Premium
P3‐412: 123 I‐MIBG MYOCARDIAL SCINTIGRAPHY CAN BE A MARKER OF LANGUAGE FUNCTION IN DE NOVO PARKINSON'S DISEASE
Author(s) -
Murakami Hidetomo,
Owan Yoshiyuki,
Futamura Akinori,
Saito Yu,
Kuroda Takeshi,
Ono Kenjiro
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.2018.06.1775
Subject(s) - scintigraphy , medicine , cognition , correlation , cognitive test , cardiology , audiology , nuclear medicine , psychology , psychiatry , geometry , mathematics
Background:Current methods of amyloid-PET interpretation fail to identify early phases of amyloid deposition. Grothe et al. recently used Florbetapir-PET data from the ADNI cohort to develop a hierarchical stage model of PET-evidenced amyloid deposition (Fig. 1) resembling the estimation reported in neuropathologic studies. This in vivo hierarchical stage allowed classifying over 95% of the individual amyloid deposition profiles into one of four amyloid stages (Grothe et al., Neurology, 2017). Here we evaluated the replicability of Grothe in-vivo amyloid staging in an independent cohort of the INSIGHT-preAD study. We further explored potential benefits of this in-vivo amyloid staging approach for predicting incipient cognitive decline in this preclinical cohort.Methods:The monocentric INSIGHT-preAD cohort includes florbetapir-PET data from 318 cognitively intact older individuals with subjective memory complaints (SMC). All individuals underwent extensive neuropsychological testing at baseline, and a subset (N1⁄4265) was repeatedly tested at 6-months intervals for 2 years of follow-up. After initial pre-processing of the Florbetapir-PET data, we projected it into the previously proposed four-stage model of amyloid progression. Associations between in-vivo amyloid stage and cognitive decline were assessed cross-sectionally using ANCOVA, as well as longitudinally using a latent class growth modeling (LCGM) approach. Results obtained using the regional staging model were compared to the conventional dichotomization based on a global signal cutoff (SUVRcereb 1⁄4 1.1). Results: 38.7% of individuals were identified as having detectable amyloid load, and only 6 (4.9%) of these violated the proposed regional hierarchy. Compared to conven-