Video-EEG polygraphy in the clinical characterization of hyperkinetic movement disorders: a tertiary referral cohort

Background: Hyperkinetic movement disorders (MDs) in children, including tremor, myoclonus, dystonia, chorea, and tics, often present with overlapping or evolving clinical features, making classification based on observation alone challenging. Neurophysiological techniques such as video-EEG polymyography (P-VEEG) and somatosensory evoked potentials (SEPs) may provide objective markers to improve diagnostic accuracy, yet their application in pediatric populations remains limited. Objective: To describe the polygraphic characteristics of hyperkinetic MDs in a large pediatric cohort and to assess whether specific EMG patterns can support movement disorder classification beyond clinical inspection. Methods: Seventy-three patients with pediatric-onset hyperkinetic MDs underwent standardized P-VEEG recordings and SEPs at Besta Neurological Institute between January and October 2024. EMG activity was recorded from antagonist muscle pairs during rest, posture maintenance, and specific tasks according to age. Polygraphic findings were compared with clinical diagnoses. Associations between EMG patterns and clinical variables were analyzed using Chi-square tests and regression analyses. Results: Mean age at MD onset was 9.4 +/- 6.1 years and mean age at recording was 13.5 +/- 5.1 years. Polygraphic analysis identified four distinct EMG patterns: (i) regular rhythmic oscillatory pattern (RRO), consistent with tremor, in 40 patients (56%); (ii) repetitive myoclonic pattern (RM) in 11 (15%); (iii) complex mixed repetitive pattern (CR) in 9 (12%); and (iv) complex non-rhythmic pattern (CNR), encompassing chorea, dystonia, tics and other dyskinesias, in 11 (15%). Significant differences among repetitive patterns were observed in rhythmicity (p < 0.001), burst duration (p < 0.001), agonist-antagonist synchrony (p < 0.001), and SEP hyperexcitability (p < 0.03). RM was significantly associated with enhanced amplitude SEPs (p = 0.023), supporting a cortical origin of myoclonus in most cases. A significant association between clinical diagnosis and EMG pattern was found (p = 0.003), yet discrepancies between clinical and polygraphic classification was found in 22 cases (30%), particularly among tremor and myoclonus presentations. RM and CR were prevalent in older patients compared with CNR. Conclusion: P-VEEG proved to be an easily applicable tool providing valuable objective markers for the characterization of pediatric hyperkinetic MDs, revealing distinct EMG patterns that complement clinical evaluation and improve diagnostic precision. Systematic integration of neurophysiological assessment may enhance classification, diagnostic accuracy and help guide management in childhood movement disorders.