Premium
IC‐P‐078: Correlation between metabolic and CSF biomarkers in Alzheimer's disease patients with early cognitive decline
Author(s) -
Jensterle Luka,
Tomše Petra,
Rot Uroš,
Emeršič Andreja,
Pirtošek Zvezdan,
Tang Chris C.,
Eidelberg David,
Trošt Maja
Publication year - 2015
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.2015.06.099
Subject(s) - dementia , medicine , montreal cognitive assessment , disease , correlation , cognition , cognitive impairment , pathology , psychology , cardiology , psychiatry , geometry , mathematics
P1-157) and predicted conversion from mild cognitive impairment (MCI) to AD (AAIC 2013, P3-085). However, it is not known which biological phenomena of the disease process texture reflect. The purpose of this study was to investigate a potential relationship between texture and tau-mediated neuronal injury as reflected by reduced glucose metabolism in FDG-PET.Methods:The study dataset consisted of the 215 baseline T1-weighted structural MRI scans from the “complete annual year 2 visit” 1.5T standardized Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset [62 normal controls (NC), 101 MCI subjects, and 52 AD patients] that had an associated FDG-PET measurement of hippocampus metabolic rate of glucose (MRglc) available from the Center for Brain Health, NYU School of Medicine, New York with a maximum of 60 days between scans. MRI analysis consisted of segmenting the hippocampi using cross-sectional FreeSurfer (v.5.1.0), computation of the hippocampal fraction (HF, hippocampal volume divided by intra-cranial volume), and texture scoring of the hippocampus using our in-house method. Left and right hippocampal MRglc were downloaded directly from the ANDI website and subsequently averaged to obtain a single score. Results:A scatter plot revealed a linear relationship between texture and MRglc (see figure). Three linear regression models with different combinations of MRI biomarkers as covariates were fitted, all with hippocampal MRglc as outcome, and with age and sex as additional covariates (see table). Both HF (model 1) and texture (model 2) explained hippocampal MRglc, and the model using texture resulted in a better overall fit. Adding both HF and texture as covariates (model 3) resulted in an even better overall model fit, and both texture and HF had a significant contribution. Conclusions:There was a significant relationship between hippocampal MRI texture and hippocampal glucose metabolism demonstrating that texture potentially is related to the synaptic dysfunction that accompanies neurodegeneration in AD. The relationship persisted when controlling for volume, i.e., texture carries additional information. Interestingly, the combination of volume and texture resulted in an even better overall model fit. IC-P-078 CORRELATION BETWEEN METABOLIC AND CSF BIOMARKERS IN ALZHEIMER’S DISEASE PATIENTS WITH EARLY COGNITIVE DECLINE Luka Jensterle, Petra Tom se, Uro s Rot, Andreja Emer si c, Zvezdan Pirto sek, Chris C. Tang, David Eidelberg, Maja Tro st, University Medical Centre Ljubljana, Ljubljana, Slovenia; The Feinstein Institute for Medical Research, Manhasset, NY, USA. Contact e-mail: maja. trost@kclj.si