The accuracy, reproducibility and reliability of CT-guided functional stereotaxis using Laitinen's Stereoadapter(TM) were assessed in 19 patients who underwent 14 thalamotomies and 5 pallidotomies for the treatment of essential tremor or Parkinson's disease. The coordinates of the target were obtained from a CT study with the Stereoadapter. At surgery, the Stereoadapter was remounted to the patient's head and the CT coordinates were transferred to Laitinen's Stereoguide without ventriculography. 3-12 months after surgery, a postoperative CT study with Stereoadapter was done. The coordinates of the stereotactic lesion were compared to the preoperative target coordinates. A new similar target was simulated on the postoperative CT study and its coordinates were compared to those of the preoperative target. The coordinates of the lesion versus preoperative target showed a maximal difference of 3.5 mm for the lateral (x-) coordinate, 3.5 mm for the anteroposterior (y-) coordinate, and 3.75 mm for the height (z-) coordinate. The differences were statistically significant only for the x-coordinate: The lesions lay 1.25 mm, on the average, medial to the preoperative target. This medial displacement of the lesions was presumably due to a 10% inborn magnification error of the CT picture for the lateral direction. The maximal differences between the coordinates of the 'new' simulated target on the postoperative CT and those of the preoperative target were 2.5 mm for the x-, 2.25 mm for the y- and 3.75 mm for the z-coordinates. The differences were not statistically significant.
