Olfactory identification deficit predicts white matter tract impairment in Alzheimer's disease

Olfactory identification deficit (OID) has been associated with both aging and Alzheimer's disease (AD). In the context of an amnestic disorders, OID predicts conversion to AD. Neuroanatomical correlates could increase specificity and sensitivity and elucidate the mechanistic differences between OID in AD and aging. Cross-sectional analysis of white matter microstructural changes was performed using diffusion tensor imaging (DTI) and tract-based-spatial-statistics in amnestic mild cognitive impairment (aMCI), AD and normal controls (NC) in 66 subjects (26 AD, 15 aMCI, 25 NC). DTI 3-Tesla MRI scans were analyzed and subject level means for fractional anisotropy (FA), mean diffusivity (MD), radial and axial diffusivity (lambda D-1 and lambda D-2,D-3) were calculated. Linear regression models were applied using DTI markers as predictor and OID as outcome. OID was associated with increased lambda D-1 in aMCI and increased MD, lambda D-1 and lambda D-2,D-3 in AD. Voxel-wise analyses revealed widespread differences in all markers in AD. There were significant differences in lambda D-1 in aMCI, particularly in the olfactory tract. OID is correlated with microstructural white matter changes as early as in aMCI. This study may help elucidate the biological basis for olfactory impairment in Alzheimer's disease. Neuroanatomical correlates could help distinguish OID associated with AD and that associated with aging.