Article 2021

Association of Hippocampal Subfields, CSF Biomarkers, and Cognition in Patients With Parkinson Disease Without Dementia

{Objectives: To examine whether hippocampal volume loss is primarily associated with cognitive status or pathologic Amyloid-$\beta$ 1-42 (A$\beta$42) levels, this study compared hippocampal subfield volumes between both Parkinson\textquotesingles Disease (PD) patients with (PD-MCI) and without (PD-CN) cognitive impairment and between patients with low and high A$\beta$42 levels, in addition exploring the relationship between hippocampal subfield volumes, CSF biomarkers (A$\beta$42, phosphorylated and total tau), neuropsychological tests, and activities of daily living. Methods: Forty-five non-demented PD patients underwent CSF analyses and magnetic resonance imaging as well as comprehensive motor and neuropsychological examinations. Hippocampal segmentation was conducted using FreeSurfer image analysis suite 6.0. Regression models were used to compare hippocampal subfield volumes between groups, and partial correlations defined the association between variables while controlling for intracranial volume (ICV). Results: Linear regressions revealed cognitive group as a statistically significant predictor of both the hippocampal-amygdaloid transition area (HATA; $\beta$ \textequals -0.23, 95\textpercent CI: -0.44 to -0.02) and the Cornu Ammonis 1 region (CA1; $\beta$ \textequals -0.28, 95\textpercent CI: -0.56 to -0.02), independent of disease duration and ICV, with PD-MCI patients showing significantly smaller volumes than PD-CN. In contrast, no subfields were predicted by A$\beta$42 levels. Smaller hippocampal volumes were associated with worse performance on memory, language, spatial working memory and executive functioning tests. The subiculum was negatively correlated with total tau levels (r \textequals -0.37, 95\textpercent CI: -0.60 to -0.09). Conclusion: Cognitive status, but not CSF A$\beta$42, predicted hippocampal volumes, specifically the CA1 and HATA. Hippocampal subfields were associated with various cognitive domains, as well as with tau pathology.}

Author(s): Becker, S and Granert, O and Timmers, M and Pilotto, A and Van Nueten, L and Roeben, M and Salvadore, G and Galpern, WR and Streffer, J and Scheffler, K and Maetzler, W and Berg, D and Liepelt Scarfone, I
Journal: {Neurology}
Volume: 96
Number (issue): 6
Pages: e904--e915
Year: 2021
Publisher: Advanstar Communications [etc.]
Bibtex Type: Article (article)
DOI: 10.1212/WNL.0000000000011224
Address: Cleveland, Ohio [etc.]
Electronic Archiving: grant_archive

BibTex

@article{item_3265148,
  title = {{Association of Hippocampal Subfields, CSF Biomarkers, and Cognition in Patients With Parkinson Disease Without Dementia}},
  journal = {{Neurology}},
  abstract = {{Objectives: To examine whether hippocampal volume loss is primarily associated with cognitive status or pathologic Amyloid-$\beta$ 1-42 (A$\beta$42) levels, this study compared hippocampal subfield volumes between both Parkinson\textquotesingles Disease (PD) patients with (PD-MCI) and without (PD-CN) cognitive impairment and between patients with low and high A$\beta$42 levels, in addition exploring the relationship between hippocampal subfield volumes, CSF biomarkers (A$\beta$42, phosphorylated and total tau), neuropsychological tests, and activities of daily living. Methods: Forty-five non-demented PD patients underwent CSF analyses and magnetic resonance imaging as well as comprehensive motor and neuropsychological examinations. Hippocampal segmentation was conducted using FreeSurfer image analysis suite 6.0. Regression models were used to compare hippocampal subfield volumes between groups, and partial correlations defined the association between variables while controlling for intracranial volume (ICV). Results: Linear regressions revealed cognitive group as a statistically significant predictor of both the hippocampal-amygdaloid transition area (HATA; $\beta$ \textequals -0.23, 95\textpercent CI: -0.44 to -0.02) and the Cornu Ammonis 1 region (CA1; $\beta$ \textequals -0.28, 95\textpercent CI: -0.56 to -0.02), independent of disease duration and ICV, with PD-MCI patients showing significantly smaller volumes than PD-CN. In contrast, no subfields were predicted by A$\beta$42 levels. Smaller hippocampal volumes were associated with worse performance on memory, language, spatial working memory and executive functioning tests. The subiculum was negatively correlated with total tau levels (r \textequals -0.37, 95\textpercent CI: -0.60 to -0.09). Conclusion: Cognitive status, but not CSF A$\beta$42, predicted hippocampal volumes, specifically the CA1 and HATA. Hippocampal subfields were associated with various cognitive domains, as well as with tau pathology.}},
  volume = {96},
  number = {6},
  pages = {e904--e915},
  publisher = {Advanstar Communications [etc.]},
  address = {Cleveland, Ohio [etc.]},
  year = {2021},
  slug = {item_3265148},
  author = {Becker, S and Granert, O and Timmers, M and Pilotto, A and Van Nueten, L and Roeben, M and Salvadore, G and Galpern, WR and Streffer, J and Scheffler, K and Maetzler, W and Berg, D and Liepelt Scarfone, I}
}