There is a renewal of interest in surgical approaches including lesions and deep brain stimulation directed at motor subcorticofrontal loops. Bilateral lesioning presents a far greater risk of adverse effects, especially cognitive impairment. Furthermore, the main advantages of the stimulation procedure over lesioning are adaptability and reversibility of effects. The aim of this study was to assess the influence of bilateral stimulation of the subthalamic nucleus or internal globus pallidus on memory and executive functions in Parkinson's disease. Sixty‐two patients were assessed before and after 3 to 6 months of chronic bilateral stimulation of the subthalamic nucleus (n = 49) or internal globus pallidus (n = 13). The neuropsychological tests used were the Mattis Dementia Rating Scale, the Grober and Buschke Verbal Learning Test, the Wisconsin Card Sorting Test, category and literal fluency, graphic and motor series, the Stroop Test, and the Trail Making Test. Mood was evaluated by the Beck Depression Inventory. Only 4 of 25 cognitive variables were affected by deep brain stimulation. Under stimulation, performance improved for Parts A and B of the Trail Making Test, but there was a deterioration in literal and total lexical fluency. There was also a mild but significant improvement in mood. It may therefore be concluded that stimulation of the subthalamic nucleus or internal globus pallidus does not change the overall cognitive performance in Parkinson's disease and does not greatly affect the functioning of subcorticofrontal loops involved in cognition in humans. This relative absence of cognitive impairment in bilateral deep brain stimulation is likely because of the accurate positioning of the electrodes, allowing the effects of stimulation to be confined to sensorimotor circuits. Ann Neurol 1999;46:217–223

The beneficial effects of high-frequency stimulation result from a change in the firing pattern of cellular discharge and a blockade of the spontaneous overactivity of STN neurons.

The beneficial effects of subthalamic stimulation depend on (1) the criteria used for patient selection, (2) the precision with which the subthalamic nucleus is targeted (dependent on the 3-dimensional magnetic resonance imaging and the intraoperative electrophysiologic and clinical assessments), and (3) the long-term postoperative adjustment of stimulation variables.

The improvement of segmental akinesia and the observation of dyskinesias provoked by stimulation during the operation predict the best postoperative effects of bilateral STN stimulation on parkinsonian motor disability.