Clinical characteristics of patients with dystonic crisis, also known as status dystonicus (SD).

Table outlines clinical characteristics of 30 patients admitted to Texas Children’s Hospital from 2015-2019 with acquired etiologies of dystonic crisis. Top grouping includes patient with only supratentorial findings, while bottom grouping includes those with cerebellar findings. Summarized imaging findings are as indicated in the respective columns. Ages are shown in years. “Dystonia phenomenology” refers to the description of dystonia based on consensus classifications while “Description of phenomenology” indicates the patient specific presentation of dystonia during SD. Patients with imaging depicted in figure 1 are as follows: # = top left, $ = bottom left, € = top middle, % = bottom middle, £ = top right, & = bottom right. Abbreviations of treatments with clinical benefit: BZ = benzodiazepine; α-2 = Alpha-2 agonists; AS = antispasmodic agents; ChemD = chemodenervation; VMAT2 = VMAT-2 inhibitor; None = no therapy with clinical benefit; Dopa = dopaminergic agonism; ITB = intrathecal baclofen pump; NmB = neuromuscular blockade.

Cerebellar abnormalities and inhibitory neural function are implicated in the incidence and treatment of dystonic crisis.

(A) Patient flow chart showing the division of 49 patients with dystonic crisis into groups based on the etiology of their dystonia: genetic/progressive (19 patients) or acquired injury (30 patients). (B) Percentage of patients admitted for dystonic crisis that had abnormalities in the cerebellum with or without supratentorial findings or only supratentorial abnormalities (blue) on imaging. (C) Examples of imaging in patients with supratentorial only injury (left column, outlined in blue) or cerebellar findings (middle and right column, outlined in yellow with abnormalities encircled). The top row depicts the axial plane through the cerebellum on T1 weighted imaging. The top left depicts the cerebellum at the level of the superior cerebellar peduncles, the top middle depicts a sequela of isolated left cerebellar hemispheric stroke, the top right depicts bilateral cerebellar hemispheric encephalomalacia (outlined in blue). The bottom row depicts the coronal plane through the cerebellum and the parietal lobe on T1 weighted imaging. The bottom right shows the normal cerebellum with extensive bilateral periventricular encephalomalacia. The bottom middle and right show patients with extensive bilateral cerebellar encephalomalacia. (D) Comparison of therapies that had clear clinical benefit across the 30 patients with acquired injury associated with dystonic crisis. Benzodiazepines and alpha-2 adrenergic agonists were the most widely used and successful. Patients with cerebellar abnormalities responded preferentially to alpha-2 adrenergic agonists rather than benzodiazepines. (E) Patients with cerebellar abnormalities tended to have longer lengths of hospitalization (Unpaired t-test with Welch’s correction p=0.167, f=0.001) and days to improvement (Unpaired t-test with Welch’s correction p=0.104, f=0.004) following their admission with dystonic crisis.

Photoactivation of the inhibitory cerebellar nuclei neuron (iCNN) pathway induces dystonic crises reminiscent of the human condition in Ptf1aCre;Vglut2fx/fx;ROSAlsl-Chr2/EYFPmice with existing dystonia, but not in otherwise healthy mice.

(A) Schematic showing the genetic manipulation in Ptf1aCre;Vglut2fx/fx mice that selectively silenced olivocerebellar synapses, which causes dystonia. (B) Using the Ptf1aCre driver, exon 2 of Vglut2 was selectively removed and Vglut2 deleted from the inferior olive. Concurrently, the loxP sites surrounding a stop signal preceding ChR2-YFP were removed, allowing expression of ChR2-YFP in Ptf1a-expressing neurons including inferior olivary neurons, Purkinje cells, and iCNNs. (C) In situ image showing co-labeling of YFP and Vgat in the cerebellar nuclei. (D) Schematic showing the selective photoactivation iCNNs through their projections in the superior cerebellar peduncle. (E) Photoactivation of the iCNN pathway induced a dystonic crisis on-demand. (F) In Ptf1aCre;Vglut2fx/fx;ROSAlsl-Chr2/EYFPmice, photoactivation of the iCNN pathway significantly increased the percentage of time spent in dystonic crisis when comparing before and during stimulation (n=9 animals; p=0.0039, Wilcoxon matched-pairs signed rank test, two-tailed). (G) Photoactivation of the iCNN pathway in healthy Ptf1aCre;ROSAlsl-Chr2/EYFPmice resulted in abnormal movements, but not dystonic crises. (H) Photoactivation of the iCNN pathway did not influence the percentage of time spent in dystonic crisis when comparing before and during stimulation (n=9 animals). Error bars are defined as standard error of the mean. SCP, superior cerebellar peduncle; PC, Purkinje cell; iCNN, inhibitory cerebellar nuclei neuron; IO, inferior olive; DN, dentate cerebellar nucleus.

Photoinhibition of the inhibitory cerebellar nuclei neuron (iCNN) pathway alleviates the spontaneous dystonic crises in Ptf1aCre;Vglut2fx/fx;ROSAlsl-Arch/EYFPmice with existing dystonia.

(A) Schematic showing the selective photoinhibition of the iCNN pathway by implanting optical fibers into the superior cerebellar peduncle of Ptf1aCre;Vglut2fx/fx;ROSAlsl-Arcmice. (B) Schematic showing the four-day experimental paradigm. (C) Photoinhibition of the iCNN pathway alleviated dystonic crises in Ptf1aCre;Vglut2fx/fx;ROSAlsl-Arch mice as shown in (D) by a significant decrease in the percentage of time spent in crisis when comparing pre versus during photostimulation on Day 1 (p=0.0156), Day 2 (p=0.0156), and Day 3 (p=0.0156) (E) Photoinhibition on Day 1 reduced dystonic crises across all time periods (for 1-5 vs 6-20, p=0.0312; for 1-5 vs 11-15, p=0.0156; for 1-5 vs 16-20, p=0.0156; for 1-5 vs 21-25, p=0.0156; for 1-5 vs 26-30, p=0.0156; for 1-5 vs 31-35, p=0.0156). Photoinhibition on Day 2 alleviated crises across several time periods (for 1-5 vs 6-20, p=0.0156; for 1-5 vs 11-15, p=0.2345; for 1-5 vs 16-20, p=0.0156; for 1-5 vs 21-25, p=0.2969; for 1-5 vs 26-30, p=0.0156; for 1-5 vs 31-35, p=0.0469). Photoinhibition on Day 3 reduced crises across several time periods (for 1-5 vs 6-20, p=0.0156; for 1-5 vs 11-15, p=0.2188; for 1-5 vs 16-20, p=0.0156; for 1-5 vs 21-25, p=0.1562; for 1-5 vs 26-30, p=0.0156; for 1-5 vs 31-35, p=0.0781). For all statistics: n=7 animals, Wilcoxon matched-pairs signed rank test, two-tailed, error bars are defined as standard error of the mean.

Inhibitory cerebellar nuclei neurons (iCNNs) interact with the centrolateral nucleus of the thalamus (CL) during dystonic crisis.

(A) Immunohistochemical staining in Ptf1aCre;Vglut2fx/fx;ROSAlsl-Chr2/EYFPmice revealed ChR2-YFP-labeled projections in the CL. (B) Schematic of genetics in Ptf1aCre;Vglut2fx/fx;ROSAlsl-Sun1-GFP mice that allowed localized GFP expression in the somas of Ptf1a-expressing neurons. (C) Experimental setup to identify iCNNs projecting to the CL. Retrograde BDA 3K tracer was injected into CL of Ptf1aCre;Vglut2fx/fx;ROSAlsl-Sun1-GFPmice as in (D), and iCNNs were co-labeled with BDA 3K and GFP as shown in (E, inset i) (n=4 mice). (F) Schematic showing the selective photoactivation of iCNN projections in the SCP and DBS of the CL. (G) Four-day experimental setup showing the optogenetic and DBS paradigms during the five distinct time periods for each day. (H) Representative images showing how DBS alleviated dystonic crisis induced by photoactivation of the iCNN pathway. (I) Dystonic crises induced by photoactivation of the iCNN pathway did not change greatly across time periods on Day 0 though a statistical difference was detected when comparing the photobaseline to immediately post-DBS (p=0.0312). Induced crises were substantially reduced on Day 1 when comparing the photobaseline to immediately post-DBS (p=0.0312), 30 minutes post-DBS (p=0.0312), and 1 hour post-DBS (p=0.0312) and on Day 2 when comparing the photobaseline to immediately post-DBS (p=0.0312). For all statistics: n=6 animals, Wilcoxon matched-pairs signed rank test, two-tailed, error bars are defined as standard error of the mean. CL, centrolateral nucleus of the thalamus; HP, hippocampus; VL, ventrolateral thalamus; VPM, ventroposterior thalamus; Pfl, paraflocculus; DN, dentate cerebellar nucleus. SCP, superior cerebellar peduncle; DBS, deep brain stimulation; iCNN, inhibitory cerebellar nuclei neuron.

Ptf1a-cre induced expression of cre-dependent ChR2-YFP in inhibitory cerebellar nuclei neurons, but not excitatory cerebellar nuclei neurons, in Ptf1aCre;Vglut2fx/fx;ROSAlsl-Chr2/EYFP mice.

(A) In situ hybridization in a coronal section showing mRNA expression Vgat and YFP as well as DAPI staining (n=3 animals; 3 sections per animal). The cerebellar nuclei are outlined in white. Inset (i) is a higher magnification image from (A) with white arrows marking cerebellar nuclei neurons that were co-labeled with Vgat and YFP probes. (B) In situ hybridization in a coronal section showing mRNA expression Vglut2 and YFP as well as DAPI, with the cerebellar nuclei outlined in white (n=3 animals; 3 sections per animal). Inset (ii) is a higher magnification image of (B) showing that there was no co-labeling of Vgut2 and YFP.

Anatomical verification of optical fibers over the superior cerebellar peduncle (SCP), which houses extracerebellar projections from inhibitory cerebellar nuclei neurons as labeled with ChR2-YFP in Ptf1aCre;Vglut2fx/fx;ROSAlsl-Chr2/EYFPmice.

(A) Coronal section demonstrating location of optic fiber over the SCP (n=9 animals, 3 sections per animal). Note the punctate labeling of axonal projections expressing ChR2-YFP in the SCP. (B) Sagittal section demonstrating location of optic fiber over the SCP (n=3 animals, 3 sections per animal). Insets (i and ii) show higher magnification of the SCP from (B), highlighting the expression of ChR2-YFP in axonal projections with white arrows. SCP, superior cerebellar peduncle; Ant, anterior cerebellar lobe; KF, Kolliker-Fuse nucleus; Su5, supratrigeminal nucleus; Mo5, trigeminal motor nucleus; Int, interposed cerebellar nucleus; FN, fastigial cerebellar nucleus.

Photoactivation of the inhibitory cerebellar nuclei neuron pathway through the superior cerebellar peduncle induced dystonic crises that occurred within a short latency from light onset and that persisted many seconds after stimulation ends in Ptf1aCre;Vglut2fx/fx;ROSAlsl-Chr2/EYFPmice.

(A) Dystonic crises were induced at around 10 seconds from light onset. (B) After the light stimulation ends, the induced dystonic crisis persisted for around 40 seconds. n=9 animals.

Photostimulation over several days resulted in a reduction in the time spent in dystonic crisis.

(A) For the pre stimulation periods when comparing Day 1 (the first day of stimulation) against Day 2 (p=0.0625) and Day 3 (p=0.0781), there was a trend toward a reduction in dystonic crisis, but it was not statistically significant. (B) For the during stimulation periods, there was a significant reduction in dystonic crises when comparing Day 1 to Day 2 (p=0.0312) and Day 3 (p=0.0156). For all statistics, n=7 animals, Wilcoxon matched-pairs signed rank test, two tailed.

Inhibitory cerebellar nuclei neurons from all three major cerebellar nuclei (fastigial, interposed, dentate) project to the centrolateral nucleus of the thalamus (n=4 animals, 3 sections per animal).

(A) Coronal section showing that neurons from the interposed and dentate cerebellar nucleus were filled with BDA 3K following injection of BDA 3K, a predominantly retrograde tracer, in the centrolateral nucleus of the thalamus. Insets (i) and (ii) are higher magnification images of (A) showing the co-labeling of DAPI, Sun1-GFP, and BDA 3K in the interposed and dentate cerebellar nucleus, respectively. (B) Coronal section showing that neurons from the fastigial cerebellar nucleus were filled with BDA 3K following injection of BDA 3K, a predominantly retrograde tracer, in the centrolateral nucleus of the thalamus. Inset (iii) is a higher magnification image of (B) showing a co-labeling of DAPI, Sun1-GFP, and BDA 3K. Inset (iv) is a higher magnification image of (B) showing co-labeling of DAPI and BDA 3K, but not Sun1-GFP. Int, interposed cerebellar nucleus; DN, dentate cerebellar nucleus; FN, fastigial cerebellar nucleus; 4V, fourth ventricle.

Anatomical verification of the surgical targeting of deep brain stimulation electrodes in the centrolateral nucleus of the thalamus.

(A) Immunohistochemical staining of DAPI showing the tracks formed by the bilateral implantation of deep brain stimulation electrodes in the centrolateral nucleus of the thalamus (n=6 animals, 3 sections per animal). Yellow triangles mark the tracks formed by the DBS electrodes above and with some penetration into the centrolateral nucleus of the thalamus. (B) Inset of A showing a higher magnified image of the DBS electrode track in the centrolateral nucleus of the thalamus. CL, centrolateral nucleus of the thalamus; hb, habenula; pv, paraventricular thalamic nucleus.

Deep brain stimulation of the centrolateral nucleus of the thalamus over several days reduced the time in dystonic crises induced by photoactivation of the inhibitory cerebellar nuclei neuron pathway.

(A) Comparing the induced dystonic crises during photobaseline on Day 0 against Day 3 revealed a significant reduction in dystonic crises (p=0.0312). Photoinduced dystonic crises during DBS on Day 0 were reduced on Day 2 (p=0.0312) and Day 3 (p=0.0312). Photoinduced dystonic crises immediately post-DBS on Day 0 were decreased on Day 1 (p=0.0312), Day 2 (p=0.0312), and Day 3 (p=0.0312). For 30 minutes post-DBS, photoinduced crises were reduced on Day 1 (p=0.0312) and Day 3 (p=0.0312). For 1 hour post-DBS, photoinduced crises decreased on Day 1 (p=0.0156), Day 2 (p=0.0156), and Day 3 (p=0.0156). For all statistics: n=6 animals, Wilcoxon matched-pairs signed rank test, two-tailed, error bars are defined as standard error of the mean.