To wake or not to wake? The two-sided nature of the human K-complex
Highlights
► Human in vivo findings support animal studies: thalamic involvement in KC mediation. ► KC reflects subcortical arousal and low level information processing. ► Arousal process is limited by disengagement of saliency network (anterior insula). ► Sleep preserving counteraction is promoted by cerebral midline structures.
Introduction
Sleep describes a period of reduced consciousness. The ability of the organism to react to stimuli is preserved making sleep a reversible physiological state in contrast to states of pathologically reduced responsiveness such as coma or the vegetative state (Cirelli and Tononi, 2008). Sleep is subdivided into stages with different arousal thresholds (drowsiness, light sleep, deep sleep) and categorized based on features of the electroencephalogram (EEG) changing from high-frequency, low-voltage oscillations during wakefulness over higher voltage, slower waves representing light non-rapid eye movement (NREM) sleep to high-voltage, slow wave activity during deep sleep (Saper et al., 2010).
The K-complex (KC) is one major graphoelement of the adult non-REM sleep EEG. It characterizes sleep stage 2, an intermediate epoch between drowsiness and deep sleep. Loomis and colleagues first described that this “K wave or complex appears […] either spontaneously or as the result of stimulation” (Loomis et al., 1938). Neurophysiologically, the KC indeed represents a multimodal fronto-central evoked response with a longer refractory period and different late components (N350, N550 and P900) than primary sensory evoked potentials (Colrain, 2005, Colrain et al., 2000, Gora et al., 1999, Halasz, 2005).
Functionally, it is under debate whether the KC plays a sleep protective role (Bastien et al., 2000, Cash et al., 2009, Colrain, 2005, Nicholas et al., 2002, Wauquier et al., 1995) promoting deeper sleep, or whether it is correlated with a cortical arousal (Ehrhart et al., 1981, Halasz, 1993). Accordingly, the KC has been ascribed a janus-faced role (Bastien et al., 2000, Halasz, 2004, Halasz, 2005, Terzano and Parrino, 2000, Terzano et al., 2005). Sleep stage transitions are thought to happen relatively rapidly when the system passes a critical state (Beenhakker and Huguenard, 2009), and it is conceivable that the KC marks instances when the system approaches but eventually does not cross the critical line.
For the understanding of the KC, we will distinguish a subcortical from a cortical arousal level. i) At a level confined to the brainstem, arousal can consist of only a change of autonomic functions. As a consequence of especially mesencephalic brain stem activity the thalamic relay neuron membrane potential can be influenced and thalamic signal transmission thereby reinstalled (Evans, 2003). ii) Additionally, an unspecific broad cerebral excitation can be brought about by pontomesencephalic nuclei via the intrathalamic and midline thalamic nuclei. Cortically, primary and maybe secondary sensory areas might resume information processing which the KC has been related to (Bastuji and Garcia-Larrea, 1999, Cash et al., 2009, Halasz, 2005, Niiyama et al., 1995). The resumption of information processing in primary sensory areas should also include a change in thalamic activity from the oscillatory burst mode to the relay mode allowing for accurate signal transmission (Evans, 2003). iii) At a higher level of processing external or self-awareness might be reinstalled. In this context, apart from the brainstem, thalamus and sensory areas, the lateral frontal–parietal regions subserving external awareness as well as the default mode network (DMN), related to self-awareness, are of particular interest (Vanhaudenhuyse et al., 2011, Boly et al., 2008, Larson-Prior et al., 2009, Raichle et al., 2001).
The thalamus is centrally involved in arousal mechanisms and sensory signal transduction, and hence a thalamic association with the KC can be expected. Furthermore, electrophysiological data support the assumption that the thalamus plays a crucial role in the mediation of graphoelements (sleep spindles, delta waves) following the KC (Amzica and Steriade, 2002, Nicholas et al., 2006, Steriade and Amzica, 1998).
We used electroencephalography (EEG)-correlated functional magnetic resonance imaging (fMRI) to elucidate the function of the KC by relating it to brain areas exhibiting associated activity changes and examining the effective connectivity between these regions. Specifically, we hypothesized that the KC reflects a subcortical arousal with a limited cortical arousal in the form of low level cortical information processing. Therefore, we expected to find KC-associated positive blood oxygen level-dependent (BOLD) signal changes in the brainstem, the thalamus and primary sensory areas, especially the primary auditory cortex, and an associated connectivity precluding higher level processing.
Section snippets
Subjects
37 (age range 20–31, mean age 24 years) of 149 healthy, non-sleep-deprived subjects who participated in the study (written informed consent, approval by the local ethics committee) were analyzed. A sub-selection of subjects was necessary to ensure valid fMRI designs at the single subject level as well as a balanced design at the second level (group analysis). Subjects were included if they met a priori selection criteria which provided a sufficient number of data points (EPI volumes and KCs) for
EEG
A total of 37 subjects were included in the analysis, the number of N2 segments, stages and K-complexes included into further analysis alongside two example hypnograms are displayed in Fig. 1.
fMRI group analysis
The second level random effects group analysis revealed KC-correlated BOLD cerebral gray matter signal changes in bilateral primary and associative auditory as well as visual cortices, para- and postcentral (Area 4 d and 6) gyrus including the supplementary motor area, the fusiform gyrus, the precentral,
Discussion
In 37 healthy adults studied with EEG/fMRI during stage 2 sleep we found KC-associated positive BOLD signal changes in subcortical (brainstem, thalamus), cerebellar, sensory and motor (auditory, and visual, sensorimotor cortices), midline (anterior and midcingulate gyrus, precuneus) and regions which form part of the default mode network (prefrontal cortex, inferior parietal lobule, precuneus). Negative BOLD signal changes were detected in the anterior insula bilaterally. The effective
Conclusion
Our human in vivo multi-modal data corroborate previous animal studies demonstrating a thalamic involvement in the mediation of the KC and provide evidence for a two-sided nature of the evoked KC: (1) association with a subcortical arousal with low level information processing limited by disengagement of saliency network regions; and (2) a sleep preserving counteraction promoted by cerebral midline structures. The KC may hence reconcile two vital though seemingly contradictory processes:
Acknowledgments
This work was funded by the Bundesministerium für Bildung und Forschung (grant 01 EV 0703) and LOEWE Neuronale Koordination Forschungsschwerpunkt Frankfurt (NeFF). We thank Sandra Anti, Ralf Deichmann and Steffen Volz for extensive MRI support, Christian Kell for his important input regarding the saliency network, David E. Linden for a very helpful discussion on the P300, our reviewers for their intellectual input, Torben E. Lund for providing a matlab implementation of the RETROICOR method,
References (70)
- et al.
Identification of EEG events in the MR scanner: the problem of pulse artifact and a method for its subtraction
Neuroimage
(1998) - et al.
The functional significance of K-complexes
Sleep Med Rev
(2002) - et al.
Evoked potentials as a tool for the investigation of human sleep
Sleep Med Rev
(1999) - et al.
Neurons that fire together also conspire together: is normal sleep circuitry hijacked to generate epilepsy?
Neuron
(2009) - et al.
The use of evoked potentials in sleep research
Sleep Med Rev
(2007) Sleep, consciousness and the spontaneous and evoked electrical activity of the brain. Is there a cortical integrating mechanism?
Neurophysiol Clin
(2003)- et al.
Dynamic causal modelling
Neuroimage
(2003) - et al.
fMRI at 7 T: whole-brain coverage and signal advantages even infratentorially?
Neuroimage
(2007) Arousals without awakening — dynamic aspect of sleep
Physiol Behav
(1993)The microstructure of sleep
Suppl Clin Neurophysiol
(2004)
K-complex, a reactive EEG graphoelement of NREM sleep: an old chap in a new garment
Sleep Med Rev
Top–down control–signal dynamics in anterior cingulate and prefrontal cortex neurons following task switching
Neuron
Recent advances in recording electrophysiological data simultaneously with magnetic resonance imaging
Neuroimage
The impact of slow wave sleep proximity on evoked K-complex generation
Neurosci Lett
K-complex evoked in NREM sleep is accompanied by a slow negative potential related to cognitive process
Electroencephalogr Clin Neurophysiol
Electrophysiological evidence suggesting that sensory stimuli of unknown origin induce spontaneous K-complexes
Electroencephalogr Clin Neurophysiol
Comparing dynamic causal models
Neuroimage
Auditory processing across the sleep–wake cycle: simultaneous EEG and fMRI monitoring in humans
Neuron
Sleep state switching
Neuron
Bayesian model selection for group studies
Neuroimage
Origin and significance of the Cyclic Alternating Pattern (CAP). Review article
Sleep Med Rev
Intracranial cortical localization of the human K-complex
Clin Neurophysiol
The AASM manual for the scoring of sleep and associated events-rules
Terminology and Technical Specifications
The K-complex: its slow (< 1-Hz) rhythmicity and relation to delta waves
Neurology
EEG-fMRI of focal epileptic spikes: analysis with multiple haemodynamic functions and comparison with gadolinium-enhanced MR angiograms
Hum Brain Mapp
EEG characteristics prior to and following the evoked K-Complex
Can J Exp Psychol
Intrinsic brain activity in altered states of consciousness: how conscious is the default mode of brain function?
Ann N Y Acad Sci
Breakdown of within- and between-network resting state functional magnetic resonance imaging connectivity during propofol-induced loss of consciousness
Anesthesiology
The brain's default network: anatomy, function, and relevance to disease
Annals of the New York Academy of Sciences
Neuronal oscillations in cortical networks
Science
Event-related potentials as a measure of sleep disturbance: a tutorial review
Noise & Health
The human K-complex represents an isolated cortical down-state
Science
Resonse to comment on “The human K-complex represents an isolated cortical down-state”
Science
The precuneus: a review of its functional anatomy and behavioural correlates
Brain
Is sleep essential?
PLoS Biol
Cited by (94)
Non-REM sleep electrophysiology in REM sleep behaviour disorder: A narrative mini-review
2022, Neuroscience and Biobehavioral ReviewsAn energy screening and morphology characterization-based hybrid expert scheme for automatic identification of micro-sleep event K-complex
2021, Computer Methods and Programs in BiomedicineSignature of consciousness in brain-wide synchronization patterns of monkey and human fMRI signals
2021, NeuroImageCitation Excerpt :We simultaneously recorded EEG via a cap (modified BrainCapMR, Easycap) and fMRI (1505 vol of T2*-weighted echo planar images, TR/TE = 2080 ms/30 ms, matrix 64 × 64, voxel size 3 × 3 × 2 mm3, distance factor 50%; FOV 192 mm2) with a 3T Siemens Trio scanner. Polysomnography was performed, recording chin and tibial EMG, ECG, EOG bipolarly (sampling rate 5 kHz, low pass filter 1 kHz) and EEG with 30 channels using FCz as the reference (sampling rate 5 kHz, low pass filter 250 Hz); We recorded pulse oxymetry and respiration with sensors from the Trio (sampling rate 50 Hz) and MR scanner compatible devices (BrainAmpMR+, BrainAmpExG; Brain Products), which aided sleep scoring during fMRI recordings (Jahnke et al., 2012). MRI and pulse artifacts were removed with the average artifact subtraction (AAS) method (Allen et al., 1998) as implemented in Vision Analyzer2 (Brain Products) and thereafter by objective (CBC parameters, Vision Analyzer) ICA-based rejection of residual artifact components, after which the EEG had a sampling rate of 250 Hz (Jahnke et al., 2012).