Dystonia is a form of elevated involuntary muscle tone that substantially impairs function and quality of life in children with cerebral palsy (CP). Multiple neurosurgical approaches are used to reduce dystonia, most commonly intrathecal baclofen (ITB) for moderate to severe generalized dystonia. Ventral-dorsal rhizotomy (VDR) is increasingly used, but outcome data remain sparse. Reported evidence for ITB effectiveness is limited by patient heterogeneity and prior study design constraints. The PREDICT-Dystonia protocol aims to prospectively evaluate how ITB and VDR affect quantitative and functionally meaningful dystonia as well as related domains such as spasticity, motor function, mood and quality of life.
This is a prospective observational study enrolling 52 children with CP and moderate to severe dystonia, defined here by a Barry Albright Dystonia Scale (BADS) score greater than 14. Participants will be recruited at three US pediatric tertiary centers: Texas Children's Hospital (Houston), Lurie Children's Hospital (Chicago) and C.S. Mott Children's Hospital (Ann Arbor). Treatments are assigned according to clinical practice rather than randomization; planned group sizes reflect anticipated treatment decisions with ITB (n=35) and VDR (n=17).
Eligible children meet the clinical definition of CP with dystonia meeting the BADS threshold (>14). Primary caregivers provided informed consent for participation; participants with capacity were invited to provide assent. Further inclusion and exclusion specifics were not reported in the source document beyond the dystonia severity requirement and the planned enrollment numbers.
The study observes two neurosurgical interventions used in routine clinical care:
Detailed procedural protocols, perioperative care, device programming (for ITB) and surgical technique variations were not described in the source text and therefore are not reported here.
Primary outcome:
Secondary outcomes:
Additional data collection:
A multidimensional or composite measure integrating relevant domains will also be constructed as part of secondary analyses.
The primary analysis will use repeated measures models to estimate change from baseline in BADS scores at 3, 6 and 12 months following the intervention. The same repeated-measures approach will be applied to secondary outcome measures to quantify temporal changes.
Planned subgroup analyses will investigate characteristics of children who show a positive response to either ITB or VDR. The protocol intends to use clinical, imaging and other baseline measures to identify predictors of beneficial response and to inform patient selection strategies.
Details on model covariates, handling of missing data, power calculations and prespecified thresholds for clinically meaningful change were not provided in the source summary and are therefore not reported here.
Secondary analyses will mirror the primary repeated-measures strategy for other outcome domains and will evaluate whether particular child characteristics or brain injury patterns are associated with better outcomes after ITB or VDR. The study will also develop and evaluate a composite or multidimensional outcome measure that captures dystonia alongside function and patient-reported domains to better reflect clinically meaningful benefit.
Primary ethics approval for the study was provided by the Baylor College of Medicine Institutional Review Board (H-54449). Primary caregivers of all participants provided consent; participants with capacity were asked to assent. Study investigators plan to disseminate results via presentations at scientific conferences and through open access publication in peer-reviewed venues.
The study is registered under ClinicalTrials.gov identifier NCT06606574.