Hippocampal functional connectivity in Alzheimer’s disease: a resting state 7T fMRI study

P. Gowland,R. Dury,L. Velayudhan,S. Francis,Sana Bestwn,S. Bhattacharyya,Subhadip Paul

DOI: https://doi.org/10.1017/S1041610220003440

IF: 7.191

2020-10-20

International Psychogeriatrics

Abstract:Among the range of neuroimaging techniques available that may provide insight into Alzheimer’s disease (AD), resting-state functional magnetic resonance imaging (rsfMRI) and functional connectivity has emerged as particularly useful in the early diagnosis of AD (de Vos et al., 2018). Altered functional connectivity of the hippocampus is recognized as an important feature of preclinical AD and found in the early stages of AD (Franzmeier et al., 2019; Scherr et al., 2019; Wang et al., 2006). The hippocampus is one of the earliest affected brain regions in AD (Braak et al., 1993), with its degradation being a hallmark of AD inmany structural and functional imaging studies. Thus, we aimed to investigate whether the connectivity of the hippocampus with the rest of the brain is altered inADpatients compared to healthy individuals using a 7-Tesla MRI imaging. Eleven individuals meeting a NINCDS-ADRDA (McKhann et al., 1984) diagnosis ofmild-to-moderate probable AD and 10 non-demented controls (NDC) who were similar in age were recruited for this cross-sectional study frommemory services in Leicestershire, United Kingdom. The study had regional Research Ethics Committee approval (12/EM/0007) and all participants provided written informed consent. Mini-mental state examination (MMSE) (Folstein et al., 1975) was used for cognitive assessment. Resting-state fMRI data acquired using Philips 7-Tesla Achieva scanner (University of Nottingham), with eyes closed and a 2D gradient-echo, echo-planar imaging (GE-EPI) sequence (repetition time= 2000 ms, echo time= 25ms, nominal flip angle= 75°, 36 slices with 2× 2× 2 mm3 voxels; scan duration= 5:12 min) was preprocessed (realignment, motion correction, and slice timing correction) and functional connectivity analyses were carried out using CONN: Functional Connectivity Toolbox v17 (WhitfieldGabrieli and Nieto-Castanon, 2012) running in SPM12 (http://www.fil.ion.ucl.ac.uk/spm). rsfMRI volumes were normalized to the standard Montreal Neurological Institute space template, segmented and realignment parameters, white matter, and CSF signals used as covariates-of-no interest in the CONN toolbox(Behzadi et al., 2007). Residual time series of rsfMRI images were then band-pass filtered (0.008< f< 0.09Hz) and a seed-based region of interest (ROI) analysis was conducted using the CONN Toolbox. Resting-state BOLD signals from the left and right hippocampal ROIs were defined and correlatedwith 132 brain regions from theCONNatlas in a 2x132 correlation matrix. Individuals with AD (36% women; 54–78 years; mean: 61.6 ± 6.3 years) andNDCparticipants (70% women; 54–77 years; mean: 64.9 ± 6.3 years) were studied. The NDC group (Mean: 29.7 ± 0.7; Median: 30.0 [interquartile range: 29.75, 30.0]) had higher (Mann–Whitney U= 5.0, Z= -3.64, p< 0.001) MMSE score than AD individuals (Mean: 24.9 ± 4.8; Median: 27.0 [interquartile range: 23.0, 29.0]). There was significantly greater functional connectivity (beta= 0.31; T19= 4.7; punc= 0.000154; false discovery rate (FDR) corrected p= 0.020239; Figure 1; FDR correction applied for each seed separately) of the left hippocampal seed with left superior frontal gyrus (SFG) in NDC compared to AD individuals. MMSE scores positively correlated with hippocampalSFG functional connectivity (Spearman’s r= 0.629, n= 21, p ≤ 0.002) in the whole group of participants suggesting lower MMSE scores were associated with reduced connectivity between the left hippocampus and the left SFG. Given the modest sample size, these results are to be considered preliminary and need independent replication in larger samples. To our knowledge, this is the first functional connectivity report in AD using ultra-high-field strength, 7-Tesla, rsfMRI which provides increased contrast-to-noise ratio and facilitates increased spatial resolution (Li et al., 2012), thus providing increased sensitivity and spatial specificity for the detection of pathological changes that might not be detectable at lower field strengths. These preliminary findings are consistent with previous evidence of altered hippocampal FC in individuals with AD compared to NDC, in particular altered FC between the hippocampus and frontal lobes (de Vos et al., 2018; Franzmeier et al., 2019; Scherr et al., 2019; Wang et al., 2006). The SFG is thought to be involved in higher cognitive domains and is a critical component of the working memory network. Based on much of the previous literature that has found the hippocampus to be one of the earliest brain areas affected by the accumulation of AD lesions, onemay hypothesize that the integrity of functional connectivity between the hippocampus and other brain regions must be compromised in individuals with AD. If these preliminary results can International Psychogeriatrics (2021), 33:1, 95–96 © International Psychogeriatric Association 2020

Medicine