Nps59(ssssc).indd

The abstracts are only available online, free of charge, at www.karger.com/doi/10.1159/000230666 Annual Conference of the Swiss Society of Sleep Research, Sleep Medicine and Chronobiology (SSSSC) and the Swiss Society of Biological Psychiatry (SSBP) Chairmen (SSSSC)Christian Cajochen, BaselJohannes Mathis, Bern Basel · Freiburg · Paris · London · New York · Bangalore ·Bangkok · Shanghai · Singapore · Tokyo · Sydney 3. Tobler I, Murison R, Ursin R, Ursin H, Borbély AA: The effects of sleep deprivation and recovery on plasma corticosterone in the rat. Neurosci Lett 1983;35:297-300.
Research supported by the University of Lausanne, the NIH (MH67752), FNS (3100A0-111974), and an NSERC fellowship. No Rescue of the Impaired Homeostasic Sleep Rebound in DBA/2J Mice with Attentuation of the Stress Response V. Mongrain1, T. Curie1, P. Gip2, H.C. Heller2, P. Franken1,2 Increase in Slow Wave Sleep During Multiple Naps in Women with Major Depression: 1CIG, University of Lausanne, Lausanne, Switzerland, 2Stanford University, Stanford, Calif., USA S. Frey, A. Birchler-Pedross, P. Brunner, T. Götz, Introduction: We have previously shown that both the change in EEG delta power and in clock-gene expression after Centre for Chronobiology, Psychiatric University Clinics, sleep deprivation (SD) in mice depends on genetic background. Among several inbred strains, DBA/2J (D2) mice showed the smallest increase in delta power and the largest changes in clock- Homeostatic sleewake regulation (Process S) has been genes expression in the forebrain [1]. SD in rodents induces a hypothesized to be impaired in depression. It is not yet clear stress response as it increases corticosterone secretion [2,3]. whether the saturation of process S during wakefulness or its This study aimed at determining the contribution of changes in decline rate during sleep or both may be altered in depression. In corticosterone to the homeostatic response of EEG delta power an ongoing study, we aim at quantifying process S in women suf- and clock-gene expression. Methods: Experiment 1: Mice fering from major depression and healthy controls during short from C57BL/6J, AKR/J, and D2 inbred strains were submitted sleewake cycles under constant routine conditions. Eight females to SD by gentle handling (ZT0-6) and killed immediately after with major depression (mean age 23y, SD +/- 4.12y) and nine for plasma corticosterone measurements. Experiment 2: Sleep healthy young females (mean age 25.3, SD +/- 4.03y) followed of adrenalectomized (ADX) and sham-lesioned (sham) D2 mice a protocol over 40-h with 10 short sleewake cycles (“nap proto- was recorded for a baseline day and during recovery from SD, col”) in a chronobiology facility under constant light and tem- before and after surgery. Experiment 3: ADX and sham D2 mice perature conditions. Polysomnographic recordings were carried were submitted to either baseline condition or SD and killed at out continuously (12 EEGs, EOG, EMG). Additionally, subjec- ZT6 when brains were collected for analysis of forebrain clock- tive sleepiness was assessed by the Karolinska Sleepiness Scale. gene expression using qPCR. Results: SD induced a higher Slow wave sleep, as a marker for the decrease of process S, and increase in corticosterone secretion in D2 mice compared to the subjective sleepiness were each subjected to a repeated twway other two strains. Nevertheless, the delta power rebound after mixed ANOVA with the factors “group” (depressed vs. healthy SD was similar in ADX and sham D2 mice. ADX attenuated the subjects) and “nap” (sleep episodes) or “time” (for sleepiness rat- SD-mediated increase in mPer1-3 expression but did not change ings) respectively. Results disclosed a main effect of the factors the SD-dependant decrease in mDbp mRNA. Discussion: While “nap” (p<0.0001, “group” (p<0.01), and an interaction effect of the larger SD-mediated increase in corticosterone observed in D2 “nap x group” (p<0.05) on slow wave sleep. Post-hoc analysis mice could partly explain the larger increase in mPer expression revealed that women with major depression had significantly in this strain, a higher stress response seems not to underlie the more slow wave sleep in naps which occurred in the evening lower EEG delta power rebound after SD in this strain. The find- during the wake maintenance zone (p<0.05). Results for sub- ings also indicate that stress does not importantly contribute to the jective sleepiness showed a main effect of the factors “session” homeostatic regulation of EEG delta power.
(p<0.001) and “group” (p<0.01) whereas no interaction effect “session x group” was observed. Post-hoc analysis on subjective sleepiness ratings yielded significantly higher scores for almost 1. Franken P, Thomason R, Heller HC, O’hara BF: A non-circadian role for all time points during the 40-h nap protocol for women with clock-genes in sleep homeostasis: a strain comparison. BMC Neurosci major depression compared to healthy women. Our data indi- cate that women with major depression sleep on a higher sleep 2. Hairston IS, Ruby NF, Brooke S, Peyron C, Denning DP, Heller HC, pressure level despite the fact that multiple naps encourage low Sapolsky RM: Sleep deprivation elevates plasma corticosterone levels in neonatal rats. Neurosci Lett 2001;315:29-32.
sleep pressure conditions. This can be explained by either a faster 2009 S. Karger AG, Basel0302–282X/09/0594–0246$26.00/0 Fax ϩ41 61 306 12 34E-Mail [email protected] buildup of process S or a weaker circadian alerting signal, partic-ularly during the wake maintenance zone in depressed women.
EEG Spectra in Healthy Older Adults Across Research supported by the Swiss National Science Foundation M. Münch1, 2, E.J. Silva1, J.M. Ronda1, C.A. Czeisler1,
J.F. Duffy1
1Division of Sleep Medicine, Brigham & Women’s Hospital, Harvard Medical School Boston, Mass., USA, 2Current Address: Swiss Federal Institute of Technology Lausanne, Neonatal Care on Sleep and Crying Pattern Laboratory for Solar Energy and Building Physics, C. Guyer1, J. Fontijn2, R. Huber1, H.U. Bucher2, O.G. Jenni1 Healthy aging is associated with changes in sleewake regu- 1Child Development Center, University Children’s Hospital lation, and those changes often lead to problems sleeping, both Zurich, 2Department Of Neonatology, University Hospital during the night and during daytime. The aim of our study was to examine the EEG sleep spectra during non-rapid eye move-ment (NREM) sleep when sleep was scheduled at all times of Aims: The influence of a 24h light-dark rhythm in neonatal day. Twenty-four healthy older subjects (64.2±2.3 yrs; 13 f, 11 care on sleep and crying patterns of preterm infants was examined, m) participated in an inpatient protocol. After 3 baseline (BL) using standard diary measures over 3 consecutive days and actig- days, they were scheduled to live on 20-h “days” consisting of raphy data at corrected ages of 5,11 and 25 weeks. Methods: 41 6.7h of bed rest and 13.3h of wakefulness for 12 consecutive days preterm infants (< 32 weeks gestational age) were either nursed in (forced desynchrony, FD). The electroencephalogram (EEG) was the standard 24h dim light condition (control group; n=21) or in a recorded during all scheduled sleep episodes, and all EEGs were cycled light condition (7am-7pm light, 7pm-7am dark; interven- visually scored and subjected to spectral analysis. Compared to tion group; n=20). Sleecrying diary at 5 and 11 weeks corrected BL nights, EEG activity in the slow-wave (0.5–5.25 Hz), theta age were analysed in four 6h intervals starting at midnight, by (6-6.25, 7 Hz), alpha (10-11.25 Hz) and high spindle range (14.5-following variables: sleep, wake and content, fussing, crying and 15.5 Hz) was significantly greater during FD, when subjects slept unsoothable crying. Actigraphy data were analysed at all three across many times of day. During FD, there was a significant time points using periodogramm analysis. Furthermore an overall interaction between homeostatic and circadian dependent factors control group of 19 term infants were included, using the same such that EEG delta activity (0.5-1.5 Hz) was higher in the bio-methodological and analytical measures. Results: Sleecrying logical morning/early afternoon than at other times of day. EEG diary: Using repeated measure ANOVA fussing and crying activity was significantly increased in almost all frequency ranges showed an age (fussing: P < 0.008, crying: P < 0.00) and interval (0.5-20.5 Hz) during the biological day, when endogenous mela- (fussing: P < 0.00, crying: P < 0.001) effect with more fussing tonin was not present, except for the lower EEG spindle range and crying at 5 weeks and in the evening hours. Furthermore cry- (12.25-14 Hz). Our results show age-related changes in the sleep ing revealed an interaction between age, interval and interven- EEG spectra in healthy older subjects and provide evidence for tion. (Cycled condition: less crying between 6:00 – 12:00 at 5 why it is more difficult for older adults to maintain consolidated weeks and between 6:00 – 18:00 at 11 weeks of age; P < 0.021). sleep, especially when sleep occurs during the daytime.
Actigraphy data: Repeated measure ANOVA showed a steadily increase in 24h day-night rhythm with age (P < 0.00). No group This study was supported by NIH grant P01 AG09975 (to C.A.C.) and effect in the preterm infants could be seen, but in comparison with was conducted in the BWH GCRC supported by M01 RR02635; M.M. was supported by fellowships from the Novartis Foundation (Switzerland), the W. term infants, preterm infants showed an earlier and stronger 24h & T. La-Roche Foundation (Switzerland), and Jazz Pharmaceuticals (Palo day-night rhtyhm. Summary and Conclusion: Nursing in a Alto, CA, USA).
cycled light condition reduced crying episodes at certain times of day at 5 and 11 weeks corrected age, detected by sleecrying diary. Entrainment of a 24h day-night rhythm could not be accelerated by cycled light. Preterm infants showed an earlier and stronger 24h day-night rhythm than term infants. Mechanisms Underlying the Effects of GABAA Sedative-Hypnotic Drugs Institute of Pharmacology and Toxicology, University of Zurich, Switzerland The predominant inhibitory transmitter, gamma-aminobutyric acid (GABA), and its GABAA receptors play an important role in the neuronal systems regulating sleep. In particular, the most effective hypnotics, benzodiazepines (BZ) and BZ-like com- experience a further destabilization of their ventilation by hypoxic pounds, target GABAA receptors. However, many hypnotics ventilatory stimulation causing hypocapnia and a reduced CO2 display side-effects. Therefore, there is a need to develop new reserve. Consistent with ventilatory control theory, central apnea compounds to treat insomnia. Interestingly, the effects of GABAA at altitude prevails mostly in NREM sleep. Hypoxia, pronounced receptor agonists on sleep are substantially different from those oscillatory breathing, disturbed sleep and enhanced cardiovascu-evoked by BZ or BZ-like compounds. One of our aims is to deci- lar stress with water retention are all potentially serious adverse pher in mice the mechanisms underlying the specific effects of effects that occur even during a short-term sojourn at moderate GABAA agonists, e.g. THIP/Gaboxadol and muscimol, on behav- altitude in untreated OSA patients. Therefore, effectiveness of ior and the electroencephalogram (EEG). Traditional hypnotics CPAP and other treatment modalities should be evaluated in this target synaptic GABAA receptors mediating “phasic” inhibition. We were able to demonstrate in vivo that THIP acts at extrasyn-aptic GABAA receptors mediating a non-desensitizing “tonic” inhibition in the brain. In addition, we demonstrated a relation- Effect of altitude exposure in untreated OSA patients ship between regional electroencephalogram synchronization and the alteration of behavior induced by muscimol. Understanding the mechanisms underlying physiological and “pharmacological” sleep could provide a better basis to develop new compounds to treat sleep disorders, as well as to characterize their effects on the Forschungskredit from the University of Zurich; EU MC MCRTN-CT-2004- Untreated Patients with Obstructive Sleep Apnea. A Randomized, Controlled Study n=34, medians (quartiles) of polysomnographicdata. Variables marked with # were measured in the following morning. AHI=apnea/hypopnea index, Y. Nussbaumer1, N. Schuepfer2, S. Ulrich1, K.E. Bloch1 SpO2=oxygen saturation, PtcCO2=transcutaneous carbon dioxide tension, NREM/REM=(non)rapid eye movement sleep, TIB=time in bed, BP=blood 1Pulmonary Division, University Hospital, Zurich, 2Institute of human movement sciences and sport, ETH Zurich, Switzerland Background: Many patients with obstructive sleep apnea syndrome (OSA) travel to the mountains for recreational and professional activities while temporarily discontinuing CPAP. Since the physiologic consequences of altitude exposure in OSA Contributions in PER2 Protein in Living Mice patients are not known, we evaluated the hypothesis that sleep T. Curie, V. Mongrain, S. Dorsaz, S. Maret, Y. Emmenegger, related breathing disturbances are aggravated and cardiovascular stress is enhanced by hypoxia in these patients. Methods: 34 OSA patients, median age 62y, residing at <600m discontinued Center for Integrative Genomics, University of Lausanne, long-term CPAP therapy for 8 nights. They spent days 1 to 4 at low altitude and days 5 to 9 at high altitude in 2 mountain resorts (Davos Schatzalp, 1’850m, and Davos Jakobshorn, 2’590m), 2 Sleep is regulated by two main processes: a homeostatic pro- days and nights each. They were randomized to undergo clini- cess that regulates sleep need and a circadian process that ensures cal evaluation and polysomnographies either during the last day that sleep occur at the appropriate time of day. Although the two in Zurich (490m) before ascent to altitude and during 4 days in processes interact to shape the distribution and quality of sleep Davos, or during 4 days in Davos and on the first day after their and waking they are thought to act independently. Recent obser- return to Zurich, respectively. Results: The table summarizes vations demonstrate, however, that these processes have more in the results. Compared to Zurich, in the 1st night at Jakobshorn common that was previously thought. Our published results show (2´590m), oxygen saturation and carbon dioxide tension were that circadian clock gene expression, that of Period 2 (Per2) in reduced while the apnea/hypopnea index was nearly doubled due particular, increases during waking and decreases during sleep. to frequent central apnea in NREM sleep. Slow wave sleep was These and other results suggest a non-circadian role for clock reduced and arousals were more prevalent. Heart rate and preva- genes in sleep homeostasis. Here we investigate the dynamics of lence of premature beats were increased, blood pressure was ele- PER2 protein levels as a function of time-of-day and as a func- vated and patients had gained weight at altitude. Conclusions: tion of time-spent-awake (i.e., sleep deprivation) using C57BL/6J These findings indicate that untreated OSA patients at altitude PER2::Luciferase knock-in mice. With the Xenogen 3-D imag- Routine Protocol. The in vivo period length from 8 blind subjects ing system, we could detect bioluminescence throughout the has been measured by weekly collection of urine and analyses of body, notably the brain (using a glass cylinder positioned onto 6-sulphatoxymelatonin content for 3 weeks. From all the subjects the skull), and the liver and kidneys as verified using 3-D recon- participating to the study 2 skin punch biopsies have been taken stitution software (Living 3-D image software). The luciferase and fibroblasts have been isolated and infected with an engi-substrate, luciferin, was infused constantly using osmotic mini- neered lentiviral circadian reporter (mBmal-1::luc) to character- pumps either subcutaneously (for peripheral tissues) or ICV (for ize the circadian rhythms in vitro by bioluminescence imaging. the brain). We observed diurnal changes in PER2 protein levels Measurements were conducted over a time period of 5 days under in the brain and the periphery. After 6h sleep deprivation, we also standard experimental conditions using Dulbecco’s Modified observed an increase in bioluminescence from the brain and from Eagle’s Medium containing 10% foetal bovine serum. Results: the periphery, indicating an increase in PER2 protein consistent A statistical significant correlation has been observed between with the elevation of Per2 mRNA we reported previously. The fibroblast period length and in vivo period length measured in the increase in PER2 was still important 2 h after the end of the sleep same subjects, though fibroblast period values were on average deprivation, while mRNA immediately decreased with recovery slightly longer. The good correlation has been found both in the sleep. Preliminary western analyses in wild-type C57BL/6J mice sighted as well as in the blind group. There is a statistical sig- seem to confirm these results.These results show, for the first nificant difference between circadian period length of the sighted time, that PER2 protein levels can be followed around the clock group and of the blind group, probably reflecting the two different in the whole living mice and that PER2 protein is increased with protocols used for the determination of in vivo period. However, sleep loss. We are currently developing a system to monitor PER2 the analyses of the in vitro circadian rhythms revealed that the expression in specific brain regions (SCN and cerebral cortex) to period length of fibroblasts from the two groups of subjects is not more precisely dissect the circadian and homeostatic contribu- different. Conclusions: Our data indicate that human “free-run- ning” circadian period can be approximated by measurements in fibroblasts that can be considered as a good model for the study of the circadian rhythms. Moreover, manipulation of fibroblasts is not expensive or labor-intensive such as human in vivo protocols, fibroblasts are easy to maintain, accessible to molecular genetic tools and can be easily obtained from human subjects. The Period Length of the Human Circadian Clock Measured in Fibroblasts and in Vivo L. Pagani1, J. Izakovic2, C. Cajochen3, S.A. Brown4, L. Pagani1, C. Cajochen2, D.J. Skene3, S.A. Brown4, 1Neurobiology Laboratory for Brain Aging and Mental Health, University of Basel, Psychiatric University Clinic 1Neurobiology Laboratory for Brain Aging and Mental Basel, 2Department of Dermatology, University Hospital Health, University of Basel, Psychiatric University Clinic Basel, Basel, 3Center of Chronobiology, Psychiatric Basel, 2Center of Chronobiology, Psychiatric University University Clinic Basel, University of Basel, 4Institute of Clinic Basel, University of Basel, Switzerland, Pharmacology and Toxicology, University of Zurich, Zurich, 3Chronobiology Laboratory, School of Biological Sciences, University of Surrey, UK, 4Institute of Pharmacology and Toxicology, University of Zurich, Switzerland Aims: In animals such as in human beings ageing leads to several pathologies, including circadian rhythms disturbances. Aims: Human behaviour is governed by the period length Nevertheless, the origin of this phenomenon is unknown. To bet- of a circadian clock in the suprachiasmatic nucleus of the brain ter understand the rebound of ageing on the circadian rhythms we hypothalamus. Measurement of this period requires extensive characterized the period length of skin fibroblasts, an exploited in subject observation under controlled-light laboratory conditions. vitro model of peripheral oscillator, from 18 young and 18 elderly However the molecular mechanisms of the circadian clock are subjects. Methods: Fibroblasts were isolated from 2 skin punch conserved in most cells of the body. In this study, we have for biopsies and infected with an engineered lentiviral circadian the first time directly compared human fibroblast period length reporter (mBmal-1::luc) to characterize the circadian rhythms in with the free-running period length of human circadian behaviour vitro by bioluminescence imaging. Measurements were conducted measured in the same subjects in order to establish whether fibro- over a time period of 5 days under standard experimental condi- blasts can be a good model for the study of the circadian rhythms. tions using Dulbecco’s Modified Eagle’s Medium containing 10% Methods: In vivo circadian period length from 10 sighted sub- foetal bovine serum (FBS) or 10% of human serum. In addition, jects has been measured in a Short Sleep Wake Cycle Constant we evaluated data on the chronotype of the subjects by analysing the Munich Chronotype Questionnaire (MCTQ). Results: From Mel (from 5 nmol/ml up to 500 nmol/ml). Low Mel concentra- the analyses of the MCTQ, the young subjects recruited for the tion did not influence the circadian rhythms, whereas higher Mel study have an early chronotype compared to the subjects belong- concentrations reduced the circadian period length of peripheral ing to the older group; however no influence on the circadian oscillators. The influence of Mel on the circadian period length is period length of fibroblasts has been found upon the gender and present only after at least one day of pre-incubation of cells with the age in standard experimental conditions. To better understand Mel. To investigate the pathway through which Mel acts on the the influence of age-related factors on the circadian rhythms, circadian rhythms, cells were treated with Mel and a Gi-inhibitor. human fibroblasts have been measured in human serum contain- The treatment with Gi-inhibitor alone did not shorten the period ing medium. No difference in period length was found between length compared to untreated cells, whereas treatment with Mel treatment with FBS and “young” serum. In contrast, treatment or Mel and Gi-inhibitor shortened the fibroblasts circadian period with serum from older donors significantly reduced the circadian length of. Conclusions: From this study we can conclude that period length. To unmask the nature of the substance responsible Mel is not able to synchronize peripheral oscillators, but it seems for the influence of human serum on fibroblasts period length, able to regulate the speedy of the period length. The effect of Mel sera have been heat-inactivated. Heat-inactivation of sera from on the length of the circadian period was stronger when cells were older donors almost undid the reduction in the circadian period treated with Mel for at least 24 hours. Furthermore the effects of length suggesting that protein/s is/are responsible for the circa- Mel on fibroblasts do not seem to be mediated via Gi-coupled dian disturbances observed in elderly. Conclusions: Data of this study reveal that there is no correlation between circadian period length and age at a molecular level. Thus apparently, the interplay between the molecular components of the skin fibroblasts oscilla-tors does not change during ageing in general. However, the age-related changes in circadian rhythms in vivo are possibly related to still unkown protein/s present in the serum.
Sleep Disruption Aggravates Focal Cerebral Ischemia and Induces Expression of the Supported by grants from EU-CLOCK from E.U. #LSHM-CT-2006-018741, Growth-inhibiting Gene Neurocan in the Rat from Désireé & Niels Yde Fundation and from Fonds der Freiwilligen Akademischen Gesellschaft Basel.
B. Gao, E. Cam, H. Haeger, C. Zunzunegui, C.L. Bassetti Dept. Neurology, University Hospital Zurich, Switzerland Background/Aim: Sleewake disturbances (SWDs) are frequently observed in ischemic stroke patients and are linked with poorer function outcomes. The impact of sleep disrup-tion on stroke pathophysiology is, however, unknown. The aim of this study was to investigate two sleep disruption procedures L. Pagani1, F. Meier1, C. Cajochen2, A. Eckert1 including sleep deprivation (SD) and sleep disturbance (SDis) 1Neurobiology Laboratory, Psychiatric University Clinic on ischemia-induced brain damage and expression of neuro- Basel, University of Basel, 2Center of Chronobiology, plasticity-related genes in a rat stroke model. Methods: Focal Psychiatric University Clinic Basel, University of Basel, cerebral ischemia was induced by coagulating the distal middle cerebral artery (MCA) in Sprague Dawley rats (n=44). 12 h after the ischemia surgery, SD was carried out by gentle handling. In the SD experiment, SD was carried out once and rats were sac- Aims: Melatonin (Mel) is the principal substance synthesized rificed at the end of SD. In the SDis experiment, SD 12 h was and secreted by the pineal gland during the darkness. Despite Mel carried out for 3 consecutive days in which rats were allowed is one of the main markers of the circadian rhythms its physi- to sleep for 12h following SD. EEG was registered to evaluate ological function is still unclear. To investigate the effects of vigilance states, infarct volumetry and TUNEL-positive cells Mel as a chronobiotic substance, we characterized the circadian were determined to assess brain damage, and the Taqman® PCR rhythms of human peripheral oscillators in the presence of dif- assay was for expression of several neuroplasticity-related genes, ferent concentrations of Mel. Methods: Human fibroblasts were such as growth-promoting genes c-jun, gap43 and the growth- isolated from 2 skin punch biopsies and infected with an engi- inhibiting gene neurocan. The plasma corticosterone level was neered lentiviral circadian reporter (mBmal-1::luc) to character- determined by Radio Immuno Assay. Results: Both SD12h and ize the circadian rhythms in vitro by bioluminescence imaging. SDis significantly increased the infarct volume and number of The influence of Mel on the circadian rhythms was investigated tunnel-positive cells. Compared to the paired ischemia control in different experimental conditions. Results: The treatment of groups, the infarct volume was increased by 41% in the SD12h fibroblasts with Mel did not increase the synchronization of the (94.1+16.9 vs.66.5+18.2mm3, p=0.022) and 88% in the SDis cells. Cells that have been treated with a rather low concentra- group (74.6+40.9 vs.39.7+43.4mm3, p=0.023) respectively. The tion of Mel (5 nmol/ml) showed a dampening rate similar to cells number of TUNEL-positive cells was increased by 137% in the treated with vehicle; instead fibroblasts treated with a rather high SD (46.8+15 vs.19.7+ 7.7/mm2, p<0.001) and 219% in the SDis concentration of Mel (500 nmol/ml) showed a higher dampen- group (32.9+13.2 vs.10.3+2.5/mm2, p=0.002), respectively. SDis ing rate. To better understand the rule of Mel on the circadian also significantly elevated the expression of neurocan (14.3+0.4 rhythms, cells were measured with different concentrations of vs.6.2+0.1, p<0.001), an axonal extension inhibitory molecule, with older subjects exhibiting attenuated complexity and aug-in the injured hemisphere. Neither in the SD nor in the SDis mented non-linearity in the sympathetic domain, particularly dur-experiment was the corticosterone level increased significantly. ing REM sleep. The main implication is that this sleep stage in Conclusion: Sleep disruption aggravates ischemia-induced older subjects may represent a physiological state risk for cardio-brain injury and alters expression of neuroplaticity-related genes. The impact of these changes on function recovery needs to be fur-ther investigated. Duration And Intellectual Abilities from O.G. Jenni, L. Molinari, J. Caflisch A.U. Viola1, S.L. Chellappa1, E. Tobaldini2, A. Porta2, Child Development Center, University Children’s Hospital 1Centre for Chronobiology, University of Basel, Switzerland, 2Department of Clinical Sciences, Internal Introduction: A large body of adult research indicates that Medicine II, L. Sacco Hospital, University of Milan, Italy, certain types of learning and cognitive processes depend on spe- 3Department of Physiology, University of Rio Grande do cific characteristics of sleep. However, these issues have only rarely been examined in children and adolescents. Methods: We studied the association between sleep duration and intellectual Aim: To test the hypothesis that symbolic indices of non-lin- abilities in 493 children and adolescents of the Zurich Longitudinal ear dynamics of heart rate variability (HRV) can track down age- Studies (born between 1974 and 1993, 352 term born children and related effects and modulation by different sleep stages. Methods: 141 preterm born children). Structured face-tface interviews with Polysomnographic and electrocardiogram (ECG) recordings were parents were performed for various sleerelated habits: among performed in 12 young (21.1±0.8y) and 12 older (64.9±1.9y) other variables, bedtime, wake time, and daytime sleep duration healthy volunteers during one experimental night. Spectral analy- were asked for the 3 months preceding the consultation. Mean sis of HRV was performed as first step analysis. Symbolic analysis sleep duration (mean standard deviation scores, SDS) from age 1 (SA) was done from ECG recordings and based on an automatic to 16 years was calculated from bedtime and wake time including algorithm that constructed RR interval time series for each sleep day time sleep duration (= time in bed per 24 hours). Intellectual stage (W, NREM, REM). SA and Shannon entropy (SE) assess- abilities were assessed using two different instruments: the ments were then performed on time series of 4 epochs per sleep German version of the WISC-R at ages 7, 9 and 14 years and stage. SA encompasses conversion of time series into sequence the German AID at ages 11, 13 and 16 years. Results: A con-of symbols and subsequent construction of series of patterns with sistent negative correlation at all ages between mean SDS sleep three symbols, out of which four patterns can be identified: 0V duration and the intelligence quotient (IQ) was found. Both verbal (no variation), 1V (one variation), 2LV (two like variations) and and performance subscales contributed to the effect. Partial cor- 2UV (two unlike variations). The first is a marker of sympathetic relations (controlled for birth date, prematurity and SES) between activity, while the last refers to parasympathetic modulation. SE mean sleep duration and IQ were between -0.172 and -0.544. A measures complexity of distribution patterns and reduced values hierarchical model including the covariates gender, prematurity, indicate no relationship between subsequent patterns, which hap- SES and birth date showed that IQ was significantly higher by an pens in aging and diseases. Results: During sleep, older subjects average of 3 IQ points per hour less sleep. While birth date was exhibit a fall in absolute HRV indices and a loss of sympathova- not related to IQ, prematurity and SES both showed the expected gal sleestage dependant variation. Normalized LF/(LF+HF) ratio effect: former preterm children exhibited consistently lower intel- remains sleestage dependant, albeit increased in older subjects. lectual abilities (mean 3 IQ points) than term born children and Sleep stages show significant differences in SA and SE measures the average difference between the lowest and highest SES class in older subjects, with striking increase in 0V pattern and decrease was 20 IQ points. Conclusions: Individuals with a short sleep in 2UV pattern, together with reduced entropy rate. When con- duration showed higher intellectual capacities than those with a sidering sleep stages, these age-differences were significantly long sleep duration. Thus, individual differences in sleep duration predominant during REM. Furthermore, comparisons between and individual variation in intellectual abilities may share com-the two groups yielded a significant reduction in SE during REM, with increased 0V and decreased 2UV patterns in older subjects. Conclusions: The fall of absolute HRV indices in older subjects indicates a decrement of the autonomic nervous system. Increase in LF/(LF+HF) ratio during sleep implies that older subjects lose parasympathetic control on the autonomic balance. Non-linear dynamics and complexity of HRV undergo drastic age-changes, and dopamine D1 and D2 receptors contribute to the stimulant effects of the drug. Our laboratory recently demonstrated a role Quantification of the Periodicity of Inter for catechol-methyltransferase (COMT), an important breakdown enzyme of cortical dopamine, in the efficacy of modafinil in healthy volunteers. Specifically, the Val158Met single nucleotide C. Rummel, H. Gast, K. Schindler, J. Mathis polymorphism of COMT, responsible for a 3 - 4 fold reduction in Dept. of Neurology, Inselspital Bern, Switzerland activity of this enzyme, predicts subjective and objective efficacy of modafinil after sleep deprivation. Homozygous carriers of the Aims: Periodic leg movements in sleep (PLMS) are a fre- high-activity Val allele show better response to modafinil in allevi- quent finding in polysomnography. Most patients with restless ating impaired subjective well-being, sustained vigilant attention legs syndrome (RLS) display PLMS. However, since PLMS are and executive functioning than individuals with low-activity Met/ also often recorded in healthy elderly subjects, the clinical signifi- Met genotype. We further investigated the interaction of modafinil cance of PLMS is still discussed controversially. Leg movements with sleep wake-regulation in a double-blind, placebcontrolled are seen concurrently with arousals in obstructive sleep apnoea study in healthy men (age: 23.4 ± 0.5 years). Two matched groups (OSA) may also appear periodically. Quantitative assessment of of 10 Val/Val and 12 Met/Met allele carriers completed 40 hours the periodicity of LM/PLM as measured by inter movement inter- of prolonged wakefulness on two occasions separated by one vals (IMI) is difficult. This is mainly due to influencing factors week. Administration of 100 mg modafinil or placebo was sched-like sleep architecture and sleep stage, medication, inter and intra uled after 11 and 23 hours of wakefulness. Subjective sleepiness patient variability, the arbitrary amplitude and sequence criteria was assessed and the waking EEG recorded at 3-hour intervals. A which tend to broaden the IMI distributions or make them even baseline and recovery night in the sleep laboratory preceded and multi-modal. Methods: Here a statistical method is presented followed the sleep deprivation. Independent of COMT genotype, that enables eliminating such effects from the raw data before modafinil reduced subjective sleepiness and EEG theta activity analysing the statistics of IMI. Rather than studying the absolute (5-8 Hz) in the waking EEG, a suggested objective marker of size of IMI (measured in seconds) we focus on the shape of their sleep pressure during wakefulness. Sleep state analyses revealed distribution (suitably normalized IMI). To this end we employ no reliable effects of modafinil in recovery sleep. Nevertheless, in methods developed in Random Matrix Theory (RMT). Patients: Val/Val allele carriers, modafinil increased 2.75 - 6.25 Hz activ- The periodicity of leg movements (LM) of four patient groups (10 ity in nonREM sleep (stages 2-4) when compared to placebo. No to 15 each) showing LM without PLMS (group 1), OSA without differences between modafinil and placebo were observed in Met/ PLMS (group 2), PLMS and OSA (group 3) as well as PLMS Met subjects. Based on these results, in combination with our ear-without OSA (group 4) are compared. Results: The IMI of lier findings, we suggest that sleep deprived Val/Val allele car-patients without PLMS (groups 1 and 2) and with PLMS (groups riers benefit more from modafinil than Met/Met allele carriers. 3 and 4) are statistically different. In patients without PLMS the Wakefulness after modafinil may be more intensive in these sub- distribution of normalized IMI resembles closely the one of ran- jects when compared to placebo, and underlie the observed EEG dom events. In contrary IMI of PLMS patients show features of changes in recovery sleep. Our data further indicate a dissociation periodic systems (e.g. a pendulum) when studied in normalized between subjective and suggested EEG markers of sleep propen-manner. Conclusions: For quantifying PLMS periodicity proper normalization of the IMI is crucial. Without this procedure impor-tant features are hidden when grouping LM/PLM over whole nights or across patients. The clinical significance of PLMS might be eluded when properly separating random LM from LM that show features of periodic systems. Cognition and High Density EEG in Sleep after Paramedian Thalamic Stroke R. Poryazova1, R. Khatami1, E. Werth1, P. Brugger1, R. Huber2, C.L. Bassetti1 Department of Neurology, University Hospital Zurich, 2University Children’s Hospital Zurich, Zurich, Switzerland Val158Met Genotype Predicts Effects of Modafinil on Recovery Sleep EEG After Sleep Deprivation Objectives: Functional recovery after stroke depends on the adaptive plasticity of the human brain. Sleep contributes essen- S. Bodenmann1, V. Bachmann1,2, C. Stoll1, E. Geissler1, tially to brain plasticity and learning. The exact link between K. Jaggi-Schwarz1, H.P. Landolt1,2 sleep, EEG and cognition in stroke recovery remains unclear. 1Institute of Pharmacology and Toxicology, University of The aim of the study is to investigate the relationship between Zurich, Zurich, 2Zurich Center for Integrative Human sleep and stroke recovery by: 1) comparing EEG power in the Physiology, University of Zurich, Zurich, Switzerland slow wave (SWA) and spindle frequency ranges (SFR) in the acute phase after stroke and 3 months later, 2) correlating these EEG parameters to clinical and behavioral changes during the Although the mode of action of modafinil remains to be recovery process.Patients and methods: Seven patients, two with debated, dopaminergic mechanisms related to dopamine re-uptake bilateral paramedian thalamic stroke (PMTS) and five with uni- lateral PMTS and five matched controls underwent detailed neu- higher Epworth sleepiness score (p=0.019). There was a trend for ropsychological examination, actigraphy, 24h-polysomnography lower score on activities of daily living (p=0.08). Conclusion: (PSG) and high-density EEG (hd-EEG) during sleep. The patients RBD in PD patients is associated with various sleep disorders, were studied in the acute phase after stroke and 3 months later. including insomnia, sleep disordered breathing, restless legs, Results: Patients performed worse in verbal fluency test, had nightmares and hallucinations, leading to higher arousability and longer total sleep time in PSG and higher percentage of day rest sleep fragmentation. Patients with RBD were also sleepier than in actigraphy (acute and chronic). The patients with bilateral patients without RBD.
PMTS had more profound hypersomnia and impairment in behav-ioral tests. In hd-EEG both SWA and sleep SFR were recorded in typical locations without gross asymmetry. In comparison to healthy controls patients had significantly lower power in SWA in frontal and occipital regions after 3 months; power in SFR was lower although not statistically significant. After 3 months hyper- Sleepwalking in Patients with Parkinson’s somnia and behavioral tests improved, especially in the patients with bilateral PMTS. A significant decrease in the SWA power was found while power in SFR did not change. Conclusion: R. Poryazova, M. Oberholzer, C.L. Bassetti Behavioral and EEG changes were more profound after bilateral Department of Neurology, University Hospital Zurich, than after unilateral PMTS. Power in SWA after PMTS decreased after 3 months and was significantly lower than in controls while power in SFR was low initially and did not change. Improvement Objectives: Sleepwalking (SW) corresponds to a complex of cognition in bilateral PMTS was not accompanied by increase sleeassociated behavior, which includes locomotion, mental con- fusion, and amnesia. SW is present in about 10% of children and 2-4% of adults, but only 0.6% of adults report SW “de novo”. The study is supported by the Zurich Center for Integrative Human In a series of 165 consecutive patients with Parkinson’s disease (PD) we found SW “de novo” in 6 (3.6%). The aim of the present study was to assess the frequency and the characteristics of SW in patients with PD. Methods: A questionnaire including items on sleep quality, sleep disorders, in particular SW and REM sleep behavior disorder (RBD), PD characteristics and severity, was sent to the members of the national PD patients organization in REM Sleep Behavior Disorder in Parkinson’s Switzerland. Results: 420 questionnaires were received. Three patients had to be excluded for diagnoses other than idiopathic R. Poryazova, M. Oberholzer, F. Siclari, C.L. Bassetti PD. 36/417 patients (9%) reported adult SW, of them 22 (5% of the population studied) had SW “de novo”. Patients with SW had Department of Neurology, University Hospital Zurich, significantly longer disease duration (p=0.035), they reported more often hallucinations (p=0.004) and nightmares (p=0.003), their dream content was more variable (p=0.046) and they had Objectives: Rapid eye movement sleep behavior disorder higher scores, suggestive for RBD in a validated questionnaire (RBD), reported in up to 50% of patients with Parkinson’s dis- (p=0.001). Patients with SW also had a trend for a higher Epworth ease (PD), is characterized by loss of normal muscle atonia dur- sleepiness scale score (p=0.055). Conclusion: Our results sug- ing REM sleep which leads to increased phasic motor activity gest that SW in PD patients is more common then in the general and allows dream enactment behavior. There are three previous populations. SW appears to be a late manifestation of PD. Most large questionnaire and interview based studies addressing RBD patients with SW also had questionnaire scores suggestive for in PD including 289, 200 and 231 patients respectively. The aim RBD. The simultaneous occurrence of SW and RBD suggests a of the present study was to assess the frequency and the charac- complex disturbance of arousal, locomotion and muscle tone dur- teristics of RBD in patients with PD. Methods: A questionnaire including items on sleep quality, sleep disorders, PD character-istics and severity, was sent to the members of the national PD patients’ organization in Switzerland. A 13-item validated ques-tionnaire for RBD was included. A cut-off of five points is con-sidered suggestive for RBD. Results: 420 questionnaires were received. Three patients had to be excluded for diagnoses other than idiopathic PD. 210/417 patients (50%) had an RBD score>5. These patients reported concomitant sleep disorders like initial insomnia (p=0.03), night-time awakenings (p=0.003), apneas (p=0.008), shortness of breath (p=0.005), talking/crying in sleep (p<0.001), cursing/violent behavior in sleep (p<0.001), restless legs symptoms (p<0.001), nightmares (p<0.001) and hallucina-tions (p<0.001) significantly more often then the rest of the popu-lation studied. They also had longer disease duration (p=0.04) and Benserazide (LD/B) and Pramipexole (PPX) F. Pugin1, A.U. Viola1, S.L. Chellappa1, S. Archer2, D.J. Dijk2, 1Centre for Chronobiology, Psychiatry University Clinics, D. Eggenberger1, C.L. Bassetti2, J. Mathis1 University of Basel, Switzerland 2Surrey Sleep Research 1University of Bern, Inselspital Bern, 2University Hospital Aims: To investigate whether sleewake behaviour patterns Objective: To compare time dependent treatment effects of measured by wrist actigraphy can differ between older subjects the dopamine-agonist pramipexole (PPX; SIFROL®) with a half- homozygous for the longer allele of the gene PERIOD3 (PER35/5) duration (time with a plasma concentration higher than ½ Cmax; and heterozygous (PER34/5) individuals. Methods: Healthy t1/2) of 8-12 hours with dual-release levodopa/benserazide (LD/B; volunteers were selected solely on the basis of their genotype, MADOPAR DR®) (t1/2 = 3-4 hours) on periodic limb movements irrespective of diurnal preference or any other sleerelated charac- (PLM) in patients with idiopathic restless legs syndrome (RLS). teristics. Approximately, 150 healthy older men and women (age Methods: Post-hoc analysis of the PLM-indices of the first to 55-75) have been genotyped for the PER3 polymorphism in an eighth hour (PLM/h1-8) in bed at baseline and under treatment ongoing study, thus comprising individuals who are homozygous with PPX or LD/B obtained from 46 patients in the double-blind, for the longer (PER35/5) and for the shorter allele (PER34/4) as randomized, Swiss multicenter, comparative crossover trial per- well as heterozygous individuals (PER34/5). In parallel, subjec- formed in 2006* (submitted for publication). Results: Under tive assessment of sleep and chronotype was carried out using the baseline conditions the PLM-index was significantly larger in the Pittsburgh sleep quality index (PSQI), the Horne-Östberg and the first four hours (PLM/h1-4) compared to the second half of the Munich Chronotype Questionnaire, the Insomnia severity index night (PLM/h5-8) (p < 0.001). Under treatment PPX and LD/B and the Epworth sleepiness scale. So far, 23 healthy older sub-showed a statistically significant reduction of the PLM-indices in jects have completed sleep analysis by actigraphy and sleep dia- both halves of the night compared to baseline (p < 0.001 all-over). ries to characterize sleep timing during three consecutive weeks. For both drugs the decrease of PLM/h1-4 was significantly larger In this preliminary dataset, we present data of 12 older partici- compared to the decrease of PLM/h5-8 (p < 0.001 for PPX and p pants, whose genotypes have already been characterized. Since = 0.027 for LD/B): PPX reduced the median PLM/h1-4 from 35.8 only two PER34/4 subjects have completed this sleep analysis so to 9.0 (p < 0.001) and the median PLM/h5-8 from 15.7 to 7.4 (p far, they were not included in these results. Results: Preliminary < 0.001), whereas LD/B decreased the median PLM/h1-4 from results on 12 healthy older subject, six PER35/5 subjects and six 32.2 to 11.0 (p < 0.001) and the median PLM/h5-8 from 21.8 to PER34/5 subjects, indicate that there are no differences between 13.1 (p < 0.001). Statistical analysis of the PLM-reduction under the two genotypes with respect to age, gender, body mass index treatment demonstrated no significant difference between the two and to any of the sleepiness, sleep quality, insomnia index and drugs, neither in the first four hours of bedtime (p= 0.207), nor in chronotype questionnaires. On the other hand, the actiwatch the second part of the night (p = 0.831). Conclusion: Both drugs, analysis indicated highly significant differences between the LD/B and PPX, significantly reduce the PLM-index not only in two genotypes: PER35/5 subjects exhibited an earlier sleep time the first four hours (PLM/h1-4), but also in the second part of the (p=0.0001) and shorter sleep duration (p=0.007) in comparison to night (PLM/h5-8). The absence of any difference between the two PER34/5 subjects. Conclusions: These preliminary results point drugs supports the value of PPX as an alternative to the standard to an association between the length of the PER3 polymorphism medication LD/B in the treatment of RLS patients. Conversely, and sleep duration in older subjects, which needs to be confirmed the equal efficacy of both drugs in the second part of the night in further analysis. underlines the positive effects of the dual-release formulation of LD/B in Madopar DR® despite its shorter half-duration (t1/2).
*The Swiss multicenter RLS Study 2006 was sponsored by Boehringer- K. Hefti1, H. Van Hedel2, V. Dietz2, H.P. Landolt1,3 Latency and Sleep Inertia: an Epidemiological Data Analysis 1Institute of Pharmacology and Toxicology, University of Zurich, 2Spinal Cord Injury Center, University Hospital K. Kräuchi1, B. Gompper1, D. Anders1, J. Flammer2, Balgrist, Zurich, 3Zurich Center for Integrative Human Physiology, University of Zurich, Switzerland 1Thermophysiological Chronobiology, Centre for Chronobiology, Psychiatric University Clinics, Basel, Many patients with spinal cord injury (SCI) complain of dis- 2University Eye Clinic, Basel, Switzerland turbed nocturnal sleep and excessive daytime sleepiness. These symptoms could reflect changes in homeostatic and circadian Introduction: The primary vascular dysregulation (PVD) processes of sleewake regulation. To better understand the under- syndrome is a functional disorder of blood flow regulation in lying nature of sleewake disturbances in SCI, we are currently otherwise healthy subjects (mostly women; leading symptoms: performing sleep and quantitative sleep and waking EEG studies cold hands and feet). Women with PVD live not only with cooler in tetraplegic patients in comparison to healthy controls. To date, extremities during the day but also exhibit a circadian phase delay four patients (1 woman, 3 men; mean age: 43 years) with com- of distal skin temperature rhythm with respect to their sleewake plete tetraplegia (Asia A) and four sex- and age-matched volun- cycle (sleep timing) (Vollenweider et al., AJP 2008; Gompper et teers completed the study. Time between occurrence of injury and al., ESRS 2008). There is evidence from epidemiological, ambu- sleep studies equaled 3 months, 6 years, 19 years and 22 years. latory and controlled laboratory studies of a close relationship The study protocol includes all-night polysomnography in the between cold extremities in the evening and prolonged sleep onset hospital, 6 standardized waking EEG recordings at 3-hour inter- latency (SOL) (Kräuchi, SMR 2008). Conversely, the time course vals throughout the day following nocturnal sleep and sleepiness of distal vasoconstriction after a sleep episode is correlated with questionnaires. Saliva samples for melatonin quantification com- the disappearance of sleep inertia (SI) (Kräuchi, JSR 2004). The plete the current data collection. Our first results suggest simi- aim of present epidemiological data analysis is to explore whether lar sleep efficiency, sleep latency, REM sleep latency and sleep PVD, SOL, SI and sleep timing are interrelated. Methods: 145 architecture in patients and controls. In the nonREM sleep EEG, women were recruited from participants of a larger survey in a however, the patients show higher power in delta (2.25-2.5 and random population sample of Basel-Stadt (Kräuchi et al., JSR 4.75-5.5 Hz range) and sigma (13.75-14.5 Hz band) frequencies. 2008). Various questionnaires (questions about thermal discom- Also during wakefulness, the patients have generally higher abso- fort, SI, SOL, sleep timing etc.) were mailed to the subjects. lute power values in the 0.75-4.75 Hz and 15.25-20.25 Hz ranges. The degree of PVD was estimated from questionnaire-derived By contrast, the evolution of relative power in distinct frequency scores (feeling of cold hands and feet). Spearman rank correla- bands across consecutive waking EEG recordings appears similar tions (Rsp) were calculated. Results: The degree of PVD was in both groups. Subjective ratings indicate elevated sleepiness in positively correlated with weekly means of SOL (Rsp=0.265, the patients in comparison to the controls, except at 4 pm. To our p=0.0022), SI (Rsp=0.245, p=0.0061) and sleep timing [(lights knowledge, this is the first quantitative sleep and waking EEG on - lights off)/2; Rsp=0.234, p<0.01]. SOL correlated with SI study in patients with complete tetraplegia. Preliminary findings (Rsp=229, p=0.011), and sleep timing was weakly correlated with indicate subtle differences in the sleep EEG between patients SI (Rsp=0.170; p=0.07), but not with SOL (Rsp=0.086; p=0.35). and controls, in particular in the frequency range of sleep spin- Conclusions: These findings are in accordance with the hypoth- dles. Differences may also be present in the waking EEG. More esis that prolonged SOL and SI are associated with delayed dis- patients are needed to corroborate the preliminary findings and to tal vasodilatation before and delayed distal vasoconstriction after propose possible physiological changes contributing to the clini- sleep, respectively, leading to a phase delay of the thermoregula- cally observed sleewake symptoms in SCI.
tory system with respect to sleep timing (different internal phase of entrainment). Different external phase of entrainment (different phase angle between sleep timing and solar –cycle) seems to be less crucial for SOL and SI.
Research supported by the Schwickert-Stiftung and SNF Grant #3200B0- marker of sleep intensity. Previous research has shown a dra- matic decrease of SWA in the sleep EEG during adolescence [1, Prevalence, Severity, Clinical Relevance, 2], which seems to be paralleled by a decline in synaptic density Pathophysiology and Course of the Restless [3]. Another anatomical correlate of development, the increase Legs Syndrome in Pregnancy: A Prospective, and following reduction of grey matter volumes during adoles-Systematic Study cence [4], can be assessed in vivo on MRI images. Based on the hypothesis that sleep actively supports developmental processes A. Hübner1, K. Krafft2, E. Werth1, S. Gadient1, in the young brain, we looked for grey matter correlates of slow R. Zimmermann2, C.L. Bassetti1 wave sleep (SWS, i.e. sleep with high SWA) duration in the brain. 1Department of Neurology, University Hospital Zurich, Methods: In a sample of 14 children (6 boys; age range 9-13 2Department of Gynecology, University Hospital Zurich, years, mean=10.5y, SD=1.2y) we correlated local grey matter volumes with the percentage of SWS (mean of two nights). To achieve this, we used T1-weighted MRI images and segmented Introduction: Three retrospective studies [1-3] suggested them in a fully automatic way using the SPM5 package (http:// a high frequency (11-26%) of restless legs syndrome (RLS) in www.fil.ion.ucl.ac.uk/spm). Grey matter segments were then pregnancy. None of these studies used standardized methods for smoothed with a 12mm Gaussian kernel and compared voxel by assessment of RLS and sleep. Characteristics and determinants voxel using voxel-based morphometry [4]. We calculated whole-of RLS during pregnancy are poorly known. Methods: Women brain and voxel-wise multiple regressions accounting for sex, age during (from the 2nd trimester, 1x/month) and after pregnancy and whole-brain grey matter volumes. Results: Whole brain grey (8 weeks post-partum) are prospectively studied. Assessment matter volumes (in our sample 891±19ccm) did not correlate sig-includes 1) interview about RLS-symptoms and sleep habits/ nificantly with percentage time in slow wave sleep (Beta =-.085; disturbances: 2) standardized questionnaires (incl. the interna- p=.845). However, we found areas of grey matter correlating tional RLS-scale (IRLSS), Epworth sleepiness scale (ESS), and positively (Beta up to .98; p<.001 uncorrected) with the duration Pittsburgh Sleep Quality Questionnaire (PSQI)); 3) blood tests of slow wave sleep, namely in anterior and posterior cinguli, in (incl. hemoglobin, C- reactive protein and ferritin), 4) leg actigra- the medial temporal gyrus, lateral parietal lobe and in the ventro- phy (3rd trimester and post-partum). Results: So far 308 women medial and dorsolateral prefrontal cortex. No significant negative were included. RLS was diagnosed in 28 women (9.1%). 28% of correlations were found. Conclusions: Our preliminary anatom- them had a positive family history for RLS, 60% reported onset of ical data provide first evidence for a relationship between SWS RLS-symptoms before the 20th week, 60% had RLS-symptoms duration and local grey matter volumes in regions of the brain, daily, 60% had an IRLSS>20, and 85% had a PSQI> 5. Anemia which are highly plastic during adolescence. Ongoing work will (defined as Hb <11 g/dl in pregnant women) was found in 10% of show if SWS-associated regions shift during adolescence.
affected and unaffected women. Ferritin levels <50 were found in 88% of women. Women with (pRLS) and without RLS (nRLS) had similar values in hemoglobin CRP, and ferritin (mean Hb 11.8 1. Jenni O.G. & Carskadon M.A., Sleep 27, 774-783, 2004.
mg/dl in pRLS, 12mg/dl in nRLS; Ferritin 19.1 μg/l in pRLS, 13.3 2. Jenni O.G. et al., Sleep 28, 1446-1454, 2005.
in nRLS). Conclusion: Preliminary results of this ongoing study 3. Feinberg I, Journal of Psychiatry Research 17(4), 319-334, 1982.
suggest that RLS in pregnancy: 1) is present in 10% of women; 4. Giedd J N et al., Nature Neuroscience 2, 861-863, 1999.
2) frequently appears early in pregnancy; 3) is often severe/ fre- 5. Ashburner J & Friston K J, Neuroimage 11, 805-821, 2000.
quent; 4) may not be related (only) to anemia/low ferritin levels; 5) has a significant impact on sleep quality.
1. Manconi M.,Govoni V., De Vito A., Economou N.T., Cesnik E., Casetta I., Mollica G., Ferini-Strambi L., Granieri E., Restless legs syndrome in Behaviourally Induced Insufficient Sleep 2. Goodman J., Brodie C., Ayida G.A., Restless leg Syndrome in pregnancy, 3. Ekbom K.A., Restless Legs, Acta Scand. Med. 1945;158:1-123 E. Werth, N. Michael, C.R. Baumann, C.L. Bassetti Department of Neurology, University Hospital Zurich, Switzerland Introduction: Behaviourally induced insufficient sleep syn- drome (BIISS) occurs when an individual chonically fails to obtain the amount of sleep required to maintain normal levels of alert-ness and wakefulness. Method: This study presents the results of A. Buchmann, S. Kurth, M. Ringli, O.G. Jenni, R. Huber the post hoc evaluation (Clinical evaluation, Sleep questionnaire University Children’s Hospital Zurich, Switzerland (SQ), actigraphy, PSG, MSLT, MWT, HLA typing, hypocretin) of 47 consecutive patients who received the diagnosis of BIISS in our Center of Sleep Disorders. Results: Mean age of the BIISS Aims: Slow-wave activity (SWA; power <4.5 Hz) dur- patient was 40±12 years (mean±SD). Only 30% were females. ing nonREM sleep is a well established electrophysiological Patients mostly complain symptoms of hypersomnia with exces- sive daytime sleepiness, however, many individuals reported other analysis, all four nights from each participant were correctly clas- symptoms as sleep attacks without general daytime sleepiness, sified except for one participant, for whom the two nights at time fatigue, sleep drunkenness, concentration and attention deficits 1 and time 2 were paired but not all 4 nights, possibly due to a 0.4 or cognitive impairment. Mean ESS was 14.1±3.6. Time in bed Hz increase in the spindle frequency peak at time 2. Our results (TIB) estimation based on the SQ revealed TIB of 7:10h±1:03h indicate that - in spite of neurdevelopmental changes - trait-like during weekdays and 8:29h±1:16h on weekend. TIB estimation characteristics are preserved in 90% of our sample across this based on actigraphy recordings revealed significantly shorter TIB developmental phase. Future analyses will include a larger sample on weekdays and on weekends (weekday: 6:25h±0:57h, weekend: size and examine NREM and REM sleep separately.
7:56h±1:13h) compared to TIB taken from the SQ. In this popula-tion the PSG recording revealed short sleep latency 8.4±7.9 min-utes and high sleep efficiency (91.5±16.7%). Mean sleep latency of MSLT was 5.5±3.3 minutes. Sleep onset REM (SOREM) epi-sodes with 2 and more SOREM were present in 8 patients. Mean sleep latency of MWT was very variable. A clear reduced ability to Sleep Patterns and Cognitive Development maintain wakefulness (sleep latency < 12 min) was present in 34% of patients. Conclusion: The results of this case series indicate that there are a noticeable large number of patients who were not A. Geiger1, P. Achermann2, O.G. Jenni1 aware that their sleep duration was insufficient and that there is 1University Children’s Hospital Zurich, Child Development a substantial clinical overlap between BIISS, narcolepsy without Center, Zurich Center for Integrative Human Physiology cataplexy and idiopatic hypersomnia without long sleep. A posi- (ZIHP), 2University of Zurich, Institute of Pharmacology tive response to increased sleep time is diagnostic of BIISS and an and Toxicology, Zurich Center for Integrative Human important feature to differentiate between these three entities. Sleep and cognition are commonly linked – in terms of minimal sleep duration that is assumed for adequate cognitive functioning and in terms of developmental relevance. But even though a role for sleep in learning processes could be demonstrated in studies Stability of Spectral Characteristics in the [1], the association between sleep and cognition on the trait-level is far less clear. Furthermore, most research about the relationship between sleep and cognition focuses on clinical populations or on the effect of disturbed sleep on cognitive performance. To specify L. Tarokh1,2, M. A. Carskadon1,3, P. Achermann4,5
this association in healthy children, 60 children aged 8–11 years 1Bradley Hospital Sleep Research Laboratory, Providence, were comprehensively characterized on cognitive (German ver- R.I., 2Center for Alcohol and Addiction Studies, Brown sion of the WISC IV) and attentional dimensions (alertness and University, Providence, RI, 3Department of Psychiatry and go/nogo paradigm). Additionally, several sleep behavior param- Human Behavior Warren Alpert Medical School at Brown eters for the same children were assessed. Questionnaires, four- University, Providence, RI, USA, 4Institute of teen days of actigraphy recording and sleep diaries were used to Pharmacology and Toxicology, University of Zurich, Zurich, specify sleep behavior parameters such as daytime sleepiness, 5Zurich Center for Integrative Human Physiology, sleep duration, and chronotype. In a correlative analysis, a nega- University of Zurich, Zurich, Switzerland tive correlation of -.29 (p<.05) between full scale IQ scores and sleep duration (according to questionnaires filled out by parents) The waking and sleep EEG spectrum in adults is highly stable could be demonstrated with the main component being fluid intel- across recording sessions separated in time by several weeks. This ligence (r= -.35, p<.05, partial correlation controlled for age and phenomenon has not been studied in children and adolescents in socioeconomic status). Thus, the shorter the sleep duration on free whom brain development may affect the sleep EEG. The pres- days, the higher the scores of cognitive performance. Attentional ent longitudinal study examined whether the sleep EEG spec- parameters did neither correlate with any of the cognitive vari- trum remains stable during the transition from childhood (9/10 ables nor with sleep behavior parameters. Based on the correlation years old) to early adolescence (11 - 13 years old). Standard sleep analysis, regression analyses were conducted. We conclude that, recordings were performed in ten healthy (3 girls) children on two children displaying higher cognitive efficiency during daytime (as consecutive nights when participants were 9/10 years old (time reflected by higher IQ scores) seem to display higher night-time 1) and again 2 -3 years later (time 2; mean = 2.32 years, SD = sleep efficiency (as reflected by shorter sleep duration). In light of 0.28). EEG signals (C3/A2) were Fourier transformed (Hanning the neural efficiency hypothesis [2], the current results argue for window, average of six 5-s epochs) and artifacts rejected using a an upgrading of the original theory – referring not only to daytime semi-automated procedure; data were averaged across the night but also night-time efficient behavior. After all, sleep as well as independent of vigilance states. Spectral power from 0.6 to 16 cognition can both be considered as substrates of neuronal pro-Hz (in 0.2 Hz bins) was normalized and log transformed before cessing and thus should indeed have some common properties.
classification by hierarchical cluster analysis. At both times, the cluster analysis correctly paired consecutive recording nights 1. Walker MP, Stickgold R: Sleep, memory, and plasticity. Annu Rev for every participant. When all 40 nights were entered into the 2. Haier RJ, Siegel EV: Cortical glucose metabolism rate correlates of abstract reasoning and attention studied with positron emission tomogra- phy. Intelligence 1988, 12 (2): 199-217.
The Chronophysiological Dawn of Female Adulthood S. Frey1, S. Balu1, S. Greusing1, N. Rothen2, C. Cajochen1 1Chronobiology, Psychiatric University Clinics, Basel, 2Institute of Psychology, University of Bern, Bern, Source Localization of Slow Oscillations in the Nonrem Sleep EEG: Effect of Sleep Deprivation Most parents experience their children’s puberty as a dramatic change in their family life. This is not surprising considering the dynamics of the physical and psychosocial maturation period, Institute of Pharmacology and Toxicology, University of which occur during adolescence. Although an age- and sex- dependent turn in chronotype has been shown during adolescence chronological age may not give an accurate explanation for the Aims: Slow waves are the most prominent feature of non- physical and behavioural dynamics that occur before the transition REM sleep and are visible at the level of the scalp EEG as large from adolescence to adulthood. Hence, the aim of our study was amplitude fluctuations at frequencies ranging from slow (< 1 Hz) to assess whether there is a relationship between puberty and the to delta (1-4 Hz) activity [1]. At the cellular level they are charac- changes in sleep phase preferences during female maturation and terized by slow membrane potential fluctuations of cortical neu- adulthood. Our study is based on a cross-sectional survey among rons consisting of depolarized up states and hyperpolarized down 906 females aged 5 to 29 years. Habitual sleep timing on week- states alternating synchronously over vast cortical territories with days and free days were investigated along with the date of birth a frequency below 1 Hz [2]. Aim of the present analysis was to and the year of the occurrence of menarche. The distance mea- investigate changes in the activation of brain structures during sured in years from the actual age to menarche onset was taken slow oscillations following sleep deprivation. Methods: EEG as a biological variable that may affect the change in sleep timing recordings during baseline sleep and recovery sleep after 40 hours preference. To account for changes of sleep phase preferences the of wakefulness were analyzed (27 channels, 7 healthy young men, difference between the midpoints of sleep on free days and on slow wave sleep (stages 3 and 4) of the first nonREM sleep epi- weekdays was calculated which has also been aptly described as sode). Slow oscillations were detected based on the Hilbert trans- “social jetlag”. Sleep phase preference was then subjected to a formed EEG of derivation Cz (linked mastoids; band-pass filter one-way ANOVA with the factor “distance to menarche”. Results 0.5-4.5 Hz; peak-tpeak amplitude > 75 μV). Source localization show that in contrast to prepubertal children, female adolescents using the software LORETA [3] was performed for the down- exhibit a striking progression in delaying their sleep phase prefer- states (short time interval around the negative peak). Results: ence until 5 years after menarche. Thereafter, the mid-sleep dif- Preliminary analysis revealed significant clusters of increased ference switches to advancing. Results disclosed a main effect of activation after sleep deprivation in the right frontal lobe and the factor “distance to menarche” (p<0.0001). Post-hoc analysis decreased activation in both parietal lobes, left limbic and tempo- showed that mid-sleep difference at -7 years to menarche and ral lobes. Conclusions: Further analyses are needed to investi- 9.5, 11.5, and 13.5 years after menarche are significantly differ- gate the contribution of the ustates.
ent from the peak in mid-sleep difference at 5 years after menar-che (p<0.0001). Our data provide evidence that the downsizing 1. Massimini M, Huber R, Ferrarelli F, Hill S, Tononi G: The sleep slow of social jetlag 5 years after menarche is a step towards “adult- oscillation as a traveling wave. J Neurosci, 2004, 24: 6862-70.
like” behaviour as the ability to approach more aligned sleep/ 2. Steriade M, Nunez A, Amzica F: A novel slow (<1 Hz) oscillation of neo- cortical neurons in vivo: depolarizing and hyperpolarizing components. J wake cycles during weekdays and weekends emerges. Hence, we conclude that the distance to menarche may serve as an individual 3. Pascual-Marqui RD, Michel CM, Lehmann D: Low resolution electro- biological marker for the beginning of female adulthood.
magnetic tomography: a new method for localizing electrical activity in the brain. International Journal of Psychophysiology 1994, 18:49-65.
Research supported by the Swiss National Science Foundation and the Supported by SNSF grant 320000-112674.
Dynamics of Sleep Homeostasis: A Trait-like T. Rusterholz1,2, H.P.A. Van Dongen3, P. Achermann1,2,4
M. Ringli, S. Kurth, A. Geiger, O.G. Jenni, R. Huber 1Institute of Pharmacology and Toxicology, University of University Children‘s Hospital Zurich, Zurich, Switzerland Zurich, Zurich, 2Neuroscience Center Zurich, University and ETH Zurich, Zurich, Switzerland, 3Sleep and Introduction: Slow wave activity (SWA) during NREM Performance Research Center, Washington State sleep is homeostatically regulated and was suggested to depend on University, Spokane, Wash., USA, 4Zurich Center for synaptic changes during wakefulness [1]. In adults it was shown, Integrative Human Physiology, University of Zurich, Zurich, that a visuomotor learning task involving specific brain regions resulted in a local increase of SWA during subsequent sleep [2]. Knowing that the brain undergoes dramatic changes during devel- The homeostatic Process S reflects the prior history of sleep opment [3, 4], we wondered if the same learning task would have and wakefulness and is modeled by a saturating exponential func- similar effects on SWA in immature children. Methods: All-night tion during waking and an exponential decline during sleep. EEG high density EEG was recorded in twelve prepubertal children slow-wave activity (SWA) is a marker of nonREM sleep intensity (mean age 10.9±1.3 years, right handed) using Geodesics Sensor and serves as an indicator of sleep homeostasis. There is consid- Nets composed of 128 electrodes in two separate nights. The EEG erable inter-individual variation in responses to sleep depriva- recordings were sleep staged, subjected to semi-automatic artifact tion. The aim of the present study was to investigate whether the removal and processed using power spectral analysis. We calcu- dynamics of sleep homeostasis (i.e. the time constants of Process lated for every electrode power in the SWA frequency range (1-4.5 S) are trait-like. EEG data of healthy young adults who underwent Hz) for the first 30 minutes of NREM sleep for each night. Before a laboratory protocol with repeated sleep deprivation was used to sleep subjects completed with their right hand either a visuomotor investigate this. Each subject’s data consisted of EEG recordings learning task or a motor control condition. Results: Our children from eight sleep periods (12 h time in bed, 22:00–10:00), which successfully learned the task. During subsequent sleep, we found were interspersed with three 36 h periods of sustained wakefulness a highly reproducible pattern of SWA topography for the two con- [1]. Empirical mean SWA per nonREM sleep episode at episode ditions. However, a contrast revealed a significant local increase midpoint served as input for least-squares model parameter esti- of SWA over left central (8.3±2.5%; p<0.05) and occipital regions mation. Time constants were constrained to a physiological range (8.6±3.3%; p<0.05) after the learning condition. Discussion: As derived from average data. Parameters were estimated for 6 dif- observed in adults a preliminary analysis revealed local changes ferent pairings of the recording nights. Random-effects ANOVA in SWA triggered by the learning condition in children. Other than was used to investigate the intra- and inter-individual variation in in adults, where higher-order brain regions in the parietal cortex the estimates of the time constants. The standard deviation across were found to be involved in the underlying learning processes, in subjects for the time constant of buildup was 4.9h, and the inter- children significant increases were observed over the left motor individual differences explained 59% of the variance. The standard and visual regions. Based on evidence that higher-order brain deviation across subjects for the time constant of the decline was regions mature later during development [3], our findings may 0.3h, and inter-individual differences explained only 26% of the suggest that in children lower-order brain regions seem to execute variance. Thus, there was substantial inter-individual variability in the dynamics of Process S during wakefulness but not during sleep. It remains to be determined to what extent the inter-individual dif- 1. Tononi G, Cirelli C: Sleep and synaptic homeostasis: a hypothesis. Brain ferences during wakefulness are a byproduct of the model fitting 2. Huber R, Ghilardi MF, Massimini M, Tononi G: Local sleep and learn- procedure, or a result of systematic inter-individual differences in waking activities, or evidence of a trait-like feature.
3. Gogtay N, et al.: Dynamic mapping of human cortical development dur- ing childhood through early adulthood. Proc Natl Acad Sci U S A 2004; 1. Tucker AM, Dinges DF, Van Dongen HPA: Trait interindividual differ- ences in the sleep physiology of healthy young adults. J. Sleep Res. 2007, 4. Sowell ER, et al.: Longitudinal mapping of cortical thickness and brain growth in normal children. J Neurosci 2004; 24: 8223-31.
Supported by SNSF grant 320000-112674 and NIH grants HL70154 and Age-related Changes in Response to Sleep R. Wehrle1, J.V. Retey1, M. Adam1, H.P. Landolt1,2 S. Kurth, M. Ringli, A. Geiger, O.G. Jenni, R. Huber 1Institute of Pharmacology & Toxicology, University of Children‘s Hospital Zurich, Switzerland Zurich, 2Center for Integrative Human Physiology, University of Zurich, Switzerland In adults, numerous reports show sleep dependent perfor- mance improvement on various tasks [1]. For example, subjects Aim: Recent evidence suggests that healthy older individu- trained on a computer-based visumotor rotation task showed als are less vulnerable to performance impairment after sleep improved test performance after sleep [2]. We investigated a sim-deprivation than young individuals. Some of the age-related dif- plified version of this task in 16 children (10.4±1.2y, 8 males) ferences may be reminiscent of the response in young people and 8 adolescents/adults (20.3±3.9y, 3 males). In this task right-to the adenosine receptor antagonist caffeine. To study possible handed subjects using a computer mouse had to perform out and adenosinergic mechanisms contributing to age-related changes in back movements towards a target as accurately and fast as pos- sleewake regulation, we examined the effects of sleep deprivation sible. Unbeknown to the subjects they had to overcome a com- and caffeine on sleepiness, daytime functions and EEG variables puter imposed rotation and trying to adapt their movements to in young subjects of high and low self-rated caffeine sensitivity hit the targets. A 25-min training session followed by a first test and healthy older men. Methods: Twelve young caffeine-sen- took place in the evening before sleep. Subjects were retested on sitive (23.7 ± 3.3 years), 10 young caffeine-insensitive (25.0 ± the task next morning. We compared learning performance of the 2.1 years), and 10 older volunteers (66.1 ± 3.3 years) were stud- subjects during the training session, as well as the changes in test ied. All participants completed two 40-hour periods of prolonged performance before and after sleep. Performance was assessed wakefulness in subsequent weeks preceded by two 8-hour nights by the directional error (degrees of accuracy), the angle between in the sleep laboratory. In both conditions, two capsules were the target and actual movement direction. At the end of the train-administered after 11 and 23 hours waking, containing either 2 x ing session, similar levels of directional error were reached in 200 mg caffeine or placebo in randomized, double-blind, cross- children (25.5 degrees) and adolescents/adults (21.8 degrees). over fashion. Subjective sleepiness (Stanford Sleepiness Scale), However, adolescents/adults reduced their movement variance sustained vigilant attention (PVT), and waking EEG (eyes open, across the training sessions (i.e. the precision of hitting targets) to eyes closed) were obtained at 3-hour intervals throughout pro- a larger extent than children did (adolescents/adults, -24.5 degrees longed wakefulness. Significant effects of sleep deprivation and children -13.7 degrees, p<0.05). After sleep, both children and caffeine were identified with mixed-model ANOVAs. Results: adolescents/adults showed improved task performance. The Sleep deprivation impaired subjective sleepiness and PVT perfor- improvement was larger in adults (+35.2±13.8%) than in children mance in all groups, with the sensitive young participants being (+10.8±4.7%, p<0.05). We found no correlations between the most affected. The pharmacokinetics of caffeine in saliva did not after sleep performance improvement and sleep variables (sleep differ among the groups. Notwithstanding, caffeine most potently efficiency, NREM sleep, REM sleep). A possible explanation for improved sleepiness, and PVT measures following sleep loss in the reduced sleep dependent task performance improvement in the young sensitive group, whereas insensitive and older subjects children might be their larger movement variance during the train- showed little effects. These differences among the groups were ing session, i.e reduced movement accuracy, and/or a not yet fully not paralleled by the effects of sleep deprivation and caffeine on delta and theta power in the waking EEG. Conclusions: Sleep deprivation and caffeine have less effect on subjective sleepiness 1. Born J, Rasch B, Gais S, Sleep to remember. Neuroscientist 2006; and objective performance in older subjects than in young caf- 2. Huber R, Ghilardi MF, Massimini M, Tononi G, Local sleep and learn- feine sensitive subjects. These differences may reflect age-related changes in adenosine and adenosine receptors, which are well established in animals and humans. A dissociation among sleepi-ness, performance, and putative EEG measures of alertness dur-ing prolonged wakefulness is suggested.
Research supported by Swiss National Science Foundation and EU Marie- Sleep Regulation in Mice with Adenosine Metabolism Deficiency S. Palchykova, A. Gerling, R. Winsky-Sommerer, I. Tobler Institute of Pharmacology and Toxicology , University of Zurich, Switzerland Introduction: It is suggested that the function of sleep is associated with energy metabolism. Adenosine is a neuromodula-tor implicated in sleep promotion and regulation due to its ability to decrease neuronal activity and due to a direct link of adenos- activity (FLA) and subjective sleepiness during a 40-h episode ine to the energy metabolism in the cell. It has been shown by of sustained wakefulness in depressed in comparison to healthy microdialysis in cats and rats that adenosine increases during pro- control women. Methods: Depressive volunteers full field longed wakefulness. Moreover, adenosine, its analogs and ade- the DSM-IV criteria of a major depression assessed by a SKID nosine receptor agonists induce sleep, while adenosine receptor Interview. Twelve EEG channels were recorded continuously in antagonists induce waking. Adenosine kinase (ADK) is thought to 6 young depressive (26 mean age +/-5sd) and 8 young healthy be a primary route of adenosine metabolism because its pharma- women (25.33mean age+/-4.03sd) during a 40-h sleep depriva- cological inhibition leads to increased adenosine levels in vitro. tion protocol under constant routine conditions. The Karolinska Therefore, reduced levels of ADK may lead to an increase of drowsiness test was carried out in 1.25-h intervals for 3 minutes endogenous adenosine in the brain. To further investigate the role in order to gather artefact free EEG samples. The EEGs were of adenosine in sleep we exploited a mouse model with a disrup- offline manually scored for artefacts and subjected to spectral tion of the Adk gene (the Adk knock-out mice do not survive after analysis. Subjective sleepiness was assessed using the Karolinska 14 days postnatal). Methods: Male Adk+/- mice (n=15) and Sleepiness Scale every 30 min during the 40-h of sustained wake-wild-type (WT) littermates (n=14) were used in the study. Sleep fulness. Results: We observed a significantly faster build-up was continuously recorded during a 24-h baseline followed by 6 h of frontal-low EEG activity (FLA, 1-4.5 Hz) during the first 16 sleep deprivation (SD) and 18 h recovery. The effects of the ADK hours of the CR (group x time, p<0.01). Post-hoc analysis yielded inhibitor, 5’-iodotubercidin (ITU; 0.75-1.5 mg/kg i.p.), on sleep, significantly more FLA during the late evening in depressed than EEG spectra and brain temperature were evaluated in a cross-over healthy women (p<0.03). This was paralleled by higher sleepiness design with vehicle. Results: During baseline the amount and ratings during this time window (p<0.05). Conclusion: Rather distribution of waking, NREM sleep or REM sleep and their cor- than an S-deficiency, we have preliminary evidence for a faster responding EEG power did not differ between Adk+/- and WT increase of sleep homeostatic pressure (process S) during wake-mice. SD induced a similar marked enhancement of NREM sleep fulness in depressed women. This could reflect a use-dependent amounts and slow-wave activity (SWA) in NREM sleep above phenomenon (enhanced rumination in depression), since it was baseline levels in both genotypes during the first 6 h recovery. only observed in frontal brain regions. Alternatively, it could be No genotype difference was observed in the slow-wave energy the consequence of a less efficient circadian arousal signal in the in NREM sleep at the beginning of recovery or 18 h later. ITU evening, thus allowing more intrusion of delta in frontal brain induced mild hypothermia lasting approximately 1-2 h concomi- tant with a prominent reduction in EEG power in most frequen-cies in waking and NREM sleep during the first 2 h post-injection. This Research was supported by Swiss National Science Foundation.
SWA failed to be enhanced during 9 h following ITU treatment. Conclusion: Sleep and sleep regulation were not affected by the reduced expression levels of ADK, indicating compensa-tory mechanisms. Different kinetics of the processes underlying the build-up of sleep pressure and of ITU metabolism as well as hypothermia may have contributed to the lack of SWA increase Biomarker Distinguishing Excessive Daytime Sleepiness from Fatigue? Supported by SNSF grant nr. 3100A0-112528/1 and EU LSHM-CT-2005- U. Kallweit1, K. Aritake2, Y. Urade2, M. Uhl1, C.L. Bassetti1, C.R. Baumann1 1University Hospital Zurich, Department of Neurology, Zurich, Switzerland, 2Osaka Bioscience Institute, Molecular Behavioral Biology, Osaka, Japan Background/Aims: Endogeneous sleepromoting substances such as prostaglandins are suggested to be involved in homeostatic sleewake regulation. Lipocalin-type prostaglandin D synthase (L-PGDS) catalyzes the production of somnogenic prostaglandin A. Birchler-Pedross, S. Frey, P. Brunner, T. Götz, D, and appears to be decreased in the cerebrospinal fluid (CSF) of patients with excessive daytime sleepiness (EDS). It is, how-ever, not known whether CSF L-PGDS levels are decreased also Centre for Chronobiology, Psychiatric University Clinics, in patients with fatigue. We aimed at measuring CSF L-PGDS levels in relation to the presence of EDS and fatigue. Methods: In this prospective study, we included consecutive patients with Aims: In an ongoing study we aim at investigating whether neurological disorders, and with a clinical indication to perform women suffering from major depression show impairments in a lumbar puncture. Patients with primary sleewake disorders (e.g. the regulation of sleewake homeostasis (process S) and circadian narcolepsy) were excluded from this analysis. We performed all rhythms. Here we focus on a marker of process S during wake- lumbar punctures between 10 a.m. and 1 p.m. Thereafter, CSF fulness. Thus, we quantified the time course of frontal low-EEG was immediately frozen at -80°C and sent to Japan for L-PGDS determination by ELISA. We analyzed data from 15 patients (9 EEG data (Rétey et al., 2005), they suggest that individuals with women), mean age was 43 years (range 22-61). For the assessment genetically reduced adenosine deaminase activity (G/A-genotype) of EDS and fatigue, we administered validated questionnaires are under higher sleep pressure than subjects with unimpaired (Epworth sleepiness scale; ESS, and fatigue severity scale; FSS). adenosine metabolism (G/G-genotype).
Results: Six of 15 subjects suffered from EDS (ESS >10), 8 sub-jects from fatigue (FSS>4.0). EDS patients had lower L-PGDS Research supported by Swiss National Science Foundation and Zürich levels (16.3 ± 3.4 μg/mL) than non-EDS patients (22.6 ± 7.5 μg/ Center for Integrative Human Physiology.
mL, p=0.048). On the other hand, levels were similar between patients with (20.2 ± 9.0 μg/mL) and without fatigue (19.9 ± 3.3 μg/mL, p=0.94). An association between CSF L-PGDS and gender or age was not found. Conclusion: The first preliminary results of this ongoing bicenter study indicate L-PGDS may be the first biomarker differentiating EDS from fatigue. Per3 Polymorphism Predicts Physiological and Cognitive Performance Response to Sleep Loss Center for Chronobiology, Psychiatric University Clinics A Functional Genetic Variation of Adenosine Deaminase Modulates Waking Performance Irrespective of Self-rated Sleep Duration Aims: To investigate the physiological consequences of the PER3 polymorphism. Methods: Healthy volunteers were V. Bachmann1,2, F. Klaus1, S. Bodenmann1, P. Brugger2,3,
selected on the basis of their genotype, irrespective of diurnal preference or sleep related characteristics. Approximately 400 1Institute of Pharmacology and Toxicology, University of volunteers (age 20-35) were genotyped for the PER3 polymor- Zurich, 2Zurich Center for Integrative Human Physiology, phism. Individuals homozygous for the rarer 5-repeat VNTR University of Zurich, 3Department of Neurology, University (PER3-5/5) and for the 4-repeat VNTR (PER3-4/4) were identi- fied and matched for age, gender, ethnicity and body mass index. Twenty-eight subjects wore actigraphs and completed sleep dia- Accumulating evidence suggests that the neuromodulator ade- ries for three weeks prior to the five-day laboratory study. The“ nosine contributes to homeostatic sleewake regulation. In humans, in-lab” part of this study was completed by 24 participants and a functional polymorphism of the gene encoding the adenosine comprised two baseline sleep episodes followed by 40-h sleep metabolizing enzyme, adenosine deaminase (ADA), predicts deprivation protocol under constant routine conditions, with a duration and intensity of slow wave sleep in healthy individuals subsequent recovery sleep episode. Results: Sleep diaries and (Rétey et al., 2005). The impact of this polymorphism on wak- actigraphy analyses indicated that the two groups did not differ ing performance is unknown. Alleles and genotypes of the 22G>A with respect to their habitual sleewake timing and sleep dura- polymorphism of ADA were determined in 242 males (n=124) tion. Similarly, the two groups did not differ with respect to the and females (n=118). All subjects (age range: 18-40 years) filled timing of markers of hormonal rhythms, such as melatonin and in questionnaires about sleewake habits, and their performance on cortisol, and on a novel circadian marker, PER3 mRNA. On the tasks of executive functioning, attention and memory were sys- other hand, comparison of sleep structure between PER3-5/5 and tematically quantified. In subsequent studies, individual carriers PER3-4/4 subjects yielded significant differences: PER3-5/5 sub- of the variant allele (G/A genotype) were prospectively compared jects showed more deep slow wave sleep during NREM sleep, to homozygous G allele carriers. Self-rated sleep duration as esti- more alpha activity in REM sleep, and, under sleep deprivation, mated with the Munich Chronotype Questionnaire did not differ they experienced higher sleep pressure. Furthermore, their cogni- between ADA genotypes. By contrast, counting tendency on a tive performance on tasks that engage more frontal cortical areas, random number generation (RNG) task was higher in women with such as executive function tasks, was significantly deteriorated G/A (n=14) than G/G (n=87) genotype. No significant effect was in comparison to PER3-4/4 participants, particularly during the observed in men. However, adjacency on the RNG and overall biological night. Analyses of wake and sleep ECG data revealed performance on a d2-attention task were worse in 28 individuals a loss of parasympathetic control on autonomic balance in PER3- (13 men, 15 women) with G/A genotype compared to individually 5/5 in comparison to PER3-4/4 subjects. Conclusions: While age- and sex-matched subjects with G/G genotype. Heterozygous the clock gene PER3 does not appear to have a substantial effect carriers of the A allele (n=10) showed higher subjective sleepiness on the circadian phase; it appears to affect several aspects of sleep (Stanford Sleepiness Scale) after one night without sleep, and per- homeostasis. The main implication is that sleep homeostasis can formed worse than matched individuals with G/G genotype on the be affected by clock genes and that these effects may account for psychomotor vigilance task throughout prolonged wakefulness. the inter-individual differences in the susceptibility to sleep loss Our findings demonstrate that a functional polymorphism of ADA and circadian phase misalignment, which can happen during jet predicts waking functions in healthy adults. They further support the notion that adenosinergic mechanisms play an important role in sleep homeostasis. Together with our previous sleep and sleep A. Bader1,2, A. Riecher-Rössler2, U. Frisch2, A. Wirz-Justice1
J. Vienne1, M. Gassmann2, B. Bettler2, M. Tafti1 1Center for Chronobiology, Psychiatric University Clinics 1Center for Integrative Genomics (CIG), University of Basel, 2Psychiatric Outpatient Department, University Lausanne, Lausanne, 2Pharmazentrum, Department of Biomedicine, Institute of Physiology, University of Basel, Switzerland Objective: Affective disorder during pregnancy is a com- mon condition: one in ten pregnant women suffers from Major Introduction: Gamma-hydroxybutyrate (GHB) and its pre- Depressive Disorder (MDD). One complaint in pregnancy is sleep cursor gamma-butyrolactone (GBL) have wide-range effects on disturbance, and this is also a characteristic of MDD. However, vigilance and behavior. However, the mode of action of GHB the literature about sleep disturbances in depressed pregnant remains unclear. Evidence indicates that GHB acts mainly through women is sparse. This ongoing study compared subjective sleep in GABAB receptors that consist of two subunits; GABAB(1) and pregnant depressed and non-depressed women. Methods: Self- -B(2). GABAB(1) has two isoforms; 1a and 1b. To elucidate the reported sleep measures were collected by 35 pregnant women mechanisms by which GHB affects vigilance state and the role with a DSM-IV diagnosis of MDD and 7 non-depressed preg-of GABAB receptor subunits, the electroencephalogram (EEG) nant women (CON). All women kept a diary for one week where of different GABAB receptor subunit knock-out (KO) mice was they assessed daily, before bedtime, their feelings (relaxed-tense, studied before and after GBL administration. Methods: Adult somatically unwell-well, alert-sleepy, sated-hungry) and mood male wild-type (WT) BALB/c, GABAB(1a) -/-, GABAB(1b) (worst-best) on a bipolar scale (VAS) and noted the next morning -/-, GABAB(1) -/- and GABAB(2) -/- mice were implanted with their sleep onset time, wake-up time, and nighttime awakenings EEG and EMG electrodes. Mice were recorded during 24hr base- (number, duration). Results: The two groups did not differ in age line followed by daily injections with saline and 3 doses of GBL (M=32.4) and pregnancy week (PW) (M=20.6). Over all trimes- (50, 100 and 150 mg/kg i.p.). Results: Under baseline condi- ters, they showed significant differences in mood (p<.001), feel- tions, GABAB(1) -/-, GABAB(2) -/- and GABAB(1a) -/- mice ing tense (p<.05), and duration of nighttime awakenings (p<.05). exhibited spontaneous seizures. Sleep amount, sleep quality and Compared with CON, MDD showed in the second trimester the EEG were differentially affected in the different KO mice. worse mood (p<.05), and in the third trimester also lower mood GBL induced EEG slow waves (1-4Hz) with a spike-like EEG (p<.05), longer duration of nighttime awakenings (p<.05), and a pattern at high doses in GABAB(1b) -/-, GABAB(1a) -/- and trend for longer sleep onset latency (p<.052). Within the groups, WT mice, but no behavioral or EEG changes were observed in no significant difference between the second and third trimester mice without any functional GABAB receptors (GABAB(1) -/- was found. For all pregnant women, better mood correlated with and - B(2) -/- mice). The duration of the ‘GBL-induced state’ less tension (r = -.466, p<.05), better somatic wellbeing (r=.737, and the prevalence of EEG slow-waves (delta power) during this p<.001), fewer (r = -.371, p<.05) and shorter nighttime awaken- state increased with dose in each genotype. GABAB(1a) -/- mice ings (r = -.392, p<.05), and shorter sleep onset latency (r=-.367, showed 2-fold higher values of delta power than the 2 other gen- p<.05), when PW was controlled. For PW, no significant correla- otypes. Although GBL suppressed sleep dose-dependently after tion was found. There were no significant correlations between injection and induced a dramatic but non-physiological high delta the trimesters within the groups. Conclusion: Pregnant women activity, none of the 3 genotypes clearly compensated this sleep with MDD had a significantly lower mood, higher tension, and loss and/or modified the kinetics of delta power. Conclusion: more sleep disturbances than CON. Greater depressed mood in Our results demonstrate that GABAB(1a) plays an important role pregnancy was correlated with poorer somatic wellbeing, higher in epilepsy, sleep and thalamcortical synchronization. GBL acts tension, more and longer nighttime awakenings, and prolonged only through GABAB receptors to induce behavioral and EEG sleep onset latency. There was no increase in sleep disturbances in effects. The lack of a compensatory response to the sleep time lost the third compared with the second trimester.
due to GBL indicates that sleep need does not change during the GBL-induced state. Moreover, the slow waves that accompany this state do not seem to lower subsequent sleep need. Thus, the homeostatic regulation of sleep seems temporarily suspended dur-ing the GBL-induced state and the delta activity induced by GBL might functionally be different from delta activity expressed dur-ing ‘physiological’ sleep.
H. Burkhalter1, S. Sereika2, S. Engberg2, A. Wirz-Justice3, Child Development Center, University Children’s Hospital J. Steiger4, Sabina De Geest1 1Institute of Nursing Science, University of Basel, 2Institute of Nursing science, University of Pittsburgh, Pa., Introduction: Sleewake patterns differ substantially among USA, 3Center for Chronobiology, Psychiatric University individuals. These inter-individual differences root in the geneti- Clinics Basel, 4Division of Transplant Immunology and cally programmed circadian clock which determines the so called Nephrology, University Hospital Basel, Switzerland “chronotypes” (Aeschbach et al. 2003; Roenneberg et al. 2002). Up to now, there is very little known about the effects of chronotype Aim: Poor sleep quality (SQ) has been associated with poor differences within family members. How does parent-child adjust- clinical outcome in a number of chronically ill patient popula- ment occur when morning-oriented parents have a night-type child tions, but is not yet well studied in transplant patients. Assessing or the other way round? The fit between parents and children’s SQ in large cohort study such as the Swiss Transplant Cohort chronotypes is an important factor in understanding and predict- Study is therefore very relevant. The aim of this study was to ing conflicts in family relationships (Barber, 1994). We aimed to asses the validity of 2 items: one to assess SQ and one daytime assess chronotypes and sleewake patterns in children and parents functioning (DF) item, using the Pittsburgh Sleep Quality Index by subjective and objective methods, and to focus on the effects (PSQI) as gold standard. We used the validation criteria proposed of differences between children and parents on parental distress. by the American Psychological Association (APA). Methods: Methods: Chronotypes of healthy kindergarten children (age 5-7 We used a cross-sectional descriptive design, inclu ding a con- years) were assessed by the Children’s Chronotype Questionnaire venience sample of all adult renal transplants transplanted at the

Source: http://www.swiss-sleep.ch/dokumente/Abstracts%20of%20the%20SSSSC.pdf