K-complex associated thalamic BOLD signal changes revealed by EEG/fMRI

Introduction: Recent studies suggest that the K-complex (KC) reflects a synchronized and cortically generated state [1]. Electrophysiological data support the role of the thalamus in the mediation of graphoelements (sleep spindles, delta waves) following the K-complex [2]. We hypothesized that KC detected with scalp EEG during stage 2 sleep correlate with thalamic (volume of interest) blood oxygen level-dependent (BOLD) signal changes.

Material and Methods: We analyzed twenty-five healthy, not formally sleep-deprived, eyes-closed resting volunteers studied with functional magnetic resonance imaging at 3T and simultaneous polysomnography recordings including EEG, electrooculogram, chin/tibial electromyography (BrainAmp MRplus/ExG, Brain Products, Gilching, Germany), respiration-belt, and pulse oximetry (Siemens). MR-induced artifacts were removed from the EEG. Manually marked KC, vertex sharp waves and sleep spindles occurring in stage N2 sleep (minimum epoch length 2min, minimum total duration 5min) each served to form a common event-related general linear model (SPM5) via convolution with the canonical haemodynamic response function and its temporal derivative; cardiac, respiratory and motion-induced noise were modelled as confounds.

Results: The second level random effects analysis of KC-correlated BOLD signal changes revealed a significant (p<0.05 family-wise error corrected for the thalamic search volume of interest) activation bilaterally in the thalamus. Results for sleep spindles and vertex sharp waves will be reported elsewhere.

Conclusion: Our findings demonstrate thalamic activity during the occurrence of KC confirming our hypothesis. Further analyses will examine thalamic interplay with other, specifically frontal cortical and primary sensory areas as well as the KC's relation to other sleep-specific graphoelements to further the understanding of the role of KC in the maintenance of NREM-sleep and sleep stage transitions.

Bibliography:

[1] Cash SS, Halgren E, Dehghani N, Rossetti AO, Thesen T, Wang C, Devinsky O, Kuzniecky R, Doyle W, Madsen JR, Bromfield E, Eross L, Halász P, Karmos G, Csercsa R, Wittner L, Ulbert I. The human K-complex represents an isolated cortical down-state. Science. 2009 May 22;324(5930):1084–7.

[2] Amzica F, Steriade M., The functional significance of K-complexes. Sleep Med Rev. 2002Apr;6(2):139–49.

Fig.1: Die Gruppenanalyse (25 Probanden, random effects) zeigt signifikante (p<0,05 FWE korrigiert für thalamisches Suchvolumen) positive BOLD-Signalveränderungen im Thalamus beidseits, hier projiziert auf eine Standard-T1-Hirndarstellung.

Fig.2: Random effects group analysis (25 subjects) revealed significant (p<0.05 FWE for the thalamic search volume) positive K-complex-associated BOLD signal changes bilaterally in the thalamus which are displayed on an T1-MNI-template brain.