Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Frontal-executive and corticolimbic structural brain circuitry in older people with remitted depression, mild cognitive impairment, Alzheimer’s dementia, and normal cognition

Abstract

A history of depression is a risk factor for dementia. Despite strong epidemiologic evidence, the pathways linking depression and dementia remain unclear. We assessed structural brain alterations in white and gray matter of frontal-executive and corticolimbic circuitries in five groups of older adults putatively at-risk for developing dementia- remitted depression (MDD), non-amnestic MCI (naMCI), MDD+naMCI, amnestic MCI (aMCI), and MDD+aMCI. We also examined two other groups: non-psychiatric (“healthy”) controls (HC) and individuals with Alzheimer’s dementia (AD). Magnetic resonance imaging (MRI) data were acquired on the same 3T scanner. Following quality control in these seven groups, from diffusion-weighted imaging (n = 300), we compared white matter fractional anisotropy (FA), mean diffusivity (MD), and from T1-weighted imaging (n = 333), subcortical volumes and cortical thickness in frontal-executive and corticolimbic regions of interest (ROIs). We also used exploratory graph theory analysis to compare topological properties of structural covariance networks and hub regions. We found main effects for diagnostic group in FA, MD, subcortical volume, and cortical thickness. These differences were largely due to greater deficits in the AD group and to a lesser extent aMCI compared with other groups. Graph theory analysis revealed differences in several global measures among several groups. Older individuals with remitted MDD and naMCI did not have the same white or gray matter changes in the frontal-executive and corticolimbic circuitries as those with aMCI or AD, suggesting distinct neural mechanisms in these disorders. Structural covariance global metrics suggested a potential difference in brain reserve among groups.

Introduction

Several large cohort studies and meta-analyses have shown that both a remote or recent history of major depression increases the risk of dementia [1, 2]. After older adults with a major depressive disorder (MDD) have been successfully treated, many still experience cognitive deficits [3,4,5,6]. One-third to one-half of older individuals with MDD may have a concomitant diagnosis of MCI, a rate far exceeding MCI prevalence in the general population [7,8,9]. While MDD has been shown to be a risk factor for AD in many studies [1, 2, 7, 10], the pathophysiologic mechanisms linking these two conditions remain unclear. A decade ago, Butters et al. [11] proposed two main mechanistic hypotheses to explain the association between MDD and AD: (a) a vascular hypothesis, i.e., structural damage to frontal-executive circuitry and (b) an inflammation hypothesis, i.e., a high level of stress hormones in depression can lead to hippocampal volume loss resulting in damage to corticolimbic circuitry.

A large number of neuroimaging studies have reported alterations in brain regions involved in frontal-executive [12,13,14] and corticolimbic [15,16,17,18] circuitries [19, 20] in samples of older individuals with MDD, MCI, or AD. However, almost all included only a single diagnostic group, and a group of healthy controls typically focused on a single or small number of brain regions (e.g., hippocampus [15, 17, 21, 22], corpus callosum [23]). Only a handful of studies have included older adults both from groups with MDD and from groups with MCI or AD. These studies typically have small sample sizes, and this may be one reason why results remain inconclusive [16, 18, 24, 25]. Also, examining brain circuitry using more novel techniques such as structural covariance, may provide new insights into possible neural risk mechanisms for dementia [26,27,28]. Previous results have shown disruption of gray matter network topologies in older individuals with MDD [29,30,31], MDD with memory deficits [28], or MCI groups [26] compared with those with normal cognition.

To address these limitations in the literature, we conducted an analysis including five groups of older individuals putatively at high risk for AD; those with: (1) remitted MDD; (2) amnestic mild cognitive impairment (aMCI) and no MDD; (3) non-amnestic MCI (naMCI) and no MDD; (4) both remitted MDD and naMCI (MDD+naMCI); and (5) both remitted MDD and aMCI (MDD+aMCI). We also included a group of older non-psychiatric (“healthy”) controls (HC), and a group of individuals with mild or moderate AD. We hypothesized that structural brain alterations in the gray and white matter of frontal-executive and corticolimbic circuitries would be present among these seven groups of older individuals, with the degree of alterations ranked according to the expected degree of risk for AD: HC, MDD, naMCI, MDD+naMCI, aMCI, MDD+aMCI, AD. Implicit in this ranking is the expectation that individuals with MCI are at higher risk than those with MDD; those with aMCI are at higher risk than those with naMCI; and those with two disorders (i.e., both MDD and MCI) are at higher risk than those with one. We also hypothesized that differences in structural brain circuitry disruption among these seven groups would be observed both in terms of white matter integrity assessed with diffusion-weighted imaging (DWI) and gray matter (assessed with T1-weighted image). We also explored whether these differences would be captured in the topology of structural covariance networks.

Methods and materials

Participants

MRI data for the five groups at-risk of AD (i.e., MDD, naMCI, MDD+naMCI, aMCI, MDD+aMCI) were acquired in an ongoing clinical trial, the Prevention of Alzheimer’s Dementia with Cognitive Remediation plus Transcranial Direct Current Stimulation in Mild Cognitive Impairment and Depression (PACt-MD). This trial was approved by the institutional review board of the Centre for Addiction and Mental Health (CAMH), in Toronto, Canada. From a total of 378 participants in the PACt-MD study, we included in this analysis all who agreed to complete a baseline MRI scan, of whom five were diagnosed with dementia during the consensus conference leaving n = 284. All provided written informed consent and underwent a comprehensive baseline clinical assessment including extensive cognitive testing. Their baseline diagnoses were adjudicated according to the criteria of the Diagnosis and Statistical Manual 5th edition (DSM-5) during a consensus conference. Comorbid physical illnesses were also assessed, and their burden was quantified using the Cumulative Illness Rating Scale-Geriatric (CIRS-G) [32].

The 284 PACt-MD participants comprised five at-risk groups based on their consensus DSM-5 diagnoses. First, participants with MDD (n = 50) were 65 years and older; they had a history of one or more major depressive episodes (MDE) that occurred during their adult life (>18 y/o) and had been in remission for at least 2 months. Participants with MCI (n = 146) were 60 years and older; they met DSM-5 criteria for mild neurocognitive disorder but not for an MDE at any time during their adult life. Since previous studies have reported substantial biological heterogeneity in MCI [33, 34], they were further divided into aMCI and naMCI groups, with aMCI requiring an impairment in the memory domain during cognitive testing.

We also included a group of participants 55 years and older from two clinical trials for AD who had completed a baseline MRI scan (n = 58). They were diagnosed with mild-to-moderate AD (i.e., scores on the Mini-Mental State Examination (MMSE) [35] or Montreal Cognitive Assessment [36] higher than 15 and 10, respectively). Participants in the AD group were significantly older and had less years of education than all the other groups. Therefore, we conducted post-hoc sensitivity analyses with a subset of AD patients (n = 32) matched for age and education to the other groups.

Our HC were 60 years and older (n = 62). Each completed an MRI scan in the context of one of four studies: PACt-MD (n = 20); the two AD trials already described (n = 22); and a cohort study of normal aging (n = 20). To be included, HCs could not have a DSM-IV-TR or DSM-5 diagnosis of any minor or major neurocognitive disorder or any other lifetime diagnosis (except for simple/specific phobias). Prior to data analysis, we ruled out any differences in brain structure among the HCs from each of the four studies, using a Kruskal–Wallis test.

MRI acquisition and analysis

All participants were scanned on the same 3-T GE MR750 Echospeed (General Electric, Milwaukee, WI) research-dedicated scanner at CAMH using the same protocol. Of the 404 participants who agreed to the multimodal MRI scan, all completed the T1, and 366 completed the DWI scan.

DWI

Of the DWI scans, 66 were excluded due to poor quality, leaving 300 scans for analyses; HC (n = 30), MDD (n = 43), naMCI (n = 36), MDD+naMCI (n = 24), aMCI (n = 84), MDD+aMCI (n = 47), and AD (n = 36) (see Supplementary Fig. 1).

Diffusion-weighted images—60 gradient directions with b = 1000, 5 baseline scans with b = 0—were denoised with the MRtrix3 dwidenoise command. Next, we corrected the DWI data for motion and eddy current distortion in a single step using eddy (FSL 5.0.10) [37, 38]. The BrainSuite diffusion pipeline was used to register each participant’s b0 image to their bias-field corrected anatomical T1-weighted image [39]. Next, we used FSL’s dtifit algorithm to calculate output maps of fractional anisotropy (FA), mean diffusivity (MD), V1, and sum of squared errors [40, 41]. After applying the ENIGMA-DTI tract-based spatial statistics algorithm [42], the JHU white matter atlas was used to extract FA and MD values for each tract.

T1-weighted Image

A sagittal 3D fast spoiled gradient echo image with 0.9 mm isotropic voxels was acquired. Reconstruction of the cortical surfaces was performed using the FreeSurfer toolkit version 6.0.0. Cortical thickness maps were aligned across all scans using non-linear surface-based registration [43,44,45,46]. The mean thickness was passed to a segmentation algorithm producing a table of thickness values for all labels in “aparc” (Desikan–Killiany atlas) [47, 48]. Subcortical segmentations were labeled based on their location using a probabilistic atlas (“aseg”) [49, 50]. Of 404 T1-weighted scans, 71 were excluded due to poor quality, leaving 333 for analyses; HC (n = 56), MDD (n = 42), naMCI (n = 34), MDD+naMCI (n = 29), aMCI (n = 84), MDD+aMCI (n = 45), and AD (n = 43) (see Supplementary Fig. 1).

Statistical analysis

Primary ROI analyses

The open-source R-statistics (v.3.3.2) program was used to analyze regions of interest (ROIs) based on our a priori hypothesis. See Fig. 1a for white matter tract and 2a and 3a for gray matter ROIs.

Fig. 1: Diffusion-based measures of white matter integrity in frontal-executive and corticolimbic tracts.
figure1

a The JHU white matter atlas was used to extract FA and MD values for each tract. Tracts of interest corresponding to the frontal-executive circuitry are displayed in blue boxes: gCC (genu of corpus callosum), bCC (body of corpus callosum), sCC (splenium of corpus callosum), SLF (superior longitudinal fasciculus), and FX-ST (fornix-stria terminalis). The corticolimbic tracts are displayed in pink boxes: CGC (cingulum-cingulate gyrus part), CGH (cingulum-hippocampus part), UNC (uncinate fasciculus). *Bilateral UNC tracts are not displayed in this. b Post-hoc comparisons of fractional anisotropy (FA) in ROIs from both frontal-executive and corticolimbic circuitries. c Post-hoc comparisons of mean diffusivity (MD) in regions of interest (ROIs) from both frontal-executive and corticolimbic circuitries. Number of participants in each group: AD (n = 36), aMCI (n = 84), naMCI (n = 36), rMDD+aMCI (n = 47), rMDD+naMCI (n = 24), rMDD (n = 43), HC (n = 30).

Within each group, data from outliers (Z-score > ±3) were excluded from further analysis. An analysis of covariance (ANCOVA, type II sums of squares) was conducted for each measure (FA, MD, subcortical volume, cortical thickness) to examine the main effects of diagnosis while controlling for age, sex, and years of education (and intracranial volume in the subcortical volume analysis). We corrected for multiple comparisons using a false discovery rate (FDR) of 0.05 for all the tests [51]. Post-hoc t-tests were used to further assess regions with significant group effects after FDR correction and adjusted p values are reported.

In addition, a sex-by-diagnosis analysis was conducted in the ROIs that had a statistically significant main effect for diagnostic groups.

Finally, a Pearson correlation test was used to explore whether age of onset of depression (in MDD, MDD+naMCI, and MDD+aMCI group), influenced brain structures.

Exploratory graph theory analysis

We conducted an exploratory graph theory analysis [52] examining whole-brain structural covariance, using Braingraph 2.7.0 [53]. Residuals of 68 cortical thickness ROIs extracted from the Desikan–Killiany atlas (controlled for age, sex, and years of education) were used to construct the structural covariance networks for each group. The adjacency matrix of positive correlations in each group was binarized.

We estimated group differences in clustering coefficient and modularity (Q) (segregation), global efficiency, characteristic path length (integration) at a range of densities from 0.5 (the minimum density at which all groups were fully connected) to 0.8 with increments of 0.02 (see Supplementary Table 1 for definition of measures) to minimize the influence of different density on the network measures among groups [54]. Mean estimates of global graph-theoretical measures along with their confidence intervals for each group and density were generated with 1000 bootstrap samples. The significance of between-group differences was determined with 5000-permutation testing. FDR correction was applied for each density.

We identified high-centrality brain hubs, defined as the top 10 in more than 50% of densities for more than two centrality measures; degree, eigenvector, and edge betweenness. Brain hubs are considered critical waypoints for information flow and thus crucial to complex cognition and potential points of vulnerability [55].

Results

Baseline characteristics

Demographics and clinical (including burden of comorbid of physical illness) characteristics of the 333 participants in the T1-weighted ROI analysis are presented in Table 1 (Supplementary Table 2 for demographics of n = 300 participants in the DWI analysis).

Table 1 Demographics and clinical characteristics of participants in the T1-weighted data analysis.

Primary ROI analyses of DWI and white matter analysis

Fractional anisotropy (FA)

There was a statistically significant main effect for diagnostic group in all 13 tracts.

In frontal-executive circuitry, we found the most prominent effect of diagnosis in the left and right fornix-stria terminalis (FX-ST; F(6, 284) = 29.1, p = 5.2e−26, η2 = 0.30, and F(6, 283) = 2 4.7, p = 5.8e−25, η2 = 0.30, respectively), and the genu of the corpus callosum (GCC; F(6, 283) = 26.3, p = 2.3e−24, η2 = 0.26).

In corticolimbic circuitry, the mostprominent effects of diagnosis were within the left and right cingulum-hippocampus (CGH; F(6, 284) = 26.5, p = 1.9e−24, η2 = 0.31, and F(6, 285) = 24.9, p = 2.2e−23, η2 = 0.30, respectively). The statistical details of the remaining significant tracts are displayed in Supplementary Table 3.

In the post-hoc Tukey tests, the AD group had the lowest FA in each of the 13 ROIs compared with the other six groups (p < 0.001) (Fig. 1b and Supplementary Table 3). No other comparisons reached significance.

Mean diffusivity (MD)

There was a statistically significant main effect for diagnostic group for five of 13 tracts.

In frontal-executive circuitry, we found the most prominent main effect of diagnosis in the left and right fornix-stria terminalis; F(6, 285) = 13.0, p = 6.0e−12, η2 = 0.15, and F(6, 285) = 11.6, p = 8.6e−11, η2 = 0.13, respectively), and the left superior longitudinal fasciculus; F(6, 284) = 8.9, p= 6.4e−09, η2 = 0.05).

In corticolimbic circuitry, we found the main effect of diagnosis in the left and right cingulum bundles; F(6, 285) = 2.78, p = 1.21e−02, η2 = 0.047, and F(6, 284) = 4.7, p = 5.3e−04, η2 = 0.07, respectively).

In the post-hoc Tukey test, the AD group had the highest MD in the frontal-executive and corticolimbic tracts compared with the other six groups (p < 0.05).

The aMCI group had significantly higher MD in the right cingulum bundle compared with the HC group, and bilateral fornix-stria terminalis compared with the HC and MDD groups, and in the left superior longitudinal fasciculus compared with the MDD+aMCI and MDD+naMCI groups (p < 0.05) (Fig. 1c and Supplementary Table 4).

T1-weighted imaging and gray matter analysis

Subcortical volume

There was a significant main effect of diagnostic group in seven of the ten ROIs.

In frontal-executive circuitry, we found that the right and left nucleus accumbens (F(6, 320) = 5.2, p= 9.5e−05, η2 = 0.07), and F(6, 321) = 4.6, p= 3.2e−04, η2 = 0.06, respectively), and right thalamus (F(6, 320) = 2.8, p= 1.6e−02, η2 = 0.028) significantly differed across groups.

In corticolimbic circuitry, we found that the right hippocampus (F(6, 317) = 14.8, p = 7.6e−14, η2 = 0.12), left hippocampus (F(6, 317) = 12.7, p = 3.8e−12, η2 = 0.12), left amygdala (F(6, 317) = 11.7, p= 2.4e−11, η2 = 0.13), and right amygdala (F(6, 317) = 9.5, p = 3.7e−09, η2 = 0.11) significantly differed across groups (Fig. 2b).

Fig. 2: Subcortical volume in frontal-executive and corticolimbic regions of interest.
figure2

a The “aseg” atlas was used for subcortical volume segmentation. The frontal-executive regions are displayed in blue and corticolimbic regions are displayed in pink. *Nucleus accumbens is not displayed in  the figure. b Subcortical regions compared across the seven groups. The F-statistics of the significant regions of interest (ROIs) are displayed in a scaled color, and non-significant regions are displayed in white. Bilateral nucleus accumbens were significant but not displayed in the figure (a and b made with “ggseg” R package) [103], c Post-hoc comparisons of subcortical volume in four regions of interest (ROIs) from both frontal-executive and corticolimbic circuitries. Number of participants in each group: AD (n = 43), aMCI (n = 84), naMCI (n = 34), rMDD+aMCI (n = 45), rMDD+naMCI (n = 29), rMDD (n = 42), HC (n = 56).

In the post-hoc Tukey test, the AD group had significantly lower volumes compared with the other six groups, with the exception of the right thalamus. The aMCI group had significantly lower bilateral hippocampal volume compared with the HC, MDD, and MDD+naMCI groups; and lower left amygdala volume compared with the HC and MDD groups (p < 0.05) (Fig. 2c and Supplementary Table 5).

Cortical thickness

There was a main effect of diagnostic group in 22 of the 38 cortical thickness ROIs. The left and right entorhinal cortices had the largest effect size (F(6, 323) = 11.6, p < 3.2e−10, η2 = 0.15, and F(6, 319) = 10.6, p = 1.9e−09, η2 = 0.15, respectively), followed by the right inferior parietal (F(6, 322) = 9.08, p = 4.5e−08, η2 = 0.14), and the left temporal pole (F(6, 322) = 8.5, p = 1.08e−07, η2 = 0.12). The remaining results are displayed in Fig. 3b, with statistical details in Supplementary Table 6.

Fig. 3: Cortical thickness in frontal-executive and corticolimbic regions of interest.
figure3

a The Desikan–Killiany atlas (aparc) was used for cortical thickness segmentation. The frontal-executive regions are displayed in blue boxes and corticolimbic regions are displayed in pink boxes. b All the cortical thickness regions that were included in the model and compared across the seven groups are displayed. The F-statistics of the significant regions of interest (ROIs) are displayed in a scaled color, and non-significant regions are displayed in white (Figure made with “ggseg” R package) [103]. c Post-hoc comparisons of cortical thickness in four ROIs from both frontal-executive and corticolimbic circuitries. Number of participants in each group: AD (n = 43), aMCI (n = 84), naMCI (n = 34), rMDD+aMCI (n = 45), rMDD+naMCI (n = 29), rMDD (n = 42), HC (n = 56).

In the post-hoc Tukey test, the AD group had more cortical thinning than the other six groups (p < 0.05). The aMCI group had lower thickness in several temporal and frontal lobe regions (e.g., entorhinal cortex, superior (medial) frontal, caudal middle frontal, superior temporal gyrus) compared with the HC, MDD, naMCI, and MDD+aMCI groups. The left caudal middle frontal gyrus was significantly thinner in both aMCI and MDD+aMCI groups than the MDD+naMCI group (Fig. 3c). For further details see Supplementary Table 7.

Sensitivity analysis, correlation analyses and sex-based analyses

In the sensitivity analysis excluding 11 participants with AD who were older than 80 years and had low education levels, all results remained the same in FA, MD, in subcortical volumes analysis, except for right thalamus, and in cortical thickness analysis, except for left pars orbitalis.

Within the three MDD groups, we did not find any significant correlations between age of onset of depression and FA, MD, subcortical volume, or cortical thickness.

There was a significant effect of sex-by-diagnosis in the right amygdala F(6, 311) = 3.4, p = 0.02, η2 = 0.039 (Supplementary Fig. 2). The post-hoc Tukey results showed that in females, AD patients had more volume loss than the other six groups (p < 0.05). However, in the males, a volume loss was found in AD, aMCI, and rMDD+aMCI groups compared with the HC group only (p < 0.05).

Exploratory structural covariance and graph theory analyses

Whole-network measures

We found group differences in the characteristic path length, clustering coefficient, and modularity. However, there was no group difference in global efficiency (Fig. 4a). The MDD group had high characteristic path lengths compared with the HC group (Fig. 4b, c). Modularity was higher in the AD and aMCI groups; however, it did not reach statistical significance (Fig. 4d, e).

Fig. 4: Global properties of cortical thickness covariance networks.
figure4

Graph theory measures over range of densities; a no difference in global efficiency among the seven groups, b high clustering coefficient in the MDD group compared with the control group, c high characteristic path length (average shortest path length) in the MDD and naMCI groups compared with the HC and AD groups, d high modularity in the AD and aMCI groups, and low modularity in the MDD and naMCI groups relative to the other groups (did not reach statistical significance), e low modularity and lack of clear boundaries between the optimal community structures (modules) in the MDD and naMCI groups, and high modularity with a higher number of optimal community structures; indicating fragmented brain network into a few large isolated components, in the aMCI and AD groups. The HC group demonstrates a normal modular brain structure. Figures are displayed at density of 0.52 with BrainNet Viewer [104].

Hub regions

Within the frontal-executive and corticolimbic networks of interest, all groups had high-centrality hubs across a range of densities, reaffirming the central role of these networks. The AD group appeared to be most compromised, having fewer hub regions overall, but particularly within the frontal ROIs (see Supplementary Table 8).

Within the frontal-executive network, the HCs had four hubs in frontal regions. All other groups had fewer frontal hubs. Most notably, the MDD+aMCI, aMCI, and AD groups had only a single frontal hub-node (different among them). The MDD+naMCI group had three hubs, and the MDD group and naMCI group had two hubs.

In the corticolimbic network frontal hubs were found in the HC, MDD, naMCI, MDD+naMCI, and aMCI groups, however, the AD and MDD+aMCI groups had none. Bilateral superior temporal nodes were consistently found across the aMCI, MDD+aMCI and AD groups (see Supplementary Table 8).

Discussion

We used diffusion-weighted and T1-weighted MRI measures to assess brain differences in the white and gray matter of the frontal-executive and corticolimbic circuitries in five groups of participants at-risk of developing AD (MDD, naMCI, MDD+naMCI, aMCI, MDD+aMCI), control participants, and participants with mild-to-moderate AD. Using an omnibus comparison, we found significant differences in FA, MD, subcortical volume, and cortical thickness in frontal-executive and corticolimbic ROIs among the seven groups. As expected, the AD group [56,57,58,59,60,61] and to an extent the aMCI group [62,63,64,65] demonstrated alterations across a range of white matter and gray matter measures in our circuitries of interest compared with the other five groups. The naMCI, MDD+naMCI and MDD+aMCI groups did not exhibit the same extent of alterations as the aMCI or AD group, and surprisingly they were often not different from the HC group. Finally, the MDD group did not differ from the HC group in any region or circuitry. However, graph theory measures of cortical topology revealed additional network-based differences in cortical topology in the MDD group compared with the HC group, congruent with previous reports [26, 29, 31]. Also, the AD and aMCI groups had the highest modularity; this is consistent with neurodegenerative conditions in which whole-brain networks fragment into a few large, isolated components [66, 67]. In addition, we found differences in the pattern of hub distribution, in both frontal-executive and corticolimbic circuitries across groups pronounced in the frontal regions, where the number of high-level centrality frontal hub regions decreased with the putative risk for AD. The higher frontal centrality hubs in the HC is consistent with a cognitive reserve framework [68], as the brain hubs are metabolically costly and vulnerable to pathology in the other groups. The AD group had the highest number of temporo-parietal hub regions, followed by the aMCI and MDD+aMCI groups. This finding is congruent with previous findings of greater local deposition of amyloid protein in these regions [69] and could imply that the neuropathology of Alzheimer’s disease is taking another route.

The absence of differences in DWI measures between the MDD and HC groups was contrary to our hypothesis and some previous studies [24, 70,71,72] but it has been previously reported in several other studies of white matter [17, 23, 73,74,75,76,77]. Mechanistic models linking MDD and AD suggest that frontal-executive white matter disruption is associated with white matter hyperintensities (WMH) [70, 78, 79]. However, recent studies found no difference in WMH between an MDD and HC groups [77, 80, 81] or between MDD and aMCI groups [25]. Furthermore, our groups were not different on most comorbid physical illnesses, with the exception of systolic blood pressure, which was driven only by the aMCI group.

We also did not find differences in gray matter ROIs between the MDD groups (with or without MCI) and the HC group. This lack of differences could be associated with: (1) effects of antidepressant medications on brain structure; (2) depression status and treatment response (remitting), or (3) age of onset of depression. First, antidepressants (especially serotonergic) have been shown to alter brain structure, potentially suggesting neurotrophic effects [82, 83]. In the present study, 87% of participants with remitted MDD had been exposed to serotonergic antidepressants, and more than three-quarters (77.9%) were still taking them. Thus, it is possible that exposure to antidepressants prevented some brain alterations in our participants with MDD. Second, all our participants with MDD (with or without MCI) were in remission and had responded to treatment. However, the majority of studies in older patients with MDD have been conducted during an acute episode (i.e., lower hippocampal [16, 84,85,86,87,88], and frontal and medial temporal lobe volume [84, 89, 90]). In the limited number of studies conducted in older individuals with remitted MDD, most have reported no differences between MDD and HC [80, 91, 92]. Only one study reported significantly smaller hippocampal volume in older individuals with remitted MDD compared with HCs [93]. Those with remitted MDD vs. non-remitted MDD have been shown to have less atrophy of the hippocampus over 2 years [81], and thicker frontal gyrus [94]. Third, nearly all our participants with MDD had experienced onset of depression prior to age 60, i.e., they had early-onset depression (EOD) [95, 96]. Thus far, only a few neuroimaging studies have assessed the differences between EOD and late-onset depression (LOD). These few studies have reported that compared with LOD, EOD shows unilateral hippocampal volume loss [85], less WMH burden [91], greater cerebrum volume, and significantly slower hippocampal volume loss over 4 years [97]. Thus, the lack of differences between our MDD groups (with or without MCI) and HC groups may be due to the fact that most of our participants with MDD had EOD. However, we found no associations between age of onset and any structural brain measure in our sample.

Overall, our results show considerable heterogeneity among older individuals with MDD and those with cognitive disorders. Alterations associated with aMCI and AD in frontal-executive and corticolimbic circuitries are not present in those with remitted MDD in our sample. Also, it is possible that recovering from depression may decrease the risk of developing AD. Some recent studies have shown that persistent depressive symptoms in the general population explain the association between MDD and the risk of incident AD [98, 99]. Even if older individuals with a remitted MDD remain a higher risk for AD, our findings suggest that the mechanism underlying this risk may not be the same as the mechanism linking aMCI and AD.

The naMCI or MDD+naMCI groups did not show the same brain alterations as the aMCI or AD groups, even though one might expect the frontal-executive circuitry to be equally or more structurally compromised in the naMCI group than the aMCI group. While previous studies have reported substantial biological heterogeneity in patients with naMCI and aMCI [33, 34], only a few have addressed structural brain differences in these MCI subtypes. These studies found no consistency in structural brain alterations in naMCI, aMCI, and AD [100, 101]. We do not think our findings were the result of misdiagnosis due to our rigorous consensus conferences.

Strengths and limitations

Our study has unique strengths, the first of which is the relatively large sample size for this type of neuroimaging study. Second, we included seven groups across a putative spectrum of AD risk ranging from HC to AD. Third, the diagnosis of the at-risk participants was based on a rigorous consensus conference using the results of both a comprehensive clinical assessment and cognitive testing. Finally, we examined multiple measures of brain structure, including white and gray matter circuitry and cortical topology to provide a comprehensive picture of structural brain differences across these populations.

Our study also has several limitations. Despite best efforts to account for their effects, some variables may have influenced our results. They include effects of WMH and antidepressant medications, all MDD participants being in remission, and most of them having recurrent EOD. By including only those with remitted MDD, we could not make conclusions about those with an acute major depressive episode or treatment-resistant MDD. However, we were able to conclude that those with remitted MDD did not have brain structure alterations like those with aMCI or AD. Finally, our MDD or MCI participants may represent a ‘healthier’ subset of individuals as they were also participants in a clinical trial which required a high level of motivation. This could in part account for the lack of difference in many of the brain structural measures between these participants and HCs. Finally, several participants were excluded due to poor quality of MRIs, which is a challenge with psychiatric neuroimaging studies, particularly in older adults [102].

Conclusion

Our results in AD and aMCI on both white and gray add to the wealth of literature showing alterations in frontal-executive and corticolimbic circuitries in these disorders. Our most notable results are that individuals with remitted MDD did not have brain alterations related to aMCI or AD. Although cross-sectional, these results suggest that individuals with remitted MDD may not be as much at-risk for AD as previously thought. Future work should follow these individuals longitudinally and contrast their brain structures and course with those with an acute major depressive episode or treatment-resistant MDD. Although speculative, demonstrating that successful treatment of MDD in late-life may prevent brain changes associated with MCI or AD, would be a major advance.

Funding and disclosure

This Project has been made possible by Brain Canada through the Canada Brain Research. Fund, with the financial support of Health Canada and the Chagnon Family. NRR is funded by the Alzheimer Society of Canada Research Program (ASRP) Doctoral Fellowship. ANV has received grant support from CIHR, NIH, the Brain and Behavior Research Foundation, CFI, the CAMH Foundation, and University of Toronto. AJF has received grant support from the US National Institutes of Health, the Patient-Centered Outcomes Research Institute, the Canadian Institutes of Health Research, Brain Canada, the Ontario Brain Institute, and the Alzheimer’s Association. BHM receives compensation from the Department of Psychiatry, University of Toronto, Toronto, ON; the Centre for Addiction and Mental Health (CAMH), Toronto, ON; and the University of Pittsburgh, Pittsburgh, PA. He also receives research support from Brain Canada, the Canadian Institutes of Health Research, the CAMH Foundation, the Patient-Centered Outcomes Research Institute (PCORI), the US National Institute of Health (NIH), Capital Solution Design LLC (software used in a study founded by CAMH Foundation), and HAPPYneuron (software used in a study founded by Brain Canada). He directly owns stocks of General Electric (less than $5000). Within the past 3 years, he has also received research support from Eli Lilly (medications for a NIH-funded clinical trial) and Pfizer (medications for a NIH-funded clinical trial). CEF has received grant funding from Hoffman LaRoche and Vielight Inc. EWD has received funding from NARSAD and NIMH. SK has received grant support from Brain Canada, NIH, Brain and Behavior Foundation (NARSAD), BrightFocus Foundation, Weston Brain Institute, Canadian Centre for Ageing and Brain Health Innovation, CAMH foundation, and University of Toronto, and in-kind equipment support from Soterix Medical Inc. TKR has received research support from Brain Canada, Brain and Behavior Research Foundation, BrightFocus Foundation, Canada Foundation for Innovation, Canada Research Chair, Canadian Institutes of Health Research, Centre for Aging and Brain Health Innovation, National Institutes of Health, Ontario Ministry of Health and Long-Term Care, Ontario Ministry of Research and Innovation, and the Weston Brain Institute. TKR also received in-kind equipment support for an investigator-initiated study from Magstim, and in-kind research accounts from Scientific Brain Training PRO. NRR, LM, NH, MAB, JAEA, and BGP have no potential conflicts of interest to declare.

References

  1. 1.

    Jorm AF. History of depression as a risk factor for dementia: an updated review. Aust N Z J Psychiatry. 2001;35:776–81.

    CAS  PubMed  Google Scholar 

  2. 2.

    Diniz BS, Butters MA, Albert SM, Dew MA, Reynolds CF. Late-life depression and risk of vascular dementia and Alzheimer’s disease: systematic review and meta-analysis of community-based cohort studies. Br J Psychiatry. 2013;202:329–35. https://doi.org/10.1192/bjp.bp.112.118307.

    Article  PubMed  PubMed Central  Google Scholar 

  3. 3.

    Herrmann LL, Goodwin GM, Ebmeier KP. The cognitive neuropsychology of depression in the elderly. Psychol Med. 2007;37:1693–702.

    PubMed  Google Scholar 

  4. 4.

    Koenig AM, Delozier IJ, Zmuda MD, Marron MM, Begley AE, Anderson SJ, et al. Neuropsychological functioning in the acute and remitted states of late-life depression. Handb Depress Alzheimer’s Dis. 2015;45:95–106.

    Google Scholar 

  5. 5.

    Sheline YI, Barch DM, Garcia K, Gersing K, Pieper C, Welsh-Bohmer K, et al. Cognitive function in late life depression: relationships to depression severity, cerebrovascular risk factors and processing speed. Biol Psychiatry. 2006;60:58–65.

    PubMed  Google Scholar 

  6. 6.

    Strømnes Dybedal G, Tanum L, Sundet K, Gaarden TL, Bjølseth TM, Schmid MT, et al. Neuropsychological functioning in late-life depression. Front Psychol. 2013;4:381. https://doi.org/10.3389/fpsyg.2013.00381.

    Article  Google Scholar 

  7. 7.

    Barnes DE, Yaffe K. The projected effect of risk factor reduction on Alzheimer’s disease prevalence. Lancet Neurol. 2011;10:819–28.

    PubMed  PubMed Central  Google Scholar 

  8. 8.

    Bhalla RK, Butters MA, Mulsant BH, Begley AE, Zmuda MD, Schoderbek B, et al. Persistence of neuropsychologic deficits in the remitted state of late-life depression. Am J Geriatr Psychiatry. 2006;14:419–27.

    PubMed  Google Scholar 

  9. 9.

    Butters MA, Becker JT, Nebes RD, Zmuda MD, Benoit Mulsant BH, Pollock BG, et al. Changes in cognitive functioning following treatment of late-life depression. Am J Psychiatry. 2000;15712:1949–54.

    Google Scholar 

  10. 10.

    Gallagher D, Kiss A, Lanctot K, Herrmann N. Depression and risk of alzheimer dementia: a longitudinal analysis to determine predictors of increased risk among older adults with depression. Am J Geriatr Psychiatry. 2018;26:819–27. https://doi.org/10.1016/j.jagp.2018.05.002.

    Article  PubMed  PubMed Central  Google Scholar 

  11. 11.

    Butters MA, Young JB, Lopez O, Aizenstein HJ, Mulsant BH, Reynolds CF, et al. (2008). Pathways linking late-life depression to persistent cognitive impairment and dementia. Dialogues Clin Neurosci. https://doi.org/10.1016/j.bbi.2008.05.010.

  12. 12.

    Schroeter ML, Stein T, Maslowski N, Neumann J. Neural correlates of Alzheimer’s disease and mild cognitive impairment: a systematic and quantitative meta-analysis involving 1351 patients. Neuroimage. 2009;47:1196–206.

    PubMed  PubMed Central  Google Scholar 

  13. 13.

    Whitwell JL, Przybelski SA, Weigand SD, Knopman DS, Boeve BF, Petersen RC, et al. 3D maps from multiple MRI illustrate changing atrophy patterns as subjects progress from mild cognitive impairment to Alzheimer’s disease. Brain. 2007;130:1777–86.

    PubMed  PubMed Central  Google Scholar 

  14. 14.

    Wolk DA, Dickerson BC. Apolipoprotein E (APOE) genotype has dissociable effects on memory and attentional-executive network function in Alzheimer’s disease. Proc Natl Acad Sci USA. 2010;107:10256–61.

    CAS  PubMed  Google Scholar 

  15. 15.

    Du M, Liu J, Chen Z, Huang X, Li J, Kuang W, et al. Brain grey matter volume alterations in late-life depression. J Psychiatry Neurosci. 2014;39:397–406.

    PubMed  PubMed Central  Google Scholar 

  16. 16.

    Joko T, Washizuka S, Sasayama D, Inuzuka S, Ogihara T, Yasaki T, et al. Patterns of hippocampal atrophy differ among Alzheimer’s disease, amnestic mild cognitive impairment, and late-life depression. Psychogeriatrics. 2016;16:355–61.

    PubMed  Google Scholar 

  17. 17.

    Sexton CE, Allan CL, Le Masurier M, Bradley KM, Mackay CE, Ebmeier KP. Magnetic resonance imaging in late-life depression. 2012;69:680–9.

  18. 18.

    Shimoda K, Kimura M, Yokota M, Okubo Y. Comparison of regional gray matter volume abnormalities in Alzheimers disease and late life depression with hippocampal atrophy using VSRAD analysis: a voxel-based morphometry study. Psychiatry Res. 2015;232:71–5.

    PubMed  Google Scholar 

  19. 19.

    Sexton CE, Phil D, Mackay CE, Ph D, Ebmeier KP. A systematic review and meta-analysis of magnetic resonance imaging studies in late-life depression. Am J Geriatr Psychiatry. 2013;21:184–95.

    PubMed  Google Scholar 

  20. 20.

    Tadayonnejad R, Ajilore O. Brain network dysfunction in late-life depression: a literature review. J Geriatr Psychiatry Neurol. 2014;27(Mar):5–12. https://doi.org/10.1177/0891988713516539.

    Article  PubMed  Google Scholar 

  21. 21.

    Warren D Taylor. Hippocampus atrophy and the longitudinal course of late-life depression. Am J Geriatr Psychiatry. 2014. https://doi.org/10.1021/nl061786n.Core-Shell.

  22. 22.

    Yin Y, Hou Z, Wang X, Sui Y, Yuan Y. Association between altered resting-state cortico-cerebellar functional connectivity networks and mood/cognition dysfunction in late-onset depression. J Neural Transm. 2015;122:887–96.

    PubMed  Google Scholar 

  23. 23.

    Emsell L, Adamson C, De Winter FL, Billiet T, Christiaens D, Bouckaert F, et al. Corpus callosum macro and microstructure in late-life depression. J Affect Disord. 2017;222:63–70.

    PubMed  Google Scholar 

  24. 24.

    Li W, Muftuler LT, Chen G, Ward BD, Budde MD, Jones JL, et al. Effects of the coexistence of late-life depression and mild cognitive impairment on white matter microstructure. J Neurol Sci. 2014;338:46–56.

    PubMed  Google Scholar 

  25. 25.

    Liao W, Zhang X, Shu H, Wang Z, Liu D, Zhang ZJ. The characteristic of cognitive dysfunction in remitted late life depression and amnestic mild cognitive impairment. Psychiatry Res. 2017;251:168–75.

    PubMed  Google Scholar 

  26. 26.

    Bai F, Shu N, Yuan Y, Shi Y, Yu H, Wu D, et al. Topologically convergent and divergent structural connectivity patterns between patients with remitted geriatric depression and amnestic mild cognitive impairment. J Neurosci. 2012;32:4307–18.

    CAS  PubMed  PubMed Central  Google Scholar 

  27. 27.

    Li W, Ward BD, Liu X, Chen G, Jones JL, Antuono PG, et al. Disrupted small world topology and modular organisation of functional networks in late-life depression with and without amnestic mild cognitive impairment. J Neurol Neurosurg Psychiatry. 2015;86:1097–105.

    PubMed  Google Scholar 

  28. 28.

    Mai N, Zhong X, Chen B, Peng Q, Wu Z, Zhang W, et al. Weight rich-club analysis in the white matter network of late-life depression with memory deficits. Front Aging Neurosci. 2017;9:279. https://doi.org/10.3389/fnagi.2017.00279.

    Article  PubMed  PubMed Central  Google Scholar 

  29. 29.

    Ajilore O, Lamar M, Leow A, Zhang A, Yang S, Kumar A. Graph theory analysis of cortical-subcortical networks in late-life depression. Am J Geriatr Psychiatry. 2014;22:195–206.

    PubMed  Google Scholar 

  30. 30.

    Lim HK, Jung WS, Aizenstein HJ. Aberrant topographical organization in gray matter structural network in late life depression: a graph theoretical analysis. Int Psychogeriatr. 2013;25:1929–40.

    PubMed  Google Scholar 

  31. 31.

    Mak E, Colloby SJ, Thomas A, O’Brien JT. The segregated connectome of late-life depression: a combined cortical thickness and structural covariance analysis. Neurobiol Aging. 2016;48:212–21.

    PubMed  PubMed Central  Google Scholar 

  32. 32.

    Miller MD, Paradis CF, Houck PR, Mazumdar S, Stack JA, Rifai AH, et al. Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale. Psychiatry Res. 1992;41:237–48. https://doi.org/10.1016/0165-1781(92)90005-N.

    CAS  Article  PubMed  Google Scholar 

  33. 33.

    Allen JW, Yazdani M, Kang J, Magnussen MJ, Qiu D, Hu W. Patients with mild cognitive impairment may be stratified by advanced diffusion metrics and neurocognitive testing. J Neuroimaging. 2019;29:79–84.

    PubMed  Google Scholar 

  34. 34.

    Ezzati A, Zammit AR, Habeck C, Hall CB, Lipton RB. Detecting biological heterogeneity patterns in ADNI amnestic mild cognitive impairment based on volumetric MRI. Brain Imaging Behav. 2019. https://doi.org/10.1007/s11682-019-00115-6.

  35. 35.

    Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–98.

    CAS  PubMed  PubMed Central  Google Scholar 

  36. 36.

    Nasreddine ZS, Phillips NA, Bédirian V, Charbonneau S, Whitehead V, Collin I, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53:695–9.

    PubMed  Google Scholar 

  37. 37.

    Andersson JLR, Sotiropoulos SN. An integrated approach to correction for off-resonance effects and subject movement in diffusion MR imaging. Neuroimage. 2016;125:1063–78.

    PubMed  PubMed Central  Google Scholar 

  38. 38.

    Jenkinson M, Beckmann CF, Behrens TEJ, Woolrich MW, Smith SM. Review FSL. Neuroimage. 2012;62:782–90.

    PubMed  Google Scholar 

  39. 39.

    Bhushan C, Haldar JP, Joshi AA, Leahy RM Correcting Susceptibility-Induced Distortion in Diffusion-Weighted MRI using Constrained Nonrigid Registration. Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA). 2012. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC4708288.

  40. 40.

    Behrens TEJ, Woolrich MW, Jenkinson M, Johansen-Berg H, Nunes RG, Clare S, et al. Characterization and propagation of uncertainty in diffusion-weighted MR imaging. Magn Reson Med. 2003;50:1077–88.

    CAS  PubMed  Google Scholar 

  41. 41.

    Behrens TEJ, Berg HJ, Jbabdi S, Rushworth MFS, Woolrich MW. Probabilistic diffusion tractography with multiple fibre orientations: what can we gain? Neuroimage. 2007;34:144–55.

    CAS  PubMed  Google Scholar 

  42. 42.

    Smith SM, Johansen-Berg H, Jenkinson M, Rueckert D, Nichols TE, Klein JC, et al. Acquisition and voxelwise analysis of multi-subject diffusion data with tract-based spatial statistics. Nat Protoc. 2007;2:499–503.

    PubMed  Google Scholar 

  43. 43.

    Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis: I. Segmentation and surface reconstruction. Neuroimage. 1999;9:179–94.

    CAS  PubMed  Google Scholar 

  44. 44.

    Fischl B, Sereno MI, Dale AM. Cortical surface-based analysis: II. Inflation, flattening, and a surface-based coordinate system. Neuroimage. 1999;9:195–207.

    CAS  PubMed  Google Scholar 

  45. 45.

    Fischl B, Liu A, Dale AM. Automated manifold surgery: constructing geometrically accurate and topologically correct models of the human cerebral cortex. IEEE Trans Med Imaging. 2001;20:70–80.

    CAS  PubMed  Google Scholar 

  46. 46.

    Ségonne F, Dale AM, Busa E, Glessner M, Salat D, Hahn HK, et al. A hybrid approach to the skull stripping problem in MRI. Neuroimage. 2004;22:1060–75.

    PubMed  Google Scholar 

  47. 47.

    Fischl B, Dale AM. Measuring the thickness of the human cerebral cortex from magnetic resonance images. Proc Natl Acad Sci USA. 2000;97:11050–5.

    CAS  PubMed  Google Scholar 

  48. 48.

    Killiany RJ, Dale AM, Ségonne F, Desikan RS, Blacker D, Buckner RL, et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage. 2006;31:968–80.

    PubMed  Google Scholar 

  49. 49.

    Fischl B, Salat DH, Busa E, Albert M, Dieterich M, Haselgrove C, et al. Whole brain segmentation: Automated labeling of neuroanatomical structures in the human brain. Neuron. 2002;33:341–55.

    CAS  PubMed  Google Scholar 

  50. 50.

    Fischl B, Salat DH, Van Der Kouwe AJW, Makris N, Ségonne F, Quinn BT, et al. Sequence-independent segmentation of magnetic resonance images. Neuroimage. 2004;23. https://doi.org/10.1016/j.neuroimage.2004.07.016.

  51. 51.

    Benjamini Yoav, Hochberg Yosef. Controlling the FDR: a practical and powerful approach to multiple testing. J R Stat Soc Ser B. 1995;57:289–300.

    Google Scholar 

  52. 52.

    Bullmore E, Sporns O. Complex brain networks: graph theoretical analysis of structural and functional systems. Nat Rev Neurosci. 2009;10:186–98.

    CAS  PubMed  Google Scholar 

  53. 53.

    Christopher G, Watson (2018). brainGraph: graph theory analysis of brain MRI data. R package version 2.7.0. https://github.com/cwatson/brainGraph.

  54. 54.

    van Wijk BCM, Stam CJ, Daffertshofer A. Comparing brain networks of different size and connectivity density using graph theory. PLoS ONE. 2010;5. https://doi.org/10.1371/journal.pone.0013701.

  55. 55.

    van den Heuvel MP, Sporns O. Network hubs in the human brain. Trends Cogn Sci. 2013;17:683–96.

    PubMed  Google Scholar 

  56. 56.

    Apostolova LG, Steiner CA, Akopyan GG, Dutton RA, Hayashi KM, Toga AW, et al. Three-dimensional gray matter atrophy mapping in mild cognitive impairment and mild Alzheimer disease. Arch Neurol. 2007;64:1489–95.

    PubMed  PubMed Central  Google Scholar 

  57. 57.

    Zhang Y, Schuff N, Jahng GH, Bayne W, Mori S, Schad L, et al. Diffusion tensor imaging of cingulum fibers in mild cognitive impairment and Alzheimer disease. Neurology. 2007;68:13–19.

    CAS  PubMed  PubMed Central  Google Scholar 

  58. 58.

    Chua TC, Wen W, Slavin MJSP. Diffusion tensor imaging in mild cognitive impairment and Alzheimer’s disease: a review. Curr Opin Neurol. 2008;21:83–92.

    PubMed  Google Scholar 

  59. 59.

    Schuff N, Woemer N, Boreta L, Kornfield T, Shaw LM, Trojanowski JQ, et al. MRI of hippocampal volume loss in early Alzheimer’s disease in relation to ApoE genotype and biomarkers. Brain. 2009;132:1067–77.

    CAS  PubMed  PubMed Central  Google Scholar 

  60. 60.

    Di Paola M, Luders E, Di Iulio F, Cherubini A, Passafiume D, Thompson PM, et al. Callosal atrophy in mild cognitive impairment and Alzheimer’s disease: different effects in different stages. Neuroimage. 2010;49:141–9.

    PubMed  Google Scholar 

  61. 61.

    Yang J, Pan P, Song W, Huang R, Li J, Chen K, et al. Voxelwise meta-analysis of gray matter anomalies in Alzheimer’s disease and mild cognitive impairment using anatomic likelihood estimation. J Neurol Sci. 2012;316:21–29.

    PubMed  Google Scholar 

  62. 62.

    Gyebnár G, Szabó Á, Sirály E, Fodor Z, Sákovics A, Salacz P, et al. What can DTI tell about early cognitive impairment? – differentiation between MCI subtypes and healthy controls by diffusion tensor imaging. Psychiatry Res Neuroimaging. 2018;272:46–57.

    PubMed  Google Scholar 

  63. 63.

    Liu J, Yin C, Xia S, Jia L, Guo Y, Zhao Z, et al. White Matter Changes in Patients with Amnestic Mild Cognitive Impairment Detected by Diffusion Tensor Imaging. PLoS ONE. 2013;8. https://doi.org/10.1371/journal.pone.0059440.

  64. 64.

    Kantarci K, Avula RT, Senjem ML, Samikoglu AR, Shiung MM, Przybelski SA, et al. Diffusion tensor imaging characteristics of amnestic and non-amnestic mild cognitive impairment. Alzheimer’s Dement. 2009;5:P12.

    Google Scholar 

  65. 65.

    Zheng D, Sun H, Dong X, Liu B, Xu Y, Chen S, et al. Executive dysfunction and gray matter atrophy in amnestic mild cognitive impairment. Neurobiol Aging. 2014;35:548–55.

    PubMed  Google Scholar 

  66. 66.

    Baggio HC, Sala-Llonch R, Segura B, Marti MJ, Valldeoriola F, Compta Y, et al. Functional brain networks and cognitive deficits in Parkinson’s disease. Hum Brain Mapp. 2014;35:4620–34.

    PubMed  PubMed Central  Google Scholar 

  67. 67.

    Pereira JB, Mijalkov M, Kakaei E, Mecocci P, Vellas B, Tsolaki M, et al. Disrupted network topology in patients with stable and progressive mild cognitive impairment and Alzheimer’s Disease. Cereb Cortex. 2016;26:3476–93.

    PubMed  PubMed Central  Google Scholar 

  68. 68.

    Greenwood PM. The frontal aging hypothesis evaluated. J Int Neuropsychol Soc. 2000;6:705–26.

    CAS  PubMed  Google Scholar 

  69. 69.

    Buckner RL, Sepulcre J, Talukdar T, Krienen FM, Liu H, Hedden T, et al. Cortical hubs revealed by intrinsic functional connectivity: mapping, assessment of stability, and relation to Alzheimer’s disease. J Neurosci. 2009;29:1860–73.

    CAS  PubMed  PubMed Central  Google Scholar 

  70. 70.

    Shimony JS, Sheline YI, Angelo GD, Epstein AA, Tammie LS, Mintun MA, et al. Diffuse microstructural abnormalities of normal appearing white matter in late life depression. Biol Psychiatry. 2010;66:245–52.

    Google Scholar 

  71. 71.

    Yuan Y, Hou Z, Zhang Z, Bai F, Yu H, You J, et al. Abnormal integrity of long association fiber tracts is associated with cognitive deficits in patients with remitted geriatric depression: a cross-sectional, case-control study. J Clin Psychiatry. 2010;71:1386–90.

    PubMed  Google Scholar 

  72. 72.

    Alves GS, Karakaya T, Fußer F, Kordulla M, O’Dwyer L, Christl J, et al. Association of microstructural white matter abnormalities with cognitive dysfunction in geriatric patients with major depression. Psychiatry Res Neuroimaging. 2012;203:194–200.

    Google Scholar 

  73. 73.

    Colloby SJ, Firbank MJ, Thomas AJ, Vasudev A, Parry SW, O’Brien Michael J, et al. White matter changes in late-life depression: a diffusion tensor imaging study. J Affect Disord. 2011;135:216–20.

    PubMed  Google Scholar 

  74. 74.

    Bezerra DM, Pereira FRS, Cendes F, Jackowski MP, Nakano EY, Moscoso MAA, et al. DTI voxelwise analysis did not differentiate older depressed patients from older subjects without depression. J Psychiatr Res. 2012;46:1643–9.

    PubMed  Google Scholar 

  75. 75.

    Mettenburg JM, Benzinger TL, Shimony JS, Snyder AZ, Sheline YI. Diminished performance on neuropsychological testing in late life depression is correlated with microstructural white matter abnormalities. Neuroimage. 2012;60:2182–90. https://doi.org/10.1016/j.neuroimage.2012.02.044.

    Article  PubMed  PubMed Central  Google Scholar 

  76. 76.

    Tadayonnejad R, Yang S, Kumar A, Ajilore O. Multimodal brain connectivity analysis in unmedicated late-life depression. PLoS ONE. 2014;9. https://doi.org/10.1371/journal.pone.0096033.

  77. 77.

    Harada K, Matsuo K, Nakashima M, Hobara T, Higuchi N, Higuchi F, et al. Disrupted orbitomedial prefrontal limbic network in individuals with later-life depression. J Affect Disord. 2016;204:112–9.

    PubMed  Google Scholar 

  78. 78.

    Kohler S, Thomas AJ, Lloyd A, Barber R, Almeida OP, O’Brien John T. White matter hyperintensities, cortisol levels, brain atrophy and continuing cognitive deficits in late-life depression. Br J Psychiatry. 2010;196:143–9.

    PubMed  Google Scholar 

  79. 79.

    Diniz BS, Sibille E, Ding Y, Tseng G, Aizenstein HJ, Lotrich F, et al. Plasma biosignature and brain pathology related to persistent cognitive impairment in late-life depression. Mol Psychiatry. 2015;20:594–601. https://doi.org/10.1038/mp.2014.76.

    CAS  Article  PubMed  Google Scholar 

  80. 80.

    Weber K, Giannakopoulos P, Delaloye C, de Bilbao F, Moy G, Moussa A, et al. Volumetric MRI changes, cognition and personality traits in old age depression. J Affect Disord. 2010;124:275–82.

    PubMed  Google Scholar 

  81. 81.

    Taylor WD, McQuoid DR, Payne ME, Zannas AS, MacFall JR, Steffens DC. Hippocampus atrophy and the longitudinal course of late-life depression. Am J Geriatr Psychiatry. 2014;22:1504–12.

    PubMed  Google Scholar 

  82. 82.

    Cirrito JR, Disabato BM, Restivo JL, Verges DK, Goebel WD, Sathyan A, et al. Serotonin signaling is associated with lower amyloid- levels and plaques in transgenic mice and humans. Proc Natl Acad Sci USA. 2011;108:14968–73.

    CAS  PubMed  Google Scholar 

  83. 83.

    Ren QG, Wang YJ, Gong WG, Xu L, Zhang ZJ. Escitalopram ameliorates tau hyperphosphorylation and spatial memory deficits induced by protein kinase A activation in Sprague Dawley rats. J Alzheimer’s Dis. 2015;47:61–71.

    CAS  Google Scholar 

  84. 84.

    Andreescu C, Butters MA, Begley A, Rajji T, Wu M, Meltzer CC, et al. Gray matter changes in late life depression - a structural MRI analysis. Neuropsychopharmacology. 2008;33:2566–72.

    PubMed  Google Scholar 

  85. 85.

    Ballmaier M, Narr KL, Toga AW, Elderkin-Thompson V, Thompson PM, Hamilton L, et al. Hippocampal morphology and distinguishing late-onset from early-onset elderly depression. Am J Psychiatry. 2008;165:229–37.

    PubMed  Google Scholar 

  86. 86.

    Steffens DC, McQuoid DR, Payne ME, Potter GG. Change in hippocampal volume on magnetic resonance imaging and cognitive decline among older depressed and nondepressed subjects in the neurocognitive outcomes of depression in the elderly study. Am J Geriatr Psychiatry. 2011;19:4–12.

    PubMed  PubMed Central  Google Scholar 

  87. 87.

    Lim HK, Hong SC, Jung WS, Ahn KJ, Won WY, Hahn C, et al. Automated hippocampal subfields segmentation in late life depression. J Affect Disord. 2012;143:253–6.

    PubMed  Google Scholar 

  88. 88.

    De Winter F, Emsell L, Ph D, Bouckaert F, Claes L, Sc M, et al. No association of lower hippocampal volume with Alzheimer’s disease pathology in late-life depression. Am J Psychiatry 2017;l:237–45.

    Google Scholar 

  89. 89.

    Lim HK, Jung WS, Ahn KJ, Won WY, Hahn C, Lee SY, et al. Regional cortical thickness and subcortical volume changes are associated with cognitive impairments in the drug-naive patients with late-onset depression. Neuropsychopharmacology. 2012;37:838–49.

    PubMed  Google Scholar 

  90. 90.

    Xie C, Li W, Chen G, Douglas Ward B, Franczak MB, Jones JL, et al. The co-existence of geriatric depression and amnestic mild cognitive impairment detrimentally affect gray matter volumes: Voxel-based morphometry study. Behav Brain Res. 2012;235:244–50.

    PubMed  PubMed Central  Google Scholar 

  91. 91.

    Delaloye C, Moy G, de Bilbao F, Baudois S, Weber K, Hofer F, et al. Neuroanatomical and neuropsychological features of elderly euthymic depressed patients with early- and late-onset. J Neurol Sci. 2010;299:19–23.

    PubMed  Google Scholar 

  92. 92.

    Weber K, Giannakopoulos P, Delaloye C, de Bilbao F, Moy G, Ebbing K, et al. Personality traits, cognition and volumetric MRI changes in elderly patients with early-onset depression: A 2-year follow-up study. Psychiatry Res. 2012;198:47–52.

    PubMed  Google Scholar 

  93. 93.

    Choi WH, Jung WS, Um YH, Lee CU, Park YH, Lim HK. Cerebral vascular burden on hippocampal subfields in first-onset drug-naïve subjects with late-onset depression. J Affect Disord. 2017;208:47–53.

    PubMed  Google Scholar 

  94. 94.

    Liu J, Xu X, Luo Q, Luo Y, Chen Y, Lui S, et al. Brain grey matter volume alterations associated with antidepressant response in major depressive disorder. Sci Rep. 2017;7:10464. https://doi.org/10.1038/s41598-017-10676-5.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  95. 95.

    Brodaty H, Luscombe G, Parker G, Wilhelm K, Hickie I, Austin MP, et al. Early and late onset depression in old age: different aetiologies, same phenomenology. J Affect Disord. 2001;66:225–36.

    CAS  PubMed  Google Scholar 

  96. 96.

    Hashem AH, Nasreldin M, Gomaa MA, Khalaf OO. Late versus early onset depression in elderly patients: vascular risk and cognitive impairment. Curr Aging Sci. 2017;10. https://doi.org/10.2174/1874609810666170404105634.

  97. 97.

    Sachs-Ericsson N, Corsentino E, Moxley J, Hames JL, Collins N, Sawyer K, et al. A longitudinal study of differences in late and early onset geriatric depression: depressive symptoms and psychosocial, cognitive, and neurological functioning. Aging Ment Health. 2013;17:1–11.

    PubMed  Google Scholar 

  98. 98.

    Kaup AR, Byers AL, Falvey C, Simonsick EM, Satterfield S, Ayonayon HN, et al. Trajectories of depressive symptoms in older adults and risk of dementia. JAMA Psychiatry. 2016;73:525–31.

    PubMed  PubMed Central  Google Scholar 

  99. 99.

    Mirza SS, Wolters FJ, Swanson SA, Koudstaal PJ, Hofman A, Tiemeier H, et al. 10-year trajectories of depressive symptoms and risk of dementia: a population-based study. Lancet Psychiatry. 2016;3:628–35.

    PubMed  Google Scholar 

  100. 100.

    Kantarci K, Petersen RC, Przybelski SA, Weigand SD, Shiung MM, Whitwell JL, et al. Hippocampal volumes, proton magnetic resonance spectroscopy metabolites, and cerebrovascular disease in mild cognitive impairment subtypes. Arch Neurol. 2008;65:1621–8.

    PubMed  PubMed Central  Google Scholar 

  101. 101.

    Csukly G, Sirály E, Fodor Z, Horváth A, Salacz P, Hidasi Z, et al. The differentiation of amnestic type MCI from the non-amnestic types by structural MRI. Front Aging Neurosci. 2016;8. https://doi.org/10.3389/fnagi.2016.00052.

  102. 102.

    Madan CR. Age differences in head motion and estimates of cortical morphology. PeerJ. 2018. https://doi.org/10.7717/peerj.5176.

  103. 103.

    Mowinckel AM, Vidal-Piñeiro D. Visualisation of Brain Statistics with R-packages ggseg and ggseg3d. 2019. arXiv:1912.08200.

  104. 104.

    Xia M, Wang J, He Y. BrainNet viewer: a network visualization tool for human brain connectomics. PLoS ONE. 2013;8. https://doi.org/10.1371/journal.pone.0068910.

Download references

Acknowledgements

BGP acknowledges the Peter & Shelagh Godsoe Endowed Chair in Late-Life Mental Health. In addition, we thank other members of PACt-MD Study Group; Lillian Lourenco, Daniel M. Blumberger, Christopher R. Bowie, Damian Gallagher, Angela Golas, Ariel Graff, James L. Kennedy, Shima Ovaysikia, Mark Rapoport, Kevin Thorpe, and Nicolaas P.L.G. Verhoeff.

Author information

Affiliations

Authors

Contributions

NRR: Substantial contributions to the conception or design of the work and the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; final approval of the version to be published; and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. TKR: Substantial contributions to the conception or design of the work; the acquisition, analysis, or interpretation of data for the work; and final approval of the version to be published. SK: Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; and final approval of the version to be published. NH: Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; and final approval of the version to be published. LM: Substantial contributions to the acquisition, analysis, or interpretation of data for the work; and final approval of the version to be published. AJF Substantial contributions to the acquisition, analysis, or interpretation of data for the work; and final approval of the version to be published. CEF: Substantial contributions to the acquisition, analysis, or interpretation of data for the work; and final approval of the version to be published. MAB: Substantial contributions to the acquisition, analysis, or interpretation of data for the work; and final approval of the version to be published. BGP: Substantial contributions to the conception or design of the work; the acquisition, analysis, or interpretation of data for the work; and final approval of the version to be published. EWD: Substantial contributions to the acquisition, analysis, or interpretation of data for the work; and final approval of the version to be published. JAEA: Drafting the work or revising it critically for important intellectual content; and final approval of the version to be published. BHM: Substantial contributions to the conception or design of the work; the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; and final approval of the version to be published. ANV: Substantial contributions to the conception or design of the work; the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; final approval of the version to be published; and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Corresponding author

Correspondence to Aristotle N. Voineskos.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Rashidi-Ranjbar, N., Rajji, T.K., Kumar, S. et al. Frontal-executive and corticolimbic structural brain circuitry in older people with remitted depression, mild cognitive impairment, Alzheimer’s dementia, and normal cognition. Neuropsychopharmacol. 45, 1567–1578 (2020). https://doi.org/10.1038/s41386-020-0715-y

Download citation

Further reading

Search

Quick links