Unstable EEG-vigilance in patients with cancer-related fatigue (CRF ...

The World Journal of Biological Psychiatry, 2011; Early Online, 1–7

BRIEF REPORT

World J Biol Psychiatry Downloaded from informahealthcare.com by Zentralbibliothek Medizin on 01/19/11 For personal use only.

Unstable EEG-vigilance in patients with cancer-related fatigue (CRF) in comparison to healthy controls

SEBASTIAN OLBRICH1, CHRISTIAN SANDER1, INA JAHN1, FRANZISKA EPLINIUS1, SYLKE CLAUS2, ROLAND MERGL1, PETER SCHÖNKNECHT1 & ULRICH HEGERL1 1Department

of Psychiatry and Psychotherapy of the University Hospital Leipzig, Germany and 2Department of Medical Psychology and Medical Sociology, University of Leipzig, Leipzig, Germany

Abstract Objectives. Cancer-related fatigue (CRF) is associated with tiredness and sleepiness. It remains unclear, whether such complaints are associated with neurophysiological signs of sleep proneness or a state of neurophysiological hyperarousal in which the patient finds it difficult to relax and to initiate sleep. Therefore the goal of this study is to compare the electroencephalographic (EEG)-vigilance regulation of patients with CRF and healthy controls. Methods. A 15-min resting EEG with eyes closed was recorded in 22 patients with CRF and 22 matched healthy controls. Consecutive 1-s segments were classified into seven different vigilance stages ranging from high alertness to relaxed wakefulness (stage 0, A1, A2, A3) and further on to drowsiness (B1, B2/3) and sleep onset (stage C). Results. Results showed that patients with CRF revealed a higher number of vigilance stages A3 (mean 15.26 vs. 6.67%, P 0.004) dropped significantly earlier to vigilance levels A3 (drop after 130.8 vs. 533.3 s, P 0.000) and B2/3&C (407.8 vs. 604.1 s, P0.035) and showed significantly more transitions between vigilance stages (46.0 vs. 31.1%, P 0.003) in comparison to healthy controls. Conclusions. These findings suggest an unstable vigilance regulation in patients with CRF and provide a neurophysiological framework for the reported efficacy of psychostimulants in CRF. Key words: Neuropsychiatry, biological psychiatry, cancer-related fatigue, EEG-vigilance, depression

Introduction Cancer-related Fatigue (CRF) is characterized by an increased need to rest, sleep cycle alterations, complaints of generalized weakness, decreased motivation in usual activities and attention and memory deficits (Cella et al. 1998). As possible mechanisms involved in the development of the symptomatology several factors are discussed such as anemia which is often found in patients with CRF (Ludwig et al. 2004), disturbances of the hypothalamic-pituitary-adrenal axis (Bruera et al. 1985), increased levels of interleukin-6 and tumor necrosis factor (Patarca et al. 1994) and alterations of monoamine transmitter function (Wang 2008). Also sleep disturbances such as waking at night, early awakening in the morning and altered rest/activity cycles have been discussed to explain CRF (Stasi et al. 2003; Roscoe et al. 2007). Still it remains unclear if CRF is associated with an increased propensity to develop drowsiness and to

fall asleep or whether it corresponds more to the situation of depressed patients with a state of chronic hyperarousal (Ulrich and Fürstenberg 1999) that is associated with the feeling of exhaustion and difficulties to fall asleep as shown by prolonged sleep onset latencies in the Multiple Sleep Latency Test (Argyropoulos and Wilson 2005). This differentiation is important because opposite neurophysiological alterations of sleep-wake regulation would fortify different treatment options: 1.

The first constellation with an unstable vigilance regulation would support the use of psychostimulants in the treatment of CRF. It has been shown in randomized clinical trials that dexmethylphenidate significantly reduces fatigue symptoms in comparison to placebo after 8 weeks of treatment (Lower et al. 2009) although short-term medication

Correspondence: Sebastian Olbrich, Department of Psychiatry, University of Leipzig, Semmelweisstr. 10, 04103 Leipzig, Germany. Tel: 49 341 97 25041. Fax: 49 341 97 24539. E-mail: [email protected] (Received 25 August 2010 ; accepted 2 December 2010 ) ISSN 1562-2975 print/ISSN 1814-1412 online © 2011 Informa Healthcare DOI: 10.3109/15622975.2010.545434

2

S. Olbrich et al.


2.

for a period of 8 days did not improve the fatigue symptoms in comparison to placebo (Auret et al. 2009). Also modafinil, another psychostimulant was found to improve severe fatigue symptoms in a phase III clinical trial (Jean-Pierre et al. 2010). In the latter constellation with a hyperstable vigilance regulation, resembling more the situation in depression, antidepressants might be a better treatment option to cope with the fatigue symptomatology. Because of the mentioned underlying serotonin dysregulation in CRF (Ryan et al. 2007; Levy 2008; Wang 2008) the use of antidepressants, especially of serotonin reuptake inhibitors, has been proposed as a possible treatment option in the management of CRF (Barrière et al. 2008), although evidence for efficacy is lacking.

The aim of this study was to assess the vigilance regulation in patients with CRF using an EEG-based vigilance classification algorithm. The hypothesis is that compared to healthy controls, patients with CRF show a less stable vigilance regulation with more declines to lower vigilance stages when recorded under resting conditions with eyes closed. Parameters indicating an unstable vigilance regulation were 1. 2.

3.

increased amounts of 1-s EEG-segments classified to low vigilance stages; lower EEG-vigilance levels, defined by occurrence of over 10% of a certain vigilance stage within 1-minute blocks, occurring earlier during the recording; more transitions between different EEGvigilance stages.

Sleep quality and sleepiness were additionally assessed using questionnaires.

Methods Subjects Patients. Twenty-two patients (17 females, age 48.8 5.5 years) with a malign tumour (breast cancer n 12, ovarian cancer n 3, renal cancer n 2, testicular cancer n 2, colon cancer n 1, cervix cancer n 1, Hodgkin’s disease n 1) in remission and CRF according to the criteria proposed by the ICD-10 (Cella et al. 2001) were recruited at the outpatient ward of the Department for Occupational Medicine of the University of Leipzig. CRF was assessed by clinical evaluation of a senior physician, and by the Multiple Fatigue Inventory (MFI40 points, Lin et al. 2009). Patients were free from

psychopharmacological medication. Two patients received cancer related treatment with tamoxifen, one with anastrozole. Patients with a history of additional neuropsychiatric disorders, especially with a score 8 in the depressive subscale of the Hospital Anxiety and Depression Scale (HADS), were not included in the sample. Healthy controls. Twenty-two age- and gendermatched healthy controls (age 47.4 6.8 years; 16 females) without psychopharmacological medication were chosen from the database of the Department of Psychiatry of the University Leipzig. Controls who met the criteria of DSM-IV, axis I disorders in the Structured Clinical Interview for DSM-IV (SCID-I, Wittchen 1994) or had a history or actual symptoms of neurological or other medical disorders were not included. The study was approved by the local ethics committee. Written informed consent was obtained prior to investigation according to the declaration of Helsinki. Questionnaires Sleep quality of the preceding night was evaluated using a German sleep questionnaire (Schlaffragebogen A SF-A, Görtelmayer 1981). Sleepiness at the beginning of the EEG-recording was assessed using the Stanford Sleepiness Scale (SSS, Hoddes et al. 1973).

EEG acquisition and vigilance assessments In CRF patients and healthy controls 15 min of resting EEG were recorded between 09:00 and 13:00 h in half-reclined position with constant light (approximately 40 lux) and temperature (20–23°C) conditions, attenuated sound and eyes closed. The EEG was recorded with a 40-channel QuickAmp amplifier (Brain Products GmbH, Gilching, Germany) from 31 electrode sites (extended international 10–20 system) at a sampling rate of 1 kHz, referenced against common average. Impedances were kept below 10 kW in most cases. EEG data were processed using BrainVision Analyzer 2.0 (BrainProducts, Gilching, Germany). • EEG raw data were digitally filtered at 70 Hz (low-pass), 0.5 Hz (high-pass) and 50 Hz (notch-filter) and segmented into 1-s intervals. Eye artefacts and continuous muscle artefacts were removed using an independent component analysis (ICA)-based approach (Olbrich et al.


EEG-vigilance in CRF 2010), segments with remaining artefacts were marked but not excluded. • Because of the restricting inverse relation between segment length and frequency bins in fastFourier transformed data, complex demodulation of the EEG frequency bands δ (2–4 Hz), θ (4–8 Hz), α (8–12 Hz) and β (12–25 Hz) was computed for all EEG channels to obtain a frequency band envelope magnitude in μV² to approximate the power of the underlying signal (Schroeder and Barr 2000). Since topographical power mapping, but not EEG source solutions, depends upon the reference electrode (Lehmann 1987), average EEG source densities were estimated using low resolution brain electromagnetic tomography (LORETA) in four different regions of interest (ROIs) including the (1) occipital cortices, (2) parietal cortices, (3) inferior temporal cortices and (4) mediofrontal cortices. • Segments were classified into seven different EEG-vigilance stages (see Figure 1) using the Vigilance Algorithm Leipzig (VIGALL). To avoid misclassification of low amplitude EEG segments that reflect desynchronisation during states of high alertness, the preceding version of the algorithm (Olbrich et al. 2009) has been extended with an additional high vigilance stage 0 that is characterized by desynchronized EEG

3

without slow eye movements (SEMs). Stage A1 is marked by dominant α activity in the occipital ROI, stage A2 by increased α activity in temporal and parietal ROIs and stage A3 by increased α activity in the frontal ROI. Stage B1 is marked by low amplitude non-α EEG with SEMs and stage B2/3 by increasing θ and δ power. Sleep onset corresponds to stage C, characterized by sleep spindles and/or K-complexes. This vigilance classification is based on the work of Loomis et al. (1937) that was refined by Bente (1964) and Roth (1961) and has been confirmed by other research groups (Santamaria and Chiappa 1987; Tanaka et al. 1996, 1997; Benca et al. 1999; Corsi-Cabrera et al. 2000; De Gennaro et al. 2001a,b, 2004, 2005; Tsuno et al. 2002; De Gennaro and Ferrara 2003). • Percentages of vigilance stages were computed by dividing the amount of segments of a certain stage by the amount of non-artefact segments. A vigilance level was defined as the occurrence of a certain vigilance stage in over 10% of segments within blocks of 60 segments, i.e. 60 s, in sliding steps of one segment. Transition rates were calculated by dividing the amount of changes between different vigilance stages with the total amount of artefact-free segments. Statistics Amount of vigilance stages and transition rates were compared using two-tailed Mann–Whitney U-test (P0.05, Bonferroni corrected for multiple comparison of six stages). In order to analyze vigilance decline in CRF patients in comparison to healthy controls, Kaplan– Meier cumulative functions were computed for occurrence of all vigilance levels. SF-A item scores for patients with CRF and healthy controls were compared using two-tailed t-test (P0.05; Bonferroni corrected for multiple comparison of five main items). SSS scores for patients with CRF and healthy controls were compared using Mann–Whitney U-test (P 0.05).

Results Figure 1. Criteria of the Vigilance Algorithm Leipzig (VIGALL) for the classification of EEG-vigilance stages from high alertness to sleep onset: Stage 0 with low amplitude EEG without slow eye movements (SEMs), stage A1 with occipital dominant EEG-α power, stages A2/A3 with pronounced α activity at temporoparietal and frontal cortices, low amplitude EEG with SEMs during stage B1, increasing δ and θ power during stage B2/3 and the occurrence of K-complexes or sleep spindles at stage C. Examples of 3-s EEG/EOG recordings are given at the right panel.

Vigilance classification (a) Prevalence of different vigilance stages. For every subject, a classification into the different vigilance stages was performed for 822 consecutive one-second segments. This was the length of recording available for every subject. Stage C, indicating sleep, was reached in only two subjects in both groups. Therefore, in the

4

S. Olbrich et al.

analysis stage C was combined with stage B2/3. Analysis of the EEG-vigilance stages 0, A1, A2, A3, B1 and B2/3&C revealed significant differences only for A3. The percentage of EEG segments with vigilance stage A3 were higher in patients with CRF compared to healthy controls (see Table I).

CRF patients perceived themselves as significantly sleepier than controls in the SSS (CRF-patients: median score 2, range 1–6; controls: median score 1, range 1–3; Z –3.284; P 0.001).


Discussion (b) Stability of vigilance regulation. Kaplan–Meyer cumulative functions revealed significant differences between patients and controls concerning the time until decline to lower vigilance levels. Patients with CRF reached vigilance levels of stages A3 and B2/3 faster (estimated mean time until drop to A3: 130.8 s (CRF) vs. 533.3 s (controls); Log Rank (Mantel– Cox) P 0.000 and until drop to B2/3&C: 407.8 s (CRF) vs. 604.1 s (controls); P 0.035; see Figure 2) and more often (A3: 20 (CRF) vs. 11 (controls); B2/3&C: 13 (CRF) vs. 6 (controls)) than controls. (c) Transition rates. Comparison of transition rates between stages 0, A1, A2, A3, B1, B2/3&C revealed significantly more transitions for patients with CRF than controls (patients: mean 46.0%, SD 16.4; controls: mean 31.1; SD 15.2; Z –2.934, P 0.003). Subjective sleep & sleepiness scores The SF-A revealed significantly increased scores in patients with CRF (see Table II) for mental tiredness in the evening before EEG recording and for psychosomatic symptoms during the preceding night's sleep, comprising headache, heart complaints and sweating. No significant differences were found concerning SF-A items of sleep quality, time of falling asleep, time of waking up, total sleep duration and time spent awake before the EEG recording.

Patients suffering from CRF but free of psychopharmacological medication showed significantly more often lower EEG-vigilance stage A3 and dropped earlier to this vigilance level than healthy controls. Also declines to vigilance level B2/3&C occurred earlier and more often. In addition more changes between vigilance stages were observed. This was paralleled by increased SF-A scores of mental tiredness in the evening and psychosomatic symptoms during the preceding night as well as increased SSSscores of sleepiness before EEG-recording in comparison to healthy controls. To date, no studies have been conducted which analyze the EEG-vigilance regulation in CRF patients. However, one EEG study that focused on chronic fatigue syndrome (CFS) without an association of a malign tumour also found increased EEG measures of sleepiness and low vigilance (i.e. δ power) during rest in patients in comparison to their non-affected siblings (Sherlin et al. 2007). Other works on CFS report increased EEG δ power during different sleep stages (Guilleminault et al. 2006). The lowered vigilance revealed in this study is also in line with reported dysregulation of sleep–wake cycles in patients with CRF. While some studies found CRF to be associated with difficulties to fall asleep, others reported increased daytime sleepiness and daytime sleep to be correlated with CRF (review: Roscoe et al. 2007). The increased subjective feeling of sleepiness in patients with CRF in this study was paralleled by faster declines to low vigilance levels A3 and B2/3&C and increased transition rates

Table I. Differences in the amount of time spent in different EEG-vigilance stages ranging from full alertness after closing the eyes (stage 0) to sleep onset (stage C) between 22 patients with cancer-related fatigue (CRF) and 22 healthy controls. Shown are means and standard deviation (SD) of the percentage amount of a vigilance stage within 822 consecutive 1-s segments. After Bonferroni correction for multiple testing, CRF patients revealed significantly more vigilance stages A3 than healthy controls. Analysis of α peak frequencies showed no differences between the groups (CRF: 10.15 0.88 Hz; controls: 9.91 0.97; two-tailed t-test, P 0.41; see online supplement). Vigilance stages Stage A0 Stage A1 Stage A2 Stage A3 Stage B1 Stages B2/3 & C

CRF patients [Mean in % SD]

Controls [Mean in % SD]

Z

P

21.58 31.19 45.66 35.69 10.41 12.01 6.67 8.86 8.60 13.63 7.08 17.43

–0.658 –0.704 –1.984 –2.887 –0.426 –1.410

0.510 0.481 0.047 0.004 0.670 0.159

12.93 35.78 17.71 15.26 8.31 10.00

23.50 24.82 15.28 12.31 19.44 12.21

Group differences were tested for significance using the Mann–Whitney U-test since the data was not distributed normally. Bonferroni correction for 6 tests (0, A1, A2, A3, B1, B23&C): P (0.05) 0.008.


EEG-vigilance in CRF

5

Figure 2. Kaplan–Meier cumulative curves with significant differences in the amount of time it took patients with cancer-related fatigue syndrome (CRF) compared to healthy controls to reach low vigilance stages A3 and B2/3&C. Patients with CRF reached those stages significantly earlier and more often than controls.

between vigilance stages in comparison to healthy controls. These results suggest instability of vigilance regulation and increased propensity for development of drowsiness during rest in patients with CRF. Also the increased amount of low vigilance stage A3 in comparison to healthy controls found within this study provides further evidence for an unstable vigilance regulation in CRF. These findings indicate that the feeling of sleepiness reported by patients with CRF is different to the situation in patients with major depression, who often also report about tiredness but show neurophysiological signs of tonic hyperarousal. The latter is indicated by prolonged sleep onset latencies (Hubain et al. 2006), activation of the stress-related hypothalamus-pituitary-adrenal (HPA) axis (Pariante and Lightman 2008) and hyperstable vigilance regulation (Ulrich and Fürstenberg 1999) with reduced declines to lower vigilance stages. In the light of suggested unstable vigilance regulation in patients with CRF it appears to be plausible that drugs with vigilance stabilizing properties such

as psychostimulants have positive clinical effects: A Cochrane review reports the efficacy of dexmethylphenidate in the treatment of CRF in comparison to placebo in a meta-analysis of four randomized clinical trials (Minton et al. 2008, 2010). On the other hand, treatment with the antidepressant paroxetine, a selective serotonin reuptake inhibitor, only improved depressive symptoms in patients suffering from CRF in a randomized clinical trial while fatigue symptomatology remained unaffected (Morrow et al. 2003). Only medication with the antidepressant bupropion that has a stimulating effect via the norepinephrine/dopamine transmitter system showed improvements of fatigue symptoms in an open label study (Moss et al. 2006). A possible explanation for these findings is that CRF is associated with an altered vigilance regulation with an increased propensity to show declines to lower vigilance stages. Although interpretation is limited by the small sample size with mixed tumour types and cancer-related medication in three patients, the results of this study provide evidence for a lowered vigilance in CRF at a neurophysiological level.

Table II. Differences in subjective evaluation of sleep in the night before EEG assessment between 22a patients with cancer-related fatigue (CRF) and 22a healthy controls. Shown are the results of the five subscales of the German sleep questionnaire A (SF-A, Görtelmeyer 1981). Information about correlations between EEG-vigilance stages and SF-A/Stanford Sleepiness Scale can be found in the online supplement.

SF-A scales Sleep quality (SQ) Feeling rested after sleep (GES) Mental balance in the evening (PSYA) Mental tiredness in the evening (PSYE) Psychosomatic symptoms during the sleep period (PSS)

CRF patients Mean/ Standard deviation

Controls Mean/ Standard deviation

T

Pb

3.54/0.70 3.13/0.63 3.09/0.85 3.60/0.78 1.47/0.44

3.95/0.84 3.72/0.79 3.74/0.80 2.97/0.56 1.14/0.28

–1.657 –2.552 –2.433 2.871 2.965

0.106 0.015 0.020 0.007 0.005

aDue to missing values not all scales could be calculated in all subjects. Exact N (CFS/Controls) for the different scales was: SQ: 18/21; GES, PSYA, PSYE: 18/20 and PSS: 18/22. bDue to multiple testing a Bonferroni correction was performed: 5% significance: P 0.010; 1% significance: P 0.002.


6

S. Olbrich et al.

Still, some of the results, especially the increased frontal α activity in CRF as reflected by increased amounts of stage A3 might be related to pathomechanisms other than vigilance dysregulation. Alpha activity sources within the anterior cingulate cortex have not been found to be altered by the sedating, i.e. vigilance reducing, agent lorazepam (Connemann et al. 2005). Alpha anteriorisation has further been correlated with structural alterations in hippocampal regions (Begré et al. 2003) in schizophrenic patients. Therefore a better understanding of α anteriorisation in CRF is needed since this might yield important information about the underlying causes of CRF.

Acknowledgements We would like to thank Prof. Dr. Schwarz, who was engaged in the enrollment of the study. Prof. Schwarz died in 2009.

Statement of Interest The research group received fees according to the German medical fee schedule for recording of clinical EEG for the CRF study population from Medice. Ulrich Hegerl has been participating in advisory boards of Lilly, Wyeth, Lundbeck and SanofiAventis and received honoraria from Cephalon.

References Argyropoulos SV, Wilson SJ. 2005. Sleep disturbances in depression and the effects of antidepressants. Int Rev Psychiatr 17:237–245. Auret KA, Schug SA, Bremner AP, Bulsara M. 2009. A randomized, double-blind, placebo-controlled trial assessing the impact of dexamphetamine on fatigue in patients with advanced cancer. J Pain Symptom Manag 37:613–621. Barrière J, Cherikh F, Pringuey D, Milano G, Ferrero JM. 2008. Antidepressants in oncology: issues and clinical perspectives. B Bull Cancer: 95:1103–1111. Begré S, Federspiel A, Kiefer C, Schroth G, Dierks T, Strik WK. 2003. Reduced hippocampal anisotropy related to anteriorization of alpha EEG in schizophrenia. Neuroreport 14:739–742. Benca RM, Obermeyer WH, Larson CL, Yun B, Dolski I, Kleist KD, et al. 1999. EEG alpha power and alpha power asymmetry in sleep and wakefulness. Psychophysiology 36:430–436. Bente D. 1964. Vigilanz, dissoziative Vigilanzverschiebung und Insuffizienz des Vigilitätstonus. In: Kranz H and Heinrich K (Editors) Begleitwirkungen und Mißerfolge der psychiatrischen Pharmakotherapie. Thieme, Stuttgart. p. 13–28. Bruera E, Roca E, Cedaro L, Carraro S, Chacon R. 1985. Action of oral methylprednisolone in terminal cancer patients: a prospective randomized double-blind study. Cancer Treat Rep 69:751–754. Cella D, Peterman A, Passik S, Jacobsen P, Breitbart W. 1998. Progress toward guidelines for the management of fatigue. Oncology 12:369–377.

Cella D, Davis K, Breitbart W, Curt G. 2001. Cancer-related fatigue: Prevalence of proposed diagnostic criteria in a united states sample of cancer survivors. J Clin Oncol 19:3385–3391. Connemann BJ, Mann K, Lange-Asschenfeldt C, Ruchsow M, Schreckenberger M, Bartenstein P, et al. 2005. Anterior limbic alpha-like activity: a low resolution electromagnetic tomography study with lorazepam challenge. Clin Neurophysiol 116:886–894. Corsi-Cabrera M, Guevara MA, Del Río-Portilla Y, Arce C, Villanueva-Hernández Y. 2000. EEG bands during wakefulness, slow-wave and paradoxical sleep as a result of principal component analysis in man. Sleep 23:738–744. De Gennaro L, Ferrara M. 2003. Sleep spindles: an overview. Sleep Med Rev 7:423–440. De Gennaro L, Ferrara M, Bertini M. 2001a. The boundary between wakefulness and sleep: quantitative electroencephalographic changes during the sleep onset period. Neuroscience 107:1–11. De Gennaro L, Ferrara M, Curcio G, Cristiani R. 2001b. Anteroposterior EEG changes during the wakefulness-sleep transition. Clin Neurophysiol 112:1901–1911. De Gennaro L, Vecchio F, Ferrara M, Curcio G, Rossini PM, Babiloni C. 2004. Changes in fronto-posterior functional coupling at sleep onset in humans. J Sleep Res 13:209–217. De Gennaro L, Vecchio F, Ferrara M, Curcio G, Rossini PM, Babilon CI. 2005. Antero-posterior functional coupling at sleep onset: changes as a function of increased sleep pressure. Brain Res Bull 65:133–140. Görtelmeyer R. SF-A und SF-B. Schlaffragebogen A und B. In CIPS (Eds.), Internationale Skalen für Psychiatrie (2. Edition). 1981; Weinheim: Beltz Test. Guilleminault C, Poyares D, Da Rosa A, Kirisoglu C, Almeida T, Lopes MC. 2006. Chronic fatigue, unrefreshing sleep and nocturnal polysomnography. Sleep Med 7:513–520. Hoddes E, Zarcone V, Smythe H, Phillips R, Dement WC. 1973. Quantification of sleepiness: a new approach. Psychophysiology 10:431–436. Hubain P, Le Bon O, Vandenhende F, Van Wijnendaele R, Linkowski P. 2006. Major depression in males: effects of age, severity and adaptation on sleep variables. Psychiatr Res 145:169–177. Jean-Pierre P, Morrow GR, Roscoe JA, Heckler C, Mohile S, Janelsins M, et al. 2010. A phase 3 randomized, placebocontrolled, double-blind, clinical trial of the effect of modafinil on cancer-related fatigue among 631 patients receiving chemotherapy. Cancer 116:3513–3520. Lehmann D. 1987. Principles of spatial analysis. In: Gevins AS, Rémond A, editors. Methods of analysis of brain electrical and magnetic signals. Amsterdam: Elsevier. p 309–354. Levy MR. 2008. Cancer fatigue: a neurobiological review for psychiatrists. Psychosomatics 49:283–291. Lin JMS, Brimmer DJ, Maloney EM, Nyarko E, BeLue R, Reeves WC. 2009. Further validation of the Multidimensional Fatigue Inventory in a US adult population sample. Popul Health Metr 7:18. Loomis A, Harvey E, Hobart G. 1937. Cerebral states during sleep, as studied by human brain potentials. J Exp Psychol 21:127–144. Lower EE, Fleishman S, Cooper A, Zeldis J, Faleck H, Yu Z, Manning D. 2009. Efficacy of dexmethylphenidate for the treatment of fatigue after cancer chemotherapy: a randomized clinical trial. J Pain Symptom Manag 38:650–662. Ludwig H, Van Belle S, Barrett-Lee P, Birgegård G, Bokemeyer C, Gascón P, et al. 2004. The European Cancer Anaemia Survey (ECAS): A large, multinational, prospective survey defining the prevalence, incidence, and treatment of anaemia in cancer patients. European Journal of Cancer, 40:2293–2306.


EEG-vigilance in CRF Minton O, Richardson A, Sharpe M, Hotopf M, Stone P. 2008. A systematic review and meta–analysis of the pharmacological treatment of cancer-related fatigue. J Natl Cancer I 100: 1155–1166. Minton O, Richardson A, Sharpe M, Hotopf M, Stone P. 2010. Drug therapy for the management of cancer-related fatigue. Cochrane Database of Systematic Reviews (Online) 7. Morrow GR, Hickok JT, Roscoe JA, Raubertas RF, Andrews PLR, Flynn PJ, et al. 2003. Differential effects of paroxetine on fatigue and depression: a randomized, double-blind trial from the University of Rochester Cancer Center Community Clinical Oncology Program. J Clin Oncol 21:4635–4641. Moss EL, Simpson JSA, Pelletier G, Forsyth P. 2006. An openlabel study of the effects of bupropion SR on fatigue, depression and quality of life of mixed-site cancer patients and their partners. Psycho-Oncology 15:259–267. Olbrich S, Mulert C, Karch S, Trenner M, Leicht G, Pogarell O, et al. 2009. EEG-vigilance and BOLD effect during simultaneous EEG/fMRI measurement. Neuroimage 45:319–332. Olbrich S, Jödicke J, Sander C, Himmerich H, Hegerl U. 2011. ICA-based muscle artefact correction of EEG data: What is muscle and what is brain? Comment on McMenamin et al. Neuroimage 54:1–3. Pariante CM, Lightman SL. 2008. The HPA axis in major depression: classical theories and new developments. Trends Neurosci 31:464–468. Patarca R, Klimas NG, Lugtendorf S, Antoni M, Fletcher MA. 1994. Dysregulated expression of tumor necrosis factor in chronic fatigue syndrome: interrelations with cellular sources and patterns of soluble immune mediator expression. Clin Infect Dis 18:147–153. Roscoe JA, Kaufman ME, Matteson-Rusby SE, Palesh OG, Ryan JL, Kohli S, et al. 2007. Cancer-related fatigue and sleep disorders. Oncologist 12:35–42.

Supplementary material available online Supplement Table 1. Supplement Table 2.

7

Roth B. 1961. The clinical and theoretical importance of EEG rhythms corresponding to states of lowered vigilance. Electroencephalogr Clin Neurol 13:395–399. Ryan JL, Carroll JK, Ryan EP, Mustian KM, Fiscella K, Morrow GR. 2007. Mechanisms of cancer-related fatigue. Oncologist 12:22–34. Santamaria J, Chiappa KH. 1987. The EEG of drowsiness in normal adults. J Clin Neurophysiol 4:327–382. Schroeder MJ, Barr RE. 2000. An alpha modulation index for electroencephalographic studies using complex demodulation. Med Biol Eng Comput 38:306–310. Sherlin L, Budzynski T, Budzynski HK, Congedo M, Fischer ME, Buchwald D. 2007. Low-resolution electromagnetic brain tomography (LORETA) of monozygotic twins discordant for chronic fatigue syndrome. Neuroimage 34:1438–1442. Stasi R, Abriani L, Beccaglia P, Terzoli E, Amadori S. 2003. Cancerrelated fatigue. Cancer 98:1786–1801. Tanaka H, Hayashi M, Hori T. 1996. Statistical features of hypnagogic EEG measured by a new scoring system. Sleep 19:731–738. Tanaka H, Hayashi M, Hori T. 1997. Topographical characteristics and principal component structure of the hypnagogic EEG. Sleep 20:523–534. Tsuno N, Shigeta M, Hyoki K, Kinoshita T, Ushijima S, Faber PL, et al. 2002. Spatial organization of EEG activity from alertness to sleep stage 2 in old and younger subjects. J Sleep Res 11:43–51. Ulrich G, Fürstenberg U. 1999. Quantitative assessment of dynamic electroencephalogram (EEG) organization as a tool for subtyping depressive syndromes. Eur Psychiatry 14:217–229. Wang XS. 2008. Pathophysiology of cancer-related fatigue. Clin J Oncol Nurs 12:11–20. Wittchen HU. 1994. Reliability and validity studies of the WHO-Composite International Diagnostic Interview (CIDI): a critical review. J Psychiatr Res 28:57–84.