Chemosensory event-related potentials during sleep—A pilot study

Chemosensory event-related potentials during sleep—A pilot study

Neuroscience Letters 406 (2006) 222–226 Chemosensory event-related potentials during sleep—A pilot study Boris A. Stuck a , Heike Weitz b , Karl H¨or...

266KB Sizes 0 Downloads 47 Views

Neuroscience Letters 406 (2006) 222–226

Chemosensory event-related potentials during sleep—A pilot study Boris A. Stuck a , Heike Weitz b , Karl H¨ormann a , Joachim T. Maurer a , Thomas Hummel b,∗ a

b

Department of Otorhinolaryngology, Head and Neck Surgery, University Hospital Mannheim, Germany Smell & Taste Clinic, Department of Otorhinolaryngology, University of Dresden Medical School (“Technische Universit¨at Dresden”), Fetscherstrasse 74, 01307 Dresden, Germany Received 1 July 2006; received in revised form 26 July 2006; accepted 26 July 2006

Abstract Aim of the present pilot study was to investigate whether cortically generated chemosensory event-related potentials (ERPs) can be recorded during sleep. Chemosensory function during sleep was assessed in 14 healthy female volunteers. An overnight polysomnography was performed to assess nocturnal sleep and to classify sleep stages. Chemosensory ERPs were recorded using air-dilution olfactometry. H2 S (4 ppm) was used for olfactory and CO2 (40%, v/v) for trigeminal stimulation. Chemosensory ERPs could be recorded during sleep for both olfactory and trigeminal stimuli in some but not all subjects. Compared to baseline, latencies of olfactory ERPs were longer and amplitudes were larger during light sleep and slow wave sleep (SWS). For trigeminal stimulation N1 latencies were longest during REM sleep. These results indicate that both trigeminal and olfactory ERPs can be recorded during sleep suggesting that chemosensory stimuli are processed on a cortical level during sleep. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Chemosensation; Event related potentials; Olfaction; Trigeminal

Sensory activation during sleep has been studied extensively for visual, auditory, and somatosensory stimuli. In terms of electrophysiological measures, it was shown that the late components of auditory event-related potentials (ERPs) are significantly altered during sleep, while brainstem auditory evoked potentials were relatively unaffected [3]. During both slow wave sleep (SWS) and REM-sleep responses exhibited longer latencies while reports for amplitudes were equivocal [7,18,22,23]. However, more recent research suggests that amplitude N1 decreases while P2 actually increases [4,6]. No such data exists for olfactory ERPs. Research suggests that olfactory stimuli are processed during sleep as awakening reactions occur [2,5]. In this context, however, it is not exactly clear whether these responses are due to activation of the trigeminal system which is also stimulated by most odors [8]. Further, olfactory information is processed differently than other sensory information [10]. Aims of the present pilot study were to investigate chemosensory ERPs during sleep to assess (1) whether chemosensory ERPs can be recorded during sleep, also with regard to arousal reactions/awakenings and a good signal-to-noise ratio, and, if so, (2) whether sleep does



Corresponding author. Tel.: +49 351 458 4189; fax: +49 351 458 4326. E-mail address: [email protected] (T. Hummel).

0304-3940/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2006.07.068

alter chemosensory ERPs in the same way as it alters ERPs in other sensory modalities. The study was conducted at the Department of Otorhinolaryngolgoy, Head and Neck Surgery Mannheim. The study protocol was approved by the local ethics board of the Faculty of Clinical Medicine Mannheim of the University of Heidelberg; written informed consent was obtained from all participating subjects. Participants: 15 young healthy female volunteers were included in this prospective study (mean age 24.6 ± 1.9 years, range: 21–29 years). Exclusion criteria were present/previous history of smell or taste disorders, use of any medication known to affect chemosensory function [1], and a history of sleep disorders. At the screening visit, relevant nasal pathology such as mucosal inflammation, significant septal deviation, or nasal polyposis were ruled out with a clinical examination including nasal endoscopy. Patency of the nasal airways was also ascertained using active anterior rhinomanometry (Rhinomanometer 300, ATMOS Medizintechnik GmbH & Co.KG, Lenzkirch, Germany). Psychophysical testing of olfactory function: All of the participating subjects underwent olfactory testing using the “Sniffin’ Sticks” test kit to establish normal olfactory function [12,14]. Testing involved assessment of butanol odor thresholds, odor discrimination, and odor identification.

B.A. Stuck et al. / Neuroscience Letters 406 (2006) 222–226

223

Fig. 1. Conditions for overnight recording. (A) Control units and the olfactometer can be seen in the foreground. The subject is lying on the bed behind the screen. (B) Sleeping subject with Teflon tubing (arrow) securely taped to the left nostril. The tube is connected to the olfactometer outlet.

Sleep recordings: Analogous to routine sleep studies an overnight polysomnography was performed to assess nocturnal sleep and to classify sleep stages. Monitoring included two electroencephalograms, two electrooculograms, two submental and two leg electromyograms. Sleep stages were scored according to Rechtschaffen and Kales [20]. Chemosensory ERPs: For chemosensory stimulation a dynamic olfactometer based on air-dilution olfactometry was used (OM6b; Burghart instruments, Wedel, Germany) (Fig. 1a). This allows the presentation of odorous stimuli within a continuous airstream of 8 L/min that does not alter the mechanical or thermal conditions at the nasal mucosa [13]. Moreover, this constant airstream ensures that the influence of breathing patterns on stimulus presentation is minimized. For specific olfactory stimulation H2 S was used (200 ms stimuli of 4 ppm); to specifically evaluate trigeminal function CO2 was administered (200 ms stimuli) [11]. A concentration of 40% (v/v) was selected for CO2 as higher concentrations produce an intense nociceptive stimulus with the risk of frequent awakenings. With regard to the rapid adaptation of the chemosensory system to repeated stimuli, the interstimulus interval was set to 30 s. Stimuli were applied to the left or right nostril with regard to nasal patency. To allow some mobility during sleep, a tube of approximately 60 cm length was used to connect the subjects nostril with the olfactometer (Fig. 1b). To avoid mechanical alteration of the nasal mucosa by the tube, the tube was imbedded in a soft cylindrical foam which was placed into the nostril to secure its position during measurements. A curtain separated the subjects’ bed from the olfactometer and the investigator. Ear plugs were administered to dampen external sounds. Chemosensory ERPs were recorded at six positions according to the international 10-20 system (Cz, Pz, C3, C4, P3, and P4; referenced to linked earlobes) using an 8-channel amplifier (S.I.R., R¨ottenbach, Germany). For averaging a minimum of 8 and a maximum of 20 stimuli was used. Vertical eye movements were monitored at the Fp2 lead. The sampling frequency was 250 Hz; the pre-trigger period was 500 ms with a recording time of 2048 ms (band pass 0.02–15 Hz). Eye blink-contaminated recordings during wakefulness were discarded. Mean peak-topeak amplitudes (P1N1, N1P2) and peak latencies (N1, and P2) were measured (software BOMPE 4.1; Kobal, Erlangen, Germany). Chemosensory ERPs were assessed during wakefulness (baseline) at the beginning of the overnight recordings. Then the

hypnogram was used to detect when the subjects were falling asleep. As soon as they reached a stable sleep stage, chemosensory testing started. As long as no arousal was detected, ERPs were recorded in response to a maximum of 20 H2 S-stimuli each followed by recordings in response to 20 CO2 -stimuli each during light sleep (sleep stages 1 and 2), SWS (sleep stages 3 and 4), and REM sleep. If a change in sleep stage during the measurement was detected, the measurement was continued. In those cases where an arousal or an awakening occurred, the measurement was terminated and restarted as soon the subject reached a stable sleep stage again. When measurements were started or interrupted, marks were set in the hypnogram to be able to re-evaluate and potentially re-classify sleep stages. Thus, measurements were performed throughout the entire night, until 20 stimuli of H2 S and CO2 had been presented during each sleep stage. Accordingly, measurements were often scattered over the entire night. Presence of chemosensory ERPs at the different sleep stages was heuristically analysed [11], based on (1) the characteristic topographical distribution of response amplitudes to olfactory or trigeminal stimuli, (2) appearance of ERPs within windows of 200–600 ms for the N1 peak, and 400–900 ms for the P2 peak at position Cz, and (3) the characteristic ERPs shape with an inter-peak interval of 200–400 ms. Statistics: Peak latencies and peak-to-peak amplitudes of ERPs at the different sleep stages were compared to those at baseline using t-tests for paired samples. In order not to inflate the alpha level and due to the fact that EEG recordings in this series were technically most stable at Cz analyses were only performed for recordings at this position. The SPSS program package Version 12.0 was used for statistical analyses (SPSS Inc., Chicago, IL, USA). Fifteen subjects completed the protocol. One subject did not fall asleep during the entire night so no recordings during sleep could be obtained. Chemosensory ERPs could be obtained at baseline in 14 out of 15 subjects; in 1 of the 15 subjects interpretation of recordings was not possible due to blink artifacts. Table 1 provides an overview about the number of recordings in individual subjects. Presence of ERPs: Recordings of chemosensory ERPs were possible in all sleep stages for both olfactory and trigeminal stimulation. Nevertheless, ERPs were not consistently present at all sleep stages. Olfactory ERPs could be detected during light sleep in six subjects, during SWS in eight subjects,

224

B.A. Stuck et al. / Neuroscience Letters 406 (2006) 222–226

Table 1 Number of successful recordings at recording position Cz, separately for olfactory and trigeminal stimulation, with H2 S and CO2 , respectively

Each line represents the recordings obtained in an individual subject with grey areas denoting successful recordings (also indicated by “1”), and empty boxes unsuccessful recordings.

and in one subject only during REM sleep. Trigeminal ERPs could be detected during light sleep in two subjects, during SWS in three subjects, and in four subjects during REM sleep. Other difficulties in recording ERPs during sleep related to arousals induced by chemosensory stimulation. In total, 41 sets of measurements (20 H2 S-stimuli followed by 20 CO2 -stimuli) were performed during sleep in the 14 subjects. In 15 of those recordings no arousals were detected. During the remaining 26 recordings arousals occurred which lead to termination of the measurements. Olfactory ERPs: Overall, differences were only found between baseline measures and the various sleep stages, but not between sleep stages. Specifically, latencies N1 of olfactory ERPs (Table 2) were longer during light sleep and SWS compared to baseline (t > 3.45, p < 0.02), with olfactory ERPs

recorded during light sleep having the longest latencies (Fig. 2). Latencies P2 were found to be significantly longer during SWS (t = 5.41, p = 0.001). Compared to baseline amplitudes P1N1 were larger during light sleep and SWS sleep (t > 2.80,

Table 2 Descriptive statistics of latencies of N1 and P2 (in ms) at recording position Cz, separately for olfactory and trigeminal stimulation, with H2 S and CO2 , respectively Stimulant H2 S N1 P2 Stimulant CO2 N1 P2

Baseline n = 14

Light sleep n=6

SWS n=8

REM n=1

397 ± 55 547 ± 53

543 ± 52 719 ± 126

494 ± 83 660 ± 26

– –

Baseline n = 14

Light sleep n=2

SWS n=3

REM n=3

373 ± 33 525 ± 54

– –

379 ± 87 596 ± 111

509 ± 74 676 ± 49

Mean values ± S.D.; n: number of subjects with successful recordings (averages are only presented when more than two recordings have been made).

Fig. 2. Examples of olfactory ERPs recorded in single subjects during wakening, REM-sleep, light sleep, and slow wave sleep. Recordings started 500 ms prior to stimulus presentation (indicated by the vertical dotted line). The peaks N1 and P2 are indicated by short vertical lines. Recordings containing ERPs are plotted in bold lines, recordings during the three sleep stages where no ERP had been detected are plotted in thin lines.

B.A. Stuck et al. / Neuroscience Letters 406 (2006) 222–226 Table 3 Descriptive statistics of amplitudes of P1N1 and N1P2 (in ␮V) at recording position Cz, separately for olfactory and trigeminal stimulation, with H2 S and CO2 , respectively Stimulant H2 S P1N1 N1P2 Stimulant CO2 P1N1 N1P2

Baseline n = 14

light sleep n=6

SWS n=8

REM n=1

8.3 ± 5.0 14.1 ± 4.8

8.3 ± 2.1 13.6 ± 1.6

13.1 ± 6.4 26.4 ± 8.5

– –

Baseline n = 14

Light sleep n=2

SWS n=3

REM n=3

14.0 ± 9.8 27.7 ± 13.4

– –

10.9 ± 4.3 21.3 ± 8.9

10.6 ± 5.2 21.4 ± 7.7

Mean values ± S.D.; n: number of subjects with successful recordings (averages are only presented when more than two recordings have been made).

p < 0.039). Similarly, amplitudes N1P2 were larger during SWS (t = 3.91, p = 0.006) (Table 3). Trigeminal ERPs: As with peak latencies of olfactory ERPs, in comparison to baseline N1 latencies of trigeminal ERPs were, on average, longest during light sleep. However, these differences were not significantly different for light sleep and SWS; only for REM sleep recordings the longer N1 latencies differed (t = 4.69, p = 0.043) from baseline (Table 3). Average responses to CO2 obtained at SWS and REM sleep were smaller compared to baseline conditions. However, these differences did not reach a level of statistical significance (t < 1.52, p > 0.26). Descriptive statistics are presented in Table 3. Results from this pilot study indicate that both trigeminal and olfactory ERPs can be recorded during sleep. This suggests that chemosensory stimuli are actively processed during sleep. Importantly, recordings during baseline compared to previous research [11,21] indicating that the specific study conditions did not produce an environment that would significantly alter responses to chemosensory stimuli. Although it has to be kept in mind that the number of successful recordings was limited, chemosensory ERPs appeared to exhibit significant changes during sleep when compared to baseline. For both trigeminal and olfactory stimulation latencies for N1 and P2 increased during sleep with the longest latencies obtained during light sleep. Why that should be so remains unclear. Despite the very small sample size it may be speculated, however, that under some conditions more cortical resources were available for the processing of the olfactory stimuli resulting in prolonged latencies of the ERPs [9,19]. In this context, it should also be emphasized that circadian changes of chemosensory function might play a role when comparing recordings at baseline and recordings obtained during the night [17]. Having said that, the differences in latency indicate that the processing of olfactory information is significantly different during sleep compared to baseline which has also been demonstrated for other sensory modalities, e.g., for audition [7,18,22,23]. This observation is also consistent with previous research indicating that the cognitive processing of chemosensory information is dependent on the individual’s state [5]. More specifically, the present data appear to confirm recent

225

observations by Marukami et al. [15] suggesting that processing in the olfactory cortex, but much less so in the olfactory bulb, is gated through activation in the ascending reticular activating system (ARAS), probably in synchrony with other sensory systems. While this gating is due to the thalamic relay of most sensory systems the ARAS seems to take the place of the thalamus in case of the olfactory system which has no major relay in the thalamus [10,16]. The small number of successful recordings during sleep also seems to be an expression of the gating of the olfactory information processing. Other than in previous research [5] the present study did not reveal major differences between arousal reactions to trigeminal or olfactory stimulation which may be due to the relatively small number of observations. In fact, the number of arousal reactions was identical when evoked from slow wave sleep, light sleep, and REM sleep. Further research is currently under way to answer the question whether trigeminal or olfactory stimuli are more likely to wake subjects during sleeping. Taken together, the present data suggest that chemosensory stimulation elicits cortical activation during sleep. It can be speculated that the lower degree of activation during sleep in combination with the prolongation of response latencies may be due to state-dependent gating mechanisms. Results from this pilot data will be used to direct a future experiment with an improved design. References [1] B.H. Ackerman, N. Kasbekar, Disturbances of taste and smell induced by drugs, Pharmacotherapy 17 (1997) 482–496. [2] P. Badia, N. Wesenstein, W. Lammers, J. Culpepper, J. Harsh, Responsiveness to olfactory stimuli presented in sleep, Physiol. Behav. 48 (1990) 87–90. [3] K.B. Campbell, E.A. Bartoli, Human auditory evoked potentials during natural sleep: the early components, Electroencephalogr. Clin. Neurophysiol. 65 (1986) 142–149. [4] K.B. Campbell, I.M. Colrain, Event-related potential measures of the inhibition of information processing: II. The sleep onset period, Int. J. Psychophysiol. 46 (2002) 197–214. [5] M.A. Carskadon, R.S. Herz, Minimal olfactory perception during sleep: why odor alarms will not work for humans, Sleep 27 (2004) 402– 405. [6] K.A. Cote, L. Etienne, K.B. Campbell, Neurophysiological evidence for the detection of external stimuli during sleep, Sleep 24 (2001) 791–803. [7] H. Davies, N. Yoshie, Human evoked cortical responses to auditory stimuli, Physiologist 6 (1963) 164. [8] R.L. Doty, W.P.E. Brugger, P.C. Jurs, M.A. Orndorff, P.J. Snyder, L.D. Lowry, Intranasal trigeminal stimulation from odorous volatiles: psychometric responses from anosmic and normal humans, Physiol. Behav. 20 (1978) 175–185. [9] M.W. Geisler, C.D. Morgan, J.W. Covington, C. Murphy, Neuropsychological performance and cognitive olfactory event-related brain potentials in young and elderly adults, J. Clin. Exp. Neuropsychol. 21 (1999) 108–126. [10] J.A. Gottfried, Smell: central nervous processing, Adv. Otorhinolaryngol. 63 (2006) 44–69. [11] T. Hummel, G. Kobal, Olfactory event-related potentials, in: S.A. Simon, M.A.L. Nicolelis (Eds.), Methods and Frontiers in Chemosensory Research, CRC press, Boca Raton, Florida, USA, 2001, pp. 429–464. [12] T. Hummel, B. Sekinger, S. Wolf, E. Pauli, G. Kobal, Sniffin’ Sticks”: Olfactory performance assessed by the combined testing of odor identification, odor discrimination and olfactory threshold, Chem. Senses 22 (1997) 39–52.

226

B.A. Stuck et al. / Neuroscience Letters 406 (2006) 222–226

[13] G. Kobal, Elektrophysiologische Untersuchungen des menschlichen Geruchssinns, Thieme Verlag, Stuttgart, 1981. [14] G. Kobal, L. Klimek, M. Wolfensberger, H. Gudziol, A. Temmel, C.M. Owen, H. Seeber, E. Pauli, T. Hummel, Multicenter investigation of 1,036 subjects using a standardized method for the assessment of olfactory function combining tests of odor identification, odor discrimination, and olfactory thresholds, Eur. Arch. Otorhinolaryngol. 257 (2000) 205–211. [15] M. Murakami, H. Kashiwadani, Y. Kirino, K. Mori, State-dependent sensory gating in olfactory cortex, Neuron 46 (2005) 285–296. [16] K.R. Neville, L.B. Haberly, Olfactory cortex, in: G.M. Shepherd (Ed.), The synaptic organization of the brain, Oxford University Press, New York, 2004, pp. 415–454. [17] S. Nordin, J. Lotsch, C. Murphy, T. Hummel, G. Kobal, Circadian rhythm and desensitization in chemosensory event-related potentials in response to odorous and painful stimuli, Psychophysiology 40 (2003) 612–619. [18] E.M. Ornitz, L.M. Panman, R.D. Walter, The variabilitiy of the auditory averaged evoked responses during sleep and dreaming in children and adults, Electroenceph. Clin. Neurophysiol. 22 (1967) 514–524.

[19] B.M. Pause, B. Sojka, K. Krauel, R. Ferstl, The nature of the late positive complex within the olfactory event-related potential, Psychophysiology 33 (1996) 168–172. [20] A. Rechtschaffen, A. Kales, A Manual of Standardized Technology, Techniques and Scoring System for Sleep Stages of Human Sleep, Brain Information Service/Brain Research Institute, UCLA, Los Angeles, CA, USA, 1968. [21] B.A. Stuck, S. Frey, C. Freiburg, K. Hormann, T. Zahnert, T. Hummel, Chemosensory event-related potentials in relation to side of stimulation, age, sex, and stimulus concentration, Clin. Neurophysiol. 117 (2006) 1367–1375. [22] E.D. Weitzman, H. Kremen, Auditory evoked responses during different stages of sleep in man, Electroenceph. Clin. Neurophysiol. 18 (1965) 65–70. [23] H.L. Williams, D.I. Tepas, H.C. Morloch, Evoked responses to clicks and electroencephalographic stages of sleep in man, Science 138 (1963) 685–686.