Premium
P1‐373: βETA‐AMYLOID AND WHITE MATTER LESIONS ARE INDEPENDENTLY ASSOCIATED WITH HIPPOCAMPAL ATROPHY AND REDUCED CORTICAL TEMPORAL THICKNESS
Author(s) -
Svenningsson Anna,
Stomrud Erik,
Palmqvist Sebastian,
Hansson Oskar
Publication year - 2018
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2018.06.381
Subject(s) - atrophy , temporal lobe , white matter , pathology , magnetic resonance imaging , cerebrospinal fluid , hyperintensity , hippocampal formation , pittsburgh compound b , brain size , neuroimaging , amyloid (mycology) , medicine , psychology , neurodegeneration , alzheimer's disease , neuroscience , disease , radiology , epilepsy
amyloid positive (A+) AD subjects from ADNI1/GO/2 were used as a common data set. Results: All four methods classified more cases (35-65%) as Typical than either Cortical (8-29%) or Limbic (12-23%) (Figure 1). The Cortical subtype had younger average age for all methods. 40% of subjects were assigned the same subtype by the Charil, Risacher and Young methods, and another 52% by 2/3 of those methods. 36% of subjects were assigned the same subtype by the Charil, Risacher and staged Young methods, and another 55% by 2/3 of those methods. 17% of subjects were Nby the Charil method and 28% by the staged Young method. Confusion matrices show the overlap in specific subtype assignments (Figure 2). The Cortical subtype exhibited significantly worse baseline executive function for all methods. However, the profiles of baseline and decline on memory qualitatively differed (Figure 3). Conclusions:The four algorithms identified incomplete overlap in subtype assignments and may identify slightly different cognitive phenotypes. Further investigation of how different atrophy subtyping algorithms compare is warranted.