Deep brain stimulation of the thalamus restores signatures of consciousness in a nonhuman primate model

Abstract

Loss of consciousness is associated with the disruption of long-range thalamocortical and corticocortical brain communication. We tested the hypothesis that deep brain stimulation (DBS) of central thalamus might restore both arousal and awareness following consciousness loss. We applied anesthesia to suppress consciousness in nonhuman primates. During anesthesia, central thalamic stimulation induced arousal in an on-off manner and increased functional magnetic resonance imaging activity in prefrontal, parietal, and cingulate cortices. Moreover, DBS restored a broad dynamic repertoire of spontaneous resting-state activity, previously described as a signature of consciousness. None of these effects were obtained during the stimulation of a control site in the ventrolateral thalamus. Last, DBS restored a broad hierarchical response to auditory violations that was disrupted under anesthesia. Thus, DBS restored the two dimensions of consciousness, arousal and conscious access, following consciousness loss, paving the way to its therapeutical translation in patients with disorders of consciousness.

Fig. 1.. Experimental design.

Schematic representation of the study design, followed by the experimental steps and modalities of investigation.

Fig. 2.. DBS of the CT restores arousal and cortical entropy in anesthetized macaques.

(A) Anatomical localization of the DBS lead and active contacts. Coronal section of an anatomical MRI. Zoomed-in view of the thalamic nuclei as segmented with the CIVM atlas. DBS was delivered to either the CT (star label) or the ventral lateral thalamus (VL) (square label). 1, mediodorsal nucleus, central part; 2, mediodorsal nucleus, lateral part; 3, ventral lateral nucleus, lateral part; 4, ventral posterolateral nucleus; 5, lateral dorsal nucleus, superficial part; 6, ventral anterior nucleus, lateral part; 7, intermediodorsal nucleus; 8, mediodorsal nucleus, medial part; 9, ventral posteromedial nucleus; 10, CM nucleus, lateral part; 11, CM nucleus, medial part; 12, ventral lateral nucleus, medial part; 13, centrolateral nucleus; 14, mediodorsal nucleus, dorsal part. (B) Automated electrode reconstruction and three-dimensional (3D) rendering in monkey T (left), in monkey N (center), and in both monkeys (right). (C and D) Coronal histological section, corresponding to the level interaural 8.40 mm. Both cryostat image (C) and NeuN immunohistochemistry image (D) confirm the positioning of the active contacts within thalamic nuclei. (E) Effects of thalamic DBS on arousal in anesthetized monkeys, as a function of the electrode location and the level of stimulation (low-voltage versus high-voltage DBS). Only the stimulation of the CT could modulate arousal in the two anesthetized monkeys. (F) Modulation of spectral entropy (SE) by DBS. The figure consists of a distribution-smoothened version of a histogram, a box plot, and a representation of the data points. Each dot represents the average value of SE across epochs during one recording session. All pairs show significant group differences except for anesthesia + low CT-DBS and anesthesia + high VL-DBS. **0.001 < P ≤ 0.01, ***0.0001 < P ≤ 0.001, and ****P < 0.0001. P values are false discovery rate (FDR)–corrected.

Fig. 3.. Thalamic DBS induces remote brain activations.

(A) Effects of thalamic DBS induced on distant cortical areas. Cortical fMRI activation maps during low CT-DBS (top left), high CT-DBS (bottom left), low VL-DBS (top right), and high VL-DBS (bottom right). Individual results (monkey T), P < 0.05, family wise error (FWE)–corrected. (B) Effects of thalamic DBS on the blood oxygen level–dependent (BOLD) signal change (%) within the cortex. Red shading shows the stimulation period. High CT-DBS consistently activated a prefrontal parietal network. Only high CT-DBS activated both anterior and posterior cingulate cortices. (C) Effects of high CT-DBS on stationary FCs. Whole-brain average stationary intervoxel correlations before (blue), during high CT-DBS (green), and after high CT-DBS (yellow). (D) Average positive and negative z values before, during, and after high CT-DBS. In all plots, error bars represent 1 SEM. ACC, anterior cingulate cortex; area 9/46 [dorsolateral prefrontal cortex (PFCdl)]; area 8A [part of frontal eye field (FEF)]; area 6V [dorsolateral premotor cortex (PMCdl)]; area M1, primary motor cortex; PFG, parietal area PFG; VIP, intraparietal cortex (Pcip); PCC, posterior cingulate cortex; TPO, temporo-parieto-occipital–associated area in superior temporal sulcus.

Fig. 4.. Central thalamic stimulation restores thalamocortical and corticocortical static connectivity in anesthetized macaques.

(A) Average positive and negative z values within each experimental condition. In all plots, error bars represent 1 SEM. All pairwise comparisons between experimental conditions are significantly different (P < 0.001) both for positive and negative z values, except for the negative z value comparison between anesthesia and high VL-DBS (n.s., not significant). (B) Effects of DBS on corticocortical FCs. Average connectivity matrices displaying intervoxel correlations between selected cortical regions of interest (ROIs) of the “macaque” GNW (ACC; area 9/46; area 8A, area 6V, VIP, and PCC) under different experimental conditions. x axis, arousal state; y axis, studied regions. For each region, the matrix represents the FCs between the defined seed (in the right hemisphere) and the remaining selected regions. (C) Effects of DBS on corticocortical FCs. Schematic representation of the cortical static correlations of the macaque GNW nodes (ACC, PCC, PFCdl, FEF, PMCdl, and Pcip) and M1, S1, V1, and A1; correlation strengths higher than 0.3 of the right hemisphere for the different experimental conditions. (D) Effects of DBS on thalamocortical FCs. Average connectivity matrices displaying intervoxel correlations between selected thalamic nuclei and cortical ROIs representing the macaque GNW (ACC; area 9/46; area 8A, area 6V, VIP, and PCC) under different experimental conditions. x axis, arousal state; y axis, studied regions. For each region, the matrix represents the FC between the defined seed (in the right hemisphere) and the remaining studied areas. Area 9/46 (PFCdl); area 8A (part of FEF) and area 6V (PMCdl); area M1, primary motor cortex; area A1, primary auditory cortex; VIP: Pcip.

Fig. 5.. Central thalamic stimulation restores dynamic connectivity in anesthetized macaques.

Effects of high CT-DBS (A to C) or high VL-DBS (D and E) on dynamic connectivity of the cortex. Unsupervised clustering of the covariance FC matrix revealed seven brain states, which are sorted according to their similarity to the structural connectivity matrix. (A) Probability distributions of brain states for the awake state, anesthesia, anesthesia + high CT-DBS clustering. Each bar represents the within-condition probability of occurrence of a state. (B) Probability of occurrence of each brain state as a function of the similarity between FC and structural connectivity for the awake state, anesthesia, and anesthesia + high CT-DBS. (C) Changes of FC across the seven brain states between the GNW and its remaining areas such as PFCdl, PMCdl, Pcip, ACC, and PCC. On the x axis, brain states 1 to 7. On the y axis, studied ROIs of the macaque GNW [ACC, PFCpol (polar prefrontal cortex), FEF, PMCdl, S1, M1, Pcip, A1, TCi, V1, and PCC of the left (L) and right (R) hemisphere] connected to the seed. (D) Probability distributions of brain states for the awake state, anesthesia, anesthesia + high VL-DBS. Each bar represents the within-condition probability of occurrence of a state. (E) Probability of occurrence of each brain state as a function of the similarity between FC and structural connectivity for the awake state, anesthesia, and anesthesia + high VL-DBS. Primary somatosensory cortex (S1); primary motor cortex (M1); primary auditory cortex (A1); inferior temporal (TCi); visual area 1 (V1).

Fig. 6.. Electrical stimulation of central thalamic nuclei restores cortical hierarchical responses to local and global novelty detections in anesthetized macaques.

(A) Experimental paradigm. The task consisted of passive listening to the local-global paradigm, which probes auditory sequence processing at two hierarchical levels of deviancy. The first level of novelty detection corresponds to a deviant sound within a sequence of identical sounds, while the second level corresponds to a deviant sequence. (B) First-order novelty detection “local effect” (pitch violation). Activation maps show a restoration of first-order deviancy detection when high CT-DBS is applied in anesthetized monkeys. Group results, P < 0.05, FDR-corrected. (C) Second-order novelty detection “global effect” (sequence violation). Activation maps show a partial enhancement of second-order deviancy detection during high CT-DBS in anesthetized monkeys. Group results, P < 0.05, FDR-corrected. fMRI signal change in area 9/46V (dorsolateral prefrontal cortex), area 23c (PCC), and IPS (intraparietal sulcus). (D) Task-evoked connectivity during the global novelty effect. A seed is applied to the right auditory cortex (purple) to look for PPI during the global novelty effect. High CT-DBS resulted in an increase in PPI between auditory cortex and prefrontal parietal and cingular cortices. Group results, P < 0.05, FDR-corrected. Area 6D (dorsal), 6M (medial), 6/32, 9L (lateral), 10D (dorsal), 10M (medial), 25, 45B, 9/46D (dorsal), PRoM (promotor), 46D (dorsal), prefrontal cortex; area 23, 23c, PGM/31, PCC; IPS, 3a, 3b (somatosensory), area PFG, PEC, PG, parietal cortex; area TEa, TPO, AKM (auditory koniocortex medial part), PaAL (para auditory lateral), FST (fundus of superior temporal sulcus) Dpt (dorsoparietal), TEO (temporo-occipital), TPO, V4 (visual area 4), temporal cortex; area V2 (visual area 2), occipital cortex; Cd, caudate nucleus; Pu, putamen; LGN, lateral geniculate nucleus; Th, thalamus; HyTh, hypothalamus.

Fig. 7.. Schematic representation showing the mechanisms by which thalamic DBS restores consciousness in anesthetized macaques.

Th, intralaminar nuclei of the thalamus.