Deep brain stimulation of the internal globus pallidus in dystonia: target localisation under general anaesthesia

Deep brain stimulation (DBS) of the internal globus pallidus (Gpi) is an effective therapy for various types of dystonia. The authors describe their technical approach for securing appropriate placement of the stimulating electrodes within the Gpi under general anaesthesia, including MRI based individualised anatomical targeting combined with electrophysiological mapping of the Gpi using micro-recording (MER) as well as macrostimulation and report the subsequent clinical outcome and complications using this method.

We studied 42 patients (male-female ratio 25:17; mean age 43.6 years, range 9 to 74 years) consecutively operated at the Department of Neurosurgery, University Hospital Schleswig-Holstein, Campus Kiel, between 2001 - 2006. One patient underwent unilateral implantation after a right-sided pallidotomy 30 years before and strictly unilateral symptoms; all other implantations were bilateral. Two patients had repeat surgery after temporary removal of uni- or bilateral implants secondary to infection. Overall, 86 DBS electrodes were implanted. In 97% of the implantations, at least three microelectrodes were inserted simultaneously for MER and test stimulation. Initial anatomical targeting was based on stereotactic atlas coordinates and individual adaptation by direct visualisation of the Gpi on the stereotactic T2 or inversion-recovery MR images. The permanent electrode was placed according to the results of MER and test stimulations for adverse effects.

The average improvement from baseline in clinical ratings using either the Burke-Fahn-Marsden-Dystonia (BFMDRS) or Toronto-Western-Spasmodic-Torticollis (TWSTR) rating scale at the last post-operative follow-up (mean 16.4 ; range 3-48 months) was 64.72% (range 20.39 to 98.52%). The post-operative MRI showed asymptomatic infarctions of the corpus caudatus in three patients and asymptomatic small haemorrhages in the lateral basal ganglia in two patients. One patient died due to a recurrent haemorrhage which occurred three months after the operation. The electrodes were implanted as follows: central trajectory in 64%, medial trajectory in 20%, anterior in 9% and lateral dorsal trajectories in 3.5% each. The reduction in BFMDRS or TWSTR motor score did not differ between the group implanted in the anatomically defined (central) trajectory bilateral (-64.15%, SD 23.8) and the physiologically adopted target (uni- or bilateral) (-63.39%, SD 23.1) indicating that in both groups equally effective positions were chosen within Gpi for chronic stimulation (t-test, p > 0.4).

The described technique using stereotactic MRI for planning of the trajectory and direct visualisation of the target, intra-operative MER for delineating the boundaries of the target and macrostimulation for probing the distance to the internal capsule by identifying the threshold for stimulation induced tetanic contractions is effective in DBS electrode implantation in patients with dystonia operated under general anaesthesia. The central trajectory was chosen in only 64%, despite individual adaptation of the target due to direct visualisation of the Gpi in inversion recovery MRI in 43% of the patients, demonstrating the necessity of combining anatomical with neurophysiological information.