Synopsis

Keywords: Stroke, fMRI (resting state), deep brain stimulation

To improve reproducibility of positive outcomes from cerebellar dentate stimulation for chronic stroke, we hypothesized that the volume of tissue activated (VTA) (dentate_VTA) proximity to the volume functionally connected to ipsilesional motor-associated cortices (dentate_{FC-ipsi-motor}) would be associated with improved arm impairment/function metrics. Dentate_VTA was estimated using computed tomography and T1-weighted MRI. Dentate_{FC-ipsi-motor} was estimated from fMRI (resting state). Significant correlations with arm metric improvement were not found for dentate_VTA proximity to dentate_{FC-ipsi-motor}, but were found for dentate_VTA size, distance from center, and percentage VTA in the inferior-lateral-posterior region. These findings suggest stimulating a small target at the inferior-posterior-lateral dentate edge.

Introduction

Stroke can cause crossed cerebellar diaschisis, characterized by reduced metabolism within the contralesional cerebellum. In a phase I clinical trial, stimulation of the contralesional cerebellar dentate nucleus enhanced the effects of rehabilitation of hand motor deficits in the majority of chronic post-stroke participants.

In the clinical trial, the electrode was placed based on neurosurgical experience. We hypothesized that simulation close to the dentate region with residual functional connectivity to the ipsilesional motor-associated cortices (dentate_{FC-ipsi-motor}) could improve the reproducibility of positive outcomes. As the first step to testing this hypothesis, we measured the association between proximity of the dentate volume of tissue activated (VTA) by electrode stimulation (dentate_VTA) to dentate_{FC-ipsi-motor} and change in arm impairment and function metrics. We also measured the independent associations between dentate_{FC-ipsi-motor} and dentate_VTA and change in arm metrics. Finally, we measured the association between the VTA location within the dentate and change in arm metrics.

Methods

Data was available in 8 chronic stroke participants.

Clinical metrics: Upper-extremity impairment was quantified using the Fugl-Meyer Assessment (FM-UE), and arm function was quantified using the Arm Motor Ability Test (AMAT). The change in arm metrics ΔAMAT and ΔUEFM were calculated as the difference in score immediately after stimulation was stopped relative to prior to the start of stimulation.

MRI acquisition: T1-weighted (T1w) MRI was acquired at 3 tesla (T), using magnetization prepared rapid gradient echo imaging with 192 1mm-thick slices, in-plane resolution=1x1mm². Resting-state functional MRI (rs-fMRI) were acquired at 7T in awake patients with eyes closed, using a simultaneous multi-slice echo-planar-imaging with 81 contiguous 1.5mm-thick axial slices, in-plane resolution=1.2x1.2mm², 128 volumes.

VTA estimation: For comparisons of VTA and functional connectivity, VTA estimation was performed using the Lead-DBS toolbox¹ with the SimBio/Fieldtrip pipeline². Atlas-based³ segmentation used the pre-implant T1w MRI, and lead localization used the co-registered postoperative computed tomography imaging. Modeling parameters matched participants’ clinical settings. For measuring the association between the VTA location within the dentate and change in arm metrics, data from VTAs previously quantified using Guide XT (Boston Scientific) were used.

Dentate metrics:
· The dentate was manually segmented on the average rs-fMRI volume. The ipsilesional Human Motor Associated Template (HMAT⁴) was warped to rs-fMRI-space. For each dentate voxel, the associated z-map was calculated and significant clusters⁵ were identified. The dentate voxel was classified as dentate_{FC-ipsi-motor} if the associated z-map contained a significant cluster with a center within the ipsilesional HMAT. To minimize the effects of noise in the z-maps on voxel classification, z-maps with widespread significant clusters within and outside the ipsilesional HMAT on both ipsilesional and contralesional sides were omitted.

· The dentate VTA was warped to rs-fMRI-space to calculate overlap_FC-VTA=(dentate_{FC-ipsi-motor} ∩ dentate_VTA)/|dentate_{FC-ipsi-motor}|. To overcome noise in the voxel classification, we also measured distance_FC-VTA, the distance between the centers of dentate_{FC-ipsi-motor} and dentate_VTA.

· To characterize dentate_{FC-ipsi-motor} and dentate_VTA independently, we calculated their sizes relative to the dentate size (size_FC/Den and size_VTA/Den) and the distance between their centers from the dentate center (distance_FC-Den and distance_VTA-Den).

· Using Matlab (2020b, The MathWorks, Inc.), a principal components analysis of the dentate was performed to quantify the percentage of the VTA within 8 dentate regions: superior-anterior-medial, superior-anterior-lateral, superior-posterior-medial, superior-posterior-lateral, inferior-anterior-medial, inferior-anterior-lateral, inferior-posterior-medial, inferior-posterior-lateral.

Statistical analyses: Statistical analyses were performed using Matlab. Pearson and Spearman correlations were used to evaluate the associations between MRI metrics and change in arm function metrics.

Results

The patient demographics, stroke characteristics and change in arm metrics are shown in the Table. Figure 1 shows the contralesional dentate, VTA, and dentate_{FC-ipsi-motor} segmentations for a participant who experienced a large improvement of arm function and a participant who experienced a small improvement of arm function. There were no statistically significant correlations between overlap_FC-VTA and distance_FC-VTA and ΔUEFM or ΔAMAT (Fig. 2, top). Statistically significant correlations were found between size_VTA/Den and ΔAMAT (r=-0.867, p=0.005, Fig. 2, bottom-left) and between distance_VTA-Den and ΔAMAT (ρ=0.731, p=0.048, Fig. 2, bottom-right), but not with ΔUEFM. Correlations between size_FC/Den and distance_FC/Den and ΔUEFM or ΔAMAT were not significant. The percentage of VTA in the inferior-posterior-lateral dentate significantly correlated with ΔUEFM (r=0.879, p=0.004), but not for the other dentate regions or with ΔAMAT.

Discussion

We hypothesized that dentate_VTA proximity to dentate_{FC-ipsi-motor} would be associated with improved arm metrics. This was not supported by linear correlation, however statistically significant correlations between size_VTA/Den and distance_VTA-Den and ΔAMAT suggested that ΔAMAT was associated with a smaller-sized VTA located away from the dentate center. To improve VTA characterization, additional correlation analyses between VTA location and change in arm metrics revealed a significant correlation between the percentage of VTA in the inferior-posterior-lateral dentate and ΔUEFM. Together the findings suggest that a small-sized VTA located at the inferior-posterior-lateral dentate edge may provide positive therapeutic outcomes. These findings are limited by several factors, including: underestimation of dentate_{FC-ipsi-motor} due to noise, inaccuracies in VTA estimation, and inaccuracies in measuring overlap and distances due to inaccuracies in image registration. Future studies will quantify how the dentate_{FC-ipsi-motor} changes with stimulation.

Conclusion

These findings suggest that electrode placement in the contralesional dentate for chronic stroke therapy should avoid stimulating the dentate widely, and target a small region at the inferior-posterior-lateral dentate edge.

Acknowledgements

This work was supported by the National Institutes of Health - [UH3-NS100543] and Enspire DBS Therapy, Inc.

References

1. Horn A and Kühn AA: Lead-DBS: A toolbox for deep brain stimulation electrode localizations and visualizations. Neuroimage; 2015;107:127-135.

2. Horn A, Reich M, Vorwerk J, et al. Connectivity Predicts deep brain stimulation outcome in Parkinson disease. Ann Neurol. 2017;82(1):67-78.

3. Diedrichsen J, Maderwald S, Küper M, et al. Imaging the deep cerebellar nuclei: a probabilistic atlas and normalization procedure. Neuroimage; 2011;54(3):1786-1794.

4. Mayka MA, Corcos DM, Leurgans SE, et al. Three-dimensional locations and boundaries of motor and premotor cortices as defined by functional brain imaging: a meta-analysis. Neuroimage; 2006; 31(4):1453-1474.

5. Jenkinson M, Beckmann CF, Behrens TEJ, et al. FSL. Neuronimage; 2012; 62(2):782-790.