Stereotactic ventrolateral thalamotomy: is ventriculography necessary?

In the computed tomography/magnetic resonance imaging (CT/MRI) era, the need for ventriculography to perform ventrolateral thalamotomy accurately has been debated. We retrospectively compared CT/MRI-derived coordinates for ventrolateral thalamotomy with the final lesion coordinates that were determined by ventriculography and microelectrode recording in 74 thalamotomies performed from 1984 to 1994. The median three-dimensional distance between the CT/MRI-derived loci and the ventriculography/microelectrode loci was 4.7 mm (range, 1.0-11.7 mm). The techniques correlated least along the Y axis (median, -0.3 mm; range, -8.2 to 8.0 mm). Correlation along the X axis was most consistent (median, 0.5 mm; range, -4.2 to 5.0 mm). Since 1990, the CT/MRI-derived coordinates have been generated by a multimodality correlative imaging technique (MCIT). A comparison of thalamotomies performed with and without the MCIT revealed a significant improvement in the correlation of CT/MRI- and ventriculography/microelectrode-derived coordinates when the MCIT was employed. The greatest improvement was noted along the Y axis where the median absolute difference was reduced from 4.0 to 1.8 mm (P = 0.0001). The result was a statistically significant reduction in the median three-dimensional distance from 5.6 to 3.7 mm (P = 0.0007). The authors conclude that thalamotomies can be safely and effectively performed without ventriculography when the MCIT is employed and supported by neurophysiological monitoring.