A pooled meta-analysis of GPi and STN deep brain stimulation outcomes for cervical dystonia

Congratulations Drs. Takashi Tsuboi, Joshua K. Wong, Leonardo Almeida,  Christopher W. Hess,  Aparna Wagle Shukla,  Kelly D. Foote, Michael S. Okun and Adolfo Ramirez-Zamora, on the publication of  “A pooled meta-analysis of GPi and STN deep brain stimulation outcomes for cervical dystonia.”  This article was published in the January edition of the Journal of Neurology.



To analyze deep brain stimulation (DBS) outcomes in patients with cervical dystonia (CD), the relationships between motor and disability/pain outcomes, and the differences in outcomes between globus pallidus internus (GPi) and subthalamic nucleus (STN) DBS, and to identify potential outcome predictors.


A systematic literature search identified individual patient data of CD patients who underwent DBS and whose outcomes were assessed with the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS). Then, we performed a pooled meta-analysis on this cohort.


A review of 39 papers yielded 208 patients with individual TWSTRS scores and demographic information. At a mean follow-up period of 23.3 months after either GPi or STN DBS, the TWSTRS total (58.8%), severity (53.9%), disability (61.3%), and pain (46.6%) scores significantly improved compared to baseline status (all p < 0.001). There were no significant outcome differences between short-term (< 23.3 months) and long-term (≥ 23.3 months). The TWSTRS outcomes after GPi and STN DBS were comparable, whereas these two targets showed different adverse effect profiles. The rates of responders to DBS according to the TWSTRS total and severity (defined as ≥ 25% improvement) were both 89%. Regression analyses demonstrated motor benefits associated with disability improvement more than pain relief (R2 = 0.345 and 0.195, respectively). No clinically meaningful predictors for DBS outcomes were identified.


DBS improves motor symptoms, disability, and pain in CD patients and may provide sustained benefits over 2 years. GPi and STN appear to be equally effective targets with different adverse effect profiles.