Background: Deep brain stimulation of the globus pallidus interna (GPi-DBS) has been approved as a surgical treatment for severe dystonia.
Objective: To study and compare the efficacy of GPi-DBS for various types of dystonia, and to identify predictive factors of
surgical outcome.
Material and Method: Fifteen dystonic patients who received bilateral GPi-DBS were included in the study. Clinical outcomes
were evaluated by Burke-Fahn-Marsden Dystonia Rating Scale (BFMDRS).
Results: Seven cases with primary dystonia had consistent dramatic improvement in pain, motor and bulbar functions though all were negative for DYT1 gene mutation. The mean improvement of BFMDRS was 71.02%. Eight with secondary dystonia had inconsistent improvement of BFMDRS with a mean improvement of 30.49%. Post-stroke dystonia and tardive dystonia had significant sustained improvement of 100% and 97%, respectively. Secondary dystonia from traumatic brain injury had modest sustained improvement of 47%. Dystonia secondary to cerebral palsy, Huntington’s disease and CNS infection, showed no improvement. Patients with segmental dystonia improved greater than generalized dystonia regardless of primary or secondary type (82.45% and 20.49%, respectively). Patients whose main symptom was mobile dystonia had more improvement than mixed dystonia and fixed dystonia.
Conclusion: Primary dystonia was a strong, good predictive factor. Secondary dystonia from stroke and tardive disorder,
segmental and mobile dystonia seemed to be good predictive factors. Dystonia, secondary to cerebral palsy, Huntington’s
disease and CNS infection was a poor predictive factor.

Keywords: Dystonia, Globus pallidus internus, Deep brain stimulation