Spiritual

7 pages
6 views

A High Resolution Quantitative EEG Power Analysis of Obsessive-Compulsive Disorder

of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
A High Resolution Quantitative EEG Power Analysis of Obsessive-Compulsive Disorder
Transcript
  Reprinted from the German Journal of Psychiatry · http://www.gjpsy.uni-goettingen.de · ISSN 1433-1055  A High Resolution Quantitative EEG Power Analysis of Obsessive-Compulsive Disorder Pushpal Desarkar 1 , Vinod Kumar Sinha 2 , K. Jagadheesan 3 , S. Haque Nizamie 4   1 Senior Resident, Centre for Cognitive Neuroscience, Central Institute of Psychiatry, Ranchi 2  Associate Professor of Psychiatry, Central Institute of Psychiatry, Ranchi 3 Broadmeadows Adult Psychiatric Inpatient Unit, Melbourne, Victoria, Australia 4 Professor of Psychiatry & Director, Central Institute of Psychiatry, Ranchin Corresponding author: Pushpal Desarkar MD, DPM. Senior Resident, Centre for Cognitive Neuroscience, Central Institute of Psychiatry, Kanke (PO), Ranchi-834006, Jharkhand, India. E-mail: pushpalds@yahoo.com  Abstract Introduction: The findings of the quantitative EEG power-spectral studies in obsessive-compulsive disorder (OCD) have so far been mostly inconsistent. Moreover, none of the studies has been a high resolution one.  Aim  : The aim of the study was to examine the band power of delta, theta, alpha, beta1 and beta2 bands with high resolution EEG data in patients with obsessive- compulsive disorder and compare it with that of normal controls. It was hypothesized that there will be no significant group difference of individual band power between patient and control  group.  Methods: Raw EEG data were acquired from 64 channels using a linked ear reference. We obtained EEG power values for 20 adult OCD patients (10 males; 10 females) and 19 appropriately matched healthy controls across the abovementioned bands. We used Advanced Source Analysis (ASA, ANT software b.v. Netherlands; version- 3.0.0.5.) program for analysis of power. Results: OCD patients had significantly higher power in comparison to controls which was widespread in the theta fre- quency, predominantly left sided fronto-temporo-parietal in delta and alpha and only left frontal in beta2 bands. Conclusion: Increased band power in obsessive-compulsive disorder patients in all these bands which have been shown to be associated with cognitive processing, may reflect increased processing load in this group of patients with recruitment of wide area of cerebral hemisphere (German J Psychiatry 2007; 10: 29-35). Keywords: Obsessive-compulsive disorder, quantitative EEG, band power, increased cognitive processing Received: 9.1.2006 Published: 19.6.2006 Published: 1.4.2007 Introduction n the past two decades, there has been a considerable advance in the understanding of the biological underpin-nings of obsessive-compulsive disorder (OCD) with emergence of various hypothetical models. Currently, there is little confusion over a significant neurobiological basis of OCD. Genetic, neurochemical, neuropsychological and neu-roimaging studies have shown certain consistent abnormali-ties in OCD. After repeated failures of structural neuroima-ging methods in finding consistent structural abnormalities in psychiatric disorders, evolving functional brain-imaging tech-niques nowadays represent the most powerful tools for cha-racterizing in vivo human anatomy, neurophysiology and neurochemistry at modest temporal and spatial resolution. Recent functional neuroimaging studies in OCD population have consistently found abnormalities the frontal cortex and basal ganglia structures (Saxena & Rauch, 2000). Quantitative EEG (QEEG) being a physiological imaging technique can be used to delineate abnormal functioning of brain regions with remarkable temporal precision. However, the findings of the quantitative EEG power-spectral studies in OCD population have so far been mostly inconsistent. In one of the first quantitative EEG studies using power- I  D ESARKAR ET AL . 30 spectral analysis on 10 patients with OCD with a limited montage, Flor-Henry et al. (1979) reported relatively decrea-sed variability in the temporal region. In this study, however, no frontal lead was used. Khanna (1988) found decreased log power in the non domi-nant frontal-midline and posterior temporal regions. This study suggested right temporo-frontal hyper functioning to be associated with OCD and stressed the importance of the non-dominant fronto-temporal regions in this connection,  with regards to both the localization by changes and by the nature of the activity observed (beta activity). Bennasar et al (1991), in their study of power spectrum on OCD patients, found increased beta power in the frontal region. However, Perros et al. (1992) found significantly increased relative power in the theta-2 band in the left tem-poral and central regions and significantly reduced variability in frontal and temporal regions. Kuskowsky et al. (1993) recorded EEG from 13 unmedica-ted and nondepressed patients with DSM-III-R obsessive-compulsive disorder (OCD) and from 10 age-matched controls. Quantitative analysis of the EEG revealed lower log absolute power in the delta, beta 1, and beta 2 band- widths for OCD patients at frontal and right-hemisphere locations. Moreover, OCD patients displayed greater he-mispheric asymmetries in EEG activity based on difference measures of EEG power from homologous electrode pairs, indicative of severe right hemisphere EEG hypoactivity. Drake et al. (1996) compared EEG spectral measures in 20 patients meeting DSM-III-R criteria for obsessions and com-pulsions and 12 neurologically intact unmedicated controls.  Among the patients, 10 had comorbid Tourette's syndrome.  The EEG was recorded from 11 electrodes. The results showed that both left frontal and right frontal variability was significantly reduced in OCD patients in compared to controls. Consistent with literatures suggesting neurophysio-logical disturbances in OCD, this study, too, supported it by showing frontal lone dysfunction in OCD patients in compa-rison to neurologically intact controls.  Tot et al. (2002) did a recent study on OCD patients using quantitative EEG methods. The study included 22 right handed OCD subjects as the study group and 20 right han-ded pair-wise matched healthy subjects as control group.  The authors used 12-channels for analysis of alpha, beta, delta and theta bands. Hemispheric asymmetry and regional differences for 3 brain regions: frontal, Temporal and parie-tal, were evaluated with frequency analysis. The analysis sho- wed a considerable increase in delta and theta activity and a decrease in alpha activity of left fronto-temporal regions. A left temporal decrease in beta activity was significant. The study found that patients with OCD showed important fron-to-temporal dysfunction, predominantly in the left hemisphe-re. In another recent study, Karadag et al. (2003) obtained quan-titative analysis of the EEG in patients with OCD. The study revealed a decreased beta and an increased theta power at fronto-temporal regions. The patients who had higher scores in doubting test (Maudsley Obsessive Compulsive Question-naire) and more severely ill patients shared similar q-EEG features. The relative theta powers were significantly increa-sed and alpha powers were significantly decreased in these patients, particularly in the frontotemporal region. It was suggested that the q-EEG may be useful in investigating the OCD patients with heterogeneous characteristics. In summary, the results of different quantitative EEG studies in OCD patients have so far been mostly inconsistent except the finding of an increase in theta power, which has come up more or less consistently across studies. The reason for these inconsistencies is varied and includes lack of a control group, differences in EEG methods, medication differences and heterogeneity of the samples (one at least was largely made up of patients co-morbid for Tourette’s).  To the best of our knowledge none of the studies has been a high resolution one. All studies used limited number of e-lectrodes while EEG sampling of information. Moreover most of the study included heterogeneous population and did not strictly match healthy controls.  The aim of the study was to examine the band power (relati- ve) of delta (0.5- 3.5 Hz), theta (4-7.5 Hz), alpha (8-12Hz), beta1 (12.5-20Hz) and beta2 (20.5-30Hz) bands in patients  with obsessive-compulsive disorder and compare it with strictly matched normal controls using a high resolution EEG data. Given the inconsistencies of findings in the e-xisting literature and exploratory nature of the study using a high resolution EEG technique, we hypothesized that there  will be no significant group difference of individual band power between patient and control group. Subjects & Methods  This study was conducted at the Center for Cognitive Neu-roscience at the Central Institute of Psychiatry (CIP), Ranchi, India. The institute review committee approved the study. Participants Patients of both sexes, irrespective of medication status, aged between 18 and 50 years, giving informed consent and having a diagnosis of OCD according to DSM-IV (American Psychiatric Association 1994) were recruited using purposive sampling technique from outpatient facilities of the institute. Eleven patients (55%) were either drug naïve (5 out of 11 patients) or drug free for at least 2 weeks and 9 patients (45%) were receiving an SSRI (3 fluoxetine, 4 sertraline, and 2 fluvoxamine). Neither of them had any family history of significant neurological illness. Considering the high lifetime prevalence of co-morbid depression with OCD (Rasmussen and Eisen 1988) we also included those patients in whom onset of depressive symptoms (if present) had taken place after the onset of OC symptoms and was essentially mild (score < 18) as per the 17-item Hamilton rating scale for depression (HDRS) (Hamilton 1960). Healthy controls with nil significant personal and familial history, giving informed consent, and matched with patients for age, gender and handedness were recruited from the department personnel and post graduate students. We used  EEG   P OWER  A NALYSIS OF OCD 31general health questionnaire-5 (GHQ-5) (Shamsunder et al., 1986) score 0 for screening of controls.  All subjects were right-handed as assessed by the sidedness bias schedule (SBS) (Mandal et al., 1992). Yale Brown obsessi- ve compulsive scale (Y-BOCS) (Goodman et al., 1989) was applied on the patient population for quantification of psy-chopathology.  According to the above-mentioned inclusion criteria, 20 OCD patients and 19 healthy controls were recruited. EEG data acquisition Raw EEG data were acquired through Nihon-Kohden Neu-rofax electroencephalograph EEG-1100K from 64 channels using Ag-Ag/cl electrodes placed according to international 10/10 system using a linked ear reference. The sampling rate  was 1024Hz/sec/per channel and AD conversion was 16 bits. Skin resistance at each site was kept <5 K   Ώ . Eye mo- vement potentials were recorded using two electrodes placed 1 cm lateral to the outer canthus of each eye. All subjects  were asked not to smoke or take caffeine 3 hrs prior to the recording. Twenty minutes of continuous resting EEG was recorded with subjects sitting on a reclining chair with eyes closed in a light and sound attenuated room. EEG data processing  An experienced EEG reviewer selected more than 2 minutes of continuous, artifact free, awake-state EEG data. The EEG files were subsequently opened with Advanced Source  Analysis (ASA, ANT software b.v. Netherlands; version-3.0.0.5.) operations for analysis of power. We calculated absolute power (  μ  V  2 /Hz). Artifact rejection procedure was again carried out with the help of the artifact rejection pro-gram in the software. The study used a 50 Hz notch filter, had set a low cut filter at 0.16 Hz and a high cut filter at 70 Hz as a part of data processing. The software had an in-built  Welch periodogram for windowing. Five frequency bands  were defined as follows: delta (0.5- 3.5 Hz), theta (4-7.5 Hz), alpha (8-12Hz), beta1 (12.5-20Hz) and beta2 (20.5-30Hz) bands. Exact 2 minutes of artifact free EEG data was taken for quantitative analyses and was divided into multiple 2 secs epochs. The epochs were made to overlap for 0.5 secs, both at the beginning and at the end. Power analysis was done selecting 2 secs block lengths for already defined epoch events. We were forced to exclude 16 channels from the srcinal set of 64 channels since we could not get 2 minutes of artifact free data from them. Thus, finally, the following 48 electrodes were taken for power analysis: Fp1, Fp2, F3, F4, C3, C4, P3, P4, O1, O2, F7, F8, T7, T8, P7, P8, F1, F2, Fc1, Fc2, C1, C2, Cp1, Cp2, P1, P2, Af3,  Af4, Fc3, Fc4, Cp3, Cp4, F5, F6F, c5, Fc6, C5, C6, Cp5, Cp6, P5, P6, Af7, Af8, Ft7, Ft8, Tp7, Tp8.  All the electrodes were ear referenced. Statistical procedures Statistical analysis was done using Statistical Package for Social Sciences (SPSS, Inc., Chicago, Illinois) version 11.0.  As analysis for normal distribution of power values with Kolmogorov-Smirnov test with Shapiro-Wilk correction revealed significant non-normal distribution, all power scores  were log transformed to achieve gaussianity. Independent t test was applied to study the group difference of spectral power values for delta, theta, alpha, beta1, and beta2 bands between patient and control group across 48 channels. Since our study is the first high resolution quantita-tive EEG study on OCD population and essentially investi-gational in nature, we did not calculate regionwise averaging. Instead, we decided to apply t test on each channel power  value. However, to reduce the possibility of type I error because of multiple dependent variables, Bonferroni’s cor-rection was made to the level of significance. The new level of significance for power data taken was P < 0.01. Exponen-tial function of all the mean and standard deviation values  were calculated from the log power data. However, a level of significance (  α  ) of < 0.05 (two tailed)  was taken to consider a result statistically significant for socio-demographic and clinical data. Chi-square test (for categorocal variables) and independent t test (for continuous  variables) were applied for studying group difference of socio-demographic data.  The study population in the present study was a dichotomy as to the medication status. Therefore, subsequent Mann- Whitney U test was performed to address the impact of medication status. Results  The patient sample included 10 male and 10 female subjects  with a mean age 31.45 years (SD 8.91). The control group included 10 males and 9 females, with a mean age 29.85 years (SD 7.78). In patient group, mean age at onset of ill-ness was 23.90 years (SD 6.89) while mean duration of ill-ness was 7.60 years (SD 6.80). Mean duration of treatment in those who were taking medications was 8.28 months (SD 8.40). Mean Y-BOCS obsession score in the patient group  was 15.45 (SD 2.23), compulsion score was 12.40 (SD 4.22) and mean Y-BOCS total score was 27.85 (SD 5.44). The mean HDRS score was 11.30 (SD 3.65).  Analysis of delta (0.5-3.5 Hz) power showed that the two groups differed significantly in terms of mean power across left frontal (F5, Af7), right frontal (F2, Af8), left temporal (Fc5, Ft7, C5, Cp5, T7, Tp7), right temporal (Ft8, T8, Tp8), left parietal (C3, Cp3, P1, P3, P5) and right parietal (Fc2) regions. The OCD group had higher mean power than the control group in all regions (Table 1).  Analysis of theta (4-7.5 Hz) power revealed that the two groups differed significantly in terms of mean power across all channels except Fc1, C1, and Fc3. In theta band, too, the OCD group had higher mean power than the control group (Table 2).  Reprinted from the German Journal of Psychiatry · http://www.gjpsy.uni-goettingen.de · ISSN 1433-1055 Table 1: Group difference in delta power ( μ V2/Hz) across all channels Channel Patients (Mean ±SD) (N=20) Normals (Mean ±SD) (N=19) t (df =37) P* F3 49.84 + 2.77 23.71 + 2.11 2.584 .014 F4 60.02 + 3.08 32.72 + 2.89 1.730 .092 C3 13.62 + 3.13 5.52 + 1.95 3.000 .005* C4 13.63 + 3.11 5.56 +1.95 2.985 .005* P3 56.93 + 4.53 16.65 + 1.68 3.365 .002* P4 35.52 + 2.93 13.91 + 1.67 3.450 .001* O1 55.36 + 3.62 24.48 + 1.67 2.579 .014 O2 54.47 + 3.64 23.95 + 1.74 2.560 .015 F7 140.75 + 3.81 55.55 + 3.11 2.336 .025 F8 119.71 + 3.23 51.94 + 2.75 2.376 .023 T7 68.33 + 3.59 21.85 + 1.86 3.515 .001* T8 50.03 + 2.98 20.9 + 1.67 3.169 .003* P7 76.84 + 4.91 26.78 + 1.78 2.721 .010 P8 58.65 + 3.5 27.51 + 1.81 2.389 .022 F1 801.75 + 3.32 122.47 + 18.26 2.665 .011 F2 647.29 + 2.91 296.28 + 1.85 2.777 .009* Fc1 510.1 + 4.22 223.77 + 1.95 2.274 .029 Fc2 452.1 + 3.17 181.34 + 1.67 3.168 .003* C1 270.4 + 2.94 149.35 + 1.91 2.071 .045 C2 230.26 + 3.16 107.68 + 1.85 2.551 .015 Cp1 398.82 + 5.54 139.99 + 1.73 2.542 .015 Cp2 263.78 + 3.16 136.48 + 1.74 2.259 .030 P1 571.69 + 3.53 211.35 + 1.76 3.150 .003* P2 458.15 + 2.68 258.50 + 1.98 2.096 .043  Af3 1424.39 + 3.13 569.35 + 2.64 2.696 .011  Af4 1204.84 + 2.58 702.82 + 2.43 1.832 .075 Fc3 496.16 + 4.84 220.77 + 2.13 2.028 .050 Fc4 444.34 + 3.09 221.05 + 2.03 2.297 .027 Cp3 435.42 + 4.31 155.99 + 1.74 2.872 .007* Cp4 291.81 + 3.57 139.78 + 2.38 2.101 .042 F5 1365.26 + 3.7 416.09 + 2.42 3.309 .002* F6 1217.07 + 3.15 493.44 + 2.57 2.673 .011 Fc5 919.38 + 4.1 330.89 + 2.25 2.754 .009* Fc6 765.48 + 3.18 328.88 + 2.22 2.642 .012 C5 919.38 + 4.1 330.89 + 2.25 2.754 .009* C6 765.48 + 3.18 328.88 + 2.22 2.642 .012 Cp5 740.85 + 5.43 232.08 + 1.9 2.804 .008* Cp6 430.65 + 3.75 177.68 + 1.8 2.680 .011 P5 957 + 5.55 249.76 + 1.7 3.268 .002* P6 556.41+ 3.36 230.56 + 1.69 2.918 .006*  Af7 3787.65 + 4.07 1178.38 + 3.3 2.787 .008*  Af8 3782.73 + 3.34 1276.53 + 2.8 3.016 .005* Ft7 1520.96 + 4.42 497.75 + 2.7 2.744 .009* Ft8 1068.27 + 3.37 443.15 + 1.84 2.833 .007* Tp7 1087.79 + 5.13 335.53 + 1.86 2.939 .006* Tp8 880.33 + 4.61 299.23 + 1.7 2.912 .006* (*Bonferroni adjusted significance at P < 0.01) Table 2: Group difference in theta power ( μ V 2  /Hz) across all channels Chan-nel Patients (Mean ±SD) (N=20) Normals (Mean ±SD) (N=19) t (df =37) P* F3 7.72 + 2.76 3.62 + 1.85 2.799 .008* F4 8.03 + 2.84 3.58 + 1.71 3.015 .005* C3 2.67 + 3.53 1.02 + 1.54 3.171 .003* C4 2.68 + 3.53 1.02 + 1.54 3.162 .003* P3 11.25 + 3.33 3.42 + 1.91 3.818 .000* P4 7.99 + 3.47 3.00 + 1.97 3.030 .004* O1 14.32 + 3.05 5.94 + 1.74 3.097 .004* O2 14.32 + 2.98 5.80 + 1.84 3.166 .003* F7 12.52 + 2.73 5.31 + 1.72 3.296 .002* F8 11.36 + 2.66 4.98 + 1.62 3.312 .002* T7 10.74+ 2.67 4.1 + 1.41 4.048 .000* T8 9.39 + 2.81 4.04 + 1.48 3.336 .002* P7 16.76 + 2.91 6.16 + 1.8 3.592 .001* P8 14.75 + 2.92 6.45 + 1.94 2.886 .006* F1 123.87 + 2.72 27.24 + 5.06 3.531 .001* F2 112.54 + 2.8 49.21+ 1.8 3.060 .004* Fc1 82.83 + 3.08 37.56 + 1.96 2.648 .012 Fc2 84.67 + 3.18 32.58 + 1.81 3.216 .003* C1 42.82 + 3.53 21.06 + 1.97 2.169 .037 C2 42.12 + 3.74 17.19 + 1.69 2.758 .009* Cp1 60.69 + 3.62 20.02 + 1.67 3.501 .001* Cp2 48.13 + 3.77 19.10 + 1.62 2.858 .007* P1 116.28 + 3.45 35.73 +1.99 3.654 .001* P2 102.69 + 3.43 38.02 + 1.85 3.157 .003*  Af3 146.1 + 2.59 55.80 + 1.76 3.813 .001*  Af4 135.92 + 2.59 65.81 +1.74 2.883 .007* Fc3 68.74 + 3.35 30.01 + 1.98 2.613 .013 Fc4 65.63 + 3.16 28.19 + 1.52 3.012 .005* Cp3 68.14+ 3.46 23.41 +1.54 3.552 .001* Cp4 57.8 + 3.58 22.68 + 1.65 2.985 .005* F5 133.17 + 2.79 57.54 + 1.97 2.996 .005* F6 125.89 + 2.68 54.64 +1.61 3.333 .002* Fc5 97.01 + 3.00 36.29 + 1.53 3.645 .001* Fc6 91.09 + 2.94 39.01 + 1.52 3.202 .003* C5 81.64 + 2.93 29.85 +1.45 3.868 .000* C6 73.14 + 3.04 29.58 + 1.52 3.332 .002* Cp5 119.44 + 3.11 32.74 +1.41 4.767 .000* Cp6 83.86 + 3.2 32.81 + 1.65 3.243 .003* P5 207.04 + 3.37 60.93 + 1.99 3.842 .000* P6 130.50 + 3.11 56.28 + 2.00 2.772 .009*  Af7 234.51 + 2.79 86.25 +1.79 3.714 .001*  Af8 218.66 + 2.69 89.51 + 1.73 3.467 .001* Ft7 164.51 + 2.92 60.41 + 1.62 3.726 .001* Ft8 139.38 + 2.77 57.25 + 1.44 3.591 .001* Tp7 199.82 + 2.88 68.05 +1.54 4.125 .000* Tp8 186.74 + 3.09 66.8 + 1.73 3.588 .001* *Bonferroni adjusted significance at P < 0.01  EEG   P OWER  A NALYSIS OF OCD 33 Table 3: Group difference in alpha power ( μ V 2  /Hz) across all channels Chan-nel Patients (Mean ±SD) (N=20) Normals (Mean ±SD) (N=19) t (df =37) P* F3 15.26 + 3.23 7.13 + 3.15 2.051 .047 F4 15.54 + 3.21 6.79 + 2.95 2.297 .027 C3 6.20 + 3.35 2.26 + 2.56 2.897 .006* C4 6.24 + 3.33 2.26 + 2.57 2.921 .006* P3 41.07 + 3.29 16.12 + 4.8 2.103 .042 P4 36.22 + 3.7 15.16 + 4.65 1.910 .064 O1 47.21 + 3.19 27.77 + 4.1 1.285 .207 O2 46.74 + 3.27 28.26 + 4.12 1.205 .236 F7 18.82 + 2.91 8.31 + 2.57 2.532 .016 F8 16.83 + 2.98 7.37 + 2.51 2.549 .015 T7 21.6 + 2.75 10.00 + 2.53 2.472 .018 T8 20.22 + 2.83 9.98 + 2.84 2.113 .041 P7 53.42 + 3.18 29.58 + 4.84 1.340 .189 P8 58.55 + 3.33 36.65 + 4.64 1.064 .294 F1 237.56 + 3.29 47.42 + 9.47 2.818 .008* F2 238.53 + 3.31 100.21 + 3.35 2.249 .031 Fc1 168.11 + 3.44 78.45 + 3.55 1.903 .065 Fc2 176.78 + 3.65 71.41 + 3.3 2.267 .029 C1 91.13 + 3.59 44.64 + 3.33 1.794 .081 C2 97.76 + 3.82 36.10+ 2.88 2.565 .015 Cp1 160.53 + 3.7 50.40 + 3.08 2.960 .005* Cp2 150.42 + 3.77 57.52 + 3.27 2.381 .023 P1 436.72 + 3.43 125.61 + 4.58 2.818 .008* P2 411.91 + 3.8 145.97 + 4.16 2.346 .024  Af3 276.33 + 3.24 109.88 + 3.07 2.506 .017  Af4 269.67 + 3.22 118.68 + 3.00 2.255 .030 Fc3 132.87 + 3.19 62.00 + 3.53 1.966 .057 Fc4 132.98 + 3.4 52.85 + 2.72 2.572 .014 Cp3 191.85 + 3.41 72.26 + 3.28 2.521 .016 Cp4 176.02 + 3.52 81.79 + 3.23 1.965 .057 F5 219.36 + 3.06 88.90 + 2.65 2.684 .011 F6 213.41 + 3.03 90.03 + 2.68 2.564 .015 Fc5 168.04 + 2.98 67.84 + 2.86 2.641 .012 Fc6 152.98 + 3.26 64.92 + 2.55 2.503 .017 C5 173.28 + 2.76 65.27 + 2.62 3.077 .004* C6 145.72 + 3.15 65.22 + 2.66 2.350 .024 Cp5 298.39 + 2.98 104.91 + 3.3 2.858 .007* Cp6 238.56 + 3.31 111.97 + 3.38 1.957 .058 P5 720.83 + 3.14 293.3 + 5.31 1.970 .056 P6 568.39 + 3.33 324.86 + 5.22 1.213 .233  Af7 318.05 + 2.93 137.72 + 2.8 2.481 .018  Af8 299.74 + 2.95 122.08 + 2.65 2.718 .010 Ft7 261.6 + 2.85 108.43 + 2.75 2.668 .011 Ft8 223.65 + 2.97 93.8 + 2.49 2.693 .011 Tp7 493.29 + 2.95 243.06 + 3.91 1.800 .080 Tp8 515.89 + 3.35 254.42 + 3.97 1.705 .097 *Bonferroni adjusted significance at P < 0.01  The OCD group had significantly higher mean power across left frontal (F1), left temporal (C5, Cp5), left parietal (C5, Cp1, and P1) and right parietal (C4) electrodes in alpha (8-12 Hz) band than the control group (Table 3).  Analysis of the beta 2 (20.5-30 Hz) band revealed that the OCD group had significantly higher mean power in left frontal Af7 (t = 2.949; df = 37; P = 0.005) in comparison to the control group. In the beta 1 (12.5-20 Hz) band, no significant group diffe-rence was observed. Mann-Whitney U test revealed no significant difference of power between those patients who were drug naïve/free and those who were taking medications. Discussion Methodological Considerations Better spatial sampling is the first requirement for extracting more detailed information about cognitive processes from scalp recorded EEGs. It is well recognized that the optimum 3 dB point of the point spread function for conductance of potentials from the brain surface to the scalp averages about 2.5 cm (Gevins et al, 1995) and the 64-channel has got the benefit over the former 32-, or 19- channel record in the form of more spatial sampling of EEG data (Gevins et al, 1994). All of the quantitative EEG studies done so far on obsessive compulsive disorder have used lesser number of channels resulting in insufficient spatial resolution. The present study has improved upon the former studies by using 64-channel and thus getting satisfactory spatial resolu-tion. Unlike previous studies, the current study has strictly (pair  wise) matched age, sex, and handedness while taking controls into the study. Thus, the possible confounding influence of all these factors could be controlled.  A matter of concern in studying patients with obsessive compulsive disorder is to control the effect of comorbid depressive symptoms, which have been the most common comorbid syndrome associated with obsessive compulsive disorder (Attiullah et al, 2000) with a lifetime prevalence  which can be as high as 67% (Rasmussen & Eisen, 1988).  The current study has only included those patients in whom onset of depressive symptoms had taken place after the onset of obsessive compulsive symptoms and was essentially mild (score < 18) as per 21-item HDRS scale. Moreover, the current study has combined both visual in-spection and computer assisted algorithms for rejection of artifacts which increased the accuracy of the procedure.
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks