Skip to main content
Download PDF

Effects of glycaemic control on memory performance, hippocampal volumes and depressive symptomology

Diabetology & Metabolic Syndrome volume 16, Article number: 219 (2024) Cite this article

Abstract

Background

Diabetes and poor glycaemic control have been shown to negatively impact cognitive abilities, while also raising risk of both mood disorders and brain structural atrophy. Sites of atrophy include the hippocampus, which has been implicated in both memory performance and depression. The current study set out to better characterise the associations between poor glycaemic control, memory performance, and depression symptoms, and investigate whether loss of hippocampal volume could represent a neuropathological mechanism underlying these.

Methods

1331 participants (60.9% female, age range 18–88 (Mean = 44.02), 6.5% with likely diabetes) provided HbA1c data (as an index of glycaemic control), completed a word list learning task, and a validated depression scale. A subsample of 392 participants underwent structural MRI; hippocampal volumes were extracted using FreeSurfer.

Results

Partial correlation analyses (controlling for age, gender, and education) showed that, in the full sample, poorer glycaemic control was related to lower word list memory performance. In the MRI sub-sample, poorer glycaemic control was related to higher depressive symptoms, and lower hippocampal volumes. Total hippocampus volume partially mediated the association between HbA1c levels and depressive symptoms.

Conclusions

Results emphasise the impact of glycaemic control on memory, depression and hippocampal volume and suggest hippocampal volume loss could be a pathophysiological mechanism underlying the link between HbA1c and depression risk; inflammatory and stress-hormone related processes might have a role in this.

Background

Impaired glycaemic control has been linked to poorer cognitive function and increased risk of depressive symptomology. Haemoglobin A1c (HbA1c) in blood plasma is a commonly utilised biomarker of glycaemic control: it reflects mean blood glucose levels over the previous 2–3 months [1]. An HbA1c of 6.5% or above indicates a diabetes diagnosis according to World Health Organization (WHO) guidelines [2]. Diabetes is a pressing world health priority, with a global prevalence of 8.5% amongst adults [3]. Type 2 diabetes raises the risk of depression [4], cognitive impairment [5], and dementia [6].

Higher HbA1c levels have been linked to poorer cognitive performance in individuals with diabetes [7, 8] including delayed memory recall [9]. In individuals without diabetes, this is also the case. In healthy young adults (N = 1200, aged 22 to 35), HbA1c negatively impacted working memory and fluid intelligence scores [10]. In a sample (N = 600) including adults with normal glucose tolerance (34.6%), with pre-diabetes (34.5%), and with diabetes (30.8%) aged between 55 and 64, HbA1c correlated with word list recall performance (both immediate, and after a 15-minute delay) [11]. Further, in a longitudinal study of individuals aged 50+, higher HbA1c levels at baseline increased risk of subsequent memory decline regardless of diabetes status [12]; in participants with no diabetes aged over 75, higher baseline HbA1c levels has been linked to a greater decline in Mini-Mental State Exam scores at 32-month follow-up, even after excluding participants with suspected incipient diabetes [13]. Higher average blood glucose levels have also been shown to increase the risk of dementia in both individuals with diabetes and without diabetes, e.g. in adults over 65 followed up over 7 years [14].

However, the underlying mechanism behind the relationship between HbA1c levels and cognition is unclear. The hippocampus is a crucial structure supporting episodic encoding and retrieval [15, 16]. In participants with diabetes, studies have shown an inverse relationship between HbA1c levels and total hippocampal volume [17]. Garfield et al. [16] compared participants with diabetes/prediabetes, participants with normal HbA1c levels and participants with below-normal HbA1c levels (N = 35,418). Participants with diabetes and prediabetes had smaller hippocampal volume than participants with normal HbA1c levels, and participants with below-normal HbA1c levels had larger hippocampal volume than participants with normal HbA1c levels [18]. Research based on continuous, rather than categorical, measures of HbA1c levels is lacking, but impaired glycaemic control likely impacts hippocampal volume through a variety of biological mechanisms including the action of advanced glycation end-products (AGEs) [19,20,21], glycoxidative damage [22], and elevated cortisol levels [23].

Given the well-established role of the hippocampus in memory, some (but not all) studies have found links between hippocampal volumes and memory performance, for example in relation to delayed recall in healthy young adults [24]. A recent meta-analysis in children/adolescents found a small, but significant, positive association between total hippocampal volume and memory performance across both immediate and delayed recall [25]. A neuroimaging study spanning the entire adult age range found links between volumes of the hippocampal cornus ammonis subregions in particular, with both immediate and delayed recall performance [26]. Hardcastle et al. [27] linked larger left hippocampal volume with better working and episodic memory performance in adults aged 65–89, and loss of hippocampal volume over 4 years in older adults has been correlated with memory decline [28]. Thus, the impact of HbA1c levels on hippocampal volume could undermine memory performance and at least partially explain the observed associations between HbA1c levels and memory.

A relationship between HbA1c and depressive symptomology has been shown but primarily using diabetes groups. Meta-analyses show patients with diabetes have higher levels of both incident clinical depression and self-reported depressive symptoms [29], health complications associated with diabetes such as micro/macrovascular disease, weight gain, and neuropathy, have been theorised to contribute [30] Longitudinal studies have shown an association between HbA1c levels and self-reported depressive symptoms in diabetes patients aged > 50 [31]; depression also raises the risk of diabetes [32]. In one study including both healthy and individuals with diabetes aged > 50, increased HbA1c levels were linked to higher self-reported depressive symptoms at follow-up (eight years) [33], similar results were observed in another community-based sample spanning ages 18–72 [34]. However, more research is needed to clarify the mechanisms involved.

Again, the hippocampus might mediate the relationship between glycaemic control and depressive symptoms, as it is implicated in mood regulation [35], and hippocampal volume reduction is the most commonly reported structural imaging finding in major depressive disorder (MDD), with reductions of up to 10% (for meta-analyses see [36, 37]. Also, depressed patients with smaller hippocampal volume are less likely to respond to antidepressant treatment [38], and a longer duration of depression is associated with greater reductions in hippocampal volume [39]. Even in non-clinical samples, an inverse association between self-reported depressive symptomology and bilateral hippocampal volume has been shown [40, 41]; comparing the strength of the relationship within three different age ranges (those aged below 40, between 40 and 59, and 60 years and above) indicated it might be stronger in later life [42]. This might result from increased vulnerability of the aged hippocampus to the effect of glucocorticoids [43]. The hippocampus has a role in the regulation of the hypothalamic-pituitary-adrenal (HPA) axis, which produces stress-related glucocorticoids [44]. Reduced hippocampus volume has been linked to higher levels of circulating cortisol [45]. Animal studies suggest that this is because of impaired HPA regulation in response to stressors [46]. High levels of circulating glucocorticoids cause further dendritic shrinkage and loss of spines in hippocampus [47], setting up a feedback loop which could lead to further HPA impairment and further hippocampal damage. Reduced hippocampal volume predicts poorer outcomes in depressed patients, in terms of clinical response and relapse rate [48, 49], suggesting that lower hippocampal volume might be a vulnerability factor for MDD. Thus, hippocampal volume loss due to glucose dysregulation via the mechanisms discussed above, could represent a mediating pathway and mechanism, explaining the link between HbA1c and depressive symptoms, although this has not been explicitly studied.

In sum, research suggests that higher HbA1c levels are associated with both poorer memory and depression symptomology. However, previous work has tended to focus on older people with diabetes. The question of whether these associations are generalisable across the age range remains unanswered. Also, while higher HbA1c levels impact hippocampal volume, and its effects on this brain structure could be mediate its associations with cognition and depressive symptoms, this has not been explicitly investigated. Thus, the present study examined associations between HbA1c levels and both memory performance and depressive symptoms in a sizeable community-based sample spanning a wide adult age range. A subsample also provided magnetic resonance imaging (MRI) data. The data came from the Baependi Heart Study, a longitudinal cohort study that began in 2005 [50]. Hippocampal volume was extracted and associations with HbA1c levels, memory performance, and depressive symptoms were assessed. In contrast to most previous work, we took a dimensional rather than a categorical approach to quantifying depression and glycaemic control, with continuous measures of these employed in all analyses. We hypothesised that higher levels of HbA1c would be correlated with lower immediate memory recall performance, higher depressive symptoms, and smaller left, right, and total hippocampal volumes. In addition, given that some of the studies outlined above point to a possible effect of age, as a follow-up analysis, we examined whether the relationship between HbA1c levels and hippocampus volume, and depressive symptoms, might be stronger in those amongst the sample aged over 50. We used this age point as a cut-off so as to align with various previous relevant studies focusing on this age range, for example work exploring effects of diabetes and HbA1c level on depression symptoms [51], and impact of HbA1c on cognition and brain health [52], to specifically test the possibility that, in the ageing brain, the hippocampus might be more vulnerable to the toxic effects of HbA1c. Then, a series of mediation models were run to test the hypotheses that hippocampus volume mediates the association between HbA1c and depressive symptoms and memory performance.

Method

Participants

The data were obtained from the Baependi Heart Study, a longitudinal cohort study focused on cardiovascular risk factors amongst individuals living in a small town in Brazil [50]. See [53] for details regarding the population and sampling methods. We used cross-sectional data from a single wave of data collection since only this wave contained measures of HbA1c, memory, depression, and neuroimaging. All measures were collected from 2015 to 2017. Owing to limited resources, MRI scans were only obtained for a subset of the cohort; also, depression data was only available for this subset. Therefore, in the present study, analyses were conducted on two samples: the ‘whole sample’ and the ‘MRI sample’. The MRI sample was comprised of a subset of participants from the full sample who were randomly invited for an MRI. Informed consent was obtained, consistent with the Helsinki Declaration. Ethical approval was granted by the ethics committee of the Hospital das Clinicas – Universidade de So Paulo, Brazil. All participants gave informed written consent before participation. Exclusion criteria for the whole sample: we excluded participants with a history of head trauma, stroke, or psychiatric diagnosis other than anxiety/depression. We also excluded 5 participants with unrealistically low values of HbA1c (below 4% [54]). We did not exclude participants with low cognitive performance as it could be confounded with low education [55]. Each analysis was conducted with the maximum N of available data.

Whole Sample

The whole sample consisted of 1331 (60.9% female) individuals. The participants’ ages ranged between 18 and 88 (M = 44.02, SD = 15.17). According to the WHO (2011), an HbA1c level of 6.5% and above is recommended as a reliable cut-off value for the diagnosis of diabetes; 86 (6.5%) participants met this criterion. The whole sample was used to test the associations between HbA1c levels and memory. Mean education years was 8.66 (see Table 1), this aligns closely with the mean for Brazil (~ 8 years according to the World Bank’s Human Capital Index 2020).

Whole Sample

MRI Sample. The MRI sample consisted of 392 individuals (63% female) aged between 18 and 85 (M = 46.51, SD = 15.08). 29 (7.4%) participants met the criterion for diabetes. The MRI sample was used to test the associations between HbA1c, depression, and hippocampal volume. Within the MRI sample, we also looked at effects amongst those participants aged 50 and over (the ‘Age 50+ MRI sample’). The Age 50+ MRI sample comprised 180 participants (57.2% female, Mage = 59.97, SD = 7.69).

Measures

Sociodemographic variables

Participants provided their age, gender, and their education(years).

HbA1c

Participants fasted for 12 h prior to blood draw: time of collection ranged between 6.30am to 8am. Levels of HbA1c in blood were extracted via high-performance liquid chromatography (HPLC), using procedures that followed the National Glycohemoglobin Standardization Program (USA).

Word list recall task

Participants’ immediate memory recall was assessed using the brief neuropsychological battery established by The Consortium to Establish a Registry for Alzheimer’s disease (CERAD). Its applicability for use in Brazilian populations has been established [56]. Due to concerns around inaccurate/missing data, the delayed recall trials were not used. In this task, participants are asked to read aloud 10 unrelated items on printed cards presented one by one to them. After all of these items have been shown, the participants recall as many items as possible. This is repeated three times, and the presentation order of the 10 words is varied each time. Correct recalls are tallied and used to construct three measures of memory performance: “Word List: Total Immediate Recall” (the total number of items recalled (max = 30)), “Word List: Trial 1” (recall on the first attempt only), “Word List: Learning” (memory improvement across trials: recall on trial 3 minus recall on trial 1).

Hospital anxiety and depression scale (HADS)

Depressive symptoms and anxiety were evaluated using a Portuguese translation of the HADS, validated in Brazilian populations [57]. The Portuguese translation has high sensitivity and specificity [58]. We only used summed scores from the depression subscale, which consists of 7 questions using a 4-point Likert scale (range from 0 to 3, with a maximum score of 21).

MRI acquisition

MRI scans were acquired at the Hospital Conego Monte Raso in Baependi on a 1.5 T MAGNETOM (Siemens, Munich, Germany). A high resolution T1-weighted structural images were obtained using a three-dimensional fast spoiled gradient echo T1-weighted sequence with the following parameters: Voxel size 1 mm3, 160 slices, Matrix Size 256 × 256, TR 1700ms, TE 5.1 ms, flip angle 120, inversion time 850 ms, scan time 5m30s. A previous analysis using this dataset to examine associations with metabolic syndrome has been published [59]. The data collection period was from March 2015 to December 2017.

MRI Data Processing

All images were visually inspected, and then cortical reconstruction and volumetric segmentation were carried out using the Freesurfer 6.0 image analysis suite (https://surfer.nmr.mgh.harvard.edu). The reconstruction pipeline employed by Freesurfer includes intensity normalization, motion correction, and the exclusion of non-brain tissue was performed using a hybrid watershed/surface deformation procedure. Images are transformed into Talairach space, and the subcortical white matter and deep grey matter structures are segmented [60, 61]. Hippocampal volumes were calculated using Freesurfer’s automated hippocampal and amygdala segmentation algorithm (included with the development version of Freesurfer 6.0) which uses a probabilistic atlas built with ultra-high resolution MRI data to segment the hippocampal substructures and nuclei of the amygdala [62]. The volumes of the whole hippocampus were extracted for each hemisphere. To account for differences in head size, we used Freesurfer’s estimated total intracranial volume (ICV) as a covariate in our statistical analyses.

Statistical analysis

All statistical analyses were performed using the Statistical Package for the Social Sciences, version 26.0 (SPSS, Chicago, IL). In this study, the whole sample was used to test the association between HbA1c and memory performance. Then, the MRI sample was used to examine the associations between HbA1c levels, depressive symptoms, and hippocampal volumes; the 50 + sample was used to test these associations in participants aged 50 and over. Comparisons between the various samples in terms of participant characteristics were carried out with t-tests or ANOVA (where appropriate for continuous variables) and Pearson`s chi-squared test (for categorical variables). We first checked for extreme outliers in the data indicating unrealistic values, none were found. Normality was then checked by calculating skewness and kurtosis: values above 2 or below − 2 were considered as indicating non-normality [63]. HbA1c levels were non-normally distributed; therefore, HbA1c levels were log10 transformed to ensure more normally distributed data in all analyses, in line with the previous work [64]. The transformation was computed in SPSS.

Pearson’s partial correlations were conducted to investigate the relationship between depressive symptoms, HbA1c, memory performance, and hippocampal volume (total, right, and left hippocampus volume). All partial correlations were adjusted for age, gender, and education. Due to non-normality in the Education (years) variable, we categorised participants into 3 education groups: Primary or less (< 5 years education), Secondary (between 5 and 10 years education), and Higher (> 10 years education), and this categorical variable was used as a covariate in the correlation and mediation analyses. The partial correlation analyses investigating the relationship between HbA1c levels and hippocampus volume also adjusted for depressive symptoms (HADS-D score) and ICV. As we had directional hypotheses, we report one-tailed results for all correlations.

In the MRI sample, we next sought to examine whether the effect of HbA1c on memory performance and depressive symptoms is mediated by total hippocampal volume. Model 1 tested whether the effect of an independent variable (IV: HbA1c) on the dependent variable (DV: Word List: Trial 1) is accounted for by the mediator (M: total hippocampus volume). Model 2 tested whether the effect of IV (HbA1c) on the DV (Word List: Total Immediate Recall) is accounted for by M (total hippocampal volume). Model 3 explored whether the effect of IV (HbA1c levels) on the DV (Depressive symptoms) is mediated by M (total hippocampal volume). We also tested these same three models in the 50 + MRI sample as follow-up analyses. We first report the mediation results adjusted for age only. Then, we present mediation results with additional covariates added to the models (gender and education, and then with ICV also). All models were simple mediation models (Model 4) with 10,000 bootstrap samples using the PROCESS macro in SPSS [65].

Results

Demographic data

Table 1 presents the participant characteristics.

Table 1 Participant characteristics
Full size table

Correlations between HbA1c and memory performance (whole sample)

As we had hypothesised an inverse relationship between HbA1c levels and memory performance, partial correlations were conducted between HbA1c levels and the Word List memory measures, controlling for age, gender, and education. HbA1c levels were related to Word List: Total Immediate Recall (r (1326) = − 0.062, p = .012, 1-tailed) and Word List: Trial 1 (r (1326) = − 0.053, p = .027, 1-tailed), see Fig. 1. There was no significant correlation between HbA1c levels and Word List: Learning (p = .222, 1-tailed). Analyses in the MRI sample (and the age 50 + MRI sub-sample) showed similar associations: significant correlations between HbA1c with Word List: Total Immediate Recall, and Trial 1, were found (see Supplementary Results 1, and Supplementary Figs. 1 and 2).

Fig. 1
figure 1

Scatterplots of the associations between HbA1c levels, (a) Word List: Trial 1, and (b) Word List: Total Immediate Recall performance, in the Whole sample, after adjusting for age, gender, and education

Full size image

Relationship between HbA1c and depressive symptoms (MRI sample)

As we had hypothesised a positive relationship between HbA1c levels and depressive symptomology, a partial correlation (controlling for age, gender, and education) was conducted: the relationship was significant, (r (387) = 0.121, p = .008, 1-tailed, see Fig. 2). In the Age 50+ MRI sample, the association was also significant (r (175) = 0.168, p = .013, 1-tailed), see Fig. 2.

Fig. 2
figure 2

Scatterplots of the positive associations between HbA1c levels and depressive symptoms in (a) the MRI sample and in (b) the Age 50+ MRI sample, after adjusting for age, gender, and education

Full size image

Correlations between HbA1c and hippocampal volumes (MRI sample)

To test whether higher HbA1c levels were associated with lower left, right, and total hippocampus volume, partial correlations were again performed (controlling for age, gender, education, ICV and depressive symptoms). Higher HbA1c levels were significantly associated with smaller left (r (385) = − 0.108, p = .017, 1-tailed), and total hippocampal volume (r (385) = − 0.101, p = 024, 1-tailed), see Fig. 3. There was a trend-level association between HbA1c levels and right hippocampal volume (r (385) = -0.077, p = .065, 1-tailed). In the Age 50+ MRI sample, higher HbA1c levels were associated with smaller left (r (173) = − 0.200, p = .004, 1-tailed), right (r [17] = − 0.143, p = .0295, 1-tailed) and total hippocampus volume (r [17] = − 0.180, p = .0054, 1-tailed), see Supplementary Fig. 3.

Fig. 3
figure 3

Scatterplots showing the negative associations of HbA1c levels with (a) left hippocampus volume and (b) total hippocampus volume (in the MRI sample) after adjusting for age, gender, education, depressive symptoms, and ICV

Full size image

Correlations between hippocampal volumes and memory performance (MRI sample)

To investigate whether Word List: Trial 1, Learning, and Total Immediate Recall were associated with hippocampus volumes, partial correlations were performed (controlling for age, gender, education, and ICV). Results showed that Word List: Trial 1 performance was not associated with left (p = .35), right (p = .31), or total (p = .48) hippocampus volume. Similarly, Word List: Total Immediate Recall performance was not associated with left (p = .35), right (p = .21), or total (p = .42) hippocampus volume. Again, Word List: Learning performance was not associated with left (p = .35), right (p = .49), or total (p = .42) hippocampus volume. In the Age 50+ MRI sample there was likewise no significant associations between any of the memory performance measures and left, right or total hippocampus volume (all p > .05).

Mediation Analyses (MRI sample)

Mediation analyses were carried out to explore whether total hippocampus volume mediates the observed relationships between HbA1c levels and two of the memory performance measures (Model 1 - Word List: Trial 1; Model 2 - Word List: Total Immediate Recall), and the observed relationship between HbA1c levels and depressive symptoms (Model 3). We first conducted partially-adjusted mediation models (adjusted only for age), then again further adjusting for gender, and education, and finally additionally adjusting for total intracranial vol­ume (ICV).

Model 1

(IV: HbA1c levels, DV: Word List: Trial 1, M: total hippocampal volume) showed no significant mediation effect of total hippocampal volume. In the age-adjusted mediation model, results showed that higher HbA1c levels were associated with lower total hippocampal volume (p < .05), however, hippocampal volume did not predict Word List: Trial 1 performance (p > .05); the indirect effect of total hippocampal volume was not significant in either the MRI (est = 0.005, 95% CI [-0.018, 0.029]) nor the Age 50+ MRI sample (est = 0.026, 95% CI [-0.016, 0.079]). Similar results were found with gender and education as additional covariates: the indirect effect was non-significant in both the MRI (est = 0.009, 95% CI [-0.010, 0.037]) and Age 50+ MRI sample (est = 0.030, 95% CI [-0.015, 0.094]) samples. Adding ICV as a further covariate, the mediation analysis showed that the indirect effect of HbA1c levels on Word List: Trial 1 was again non-significant in both the MRI (est = 0.002, 95% CI [-0.019, 0.025]) and the Age 50+ MRI samples (est = 0.017, 95% CI [-0.026, 0.070]).

Model 2

(IV: HbA1c levels, DV: Word List: Total Immediate Recall, M: total hippocampal volume) also revealed no significant mediation effect of total hippocampal volume in the age-adjusted model. The path from HbA1c to hippocampal volume was significant, but the association between hippocampal volume and Word List: Total Immediate Recall was not (p > .05); the indirect effect was also not significant in either the MRI (est = − 0.010, 95% CI [− 0.051, 0.076]) nor the50 + MRI (est = 0.095, 95% CI [− 0.022, 0.246]) samples. Again, adding gender and education yielded similar results, the indirect effect was not significant in either the MRI (est = 0.029, 95% CI [-0.022, 0.098]) nor the 50 + MRI (est = 1.989, 95% CI [-0.149, 5.053]) samples. Adding ICV as a further covariate showed that the indirect effect of HbA1c levels on Word List: Total Immediate Recall remained non-significant in both the MRI (est = 0.001, 95% CI [-0.054, 0.056]) and the Age 50+ MRI sample (est = 0.048, 95% CI [0.071, 0.184]) samples.

Thus, there was no evidence of a mediating effect of hippocampal volume for either of the memory measures (see Supplementary Tables 1, 2 and 3).

Model 3

(IV: HbA1c levels, DV: depressive symptoms, M: total hippocampal volume) showed a significant mediation effect of total hippocampal volume in the age-adjusted model, see Fig. 4. Higher HbA1c levels were associated with lower total hippocampus volume (a = -90.905, p = .011) and lower total hippocampus volume was subsequently associated with higher depressive symptoms (b = − .001, p < .001). A 95% confidence interval based on 10,000 bootstrap samples indicated that the indirect effect (est = .085, 95% CI [0.014, 0.174]) was significant. However, the direct effect was also still significant (c’ = 0.444, p = .025). Thus, total hippocampus volume partially mediated the association between HbA1c levels and depressive symptoms in the MRI sample. However, after adding gender and education as additional covariates, the indirect effect was weakened and the CIs were seen to marginally overlap with zero (est = 0.030, 95% CI [-0.010, 0.089]). After adding ICV as a further covariate, no indirect effect through the mediator was observed (est = − 0.010, 95% CI [-0.060, 0.033]).

In the Age 50+ MRI sample, total hippocampal volume fully mediated the association between HbA1c and depressive symptoms in the age-adjusted model, as the indirect effect was significant (est = .146, 95% CI [0.032, 0.289]), while the direct effect was not (c’ = 0.450, p = .063) (see Supplementary Results 2, and Supplementary Fig. 4). However, when adding gender and education as additional covariates, the indirect effect was again seen to be weakened, with the CIs marginally overlapping with zero (est = 0.069, 95% CI [-0.010, 0.176]) and the direct effect almost reached statistical significance (c’ = 0.069, p = .057) (see supplementary Results 2). After adding ICV as a further covariate, no indirect effect through the mediator was observed (est = − 0.001, 95% CI [-0.086, 0.095]); however, the direct effect of HbA1c was significant (c’ = 0.482, p = .040).

Fig. 4
figure 4

Mediation Model 3 (MRI sample, adjusted for age only). Notes: *p < .05; statistics are unstandardized; a is effect of HbA1c on total hippocampal volume; b is effect of total hippocampal volume on depressive symptoms; c’ is direct effect of HbA1c on depressive symptoms; c is total effect of HbA1c on depressive symptoms. See Supplementary Fig. 4 for the results in the Age 50+ MRI sample

Full size image

Discussion

The current cross-sectional study examined the association between HbA1c levels and memory, as well as associations amongst HbA1c levels, depressive symptoms, and hippocampal volumes. We made use of a large community sample spanning the whole adult age range, of which a subset underwent MRI. In the entire sample, higher HbA1c levels were related to lower trial 1 and total immediate recall performance word list recall; these effects were also present in the MRI subset. Also, higher HbA1c levels were associated with higher depressive symptoms, and smaller left and total hippocampal volume (an association with right hippocampal volume was additionally present in those aged 50+). We then explored whether total hippocampal volume acted as a mediator in these relationships between HbA1c and memory, and HbA1c and depression. No mediation effect was seen for immediate memory recall, but total hippocampus volume partially mediated the association between HbA1c levels and depressive symptoms, although it should be noted that this was only evident when models were adjusted for age only; adding further covariates (gender, education, ICV) undermined the mediation effect.

There is little previous work that directly explored these questions. In one longitudinal study, higher HbA1c values were associated with lower episodic memory (composite scores of immediate and delayed word recall) at baseline, and a steeper decline over six years, in a community sample aged 60 and over [66]. However, in another cross-sectional study in healthy participants aged between 50 and 75, HbA1c was not associated with performance on the California Verbal Learning Test (CVLT) or story recall (Logical Memory of the Weschler Memory Scale- III) [67], although this study excluded those whose HbA1c levels met the diagnostic criterion for diabetes, and it was severely underpowered (N = 34). Thus, the current study is novel in that we used a large sample that spanned most of the adult age range (18 to 85); in which only a small minority (~ 7%) met diabetes criteria: findings indicate that even within the normal range, poorer glycaemic control is related to reduced immediate memory recall performance, throughout adulthood, while controlling for age, gender and education.

While we cannot infer the direction of causality from this cross-sectional study, this finding suggests neural systems subserving memory are impacted by poorer glycaemic control, and this effect is not just limited to older adulthood. We also found a significant positive relationship between HbA1c levels and depressive symptoms. Again, there is a paucity of research examining the association between HbA1c levels and depressive symptoms in community samples, as most studies have investigated this in patients with diabetes: the association between diabetes and depression is well established [33, 68]. One previous longitudinal study linked higher HbA1c levels with an increase in depressive symptoms, in a mixed community sample of healthy adults and diabetes patients aged 50+ [33]. Our results provide evidence that even in healthy adults, a higher HbA1c level is associated with elevated depressive symptoms. Further, this seems to be the case throughout adulthood, and not just in later life.

The MRI data showed that higher HbA1c levels were associated with smaller left and total hippocampal volumes; an association with right hippocampal volume was additionally present in those aged 50+. Prior studies have shown similar results, but mostly by comparing diabetes patients and healthy participants [69]. One previous study used UK Biobank to test for an effect of low-to-normal HbA1c values, as compared to individuals with normal glycaemic control, prediabetes, undiagnosed and known diabetes (age range 40–69) [16]. A stepwise reduction in hippocampal volume was observed according to the diagnostic group. The current study builds on that finding by showing that this relationship holds, in a sample of a wider age range, and when HbA1c was considered as a continuous variable, rather than grouping participants by category. There are various mechanisms that could link HbA1c levels to hippocampal volume reductions. Elevated cortisol has been associated with poor glycaemic control, and the hippocampus is highly sensitive to the neurotoxic effects of cortisol [23]; glucotoxicity due to hyperglycaemia may also contribute, through the build-up of AGEs [19, 20]. These factors can induce oxidative stress, triggering neuronal apoptosis in hippocampus [70]. In previous work, hippocampal volume has been found to partially mediate the effect of HbA1c on delayed recall and learning ability (measured by the REY Auditory Verbal Learning Test) in healthy participants aged 50–80 [64]; hippocampal microstructural metrics were also found to partially mediate the relationship, thus representing an additional possible mechanism. Here, we did not find any significant mediation effect of hippocampal volume in the association between HbA1c and immediate memory performance. We included age and other covariates in our mediation models but Kerti et al. [64] did not, which could explain the differences in findings. Alternatively, it might have been due to the specific memory measure used here: the current work is limited by the fact that only immediate but not delayed recall was tested, and used word list learning only: future work should utilise a broader range of memory tasks, to clarify the role of hippocampal volume as a possible mediating factor. Indeed, we did not identify a link between hippocampal volume and memory performance in our sample: previous work on this point is conflicting, with mixed reports in both developmental and adult samples [25]; some studies have reported a significant association between total hippocampal volume and memory (e.g., 23) while others have reported no association (e.g., [71]).

We also explored whether impaired glycaemic control may impact depressive symptomology via hippocampal volume reductions. Research has indicated the detrimental effect of depression on hippocampal volume due to cortisol-related neurotoxicity [23]; studies also suggest that smaller hippocampi might also be a risk factor for depression [48, 49]. For example, in middle-aged and older people, smaller hippocampal volume at baseline was associated with higher depressive symptoms over an eight-year follow-up [72].

We found evidence that the association between HbA1c and depressive symptom severity is partly mediated by total hippocampal volume in the mediation models that adjusted for age, although this did not survive in the fully-adjusted models. To our knowledge, the present study is the first to investigate the mediating role of hippocampal volume in the association between HbA1c levels and depressive symptoms in a community sample, and this finding is important in that it points to a potentially important role for hippocampal volume loss as an explanatory variable in this association. The mediation effect was stronger in the 50 + subsample in that a full mediation occurred, but again only in the age-adjusted model. Adding gender and education weakened the mediation effects to borderline significance, while adding ICV abolished it. This could be due to the complex interactions between HbA1c, gender and global brain volumes indicated by recent work: a large study in UK Biobank participants found an association between higher HbA1c levels and lower whole brain volume [52], while a study in older adults linked HbA1c to lower whole brain volumes exclusively in females [73]. Given this preliminary evidence, it is possible that impaired glycaemic control has widespread impact on brain volumes, with females at higher risk – thus confounding and undermining the mediation effect when gender and ICV were added to the models. Further work is needed to more fully characterise the complex effects of HbA1c on hippocampal as well as whole brain volumes.

The impact of impaired glycaemic control on the hippocampus has been clearly shown in diabetes, which has been linked to neurodegeneration and abnormal signalling in human hippocampi [74]. In animals, improving glycaemic control by insulin treatment reverses the effect of induced diabetes on hippocampal volume and affective behaviour [75]. The current findings, indicating hippocampus volume as a potential mediator in the glycaemic control - depression relationship, is noteworthy as the sample largely had HbA1c levels in the normal range, suggesting that this might be a generalisable mechanism and not restricted to diabetes. The damaging effects of cortisol on the hippocampus could contribute, and should be investigated by future work since (as discussed above), increased circulating cortisol has been linked to higher HbA1c levels, as well as the risk of depressive symptoms [23, 76,77,78]. The neurotoxic effects of cortisol on the hippocampus are well established [23]; elevated cortisol levels cause both hippocampal volume loss and the inhibition of neurogenesis [79]. In the 50 + age group, the mediation effect appeared to be stronger, in line with evidence showing that the hippocampus is particularly vulnerable to the effects of ageing. However, it should be noted that our 50 + group had lower levels of education compared to the MRI sample as a whole, which could have influenced findings, for example due to lower cognitive reserve capacity.

The robust associations between HbA1c and hippocampal volumes found here are also of importance; again, these were most apparent in the 50 + age group, with associations found with both left and right hippocampal volumes in this age range. The hippocampus is one of the first brain regions to be affected by Alzheimer’s disease (AD) pathology [80]. Moreover, higher HbA1c levels and depressive symptoms have been shown to increase the risk of AD [81, 82] and memory decline [12, 83]. Therefore, our findings point to HbA1c as a modifiable risk factor that should be targeted, to minimise hippocampal atrophy and the risk of cognitive decline in later life. This is further supported by the full mediation effect identified in those aged 50+: since HbA1c appears to link to depression symptoms via hippocampal volume, and depression itself raises the risk of AD, cognitive decline, and hippocampal atrophy, maintaining healthy HbA1c levels might be of prime importance to prevent a negative feedback loop from occurring, particularly in mid-age and older individuals. However, future studies should investigate the associations longitudinally, and also characterise and account for effects on whole brain volume and potential gender interactions, since the mediation effect observed here was only present in the partially-adjusted models, as discussed above. The cross-sectional research design is a limitation of the current study and prevents us making any casual inferences: it should be emphasised that we cannot discount the possibility of reverse causation i.e., depression levels could exert influences on hippocampal volume and HbA1c levels, perhaps due to dysregulated sleep, cortisol, or poorer dietary choices. Longitudinal work should seek to clarify causal relationships and mechanisms linking impaired glycaemic control to reduced hippocampal volume, increased risk of depression, and cognitive decline. Also, it is important to note that there are various potential confounding variables (such as body mass index) which future studies should attempt to control for: missing information in the present dataset meant we could not control for these here, and this might have impacted the findings. Body mass index and blood markers associated with obesity have been shown to impact on cognitive performance [84] and cortical volumes [85], and although findings are not consistent, it would be valuable to include such measures in future work, to rule out possible confounding effects and confirm the relationships reported here. Future work should also aim to replicate the current findings in other populations, the current findings are valuable in that they were derived from an understudied, non-Western population, but need to be confirmed in other samples.

Conclusions

The current study addresses knowledge gaps around the associations between HbA1c levels, cognition, depressive symptomology, and brain volumes, and revealed the impact of poor glycaemic control on immediate memory recall, depression levels and hippocampal volumes, in a large non-clinical sample covering the whole adult age range. Also, this is the first study to investigate hippocampal volumes as a possible mediator in the association between HbA1c levels and depressive symptoms, and shows that hippocampal structural integrity might be an important mechanistic pathway in the relationship between HbA1c and depressive symptoms, particularly in older adults. Given the powerful negative consequences of poor glycaemic control for mental health, cognitive function and brain structure demonstrated by the current findings, results underline the importance of implementing policies and interventions focused on optimising HbA1c levels, to avoid the negative outcomes identified here.

Data availability

Data will be made available on reasonable requests.

References

  1. Nathan DM, Turgeon H, Regan S. Relationship between glycated haemoglobin levels and mean glucose levels over time. Diabetologia. 2007;50(11):2239–44.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. WHO. Use of glycated haemoglobin (HbA1c) in diagnosis of diabetes mellitus. World Health Organisation. 2011. Contract No.: WHO/NMH///CHP/CPM/11.1

  3. World Health Organisation; [updated 5 April 2023. https://www.who.int/news-room/fact-sheets/detail/diabetes

  4. Chen P-C, Chan Y-T, Chen H-F, Ko M-C, Li C-Y. Population-based cohort analyses of the bidirectional relationship between type 2 diabetes and depression. Diabetes Care. 2013;36(2):376–82.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Xue M, Xu W, Ou YN, Cao XP, Tan MS, Tan L, et al. Diabetes mellitus and risks of cognitive impairment and dementia: a systematic review and meta-analysis of 144 prospective studies. Ageing Res Rev. 2019;55:100944.

    Article  CAS  PubMed  Google Scholar 

  6. Ott A, Stolk RP, van Harskamp F, Pols HA, Hofman A, Breteler MM. Diabetes mellitus and the risk of dementia: the Rotterdam Study. Neurology. 1999;53(9):1937–42.

    Article  CAS  PubMed  Google Scholar 

  7. Hishikawa N, Yamashita T, Deguchi K, Wada J, Shikata K, Makino H, et al. Cognitive and affective functions in diabetic patients associated with diabetes-related factors, white matter abnormality and aging. Eur J Neurol. 2015;22(2):313–21.

    Article  CAS  PubMed  Google Scholar 

  8. Okura T, Heisler M, Langa KM. Association between cognitive function and Social Support with Glycemic Control in adults with diabetes Mellitus. J Am Geriatr Soc. 2009;57(10):1816–24.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Zhang Y-W, Zhang J-Q, Liu C, Wei P, Zhang X, Yuan Q-Y, et al. Memory dysfunction in type 2 diabetes mellitus correlates with reduced hippocampal CA1 and subiculum volumes. Chin Med J (Engl). 2015;128(4):465–71.

    Article  CAS  PubMed  Google Scholar 

  10. Repple J, Karliczek G, Meinert S, Förster K, Grotegerd D, Goltermann J, et al. Variation of HbA1c affects cognition and white matter microstructure in healthy, young adults. Mol Psychiatry. 2021;26(4):1399–408.

    Article  CAS  PubMed  Google Scholar 

  11. Luchsinger JA, Cabral R, Eimicke JP, Manly JJ, Teresi J, Glycemia. Diabetes status, and cognition in hispanic adults aged 55–64 years. Psychosom Med. 2015;77(6):653–63.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Marden JR, Mayeda ER, Tchetgen EJT, Kawachi I, Glymour MM. High hemoglobin A1c and diabetes predict memory decline in the health and retirement study. Alzheimer Dis Assoc Disord. 2017;31(1):48.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Ravona-Springer R, Moshier E, Schmeidler J, Godbold J, Akrivos J, Rapp M, et al. Changes in glycemic control are associated with changes in cognition in non-diabetic elderly. J Alzheimer’s Disease: JAD. 2012;30(2):299–309.

    Article  CAS  Google Scholar 

  14. Crane PK, Walker R, Hubbard RA, Li G, Nathan DM, Zheng H, et al. Glucose levels and risk of dementia. N Engl J Med. 2013;369(6):540–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Eichenbaum H, Hippocampus. Cognitive processes and neural representations that underlie declarative memory. Neuron. 2004;44(1):109–20.

    Article  CAS  PubMed  Google Scholar 

  16. Squire LR. Memory and the hippocampus: a synthesis from findings with rats, monkeys, and humans. Psychol Rev. 1992;99(2):195.

    Article  CAS  PubMed  Google Scholar 

  17. Gold SM, Dziobek I, Sweat V, Tirsi A, Rogers K, Bruehl H, et al. Hippocampal damage and memory impairments as possible early brain complications of type 2 diabetes. Diabetologia. 2007;50(4):711–9.

    Article  CAS  PubMed  Google Scholar 

  18. Garfield V, Farmaki A-E, Eastwood SV, Mathur R, Rentsch CT, Bhaskaran K, et al. HbA1c and brain health across the entire glycaemic spectrum. Diabetes Obes Metabolism. 2021;23(5):1140–9.

    Article  CAS  Google Scholar 

  19. Brownlee M. Biochemistry and molecular cell biology of diabetic complications. Nature. 2001;414(6865):813–20.

    Article  CAS  PubMed  Google Scholar 

  20. Srikanth V, Maczurek A, Phan T, Steele M, Westcott B, Juskiw D, et al. Advanced glycation endproducts and their receptor RAGE in Alzheimer’s disease. Neurobiol Aging. 2011;32(5):763–77.

    Article  CAS  PubMed  Google Scholar 

  21. Wang S-h, Sun Z-l, Guo Y-j, Yuan Y, Yang B-q. Diabetes impairs hippocampal function via Advanced Glycation End product mediated New Neuron Generation in animals with diabetes-related Depression. Toxicol Sci. 2009;111(1):72–9.

    Article  CAS  PubMed  Google Scholar 

  22. Ahmad S, Moinuddin, Shahab U, Habib S, Salman Khan M, Alam K, et al. Glycoxidative damage to human DNA: neo-antigenic epitopes on DNA molecule could be a possible reason for autoimmune response in type 1 diabetes. Glycobiology. 2013;24(3):281–91.

    Article  PubMed  Google Scholar 

  23. Lehrer HM, Dubois SK, Maslowsky J, Laudenslager ML, Steinhardt MA. Hair cortisol concentration and glycated hemoglobin in African American adults. Psychoneuroendocrinology. 2016;72:212–8.

    Article  CAS  PubMed  Google Scholar 

  24. Pohlack ST, Meyer P, Cacciaglia R, Liebscher C, Ridder S, Flor H. Bigger is better! Hippocampal volume and declarative memory performance in healthy young men. Brain Struct Function. 2014;219(1):255–67.

    Article  Google Scholar 

  25. Botdorf M, Canada KL, Riggins T. A meta-analysis of the relation between hippocampal volume and memory ability in typically developing children and adolescents. Hippocampus. 2022;32(5):386–400.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Zheng F, Cui D, Zhang L, Zhang S, Zhao Y, Liu X et al. The volume of hippocampal subfields in Relation to decline of memory recall across the adult lifespan. Front Aging Neurosci. 2018;10.

  27. Hardcastle C, O’Shea A, Kraft JN, Albizu A, Evangelista ND, Hausman HK et al. Contributions of hippocampal volume to Cognition in healthy older adults. Front Aging Neurosci. 2020;12.

  28. Kramer JH, Mungas D, Reed BR, Wetzel ME, Burnett MM, Miller BL, et al. Longitudinal MRI and cognitive change in healthy elderly. Neuropsychology. 2007;21(4):412–8.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Nouwen A, Winkley K, Twisk J, Lloyd CE, Peyrot M, Ismail K, et al. Type 2 diabetes mellitus as a risk factor for the onset of depression: a systematic review and meta-analysis. Diabetologia. 2010;53(12):2480–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Liu X, Li Y, Guan L, He X, Zhang H, Zhang J et al. A systematic review and Meta-analysis of the prevalence and risk factors of Depression in type 2 diabetes patients in China. Front Med. 2022;9.

  31. Kraal AZ, Ellingrod VL, Zahodne LB. Depressive symptoms longitudinally mediate the effect of hyperglycemia on memory decline in type 2 diabetes. Diabetes Care. 2023;46(9):1673–80.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Rotella F, Mannucci E. Depression as a risk factor for diabetes: a meta-analysis of longitudinal studies. J Clin Psychiatry. 2013;74(1):31–7.

    Article  PubMed  Google Scholar 

  33. Schmitz N, Deschênes S, Burns R, Smith KJ. Depressive symptoms and glycated hemoglobin A1c: a reciprocal relationship in a prospective cohort study. Psychol Med. 2016;46(5):945–55.

    Article  CAS  PubMed  Google Scholar 

  34. Wium-Andersen IK, Hengeveld EM, Rungby J, Jorgensen MB, Osler M, Wium-Andersen MK. Hemoglobin A1c-levels and subsequent risk of depression in individuals with and without diabetes. J Diabetes Complications. 2021;35(8):107946.

    Article  CAS  PubMed  Google Scholar 

  35. Sheline YI. Neuroimaging studies of mood disorder effects on the brain. Biol Psychiatry. 2003;54(3):338–52.

    Article  PubMed  Google Scholar 

  36. Arnone D, McIntosh A, Ebmeier K, Munafò M, Anderson I. Magnetic resonance imaging studies in unipolar depression: systematic review and meta-regression analyses. Eur Neuropsychopharmacol. 2012;22(1):1–16.

    Article  CAS  PubMed  Google Scholar 

  37. Cole J, Costafreda SG, McGuffin P, Fu CH. Hippocampal atrophy in first episode depression: a meta-analysis of magnetic resonance imaging studies. J Affect Disord. 2011;134(1–3):483–7.

    Article  PubMed  Google Scholar 

  38. Colle R, Dupong I, Colliot O, Deflesselle E, Hardy P, Falissard B, et al. Smaller hippocampal volumes predict lower antidepressant response/remission rates in depressed patients: a meta-analysis. World J Biol Psychiatry. 2018;19(5):360–7.

    Article  PubMed  Google Scholar 

  39. Roddy DW, Farrell C, Doolin K, Roman E, Tozzi L, Frodl T, et al. The hippocampus in depression: more than the sum of its parts? Advanced hippocampal substructure segmentation in depression. Biol Psychiatry. 2019;85(6):487–97.

    Article  PubMed  Google Scholar 

  40. Ezzati A, Zimmerman ME, Katz MJ, Lipton RB. Hippocampal correlates of depression in healthy elderly adults. Hippocampus. 2013;23(12):1137–42.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Spalletta G, Piras F, Caltagirone C, Fagioli S. Hippocampal multimodal structural changes and subclinical depression in healthy individuals. J Affect Disord. 2014;152–154:105–12.

    Article  PubMed  Google Scholar 

  42. Brown ES, Hughes CW, McColl R, Peshock R, King KS, Rush AJ. Association of depressive symptoms with hippocampal volume in 1936 adults. Neuropsychopharmacology. 2014;39(3):770–9.

    Article  PubMed  Google Scholar 

  43. Seckl J, Olsson T. Glucocorticoid hypersecretion and the age-impaired hippocampus: cause or effect? J Endocrinol. 1995;145(2):201–11.

    Article  CAS  PubMed  Google Scholar 

  44. Jacobson L, Sapolsky R. The role of the hippocampus in feedback regulation of the hypothalamic-pituitary-adrenocortical axis. Endocr Rev. 1991;12(2):118–34.

    Article  CAS  PubMed  Google Scholar 

  45. Tessner KD, Walker EF, Dhruv SH, Hochman K, Hamann S. The relation of cortisol levels with hippocampus volumes under baseline and challenge conditions. Brain Res. 2007;1179:70–8.

    Article  CAS  PubMed  Google Scholar 

  46. Lyons DM, Parker KJ, Zeitzer JM, Buckmaster CL, Schatzberg AF. Preliminary evidence that hippocampal volumes in monkeys predict stress levels of adrenocorticotropic hormone. Biol Psychiatry. 2007;62(10):1171–4.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. McEwen BS. Stress-induced remodeling of hippocampal CA3 pyramidal neurons. Brain Res. 2016;1645:50–4.

    Article  CAS  PubMed  Google Scholar 

  48. Frodl T, Jäger M, Smajstrlova I, Born C, Bottlender R, Palladino T, et al. Effect of hippocampal and amygdala volumes on clinical outcomes in major depression: a 3-year prospective magnetic resonance imaging study. J Psychiatry Neurosci. 2008;33(5):423.

    PubMed  PubMed Central  Google Scholar 

  49. Kronmüller KT, Pantel J, Köhler S, Victor D, Giesel F, Magnotta VA, et al. Hippocampal volume and 2-year outcome in depression. Br J Psychiatry. 2008;192(6):472–3.

    Article  PubMed  Google Scholar 

  50. Egan KJ, von Schantz M, Negrao AB, Santos HC, Horimoto AR, Duarte NE, et al. Cohort profile: the Baependi Heart Study-a family-based, highly admixed cohort study in a rural Brazilian town. BMJ Open. 2016;6(10):e011598.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Foran E, Hannigan A, Glynn L. Prevalence of depression in patients with type 2 diabetes mellitus in Irish primary care and the impact of depression on the control of diabetes. Ir J Med Sci. 2015;184(2):319–22.

    Article  CAS  PubMed  Google Scholar 

  52. Ranglani S, Ward J, Sattar N, Strawbridge RJ, Lyall DM. Testing for associations between HbA1c levels, polygenic risk and brain health in UK Biobank (N = 39 283). Diabetes Obes Metabolism. 2023;25(11):3136–43.

    Article  CAS  Google Scholar 

  53. de Oliveira CM, Pereira AC, de Andrade M, Soler JM, Krieger JE. Heritability of cardiovascular risk factors in a Brazilian population: Baependi Heart Study. BMC Med Genet. 2008;9:32.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Wang P. What clinical laboratorians should do in response to extremely low hemoglobin A1c results. Lab Med. 2017;48(1):89–92.

    Article  PubMed  Google Scholar 

  55. Fillenbaum GG, Burchett BM, Unverzagt FW, Rexroth DF, Welsh-Bohmer K. Norms for CERAD Constructional Praxis Recall. Clin Neuropsychol. 2011;25(8):1345–58.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Bertolucci PHF, Okamoto IH, Brucki SMD, Siviero MO, Toniolo Neto J, Ramos LR. Applicability of the CERAD neuropsychological battery to Brazilian elderly. Arq Neuropsiquiatr. 2001;59:532–6.

    Article  CAS  PubMed  Google Scholar 

  57. Botega NJ, Bio MR, Zomignani MA, Garcia C Jr., Pereira WA. [Mood disorders among inpatients in ambulatory and validation of the anxiety and depression scale HAD]. Rev Saude Publica. 1995;29(5):355–63.

    CAS  PubMed  Google Scholar 

  58. Soares-Filho GLF, Freire RC, Biancha K, Pacheco T, Volschan A, Valença AM, et al. Use of the hospital anxiety and depression scale (HADS) in a cardiac emergency room: chest pain unit. Clinics. 2009;64:209–14.

    Article  PubMed  PubMed Central  Google Scholar 

  59. Alkan E, Taporoski TP, Sterr A, von Schantz M, Vallada H, Krieger JE, et al. Metabolic syndrome alters relationships between cardiometabolic variables, cognition and white matter hyperintensity load. Sci Rep. 2019;9(1):4356.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. 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(3):341–55.

    Article  CAS  PubMed  Google Scholar 

  61. Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis: I. Segmentation and Surface Reconstruction. NeuroImage. 1999;9(2):179–94.

    Article  CAS  PubMed  Google Scholar 

  62. Iglesias JE, Augustinack JC, Nguyen K, Player CM, Player A, Wright M, et al. A computational atlas of the hippocampal formation using ex vivo, ultra-high resolution MRI: application to adaptive segmentation of in vivo MRI. NeuroImage. 2015;115:117–37.

    Article  PubMed  Google Scholar 

  63. Field A. Discovering statistics using IBM SPSS statistics: sage; 2013.

  64. Kerti L, Witte AV, Winkler A, Grittner U, Rujescu D, Flöel A. Higher glucose levels associated with lower memory and reduced hippocampal microstructure. Neurology. 2013;81(20):1746–52.

    Article  CAS  PubMed  Google Scholar 

  65. Hayes AF. Introduction to Mediation, Moderation, and conditional process analysis: a regression-based Approach. Guilford; 2021.

  66. Pappas C, Andel R, Infurna FJ, Seetharaman S. Glycated haemoglobin (HbA1c), diabetes and trajectories of change in episodic memory performance. J Epidemiol Commun Health. 2017;71(2):115–20.

    Article  Google Scholar 

  67. Skinner JS, Morgan A, Hernandez-Saucedo H, Hansen A, Corbett S, Arbuckle M et al. Associations between markers of glucose and insulin function and cognitive function in healthy African American elders. J Gerontol Geriatr Res. 2015;4(4).

  68. Beran M, Muzambi R, Geraets A, Albertorio-Diaz JR, Adriaanse MC, Iversen MM, et al. The bidirectional longitudinal association between depressive symptoms and HbA1c: a systematic review and meta-analysis. Diabet Med. 2022;39(2):e14671.

    Article  CAS  PubMed  Google Scholar 

  69. Zhang T, Shaw M, Cherbuin N. Association between type 2 diabetes Mellitus and Brain Atrophy: a Meta-analysis. Korean Diabetes J. 2022.

  70. Mattson MP, Magnus T. Ageing and neuronal vulnerability. Nat Rev Neurosci. 2006;7(4):278–94.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  71. Clark IA, Monk AM, Hotchin V, Pizzamiglio G, Liefgreen A, Callaghan MF, et al. Does hippocampal volume explain performance differences on hippocampal-dependant tasks? NeuroImage. 2020;221:117211.

    Article  PubMed  Google Scholar 

  72. Buddeke J, Kooistra M, Zuithoff N, Gerritsen L, Biessels G, Van der Graaf Y, et al. Hippocampal volume and the course of depressive symptoms over eight years of follow-up. Acta Psychiatrica Scandinavica. 2017;135(1):78–86.

    Article  CAS  PubMed  Google Scholar 

  73. Fatih N, Chaturvedi N, Lane CA, Parker TD, Lu K, Cash DM, et al. Sex-related differences in whole brain volumes at age 70 in association with hyperglycemia during adult life. Neurobiol Aging. 2022;112:161–9.

    Article  PubMed  Google Scholar 

  74. Ho N, Sommers MS, Lucki I. Effects of diabetes on hippocampal neurogenesis: links to cognition and depression. Neurosci Biobehav Rev. 2013;37(8):1346–62.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  75. Ho N, Balu DT, Hilario MR, Blendy JA, Lucki I. Depressive phenotypes evoked by experimental diabetes are reversed by insulin. Physiol Behav. 2012;105(3):702–8.

    Article  CAS  PubMed  Google Scholar 

  76. Joseph JJ, Wang X, Spanakis E, Seeman T, Wand G, Needham B, et al. Diurnal salivary cortisol, glycemia and insulin resistance: the multi-ethnic study of atherosclerosis. Psychoneuroendocrinology. 2015;62:327–35.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  77. Iob E, Kirschbaum C, Steptoe A. Persistent depressive symptoms, HPA-axis hyperactivity, and inflammation: the role of cognitive-affective and somatic symptoms. Mol Psychiatry. 2020;25(5):1130–40.

    Article  PubMed  Google Scholar 

  78. Geerlings MI, Gerritsen L, Late-Life, Depression. Hippocampal volumes, and hypothalamic-pituitary-adrenal Axis Regulation: a systematic review and Meta-analysis. Biol Psychiatry. 2017;82(5):339–50.

    Article  CAS  PubMed  Google Scholar 

  79. Herbert J, Goodyer IM, Grossman AB, Hastings MH, de Kloet ER, Lightman SL, et al. Do corticosteroids damage the brain? J Neuroendocrinol. 2006;18(6):393–411.

    Article  CAS  PubMed  Google Scholar 

  80. Vijayakumar A, Vijayakumar A. Comparison of hippocampal volume in dementia subtypes. ISRN Radiol. 2012;2013:174524.

    PubMed  PubMed Central  Google Scholar 

  81. Saczynski JS, Beiser A, Seshadri S, Auerbach S, Wolf PA, Au R. Depressive symptoms and risk of dementia. Framingham Heart Study. 2010;75(1):35–41.

    CAS  Google Scholar 

  82. Ramirez A, Wolfsgruber S, Lange C, Kaduszkiewicz H, Weyerer S, Werle J, et al. Elevated HbA 1c is Associated with increased risk of Incident Dementia in Primary Care patients. J Alzheimers Dis. 2015;44:1203–12.

    Article  CAS  PubMed  Google Scholar 

  83. Zahodne LB, Stern Y, Manly JJ. Depressive symptoms precede memory decline, but not Vice Versa, in non-demented older adults. J Am Geriatr Soc. 2014;62(1):130–4.

    Article  PubMed  PubMed Central  Google Scholar 

  84. Cohen R, Diabetes. Obesity-Associated comorbidities and NIH-Toolbox Neurocognitive performance. Diabetes Obes Int J. 2021;6.

  85. Fernandez-Andujar M, Morales-Garcia E, Garcia-Casares N. Obesity and Gray Matter volume assessed by neuroimaging: a systematic review. Brain Sci. 2021;11:8.

    Article  Google Scholar 

Download references

Acknowledgements

We are grateful to the population of Baependi for their participation in the Baependi Heart Study.

Funding

This study was supported by grants from Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP, 2013/17368-0, to A.C.P.). G.Y.S. is supported by a studentship from the Government of Turkey.

Author information

Authors and Affiliations

  1. Faculty of Health and Medical Sciences, School of Psychology, University of Surrey, Guildford, Surrey, GU2 7XH, UK

    Gulin Yatagan Sevim & Simon L. Evans

  2. Faculty of Health, Science, Social Care and Education, Kingston University, London, UK

    Erkan Alkan

  3. Harvard Center for Population and Development Studies, Harvard University, Cambridge, Massachusetts, USA

    Tamara P. Taporoski

  4. University of São Paulo School of Medicine, São Paulo, Brazil

    Jose E Krieger & Alex C Pereira

Authors
  1. Gulin Yatagan Sevim
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Erkan Alkan
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Tamara P. Taporoski
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Jose E Krieger
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Alex C Pereira
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Simon L. Evans
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

G.Y.S. Formal analysis, Writing – original draft; S.E. Formal analysis, Writing – original draft, E.A. Data preparation, Writing - review & editing; T.P.T. Data collection, Data preparation, Writing - review & editing; J.E.K Study Design, Data collection; A.C.P Study Design, Writing - review & editing.

Corresponding author

Correspondence to Simon L. Evans.

Ethics declarations

Ethical approval

Informed consent was obtained, consistent with the Helsinki Declaration. Ethical approval was granted by the ethics committee of the Hospital das Clinicas – Universidade de So Paulo, Brazil. Participation in this study was voluntary and all participants gave informed written consent before participation.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

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

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yatagan Sevim, G., Alkan, E., Taporoski, T.P. et al. Effects of glycaemic control on memory performance, hippocampal volumes and depressive symptomology. Diabetol Metab Syndr 16, 219 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13098-024-01429-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13098-024-01429-2

Keywords

Diabetology & Metabolic Syndrome

ISSN: 1758-5996

Contact us