Onl_er_ene_2100 1.8

Event-related delta oscillatory responses of Alzheimer patients G. Yenera, B. Gu¨ntekinb and E. Bas¸arbaDepartments of Neurology and Neurosciences, Brain Dynamics and Multidisciplinary Research Center, Dokuz Eylul University, Izmir, Turkey; bBrain Dynamics, Cognition and Complex Systems Research Unit, Faculty of Science and Letters, Istanbul Ku¨ltu¨r University, Background and purpose: Alzheimer type of dementia (AD) is the most common neuropsychiatric morbidity in elderly individuals. Event-related oscillations (ERO) provide an useful tool for detecting subtle abnormalities of cognitive processes with high temporal resolution. Methods: In the present report, event-related oscillations of patients with AD were analyzed by using a visual oddball paradigm. A total of 22 mildprobable AD subjects according to NINCDS-ADRDA criteria and 20 age-, gender-, and education-matched healthy control subjects were compared. AD group consisted from 11 untreated patients and 11 patients treated with cholinesterase inhibitor.
Oscillatory responses were recorded from 13 scalp electrodes. Results: Significantdifferences in delta frequency range were seen between the groups by using repeatedmeasures of ANOVA analysis [F(9.120) = 2.228; P = 0.022]. Post-hoc analyses usingWilcoxon test showed that at mid- and left central regions, (Cz, C3) peak amplitudesof delta responses of healthy subjects were significantly higher than either group.
Also cholinesterase inhibitors did not have effect on delta oscillatory responses.
Conclusions: Our findings imply that the delta oscillatory responses at central loca-tions are highly instable in mild probable AD patients regardless of treatment whencompared to the healthy aged controls. This study supports the importance of oscil-latory event-related potentials for investigating AD brain dynamics.
Event-related theta oscillatory responses have been proposed to be related to the memory processes [12,17].
One of the leading neurological conditions most In subjects with ParkinsonÕs disease or schizophrenia, responsible for neuropsychiatric morbidity in elderly theta oscillations seem to be less than controls, indi- individuals is Alzheimer type of dementia (AD). Event- cating that these oscillations appear to be involved in related oscillations (ERO) provide a powerful tech- mnemonic networks [18,19]. Also theta responsiveness nique, with high temporal resolution, which can be used in frontal lobes is interpreted as an indication of the as a tool for detecting subtle abnormalities of cognitive function of the hippocampo-fronto-parietal system processes [1,2]. It has been well known for several during cognitive processes [20,21]. Our recent report decades that P300 is attenuated in AD. However, the evaluating phase locking of the visual event-related full potential of electrophysiological methods in helping theta oscillations indicated that untreated AD group to predict [3–5], to diagnose [6–10], and to monitor has lower phase locking than controls at left frontal either treatment or progress [11] in AD patients has not region and the cholinesterase inhibitor treatment in- been reflected into routine clinical practice.
creases phase locking in theta frequency ranges similar Event-related oscillatory activity in various frequency to controls [22]. The question whether cholinergic bands may reflect different aspects of information mechanisms affect or modulate event-related oscilla- processing [1,2]. Alpha oscillatory responses increase tions in other frequency ranges still remains to be with simple memory tasks and decrease with demand- clarified. Investigating these oscillations may help to ing memory tasks [12,13]. Beta oscillatory responses are understand differences in brain dynamics of AD important in attention related tasks in cats and [14], recognition of facial expression in humans [15,16].
We hypothesized that the AD group would show lower oscillatory responses than controls. In thisreport, we aimed to compare the peak amplitudes ofevent-related Correspondence: Go¨rsev G. Yener, M.D., Department of Neurology frequency ranges in AD subjects, either the untreated and Neurosciences, Dokuz Eylul University, Izmir 35340, Turkey or those on cholinesterase treatment, to those of (tel.: +90 232 412 4050; fax: + 90 232 277 7721; e-mail: gorsev.
[email protected]).
Ó 2008 The Author(s)Journal compilation Ó 2008 EFNS went a cognitive and a complete neurological, neuro- imaging (CT or MRI) and laboratory examinationincluding blood glucose, electrolytes, liver and kidney function tests, full blood count, erythrocyte sedimen- We conducted a prospective open study. Twenty-two tation rate, thyroid hormone, vitamin B12, HIV, consecutive, community-dwelling patients suffering VDRL. Healthy controls were recruited from various from dementia according to the DSM IV criteria and community sources; none of them were consanguineous also with the diagnosis of probable Alzheimer disease to the patients. The study was approved by the local according to the NINCDS-ADRDA criteria [23] were ethics committee. All subjects and relatives gave written included in the study. AD group was divided into two groups as the treated and the untreated. In the treatedAD group, eleven subjects (four males, seven females) were taking only cholinesterase inhibitors (AChEI) as apsychotropic agent for 3–6 months including the titra- A classical visual oddball paradigm was used in the tion period (eight subjects were on donepezil 10 mg/day experiments. Two types of stimuli were used: the stan- with the initial dose of 5 mg/day that was titrated to dards and the deviants. The probability of the deviant 10 mg/day by 4 weeks, and three subjects were on riv- stimuli was 0.20 and that of standard stimuli 0.80. As astigmine 6–9 mg/day with the initial dose of 3 mg/day, stimulation we used a white screen with a luminance of titrated by every 4 weeks either to 6 mg/day or to 9 mg/ 35 cd/cm2 for standard signals. The luminance of the day depending on the tolerance of the drug) and eleven deviant stimuli were 20% lower (i.e. 28 cd/cm2). The AD patients (four males, seven females) not taking any rise-time of the stimulation signal was 10 ms, the psychotropic medication comprised the untreated AD duration of the stimulation was 1 second. In all the group. Both AD groups did not differ from each other paradigms, the deviant stimuli were embedded ran- regarding FolsteinÕs Mini-Mental State Examination domly within a series of standard stimuli. The appli- (MMSE) scores, ReisbergÕs Global Deterioration Scale cation of the signal including the rise-fall time and the (GDS), gender, education, age, or handedness as shown duration occurs electronically and is supported by a in Table 1. Time from the onset of symptoms was be- MATLAB program. Further, the rise time and the dura- tween one and two years in both AD groups. The tion of the signal were also checked by means of a MMSE scores of all AD subjects ranged between 20 photo-sensor recorded in a storage oscilloscope. The and 24, whereas those of healthy subjects were between task required was mental counting of the target stimuli.
28 and 30 points. All of the AD subjects were on stage 4 These stimulation signals were applied randomly with according to the GDS. In treated AD group, the the inter-stimulus intervals varied between 3 and 7 s.
majority (8 out of 11 subjects) was ÔresponderÕ defined During the elicitation period of event-related oscilla- as Ôat least 1 MMSE point increaseÕ 3 months after tions, all subjects had displayed enough accuracy of onset of treatment, while three showed decrease in their mental count of target stimuli, with being slightly worse scores. Twenty-two healthy elderly control subjects in both groups of AD than that of controls.
volunteered for the study, two subjects were excludedfor motor artifacts, remaining 20 control subjects (12 males, 8 females) were not significantly differentfrom both AD groups regarding age, gender, handed- The EEG was recorded from F3, F4, Cz, C3, C4, T3, T4, ness and education (Table 1). All AD subjects under- T5, T6, P3, P4, O1 and O2 locations according to the SD, standard deviation; NS, non-significant; M, male; F, female; L, left; R, right; GDS,ReisbergÕs Global Deterioration Scale. aChi-square test; bKruskal–Wallis and post-hoc LSDtests.
Journal compilation Ó 2008 EFNS European Journal of Neurology Event-related oscillation of Alzheimer patients International 10–20 system. For the recordings an As oscillatory responses, we measured the peak-to- EEG-CAP was used. For the reference, EMG and EOG peak amplitudes of each subjectÕs averaged responses recordings Ag/AgCl electrodes were used. Linked ear- filtered in the frequency ranges of delta, theta, alpha, lobe electrodes (A1 + A2) served as reference. EOG and beta. The post-stimulus time intervals for peak from medial upper and lateral orbital rim of the right amplitudes of oscillatory responses were chosen as eye was also registered. The EEG was amplified by follows: frequency ranges of delta and theta, 0– means of a Nihon Kohden EEG-4421 G apparatus with 600 msn; of alpha and beta 0–250 msn. According to band limits 0.1–100 Hz 24 dB/octave. The EEG was the literature of brain oscillations and basic principles digitized on-line with a sampling rate of 512 Hz and a of systems theory, the range of oscillatory signals are total recording time of 2000 ms, 1000 ms of which chosen in correlation to the frequency signal studied served as the pre-stimulus baseline.
Computation of selectively averaged ERPs and digital Statistical Package for the Social Sciences (SPSS) was Before the averaging procedure, the epochs with arti- used for statistical analysis. Peak-to-peak maximum facts were rejected by an off-line technique. In the off- amplitude responses were separately analyzed for each line procedure, single sweep EOG recordings were frequency band by means of a repeated measure visually studied and trials with eye-movement or blink ANOVA including the between subjects factor as artifacts were rejected. Subject averages and grand groups (healthy aged controls, untreated AD, treated averages were calculated for each electrode site and AD) and the within subject factor location (F3, F4, Cz, experimental condition. The data was digitally filtered C3, C4, T3, T4, T5, T6, P3, P4, O1 and O2). Greenhouse- according to determined frequency bands of interest.
Geisser corrected P-values have been taken into con- In the present study, two approaches were taken in sideration. Post-hoc analysis was conducted using Wil- determining the frequency responses of the system: the transient response frequency characteristics (TRFC)method and digital filtering (DF) method.
Filtering produces visual displays of the time courses of oscillatory components within the frequency limits of The peak amplitudes of oscillatory responses in delta, the utilized filters. The digital filters are advantageous theta, alpha and beta frequency ranges were measured.
because they do not produce the phase shifts that are a The only difference in peak amplitudes between groups characteristic of electronic filters. The digital filtering was seen in delta oscillations (Table 2).
was employed in the present study for the digital pass- For the delta frequency range digital filtering was band filtering of the event-related potentials (ERPs) determined between 0.5 and 3.5 Hz according to the and thus to demonstrate the event-related oscillations AFC. Oscillatory delta responses showed significant (EROs) in selected frequency-bands (delta: 0.5–3.5 Hz, differences between event-related responses of healthy theta: 4–7 Hz, alpha: 8–13 Hz, and beta: 15–30 Hz) [2].
aged controls, untreated and treated AD subjects. The The numerical evaluation of the frequency charac- main differences are observed at the central electrode teristics was accomplished using a Fast Fourier trans- locations: From the values of Table 3, one can recog- form (FFT) of the following form: Let Xn be a discrete nize that in central locations the peak-to-peak ampli- time series (Xn = X (nDt), T = ((N - 1) Dt). Then the tudes can be 50–100% larger for controls than for AD The ANOVA on delta oscillatory responses revealed [F(9.120) = 2.228; P = 0.022] indicating higher deltaresponse in controls. Post-hoc comparisons using the where Yk = ak + ibk are the complex Fourier Wilcoxon paired sample test revealed that the peak-to- coefficients whose geometric mean is the amplitude peak delta response was significantly larger for controls spectrum. According to the results of the amplitude than for either treated or untreated AD subjects over frequency characteristics (AFC) the frequencies of left and mid-central electrodes (P < 0.05 for all com- interest were determined and the frequency ranges for parisons) (Figs 1 and 2 and Table 2).
the digital filtering defined. For the frequency ranges The single sweeps of a typical healthy aged control grand averages were computed based on single subjectsÕ subject show good congruence and accordingly phase averages of the AFCs for each condition and location.
locking of delta responses at Cz, whereas responses of Ó 2008 The Author(s)Journal compilation Ó 2008 EFNS European Journal of Neurology Table 2 The mean (SD) peak amplitudes (lV) of oscillatory activities in specific frequency bands in treated AD (t-AD), untreated AD (u-AD) andhealthy elderly controls (Cont). The statistically significant (P < 0.05) results were indicated in bold style 5.10 (1.91) 4.32 (2.32) 4.45 (2.37) 5.27 (2.76) 4.57 (1.90) 5.34 (3.15) 5.05 (2.79) 4.16 (1.53) 3.93 (1.74) 6.19 (2.86) 5.90 (2.09) 8.08 (3.21) 5.02 (2.48) 4.40 (1.60) 4.83 (2.26) 5.76 (2.66) 5.21 (2.55) 5.45 (3.02) 5.74 (3.18) 4.05 (1.83) 3.74 (1.75) 6.56 (2.86) 5.43 (2.09) 7.51 (3.11) 4.01 (1.14) 3.56 (1.27) 3.78 (1.70) 5.53 (3.46) 4.60 (2.06) 5.79 (2.85) 5.74 (3.13) 4.43 (1.05) 4.45 (2.22) 5.52 (1.64) 4.77 (1.93) 8.38 (3.38) 4.28 (2.02) 3.97 (1.27) 3.79 (2.05) 5.64 (3.22) 4.48 (2.04) 5.52 (2.63) 4.97 (2.93) 3.84 (1.09) 3.98 (2.03) 4.48 (2.03) 3.45 (1.76) 7.25 (3.15) 4.40 (1.60) 5.02 (2.48) 4.83 (2.26) 6.12 (2.58) 6.11 (3.09) 5.94 (3.53) 4.74 (2.61) 4.10 (1.73) 4.17 (2.37) 5.97 (2.42) 4.91 (1.06) 7.28 (3.92) 3.85 (1.25) 4.37 (2.55) 3.85 (2.30) 3.42 (2.27) 3.49 (2.06) 3.75 (2.38) 3.00 (1.82) 2.18 (0.69) 2.66 (1.75) 2.94 (1.03) 2.73 (1.15) 4.52 (2.21) 4.90 (2.12) 4.96 (4.01) 3.67 (1.76) 3.97 (1.23) 4.35 (3.38) 4.38 (2.27) 3.15 (1.70) 2.26 (1.40) 2.59 (1.17) 3.96 (2.13) 2.70 (0.85) 4.35 (2.34) 4.05 (1.90) 4.35 (2.05) 4.35 (2.93) 4.14 (1.74) 4.21 (2.09) 3.68 (2.04) 4.39 (2.46) 2.93 (1.10) 3.01 (1.83) 4.40 (1.92) 3.49 (2.11) 5.39 (2.08) 4.34 (2.06) 4.63 (5.77) 4.95 (3.10) 6.87 (6.05) 4.21 (4.17) 5.16 (3.67) 4.40 (2.58) 2.97 (0.55) 3.80 (2.46) 4.19 (1.50) 3.91 (1.57) 5.56 (3.26) 3.67 (0.87) 4.52 (2.29) 4.54 (2.44) 6.39 (3.22) 4.61 (2.60) 5.00 (2.75) 5.94 (1.76) 2.71 (1.76) 4.05 (2.61) 3.63 (1.42) 3.61 (1.36) 4.98 (2.75) 7.02 (4.70) 4.51 (2.62) 5.05 (2.53) 5.04 (1.72) 2.85 (1.20) 4.34 (3.14) 5.00 (1.63) 4.90 (2.33 6.29 (4.37) O1 4.71 (3.38) 5.45 (2.59) 4.79 (2.82) 5.63 (2.53) 4.65 (2.22) 5.17 (2.84) 6.70 (2.93) 4.39 (2.34) 5.12 (3.28) 8.40 (5.91) 5.74 (2.67) 5.69 (3.25)O2 4.54 (2.04) 4.15 (1.83) 4.87 (2.77) 5.43 (2.03) 4.29 (1.85) 4.98 (2.45) 5.91 (3.60) 4.06 (1.52) 5.20 (3.12) 8.91 (6.34) 5.58 (2.40) 5.92 (3.81) Table 3 Mean values and standard deviations of delta oscillatory may play an integrative role in cognitive information responses and comparisons of the treated AD (t-AD), untreated AD and in generation of P300 [30]. The possibility of vol- (u-AD) and healthy elderly controls (Cont) groups with post-hoc ume conduction from C3 layer of the hippocampus to the cortex in the generation of P300 is excluded as animal experiments showed [31]. In human intracranial recordings, visual working memory task activates briefly visual association cortex and then activation soon spreads at once to multiple occipital, parietal andfrontal sites, which all remains active for the entire C3 7.25 (3.15) 4.48 (2,03) 3.45 (1.76) )2.134 0.033 )2.490 0.013 epoch. Phase-locked oscillations in theta and alpha 8.38 (3.38) 5.51 (1.64) 4.77 (1.93) )2.045 0.041 )2.667 0.008 frequency ranges are prominent in multiple structuresincluding the prefrontal cortex [32].
an AD subject do not (Fig. 1). Phase locking indicates a Further, according to several authors event-related stronger response to a given stimulus [25].
potentials arise by superposition of event-related oscil- In the grand-averages of delta oscillatory responses lations in various frequency ranges [1]. These ap- at Cz, it is shown that the control group has larger proaches hypothesize that the EEG consists of the amplitude in comparison to either treated or untreated activity of an ensemble of generators producing oscil- AD groups (Fig. 2). Peak-to-peak amplitudes of the latory activity in several frequency ranges. These control group are 7.25 (3.15) and 8.38 (3.38) lV in C3 oscillators are active usually in a random way. How- and Cz locations, showing a regular oscillatory pattern.
ever, by application of sensory stimulation these gen- On the contrary, treated and untreated AD subjects erators couple and act together in a coherent way.
have smaller amplitudes with an irregular shape Evoked potentials representing ensembles of neural population responses were considered as a result oftransition from a disordered to an ordered state [25].
Among event-related oscillations, theta (4–8 Hz) oscil- lations are correlated with memory load, task difficultyor recognition of previous stimuli [12,13,19,33]. Oscil- lations at delta frequency range are related to Ôfocused According to a group of authors, ERPs or P300 re- attentionÕ, Ôsignal detectionÕ, ÔrecognitionÕ and Ôdecision sponses are generated in the neocortex, especially in makingÕ [29,34,35]. In these reports late theta responses frontal locations [26] or centroparietal/temporoparietal behave similarly to delta oscillatory responses. Brain association cortices [27]. Involvement of limbic system oscillations in lower frequencies are proposed to play a or hippocampal formation in the generation of P300 role in mediating long range interactions [36]. In has been also proposed [28,29]. Intracranial recordings agreement with this, simulation studies have indicated also suggested that basal ganglia, especially putamen, that lower frequencies such as delta or theta oscillations Journal compilation Ó 2008 EFNS European Journal of Neurology Event-related oscillation of Alzheimer patients Figure 2 Grandaverages of delta oscillatory response of eachgroup to the target stimuli elicited by a classical visual oddballparadigm recorded from electrodes of C3 and Cz. (a) The healthyelderly control group (n = 20). (b) The untreated Alzheimergroup (n = 11). (c) The treated (cholinesterase inhibitor) Alzhei-mer group (n =11).
oddball paradigms and may be related to signal detec-tion and decision making [20]. As the major shapedetermining oscillatory activity of P300, delta responsesare related to basic information processing mechanismsof attention allocation and immediate memory [39].
Since memory and complex attention functions arehighly reduced in AD [40], our results are in accordancewith cognitive deficits from the psychophysiologicviewpoint.
Topologic distribution and frequency ranges of brain Earlier reports have shown that the P300 amplitudesare decreased in AlzheimerÕs disease [8,10,39]. Reportson AD using other functional methods such as PET,SPECT or f-MRI also have a tendency to show deficitsat left centro-frontal, left temporo-parietal locations. Arecent report on the event-related oscillatory activity inAD has reported that the significant differences havebeen noted in peak amplitudes of alpha oscillatoryactivity (7–17 Hz) over frontal, central and left tem-poral electrodes [9].
In the present paper, we also found significant dif- ferences between healthy controls and two groups ofAD subjects in delta oscillatory responses regardless ofcholinergic medication. This difference was insisting Figure 1 Examples from each group showing single sweeps in prominently over C3 and Cz in both AD groups in delta oscillatory frequency range, to the target stimuli elicited by a classical visual oddball paradigm recorded from the scalp elec- In our study, classical oddball paradigm was used. In trode of Cz. The black and thick line indicates the average of single this task, mental counting of visual target stimuli is sweeps, and the grey and thin lines show each single sweep for the considered to be related to memory and complex subject. (a) An elderly healthy control. (b) An untreated Alzheimer attention functions. Major reduction in working mem- subject. (c) A treated (cholinesterase inhibitor) Alzheimer subject.
ory and complex attention observed in Alzheimer pa-tients may be possibly correlated with reduction of are better suited to sustain long range synchronization electrical response in central regions. However we [37]. Not only thalamic neurons but also cortical neu- cannot exclude the possibility of higher error rates in rons may discharge in the slow frequency range as delta detecting target stimuli of AD group may have lead to [38]. The amplitude of delta response increases during reduced responses in delta activity.
Ó 2008 The Author(s)Journal compilation Ó 2008 EFNS European Journal of Neurology Delta difference in left and mid-central positions are mechanism is yet obscure. Although a great number of also in accordance with earlier reports of AD subjects studies establish that both components are physiologi- studied with fMRI or PET reflecting mainly frontal or cally separable, we have another argument: This is the cingulate regions of left hemisphere. Our recent and selectivity based on the application of a pharmacolog- other earlier reports imply that in AD, either effects of ical agent enhancing cholinergic transmission. We also disease or response to treatment can be more readily mention here the selectivity of another pharmacological seen over the left frontal hemisphere.
agent such as valproate with GABAergic or glutama-tergic activity which reduces delta responses in bipolaraffective disorder [44].
Differentiation of delta and theta oscillations in AD Certainly, it can be stated that in cognitive functions of Alzheimer patients there is a high decline in working Earlier functional imaging studies in AD showed that memory and complex attention [37,45,46]. Since cho- after administration of AChEI, clinical responders to linergic medication improves theta phase locking, but treatment selectively display improvements mainly over not delta oscillatory response; other transmitters such left prefrontal areas or left anterior cingulate [41].
as serotonin which increases delta activity in animal Cortical acetylcholine (ACh) is hypothesized to mediate studies [47] may help to enhance delta oscillatory the subjectsÕ abilities to select stimuli and associations for further processing. The ability of prefrontal cortex The present paper opens new conjecture to search a to regulate transmission in more posterior cortical re- new type of physiological intervention to restore gions may represent a Ôtop-downÕ mechanism to control reduction of delta response in central regions. This attention [42]. Basal forebrain is the main source of question cannot be answered by this study, but re- ACh in the neocortex and Alzheimer patients show depletion in cortical ACh due to degeneration of basal There are a few conclusions and remarks related to our forebrain early in the course of illness [43].
findings on AD patients at the initial phase of disease: We believe that delta oscillatory responses are not 1. Amplitudes of delta oscillatory responses are lower affected by cholinergic agents, because in our AD in Alzheimer disease regardless of medication over group, the degree of clinical impairment did not differ left and mid-central regions. Activation upon between the treated and untreated; and subjects in the treated group was not more advanced than the un- oscillatory response but not in delta frequency treated. Finally, majority of the treated group was considered as responder to treatment. However, a 2. In a way, these two slow oscillatory activities behave randomized controlled study can give a more probabi- separately upon application of cholinergic agents.
Possibly, the separation of delta and theta oscilla- Our recent report evaluating phase locking of the tory response in AD patients on cholinergic medi- visual event-related theta oscillations indicated that cation will gain high importance in future similar untreated AD group has lower phase locking than controls at left frontal region. However, the treated AD 3. The studies on event-related oscillations may help group showed phase locking in theta frequency range for the diagnostic purposes and also for monitor- similar to controls [22]. In the present report, peak ing the effects of pharmacological agents, therefore amplitudes of delta oscillatory responses are highly re- in evaluating the transmitter effects.
duced in AD regardless of cholinergic treatment.
Therefore cholinergic agents seem to have differentiated effect on delta and theta responses in AD subjects. Inother words, phase locking in theta oscillatory response 1. Bas¸ar E. EEG-Brain Dynamics. Relation between EEG and Brain Evoked Potentials. Amsterdam: Elsevier, 1980.
may be sensitive to cholinergic interventions in AD, 2. Bas¸ar E. Memory and Brain Dynamics. Oscillations whereas amplitudes of delta oscillatory responses are Integrating Attention, Perception, Learning, and Memory.
One might question whether the separation of delta 3. Babiloni C, Ferri R, Binetti G, et al. Fronto-parietal and theta responses is a natural way of decomposition.
coupling of brain rhythms in mild cognitive impairment: amulticentric EEG study. Brain Research Bulletin 2006; 69: If delta and theta responses would behave in a similar way such a separation could not be strongly assumed.
4. Cichocki A, Shishkin SL, Musha T, et al. EEG filtering These selective responses to pharmacological agents based on blind source separation (BSS) for early detection demonstrate the independent functional correlates of of AlzheimerÕs disease. Clinical Neurophysiology 2005; delta and theta responses. However, the underlying Journal compilation Ó 2008 EFNS European Journal of Neurology Event-related oscillation of Alzheimer patients 5. Rossini PM, Del Percio C, Pasqualetti P, et al. Conver- Alzheimer patients treated with cholinesterase inhibitors.
sion from mild cognitive impairment to AlzheimerÕs International Journal of Psychophysiology 2007; 64: 46–52 disease is predicted by sources and coherence of brain electroencephalography rhythms. Neuroscience 2006; 143: 23. McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of AlzheimerÕs disease: report of the NINCDS 6. Yener GG, Leuchter AF, Jenden D, et al. Quantitative ADRDA Work Group under the auspices of Department EEG in frontotemporal dementia. Clinical Electroen- of Health and Human Services Task Force on AlzheimerÕs Disease. Neurology 1984; 34: 939–44.
7. Babiloni C, Benussi L, Binetti G, et al. Genotype (cystatin 24. Bas¸ar E. Brain Function and Oscillations: I. Brain Oscil- C) and EEG phenotype in Alzheimer disease and mild lations. Principles and Approaches. Heidelberg, New York: cognitive impairment. Neuroimage 2006; 29: 9948–964.
8. Polich J, Herbst KL. P300 as a clinical assay: rationale, 25. Yordanova J, Kolev V. Single sweep analysis of the theta evaluation, and findings. International Journal of Psycho- frequency band during an auditory oddball task. Psy- 9. Karrasch M, Laine M, Rinne JO, et al. Brain oscillatory 26. McCarthy G, Wood CC. Scalp distributions of event–re- responses to an auditory-verbal working memory task in lated potentials: an ambiguity associated with analysis of mild cognitive impairment and AlzheimerÕs disease.
variance models. Electroencephalography and Clinical International Journal of Psychophysiology 2006; 59: 168– Neurophysiology 1985; 62: 203–208.
27. Verleger R, Heide W, Butt C, et al. Reduction of P3b in 10. Jeong J. EEG dynamics in patients with AlzheimerÕs dis- patients with temporo-parietal lesions. Cognitive Brain ease. Clinical Neurophysiology 2004; 115: 1490–1505.
11. Jelic V, Johansson SE, Almkvist O, et al. Quantitative 28. Wood CC, Allison T, Goff WR, et al. On the origin electroencephalography in mild cognitive impairment: of P300 in man. Progress in Brain Research 1980; 54: longitudinal changes and possible prediction of Alzhei- merÕs disease. Neurobiology Aging 2000; 21: 533–540.
29. Halgren E, Smith ME. Cognitive evoked potentials as 12. Jensen O, Tesche CD. Frontal theta activity in humans modulatory processes in human memory formation and increases with memory load in a working memory task.
retrieval. Human Neurobiology 1987; 6: 129–139.
European Journal of Neurosciences 2002; 15: 1395–1399.
30. Rektor I, Bares M, Kanovsky P, et al. Cognitive poten- 13. Gevins A, Smith ME, McEvoy L, et al. High resolution tials in the basal ganglia-frontocortical circuits. An EEG mapping of cortical activation related to working intracerebral recording study. Experimantal Brain Re- memory: effects of task difficulty type of processing, and practice. Cerebral Cortex 1997; 7: 374–385.
31. Bas¸ar-Eroglu C, Bas¸ar E. A compound P300-40 Hz re- 14. Wrobel A. Beta activity: a carrier for visual attention.
sponse of the cat hippocampus. International Journal of Acta Neurobiologiae Experimentalis 2000; 60: 247–260.
15. Gu¨ntekin B, Bas¸ar E. Emotional face expressions are 32. Halgren E, Boujon C, Clarke J, et al. Rapid distributed differentiated with brain oscillations. International Journal fronto-parieto-occipital processing stages during working of Psychophysiology 2007; 64: 91–100 (e-pub: 5 December memory in humans. Cerebral Cortex 2002; 12: 710–728.
33. Klimesch W, Hanslmayr S, Sauseng P, et al. Oscillatory 16. O¨zgo¨ren M, Basar-Eroglu C, Bas¸ar E. Beta oscillations in EEG correlates of episodic trace decay. Cerebral Cortex face recognition. International Journal of Psychophysiol- 34. Stampfer HG, Bas¸ar E. Does frequency analysis lead 17. Klimesch W, Doppelmayr M, Pachinger T, et al. Brain to better understanding of human event related poten- oscillations and human memory performance: EEG cor- tials. International Journal of Neuroscience 1985; 26: relates in the upper alpha and theta bands. Neuroscience 35. Demiralp T, Bas¸ar E. Theta rhythmicities following ex- 18. Schmiedt C, Meistrowitz A, Swendemann G, et al. Theta pected visual and auditory targets. International Journal of and alpha oscillations reflect differences in memory Psychophysiology 1992; 13: 147–160.
strategy and visual discrimination performance in patients 36. von Stein A, Sarnthein J. Different frequencies for dif- with ParkinsonÕs disease. Neuroscience Letters 2005; 388: ferent scales of cortical integration: from local gamma to long range alpha-theta synchronization. International 19. Schmiedt C, Brand A, Hildebrandt H, et al. Event-related Journal of Psychophysiology 2000; 38: 301–313.
theta oscillations during working memory tasks in pa- 37. Kopell N, Ermentrout GB, Whittington MA, et al.
tients with schizophrenia and healthy controls. Cognitive Gamma rhythms and beta rhythms have different syn- chronization properties. Proceedings of the National 20. Bas¸ar-Erog˘lu C, Bas¸ar E, Demiralp T, et al. P300-re- Academy of Sciences USA 2000; 97: 1867–1872.
sponse: possible psychophysiological correlates in delta 38. Steriade M, Gloor P, Llinas RR, et al. Basic mechanisms and theta frequency channels. A review. International of cerebral rhythmic activities. Electroencephalography Journal of Psychophysiology 1992; 13: 161–179.
and Clinical Neurophysiology 1990; 76: 481–508.
21. Demiralp T, Bas¸ar-Erog˘lu C, Rahn E, et al. Event-related 39. Polich J, Kok A. Cognitive and biological determinants of theta rhythms in cat hippocampus and prefrontal cortex P300: an integrative review. Biological Psychology 1995; during an omitted stimulus paradigm. International Journal of Psychophysiology 1994; 18: 35–48.
40. Cummings JL, Miller BL, Hill MA, et al. The neuropsy- 22. Yener GG, Gu¨ntekin B, O¨niz A, et al. Increased frontal chiatric aspects of multi-infarct dementia and Alzheimer type. Archives of Neurology 1987; 44: 389–393.
Ó 2008 The Author(s)Journal compilation Ó 2008 EFNS European Journal of Neurology 41. Mega MS, Dinov ID, Porter V, et al. Metabolic patterns stimuli in a group euthymic bipolar patients in comparison associated with the clinical response to galantamine to healthy controls. 62nd Annual Scientific Convention and therapy. Archives of Neurology 2005; 62: 721–728.
Program, May 17–May 19, 2007, San Diego, California, 42. Sarter M, Hasselmo ME, Bruno JP, et al. Unraveling the attentional functions of cortical cholinergic inputs: inter- 45. Miller BL, Read SL, Mahler ME, et al. Altered mental actions between signal- driven and cognitive modulation status in the elderly. Primary Care 1984; 11: 653–665.
of signal detection. Brain Research Reviews 2005; 48: 98– 46. Mesulam M-M. Patterns in behavioral neuroanatomy; association areas, the limbic system, and hemispheric spe- 43. Perry EK, Irving D, Kerwin JM, et al. Cholinergic cialization. In: Mesulam M-Med. Principles of Behavioral transmitter and neurotrophic activities in Lewy body Neurology. Philedelphia, F. A. Davis, 1985: 1–70.
dementia: similarity to ParkinsonÕs and distinction from 47. Schu¨tt A, Bas¸ar E. The effects of acetylcholine, dopamine Alzheimer disease. Alzheimer Disease and Associated and noradrenaline on the visceral ganglion of Helix pomatia. II. Stimulus evoked field potentials. Comparative 44. O¨zerdem A, Kocaaslan S, Tunca Z, et al. Effect of val- Biochemistry and Physiology – Part C: Toxicology and proate on oscillatory delta frequency responses to visual Journal compilation Ó 2008 EFNS European Journal of Neurology

Source: http://braindynamics.iku.edu.tr/Pdf/Alzdelta.pdf