Determinants of gait dystonia severity in cerebral palsy

Quantifying Arm and Leg Movements in 3-Month-Old Infants Using Pose Estimation: Proof of Concept

BACKGROUND

Pose estimation (PE) has the promise to measure pediatric movement from a video recording. The purpose of this study was to quantify the accuracy of a PE model to detect arm and leg movements in 3-month-old infants with and without (TD, for typical development) complex congenital heart disease (CCHD).

METHODS

Data from 12 3-month-old infants (N = 6 TD and N = 6 CCHD) were used to assess MediaPipe's full-body model. Positive predictive value (PPV) and sensitivity assessed the model's accuracy with behavioral coding.

RESULTS

Overall, 499 leg and arm movements were identified, and the model had a PPV of 85% and a sensitivity of 94%. The model's PPV in TD was 84% and the sensitivity was 93%. The model's PPV in CCHD was 87% and the sensitivity was 98%. Movements per hour ranged from 399 to 4211 for legs and 236 to 3767 for arms for all participants, similar ranges to the literature on wearables. No group differences were detected.

CONCLUSIONS

There is a strong promise for PE and models to describe infant movements with accessible and affordable resources-like a cell phone and curated video repositories. These models can be used to further improve developmental assessments of limb function, movement, and changes over time.

Preterm birth is associated with dystonic features and reduced cortical parvalbumin immunoreactivity in mice

BACKGROUND

Preterm birth is a common cause of dystonia. Though dystonia is often associated with striatal dysfunction after neonatal brain injury, cortical dysfunction may best predict dystonia following preterm birth. Furthermore, abnormal sensorimotor cortex inhibition is associated with genetic and idiopathic dystonias. To investigate cortical dysfunction and dystonia following preterm birth, we developed a new model of preterm birth in mice.

METHODS

We induced preterm birth in C57BL/6J mice at embryonic day 18.3, ~24 h early. Leg adduction variability and amplitude, metrics we have shown distinguish between dystonia from spasticity during gait in people with CP, were quantified from gait videos of mice. Parvalbumin-positive interneurons, the largest population of cortical inhibitory interneurons, were quantified in the sensorimotor cortex and striatum.

RESULTS

Mice born preterm demonstrate increased leg adduction amplitude and variability during gait, suggestive of clinically observed dystonic gait features. Mice born preterm also demonstrate fewer parvalbumin-positive interneurons and reduced parvalbumin immunoreactivity in the sensorimotor cortex, but not the striatum, suggesting dysfunction of cortical inhibition.

CONCLUSIONS

These data may suggest an association between cortical dysfunction and dystonic gait features following preterm birth. We propose that our novel mouse model of preterm birth can be used to study this association.

IMPACT

Mouse models of true preterm birth are valuable for studying clinical complications of prematurity. Mice born preterm demonstrate increased leg adduction amplitude and variability during gait, suggestive of clinically observed dystonic gait features. Mice born preterm demonstrate fewer parvalbumin-positive interneurons and reduced parvalbumin immunoreactivity in the sensorimotor cortex, suggesting dysfunction of cortical inhibition. Mice born preterm do not demonstrate changes in parvalbumin immunoreactivity in the striatum. Cortical dysfunction may be associated with dystonic gait features following preterm birth.

Standardized clinical data capture to describe cerebral palsy

Objective

To describe a standardized methodology for capturing clinically valuable information on young people with cerebral palsy (CP) from caregivers and clinicians during routine clinical care.

Methods

We developed a caregiver-facing intake form and clinician-facing standardized note template and integrated both into routine clinical care at a tertiary care CP center ( https://bit.ly/CP-Intake-Methodology ). We extracted this caregiver and clinician-entered data on people with an ICD10 diagnosis of CP seen between 3/22/23 and 12/28/23. We used this data to describe how CP manifests in this group and which medical features affected the odds of walking, oral feeding, and speech by age 5.

Results

Of 686 visits, 663 (97%) had caregiver- and clinician-entered data and 633 had a clinician-confirmed CP diagnosis (mean age 9.1, 53.4% Male, 78.5% White). It was common to have quadriplegia (288/613, 47.0%), both spasticity and dystonia (257/632, 40.7%), walk independently (368/633, 58.1%), eat all food and drink safely by mouth (288/578, 55.9%), and produce understandable speech (249/584, 42.6%). Cortical grey matter injury and duration of initial critical care unit stay affected the odds of walking, oral feeding, and speech (binary logistic regression, p<0.001).

Conclusions

We comprehensively captured caregiver and clinician-entered data on 97% of people seen in a tertiary care CP Center and used this data to determine medical features affecting the odds of three functional outcomes. By sharing our methodology, we aim to facilitate replication of this dataset at other sites and grow our understanding of how CP manifests in the US.

Article summary

Using caregiver and clinician-entered data on people seen in a tertiary-care CP center, we determined medical features affecting the odds of three functional outcomes.

What’s known on this subject

Detailed CP characterization can be limited if using population-based registries and retrospective chart review alone, including limited data on recently validated functional classification systems for CP.

What this study adds

We comprehensively captured caregiver and clinician-entered data on 97% of people seen in our CP Center to describe how CP manifests and show that cortical injury and initial ICU stay duration affect the odds of walking, oral feeding, and speech.

Contributors Statement

Susie Kim helped design the study, aggregated data, carried out data analyses, and critically reviewed and revised the manuscript.Kelsey Steffen helped conceptualize and design the study and critically reviewed and revised the manuscript.Lauren Gottschalk, Jennifer Miros, Katie Leger, Amy Viehoever, and Karen Taca helped design the study and critically reviewed and revised the manuscript.Bhooma Aravamuthan conceptualized and designed the study, supervised data collection and analysis, drafted the initial manuscript, and critically reviewed and revised the manuscript.

A Novel Video-Based Methodology for Automated Classification of Dystonia and Choreoathetosis in Dyskinetic Cerebral Palsy During a Lower Extremity Task

BACKGROUND

Movement disorders in children and adolescents with dyskinetic cerebral palsy (CP) are commonly assessed from video recordings, however scoring is time-consuming and expert knowledge is required for an appropriate assessment.

OBJECTIVE

To explore a machine learning approach for automated classification of amplitude and duration of distal leg dystonia and choreoathetosis within short video sequences.

METHODS

Available videos of a heel-toe tapping task were preprocessed to optimize key point extraction using markerless motion analysis. Postprocessed key point data were passed to a time series classification ensemble algorithm to classify dystonia and choreoathetosis duration and amplitude classes (scores 0, 1, 2, 3, and 4), respectively. As ground truth clinical scoring of dystonia and choreoathetosis by the Dyskinesia Impairment Scale was used. Multiclass performance metrics as well as metrics for summarized scores: absence (score 0) and presence (score 1-4) were determined.

RESULTS

Thirty-three participants were included: 29 with dyskinetic CP and 4 typically developing, age 14 years:6 months ± 5 years:15 months. The multiclass accuracy results for dystonia were 77% for duration and 68% for amplitude; for choreoathetosis 30% for duration and 38% for amplitude. The metrics for score 0 versus score 1 to 4 revealed an accuracy of 81% for dystonia duration, 77% for dystonia amplitude, 53% for choreoathetosis duration and amplitude.

CONCLUSIONS

This methodology study yielded encouraging results in distinguishing between presence and absence of dystonia, but not for choreoathetosis. A larger dataset is required for models to accurately represent distinct classes/scores. This study presents a novel methodology of automated assessment of movement disorders solely from video data.

Evolving understanding of CP phenotypes: the importance of dystonia

Cerebral palsy (CP) is the core neurodevelopmental disorder affecting movement. Several distinct movement disorders can occur in people with cerebral palsy. Dystonia is a movement disorder that causes non-velocity-dependent hypertonia and/or abnormal, often repetitive, twisting movements, and/or postures. Dystonia occurs more frequently in patients with CP than has been recognized previously, and is treated differently than other aspects of CP. Dystonia is an important cause of chronic pain, hospitalization, and musculoskeletal complications. We describe recent advances in dystonia diagnosis in patients with cerebral palsy and highlight focus areas for ongoing research and clinical care. IMPACT: Dystonia is a movement disorder that is more common in people with cerebral palsy (CP) than previously thought. Dystonia contributes to hospitalization, chronic pain, and complications in CP patients. People with dystonic CP require different tools to diagnose and treat their condition. We summarize current state of the art in dystonia in CP and identify areas of focus for future work.

Mice born preterm develop gait dystonia and reduced cortical parvalbumin immunoreactivity

Preterm birth leading to cerebral palsy (CP) is the most common cause of childhood dystonia, a movement disorder that is debilitating and often treatment refractory. Dystonia has been typically associated with dysfunction of striatal cholinergic interneurons, but clinical imaging data suggests that cortical injury may best predict dystonia following preterm birth. Furthermore, abnormal sensorimotor cortex inhibition has been found in many studies of non-CP dystonias. To assess the potential for a cortical etiology of dystonia following preterm birth, we developed a new model of preterm birth in mice. Noting that term delivery in mice on a C57BL/6J background is embryonic day 19.1 (E19.1), we induced preterm birth at the limits of pup viability at embryonic day (E) 18.3, equivalent to human 22 weeks gestation. Mice born preterm demonstrate display clinically validated metrics of dystonia during gait (leg adduction amplitude and variability) and also demonstrate reduced parvalbumin immunoreactivity in the sensorimotor cortex, suggesting dysfunction of cortical parvalbumin-positive inhibitory interneurons. Notably, reduced parvalbumin immunoreactivity or changes in parvalbumin-positive neuronal number were not observed in the striatum. These data support the association between cortical dysfunction and dystonia following preterm birth. We propose that our mouse model of preterm birth can be used to study this association and potentially also study other sequelae of extreme prematurity.

Caregiver descriptions of dystonia in cerebral palsy

OBJECTIVE

To determine how caregivers describe dystonia in people with cerebral palsy (CP).

METHODS

In this prospective cohort study, paper surveys were administered to caregivers between September 7, 2021 and October 28, 2021 during CP Center visits at a large tertiary care center. Caregivers were asked to describe involuntary movements triggered by voluntary movement or triggered by tactile stimulation in the people with CP they cared for. Their CP Center medical provider separately assessed people with CP for dystonia. Movement features described exclusively by caregivers of people with CP and dystonia were determined using conventional content analysis.

RESULTS

113 caregivers responded on behalf of 56 people with and 57 people without dystonia. If caregivers noted that both voluntary movement and tactile stimulation triggered involuntary movements, that had a 92% positive predictive value for a dystonia diagnosis. Movement features exclusively described in people with CP and dystonia included: (1) stiffening, tensing, or tightening (15% of respondents); (2) involvement of the head (10%), torso (5%), or feet (5%); and (3) triggers of stretching (12.5%), excitement (5%), or transfers (5%).

INTERPRETATION

In addition to a thorough exam, asking caregivers of people with CP to describe involuntary movements triggered by voluntary movement or tactile stimulation may inform clinical dystonia diagnosis.

Upper Extremity Dystonia Features in People With Spastic Cerebral Palsy

Background and Objectives

Dystonia in cerebral palsy (CP) is debilitating and common, but underdiagnosed, especially when coexistent with spasticity. With dedicated research-based assessment, dystonia is found in most people with spastic CP but is only clinically diagnosed in the minority. To begin addressing the high rates of dystonia underdiagnosis in this population, we determined the key feature experts use to assess upper extremity dystonia in people with spastic CP.

Methods

In this prospective cohort study, 3 pediatric movement disorder specialists assessed upper extremity dystonia in neurologic examination videos of people with spastic CP and isolated periventricular leukomalacia (PVL) on brain MRI (i.e., those with a brain injury pattern typical for spastic CP). Dystonia severity was rated using the 10-point Global Dystonia Severity Rating Scale, first by each expert independently and then again after consensus-building discussion. Conventional content analysis of these discussions revealed salient features ("codes") that experts used to assess upper extremity dystonia. Code frequency distributions were compared between dystonia severity categories using χ2 tests.

Results

We identified 96 people with spastic CP with isolated PVL on brain MRI seen in the St. Louis Children's Hospital CP Center between 2005 and 2018. Of them, 26 people were able and willing to be recorded while doing a standardized set of upper extremity examination maneuvers (age 4-25 years; 28% nonambulatory, 77% White). When assessing their videos, experts cited the "hand" less often and "shoulder" more often with increasing dystonia severity (p < 0.005, χ2 test). "Mirror movements" and the "hand open/close" examination maneuver were cited significantly more frequently in videos when experts were attempting to distinguish between no dystonia and mild dystonia (p < 0.005).

Discussion

Expert clinicians use distinct movement features to assess upper extremity dystonia in people with spastic CP and PVL. Attention to involuntary shoulder (vs hand) movements can help gauge dystonia severity. Differentiation between mirror movements and dystonia, particularly during the hand open/close examination maneuver, may help identify mild dystonia. These results can help guide upper extremity dystonia assessment in people with spastic CP, thus potentially helping mitigate dystonia underdiagnosis.

Chronic striatal cholinergic interneuron excitation induces clinically-relevant dystonic behavior in mice

Dystonia is common, debilitating, often medically refractory, and difficult to diagnose. The gold standard for both clinical and mouse model dystonia evaluation is subjective assessment, ideally by expert consensus. However, this subjectivity makes translational quantification of clinically-relevant dystonia metrics across species nearly impossible. Many mouse models of genetic dystonias display abnormal striatal cholinergic interneuron excitation, but few display subjectively dystonic features. Therefore, whether striatal cholinergic interneuron pathology causes dystonia remains unknown. To address these critical limitations, we first demonstrate that objectively quantifiable leg adduction variability correlates with leg dystonia severity in people. We then show that chemogenetic excitation of striatal cholinergic interneurons in mice causes comparable leg adduction variability in mice. This clinically-relevant dystonic behavior in mice does not occur with acute excitation, but rather develops after 14 days of ongoing striatal cholinergic interneuron excitation. This requirement for prolonged excitation recapitulates the clinically observed phenomena of a delay between an inciting brain injury and subsequent dystonia manifestation and demonstrates a causative link between chronic striatal cholinergic interneuron excitation and clinically-relevant dystonic behavior in mice. Therefore, these results support targeting striatal ChIs for dystonia drug development and suggests early treatment in the window following injury but prior to dystonia onset.

One Sentence Summary

Chronic excitation of dorsal striatal cholinergic interneuron causes clinically-relevant dystonic phenotypes in mice.