Voluntary saccade inhibition deficits correlate with extended white-matter cortico-basal atrophy in Huntington's disease

The ability to inhibit automatic versus voluntary saccade commands in demanding situations can be impaired in neurodegenerative diseases such as Huntington's disease (HD). These deficits could result from disruptions in the interaction between basal ganglia and the saccade control system. To investigate voluntary oculomotor control deficits related to the cortico-basal circuitry, we evaluated early HD patients using an interleaved pro- and anti-saccade task that requires flexible executive control to generate either an automatic response (look at a peripheral visual stimulus) or a voluntary response (look away from the stimulus in the opposite direction). The impairments of HD patients in this task are mainly attributed to degeneration in the striatal medium spiny neurons leading to an over-activation of the indirect-pathway thorough the basal ganglia. However, some studies have proposed that damage outside the indirect-pathway also contribute to executive and saccade deficits. We used the interleaved pro- and anti-saccade task to study voluntary saccade inhibition deficits, Voxel-based morphometry and Tract-based spatial statistic to map cortico-basal ganglia circuitry atrophy in HD. HD patients had voluntary saccade inhibition control deficits, including increased regular-latency anti-saccade errors and increased anticipatory saccades. These deficits correlated with white-matter atrophy in the inferior fronto-occipital fasciculus, anterior thalamic radiation, anterior corona radiata and superior longitudinal fasciculus. These findings suggest that cortico-basal ganglia white-matter atrophy in HD, disrupts the normal connectivity in a network controlling voluntary saccade inhibitory behavior beyond the indirect-pathway. This suggests that in vivo measures of white-matter atrophy can be a reliable marker of the progression of cognitive deficits in HD.

Protocol publication

[…] T1 images were analyzed through the Voxel-based morphometry technique implemented on FMRIB Software Library (FSL) (http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSLVBM). First, voxels that did not belong to the images of the cerebral tissue were eliminated using (Brain Extraction Tool) BET software. Afterwards, segmentation was done according to the type of tissue sampled (grey-matter, white-matter and cerebrospinal fluid) using FAST4. The images corresponding to the grey-matter were aligned to the MNI152 standard space by means of a linear followed by a non-linear co-registration in order to obtain optimal co-registration. The average of these co-registered images was obtained to generate the specific standard for this study. The grey-matter images were co-registered to this specific standard through a non-linear co-registration and local changes in expansion or contraction were corrected by modulation. A smoothing process was applied with a Gaussian isotropic kernel with a sigma of 4 mm. An independent two-sample t-test was applied through the general linear model correcting for multiple comparisons using randomized permutation methods (n = 10,000). We chose the appropriate smoothing (sigma 4 mm) for the Threshold-Free Cluster Enhancement (TFCE) based analysis. The family-wise error (FWE) was controlled and only corrected 1 – p values ≤ 0.095 were accepted. […]