MEASUREMENT OF MENTAL FATIGABILITY BY TASK RELATED SPECTRAL EEG. A PILOT STUDY

ELÕZETES KÖZLEMÉNY MEASUREMENT OF MENTAL FATIGABILITY BY TASK RELATED SPECTRAL EEG. A PILOT STUDY Peter Rajna 1, Zoltan Hidasi 1, Ivan Pal 2, Eva Csibri 1, Judit Veres 1, Balint Szuromi 1 1 Semmelweis University, General Medical Faculty, Department of Psychiatry and Psychotherapy, Budapest 2 Doctoral School No.4., Semmelweis University, Budapest A MENTÁLIS FÁRADÉKONYSÁG VIZSGÁLATA FELADATFÜGGÔ SPEKTRÁLIS EEG-MÓDSZERREL. ELÔZETES KÖZLEMÉNY Rajna P, MD, PhD, DSc; Hidasi Z, MD; Pal I, MD; Csibri E, MD; Veres J; Szuromi B, MD Ideggyogy Sz 2009;61(1 2):36 40. Background Task related EEG spectra are promising markers of mental activity. But the cooperation of the patients necessary for the registration limits its application in the neuro-psychiatry. Methods EEG difference spectra on counting (EDSC) was developed to detect the effect of a short calculation task on the spectral EEG. The originality of the task situation is a continuous mental work in a very short period of time, while the level of task difficulty is adapted to the patient s actual mental capacity. While the rest pre-task and the post task EEG sections were compared, the results show the mental EEG fatigability caused by the short intensive cognitive activity. The first preliminary results have been demonstrated by a comparative study of two healthy and three patient (probable Alzheimer disease, post-stroke state without mental deficit and mixed type of dementia) groups. Results Similarly to the findings of other authors, in addition to the differences of the alpha band seen on the temporo-parieto-occipital regions, the frontal localization and the beta band seem to be prominent, too. Demented patients had stronger EEG reactions than post-stroke patients without mental deficits and healthy elder persons had more extensive changes than the younger ones. Conclusions The test can be considered as indirect marker showing the different mental fatigability in diverse pathological conditions and during the aging process. Effect of therapeutic processes can also be followed based on key-lock principle. Standardization of the test is essential for the introduction of EDSC to the every-day routine of clinical neuropsychiatry. Keywords: mental fatigability, dementia, EEG spectra, Alzheimer disease, calculation Bevezetés A feladatfüggô spektrális EEG-változások alkalmasak a mentális aktivitás regisztrálására. A betegek együttmûködésének akadályozottsága miatt azonban a módszer neuropszichiátriai alkalmazása korlátozott. Vizsgálati személyek és módszerek Számolási feladat során rögzített EEG-különbség-spektrummódszert (EDSC) fejlesztettünk ki, amely abban különbözik az eddig alkalmazottaktól, hogy a feladat rövid tartamú, és nehézsége adaptálható a betegek aktuális mentális állapotához. Mivel a feladat elôtti és utáni spektrális EEG-változásokat vetjük össze, az eredmények valójában a rövid intenzív kognitív feladat közben megjelenô mentális fáraszthatóságot tükrözik. Elsô elôzetes eredményeinket ismertetjük két egészséges és három betegcsoport összehasonlító vizsgálatával. Eredmények Az irodalmi adatokkal megegyezôen a domináns változások a temporo-parieto-occipitalis alfa-teljesítményben, ezenkívül a béta-tartományban, illetve a frontális területeken alakultak ki. A demens betegek, illetve az idôs egészségesek markánsabb változásokat mutattak, mint a stroke-ot elszenvedett, de mentálisan ép betegek, illetve a fiatal egészséges személyek. Következtetések Elôzetes eredményeink felvetik, hogy a fenti módszerrel követni lehet a mentális fáradékonyság alakulását patológiás állapotokban és az idõsödés során. A kulcs-zár elv szerint a differenciaspektrumok akár a terápiás folyamatok eredményességének mutatói is lehetnek. A módszer klinikai rutineljárássá fejlesztéséhez azonban további standardizációs vizsgálatokra van szükség. Kulcsszavak: mentális fáradékonyság, dementia, spektrális EEG, Alzheimer-kór, számolás Levelezô szerzô (correspondent): Dr. Rajna Péter, Semmelweis Egyetem, Általános Orvostudományi Kar, Pszichiátriai és Pszichoterápiás Klinika; H-1083 Budapest, Balassa u. 6. Telefon: (1) 210-0330. E-mail: rajna@psych.sote.hu Érkezett: 2008. október 17. Elfogadva: 2008. október 30. www.lam.hu 36 Rajna: Mental fatigability by EEG spectra

Healthy and ill organs should react differently to the everyday exercise. This statement is completely available and already evidenced concerning the cardiac regulation for ages. Consequently the comparison of the performance of electrocardiogram before and after physical exercise is a simple and favourite method in cardiology. Because of their high sensitivity to functional or metabolic changes, spectral EEG parameters are at least theoretically able to demonstrate changes of the electric activity of the cerebral cortex caused by cognitive tasks. Although the inhibition of the alpha activity due to eye closure is a well known physiological reaction from the beginnings, application of any task related EEG test (TRET) still does not exist in the daily routine. One of the most difficult technical problems of TRET is the elimination of interindividual variability and the effect of external stimuli. While chemical effects and alcohol consumption can unavoidably lead to spectral EEG alterations 1, power spectra analysis is able to show changes to biological effects like fatigue as well 2. Therefore the question arises whether the transitory EEG changes would be aspecific. But according to other authors just the difference spectra are able to eliminate the majority of long term and actual interindividual variability of the cognitive EEG 3. The metaanalysis of TRET is mainly limited by the big differences of the tasks in the studies. In our recent preliminary study we introduce our new method of mental exercise and its spectral EEG as a marker of mental fatigability to the clinical practice. We evaluate our preliminary results in healthy persons and patients. Methods We demonstrate here the EEG difference spectra on counting (EDSC) in two healthy [of younger (YH) and elder age (EH)] and three patient [probable Alzheimer disease (ALZ), mixed type of dementia (MTD) and post-stroke without mental deficit (STR)] groups. Each of them contains five persons, four of alpha and one of non-alpha background activity. Table 1. demonstrates the summarizing data of age and gender in the five groups and the inclusion criteria of the three patient populations. REVERSE COUNTING TASK METHOD AND EEG RECORDING It was found that reverse counting of constant duration but a facultative level of difficulty satisfy the requirements outlined above. The subject has to count backwards strictly for 45 seconds. Details of the counting tasks, levels of difficulty, also the EEG registration protocol and the off line analysis for constructing EDSC were published in the previous papers of our team 4 6. Table 1. Demographic data of the investigated populations and inclusion criteria of patients groups including type of EEG background YH EH STR ALZ MTD No. of persons 5 5 5 5 5 Age (min-max) y 32 37 56 84 34 77 58 75 41 85 Age (average) y (34) (63.8) (67.4) (63.2) (64.8) Gender (f/m) 0/5 3/2 2/3 3/2 2/3 MMSE >24 >24 >24 <20 <20 Skull CT/MRI 1 not done not done done 2 done 3 done 4 Stroke in the history 5 + +/ Regular usage of psychoactive medicines +/ +/ +/ Progression of symptoms Background activity on the EEG α-type 4 4 4 4 4 Non-α-type 1 1 1 1 1 Abbreviations: YH = younger healthy, EH = elder healthy, STR = post stroke state without mental deficit, ALZ = probable Alzheimer disease, MTD = mixed type of dementia but not Alzheimer disease. Index numbers: 1 in one year 2 proving vascular lesion 3 no more than one small focal (vascular) lesion 4 abnormal 5 for at least 3 months prior to the EEG Ideggyogy Sz 2009;62(1 2):36 40. 37

YH EH YEH ALZ MTD STR Figure 1. Absolute power spectra of the rest (pre-task) EEG. Average maps of the five groups for four frequency bands (delta 1.17 3.91 Hz, theta 4.30 7.81 Hz, alpha 8.20 12.89 Hz, beta 13.28 30.86 Hz) and for the total power Abbreviations: YH = younger healthy, EH = elder healthy, YEH = unified healthy*, STR = post stroke state without mental deficit, ALZ = probable Alzheimer disease, MTD = mixed type of dementia. The EEG sections used for the spectral analysis were taken from the pre-task rest EEG record. *For achieving an age matched control group to the investigation of the STR and MTD groups, we constructed a unified healthy group (YEH), too. Results Figure 1. shows the traditional absolute power spectra of the control (upper line) and the patient (lower line) groups. For achieving an age matched control group to the investigation of the STR and MTD EH YH MTD YEH ALZ EH STR YEH Figure 2. Absolute power spectra of the rest (pre-task) EEG. Difference maps of paired groups. For details and abbreviations: see Figure 1. groups, we constructed a unified healthy group (YH+EH), too. The EEG sections used for the spectral analysis were taken from the pre-task rest EEG record. Figure 2. demonstrates the results of the paired-group-analyses of the rest EEG spectra. We obtained the difference maps by comparison of the two healthy groups of different ages (EH-YH), the ALZ to the EH (ALZ-EH), the STR to the unified healthy group (STR-YEH) and the MTD to the unified healthy (MTD- YEH). We did not see meaningful alteration between the spectra of the healthy elder (EH) and younger (YH) groups. Only mild changes can be seen also on the maps of the healthy and patient groups: 1. decrease of occipital alpha rhythm and increase of total power in the central regions in ALZ group, and 2. increase of total power in the frontal region (based mainly on the increase of delta power) and a very mild increase of alpha power on the left temporal region. On the Figure 3. the group averages of the pre- and post-task difference spectra are drawn by the 38 Rajna: Mental fatigability by EEG spectra

YH EH YEH ALZ MTD STR Figure 3. Post-minus pre-task EEG: difference maps for the groups. For details and abbreviations: see Figure 1. groups. The group averages are more pronounced here. In the healthy groups the powers increased frontally, remained unchanged centrally and decreased parieto-occipitally (mainly in the alpha spectrum). The Figure 4. shows the results of the paired group analysis of the above group-spectra similarly to the order seen on the Figure 2. EDSC demonstrate craggy distinctions even between the healthy groups: an increase of frontal delta power and a decrease of alpha performance left parietally. An opposite change appeared in the ALZ group. Changes are exhibited by the other groups (STR and MTD) are between these edges. main limitations were discussed above. So the collection of clinical experiences are still in a very initial phase. There are only few observations in only some patient populations. Simple and complicated acoustic stimuli in schizophrenic patients resulted inconsistent alterations in the EEG difference spectra but the changes of the spectra were consistent if compared to the normal group 7. Compared to healthy controls depressive patients did not react to spatial tasks with general activation in the right posterior regions but they had nor- Discussion We did not evaluate the rest power spectra of our groups separately. By the consonant experiences of the literature and supported also by our earlier data 6 in case of cognitive impairment the decrease of the alpha per slow wave ratio might be expected as main alteration. EH YH ALZ EH RESULTS OF CLINICAL STUDIES After the first very promising basic applications of TRETs performed in the eighties and nineties the efforts for its introduction in the clinical routine did not lead to success. The MTD YEH STR YEH Figure 4. Post-minus pretask EEG: difference maps of paired groups. For details and abbreviations: see Figure 1. Ideggyogy Sz 2009;62(1 2):36 40. 39

mal EEG answers (left-sided activation) during the solution of verbal tasks 8. We have practically did not find any systematic TRET study of larger patient population suffering from any kind of dementia. It is not at all surprising because the cooperation of the demented patient in such kind of studies are very limited. The first application of our paradigm tested in Alzheimer group resulted a higher theta power both in the pre- and posttask records while the decrease of the relative alpha power after the mental activity was smaller in comparison to that of the age-matched normal controls. Fast alpha activity of the patient group showed the main difference also in coherence analysis 5. We reproduced the increase of the relative power of the fast alpha band to our calculation task also in pts suffering from early onset Alzheimer disease 6. EDSC MIGHT CONTAIN INFORMATION ON THE MENTAL FATIGABILITY We did not evaluate the real cognitive EEG (meaning the EEG during the reverse counting which has also been recorded) but we rather dealt with the exhaustive effect of the task. Therefore the differences of the pre- and post task EEG rather represent the fatigability of the cortical areas involved in the task performance. The fixed duration of the task independent from the level of difficulty made the exhaustive effect of the task comparable also interindividually. While as we have expected the EEG spectra are not widely different between the healthy groups, the EDSC show remarkable alpha deficit on the left parietal region. The result seems to be promising concerning the investigation of the mental fatigability by age. The mental fatigability in dementia of Alzheimer type comparing to that of the mixed dementia group also seems to be different (Figure 3.). It is more pronounced than the differences seen on the traditional rest spectra (Figure 1.) The EDSC of the five small groups showed a lot of post-task EEG changes. Similarly to the findings of other authors (detailed above), in addition to the differences of the alpha band seen on the temporo-parieto-occipital regions, the frontal localization and the beta band also seem to be prominent. We found changes in the total power, too. Another promising experience proving the value of this method in the psychopathological disorders is that demented patients had stronger EEG reactions than post-stroke patients without mental deficits and healthy elder persons had more extensive changes than the younger ones. FUTURE PLANS Interestingly, the EDSC show differences towards the opposite direction when the healthy and patient groups were compared. These results allow the application of this method to follow the pathological or even the therapeutic processes (based on the key-lock principle 9 ). A very exciting further application is the investigation of the mental aging process by EDSC. Actually the small number of persons in the groups do not allow statistical analysis, therefore we do not consider this demonstrative mini-group-analysis as a completed study on the EEG changes resulted by our task paradigm. We only want to demonstrate the probable usefulness of our method in the investigation of the mental fatigability among different (healthy and pathological) conditions even in severe psychiatric patient groups. Further studies are necessary to clear the clinical message of these group differences and their importance for evaluation of the pathological process and the applied treatment. REFERENCES 1. Cohen HL, Porjesz B, Begleiter H. The effect of ethanol on EEG activity in males at risk for alcoholism. Electroencephalogr Clin Neurophysiol 1993;86:368-76. 2. Makeig S, Jung TP. Changes in alertness are a principal component of variance in the Eeg spectrum. NeuroReport 1995;7:213-6. 3. Fernández T, Harmony T, Rodriguez M, Bernal J, Silva J, Reyes A, et al. EEG activation patterns during the performance of tasks involving different componentes of mental calculation. Electroenceph Clin Neurophysiol 1995;94:175-82. 4. Rajna P, Csibri É, Pal I, Szelenberger W. Task related difference EEG spectrum a new diagnostic method for neuropsychiatric disorders. Medical Hypotheses 2003;61:390-97. 5. Hidasi Z, Czigler B, Salacz P, Csibri E, Molnár M. Changes of EEG spectra and coherence following performance in a cognitive task in Alzheimer s disease. Int J Psychophysiol 2007;65:252-60. 6. Czigler B, Csikós D, Hidasi Z, Gaál A, Csibri Z, Kiss E, et al. Quantitative EEG in early Alzheimer s disease patients power spectrum and complexity features. Int J Psychophysiol 2008;68:75-80. 7. Stevens A, Mattes R, Günther W, Müller R, Trapp, W. First episode schizophrenics show normal duration and topography of quasistationary EEG segments as compared to controls, during rest as well as during active tasks. Psychiatry Res Neuroimaging 1999;91:111-20. 8. Henriques JB, Davidson RJ. Brain electrical asymmetries during cognitive task performance in depressed and nondepressed subjects. Biol Psychiatry 1997;42:1039-50. 9. Saletu B, Anderer P, Pascual-Marqui RD. EEG mapping and tomography providing evidence for a key-lock principle in diagnosis and treatment of mental disorders. Progr Nat Sci, International Congress Series 2002;1232:627-34. 40 Rajna: Mental fatigability by EEG spectra