Pharmacological and neurosurgical interventions for individuals with cerebral palsy and dystonia: a systematic review update and meta-analysis.

All evidence is rated low to very low certainty by GRADE; most studies are uncontrolled before-and-after designs at high risk of bias. The only RCT for ITB had 3-month follow-up, which the authors note may have been insufficient to reach optimal dosing, and its dystonia results conflicted with the non-randomized evidence showing improvement. Outcome measures varied widely across studies, preventing direct comparison and limiting meta-analysis in several cases. Minimal clinically important differences used for interpretation were established in primary dystonia populations, not in CP with dystonia, so their applicability is uncertain. Some participants had mixed dystonia and spasticity, meaning part of any observed improvement may reflect spasticity reduction rather than a dystonia-specific effect. Only English-language studies were included, potentially excluding relevant work. No studies were found for four of the ten interventions reviewed (oral baclofen, benzodiazepines, gabapentin, medical cannabis), leaving those options without any direct evidence in this population.

Funded by the Ontario Brain Institute. No author conflicts of interest are stated in the text provided.

Do not read the ITB and DBS findings as proof these interventions work for dystonia in CP. Every outcome is rated very low certainty, the evidence is drawn mostly from uncontrolled before-and-after studies, and the effect sizes for ITB and DBS come from different outcome measures and cannot be compared directly. The one ITB RCT showed little to no difference in dystonia, and its 3-month follow-up may have been too short to capture a real effect.