Interventional neurorehabilitation for promoting functional recovery post-craniotomy: a proof-of-concept

Abstract

The human brain is a highly plastic 'complex' network-it is highly resilient to damage and capable of self-reorganisation after a large perturbation. Clinically, neurological deficits secondary to iatrogenic injury have very few active treatments. New imaging and stimulation technologies, though, offer promising therapeutic avenues to accelerate post-operative recovery trajectories. In this study, we sought to establish the safety profile for 'interventional neurorehabilitation': connectome-based therapeutic brain stimulation to drive cortical reorganisation and promote functional recovery post-craniotomy. In n = 34 glioma patients who experienced post-operative motor or language deficits, we used connectomics to construct single-subject cortical networks. Based on their clinical and connectivity deficit, patients underwent network-specific transcranial magnetic stimulation (TMS) sessions daily over five consecutive days. Patients were then assessed for TMS-related side effects and improvements. 31/34 (91%) patients were successfully recruited and enrolled for TMS treatment within two weeks of glioma surgery. No seizures or serious complications occurred during TMS rehabilitation and 1-week post-stimulation. Transient headaches were reported in 4/31 patients but improved after a single session. No neurological worsening was observed while a clinically and statistically significant benefit was noted in 28/31 patients post-TMS. We present two clinical vignettes and a video demonstration of interventional neurorehabilitation. For the first time, we demonstrate the safety profile and ability to recruit, enroll, and complete TMS acutely post-craniotomy in a high seizure risk population. Given the lack of randomisation and controls in this study, prospective randomised sham-controlled stimulation trials are now warranted to establish the efficacy of interventional neurorehabilitation following craniotomy.

Figure 1

The connectome construction pipeline used in this study. (A) A standard Glasser atlas was established using 300 healthy individuals from the Human Connectome Project (HCP). A supervised machine learning algorithm was employed to recognise connectivity patterns for each of the 360 HCP parcels in a healthy cohort. (B) Using diffusion sequences, we applied constrained spherical deconvolution (CSD) tractography to our patient cohort. Using these images, our algorithm was applied to recognise and adjust the locations of HCP parcels in highly atypical brains. (C) After establishing maximal likely structural connectivity, we used this data to inform and constrain functional connectivity using resting-state fMRI. (D) Finally, structural and functional anomaly matrices were generated to compare network connectivity differences (i.e. language) between our patient and a normative atlas. Adopted with permission from Reference.

Figure 2

Demonstration of proprietary machine learning algorithm (Omniscient) that assigns parcellations to very distorted brains. Patient with a frontal lobe GBM and resected regions resulting in total anterior brain shift. (a) Displays the modified location of the caudate nucleus and the putamen. (b) Displays the modified location of the GP. (c) Displays the modified location of the basal forebrain. (d) Displays the modified location of right 55b parcellation. (e) Displays the modified location of the right PBelt. This allows for the creation of a connectivity matrix of any brain despite.

Figure 3

(a) compares connectivity matrices of the left and right visual networks in a patient with hemianopia. The left visual network is dotted mostly blue, which means that areas of the visual system are not well synchronized to one another. By comparison, the right visual network displays strong intra-network connectivity. (b) compares the connectivity matrix of the language area of a healthy control on the left with the language area connectivity of an aphasic patient on the right. This aphasic matrix has the parcellations within the language system anticorrelated, therefore, predominantly blue, suggestion loss of connectivity within the language network. Note that columns 55b, 45 and STSdp are blue representing that they are isolated. We hypothesized that this is in part due to problems with the superior longitudinal fasciculus/ arcuate fasciculus system which links different components of the language system. Conducting connectomics analysis by comparing connectivity matrices enables us to generate potential targets for TMS treatment.

Figure 4

Neurological data (both language and motor) pre-rTMS treatment and 1-week post-rTMS treatment. (A) demonstrates a significant decrease in the ART 1-week following frontoparietal rTMS (p = 1.48 × 10^–3. (B) demonstrates a significant increase in the MRC motor strength scale 1-week following sensorimotor rTMS (p = 8.0 × 10^–5). (C–E) represents the data with patients with only lower-limb (C) or upper-limb (D), or both (E) limb deficits.

Figure 5

TMS strategy for patient presenting with aphasia and near complete hemiplegia secondary to glioblastoma. (a) Postoperative MRI of patient demonstrating resection cavity. (b) Independent right sided (green) and left sided (orange) sensorimotor networks. Although presented on the same image, these networks appeared as separate networks on connectomic analysis. The anterior satellite areas in the left (orange) dysfunctional sensorimotor network. (c) Left frontoparietal network demonstrating clear Broca’s area and area 55b. The temporal component of the network is disorganized. (d) cTBS was administered to both the middle of the sensorimotor network and the right posterior frontal component of the right frontoparietal network. (e) iTBS was administered to the disorganized temporal component of the left frontoparietal network and the (f) anterior areas of the pathological left sensorimotor network.

Figure 6

TMS strategy for patient presenting with moderate expressive aphasia secondary to glioblastoma. (a) Preoperative MRI (left) demonstrating left insula glioblastoma and postoperative MRI (right) demonstrating complete resection. (b) Network analysis demonstrating a strongly organized posterior temporal region that is not in in the same network as Broca’s area. This was the area that was selected for treatment with iTBS. (c). Further network analysis demonstrating Broca’s area with bilateral representation that is not in the same network as the posterior temporal region.