Acta Neurol Belg DOI 10.1007/s13760-014-0415-7

ORIGINAL ARTICLE

Electroencephalographic characteristics of Iranian schizophrenia patients Irman Chaychi • Mohsen Foroughipour • Hossein Haghir • Ali Talaei • Ashkan Chaichi

Received: 5 September 2014 / Accepted: 17 December 2014 Ó Belgian Neurological Society 2015

Abstract Schizophrenia is a prevalent psychiatric disease with heterogeneous causes that is diagnosed based on history and mental status examination. Applied electrophysiology is a non-invasive method to investigate the function of the involved brain areas. In a previously understudied population, we examined acute phase electroencephalography (EEG) records along with pertinent Positive and Negative Syndrome Scale (PANSS) and Mini Mental State Examination (MMSE) scores for each patient. Sixty-four hospitalized patients diagnosed to have schizophrenia in Ebn-e-Sina Hospital were included in this study. PANSS and MMSE were completed and EEG tracings for every patient were I. Chaychi (&) Psychiatry and Behavioral Sciences Research Center (PBSRC), Mashhad University of Medical Sciences (MUMS), Mashhad, Iran e-mail: [email protected] M. Foroughipour Department of Neurology, Ghaem Hospital, Mashhad University of Medical Sciences (MUMS), Mashhad, Iran H. Haghir Department of Anatomy and Cell Biology, School of Medicine, Mashhad University of Medical Sciences (MUMS), Mashhad, Iran H. Haghir Medical Genetic Research Center (MGRC), School of Medicine, Mashhad University of Medical Sciences (MUMS), Mashhad, Iran A. Talaei Psychiatry and Behavioral Sciences Research Center (PBSRC), Journal of Fundamentals of Mental Health (JFMH), Mashhad University of Medical Sciences, Mashhad, Iran A. Chaichi Veterinary Faculty, Bahonar University, Kerman, Iran

recorded. Also, EEG tracings were recorded for 64 matched individuals of the control group. Although the predominant wave pattern in both patients and controls was alpha, theta waves were almost exclusively found in eight (12.5 %) patients with schizophrenia. Pathological waves in schizophrenia patients were exclusively found in the frontal brain region, while identified pathological waves in controls were limited to the temporal region. No specific EEG finding supported laterality in schizophrenia patients. PANSS and MMSE scores were significantly correlated with specific EEG parameters (all P values \0.04). Patients with schizophrenia demonstrate specific EEG patterns and show a clear correlation between EEG parameters and PANSS and MMSE scores. These characteristics are not observed in all patients, which imply that despite an acceptable specificity, they are not applicable for the majority of schizophrenia patients. Any deduction drawn based on EEG and scoring systems is in need of larger studies incorporating more patients and using better functional imaging techniques for the brain. Keywords Schizophrenia
Electroencephalography (EEG)
Positive and Negative Syndrome Scale (PANSS)
Mini Mental State Examination (MMSE)
Iran

Introduction Schizophrenia is a prevalent psychiatric disorder usually beginning before the age of 25 years [1]. Hallucinatory thought is a hallmark of the disease [2]. The etiology of schizophrenia is not clear [3]. The clinical diagnostic category encompasses a group of disorders with heterogeneous causes and somehow similar behavioral symptoms. There is no laboratory test for schizophrenia. The diagnosis is based on psychiatric history and mental status examination [4].

123

Acta Neurol Belg

In the past two decades, pathophysiological roles of certain brain regions have been implicated in schizophrenia. A non-invasive method to study such dysfunctions is applied electrophysiology [5]. Electroencephalography (EEG) studies in schizophrenia patients show abnormal records. Reported abnormalities include increased sensitivity to activation procedures, more epileptiform activity [6], decreased alpha activity [7], increased theta and delta activity and more left-sided abnormalities [8, 9]. These findings were not consistent across different studies probably due to differences in the studied population, progression of disease or other confounding factors [10]. The population of psychiatric patients in Iran is understudied and brain electrophysiological records of schizophrenia patients had never been investigated in Iranian patients before [11]. In this study, we examined the EEG tracing records of Iranian schizophrenia patients, by using the standard (visual) interpretation method, in the acute phase along with each patient’s Positive and Negative Syndrome Scale (PANSS) and Mini Mental State Examination (MMSE) scores. The same procedure was performed in the matched individuals of the control group. These data were compared to identify the distinguishing characteristics in schizophrenia patients.

Materials and methods Patients admitted to the Ebn-e-Sina Psychiatric Hospital in Mashhad (north-eastern part of Iran) in 2010, who were diagnosed with a relapse attack of schizophrenia, were included in this study. All of them were chronic schizophrenia patients with a period of disorder between 3 and 10 years. The early diagnosis of schizophrenia at the time of admission by the psychiatric resident was confirmed by a psychiatrist using a semi-structured interview according to Diagnostic And Statistical Manual of Mental Disorders 4th edition Text Revised (DSM-IV-TR). After gathering demographic data, standard EEG tests (closed eyes, hyperventilation and optical excitation) were done at the EEG department and subsequently evaluated by a neurologist who also filled the EEG forms. (It will be beneficial if two independent neurologists interpret EEGs in future studies.) The main EEG characteristics were dominant wave pattern, dysrhythmia, pathological waves and voltage. Risperidone tablets (3–4 mg/day) were commonly used in their treatment, which have no prominent effect on the recorded EEGs [17]. PANSS and MMSE questionnaires were completed to calculate these scores. Positive sign score of PANSS is calculated from delusion, clarity disorder, hallucinatory behavior, agitation, grandiosity, suspicion and hostility. Negative sign score of PANSS is calculated from blunt affect, agitated compassionate

123

withdrawal, fragile relationships, social withdrawal (passive, inactive), abstract thinking malfunction, lack of spontaneous speech, speech fluency and stereotyped thinking. The fatigue score is formed from blind affect, agitated compassionate withdrawal, lethargic movement and disorientation. The general psychopathological score is calculated from physical concerns, stress, feeling of guilt, tension, imposed manners, depression, lethargic movement, lack of cooperation, unusual thought contents, disorientation, poor attention, lack of judgment and insight, disturbance of volition, poor impulse control, mind occupation and active social avoidance. The thought disorder score is calculated from clarity disorder, hallucinatory behavior, grandiosity and unnatural thoughts. The paranoia score is formed by suspicion (injury and harm), aggression and lack of cooperation. The depression score is calculated based on physical concerns, stress, feeling of guilt and depression. The complementary score is formed from agitation, hostility, depression, rage, delay malfunction in desire contours and emotional variability. The activation score is calculated from agitation, tension and imposed manners. The matched group’s individuals were selected from the outpatients who were referred by other neurologists to Ghaem Hospital EEG Clinic with headache. The individuals who had normal neurological examination and normal CT scans with them (as inclusion criteria) were chosen among them to accomplish the above-mentioned procedure. SPSS 11.5 was used to do the statistical analyses. Chi-square and ANOVA are used to make the comparison between the two groups. P values \0.05 are considered to be statistically significant.

Results Sixty-four patients were recruited in this study. The same number (64) formed the matched controls. There is no statistically significant difference between the age and gender of patients and matched controls. As shown in Table 1, the dominant wave pattern of EEG is different between the two groups (P value = 0.009). Also, data are provided in Table 2, which shows that there is a statistically significant difference between the dysrhythmia recorded in the patient and control individuals (P value = 0.033). Table 3 shows that the two groups are not different in the recorded localized slow waves (P value = 0.054). Table 1 Dominant wave pattern in the EEG tracings of the two studied groups Alpha

Beta

Theta

Patient

52 (81.3 %)

4 (6.3 %)

8 (12.5 %)

Control

56 (87.5 %)

8 (12.5 %)

0

Acta Neurol Belg Table 2 Dysrhythmia recorded in the EEG tracings of patients and controls No dysrhythmia

Weak dysrhythmia

Moderate dysrhythmia

Total dysrhythmia

Patient

52 (81.3 %)

4 (6.3 %)

8 (12.5 %)

12 (18.8 %)

Control

60 (93.8 %)

4 (6.3 %)

0

Table 3 Localized slow wave recordings of the two studied groups Frontal

Temporal

Patient

2 (3.1 %)

0

Control

2 (3.1 %)

2 (3.1 %)

4 (6.3 %)

Table 5 The anatomical origin of pathological waves in the two populations Frontal

The majority of our patients (60; 93.8 %) and controls (52; 81.3 %) had a normal voltage recorded in their EEG tracings. Low voltage was observed in 4 (6.3 %) patients and 12 (18.8 %) controls, which is significantly different (P value = 0.033). We also identified the pathological waves in patient and control tracings. As shown in Table 4, although a higher number of pathological waves were identified in our patients, this difference did not reach the level of statistical significance (P value = 0.17). On the contrary, the anatomical regions of pathological waves showed a significant difference (P value = 0.049) between the patient and control population with patients’ pathological waves almost exclusively originating from the frontal region (Table 5). The laterality of pathological waves was not different (P value = 0.083) between the two groups as shown in Table 6. We further analyzed the correlation between age and dysrhythmia, but no correlation was seen in our studied population (Fig. 1). As shown in Fig. 2, the observed dysrhythmia was seen in paranoid-type schizophrenia in one-third of detected dysrhythmias and 20 % in disorganized-type schizophrenia. Among 40 patients with no family history of psychiatric disorders, dysrhythmia was detected in 8 (20 %) while this figures decreased to 4 (18.2 %) in 22 patients with a known family history of psychiatric diseases (2 missing). Dysrhythmia was more prevalent in patients without family history of psychiatric diseases, yet this figure did not reach a statistically significant level (P value = 0.86). No effect was seen for the history of head trauma on detection of dysrhythmia. 6 patients (17.6 %) out of 34 with a negative history of head trauma had dysrhythmias and this figure was 6 out of 30 (20 %) in those with a positive history of head trauma, which bears no statistical significance (P value = 0.8).

Temporal

Patient

4 (6.3 %)

0

Control

0

2 (3.1 %)

Table 6 Laterality of pathological waves identified in the studied population Bilateral

Generalized

Patient

4 (6.3 %)

4 (6.3 %)

Control

2 (3.1 %)

0

We further analyzed the calculated scores against the dominant wave pattern of the recorded EEGs. Table 7 lists the scores that were significantly correlated with the identified wave patterns. No statistically significant correlation were identified between the dominant wave and general psychopathological score, thought disorder score, paranoia score, depression score, complementary score, PANSS test score, recollection score and language score (all P values were [0.05). As Table 8 shows, significant correlation was found between dysrhythmia and positive PANSS, paranoia and complementary scores. All other scores had no statistically significant correlation with dysrhythmia. Finally, we found that pathological waves had statistically significant correlation with only two scores, namely, negative sign score (P value = 0) and general psychopathological score (P value = 0.026). The mean values of these scores were 23.9 and 58.7 in 56 patients with no pathological waves, 26.7 and 55 in 6 patients with sharp waves and 16 and 47 in 2 patients with spike waves, respectively.

Discussion This is the first study addressing the EEG characteristics of Iranian schizophrenia patients in the acute phase. It also

Table 4 Pathological waves found in EEG recordings No pathological waves

Sharp waves

Spike waves

Long theta waves

Total pathological waves

Patient

56 (87.5 %)

4 (6.3 %)

2 (3.1 %)

2 (3.1 %)

8 (12.5 %)

Control

62 (96.9 %)

2 (3.1 %)

0

0

2 (3.1 %)

123

Acta Neurol Belg

40 35 30 25

no dysrhythmy

20

mild dysrhythmy

15

moderate dysrhythmy

10 5 0 mean value standard of age deviation of age Fig. 1 Age and dysrhythmia correlation

dysrhythmy 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00%

d

d ffe di un

di

ss

or

re

ga

nt

ni

ia

te

ze

al du si re

pa

ra

no

id

dysrhythmy

Fig. 2 Correlation between types of schizophrenia and dysrhythmia

investigates the correlation of PANSS and MMSE scores with the EEG findings in this understudied psychiatric population. Theta waves make the distinguishing pattern in the patient group, while alpha and beta wave patterns are shared between patients and controls, which makes their distinguishing characteristics weak. Before the application of clinical quantitative EEG, some researchers had reported that chronic schizophrenia patients had less alpha wave (SEEG) compared to healthy controls [12, 13]. In the study by Fallani and colleagues (Q-EEG), schizophrenia patients had more generalized theta and slower moderate alpha waves before treatment [14]. In another study, noticeable decrease in beta and gamma frequencies in the Q-EEG was observed in schizophrenia patients [15]. Our study had also found the same results in EEG wave patterns. The identified slow waves in our study are of generalized type, which indicates a widespread dysfunction in the brain cells. Moderate dysrhythmia that mirrors the severity of the dysfunction of the brain cells was found in eight (12.5 %) patients. The significantly higher number of controls with a low EEG voltage could be partly attributed to the significantly higher number of healthy individuals with beta waves

123

dominating their EEG tracings. The higher numbers of pathological waves in our patient population, which do not reach statistically significant level, might merely imply a small patient group size and, by increasing the patient population, significance will be reached especially because some of these pathological patterns are unique to our patient population. Given the known fact that dysrhythmia increases with increase of age [16], the absence of such correlation in our studied population shows that the dysrhythmia observed in our patients is a result of the disorder. Moreover, according to some studies, the difference in EEG abnormality risk between antipsychotic-treated patients and hospitalized patients not treated with an antipsychotic drug was not statistically significant [17]. This issue demonstrates that the dysrhythmia recorded in the patients is not related to risperidone consumption. In a study by Takahashi and colleagues (Q-EEG), lower frequencies were found in fronto–centro–temporal regions of schizophrenia patients, which were not found in the parieto-occipital brain regions of their patients [18]. Contrary to our findings, Knott and collaborators found generalized theta and slower alpha frequencies (Q-EEG) in the frontal brain region of their schizophrenia patients [7]. Our study could not localize and distinguish between the slow waves recorded in EEGs of patients and controls. This may reflect that brain cell dysfunction in our patients is not limited to a particular brain region and there is widespread involvement of brain regions. Identification of pathological waves including sharp waves, spike waves and high voltage theta waves, which are epileptiform, implies applied electrophysiological similarities between epilepsy and schizophrenia [6, 19, 20]. Although identification of epileptiform waves was previously reported in schizophrenia patients, any functional and diagnostic deduction needs larger studies incorporating more schizophrenia patients that should also enjoy fine brain functional imaging techniques. Exclusive identification of pathological waves in the frontal region of our patients not only nicely fits the personality and thought impairment of schizophrenia patients, but also opens the door to future studies tackling the similarities of epilepsy and schizophrenia with potential medical interventions that might help in better understanding both diseases and finding better treatment modalities. Both PANSS positive and negative sign scores were significantly correlated with the dominant wave pattern of EEG. A similar correlation was found in a study done in Russia between the reduction of psychopathologic signs (calculated PANSS scores) and an increase in the background Q-EEG spectrum density in the delta, theta, alpha1 (8–8.9 Hz) and alpha 3 (11–12.9 Hz) waves [21, 22]. Contrary to our findings, Gross and colleagues (Q-EEG) have reported a positive correlation between delta waves

Acta Neurol Belg Table 7 Scores significantly correlated with the dominant wave in recorded EEGs in the studied patients Beta wave dominance (4 patients)

Alpha wave dominance (52 patients)

Theta wave dominance (8 patients)

P value

Mean PANSS positive sign score

29

25.5

21.2

0.016*

Mean PANSS negative sign score

24

24.3

21

0.019*

Mean fatigue score

10.5

13.4

12.5

0.017*

8.5

10.9

9.5

0.029*

Mean activation score Mean orientation score

6

8.04

9.5

0.008*

Mean attention and calculation score

0

2.7

3.25

0.004*

0.5 18

0.88 24.5

1 27.25

0.04* 0.002*

Mean development score Mean MMSE score

Table 8 Correlation between the scores and dysrhythmia No dysrhythmia (52 patients)

Minor dysrhythmia (10 patients)

Moderate dysrhythmia (2 patients)

P Value

Mean positive PANSS score

25

27.6

15

0.002*

Mean paranoia score

10.65

12.2

9

0.047*

Mean complementary score

21.4

24.2

21

0.019*

and PANSS negative sign score [23]. Interestingly, no positive correlation between PANSS positive and negative sign scores and Q-EEG was reported in a much earlier study [24]. The main underlying reason behind the discrepancy between the reported observations is the difference in patient population and patient recruitment. A closer look at the patient population in the reported studies shows that they are at different stages of the disorder and under different medications which makes the comparison rather impossible. Current literature on the correlation between fatigue and dominant Q-EEG wave patterns is scarce [25]. We found that fatigue is seen in patients with dominant alpha, theta and beta wave patterns in a decreasing order. In our population, a decrease in frequency was seen to be associated with an increase in orientation, and attention and calculation scores. The study by Siekmeire and colleagues (Q-EEG) had observed a positive correlation between temporal lobe theta activity and positive schizophrenic symptoms, but in our study no pathological wave was recorded in the temporal lobe of patients to allow a comparison [26]. We found the highest positive sign score in patients with mild dysrhythmia. A significant correlation was observed between dysrhythmia and paranoia and complementary scores in our patients. Since the patient populations and disease characteristics in the available reported studies on the correlation between EEG parameters and calculated scores are not similar, it is practically impossible to draw a clear conclusion on such correlations. We are currently in need of larger and

preferably multi-centric studies incorporating a large number of patients with comparable symptoms, disease stage and medication history to be able to provide better reproducible data on the correlation between EEG parameters and clinical scores. The ensuing steps in such studies should also use more intricate functional brain imaging techniques allowing us to focus on specific regions of the brain found to play a role in this disorder. Conflict of interest

None.

References 1. Freedman R (2003) Schizophrenia. N Engl J Med 349(18): 1738–1749 2. Carpenter WT Jr, Buchanan RW (1994) Schizophrenia. N Engl J Med 330(10):681–690 3. Mortensen PB, Pedersen CB, Westergaard T, Wohlfahrt J, Ewald H, Mors O et al (1999) Effects of family history and place and season of birth on the risk of schizophrenia. N Engl J Med 340(8):603–608 4. Kane JM (1996) Schizophrenia. N Engl J Med 334(1):34–42 5. Galderisi S, Mucci A, Volpe U, Boutros N (2009) Evidencebased medicine and electrophysiology in schizophrenia. Clin EEG Neurosci 40(2):62–77 6. Malow BA, Reese KB, Sato S, Bogard PJ, Malhotra AK, Su TP et al (1994) Spectrum of EEG abnormalities during clozapine treatment. Electroencephalogr Clin Neurophysiol 91(3):205–211 7. Knott V, Labelle A, Jones B, Mahoney C (2001) Quantitative EEG in schizophrenia and in response to acute and chronic clozapine treatment. Schizophr Res 50(1–2):41–53 8. Nagase Y, Okubo Y, Toru M (1996) Electroencephalography in schizophrenic patients: comparison between neuroleptic-naive state and after treatment. Biol Psychiatry 40(6):452–456

123

Acta Neurol Belg 9. Omori M, Koshino Y, Murata T, Murata I, Nishio M, Sakamoto K et al (1995) Quantitative EEG in never-treated schizophrenic patients. Biol Psychiatry 38(5):305–309 10. Harris AW, Bahramali H, Slewa-Younan S, Gordon E, Williams L, Li WM (2001) The topography of quantified electroencephalography in three syndromes of schizophrenia. Int J Neurosci 107(3–4):265–278 11. Javanbakht A, Sanati M (2006) Psychiatry and psychoanalysis in Iran. J Am Acad Psychoanal Dyn Psychiatry 34(3):405–414 12. Matsuura M, Yoshino M, Ohta K, Onda H, Nakajima K, Kojima T (1994) Clinical significance of diffuse delta EEG activity in chronic schizophrenia. Clin Electroencephalogr 25(3):115–121 13. Sponheim SR, Clementz BA, Iacono WG, Beiser M (1994) Resting EEG in first-episode and chronic schizophrenia. Psychophysiology 31(1):37–43 14. De Vico Fallani F, Maglione A, Babiloni F, Mattia D, Astolfi L, Vecchiato G et al (2010) Cortical network analysis in patients affected by schizophrenia. Brain Topogr 23(2):214–220 15. Magomedov RA, Garakh ZhV, Orekhov IuV, Zaitseva IuS, Strelets VB (2010) Gamma-rhythm, positive, negative symptoms and cognitive dysfunction in schizophrenia. Zhurnal nevrologii i psikhiatrii imeni SS Korsakova/Ministerstvo zdravookhraneniia i meditsinskoi promyshlennosti Rossiiskoi Federatsii, Vserossiiskoe obshchestvo nevrologov [i] Vserossiiskoe obshchestvo psikhiat 110(1):78–83 16. Widdess-Walsh P, Sweeney BJ, Galvin R, McNamara B (2005) Utilization and yield of EEG in the elderly population. J Clin Neurophysiol 22(4):253–255 17. Centorrino E et al (2002) EEG abnormalities during treatment with typical and atypical antipsychotics. Am J Psychiatry 159:109–115 18. Takahashi T, Cho RY, Mizuno T, Kikuchi M, Murata T, Takahashi K et al (2010) Antipsychotics reverse abnormal EEG

123

19.

20.

21.

22.

23.

24.

25.

26.

complexity in drug-naive schizophrenia: a multiscale entropy analysis. Neuroimage 51(1):173–182 Bob P, Glaslova K, Susta M, Jasova D, Raboch J (2006) Traumatic dissociation, epileptic-like phenomena, and schizophrenia. Neuro Endocrinol Lett 27(3):321–326 Sachdev P, Brodaty H, Roubina S, Mackenzie RA (1999) An electroencephalographic investigation of late-onset schizophrenia. Int Psychogeriatr 11(4):421–429 Wu X, Liu XQ (1995) Study of the alpha frequency band of healthy adults in quantitative EEG. Clin Electroencephalogr 26(2):131–136 Lebedeva IS, Kaleda VG, Tsutsul’kovskaia M (2003) The relationship between clinical changes in schizophrenia, acoustic evoked potentials, and background electroencephalogram. Zhurnal nevrologii i psikhiatrii imeni SS Korsakova/Ministerstvo zdravookhraneniia i meditsinskoi promyshlennosti Rossiiskoi Federatsii, Vserossiiskoe obshchestvo nevrologov [i] Vserossiiskoe obshchestvo psikhiat 103(7):16–20 Gross A, Joutsiniemi SL, Rimon R, Appelberg B (2006) Correlation of symptom clusters of schizophrenia with absolute powers of main frequency bands in quantitative EEG. Behav Brain Funct 2:23 Gerez M, Tello A (1995) Selected quantitative EEG (QEEG) and event-related potential (ERP) variables as discriminators for positive and negative schizophrenia. Biolog Psychiatry 38(1):34–49 Higashi Y, Momotani Y, Suzuki E, Kaku T (1987) Clinical and EEG studies of zotepine, a thiepine neuroleptic, on schizophrenic patients. Pharmacopsychiatry 20(1 Spec No):8–11 Siekmeier PJ, Stufflebeam SM (2010) Patterns of spontaneous magnetoencephalographic activity in patients with schizophrenia. J Clin Neurophysiol 27(3):179–190