|Home | About | Journals | Submit | Contact Us | Français|
Brain’s alpha activity and alpha responses belong to major electrical signals that are related to sensory/cognitive signal processing. The present study aims to analyze the spontaneous alpha activity and visual evoked alpha response in drug free euthymic bipolar patients. Eighteen DSM-IV euthymic bipolar patients (bipolar I n = 15, bipolar II n = 3) and 18 healthy controls were enrolled in the study. Patients needed to be euthymic at least for 4 weeks and psychotrop free for at least 2 weeks. Spontaneous EEG (4 min eyes closed, 4 min eyes open) and evoked alpha response upon application of simple visual stimuli were analyzed. EEG was recorded at 30 positions. The digital FFT-based power spectrum analysis was performed for spontaneous eyes closed and eyes open conditions and the response power spectrum was also analyzed for simple visual stimuli. In the analysis of spontaneous EEG, the ANOVA on alpha responses revealed significant results for groups (F(1,34) = 8.703; P < 0.007). Post-hoc comparisons showed that spontaneous EEG alpha power of healthy subjects was significantly higher than the spontaneous EEG alpha power of euthymic patients. Furthermore, visual evoked alpha power of healthy subjects was significantly higher than visual evoked alpha power of euthymic patients (F(1,34) = 4.981; P < 0.04). Decreased alpha activity in spontaneous EEG is an important pathological EEG finding in euthymic bipolar patients. Together with an evident decrease in evoked alpha responses, the findings may lead to a new pathway in search of biological correlates of cognitive impairment in bipolar disorder.
Since the 1980s, the concept of “oscillatory brain dynamics” has achieved prominent progress in neuroscience research. During last decade, applications of the oscillatory activity in clinical pathology have grown rapidly, as extensively explained in a recent review (Başar and Güntekin 2008). Our research group initiated research on brain oscillatory responses and cognitive processes in schizophrenia (Başar-Eroğlu et al. 2008), Alzheimer’s disease (Yener et al. 2007, 2008; Güntekin et al. 2008; Başar et al. 2010) and in two different states of bipolar disorders (Özerdem et al. 2008a, b).
The present study analyzes the spontaneous alpha activity and visual alpha responses in a group of drug-free euthymic bipolar patients, which was already shown in manic bipolar patients (Özerdem et al. 2008a). Further, it is noteworthy to examine the degree of change of alpha activity in bipolar euthymic disease in comparison to schizophrenia.
Alpha rhythm, which was first observed by Hans Berger (1929), was initially considered as the “Brain’s Idling Rhythm”. Later, several authors described that EEG is not noise and that selectively synchronized alpha oscillations in the mammalian and human brain are part of the fundamental functional signaling of CNS (Başar 1980; Lehmann 1989; Klimesch 1999; Başar et al. 2001; Nunez et al. 2001). The alpha response is a fundamental component in sensory/cognitive tasks of healthy subjects. Following the publication of papers in the alpha special issue (Başar et al. 1997) research on cognitive processes manifesting alpha oscillations is rapidly increasing.
Bipolar disorder is a chronic mental illness with a relapsing and remitting course. Relapses are manic or depressive in nature. Patients suffer from a wide range of cognitive deficits (Martinez-Aran et al. 2004; Clark et al. 2002; Wilder-Willis et al. 2001) even when they are euthymic. There have been a relatively limited number of earlier electrophysiology studies of bipolar disorder, both in symptomatic and euthymic states (Bruder et al. 1992; Muir et al. 1991; Souza et al. 1995; Salisbury et al. 1998, 1999; O’Donnell et al. 2004). Despite different stimulus modalities, mostly being auditory, the common finding was prolonged P300 latency and reduced P300 amplitude, which was equivocal, mostly found to be related to psychosis and suggested to have an association with underlying frontal lobe pathology (Salisbury et al. 1999). More recent studies showed disturbed resting EEG activity in euthymic bipolars (El-Badri et al. 2001) and abnormal high frequency synchronization1 in response to auditory stimuli (O’Donnell et al. 2004) in symptomatic bipolar patients. Özerdem et al. (2008a) observed increased occipital beta activity in manic states in response to visual oddball paradigm.
Cortical alpha rhythms are reduced in different pathologies as Schizophrenia, Mild Cognitive Impairment and Alzheimer’s disease. In most of the studies spontaneous EEG rhythms were analyzed. There are also studies that analyzes evoked/event related oscillations in different pathologies (Başar and Güntekin 2008). In the Schizophrenia studies auditory stimuli or visual steady inputs were mostly used (See reviews: Brenner et al. 2009; Krishnan et al. 2005; Kwon et al. 1999). Alzheimer’s disease (AD) patients have been characterized by low power of posterior alpha and/or beta (13–30 Hz) rhythms similar to healthy elderly subjects. Posterior alpha rhythms showed a power decrement in subjects with mild cognitive impairment compared with healthy elderly subjects that are in a resting-state condition (Zappoli et al. 1995; Huang et al. 2000; Jelic et al. 2000; Koenig et al. 2005; Babiloni et al. 2006, 2007; Rossini et al. 2007). EEG Alpha activity was also found to be decreased in schizophrenia (Alfimova and Uvarova 2008; Iacono 1982; Itil et al. 1972, 1974; Miyauchi et al. 1990; Sponheim et al. 1994, 2000). Başar-Eroğlu et al. (2009) observed significantly reduced alpha responses in the range of 15% and Ford et al. (2008) also reported reduced alpha response in schizophrenia.
To the best of our knowledge, there is only one prior study assessing the alpha activity in bipolar disorder (Clementz et al. 1994) where alpha activity was shown to be reduced in bipolar patients with psychotic characteristics and in patients with schizophrenia in comparison to healthy controls. To date, no published study has investigated the alpha activity in drug-free euthymic bipolar patients.
The aims of the present study are as follows:
Eighteen DSM-IV (Diagnostic and Statistical Manual of Psychiatric Disorders, fourth edition) euthymic bipolar I (n = 15) patients, euthymic bipolar II (n = 3) patients (13 female, 5 male), aged between 28 and 44 years (mean age 31.66 ± 5.99 SD) and an equal number (n = 18) of sex, age (mean age 29.83 ± 7.77 SD) and educationally matched healthy controls were enrolled in the study. All subjects were followed by the psychiatrics of Bakirkoy State Hospital of Mental Health Research Training and Education Center, which has the largest psychiatric pathology population in Turkey. The possibility to recruit the 18 drug-free euthymic patients was possible, since the Bakirkoy State Hospital of Mental Health Research, Training and Education Center; Istanbul, Turkey is the largest psychiatric hospital in Turkey. This hospital has the possibility to recruit drug-free patients from all parts of Istanbul City, which has a 13 million population. All subjects were interviewed with SCID-I (Structured Interview for DSM-IV) (First et al. 1996). The study was approved by the local Ethics Committee of Bakirkoy State Hospital of Mental Health Research, Training and Education Center, Istanbul, Turkey. All participants provided written informed consent. Patients needed to be euthymic at least for 6 months, patients were psychotropic-free for at least 2 weeks prior to study enrollment and none of them were using benzodiazepines. Only one of the subjects was drug-free for 2 weeks prior to study. One of the subjects was drug-free for 3 weeks prior to study. All other 16 patients were drug-free for at least 6 weeks prior to study. Patients needed to score 7 or less on the reliable and validated Turkish versions of the Young Mania Rating Scale (YMRS) (Karadağ et al. 2002), Hamilton Depression Rating Scale (HAM-D 21) (Aydemir and Deveci 2003); to have no co-morbid axis I diagnosis, and be medically healthy, as shown through physical examination and routine laboratory tests. Volunteers who proved to have no present or past psychiatric condition and to be medically healthy on physical examination were enrolled as the control group.
The subjects were sat in a dimly-lit isolated room. Two different experimental set ups were performed: (1) Spontaneous EEG of the subject were recorded for 4 min for “eyes open” and 4 min for “eyes closed” conditions. (2) The EEG was recorded upon application of simple light stimuli for analyzing simple evoked oscillations. A visual sensory paradigm was used in the experiments. Stimulation consisted of a white screen with 10 cd/cm2 luminance. A series of 60 stimulation signals (1,000 ms duration) were applied randomly, with the inter-stimulus intervals varying between 3 and 7 s.
EEG was recorded by using the BrainAmp EEG amplifier, Brain Vision Recorder software (Brainproducts, Munich, Germany), and the BrainCap electrode cap at 30 positions. The EEG was amplified by means of a BrainAmp with band limits of 0.01–250 Hz. The EEG was digitized on-line at a sampling rate of 500 Hz. All electrode impedances were less than 10 kΩ. In order to maintain a constant level of vigilance, a researcher controlled on-line the subject and the EEG traces; The subject was verbally alerted any time there were signs of behavioral and/or EEG drowsiness. Two linked earlobe electrodes (A1 + A2) served as references. The EOG from the medial upper and lateral orbital rim of the right eye was also registered. For the reference electrodes and EOG recordings, Ag–AgCl electrodes were used. The EEG and EOG signals were visually scored and portions of the data that contained aberrant eye movements, muscle movements or artifacts were removed.
The spontaneous EEG was recorded for 4 min eyes closed and 4 min eyes open conditions and subsequent analyses were performed separately for eyes closed and eyes open conditions. The recorded EEG data were analyzed and fragmented off-line in consecutive epochs of 1 s. The digital FFT-based power spectrum analysis was performed. (10% Hanning windowing function was evaluated in order to calculate alpha frequency peak). These power values were averaged across the epochs of a given trial. The standard frequency band of interest was alpha (8–13 Hz). The maximum individual alpha frequency value for each subject was included, for the purpose of statistical analysis, as the maximum individual alpha frequency value of that subject.
The epochs (between 0 and 1,000 ms) of each subject were averaged and then the digital FFT-based power spectrum analysis was performed. (10% Hanning windowing function was evaluated in order to calculate the alpha frequency peak). The standard frequency band of interest was alpha (8–13 Hz). The maximum individual alpha frequency value for each subject was included, for the purpose of statistical analysis, as the maximum individual alpha frequency value of that subject.
SPSS was used for statistical analysis. A repeated measure ANOVA was used to determine the statistical significance of differential alpha responses over different conditions, locations; and between patients and controls. Two separate ANOVA were used for two different experimental set ups (Spontaneous EEG, Visual Evoked Oscillations): In the analysis of spontaneous EEG alpha power differences, repeated measures of ANOVA included the between-subjects factor as healthy subjects and euthymic patients; repeated measure ANOVA included the within-subject factors as condition (eyes open, eyes closed); location (Frontal, Central, Temporal 1 (T7–T8), Temporal 2 (TP7–TP8), Parietal, Occipital) and hemisphere (right, left). In the analysis of Visual Evoked alpha power differences, repeated measures of ANOVA included the between-subjects factor as healthy subjects and euthymic patients; repeated measure ANOVA included the within-subject factors as location (Frontal, Central, Temporal1 (T7–T8), Temporal 2 (TP7–TP8), Parietal, Occipital) and hemisphere (right, left). Post-hoc comparisons for between-subject effects and within-subject effects were analyzed using the t test. Greenhouse-Geisser corrected P values are reported, and the level of significance was set to P < 0.05 for post-hoc comparisons.
Figure 1 shows the grand averages of power spectrum of the alpha frequency range in occipital locations (O1, Oz, and O2) in 18 healthy and 18 euthymic bipolar participants for the eyes open condition. Here, the alpha frequency range power spectrum of healthy controls can be as high as 0.90 μV2 for all occipital electrodes, while the power spectrum of the euthymic patients reaches 0.40 μV2.
Figure 2 represents the grand averages of power spectra of 18 healthy and 18 euthymic subjects in the alpha frequency range for the eyes closed recording session for occipital locations (O1, Oz, and O2). While the power spectrum of the alpha frequency range reached 4.80 μV2 for O1; 4.0 μV2 for Oz and 4.50 μV2 for O2 electrode in healthy controls, it remained at 1.0 μV2 across all occipital electrodes in the euthymic patients.
Figure 3 represents the grand average of the evoked response power spectra for 18 healthy and 18 euthymic subjects in the alpha frequency range upon application of simple light stimuli (for O1, Oz, and O2 electrodes). The alpha frequency power spectrum of evoked response reached 0.04 μV2 in healthy controls, whereas the power spectrum of evoked response of euthymic patients only reached 0.015 μV2.
The observed differences between healthy subjects and euthymic patients are in accordance with the statistical findings described below.
The ANOVA of alpha responses revealed significant differences between groups (F(1,34) = 8.703; P < 0.007). Post-hoc comparisons showed that the spontaneous EEG alpha power of healthy subjects was significantly higher than that of euthymic patients (P < 0.0001). The ANOVA of alpha responses revealed significant differences between experimental conditions (eyes open, eyes closed; (F(1,34) = 33.043; P < 0.0001). Post-hoc comparisons showed that spontaneous EEG alpha power for the eyes closed condition were significantly higher than for the eyes open condition (P < 0.0001). The ANOVA of alpha responses revealed significant results for condition × group (F(1,34) = 5.726; P < 0.03). Post-hoc comparisons showed that spontaneous EEG alpha power of healthy subjects was significantly higher than euthymic patients for the eyes closed condition (P < 0.0001) and eyes open condition (P < 0.002). The ANOVA of alpha responses revealed significant differences between locations (F(5,17) = 17.129; P < 0.0001). Post-hoc comparisons showed that spontaneous EEG alpha power at occipital electrodes was higher than that of frontal, central, temporal and parietal electrodes (P < 0.001 for all electrode sites). Furthermore, spontaneous EEG alpha power at parietal electrodes was higher than that of frontal, central and temporal electrodes (P < 0.001 for all electrode sites). The ANOVA of alpha responses revealed significant results for condition (eyes closed, eyes open) × location (F(5,17) = 14.644; P < 0.0001). The post hoc comparisons showed that the eyes closed alpha power means at frontal, central, temporal, parietal and occipital electrodes were higher than the corresponding eyes open alpha mean for these electrodes (P < 0.001 for all electrode sides).
Figure 4a presents bar graphs of mean maximum alpha power value of healthy and euthymic subjects for the eyes open recording session for all electrode pairs. Figure 4b presents bar graphs of mean maximum alpha power values of healthy and euthymic subjects for the eyes closed recording session for all electrode pairs. As seen from Fig. 4a, during the eyes open recording session, the mean power spectrum of the healthy subjects varies between 0.29 and 1.10 μV2, while the mean power spectrum of the euthymic subjects is within the range 0.21–0.72 μV2. As seen from Fig. 4b, during the eyes closed recording session, the mean power spectrum of the healthy subjects is between 0.95 and 8.60 μV2, while that for the euthymic subjects is only between 0.51 and 3.63 μV2.
ANOVA of the alpha responses revealed significant differences between groups (F(1,34) = 4.981; P < 0.04). Post-hoc comparisons showed that visual evoked alpha power of healthy subjects was significantly higher than that of euthymic patients (P < 0.0001). The ANOVA of alpha responses revealed significant results for location (F(5,170) = 8.966; P < 0.0001). Post-hoc comparisons showed that the visual evoked alpha power of occipital and parietal electrodes were higher than for frontal and temporal electrodes (P < 0.001 for all electrode sites).
Figure 4c presents bar graphs of mean maximum evoked alpha power response value of healthy and euthymic subjects for all electrode pairs. As seen from Fig. 4c, the mean evoked power spectra of the healthy subjects are between 0.024 and 0.095 μV2, while those of the euthymic subjects are only between 0.017 and 0.040 μV2.
The literature includes several previous investigations of spontaneous EEG in bipolar patients. (Clementz et al. 1994; Cook et al. 1986; Dewan et al. 1988; Gerez and Tello 1992; Kano et al. 1992; Koles et al. 1994; Souza et al. 1995; Small et al. 1989, 1998; Schulz et al. 2000; El-Badri et al. 2001; Ikeda et al. 2002). The present study differs from other studies since none of the other groups analyzed the spontaneous EEG jointly with visual evoked oscillations in groups of drug-free euthymic patients. Clementz et al. (1994) investigated alpha activity in a group of bipolar psychosis patients, schizophrenia patients and their first-degree relatives. EEG data obtained from patients and their first-degree relatives showed that patients with schizophrenia and bipolar disorder had reduced alpha in comparison to healthy subjects. Our results indicated a reduction in alpha activity in the range of 70% within a group of euthymic patients, compared with healthy controls. This was not observed in earlier studies and can even be considered a breakdown of alpha activity and visual alpha response.
Clementz et al. (1994) included a mix patient group in their study and the bipolar patient group was not all in eutymic stage and was not all drug free. On the contrary, the patient group included in our analysis provides strongest advantages and makes the study unique in the literature. (2) On the other hand the mix subject groups included in Clementz’s study give the chance to the authors to compare different subject groups. In future, a possible comparison of drug free eutymic patients with drug free schizophrenic patients could be of major importance to detect the differences of alpha activity between these two groups. (2) Clementz et al. (1994) analyzed only the Central electrodes (C3, Cz, C4), we have analyzed Frontal (F3, F4), Central (C3, C4), Temporal 1 (T7–T8), Temporal 2 (TP7–TP8), Parietal (P3, P4), Occipital (O1, O2). (3) Clementz et al. (1994) recorded EEG during eyes closed recording session. In extension we have also recorded EEG during eyes closed, eyes open and upon application of basic visual stimuli.
To our knowledge the present study is the first one reporting decrease of alpha activity in drug-free euthymic patients. Clementz et al. (1994) reported that the bipolar patients had reduced alpha activity in central electrodes. Until now no other groups have reported such a difference between healthy subjects and bipolar patients. The present study further emphasizes the importance of including all electrodes and comparing different recording sessions.
Furthermore, increased occipital beta response and, in contrast to this, decreased alpha response in response to visual oddball paradigm was observed by Özerdem et al. (2008a, b) in manic state. According to Başar et al. (1997), if the ongoing activity has decreased within a specific frequency, then the evoked or event related responses within this frequency are also low; this was also the case in the present study.
A number of studies have shown significant differences in alpha asymmetry in bipolar individuals during an application of several paradigms (Kano et al. 1992; Allen et al. 1993; Harmon-Jones et al. 2008). In comparison with nonbipolar individuals, bipolar disorder patients showed greater relative left frontal cortical activation in preparation for the hard/win trials. In our study we did not find any lateralization effects as it is seen in Fig. 4. The present study clearly shows that drug-free euthymic subjects do not have any frontal alpha asymmetry during a spontaneous EEG recording. Application of different paradigms during the recordings may have changed the results. It is to note that in the present study there is no drug effect on the alpha activity. On the other hand in the studies by Kano et al. (1992), Allen et al. (1993) and Harmon-Jones et al. (2008) the patients were not all drug free. It is to note that drug applications have effects on oscillatory dynamics2 (Özerdem et al. 2008a; Yener et al. 2007; Başar and Güntekin 2008). Future studies are needed for tenable conclusions related to the alpha asymmetry by taking into consideration the application of different paradigms and different drug therapies.
According to several previous studies, schizophrenia patients display both reduced spontaneous alpha activity and event related alpha (Alfimova and Uvarova 2008; Başar-Eroğlu et al. 2009; Iacono 1982; Itil et al. 1972, 1974; Miyauchi et al. 1990; Sponheim et al. 1994, 2000). Few studies discuss the common and extinct parameters of schizophrenia and bipolar disorder by means of anatomical and genetically analysis (Lim et al. 1999; Roy et al. 1998; McDonald et al. 2004; Hoge et al. 1999; Campbell et al. 2004; Wright et al. 2000). According to Selemon (2004), there is a reduction of synaptic elements and neuronal connections in the cortex of schizophrenic patients without a decrease in total neuronal number (Table 1).
McDonald et al. (2004) performed a meta-analysis with 26 subjects and concluded that bipolar disorder is associated with mild ventricular enlargement. Sponheim et al. (2000) analyzed the brain alpha activity and its relation with brain morphology. These authors examined the power characteristics of resting electroencephalograms in 112 schizophrenic patients and seventy-eight non-schizophrenic psychosis patients (e.g. mood disorder patients, 33 bipolar) were included for comparison. Schizophrenic patients whose electroencephalograms were characterized by diminished alpha-band power had more negative symptoms, larger third ventricles, larger frontal horns of the lateral ventricles, increased cortical sulci widths, and greater ocular motor dysfunction compared with schizophrenic patients without these electroencephalogram characteristics. In non-schizophrenic psychosis patients, augmented low-frequency and diminished alpha band powers were not associated with any clinical or biological indices. In their study, psychosis bipolar patients were only one part of the control group. We will come back to the consequences of this comparison at the end of section “Future research toward determination of electrical biomarkers and need for standardization”.
Degabriele and Lagopoulos (2009) published a review of EEG and ERP studies in bipolar disorder. They reported the lack of a systematic approach towards experimental design and medication status. Further to the comments of these authors, in the following other issues are to be discussed in studies of Bipolar disorder.
1Stabilization of frequency response in a narrow frequency window.
2Brain’s temporal response within EEG frequency channels.