low. In clinical practice, this becomes an impor-tant issue mainly in those who are seizure free, in whom therecurrence of seizures has a generally greater impact.

Further problems associated with antipsychotic drugtreatment in patients with epilepsy include: (a) variation inthe individually inherited seizure threshold (for a thoroughreview see Pisani et al., 2002), (b) pharmacokinetic inter-actions due to common metabolism via cytochrome p-450isoenzymes, (c) side effects, in particular sedation, weightgain, worsening of glycaemic control, haematotoxicity andhepatotoxicity, some of which can be moderated by carefuldose titration, (d) a relative lack of studies investigating theefficacy of antipsychotic drugs in the psychoses of epilepsy.A summary of some frequently used antipsychotic drugs andtheir suitability for patients with epilepsy is provided inTable 3.

8. Discussion

In both this paper and its pair we have provided adetailed review of the literature pertaining to the clinicalfeatures and electrophysiological underpinnings of halluci-natory experience in patients with epilepsy. Whilst the firstpaper dealt predominantly with more elementary psycholog-ical experiences, this second paper reviews more complexhallucinatory or ‘psychotic’ states especially from the clin-ical and stereo-EEG perspectives.

A number of conclusions can be drawn. First, it is clearthat prolonged epileptic seizures can result in hallucina-tory symptoms which are indistinguishable from those inthe primary psychoses. These are an example of ictal psy-chosis and occur in complex partial status and in some casesof (wrongly termed) postictal psychosis. The hallucinatorystates are often indistinguishable from those in the primarypsychoses, although almost always have additional epilep-tic features such as confusion or altered awareness. Limbiccomplex partial status is more likely to result in psychoticfeatures than the more common frontal lobe complex partialstatus, although hallucinatory symptoms in both overlap.

Second, interictal hallucinosis is probably more com-mon in epilepsy than is appreciated, but is ill studied. Towhat extent these are caused by subclinical epileptic lim-bic discharges (i.e., auras), and are therefore ‘ictal’, isnot known. Other interictal mental changes include theinterictal dysphoric disorder defined by Blumer (2000) orthe Astheno-Emotional Syndrome described by Lindqvist andMalmgren (1993) the latter being also particularly observedafter temporal lobectomy. Evidence from Silberman et al.(1994) suggests that ‘interictal’ psychopathology (bearing inmind these could well be aura phenomena) is correlated withdepression and anxiety as well as time spent in psychiatrictreatment and propose that ‘kindling’ by chronic seizuredischarges could alter the individual’s affective response toenvironmental stimuli.

In many ways the most interesting of all these condi-tions is the interictal psychosis of epilepsy. In this condition,hallucinatory and other psychotic features occur in clearconsciousness and are often chronic. From the examplesgiven it is now likely that at least some such cases have con-

iwric

183

inuous or semi-continuous limbic seizure activity associatedith the hallucinatory state. The clinical phenomenologyf this condition is remarkably similar to those seen in theunctional psychoses. Other interesting features are the facthat the interictal psychosis most commonly occurs in tem-oral lobe (limbic) epilepsy and is more common in thoseith frequent seizures, that the temporal lobe epilepsy islmost always present and active for many years (more than0 years) before the onset of the psychotic symptoms, andhat the psychosis can develop following a period of ‘pos-ictal psychosis’ and after temporal lobectomy, especially ifurgery fails to control seizures. Laterality would appear toave less influence than sometimes claimed, but psychosiss much more likely to occur when there are bilateral limbicischarges. The most intriguing questions in this field relateo the possibility that some cases of apparently ‘interictal’sychosis may in fact be driven by ictal discharges in limbictructures. This is an inference supported by a small bodyf SEEG recordings in psychotic patients outlined earlierWieser, 1983; Wieser et al., 1985; Kristensen and Sindrup,978; Sindrup and Kristensen, 1979). It is our contention thathere is a sub-group of patients, in whom chronic networkhanges evolve over time causing complex psychotic statesnderpinned by recurrent seizure discharges. One possibilitys that abnormal neuronal network development is the causef the epileptic psychosis, although other work in bipolarisorder (but not epilepsy) considers the possible sharedeurochemical underpinnings with epilepsy with a particu-ar focus on neurotransmitter and second messenger systemse.g., Amann and Grunze, 2005).

If schizophrenia like states in patients with epilepsy canrise as a result of these neurophysiological processes its interesting to speculate on whether primary schizophre-ia might share a similar underlying pathophysiology. Thelinical features of epileptic psychosis are similar, if notdentical to, those of the primary psychoses, but clinicalimilarity does not necessarily imply similar mechanisms.urthermore, most of the evidence to support this idea isoo sparse, old, or ethically contentious to quote. How-ver, Kendrick and Gibbs (1957) series and Hanley et al.’s1972) case report of intracranial SEEG in patients withchizophrenia does demonstrate spike foci in those samereas electrophysiologically implicated in the ictal, postictalnd interictal psychoses as well as in the neuropathologicalnd neuroimaging series described above.

There are various fruitful areas for further research inhis field which is at the interface between psychiatry andeurology. As Wieser (1983) wrote: ‘Paralleled by clear cutime related epileptic discharges in hidden brain areas thehasic psychical disturbance of some patients would notave been accepted as an epileptic dysfunction withoutepth recordings. Such observations are perhaps first stepsn building a long desired bridge to psychiatry, because itan hardly be denied that in innumerable patients with tran-ient mental changes, the same might be true if they werexplored with depth electrodes.’

We concur with this view (although it is clear that

ntracranial SEEG is barred by ethical concerns in patientsith primary psychoses) and consider this an area of

esearch which is potentially of exceptional interest andmportance. There is little doubt that electrophysiologicalhanges, sometimes frankly epileptiform changes amounting

1

tniipiper

fdstonocisseaanai

R

A

A

A

A

A

S

A

B

B

B

B

B

B

C

C

C

C

C

D

F

F

F

F

F

F

F

F

F

G

G

G

84

o prolonged epileptic seizures, underpin hallucinatory phe-omenon in many persons with epilepsy, and possibly alson some psychotic patients without epilepsy. The challenges to determine what neurophysiological or neurochemicalrocesses are responsible for their development, what sim-larities these processes have in epileptic and non-epilepticatients, and to what extent these processes are directlypileptic in nature or indirectly caused by network or neu-ochemical changes induced by epilepsy.

Further work is clearly required but there are severalactors that make research into the role played by epilepticischarges in the genesis of psychotic states difficult. Thecalp EEG has been widely used but is particularly insensi-ive at detecting activity in medial limbic structures, it isften normal despite widespread activity on SEEG. Intracra-ial electrodes however are invasive, carry a 1—4% riskf significant morbidity and mortality and are only indi-ated when precise localisation of an epileptogenic zones required to plan a surgical resection. Furthermore, theirpatial resolution is highly limited to the extent that verymall adjustments in position can lead to markedly differ-nt results. To further complicate the picture epileptiformctivity is often a widely distributed network phenomenon,ctivity recorded at one point in time and space mayot be present at a later time as it is possible that notll of a particular network is active at any one pointn time.

eferences

dachi, N., Onuma, T., Nishiwaki, S., Murauchi, S., Akanuma, N.,Ishida, S., Takei, N., 2000. Inter-ictal and post-ictal psychoses infrontal lobe epilepsy: a retrospective comparison with psychosesin temporal lobe epilepsy. Seizure 9, 328—335.

lldredge, B.K., 1999. Seizure risk associated with psychotropicdrugs: clinical and pharmacokinetic considerations. Neurology53 (Suppl. 2), S68—75.

llen, P., Laroi, F., MsGuire, P.K., Aleman, A., 2008. The hallucinat-ing brain: a review of structural and functional neuroimagingstudies of hallucinations. Neurosci. Biobehav. Rev. 32 (1),175—191.

mann, B., Grunze, H., 2005. Neurochemical underpinnings in bipo-lar disorder and epilepsy. Epilepsia 46 (Suppl. 4), 26—30.

ndreasen, N.C., O’Leary, D.S., Cizadlo, T., Arndt, S., Rezai, K.,Ponto, L.L., et al., 1996. Schizophrenia and cognitive dysme-tria: a positron emission tomography study of dysfunctionalprefrontal-thalamic-cerebellar circuitry. Proc. Natl. Acad. Sci.U.S.A. 93, 9985—9990.

tephane, M., Barton, S., Boutros, N.N., 2001. Auditory verbal hal-lucinations and dysfunction of the neural substrates of speech.Schizophr. Res. 50, 61—78.

nton-Stephens, D., 1953. Preliminary observations on the psychi-atric use of chlorpromazine. J. Ment. Sci. 100, 543—547.

arczak, P., Edmunds, E., Betts, T., 1988. Hypomania following com-plex partial seizures. Br. J. Psychiatry 152, 137—139.

arta, P.E., Pearlson, G.D., Powers, R.E., Richards, S.S., Tune,L.E., 1990. Auditory hallucinations and smaller superior tem-poral gyral volume in schizophrenia. Am. J. Psychiatry 147,1457—1462.

enkert, O., Hippius, H., 1998. Kompendium der psychiatrischenpharmackotherapie. Springer, Berlin.

lumer, D., 2000. Dysphoric disorders and paroxysmal affects:recognition and treatment of epilepsy related psychiatric dis-orders. Har. Rev. Psychiatry 8 (1), 8—17.

G

B. Elliott et al.

reillmann, R.S., Kalnins, R.M., Hopwood, M.J., Ward, C., Berkovic,S.F., Jackson, G.D., 2000. TLE patients with postictal psychosis:mesial dysplasia and anterior hippocampal preservation. Neurol-ogy 55 (7), 1027—1030.

yrne, A., 1988. Hypomania following increased epileptic activity.Br. J. Psychiatry 153, 573—574.

almeil, L.F., 1824. De L’Epilepsie: Etude Sous la Rapport de sonSeige et de son Influence sur le Production de L’Alienation Men-tale. Thése de Paris.

apgras, J., Reboul-Lachaux, J., 1923. L’illusion des ‘sosies’ dansun delire systematise chronique. Bull. Soc. Clin. Med. Ment. 11,6—16.

randall, P.H., Hanley, J., Kales, A., Namerow, N.S., Ornitz,E.M., Rand, R.W., Rickles Jr., W.H., Ritvo, E.R., Walter, R.D.,1969. Clinical neurophysiology: newer diagnostic and therapeu-tic methods in neurological disease and behaviour disorders.Ann. Int. Med. 71, 619—645.

respo-Facorro, B., Kim, J.J., Andreasen, N.C., O’Leary, D.S., Bock-holt, J., Magnotta, V., 2000. Insular cortex abnormalities inschizophrenia: a structural MRI study of first episode patients.Schizophr. Res. 46, 35—43.

rowe, S.F., Kuttner, M., 1991. Differences between schizophreniaand the schizophrenia-like psychosis of temporal lobe epilepsy.Neuropsychiatry Neuropsychol. Behav. Neurol. 4 (2), 127—135.

ongier, S., 1959. Statistical study of clinical and electroencephalo-graphic manifestations of 536 psychotic episodes occurring in516 epileptics between clinical seizures. Epilepsia 1, 117—142.

alconer, M., 1958. In: Baldwin, M., Bailley, P. (Eds.), Temporal LobeEpilepsy. Discussion. Thomas CC, Springfield, IL, pp. 537—558.

arkas, M., Baran, B., Karpati, R., Rajna, P., 2002. Utility of elec-troshock therapy in epilepsy-associated psychosis. Ideggyogy Sz.55 (11—12), 400—405.

enton, G.W., 1978. Epilepsy and psychosis. J. Ir. Med. Assoc. 71,315—324.

ilho, G.M., Rosa, V.P., Lin, K., Caboclo, L.O., Sakamoto, A.C., Yacu-bian, E.M., 2008. Psychiatric comorbidity in epilepsy: a studycomparing patients with mesial temporal sclerosis and juvenilemyoclonic epilepsy. Epilepsy Behav. 13 (1), 196—201.

lor-Henry, P., 1969. Psychosis and temporal lobe epilepsy. Epilepsia10, 363—395.

lugel, D., Cercignani, M., Symms, M.R., Koepp, M.J., Foong, J.,2006b. A magnetisation transfer imaging study in patients withtemporal lobe epilepsy and interictal psychosis. Biol. Psychiatry59 (6), 560—567.

lugel, D., Cercignani, M., Symms, M.R., O’Toole, A., Thompson,P.J., Koepp, M.J., Foong, J., 2006a. Diffusion tensor imagingfindings and their correlation with neuropychological deficits inpatients with temporal lobe epilepsy and interictal psychosis.Epilepsia 47 (5), 941—944.

ong, G.C., Fong, K.Y., Mak, W., Tsang, K.L., Chan, K.H., Che-ung, R.T., Ho, S.L., 2000. Postictal psychosis related to regionalcerebral hyperperfusion. J. Neurol. Neurosurg. Psychiatry 68,100—101.

rith, C.D., Dolan, R.J., 1997. Brain mechanisms associated withtop-down processes in perception. Philos. Trans. R. Soc. Lond.Biol. 352, 1221—1230.

astaut, H., Roger, J., Lob, H. (Eds.), 1967. Xth Colloque de Mar-seille 1962. Masson, Paris.

astaut, H., Roger, J., Roger, A., 1956. Sur la signification decertaines fugues épileptiques. A propos d’une observation élec-troclinique d’état de mal temporal. Rev. Neurol. 94, 298—301.

loor, P., Olivier, A., Quesney, L.F., Andermann, F., Horowitz, S.,1982. The role of the limbic system in experiential phenomena

of temporal lobe epilepsy. Ann. Neurol. 12 (2), 129—144.

uarnieri, R., Wichert-Ana, L., Hallack, J.E.C., Velasco, T.R., Walz,R., Kato, M., Alexandre, V., Terra-Bustamente, V.C., Bianchin,M.M., Zuardi, A.W., Deakin, J.F.W., Sakamoto, A.C., 2005. Inter-ictal SPECT in patients with mesial temporal lobe epilepsy and

M

M

M

M

M

M

M

M

M

M

M

M

O

P

P

P

P

Q

R

R

R

Delusions, illusions and hallucinations in epilepsy

psychosis: a case—control study. Psychiatry Res.: Neuroimaging138, 75—84.

Hanley, J., Rickles, W.R., Crandall, P.H., Walter, R.D., 1972. Auto-matic recognition of EEG correlates of behaviour in a chronicschizophrenic patient. Am. J. Psychiatry 128, 1524—1528.

Harrison, P.J., 1999. The neuropathology of schizophrenia: a criticalreview of the data and their interpretation. Brain 122 (Part 4),593—624.

Harrison, P.J., Weinberger, D.R., 2005. Schizophrenia genes, geneexpression, and neuropathology: on the matter of their conver-gence. Mol. Psychiatry 10 (1), 40—68.

Hughes, J.R., 1985. Long-term clinical and EEG changes in patientswith epilepsy. Arch. Neurol. 42, 213—223.

Itil, T.M., Soldatos, C., 1980. Epileptogenic side effects of psy-chotropic drugs. JAMA 244 (13), 1460—1463.

Jackson, J.H., 1875. On temporary mental disorders after epilep-tic paroxysm. West Riding Lunatic Assylum Med. Rep. 5,105—129.

Jibiki, I., Maeda, T., Kubota, T., Yamaguchi, N., 1993. 123I-IMP SPECTBrain imaging in epileptic psychosis: a study of two cases oftemporal lobe epilepsy with schizophrenia-like syndrome. Neu-ropsychobiology 28, 207—211.

Kanemoto, K., Kawasaki, J., Kawai, J., 1996. Postictal psychosis: acomparison with acute interictal and chronic psychoses. Epilep-sia 37, 551—556.

Kanemoto, K., Tsuji, T., Kawasaki, J., 2001. Re-examination ofinterictal psychoses based on DSM IV psychosis classification andinternational epilepsy classification. Epilepsia 42 (1), 98—103.

Kanemoto, K., 1997. Periictal Capgras syndrome after clusteredictal fear: depth-electroencephalogram study. Epilepsia 38 (7),847—850.

Kanner, A.M., Barry, J.J., 2001. Controversies in epilepsy andbehaviour: is the psychopathology of epilepsy different from thatof nonepileptic patients? Epilepsy Behav 2, 170—186.

Kanner, A.M., Stagno, S., Kotagal, P., Morris, H.H., 1996.Postictal psychiatric events during prolonged video-electroencephalographic monitoring studies. Arch. Neurol.53, 258—263.

Kendrick, J.F., Gibbs, F.A., 1957. Origin, spread and neurosurgi-cal treatment of the psychomotor type of seizure discharge. J.Neurosurg. 14, 270—284.

Koch-Stoecker, S., 2002. Antipsychotic drugs and epilepsy: indica-tions and treatment guidelines. Epilepsia 43 (Suppl. 2), 19—24.

Kraeplin, E., 1913. Psychiatrie: Ein Lehrbuch fur Studierended undArzte. Verlag von Johann Ambrosius Barth, Achten Auflage,Leipzig.

Kristensen, O., Hein Sindrup, E., 1978. Psychomotor epilepsy andpsychosis: II. Electroencephalographic findings (Sphenoidal elec-trode recordings). Acta Neurol. Scand. 57, 370—379.

Langosch, J.M., Trimble, M.R., 2002. Epilepsy, psychosisand clozapine. Hum. Psychopharmacol. 17 (2), 115—119.

Lawrie, S.M., Abukmeil, S.S., 1998. Brain abnormality in schizophre-nia. A systematic and quantitative review of volumetricmagnetic resonance imaging studies (review). Br. J. Psychiatry172, 110—120.

Levin, S., 1952. Epileptic clouded states. J. Nerv. Ment. Dis. 116,215—225.

Lindqvist, G., Malmgren, H., 1993. Organic mental disorders ashypothetical pathogenetic processes. Acta Psychiatr. ScandSuppl 373, 5—17.

Logsdail, S.J., Toone, B.K., 1988. Post-ictal psychoses: a clinical andphenomenological description. Br. J. Psychiatry 152, 246—252.

Magnusson, A., Stordal, E., Brodtkorb, E., Steinlein, O., 2003.Schizophrenia, psychotic illness and other psychiatric symp-toms in families with autosomal dominant nocturnal frontal lobeepilepsy caused by different mutations. Psychiatr. Genet. 13 (2),91—95.

R

185

anchanda, R., Miller, H., McLachlan, R.S., 1993. Post-ictal psy-chosis after right temporal lobectomy. J. Neurol. Neurosurg.Psychiatry 56, 277—279.

arsh, L., Sullivan, E.V., Morrell, M., Lim, K.O., Pfefferbaum, A.,2001. Structural brain abnormalities in patients with schizophre-nia, epilepsy, and epilepsy with chronic interictal psychosis.Psychiatry Res. 108 (1), 1—15.

arshall, E.J., Syed, G.M., Fenwick, P.B., Lishman, W.A., 1993. Apilot study of schizophrenia like psychosis in epilepsy using singlephoton emission computerised tomography. Br. J. Psychiatry 163,32—36.

athern, G.W., Pretorius, J.K., Babb, T.L., Quinn, B., 1995. Unilat-eral hippocampal mossy fiber sprouting and bilateral asymmetricneuron loss with episodic postictal psychosis. J. Neurosurg. 82(2), 228—233.

atsuura, M., Trimble, M.R., 2000. Psychosis in epilepsy: a reviewof Japanese studies. Epilepsy Behav. 1, 315—326.

atsuura, M., 1997. Psychosis of epilepsy, with special reference toanterior temporal lobectomy. Symposium 1. Epilepsia 38 (Suppl.6), 32—34.

cLean, P.D., 1952. Some psychiatric implications of physio-logical studies on frontotemporal portion of limbic system(visceral brain). Electroencephalogr. Clin. Neurophysiol. 4,407—418.

eldrum, B., Anlezark, G., Trimble, M., 1975. Drugs modifyingdopaminergic activity and behaviour, the EEG and epilepsy inPapio papio. Eur. J. Pharmacol. 32, 203—213.

ellers, J.D., Adachi, N., Takei, N., Cluckie, A., Toone, B.K., Lish-man, W.A., 1998. SPET study of verbal fluency in schizophreniaand epilepsy. Br. J. Psychiatry 173, 69—74.

endez, M.F., Grau, R., Doss, R.C., Taylor, J.L., 1993. Schizophreniain epilepsy: seizure and psychosis variables. Neurology 43 (6),1073—1077.

oran, N.F., Lemieux, L., Kitchen, N.D., Fish, D.R., Shorvon, S.D.,2001. Extrahippocampal temporal lobe atrophy in temporal lobeepilepsy and mesial temporal sclerosis. Brain 124 (1), 167—175.

orimoto, K., Fahnestock, M., Racine, R.J., 2004. Kindling and sta-tus epilepticus models of epilepsy: rewiring the brain. Prog.Neurobiol. 73, 1—60.

shima, T., Tadokoro, Y., Kanemoto, K., 2006. A prospective study ofpostictal psychoses with emphasis on the periictal type. Epilep-sia 47 (12), 2131—2134.

accia, S.V., Devinski, O., 1994. Clozapine-related seizures: expe-rience with 5629 patients. Neurology 44, 2247—2249.

erez, M.M., Trimble, M.R., Murray, N.M.F., 1985. Epileptic psy-chosis: an evaluation of PSE profiles. Br. J. Psychiatry 146,155—163.

isani, F., Oteri, G., Costa, C., Di Raimondo, G., Di Perri, R., 2002.Effects of psychotropic drugs on seizure threshold. Drug Safety25 (2), 91—110.

ond, D.A., 1974. Epilepsy and personality disorders. In: Vinken,P.L., Bruyn, C.W. (Eds.), Handbook of Clinical Neurology. NorthHolland Publishing Co., Amsterdam.

in, P., Laursen, T.M., et al., 2005. Epilepsy or a family history ofepilepsy increases the risk of schizophrenia or schizophrenia likepsychosis. BMJ 331, 23 (In: Evid Based Ment. Health 9 (2006) 23).

ajarethinam, R.P., DeQuardo, J.R., Nalepa, R., Tandon, R., 2000.Superior temporal gyrus in schizophrenia: a volumetric mag-netic resonance imaginging study. Schizophr. Res. 41, 303—312.

eynolds, E.H., Trimble, M.R. (Eds.), 1981. Epilepsy and Psychiatry.Churchill Livingstone, Edinburgh.

oberts, G.W., Done, D.J., Bruton, C., Crow, T.J., 1990. A ‘mock-

up’ of schizophrenia: temporal lobe epilepsy and schizophrenialike psychosis. Biol. Psychiatry 28, 127—143.

ohr-Le Floch, J., Gauthier, G., Beaumanoir, A., 1988. Etats confu-sionnels d’origine épileptique intérêt de l’EEG fait en urgence.Rev. Neurol. 144, 425—436.

1

R

S

S

S

S

S

S

S

S

S

S

S

S

S

S

S

S

S

S

S

S

T

T

T

T

T

T

W

W

W

W

Williams, D., 1959. The structure of emotions reflected in epilepticexperiences. Brain 79 (1), 29—57.

Winawer, M.R., Ottman, R., Hauser, W.A., Pedley, T.A., 2000. Auto-

86

üsch, N., Tebartz-van-Elst, L., Baeumer, D., Ebert, D., Trimble,M.R., 2004. Absence of cortical gray matter abnormalities inpsychosis of epilepsy: a voxel-based MRI study in patients withtemporal lobe epilepsy. J. Neuropsychiatry Clin. Neurosci. 16(2), 148—155.

achdev, P., 1998. Schizophrenia-like psychosis and epilepsy: thestatus of the association. Am. J. Psychiatry 155, 325—336.

eeman, P., Schwarz, J., Chen, J.-F., Szechtman, H., Perreault, M.,McKnight, G.S., Roder, J.C., Quirion, R., Boska, P., Srivastava,L.K., Yanai, K., Weinshenker, D., Sumiyoshi, T., 2006. Psychosispathways converge via D2HIGH dopamine receptors. Synapse 60,319—346.

engoku, A., Toichi, M., Murai, T., 1997. Comparison of psychoticstates in patients with idiopathic generalised epilepsy and tem-poral lobe epilepsy. Epilepsia 38 (Suppl. 6), 22—25.

engoku, A., Yagi, K., Seino, S., Wada, T., 1983. Risks of occurrenceof psychosis in relation to there types of epilepsies and epilepticseizures. Folia Psychiatr. Neurol. Jpn. 37, 221—226.

hapleske, J., Rossell, S.L., Chitnis, X., Suckling, J., Simmons, A.,Bullmore, E.T., Woodruff, P.W.R., David, A.S., 2002. A compu-tational morphometric MRI study of schizophrenia: effects ofhallucinations. Cerebral Cortex 12, 1331—1341.

haw, P., Mellers, J., Henderson, M., Polkey, C., David, A.S., Toone,B.K., 2004. Schizophrenia-like psychosis arising de novo follow-ing a temporal lobectomy: timing and risk factors. J. Neurol.Neurosurg. Psychiatry 75, 1003—1008.

hihabuddin, L., Buchsbaum, M.S., Hazlett, E.A., Haznedar, M.M.,Harvey, P.D., Newman, A., et al., 1998. Dorsal striatal size,shape, and metabolic rate in never-medicated and previouslymedicated schizophrenics performing a verbal learning task.Arch. Gen. Psychiatry 55, 235—243.

horvon, S., 2007. What is nonconvulsive status, and what are itssubtypes? Epilepsia 48 (Suppl. 8), 35—38.

horvon, S.D., 1994. Status Epilepticus: Its Clinical Features andTreatment in Children and Adults. Cambridge University Press,Cambridge, New York.

ilberman, E.K., Post, R.M., Nurnberger, J., Theodore, W.,Boulenger, J.P., 1985. Transient sensory, cognitive and affec-tive phenomena in affective illness. A comparison with complexpartial epilepsy. Br. J. Psychiatry 146, 81—89.

ilberman, E.K., Sussman, N., Skillings, G., Callanan, M., 1994. Auraphenomena and psychopathology: a pilot investigation. Epilepsia35 (4), 778—784.

immel, M.L., Counts, S., 1958. Clinical and psychological resultsof anterior temporal lobectomy in patients with psychomo-tor epilepsy. In: Baldwin, M., Bailley, P. (Eds.), Temporal LobeEpilepsy. Thomas CC, Springfield, IL, pp. 530—550.

indrup, E.H., Kristensen, O., 1979. Psychosis and temporal lobeepilepsy. In: Kulig, B.M., Meinardi, H., Stores, G. (Eds.), Epilepsyand Behaviour, vol. 1980. Swets and Zeitlinger BV, Lisse, pp.133—139.

later, E., Beard, A.W., Glithero, E., 1963. The schizophrenia like

psychosis of epilepsy. V. Discussion and conclusions. Br. J. Psy-chiatry 109, 95—150.

mith, P.F., Darlington, C.L., 1996. The development of psychosisin epilepsy: a re-examination of the kindling hypothesis. Behav.Brain Res. 75, 59—66.

B. Elliott et al.

o, N.K., Savard, G., Andermann, F., Olivier, A., Quesney, L.F., 1990.Acute postictal psychosis: a stereo EEG study. Epilepsia 31 (2),188—193.

tevens, J.R., Livermore, A., 1978. Kindling of the mesolimbicdopamine system: animal model of psychosis. Neurology 28,36—46.

tevens, J.R., 1990. Psychiatric consequences of temporal lobec-tomy for intractable seizures: a 20—30 year follow-up of 14cases. Psychol. Med. 20, 529—545.

tevens, J.R., 1991. Psychosis and the temporal lobe. In: Smith, D.,Treiman, D., Trimble, M. (Eds.), Advances in Neurology, vol. 55.Raven Press, New York.

uckling, J., Roberts, H., Walker, M., Highley, J.R., Fenwick, P.,Oxbury, J., Esiri, M.M., 2000. Temporal lobe epilepsy with andwithout psychosis: exploration of hippocampal pathology includ-ing that in subpopulations of neurons defined by their contentof immunoreactive calcium-binding proteins. Acta Neuropathol.99 (5), 547—554.

akeda, Y., Inoue, Y., Tottori, T., Mihara, T., 2001. Acute psychosisduring intracranial EEG monitoring: close relationship betweenpsychotic symptoms and discharges in the amygdala. Epilepsia42 (6), 719—724.

aylor, D.C., 1972. Mental state and temporal lobe epilepsy. Epilep-sia 13, 727—765.

aylor, D.C., 1975. Factors influencing the development ofschizophrenia-like psychosis in patients with temporal lobeepilepsy. Psychol. Med. 5, 249—254.

ebartz van Elst, L., Baeumer, D., Lemieux, L., Woermann, F.G.,Koepp, M., Krishnamoorthy, S., Thompson, P.J., Ebert, D., Trim-ble, M.R., 2002. Amygdala pathology in psychosis of epilepsy:a magnetic resonance imaging study in patients with temporallobe epilepsy. Brain 125, 140—149.

rimble, M.R., 1991. The Psychoses of Epilepsy. Raven Press, NewYork.

rimble, M.R., 1992. Behaviour changes following temporal lobec-tomy, with special reference to psychosis. J. Neurol. Neurosurg.Psychiatry 55, 89—91.

ieser, H.G., Hailemariam, S., Regard, M., Landis, T., 1985. Uni-lateral limbic epileptic status activity: stereo EEG, behaviouraland cognitive data. Epilepsia 26 (1), 19—29.

ieser, H.G., 1998. Electrophysiological aspects of forced nor-malization. In: Trimble, M.R., Bettina Schmitz (Eds.), ForcedNormalization and Alternative Psychoses of Epilepsy. WrightsonBiomedical Publ., Petersfield, UK, pp. 95—119.

ieser, H.G., 1980. Temporal lobe or psychomotor status epilepti-cus. A case report. Electroencephalogr. Clin. Neurophysiol. 48,558—572.

ieser, H.G., 1983. Depth recorded limbic seizures and psy-chopathology. Neurosci. Biobehav. Rev. 7, 427—440.

somal dominant epilepsy with auditory features: defining thephenotype. Neurology 54, 2173—2176.