Functional hemispherectomy: postoperative motor state and correlation to preoperative DTI

Functional hemispherectomy (FH) is an infrequent method to reduce seizure frequency in patients with intractable epilepsy. The risk that hemispherotomy injures brain structures involved in residual motor function is challenging to predict. Our purpose was to evaluate MR diffusion tensor imaging (DTI) to preoperatively assess residual ipsilateral motor function prior to FH.

We applied DTI in 34 patients scheduled for FH to perform fiber tracking in healthy and damaged hemispheres of the corticospinal tracts (CSTs) and of the corpus callosum. We assessed the CSTs and the commissural fibers for streamline count, for fractional anisotropy (FA), and for respective ratios (affected/unaffected side). We correlated these DTI values to post-to-prior changes of muscle strength and evaluated their diagnostic accuracy.

FA of the affected CSTs and of commissural fibers was significantly higher in patients with postoperative loss of muscle strength compared to patients without (p = 0.014 and p = 0.008). In contrast, CST FA from healthy hemispheres was not different between both groups. Ratios of streamline counts and FA from CSTs were higher in patients with postoperative reduced muscle strength compared to those without (1.14 ± 0.22 vs. 0.58 ± 0.14, p = 0.040; 0.93 ± 0.05 vs. 0.74 ± 0.03, p = 0.003). CSTs' normalized FA ratio greater than -0.085 predicted loss of muscle strength with 80 % sensitivity and 69.6 % specificity.

Preoperative tracking of the CST and of commissural fibers contributes to the prediction of postoperative motor outcome after functional hemispherectomy.