Interictal SPECT in patients with mesial temporal lobe epilepsy and psychosis: a case-control study

Interictal SPECT in patients with mesial temporal lobe epilepsy and psychosis: a case-control study

Psychiatry Research: Neuroimaging 138 (2005) 75 – 84 Interictal SPECT in patients with mesial temporal lobe epilep...

202KB Sizes 0 Downloads 30 Views

Psychiatry Research: Neuroimaging 138 (2005) 75 – 84

Interictal SPECT in patients with mesial temporal lobe epilepsy and psychosis: a case-control study Ricardo Guarnieria,*, Lauro Wichert-Anaa,b, Jaime E.C. Hallakc,d, Tonicarlo R. Velascoa,c, Roger Walza,c, Mery Katob, Veriano Alexandre Jr.a,c, Vera C. Terra-Bustamantea,c, Marino M. Bianchina,c, Antonio W. Zuardic, John F.W. Deakind, Ame´rico C. Sakamotoa,c a

Center for Epilepsy Surgery-CIREP, Ribeira˜o Preto Medical School, University of Sa˜o Paulo, Av. Bandeirantes, 3900, CEP 14.048-900, Ribeira˜o Preto, SP, Brazil b Department of Nuclear Medicine, Ribeira˜o Preto Medical School, University of Sa˜o Paulo, Brazil c Department of Psychiatry, Neurology and Psychology, Ribeira˜o Preto Medical School, University of Sa˜o Paulo, Brazil d University of Manchester, Manchester, United Kingdom Received 17 August 2004; accepted 22 October 2004

Abstract Psychosis is commonly observed in patients with mesial temporal lobe epilepsy related to hippocampal sclerosis (MTLEHS). Interictal single photon emission computed tomography (SPECT) was performed to compare regional cerebral blood flow (rCBF) pattern of MTLE-HS patients with psychosis of epilepsy (POE) comorbidity and MTLE-HS patients without any psychiatric disorders (Control group). For this, 21 patients with POE and 23 Control patients were matched by educational level, clinical, demographic, electrophysiological, and MRI data. SPECT scans were acquired using 99mTc and interpreted with a semiquantitative method. We analyzed brain regions of interest (ROI) of frontal, temporal, and parietal cortex, in addition to subcortical structures. There were no significant statistical differences of ROI between the POE group and the Control group after Bonferroni adjustment. However, we observed a trend for rCBF increase of right posterior cingulate in the POE Group. This increase would be in accordance with recent findings of cingulate abnormalities in schizophrenia, suggesting that abnormal function in this region might be associated with the psychotic phenomena. D 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Psychotic disorder; Single photon emission computed tomography; Schizophrenia; Subcortical circuits; Cingulate gyrus; Neurologic disorders

* Corresponding author. Tel.: +55 16 602 2613; fax: +55 16 633 0760. E-mail address: [email protected] (R. Guarnieri). 0925-4927/$ - see front matter D 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pscychresns.2004.10.003


R. Guarnieri et al. / Psychiatry Research: Neuroimaging 138 (2005) 75–84

1. Introduction

2. Patients and methods

Psychiatric disorders are frequently encountered in patients with epilepsy. It is accepted that 30–50% of the epileptic population has psychiatric comorbidity (Torta and Keller, 1999). Of all psychiatric disorders associated with epilepsy, few require more attention and care than the psychosis of epilepsy (POE), which is not uncommon, with incidences ranging from 7% to 27%, depending upon the series reported (Devinsky and Vazquez, 1993). POE has been classified into four main patterns: ictal POE, postictal POE, interictal POE and alternative POE (Sachdev, 1998; Devinsky, 2003). Although postictal POE and interictal POE have their onsets in different time frames in relation to the epileptic seizures, both types of psychosis share a common phenomenology, characterized by affective symptoms, delusions, and auditory hallucinations. Furthermore, there is a possibility that recurrent postictal POE could be a risk factor for the development of interictal POE (Tarulli et al., 2001). Functional neuroimaging has been extensively used to investigate psychiatric disorders. Regarding schizophrenia, the most consistent observation has been a decrease in regional cerebral blood flow (rCBF) and metabolism in the the prefrontal area. These findings have been reported in at least two thirds of studies (Wu et al., 2000). Divergent results may possibly be due to methodological bias, including small samples, inconsistent cerebral activation depending upon the neuropsychological task (Ebmeier et al., 1995), and effects of neuroleptic medication (Miller et al., 1997). Furthermore, the wide spectrum of schizophrenic symptoms may also influence the results obtained (Sabri et al., 1997). Previous SPECT studies in individuals with POE have shown perfusional abnormalities in the left temporal area, suggesting a dysfunction in the temporal–limbic region (Jibiki et al., 1993; Marshall et al., 1993; Mellers et al., 1998). The aim of the present investigation was to analyze the interictal perfusion patterns of individuals with mesial temporal lobe epilepsy related to hippocampal sclerosis (MTLE-HS) associated with interictal POE or postictal POE, and to compare these perfusion patterns to those from individuals with MTLE-HS but without any psychiatric comorbidity.

2.1. Patients Twenty-one consecutive adult MTLE-HS patients with interictal or postictal POE evaluated at the Ribeira˜o Preto Epilepsy Surgery Program between January 1997 and July 2001 were included. As the Control group, we selected 23 adult MTLE-HS patients without any previous history of psychiatric disorder and seen during the same time interval, matched for gender, age, educational level, handedness, and laterality of the epileptogenic focus. Presurgical evaluation was performed by an experienced multidisciplinary team and included a detailed clinical history and neurological examination, interictal and ictal video-EEG analysis, structural and functional imaging, psychiatric evaluation, neuropsychological testing, and, whenever appropriate, the intracarotid amobarbital test (Velasco et al., 2002; Wichert-Ana et al., 2001). The characterization of MTLE-HS was according to Engel’s criteria (Engel, 1996): (1) seizure semiology consistent with MTLE, usually with epigastric, autonomic or psychic auras, followed by behavioral arrest, progressive clouding of consciousness, oroalimentary and manual automatisms, and autonomic phenomena; (2) anterior and mesial temporal interictal spikes; (3) no lesion other than uni- or bilateral hippocampal atrophy, and increased signal in the hippocampal formation on high resolution MRI (1.5 T); (4) histopathological examination compatible with hippocampal sclerosis (HS); (5) absence of dual pathology that could be identified by any of the available methods (clinical, electrophysiological, neuroimaging and histopathological); (6) presurgical investigation compatible with seizure onset zone localized in the temporal lobe. Exclusion criteria for MTLE-HS were as follows (Engel, 1996): (1) focal neurological abnormalities on physical examination; (2) generalized or extratemporal EEG spikes; and (3) marked cognitive impairment (IQ lower than 70). Psychiatric diagnosis was independently established during the presurgical evaluation by two psychiatrists with experience in psychiatric disorders associated with epilepsy (RG and JECH), and according to the Diagnostic and Statistical Manual

R. Guarnieri et al. / Psychiatry Research: Neuroimaging 138 (2005) 75–84

of Mental Disorders, 4th edition. Patients with interictal or postictal POE had no psychotic disorder associated with ictal POE, changes of antiepileptic medications, epileptic status, delirium, and psychosis for paradoxical normalization. The postictal POE group (n=10) was defined by psychotic states characterized by hallucinations, delusions or thought disorders, sometimes associated with mood changes, and occurring within 24 h to 7 days after the last epileptic seizure. During the psychotic state (ranging from a few days to 90 days), conciousness was fully preserved (Logsdail and Toone, 1988). The interictal POE (n=11) group was defined by prolonged psychotic state not related to the epileptic seizures (Sachdev, 1998). Patients with POE and Controls had no history of previous psychiatric disorders or substance dependence. All patients were referred for presurgical assessment due to seizures refractory to pharmacological treatment. All patients had their antiepileptic medications regimens kept unchanged during their SPECT scan period. The clinical and demographic characteristics of all patients included gender, age, handedness, years of education, marital and employment status, age at epilepsy onset (recurrent seizures), epilepsy duration, history of an initial precipitating insult (IPI), epilepsy duration until the psychiatric evaluation, complex partial seizure monthly frequency in the year before surgery, MRI findings, side of the ictal EEG onset, and MRI findings. Additional data analyzed in POE patients included age at psychosis onset, duration of psychosis, presence of psychotic symptoms, and use of neuroleptic medication during SPECT acquisition. The Ethics Committee of our institution approved the study, and informed consent was obtained from all patients. 2.2. SPECT methodology All patients had interictal SPECT scans performed as previously described by our group (Wichert-Ana et al., 2001). The radiotracer ethyl cysteinate dimer (ECD) was labeled with 99mtechnetium, at a maximum dose of 1295 MBq (35 mCi), and was injected in the interictal state, separated at least for 24 h from the last seizure. All subjects were instructed to remain at rest, with their eyes open, in a quiet and dark room,


and were required to refrain from talking and listening. The time between the intravenous injection of 99mtechnetium and the beginning of the SPECT scan was 30 min. SPECT scans were carried out in a single-headed rotating gamma camera (SiemensR Orbiter) with a low-energy high-resolution collimator (LEHR). Sixtyfour projections were recorded over 3608 on a 6464 matrix, with acquisition time of 30 min, and 70,000 counts/frame. Projections were filtered with a Butterworth filter (order 4, cutoff frequency 0.25). Scans were reconstructed with a SOPHYR NXT workstation in transaxial slices parallel to a line drawn from the base of the frontal lobe to the occipital lobe (orbitomeatal line—OM), and parallel to the long axis of the temporal lobe, from which transverse sections were produced. Attenuation correction was performed using the first-order method of Chang (coefficient 0.12/cm) and a pixel size of 6.338 mm. Images were blindly analyzed by two experienced nuclear medicine specialists (MK and LWA). Semiquantitative analysis was performed on three SPECT slices, including those brain regions more commonly reported to be involved in schizophrenia and POE (Miller et al., 1997; Sachdev, 1998; Wu et al., 2000). More specifically, the brain regions of interest included the frontal, temporal, and subcortical areas. We also analyzed other less commonly investigated areas including the parietal and occipital regions. The anatomical reference points were set on the basis of MRI (Talairach and Tournoux, 1988) and SPECT (Van Heertum and Tikofsky, 1989) atlases of neuroanatomy. Regions of interest (ROI) were outlined by hand, drawing around the 60% isocontour of the brain to define the outer edge of the cortex, as illustrated in Fig. 1. The first two slices were obtained on the orbitomeatal transaxial orientation plane. The first one comprised the frontal pole and basal ganglia (Fig. 1A) and the second one, on a higher plane, the prefrontal, anterior parietal and posterior cingulate cortex (Fig. 1B). Finally, a third slice was obtained along the long plane of the temporal lobes (Fig. 1C). Semiquantitative values for each ROI were computed as mean counts per pixel (count density), normalized to the cerebellum (Fig. 1D). Thus, the relative tracer uptake within each ROI was expressed as the ratio of count density in each ROI divided by cerebellum count density.


R. Guarnieri et al. / Psychiatry Research: Neuroimaging 138 (2005) 75–84

Fig. 1. Slices in four planes showing the hand-drawn ROI (colored contours) for all 44 subjects analyzed: (A) lower OM line: 1—frontal pole; 2—low prefrontal; 3—basal ganglia; 4—thalamus; 5—occipitotemporal convergence; (B) higher OM line: 1—high prefontal; 2—anterior cingulate; 3—posterior cingulate; 4—high parietal; (C) temporal lobe: 1—anterior temporal; 2—lateral temporal; 3—posterior temporal; 4— inner temporal; 5—cuneus; (D) cerebellum.

Statistical analyses were performed by means of the Statistical Package for the Social Sciences (SPSS for WINDOWS software), version 9.0 (SPSS, Chicago, IL). Numerical data were tested with the Kolmogorov–Smirnov Z-test. Continuous data with normal distribution were analyzed by the two-tailed Student’s test (t) and univariate analysis of variance (ANOVA). Non-normal data were analyzed with the Mann–Whitney U-test. Quantitative data were compared by the chi-square test (v 2) and Fisher’s exact test ( F). A multivariate analysis was carried out to correlate demographic features of the groups with SPECT findings. The level of significance was set at Pb0.05. The Bonferroni test was applied to correct for multiple comparisons.

3. Results The clinical and demographic distributions of all subgroups of patients are shown on Table 1. No significant differences were found among postictal POE, interictal POE and Controls in terms of gender, age, handedness, years of education, marital and employment status, age at epilepsy onset (recurrent seizures), epilepsy duration, positive IPI (initial precipitating injury) history, epilepsy duration until the psychiatric evaluation, monthly complex partial seizure frequency in the year before surgery, side of ictal EEG onset, and MRI findings. Although all patients were on antiepileptic medications during the SPECT, there were no significant differences when

R. Guarnieri et al. / Psychiatry Research: Neuroimaging 138 (2005) 75–84


Table 1 Demographic and clinical characteristics of patients with MTLE-HS without psychopathology (MTLE-HS-Controls), MTLE-HS with psychosis (POE), and the types of psychosis: interictal and postictal Variable

Gender Male Female Age, m (S.D.)a Education years, m (S.D.) Marital status Single or divorced Married Remunerated occupation (yes/no) Epilepsy Age at onset, mean, m (S.D.) Duration, m (S.D.) Seizure frequency/month, m (S.D.) IPI (positive/negative)b Anticonvulsant drugs on SPECT Phenytoin Carbamazepine Phenobarbital and other drugs Resonance laterality Right Left Bilateral EEG ictal laterality Right Left Bilateral Handedness (right/left) a b

MTLE-HS with psychosis


Total POEControls Test value


P value

v 2=0.349



t=0.628 t=0.102 v 2=1.485

42 42 2

0.53 0.91 0.47




t=0.046 t=0.480 U=214.0 v 2=0.02 v 2=2.381

42 42 z=0.6 1 2

0.96 0.63 0.51 0.60 0.30

v 2=0.380



v 2=0.509



v 2=0.934



Postictal (n=10)

Interictal (n=11)

Total POE (n=21)

Controls (n=23)

7 3 36.7 (5.1) 5.4 (2.4)

3 8 38.5 (8.6) 6.4 (4.1)

10 11 37.6 (7.0) 5.9 (3.4)

13 10 36.4 (5.5) 6.0 (2.8)

5 5 3/7

9 2 3/8

14 7 6/15

13 10 12/11

10.2 (5.3) 26.5 (5.8) 8.7 (5.3) 4/6

11.7 (5.2) 26.7 (8.7) 7.6 (5.3) 8/3

11.0 (5.2) 26.6 (7.3) 8.14 (5.2) 12/09

10.9 (7.1) 25.5 (7.8) 10.6 (13.4) 13/10

7 3 0

5 1 5

12 4 5

11 9 3

4 3 3

6 5 0

10 8 3

11 10 2

5 3 2 10/0

6 5 0 11/0

11 8 2 21/0

12 10 1 22/1

m (S.D.)=mean and standard deviation. IPI=initial precipitating injury.

ANOVA was carried out controlling for antiepileptic drugs (phenytoin, carbamazepine and phenobarbital and other drugs) for each group—postictal POE,

interictal POE, and Control. Furthermore, there were no differences between interictal and postictal POE patients regarding the age at psychosis onset, duration

Table 2 Demographic and clinical characteristics of patients with MTLE-HS without psychopathology (MTLE-HS-Controls), MTLE-HS with psychosis (POE), and the types of psychosis: interictal and postictal Variable

Age at onset of psychosis, m (S.D.)a Duration of psychosis, m (S.D.) Psychotic symptoms on SPECT (yes/no) Neuroleptic medication on SPECT Haloperidol Without neuroleptic medication a

m (S.D.)=mean and standard deviation.

Postictal POEInterictal POE

MTLE-HS with psychosis Postictal (n=10)

Interictal (n=11)

Total POE (n=21)

Test value


P value

31.8 (4.4) 4.9 (4.4) 1/9

29.5 (8.1) 8.9 (9.5) 6/5

30.6 (6.5) 7.0 (7.6) 7/14

t=0.782 t=1.219 F=4.67 v 2=1.530

19 19 1 1

0.44 0.23 0.06 0.36

2 8

5 6

7 14


R. Guarnieri et al. / Psychiatry Research: Neuroimaging 138 (2005) 75–84

Table 3 Group means (S.D.) of normalized count rates on SPECT of patients with MTLE-HS without psychopathology (MTLE-HS-Controls), MTLEHS with psychosis (POE), and the types of psychosis: interictal and postictal Variable

MTLE-HS with psychosis


Total POEControls*

Postictal (n=10)

Interictal (n=11)

Total POE (n=21)

Controls (n=23)

t test

P value

0.948 (0.07) 0.960 (0.08)

0.968 (0.08) 0.994 (0.09)

0.958 (0.07) 0.977 (0.09)

0.943 (0.10) 0.964 (0.11)

0.58 0.45

0.567 0.652

High prefrontal Right 0.945 (0.04) Left 0.941 (0.06)

0.939 (0.09) 0.947 (0.09)

0.942 (0.07) 0.944 (0.07)

0.925 (0.09) 0.926 (0.10)

0.68 0.70

0.500 0.486

Low prefrontal Right 0.915 (0.05) Left 0.905 (0.06)

0.926 (0.08) 0.933 (0.09)

0.920 (0.07) 0.920 (0.08)

0.921 (0.09) 0.909 (0.10)

0.02 0.41

0.984 0.687

Anterior cingulate Right 0.931 (0.07) Left 0.999 (0.07)

0.984 (0.07) 1.030 (0.10)

0.958 (0.07) 1.015 (0.09)

0.942 (0.10) 0.979 (0.11)

0.61 1.2

0.535 0.245

Posterior cingulate Right 1.017 (0.07) Left 1.050 (0.08)

1.032 (0.09) 1.067 (0.09)

1.025 (0.08) 1.058 (0.08)

0.960 (0.11) 1.010 (0.12)

2.2 1.5

0.034** 0.152

Basal ganglia Right Left

1.092 (0.07) 1.069 (0.07)

1.108 (0.09) 1.102 (0.10)

1.100 (0.08) 1.086 (0.09)

1.087 (0.11) 1.062 (0.12)

0.45 0.76

0.659 0.454

Thalamus Right Left

0.989 (0.08) 1.006 (0.09)

1.072 (0.13) 1.038 (0.12)

1.032 (0.11) 1.022 (0.11)

1.023 (0.15) 1.026 (0.12)

0.23 0.11

0.822 0.916

High parietal Right Left

0.960 (0.05) 0.945 (0.05)

0.960 (0.09) 0.984 (0.10)

0.953 (0.07) 0.966 (0.08)

0.921 (0.10) 0.939 (0.10)

1.2 0.96

0.226 0.343

Anterior temporal Right 0.816 (0.03) Left 0.792 (0.06)

0.855 (0.07) 0.806 (0.07)

0.837 (0.06) 0.800 (0.06)

0.810 (0.09) 0.787 (0.11)

1.1 0.46

0.258 0.650

Inner temporal Right 0.809 (0.08) Left 0.813 (0.09)

0.880 (0.10) 0.875 (0.09)

0.846 (0.09) 0.846 (0.10)

0.838 (0.11) 0.818 (0.11)

0.28 0.86

0.781 0.394

Lateral temporal Right 0.941 (0.06) Left 0.935 (0.06)

0.981 (0.08) 0.966 (0.10)

0.962 (0.08) 0.951 (0.08)

0.948 (0.09) 0.935 (0.10)

0.54 0.61

0.590 0.549

Posterior temporal Right 0.938 (0.06) Left 0.931 (0.07)

0.943 (0.07) 0.991 (0.10)

0.940 (0.06) 0.962 (0.09)

0.929 (0.10) 0.946 (0.11)

0.48 0.55

0.631 0.585

OTC a Right Left

0.960 (0.08) 0.987 (0.10)

0.955 (0.07) 0.962 (0.09)

0.927 (0.10) 0.948 (0.11)

1.1 0.45

0.266 0.654

Frontal pole Right Left

0.950 (0.06) 0.934 (0.07)

R. Guarnieri et al. / Psychiatry Research: Neuroimaging 138 (2005) 75–84


Table 3 (continued) Variable

Cuneus Right Left

MTLE-HS with psychosis


Total POEControls*

Postictal (n=10)

Interictal (n=11)

Total POE (n=21)

Controls (n=23)

t test

P value

1.148 (0.09) 1.153 (0.10)

1.181 (0.11) 1.182 (0.10)

1.166 (0.10) 1.168 (0.10)

1.168 (0.15) 1.162 (0.15)

0.06 0.12

0.955 0.909


OTC=occipitotemporal convergence. * Df = 42. ** Significant difference between POE and Control (t test). After Bonferroni adjustment, this difference was not maintained.

of psychosis, or type of neuroleptic medication used during SPECT acquisition (Table 2). Regarding the possible influence of neuroleptic medications on SPECT, two postictal POE and five interictal POE patients were taking haloperidol. We performed separate ANOVA in the psychotic patients on neuroleptic medications and not being treated with neuroleptics at the time of the SPECT. We compared postictal POE, interictal POE, and Control groups, and no significant differences were found. There was a trend toward a higher prevalence of psychotic symptoms in the interictal POE patients compared with the postictal POE patients ( F=4.67; df=1; P=0.06). The averages of the normalized counts at the ROI (representing rCBF) are presented in Table 3. Regarding SPECT findings, no correlation between gender and other demographic variables with SPECT findings emerged. We performed a three-way ANOVA with postictal POE, interictal POE, and Control as an independent group dimension and temporal lobe region (anterior, inner, posterior and lateral) and hemispherical sides (right and left) as repeated measures, but we also did not find any significant statistical difference. Hippocampal and anterior temporal ROI for atrophy were correlated on right and left sides separately, and no significant correlation was found. In addition, we tested the temporal lobe ROI (anterior, inner, posterior, and lateral) in the postictal POE group and the interictal POE group versus the Control group (ANOVA), but no significant statistical difference was found. When the total POE group was compared with the Control group, a significant rCBF increase was observed in the right posterior cingulate cortex (t=2.2; df=42; P=0.034). However, after the Bonferroni adjustment was applied, we had to lower the alpha for each test to 0.002 to bring the overall alpha level back to 0.05. No other rCBF differences were found when the POE group was compared with

the Control group for the other analyzed regions. Also, there were no ROI differences when we compared postictal and interictal POE subgroups with the Control group.

4. Discussion In this study, we specifically addressed perfusional changes in a homogeneous and well-controlled group of MTLE-HS patients with and without psychotic symptoms. However, contrary to previous interictal SPECT reports, we did not find any significant rCBF changes in the POE group compared with the Control group, after adjustments to correct for multiple comparisons. Marshall et al. (1993) previously analyzed interictal SPECT and observed left mesial temporal hypoperfusion in a group of five individuals with interictal POE in comparison to five other epileptic patients without psychiatric comorbidity. Mellers et al. (1998) performed interictal SPECT during a verbal fluency test and observed left superior temporal hypoperfusion in 12 POE subjects compared with 16 epileptic patients without psychiatric disorders and 11 individuals with schizophrenia. In the same study, they also found during the verbal fluency test an increased rCBF in the anterior cingulate cortex of the schizophrenic patients. In another SPECT study, Jibiki et al. (1993) observed left temporal hyperperfusion in two MTLE-HS patients with acute interictal POE. Taken together, these observations suggest a dominant hemisphere dysfunction in patients with interictal POE, which is in line with previous EEG studies that have shown a positive correlation between POE and dominant temporal lobe epilepsy (Flor-Henry, 1969; Perez and Trimble, 1980). However, other authors failed to observe any association between POE and laterality


R. Guarnieri et al. / Psychiatry Research: Neuroimaging 138 (2005) 75–84

of the epileptic focus (Ring et al., 1994; Fong et al., 2000) and this issue, therefore, remains controversial. Although our finding of right posterior cingulate hyperperfusion in the POE group before Bonferroni adjustments could simply reflect a type I error, other similar reports from the literature would support our findings. Functional neuroimaging studies revealing cingulate gyrus involvement during psychotic states have been reported in individuals with schizophrenia. Liddle et al. (1992) in a group of 30 individuals with schizophrenia observed a negative correlation between the analysis factor breality distortionQ (consisting of delusions and hallucinations) and the rCBF in the posterior cingulate cortex, posterior temporal region, head of caudate, and adjacent supramarginal gyrus of the right hemisphere. Haznedar et al. (1997), in a positron emission tomography (PET) study, observed a decrease of glucose metabolism in the anterior cingulate gyrus and an increase in the posterior cingulate gyrus of 50 male individuals with schizophrenia taking no medications. These authors attributed the differences in the perfusion pattern between the anterior and posterior cingulate regions to a possible decrease of the cellular activity in the frontal areas, including a decrement of the GABAergic inhibitory tonus cells of the anterior cingulate, with consequent disinhibition and increased activity of the posterior brain areas. Elkashef et al. (2000), also using PET scanning, reported a significant increase in 18F-DOPA—an l-DOPA analogue neurotransmitter—uptake in the posterior cingulate cortex of nine subjects with schizophrenia taking no medications. A decrease in endogenous dopamine could possibly be related to an increase in the activity of the aromatic l-amino acid decarboxylase enzyme (AADC), provoking larger 18F-DOPA uptake in the posterior cingulate cortex. Finally, Potkin et al. (2002), in another PET study, compared a group of 14 schizophrenic subjects (7 with negative symptoms and 7 with positive symptoms) with seven healthy control volunteers and found an increased metabolism in the right posterior cingulate, in the dorsal prefrontal cortex (Brodmann’s area 9), frontal eye fields (Brodmann’s area 8), motor and premotor cortices (Brodmann’s area 4, 6), supramarginal/ angular gyrus (Brodmann’s area 40), and posterior cingulate (Brodmann’s area 23) of the right hemisphere. Interestingly, when Potkin et al. (2002)

compared the two schizophrenic subgroups with positive and negative symptoms, they found an increased metabolism only in the first subgroup. In spite of methodological differences between PET and SPECT, the finding of increased cerebral metabolism in the posterior cingulate area described by Potkin et al. (2002) parallels the main finding of our study, thus suggesting that the posterior cingulate area may be associated with the modulation of psychotic symptoms in patients with POE and schizophrenia. Neuroimaging abnormalities in the cingulate cortex of patients with schizophrenia are in good agreement with postmortem findings that show decreases of neuronal and glial cell density, inhibitory interneuron number (Benes, 1993), glutamatergic neurons (Squires et al., 1993), and the dopamine metabolite 3,4-dihydroxyphenylacetic acid (Wyatt et al., 1995) in the anterior cingulate cortex. Abnormalities in the anterior cingulate support the blimbic system hypothesis of psychosesQ proposed by Tamminga et al. (2000). According to this hypothesis, during schizophrenia, primary abnormalities in hippocampal function, associated with reduced NMDAmediated glutamatergic hippocampal efferent pathways, may affect all regions of the Papez circuit, including the mammillary bodies, anterior thalamus, anterior cingulate cortex, and several other regions of the neocortex. In healthy humans, antagonistic NMDA receptors such as phencyclidine (PCP) and ketamine are capable of inducing psychotic symptoms similar to those observed in subjects with schizophrenia. These substances have been used as experimental models of psychosis. In patients with schizophrenia, subanesthetic doses of ketamine can produce short and discrete activations of psychotic symptoms that are not different from the real ones. In a PET study, schizophrenic patients had increased metabolism in the anterior cingulate cortex associated with reduction of the metabolism in the hippocampal and primary visual cortex during ketamine administration (Lahti et al., 1995). Other authors have observed induction of c-fos expression in the posterior cingulate and retrosplenial cortex of rats during ketamine injection. Olney et al. (1989) demonstrated that NMDA receptor antagonists, such as MK-801, PCP, tiletamine, and ketamine, caused vacuolar damage in posterior cingulate and retrosplenial neurons when subcutane-

R. Guarnieri et al. / Psychiatry Research: Neuroimaging 138 (2005) 75–84

ously applied to adult rats. Duncan et al. (1998) also reported that subanesthetic doses of ketamine not only induced c-fos expression but also promoted an increase of 2-deoxyglucose uptake in certain limbic regions such as the posterior cingulate and retrosplenial cortices. Taken together, these findings suggest that functional modifications in the cingulate cortex might be related to psychotic symptoms. In conclusion, our study did not find any statistical difference between POE and a non-psychotic epileptic group in the interictal SPECT scan after corrections. However, we observed a statistical trend that is suggestive given its biological plausibility. In fact, we reported a trend of perfusion abnormalities in the limbic system, particularly in right posterior cingulate gyrus. It is also interesting to observe that, despite the use of different methodologies, similar abnormal patterns of limbic system abnormalities have been found in functional imaging studies of individuals with schizophrenia.

Acknowledgments This study was supported by CNPq-Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico (Dr. Guarnieri); Programa Tema´tico Multiinstitucional em Cieˆncia da Computac¸a˜o-PROTEM-CC/ TeleMed/04-2000-CNPq (Drs. Wichert-Ana and Walz); FAPESP-Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo (Drs. Walz and Bianchin); CAPES-Coordenac¸a˜o de Aperfeic¸oamento de Pessoal de Nı´vel Superior CAPES (Dr. Wichert-Ana). We thank Ca´ssio Geraldo dos Reis for statistical review.

References Benes, F.M., 1993. Neurobiological investigations in cingulate cortex of schizophrenic brain. Schizophrenia Bulletin 19, 537 – 549. Devinsky, O., 2003. Psychiatric comorbidity in patients with epilepsy: implications for diagnosis and treatment. Epilepsy & Behavior 4, S2 – S10. Devinsky, O., Vazquez, B., 1993. Behavioral changes associated with epilepsy. Neurologic Clinics 11, 127 – 149. Duncan, G.E., Moy, S.S., Knapp, D.J., Mueller, R.A., Breese, G.R., 1998. Metabolic mapping of the rat brain after subanesthetic doses of ketamine: potential relevance to schizophrenia. Brain Research 787, 181 – 190.


Ebmeier, K.P., Lawrie, S.M., Blackwood, D.H., Johnstone, E.C., Goodwin, G.M., 1995. Hypofrontality revisited: a high resolution single photon emission computed tomography study in schizophrenia. Journal of Neurology, Neurosurgery and Psychiatry 58, 452 – 456. Elkashef, A.M., Doudet, D., Bryant, T., Cohen, R.M., Li, S.H., Wyatt, R.J., 2000. 18F-DOPA PET study in patients with schizophrenia. Positron emission tomography. Psychiatry Research: Neuroimaging 100, 1 – 11. Engel Jr., J., 1996. Surgery for seizures. New England Journal of Medicine 334, 647 – 652. Flor-Henry, P., 1969. Psychosis and temporal lobe epilepsy. A controlled investigation. Epilepsia 10, 363 – 395. Fong, G.C., Fong, K.Y., Mak, W., Tsang, K.L., Chan, K.H., Cheung, R.T., Ho, S.L., 2000. Postictal psychosis related to regional cerebral hyperfusion. Journal of Neurology, Neurosurgery and Psychiatry 68, 100 – 101. Haznedar, M.M., Buchsbaum, M.S., Luu, C., Hazlett, E.A., Siegel Jr., B.V., Lohr, J., Wu, J., Haier, R.J., Bunney Jr., W.E., 1997. Decreased anterior cingulate gyrus metabolic rate in schizophrenia. American Journal of Psychiatry 154, 682 – 684. Jibiki, I., Maeda, T., Kubota, T., Yamaguchi, N., 1993. 123I-IMP SPECT brain imaging in epileptic psychosis: a study of two cases of temporal lobe epilepsy with schizophrenia-like syndrome. Neuropsychobiology 28, 207 – 211. Lahti, A.C., Holcomb, H.H., Medoff, D.R., Tamminga, C.A., 1995. Ketamine activates psychosis and alters limbic blood flow in schizophrenia. NeuroReport 6, 869 – 872. Liddle, P.F., Friston, K.J., Frith, C.D., Hirsch, S.R., Jones, T., Frackowiak, R.S., 1992. Patterns of cerebral blood flow in schizophrenia. British Journal of Psychiatry 160, 179 – 186. Logsdail, S.J., Toone, B.K., 1988. Post-ictal psychoses. A clinical and phenomenological description. British Journal of Psychiatry 152, 246 – 252. Marshall, E.J., Syed, G.M., Fenwick, P.B., Lishman, W.A., 1993. A pilot study of schizophrenia-like psychosis in epilepsy using single-photon emission computerised tomography. British Journal of Psychiatry 163, 32 – 36. Mellers, J.D., Adachi, N., Takei, N., Cluckie, A., Toone, B.K., Lishman, W.A., 1998. SPET study of verbal fluency in schizophrenia and epilepsy. British Journal of Psychiatry 173, 69 – 74. Miller, D.D., Rezai, K., Alliger, R., Andreasen, N.C., 1997. The effect of antipsychotic medication on relative cerebral blood perfusion in schizophrenia: assessment with technetium-99m hexamethyl-propyleneamine oxime single photon emission computed tomography. Biological Psychiatry 41, 550 – 559. Olney, J.W., Labruyere, J., Price, M.T., 1989. Pathological changes induced in cerebrocortical neurons by phencyclidine and related drugs. Science 16;244 (4910), 1360 – 1362. Perez, M.M., Trimble, M.R., 1980. Epileptic psychosis–diagnostic comparison with process schizophrenia. British Journal of Psychiatry 137, 245 – 249. Potkin, S.G., Alva, G., Fleming, K., Anand, R., Keator, D., Carreon, D., Doo, M., Jin, Y., Wu, J.C., Fallon, J.H., 2002. A PET study of the pathophysiology of negative symptoms in


R. Guarnieri et al. / Psychiatry Research: Neuroimaging 138 (2005) 75–84

schizophrenia. Positron emission tomography. American Journal of Psychiatry 159, 227 – 237. Ring, H.A., Trimble, M.R., Costa, D.C., Moriarty, J., Verhoeff, N.P., Ell, P.J., 1994. Striatal dopamine receptor binding in epileptic psychoses. Biological Psychiatry 35, 375 – 380. Sabri, O., Erkwoh, R., Schreckenberger, M., Owega, A., Sass, H., Buell, U., 1997. Correlation of positive symptoms exclusively to hyperperfusion or hypoperfusion of cerebral cortex in nevertreated schizophrenics. Lancet 349, 1735 – 1739. Sachdev, P., 1998. Schizophrenia-like psychosis and epilepsy: the status of the association. American Journal of Psychiatry 155, 325 – 336. Squires, R.F., Lajtha, A., Saederup, E., Palkovits, M., 1993. Reduced [3H]flunitrazepam binding in cingulate cortex and hippocampus of postmortem schizophrenic brains: is selective loss of glutamatergic neurons associated with major psychoses. Neurochemical Research 18, 219 – 223. Talairach, J., Tournoux, P., 1988. Co-planar Sterotaxic Atlas of the Human Brain. Thieme, Stuttgart. Tamminga, C.A., Vogel, M., Gao, X., Lahti, A.C., Holcomb, H.H., 2000. The limbic cortex in schizophrenia: focus on the anterior cingulate. Brain Research Reviews 31, 364 – 370. Tarulli, A., Devinsky, O., Alper, K., 2001. Progression of postictal to interictal psychosis. Epilepsia 42, 1468 – 1471.

Torta, R., Keller, R., 1999. Behavioral, psychotic, and anxiety disorders in epilepsy: etiology, clinical features, and therapeutic implications. Epilepsia 40 (Suppl. 10), S2 – S20. Van Heertum, R.L., Tikofsky, R.S., 1989. Cerebral SPECT Imaging. Raven Press, New York. Velasco, T.R., Wichert-Ana, L., Leite, J.P., Araujo, D., TerraBustamante, V.C., Alexandre Jr., V., Kato, M., Assirati Jr., J.A., Machado, H.R., Carlotti Jr., C.G., Sakamoto, A.C., 2002. Accuracy of ictal SPECT in mesial temporal lobe epilepsy with bilateral interictal spikes. Neurology 59, 266 – 271. Wichert-Ana, L., Velasco, T.R., Terra-Bustamante, V.C., Araujo Jr., D., Junior, V.A., Kato, M., Leite, J.P., Assirati, J.A., MacHado, H.R., Bastos, A.C., Sakamoto, A.C., 2001. Typical and atypical perfusion patterns in periictal SPECT of patients with unilateral temporal lobe epilepsy. Epilepsia 42, 660 – 666. Wu, J.C., Amen, D., Bracha, H.S., 2000. Neuroimaging in clinical pratice. In: Sadock, B.J., Sadock, V.A. (Eds.), Kaplan and Sadock’s Comprehensive Textbook of Psychiatry, 7th ed. Williams and Wilkins, Baltimore, pp. 373 – 385. Wyatt, R.J., Karoum, F., Casanova, M.F., 1995. Decreased DOPAC in the anterior cingulate cortex of individuals with schizophrenia. Biological Psychiatry 38, 4 – 12.