Depression, stress, and risk for Alzheimer's disease

Abstract

Listen

Translate article

It has frequently been hypothesized that depression can cause Alzheimer's disease through hypersecretion of glucocorticoids, which would have deleterious effects on hippocampal structures. However, no large-scale studies in humans examined the direct relation between depression, hypothalamic-pituitary-adrenal (HPA) axis activity and hippocampal structures. The aim of this study was to examine the associations between depression, HPA-axis activity, and hippocampal and entorhinal cortex volumes. Cross-sectional analyses within 636 participants (mean age 62±9 years, 81% male) from the Second Manifestation of ARTerial disease-Memory depression and aging (SMART-Medea) study, an ancillary study to the SMART-MR study, aimed to investigate brain changes associated with psychosocial stressors in patients with a history of atherosclerotic disease. Twelve-month diagnosis of major depressive disorder (MDD), early-onset depression (EOD) (<50 years) and late-onset depression (LOD) (≥50 years) were assessed. HPA-axis activity was measured with 6 saliva samples, collected at awakening and 30, 45 and 60 minutes thereafter, and at 10PM and 11PM. Suppression of the HPA-axis was defined as the cortisol value at awakening after ingestion of 0.5 mg dexamethasone at 11PM. Volumetric measurements of the hippocampus and entorhinal cortex were performed on a 3-dimensional FFE T1-weighted 1.5 Tesla MRI scan with isotropic voxels. Regression models adjusted for demographics, vascular risk, antidepressant use and white matter lesions showed that MDD was not significantly associated with hippocampal or entorhinal cortex volumes. However, participants with EOD had significantly smaller hippocampal volumes than those never depressed, whereas participants with LOD had smaller entorhinal cortex volumes. Participants with higher evening cortisol levels and reduced suppression after dexamethasone had smaller hippocampal volumes. Higher evening cortisol levels were also associated with smaller right, but not left, entorhinal cortex volume. The cortisol awakening response was not significantly associated with hippocampal or entorhinal cortex volumes. Cortisol levels did not, however, explain the associations found between age of onset of depression with hippocampal or entorhinal cortex volumes. Although we found several associations between depression, HPA-axis activity, and hippocampal or entorhinal cortex volumes, from our studies it seems unlikely that hypersecretion of glucocorticoids is an explanation for the frequently observed association between depression, hippocampal atrophy, and Alzheimer's disease.