Lateral Ventricle volume is associated with CDR and Hippocampus volume
Poster No:
2301
Submission Type:
Abstract Submission
Authors:
Gengsheng Chen1, Stephanie Doering1, Nelly Joseph-Mathurin1, Jingxia Liu1, Parinaz Massoumzadeh1, John Morris1, Tammie Benzinger1
Institutions:
1Washington University School of Medicine, St. Louis, MO
First Author:
Co-Author(s):
Introduction:
The use of biomarkers for early detection of Alzheimer disease (AD) is crucial for developing potential treatments. As neurons die, brain structures, including the hippocampus, shrink and the cerebrospinal fluid spaces ventricles expand. However, the hippocampus is a very small structure, and its shrinkage can be challenging to analyze. The lateral ventricle volume is larger and easier to measure than hippocampal volumes. Previous studies have validated ventricular enlargement as a possible measure of AD progression (1-3). In this study, we tested the hypothesis that baseline and longitudinal changes in lateral ventricle volume as measured by magnetic resonance imaging (MRI) associated with baseline clinical dementia rating® (CDR®) (4) and associated with baseline and longitudinal changes in hippocampus volume.
Methods:
We evaluated 518 participants enrolled in longitudinal studies at the Knight ADRC at Washington University in St. Louis. MRI data was obtained through OASIS (www.oasis-brains.org). Participants were required to have at least two CDR assessments and MRI scans. Lateral ventricular and hippocampal volumes were extracted and corrected for intracranial volume using FreeSurfer. Univariate linear regression models were used to evaluate the relationship between baseline CDR, treated as a categorical variable, and baseline lateral ventricle volumes, as well as the relationship between baseline lateral ventricle volume and baseline hippocampal volume. A two-step random coefficient model was used to evaluate the rate of change in lateral ventricle volume associated with baseline CDR or rate of change in hippocampal volume. The random coefficient model can accommodate the heterogeneous number of visits and visit intervals undergone by study participants. The first step of this two-step model was to calculate the rate of change in lateral ventricular volume (lvv_slope) and rate of change in hippocampus volume (hv_slope) after controlling for age, sex, APOE-ε4, and years of education, see equation 1. The second step evaluated the relationship between baseline CDR and rate of change in lateral ventricular volume, as well as the relationship between rate of change in lateral ventricle volume and rate of change in hippocampus volume using univariate linear regression model, see equation 2.
MRI = α1*time + α2*age + α3*sex + α4*education +α5 *APOE ε4 (1)
lvv_slope= β*hv_slope
or lvv_slope= β*CDR (2)
Here, time was set as a continuous variable (measured in years) representing the interval between the baseline visit and each subsequent visit. Within this model, time was treated as both a fixed and random effect.
MRI = α1*time + α2*age + α3*sex + α4*education +α5 *APOE ε4 (1)
lvv_slope= β*hv_slope
or lvv_slope= β*CDR (2)
Here, time was set as a continuous variable (measured in years) representing the interval between the baseline visit and each subsequent visit. Within this model, time was treated as both a fixed and random effect.
Results:
Demographic characteristics of the participants at baseline are presented in Table 1. We found that larger baseline lateral ventricle volumes were associated with worse baseline CDR (estimate β =12110, p=0.0001, figure 1A). and with smaller baseline hippocampal volumes (estimate β =-8.64, p<0.0001, figure 1B). Additionally, the rate of changes in lateral ventricle volumes were also associated with baseline CDR and rate of changes in hippocampal volume (estimates β =548.8, p<0.0001, figure 1C and estimatesβ =-6.01, p<0.0001, figure 1D, respectively).
·Figure 1: Scatterplots showing associations of the lateral ventricle volume with CDR and hippocampal volume
Conclusions:
Larger baseline lateral ventricles volumes were associated with smaller baseline hippocampal volumes and worse baseline clinical dementia scores in an AD cohort. In addition, increased rates in enlargement of lateral ventricles volumes were association with increased atrophy rates of the hippocampus and worse baseline clinical dementia scores. Baseline and longitudinal lateral ventricle volumes could potentially be used to stratify risk for AD.
Modeling and Analysis Methods:
Classification and Predictive Modeling 2
Novel Imaging Acquisition Methods:
Anatomical MRI 1
Keywords:
Aging
Cognition
Computational Neuroscience
Degenerative Disease
MRI
Statistical Methods
STRUCTURAL MRI
1|2Indicates the priority used for review
Provide references using author date format
1. Sean M. Nestor, Raul Rupsingh, Michael Borrie, Matthew Smith, Vittorio Accomazzi, Jennie L.Wells, Jennifer Fogarty, Robert Bartha and the Alzheimer’s Disease Neuroimaging Initiative (2008), “Ventricular enlargement as a possible measure of Alzheimer’s disease progression validated using the Alzheimer’s disease neuroimaging initiative database”, Brain, 131, 2443-2454.
2. Owen T. Carmichael, Lewis H. Kuller, Oscar L. Lopez, Paul M. Thompson, Rebecca A. Dutton, Allen Lu, Sharon E. Lee, Jessica Y. Lee, Howard J. Aizenstein, Carolyn Cidis Meltzer, Yanxi Liu, Arthur W. Toga, and James T. Becker (2007), “Ventricular volume and dementia progression in the Cardiovascular Health Study”, Neurobiol Aging, 28(3): 389–397.
3. Astri J. Lundervold, Alexandra Vik, Arvid Lundervold (2019), “Lateral ventricle volume trajectories predict response inhibition in older age—A longitudinal brain imaging and machine learning approach”, PLoS ONE, 14(4): e0207967.
4. Morris JC (1993), “The Clinical Dementia Rating (CDR): current version and scoring rules”, Neurology, 43, 2412-2414.