ANALYSIS OF PALLIDOTOMY LESION POSITIONS USING 3-DIMENSIONAL RECONSTRUCTION OF PALLIDAL LESIONS, THE BASAL GANGLIA, AND THE OPTIC TRACT

OBJECTIVE: To assess the position of radiofrequency pallidotomy lesion s placed using microelectrode stimulation and cellular recordings in r elation to a stereotactically defined starting point. Radiofrequency l esion locations were also evaluated in relation to the putamen, poster ior limb of the internal capsule, and optic tract. METHODS: Magnetic r esonance images obtained from 23 patients with Parkinson's disease who underwent pallidotomy at the University of Kansas Medical Center were analyzed. Using computerized techniques, lesion positions in relation to the midcommissural point and a hypothetical starting point were de termined. Data segmentation and three-dimensional reconstruction of pa llidal lesions, the internal capsule, and the optic tract allowed asse ssment of lesion position in relation to internal anatomy. Clinical ou tcome of pallidotomy was assessed using both the Unified Parkinson's D isease Rating Scale and the Dementia Rating Scale. RESULTS: Pallidal l esions were usually placed anterior and dorsal to the stereotactically defined starting point. The position of pallidal lesions in the men w ere observed, in four trials, to be significantly more dorsal than the lesions in the women. The outer zone of the lesion was usually adjace nt to the internal capsule and the putamen and relatively close to the optic tract. The inner zone of the lesion was usually several millime ters removed from anatomic boundaries of the putamen, internal capsule , and optic tract. Patients achieved favorable outcomes, with reduced dyskinesias and ''off'' time and improvement of their Parkinsonian sym ptoms, as evidenced by clinical assessment, the Unified Parkinson's Di sease Rating Scale, and the Dementia Rating Scale. CONCLUSION: Microel ectrode stimulation and cellular recordings usually led to a final pal lidotomy lesion position that deviated from the stereotactically defin ed starting point. The pallidotomy lesions in the men were observed to be more dorsal than the lesions in the women. Clinical outcomes were not correlated with either lesion location relative to the starting po int or distances between the pallidal lesion and the putamen, internal capsule, or optic tract. Kinesthetically responsive cells may be loca lized generally more anterior and dorsal to the starting point (within the globus pallidus) and may be grouped variably from patient to pati ent in relation to other basal ganglia structures. Although the primar y lesion site is most likely within the sensorimotor region of the glo bus pallidus internus, the more dorsal lo cations of responsive cell g roups may indicate that some lesion sites may be localized within the globus pallidus externus.