Electrical Stimulation of the Anterior Thalamus for Epilepsy: Clinical Outcome and Analysis of Efficient Target

OBJECTIVE

Deep brain stimulation (DBS) of the anterior thalamic complex (ANT) is an adjunctive therapy for pharmacoresistant epilepsy. To define the most efficient target in DBS for epilepsy, we investigate clinical data, position of leads, usability of atlas data compared to electric field modeling based on programming parameters.

METHODS

Data from ten consecutive patients who underwent ANT-DBS were analyzed. The mammillothalamic tract (MTT), an internal landmark for direct stereotactic targeting, was segmented from MRI. Centers of stimulation were determined and their positions relative to ventricles and the MTT were analyzed. Two 3D thalamus atlases were transformed to segmented patient's thalami and proportions of activated nuclei were calculated.

RESULTS

Our data indicate higher response rates with a center of stimulation 5 mm lateral to the wall of the third ventricle (R² for reduction of focal seizure frequency and distance to the wall of the third ventricle = 0.48, p = 0.026). For reduction of focal seizures, a strong positive correlation with the dorsal distance to the midcommissural plane was found (R² = 0.66, p = 0.004). In one 3D atlas, stimulation of internal medullary lamina (IML) correlated strongly positive with response rates, which, however, did not reach statistical significance (R² = 0.69, p = 0.17 for tonic-clonic seizures). All electrical fields covered the diameter of the MTT. The position of the MTT in the thalamus was highly variable (range: x-coordinate 4.0 to 7.3 mm, y-coordinate -1.3 to 5.1 mm in AC-PC space).

CONCLUSIONS

The distance of the active contact to the lateral wall of the third ventricle, MTT and the ventrodorsal distance to midcommissural plane appear to be relevant for optimal target planning. For reduction of focal seizure frequency, we found best response rates with a center of stimulation 5 mm lateral to the wall of the third ventricle, and a lead tip 10 mm dorsal of the midcommissural plane.

Comparison of three transmittance oximeters

Estimates of oxygenation from three transmittance oximeters, Hewlett-Packard 47201A, Nellcor, and Biox III, were obtained simultaneously in ten healthy human subjects during isocapnic progressive hypoxia and under conditions of episodic hypoxemia. In the oxygen saturation range studied (approximately 95-75%), results showed that values from the Nellcor finger oximeter were 2-4% lower than values from the Hewlett-Packard ear oximeter. Values from the Biox III oximeter, using either the ear or finger probe, were 3-7% higher than those from the Hewlett-Packard, and the biggest discrepancies were seen as oxygen saturation fell below 80%. This often resulted in an estimation of the ventilatory response to hypoxia as measured using the Biox III oximeter that was higher than estimates using the Nellcor or the Hewlett-Packard (P less than 0.005). Hypoxic ventilatory responses using the Nellcor and the Hewlett-Packard oximeters gave similar values. During rapidly changing oxygenation, the distribution of values for oxygen saturation were significantly different for the three oximeters. For example, the median of the distribution for Nellcor values was 89 +/- 1.5% (mean +/- SEM); for the Hewlett-Packard, it was 92 +/- 1.3%; and for the Biox III, it was 93 +/- 0.8% (finger) and 95 +/- 1.6% (ear). The authors conclude that differences between the three non-invasive transmittance oximeters will affect clinical measurement of the ventilatory responsiveness to hypoxia and the estimation of the stimulus for hypoxic complications in respiratory diseases.