Predictors of deep brain stimulation outcome in tremor patients

Abstract Background Deep brain stimulation of the ventro-intermedius nucleus of the thalamus is an established treatment for tremor of differing etiologies but factors that may predict the short- and especially long-term outcome of surgery are still largely unknown. Methods We retrospectively investigated the clinical, pharmacological, electrophysiological and anatomical features that might predict the initial response and preservation of benefit in all patients who underwent deep brain stimulation for tremor. Data were collected at the following time points: baseline (preoperative), one-year post-surgery, and most recent visit. Tremor severity was recorded using the Fahn-Tolosa-Marin Tremor Rating Scale and/or the Unified Parkinson's Disease Rating Scale. Results A total of 52 patients were included in the final analysis: 31 with essential tremor, 15 with cerebellar tremor of different etiologies, and 6 with Parkinson's disease. Long-term success (mean follow-up duration 34.7 months, range 1.7–121.1 months) was reported in 63.5%. Predictors of long-term benefit were: underlying tremor etiology (best outcome in Parkinson's disease, worst outcome in cerebellar tremor); age at surgery (the older the better); baseline tremor severity (the greater the better); lack of response to benzodiazepines; a more anterior electrode placement and single-unit beta power (the greater the better). Conclusions Specific patients' features (including single unit beta activity) and electrode locations may predict the short- and long-term benefit of thalamic stimulation for tremor. Future prospective studies enrolling a much larger sample of patients are needed to substantiate the associations detected by this retrospective study.