Personalized Connectomic Targeting for Ventral Intermediate Nucleus DBS in Essential Tremor

24 

Abstract Submission 

Ha Neul Song¹, Sohae Chung², Sema Akkus¹, Brian Kopell¹, Ki Sueng Choi¹

¹Icahn School of Medicine at Mount Sinai, New York, NY, ²New York University, New York, NY

Ha Neul Song  
Icahn School of Medicine at Mount Sinai
New York, NY

Sohae Chung  
New York University
New York, NY

Sema Akkus  
Icahn School of Medicine at Mount Sinai
New York, NY

Brian Kopell  
Icahn School of Medicine at Mount Sinai
New York, NY

Ki Sueng Choi  
Icahn School of Medicine at Mount Sinai
New York, NY

Millions of individuals suffer from tremor disorders, such as essential tremor and tremor-dominant Parkinson's disease. Deep brain stimulation (DBS) of the ventral intermediate nucleus (VIM) has proven effective for treatment-resistant essential tremor. However, the clinical outcome of DBS highly depends on precise targeting, which has been challenging due to imaging limitations and inter-individual variability. Moreover, previous studies using diffusion-tractography suggest that the efficacy of VIM DBS may be associated with the dentate-rubro-thalamic tract (DRTT), but no straightforward approach exists to extract DRTT associated with a therapeutic benefit. This study aimed to define therapeutic pathway maps based on anatomical or tractography data and validate them using clinical measures. Additionally, the study evaluated the therapeutic pathway maps by measuring inter-subject similarity using human connectome project (HCP) data.

We studied 10 patients with severe tremors who underwent VIM DBS surgery (age range, 70 - 82 years). Patients were divided into two groups based on their TETRAS scores measured before and within 1 year post-surgery: responders (n = 8, improvement ≧ 50%) and non-responders (n = 2, improvement < 50%). Individual leads were reconstructed, and volumes of tissue activated (VTAs) were estimated using Lead DBS software with given stimulation settings (amplitude and contact configuration). First, therapeutic anatomical maps were defined by averaging VTAs weighted by clinical improvement. Therapeutic pathway maps were also defined by estimating VTA-based structural connectivity shared by responders. Next, overlapping voxels of anatomical and pathway maps with individual VTAs and connectivity were calculated. Lastly, the inter-subject similarity (reliability) of the therapeutic pathway was evaluated using the Human Connectome Project dataset (n = 100).

Results showed a slight difference in anatomical overlap with VTAs between responders and non-responders (p = 0.07; RES = 44.3; NON = 3.0). On the other hand, there was a significant difference in connectomic overlap between responders and non-responders (p = 0.02; RES = 161.3; NON = 64.5). This trend and significance were observed explicitly in the left brain hemisphere. Lastly, high similarity in connectivity toward therapeutic pathways was found among HCP subjects (0.7 out of 1.0).

This study defined the therapeutic pathways for VIM DBS, surpassing anatomical maps in differentiating clinical improvement and demonstrating high similarity among individuals. The results imply the importance of connectomic targeting for VIM DBS and its potential practical applications. Minor variability (0.3 out of 1.0) indicates the need for individual data in personalized DBS targeting. These findings offer valuable insights into accurate and personalized neurosurgical guidance.

Deep Brain Stimulation ¹

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) ²

White Matter Anatomy, Fiber Pathways and Connectivity

Movement Disorder

MRI

White Matter

Other - Deep Brain Stimulation; Essential Tremor