SEARCH

SEARCH BY CITATION

Keywords:

  • Temporal lobe epilepsy;
  • Postictal psychosis;
  • Intracranial EEG;
  • Amygdala

Abstract

  1. Top of page
  2. Abstract
  3. CASE REPORT
  4. DISCUSSION
  5. REFERENCES

Summary:  Purpose: This report examined the underlying mechanism of psychosis associated with epilepsy.

Methods: An adult patient with epilepsy manifesting acute psychosis during long-term EEG monitoring is presented, together with a literature review on this subject.

Results: A 25-year-old woman with intractable temporal lobe epilepsy developed acute psychosis while she underwent long-term intracranial EEG monitoring. After a clustering of seizures, she manifested psychotic symptoms including hallucination, stupor, and repeated fear. The transition of psychotic symptoms corresponded to the changes in frequency and morphology of seizure discharges restricted to the left amygdala. Improvement of psychosis coincided with disappearance of seizure discharges.

Conclusions: This case confirmed a close relationship between psychotic symptoms and seizure discharges in the left amygdala. It is suggested that paroxysmal bombardment of the medial temporal lobe structure may be a pathogenetic factor of acute psychosis associated with epilepsy.

Psychotic symptoms may occur in patients with epilepsy in an acute, chronic, or alternative form (1). Acute psychotic episodes after a bout of seizures, known as postictal psychosis, have attracted much attention in the last decade (2,3), because they are supposed to have a closer association with epileptogenesis than do other forms of epileptic psychosis. However, the pathogenetic mechanism of postictal psychosis is not well understood.

In most cases, extracranial EEG showed no significant findings during episodes of postictal psychosis (2). The case reported by So et al. (4) was one of the few studies that investigated electrical activities in deep structures involving the limbic system during postictal psychosis, and the intracranial EEG confirmed the acute psychosis to be a postictal event rather than an ictal event.

We report a patient who manifested an acute psychotic episode after a cluster of seizures during intracranial EEG monitoring. There was a close relationship between the onset of psychotic symptoms and epileptic activities in the left amygdala.

CASE REPORT

  1. Top of page
  2. Abstract
  3. CASE REPORT
  4. DISCUSSION
  5. REFERENCES

A 25-year-old right-handed woman had a history of febrile convulsion at age 2 years. At age 5 years, she was hit on the head in a traffic accident and had a convulsion lasting 30 min. An antiepileptic drug (AED) was started. She had several convulsive seizures until age 8 years, after which the seizure disappeared, and AED was discontinued at age 9 years. At age 12 years, complex partial seizures (CPSs) with oral automatism appeared, and AED treatment was restarted. There was no warning sign preceding the CPSs. CPSs continued to occur several times monthly despite active treatment with various AEDs; therefore she was evaluated for surgical therapy.

Neurologic examination was normal, and she had no previous psychiatric illness. Extracranial EEG using scalp and sphenoidal electrodes showed independent interictal bitemporal spikes. Ictal EEG discharges had higher amplitude in the right temporal region. Magnetic resonance imaging (MRI) showed slightly higher intensity at the left hippocampus on T2 and FLAIR images. Because of the dissociation between EEG and the neuroimaging findings, she was recommended to undergo an intracranial EEG study at age 25 years. Language dominance was on the left side by the Wada test.

Intracerebral electrodes for the hippocampus and amygdala, and subdural electrodes covering the basolateral temporal lobe surface and adjacent parietooccipital areas were implanted bilaterally to identify the site of seizure onset. Nineteen CPSs were recorded, three of which were followed by secondary generalization. All the seizures originated from the medial structure of the left temporal lobe (Fig. 1).

image

Figure 1. An intracranial EEG recording of a complex partial seizure with multicontact intracerebral (NA and CA) and subdural electrodes (OF, BA, BP, AT, SLT, MLT, ILT, TB). The seizure discharges originated in the medial structure of the left temporal lobe. OF, orbitofrontal lobe; NA, nucleus amygdalae; CA, cornu ammonis; BA, anterior temporal base; BP, posterior temporal base; AT, anterior temporal lobe; SLT, superior lateral temporal lobe; MLT, middle lateral temporal lobe; ILT, inferior lateral temporal lobe; TB, temporal base (anterior); L, left; R, right.

Download figure to PowerPoint

Figure 2 outlines the clinical course during the period of intracranial EEG recording. Implantation was performed on October 1, 1998, and 24-h video-EEG monitoring was started on October 8. The first seizure was recorded on October 13, which was followed by 17 more in 2 days. The seizure cluster was terminated by rectal administration of diazepam (DZP), 10 mg. Thereafter only one seizure was recorded on October 19.

image

Figure 2. Schematic representation of seizures occurring during the intracranial EEG recording, psychotic symptoms, seizure discharges in the left amygdala (duration of seizure discharges in every 4 h in percentages), and the course of medication. The recording was started on October 8 and terminated on October 21. CPS, complex partial seizure; GTC, secondarily generalized tonic–clonic seizure; CBZ, carbamazepine; PB, phenobarbital; NZP, nitrazepam.

Download figure to PowerPoint

She showed no remarkable psychiatric symptoms before and immediately after the seizure cluster. On October 15, however, she appeared elated in mood and behaved queerly, as if she were in another world. At night her facial expression became rigid, and she was engaged in a monologue. She said she heard voices of her parents, although they were not present. Haloperidol, 1.5 mg, and levomepromazine, 25 mg, were administrated that night. In the morning of October 16, she was found stuporous. She was mute, even to our questions. Haloperidol, 4.5 mg/day, had been administered since October 16. On October 17 and 18, her responses and contact with staff gradually improved. Brotizolam, 0.25–0.5 mg/night, was used as a hypnotic drug. On October 19, she became restless and frequently expressed anxiety and fear. She could not stay alone in the monitoring room and wandered around on the same floor. That night, DZP, 5 mg, was orally administered for the purpose of improving the repeated fear, and flunitrazepam, 4 mg, was used instead of brotizolam as a hypnotic drug. After October 20, she showed no apparent psychiatric symptoms, and the EEG monitoring was terminated on October 21.

Interictal EEG on October 12 (before the seizure cluster) showed spike activities in the medial temporal lobe structures predominantly on the left side (Fig. 3). After the seizure cluster, seizure discharges restricted to the left amygdala appeared and increased in frequency and duration (Fig. 4). During this period, she felt estranged from the outer world and had auditory hallucinations. On October 16, the seizure discharges in the left amygdala became almost continuous when she was stuporous. After October 17, the seizure discharges decreased in duration, and her behavior generally improved. On October 19, rhythmic spike-and-slow-wave discharges observed in the left amygdala coincided with feelings of anxiety, restlessness, and fear of the patient (Fig. 5). On October 21, when the psychosis had already improved, the EEG showed spike activities in the medial temporal lobe structures predominantly on the left side, as before the seizure cluster (Fig. 6).

image

Figure 3. Intracranial EEG recording before the psychotic episode, indicated by Fig. 3 in Fig. 2. Interictal discharges predominate in the medial structure of the left temporal lobe. Abbreviations are as in Fig. 1.

Download figure to PowerPoint

image

Figure 4. Intracranial EEG recording during the psychotic episode, indicated by Fig. 4 in Fig. 2. Frequent seizure discharges are restricted to the left amygdala. Abbreviations are as in Fig. 1.

Download figure to PowerPoint

image

Figure 5. Rhythmic spike-and-slow-wave discharges in the left amygdala. The patient looked anxious and restless, and expressed fear during the discharges. Figure 5 in Fig. 2 indicates the point at which the recording was made. Abbreviations are as in Fig. 1.

Download figure to PowerPoint

image

Figure 6. Intracranial EEG recording after the psychosis had improved, indicated by Fig. 6 in Fig. 2. Interictal discharges predominate in the medial structure of the left temporal lobe, just as in Fig. 3 (before the psychosis). Abbreviations are as in Fig. 1.

Download figure to PowerPoint

As the interictal recordings were saved automatically for 30 s every 5 min, we counted the seizure discharges in the left amygdala in the recordings thus saved. Seizure discharges were defined here as rhythmic waves, faster than 10 Hz, with amplitudes >200 μV, and duration >2 s. The duration of seizure discharges in every 4 h is indicated in percentages in Fig. 2. Before the seizure cluster, there were no seizure discharges. After the seizure cluster, the discharges appeared and increased, reaching the peak of frequency on October 16, when also the psychosis was apparently at its peak. Thereafter the frequency of the seizure discharges diminished when the psychosis got better.

The patient underwent selective left amygdalohippocampectomy on October 22, and since then has remained free of psychiatric problems. She had two CPSs at the second postoperative month, but has remained seizure free thereafter for >1 year.

DISCUSSION

  1. Top of page
  2. Abstract
  3. CASE REPORT
  4. DISCUSSION
  5. REFERENCES

This patient showed an acute psychotic episode after the occurrence of a seizure cluster during intracranial EEG monitoring. The psychotic symptoms changed rapidly and improved within a few days. The transition of psychotic symptoms corresponded to the increase or decrease of the seizure discharges in the left amygdala: psychotic symptoms appeared as the seizure discharges increased in frequency and duration, and the psychotic state reached a peak when the seizure discharges became almost continuous. After the peak of psychosis, the seizure discharges gradually decreased.

Morphology of the seizure discharges in the left amygdala also changed with transition of the psychotic symptoms. Rhythmic spike activities were prominent at the peak of psychosis. Thereafter, as the psychosis improved, rhythmic spike-and-slow-wave discharges became dominant instead. The alteration in morphology of the seizure discharges in the left amygdala might reflect the intensity of influence of the epileptic activity on brain function and the area of influence of the discharges.

In our case, the onset of psychotic symptoms occurred after the seizure cluster ended and consciousness had improved. Therefore, the psychotic episode observed in this patient represents a postictal psychosis from the clinical point of view (2,3). In most cases reported, postictal psychoses were preceded by cessation or reduction of AEDs (5,6), although the cases reported by Logsdail and Toone (2) had postictal psychoses under stable medication. So et al. (4) suggested that AED discontinuation may have a contributory role but only insofar as it promotes withdrawal seizures. In our patient, AED reduction contributed to the development of the psychotic symptoms, because the seizure cluster occurred after AED was reduced.

Only a few cases of periictal psychosis have been documented with intracranial electrodes. So et al. (4) reported a patient who manifested a postictal psychotic episode during intracranial EEG monitoring. In their case, intracranial EEG recordings confirmed that the psychosis is a postictal event rather than an ictal event. In our patient, repeated seizure discharges were observed in the left amygdala during the psychotic episode. Therefore, it may be suggested that she had an ictal psychosis rather than a postictal psychosis. In this respect, our patient resembles the case reported by Wieser (7), with prolonged anxiety, visual and auditory hallucinations, and dreamy recall during right temporal lobe status epilepticus documented by intracranial EEG electrodes.

Fear is a common affective manifestation of seizures originating in the temporal lobe. It usually has a sudden onset and may range from mild anxiety to intense terror. The amygdala has been suggested to play an important role in the experience and expression of fear. Fear can be produced in humans by stimulating the amygdala or its immediate vicinity. Sometimes fear occurs only when a seizure discharge arising elsewhere spreads to involve the amygdala (8). After the peak of the psychotic episode, our patient expressed fear corresponding to characteristic seizure discharges restricted to the left amygdala, and fear ceased with termination of the discharges. This patient had never experienced affective symptoms as an ictal event. This de novo affective symptom may underlie the psychotic symptoms biologically and/or psychologically. Increase in intensity of the seizure discharges in the amygdala may influence the brain function producing mood alteration in favor of fear, and ultimately evolve to psychotic symptoms. When the seizure discharges in the amygdala became less intense, the patient was then able to recognize and express fear as an ictal experience.

Studies other than EEG provided some evidence that hyperfunction may be involved in the pathogenesis of psychosis. A single-photon emission computed tomography (SPECT) study conducted by Jibiki et al. (9) reported two cases of periictal psychosis showing hyperperfusion in the temporolimbic system during psychotic states. Another SPECT study of four patients with temporal lobe epilepsy demonstrated medial frontal hyperperfusion during psychosis (10). We did not perform SPECT in our patient, but we may expect an altered SPECT image in accordance with the seizure discharges in the left amygdala.

In conclusion, we encountered a patient who showed an acute psychotic episode during intracranial EEG monitoring. There was a close relationship between the psychotic symptoms and seizure discharges in the medial structure of the left temporal lobe, especially in the amygdala. Hyperactivity of the medial temporal lobe may be the pathogenesis of acute psychosis associated with epilepsy.

REFERENCES

  1. Top of page
  2. Abstract
  3. CASE REPORT
  4. DISCUSSION
  5. REFERENCES
  • 1
    Sachdev P. Schizophrenia-like psychosis and epilepsy: the status of the association. Am J Psychiatry 1998;155: 32536.
  • 2
    Logsdail SJ & Toone BK. Post-ictal psychoses: a clinical and phenomenological description. Br J Psychiatry 1988;152: 24652.
  • 3
    Savard G, Andermann F, Olivier A, Rémillard GM. Postictal psychosis after partial complex seizures: a multiple case study. Epilepsia 1991;32: 22531.
  • 4
    So NK, Savard G, Andermann F, et al. Acute postictal psychosis: a stereo EEG study. Epilepsia 1990;31: 18893.
  • 5
    Demers-Desrosiers LA, Nestoros JN, Vaillancourt P. Acute psychosis precipitated by withdrawal of anticonvulsant medication. Am J Psychiatry 1978;135: 9812.
  • 6
    Sironi VA, Franzini A, Ravagnati L, Marossero F. Interictal acute psychoses in temporal lobe epilepsy during withdrawal of anticonvulsant therapy. J Neurol Neurosurg Psychiatry 1979;42: 72430.
  • 7
    Wieser HG. Temporal lobe or psychomotor status epilepticus: a case report. Electroencephalogr Clin Neurophysiol 1980;48: 55872.
  • 8
    Gloor P. Role of the amygdala in temporal lobe epilepsy. In: AggletonJP, ed. The amygdala: neurobiological aspects of emotion, memory, and mental dysfunction. New York: Wiley-Liss, 1992: 50538.
  • 9
    Jibiki I, Maeda T, Kubota T, Yamaguchi N. 123I-IMP SPECT brain imaging in epileptic psychosis: a study of two cases of temporal lobe epilepsy with schizophrenia-like syndrome. Neuropsychobiology 1993;28: 20711.
  • 10
    Baumgartner C, Podreka I, Benda N, et al. Postictal psychosis: a SPECT study [Abstract]. Epilepsia 1995;36(suppl 3):S218.