Lateralisation of subcortical functional connectivity during and after general anaesthesia

Abstract

Background: Arousal and awareness are two important components of consciousness states. Functional neuroimaging has furthered our understanding of cortical and thalamocortical mechanisms of awareness. Investigating the relationship between subcortical functional connectivity and arousal has been challenging owing to the relatively small size of brainstem structures and thalamic nuclei, and their depth in the brain.

Methods: Resting state functional MRI scans of 72 healthy volunteers were acquired before, during, 1 h after, and 1 day after sevoflurane general anaesthesia. Functional connectivity of subcortical regions of interest vs whole brain and homotopic functional connectivity for assessment of left-right symmetry analyses of both cortical and subcortical regions of interest were performed. Both analyses used high resolution atlases generated from deep brain stimulation applications.

Results: Functional connectivity in subcortical loci within the thalamus and of the ascending reticular activating system was sharply restricted under anaesthesia, featuring a general lateralisation of connectivity. Similarly, left-right homology was sharply reduced under anaesthesia. Subcortical bilateral functional connectivity was not fully restored after emergence from anaesthesia, although greater restoration was seen between ascending reticular activating system loci and specific thalamic nuclei thought to be involved in promoting and maintaining arousal. Functional connectivity was fully restored to baseline by the following day.

Conclusions: Functional connectivity in the subcortex is sharply restricted and lateralised under general anaesthesia. This restriction may play a part in loss and return of consciousness.

Clinical trial registration: NCT02275026.

Keywords: anaesthesia; arousal; consciousness; fMRI; reticular formation; subcortex; thalamus.

Copyright © 2021 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.

Figures

Fig 1

Consort diagram.

Fig 2

Thalamic functional connectivity. Seed-voxel analysis characterises functional connectivity before anaesthesia (PRE), during anaesthesia (ANA), 1 h after emergence (POST), and 1 day later (D1) for the following regions of interest (compared with the rest of the brain): (a) left thalamus, (b) centromedian nucleus, (c) ventral lateral anterior nucleus, and (D) left executive network (cortex). Under anaesthesia (ANA), the thalamic seeds show unilateral restriction of functional connectivity as compared with cortex (exemplified by the left executive network). Functional connectivity is restored at D1 as compared with PRE. Statistical threshold is set at P<0.01, false discovery rate (FDR)-corrected for multiple testing. CM, centromedian; EX, executive function network; THAL, thalmus; VLA, ventral lateral anterior nucleus; CM, centromedian nucleus.

Fig 3

Functional connectivity of ARAS loci. Seed-voxel analysis characterises functional connectivity before anaesthesia (PRE), during anaesthesia (ANA), 1 h after emergence (POST), and 1 day later (D1) for the regions of interest comprising the ascending reticular activating system (ARAS) (compared with the rest of the brain): (a) midline ventral tegmental area (VTA), (b) bilateral locus coeruleus, and (c) left locus coeruleus. As seen in a–c, before anaesthesia (PRE) these nuclei are connected to each other and to other brainstem loci, the thalamus, hippocampus, and the temporal poles cortically. Under anaesthesia (ANA) this connectivity is lost. Similar to thalamic structures, these nuclei also exhibit lateralisation under anaesthesia, as seen for the left unilateral locus coeruleus (LC). Statistical threshold is set at P<0.01, false discovery rate (FDR)-corrected for multiple testing.

Fig 4

ARAS to thalamus functional connectivity. Seed-voxel analysis characterises functional connectivity between ascending reticular activating system (ARAS) nuclei (treated as one combined seed) and the thalamus before anaesthesia (PRE), during anaesthesia (ANA), 1 h after emergence (POST), and 1 day later (D1). Connectivity between the ARAS nuclei and thalamus is largely disrupted under anaesthesia (ANA) and remains so into early recovery (POST), with the exception of connectivity between ARAS and the intrathalamic centromedian (CM) and parafascicular (PF) nuclei, highlighted by the yellow circles (2 mm spherical region of interest centred at Montreal Neurological Institute (MNI) (6, −25, −2) and crosshairs. Statistical threshold is set at P<0.01, false discovery rate (FDR)-corrected for multiple testing.

Fig 5

Homotopic functional connectivity. Correlation coefficients (averaged across subjects) in homotopic regions, normalised to values before anaesthesia. Control atlas: aMTG, middle temporal gyrus, anterior; IC, insular cortex; MidFG, middle frontal gyrus; PreCG, precentral gyrus; pSTG, superior temporal gyrus, posterior; SFG, superior frontal gyrus; sLOC, lateral occipital cortex, superior; SPL, superior parietal lobule; toITG, inferior temporal gyrus, temporo occipital. Subcortical atlas: GPe, globus pallidus external; GPi, globus pallidus internal; LC, locus coeruleus; MRF, midbrain reticular formation; MB, mammillary body; PBC, parabrachial complex; PO, oral pons; PPN, pedunculopontine nuclei; the thalamus. ∗Regions with homology that was significantly different under anaesthesia (after correction for multiple comparisons using false discovery rate (FDR)).

Supplementary figure S1

Seed voxel analysis characterizes functional connectivity before anaesthesia (PRE), during anaesthesia (ANA), 1 hour after emergence (POST), and 1 day later (D1) for the following regions of interest (compared to the rest of the brain): Left parafascicular nucleus (A), left mediodorsal nucleus (B), left parabrachial complex (C), median raphe (D), pedunculopontine nucleus (E) and the dorsal raphe (F).