Engaging dystonia networks with subthalamic stimulation

Abstract

Deep brain stimulation is an efficacious treatment for dystonia. While the internal pallidum serves as the primary target, recently, stimulation of the subthalamic nucleus (STN) has been investigated. However, optimal targeting within this structure and its surroundings have not been studied in depth. Indeed, historical targets that have been used for surgical treatment of dystonia are directly adjacent to the STN. Further, multiple types of dystonia exist, and outcomes are variable, suggesting that not all types would profit maximally from the same target. Therefore, a thorough investigation of neural substrates underlying stimulation effects on dystonia signs and symptoms is warranted. Here, we analyze a multicenter cohort of isolated dystonia patients with subthalamic implantations (N = 58) and relate their stimulation sites to improvements of appendicular and cervical symptoms as well as blepharospasm. Stimulation of the ventral oral posterior nucleus of thalamus and surrounding regions were associated with improvements in cervical dystonia, while stimulation of the dorsolateral STN was associated with improvements in limb dystonia and blepharospasm. This dissociation was matched by structural connectivity analysis, where the cerebellothalamic, corticospinal, and pallidosubthalamic tracts were associated with improvements of cervical dystonia, while hyperdirect and subthalamopallidal pathways with alleviation of limb dystonia and blepharospasm. On the level of functional networks, improvements of limb dystonia were associated with connectivity to the corresponding somatotopic regions in the primary motor cortex, while alleviation of cervical dystonia to the cingulo-opercular network. These findings shed light on the pathophysiology of dystonia and may guide DBS targeting and programming in the future.

Keywords: cervical dystonia; deep brain stimulation; limb dystonia; structural connectivity; sweet-spot analysis.

Fig. 1.

Clinical characteristics and subdivision of patients into three overlapping subcohorts. (Upper Panel) Consort flowchart for inclusion criteria. Patients from two independent cohorts (Shanghai N = 65, San Francisco N = 13) were considered as a single dataset. The final dataset consisted of 58 subjects. (Lower Panel) Classification into dystonic groups based on corresponding baseline subscores. Note that the patients are assigned to groups in nonexclusive fashion.

Fig. 2.

Appendicular dystonia. (A) Electrode reconstructions for patients included in the appendicular group. Note that proximal electrode contacts and stimulation volumes covered the ventral thalamus due to a comparably large contact spacing in the implanted electrodes. (B) Distribution of stimulation volumes, defined by electric fields thresholded at the magnitude of 0.2 V/mm. The peak intensity resides in the white matter region dorsolateral to the STN. (C) Structural connectivity statistically associated (P < 0.05) with appendicular improvement under DBS. Stimulation of the subthalamopallidal (indirect) pathway and the hyperdirect projections from the lower and upper extremity regions of the primary motor cortex (HDP-M1) positively correlated with clinical improvements, while the opposite was observed for the AL. (D) Voxel-wise correlation map of the electric field magnitude with stimulation outcome. The sweet-spot, thresholded for significance, localized to the dorsolateral aspect of STN, while the sour-spot predominantly resided in the ventral oral anterior nucleus of thalamus. (E) Clinical scores and their association with the fiber filtering and sweet-spot scores, quantified by spatial correlations between E-field distributions and sweet-spots/tractograms; Fisher Z-transformation was applied to the sweet-spot scores. To indicate robustness of the association, leave-one-out cross-validation is reported. For 10-fold cross-validations, see SI Appendix, Fig. S6. Models were also subjected to permutation tests: R = 0.69, P < 0.001 and R = 0.79, P < 0.001 for the sweet-spot and fiber filtering models, respectively.

Fig. 3.

Cervical dystonia. (A) Electrode reconstructions for patients included in the cervical group. Note that proximal electrode contacts and stimulation volumes covered the ventral thalamus due to a comparably large contact spacing in the implanted electrodes. (B) Distribution of stimulation volumes, defined by electric fields thresholded at the magnitude of 0.2 V/mm. The peak intensity resides in the white matter region dorsolateral to the STN. (C) Structural connectivity statistically associated (P < 0.05) with cervical improvement under DBS. Stimulation of the cerebellothalamic (dDRT) and corticospinal tracts (CST), as well as subthalamic afferents from the globus pallidus externus (indirect) positively correlated with clinical improvements, while the opposite was observed for the hyperdirect pathway from the primary motor cortex (HDP-M1). (D) Voxel-wise correlation map of the electric field magnitude with stimulation outcome. The sweet-spot, thresholded for significance, primarily localized to the ventral oral posterior nucleus of thalamus, while the sour-spot was identified in the central STN. (E) Clinical scores and their association with the fiber filtering and sweet-spot scores, quantified by spatial correlations between E-field distributions and sweet-spots/tractograms; Fisher Z-transformation was applied to the sweet-spot scores. To indicate robustness of the association, leave-one-out cross-validation is reported. For 10-fold cross-validations, see SI Appendix, Fig. S6. Models were also subjected to permutation tests: R = 0.50, P = 0.055 and R = 0.62, P < 0.05 for the sweet-spot and fiber filtering models, respectively.

Fig. 4.

Blepharospasm. (A) Electrode reconstructions for patients included in the blepharospasm group. Note that proximal electrode contacts and stimulation volumes covered the ventral thalamus due to a comparably large contact spacing in the implanted electrodes. (B) Distribution of stimulation volumes, defined by electric fields thresholded at the magnitude of 0.2 V/mm. The peak intensity resided in the white matter region dorsolateral to the STN. (C) Structural connectivity statistically associated (P < 0.05) with blepharospasm improvement under DBS. Stimulation of the subthalamopallidal tract (indirect), as well as the hyperdirect pathway from the primary motor cortex (HDP-M1) positively correlated with clinical improvements. (D) Voxel-wise correlation map of the electric field magnitude with stimulation outcome. The sweet-spot primarily localized to the STN proper, while the sour-spot resided in the capsule and, partially, in the ventral oral nuclei of the thalamus. (E) Clinical scores and their association with the fiber filtering and sweet-spot scores, quantified by spatial correlations between E-field distributions and sweet-spots/tractograms; Fisher Z-transformation was applied to the sweet-spot scores. To indicate the robustness of the association, leave-one-out cross-validation is reported. For 10-fold cross-validations, see SI Appendix, Fig. S6. Models were also subjected to permutation tests: R = 0.73, P < 0.001 and R = 0.60, P <0.01 for the sweet-spot and fiber filtering models, respectively.

Fig. 5.

Sweet-spot mappings in the three types of dystonia. The sweet- and sour-spots are computed by correlating distributions of electric fields with clinical outcomes across patients, coronal view. Note a partial inversion of the map for the cervical group, with the optimal stimulation site located in the thalamic and dorsal white matter region. A permutation-based similarity testing indicated a significant positive correlation of the appendicular and blepharospasm maps (R = 0.58, P < 0.001), and their nonsignificant negative correlation with the cervical map (R = −0.09, P = 0.77 and R = −0.37, P = 0.11, respectively).

Fig. 6.

Functional connectivity patterns associated with improvements in the three types of dystonia. (A) Maps of symptomatic improvement-associated functional connectivity across groups thresholded at the uncorrected significance level P < 0.05. (B) Surface projections of maps for appendicular (Top) and cervical (Bottom) improvement. While the appendicular improvement mapped onto a network that included peaks in the primary motor cortex, the cervical improvement map overlapped with the phylogenetically older motor system within the agranular prefrontal cortex and the cingulate motor area. Functionally, this network resembles the cingulo-opercular network (56) and connectivity profiles seeded from extrapyramidal motor areas such as the red nucleus (see panel D). (C) Contrast map of leg versus arm improvement (difference of z-scored symptomatic improvement maps) projected onto the primary motor cortex. This map peaked at the leg area (for lower limb improvements) and the hand knob (for upper limb improvements). The homuncular regions are adapted from ref. (licensed under CC-BY 4.0). (D) Cingulo-opercular network (56) and functional connectivity map of the red nucleus bear resemblance to the cervical improvement map (black outlines and see panel B).

Fig. 7.

Three representative cases with dystonia involving multiple body regions. Proximity of active contacts (highlighted in purple) to the sweet-spots (blue—appendicular, orange—cervical) reflects the improvement in the corresponding symptoms. (A) The first patient had improvement in both cervical and appendicular symptom clusters. Their two active contacts mapped to both sweet-spots. (B and C) The second and third patients had strong improvements only in one of the clusters, matching the sweet-spots that were stimulated. Note that the presumably optimal trajectory would coincide with the commonly employed trajectory for DBS in Parkinson’s disease.