Background Diffusion MRI is a non-invasive imaging technique to investigate white matter microstructure and has previously shown microstructural changes in the corpus callosum of Huntington’s disease (HD) patients. However, as HD is associated with grey and white matter atrophy, cerebrospinal fluid (CSF) may cause partial volume artefacts in the diffusion MRI metrics, leading to misleading results. CSF contamination is most problematic in regions close to the ventricles, such as the corpus callosum.

Aim To model the effects of partial volume contamination on diffusion MRI metrics in the corpus callosum of HD patients, and mice.

Methods The effect of CSF contamination was tested in diffusion MRI images (30 DW / 3 B0 directions, b = 1000 s/mm2) acquired in both a patient cohort (12 pre- and early-symptomatic HD patients, 8 healthy age, gender and education matched controls) and a mouse model of HD (21 HdhQ150 knock-in male mice / 23 age-matched wild-type mice). Partial volume correction was performed post-hoc (Pasternak et al, 2009), tractography was performed and mean diffusion tensor-based parameters obtained for the corpus callosum.

Results In the mice, there were no differences in the diffusion MRI metrics at 7 months old prior to symptom onset. At 19-months, there was a significant difference in mean diffusivity in the corpus callosum between the HD and wild-type mice, suggesting white matter microstructural changes occur after symptom onset. Crucially, this difference was no longer significant after correcting for CSF contamination. Analysis of the patient data is on-going.

Conclusions These findings suggest researchers should be cautious when interpreting diffusion MRI results in HD, and should perform appropriate corrections for CSF contamination.

• Download figure
• Open in new tab
• Download powerpoint

Abstract E19 Figure 1

Corpus callosum reconstruction in the human brain from a saggital view (top); Corpus callosum reconstruction in the mouse brain from a coronal view (bottom)