Divergent Longitudinal Propagation of White Matter Degradation in Logopenic and Semantic Variants of Primary Progressive Aphasia

2015 
Background: Clinico-pathological distinction of primary progressive aphasia (PPA) can be challenging at clinic presentation. In particular, cross-sectional neuroimaging signatures across the logopenic (lvPPA) and semantic (svPPA) variants are difficult to establish, with longitudinal profiles showing greater divergence. Objective: Assess longitudinal propagation of white matter degradation in lvPPA and svPPA to determine disease progression over time, and whether this reflects distinct underlying pathology. Method: A cohort of 27 patients with dementia (12 lvPPA; 15 svPPA) and 12 healthy controls were assessed at baseline and 1-year follow-up on the Addenbrooke’s Cognitive Examination-Revised and Sydney Language Battery. Diffusion weighted images were collected at both time-points and analyzed for longitudinal white matter change using DTI-TK and TBSS. Results: LvPPA patients showed a significant decline in naming and repetition, over 1 year, while svPPA patients declined in naming and comprehension. Longitudinal imaging revealed widespread bilateral degradation of white matter tracts in lvPPA over a 1-year period with early involvement of the left posterior inferior longitudinal fasciculus (ILF). SvPPA demonstrated focal left lateralized white matter degradation involving the uncinate fasciculus (UF) and anterior ILF, propagating to the right UF with disease progression. Conclusions: LvPPA and svPPA cohorts showed distinct longitudinal cognitive and white matter profiles. We propose differences in multi-centric and focal white matter dysfunction in lvPPA and svPPA, respectively, reflect underlying pathological differences. The clinical relevance of white matter degradation and mechanisms underlying disease propagation are discussed.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    31
    References
    34
    Citations
    NaN
    KQI
    []