Causes of failure of pallidal deep brain stimulation in cases with pre-operative diagnosis of isolated dystonia.

INTRODUCTION Pallidal deep brain stimulation (GPi-DBS) is an effective therapy for isolated dystonia, but 10-20% of patients show improvement below 25-30%. We here investigated causes of insufficient response to GPi-DBS in isolated dystonia in a cross-sectional study. METHODS Patients with isolated dystonia at time of surgery, and <30% improvement on the Burke-Fahn-Marsden dystonia-rating-scale (BFMDRS) after ≥6 months of continuous GPi-DBS were videotaped ON and OFF stimulation, and history, preoperative videos, brain MRI, medical records, stimulation settings, stimulation system integrity, lead location, and genetic information were obtained and reviewed by an expert panel. RESULTS 22 patients from 11 centres were included (8 men, 14 women; 9 generalized, 9 segmental, 3 focal, 1 bibrachial dystonia; mean (range): age 48.7 (25-72) years, disease duration 22.0 (2-40) years, DBS duration 45.5 (6-131) months). Mean BFMDRS-score was 31.7 (4-93) preoperatively and 32.3 (5-101) postoperatively. Half of the patients (n = 11) had poor lead positioning alone or in combination with other problems (combined with: other disease n = 6, functional dystonia n = 1, other problems n = 2). Other problems were disease other than isolated inherited or idiopathic dystonia (n = 5), fixed deformities (n = 2), functional dystonia (n = 3), and other causes (n = 1). Excluding patients with poor lead location from further analysis, non-isolated dystonia accounted for 45.5%, functional dystonia for 27.3%, and fixed deformities for 18.2%. In patients with true isolated dystonia, lead location was the most frequent problem. CONCLUSION After exclusion of lead placement and stimulation programming issues, non-isolated dystonia, functional dystonia and fixed deformities account for the majority of GPi-DBS failures in dystonia.