Deep brain stimulation (DBS) of the sub-thalamic nucleus (STN) has evolved over the past decades as a mainstream therapy for advanced Parkinson's disease (PD). The classical procedure consists in STN indirect targeting based on stereotactic atlases or statistical coordinates in AC-PC (Anterior Commissure - Posterior Commissure) referential along with target control and correction by micro-electrode recordings (MER) and awake clinical testing. To avoid potential complications and patient discomfort related to current procedure, asleep surgery without this control process has become more and more performed, essentially thanks to the progress of neuroimaging allowing to STN visualization. However, it has been reported a relative inaccuracy between the "radiological" STN delimitated on several types of MRI sequences (T2, T2*, SWI) and the per-operative electrophysiological findings. As a result, there are currently many types of STN-DBS procedures, and the lack of standardization between techniques complicates the interpretation of postoperative results on anatomical, electrophysiological and clinical points of view. Furthermore, to date, it has not been proven that asleep surgery without MER and clinical controls is as effective as the standard procedure in a prospective controlled randomized clinical trial.
Investigators hypothesize that the clinical-based 18 landmarks STN target will be precise enough to allow to perform surgery under general anesthesia without MER correction, and accurate enough to achieve non inferior clinical results compared to what is usually done in each centre.
The main objective is to compare at one year, the % of motor improvement after PARKEO 2-targeting asleep DBS without intraoperative MER versus the targeting procedure using intraoperative MER by the UPRDRS 3 (Unified Parkinson's disease rating scale 3).