Assessment of parahippocampal integrity to estimate Alzheimer's disease status in structural MRI using a Cortex-to-Sulcus-Ratio

Abstract

Parahippocampal structures are the first to be damaged by Alzheimer's disease (AD). Structural changes can be detected at predementia-phase of mild cognitive impairment (MCI). However, utility of parahippocampal integrity for early AD-detection is limited due to high anatomical variability, mainly determined by the collateral sulcus. Presenting the Cortex-to-Sulcus-Ratio (CSR) correcting for this variability, the current study takes a first step in validating a relative measurement of parahippocampal integrity cross-sectionally assessing AD-status. The CSR is assumed to be superior to standard parahippocampal volumes as it improves differentiation between parahippocampal integrity of older adults being cognitively normal (CN), MCI or AD. Contrasting CSR and absolute parahippocampal volumes derived from manually segmented brains, this was examined in a sample of 24 individuals aged between 58 and 80 years, categorised as CN, MCI or AD. Investigating which parahippocampal structures might be relevant for AD-detection, different cortical volumes were considered individually and combinedly. Descriptive comparisons supported the presumed superiority: a clearer distinction was possible and the assumed group-order of parahippocampal integrity was only maintained when looking at the CSR. Considering age as a covariate, these observations could be statistically verified by significant group*integrity-measure-interactions for the left perirhinal and entorhinal cortex and a combined measure of left parahippocampal structures. Although validation requires further research, results support the idea of including a measurement of sulcal enlargement when evaluating parahippocampal integrity in aging individuals. The CSR seems to be beneficial in estimating AD-status in cross-sectional structural magnetic resonance imaging as it overcomes weaknesses of absolute volumetric measurements.