Cortisol Reactivity to a physical stressor in Patients with Depression and Alzheimer's disease

ABSTRACT. Some prevalent mental disorders in the elderly, such as Alzheimer's disease (AD) and major depression disorder (MDD), are associated with chronic stress and consequently with possible dysregulation of hypothalamic–pituitary–adrenal (HPA) axis and cortisol levels in basal conditions or in the reactivity of an acute stressor. However, evidence of cortisol behavior after a physical stressor in patients with AD and MDD is scarce. Objective: This study aimed to investigate the cortisol reactivity to a single session of physical exercise in patients with MDD and AD and compare it to healthy control (HC) older individuals. Methods: HC individuals (n=10) and elderly with clinical diagnostic of MDD (n=08) and AD (n=13) were submitted to a single bout of aerobic exercise in a treadmill during 30 minutes of moderate intensity. Salivary cortisol was collected before and after acute stressor. A repeated-measure analysis of variance (ANOVA), spearman correlation, and linear regression were performed. Results: The repeated-measure ANOVA revealed no interaction for cortisol on the moment×group [F(2.000, 28.000)=1.285; p=0.293] and no effect for group (F=0.323; p=0.727). However, a significant effect for moment [F(1.000, 28.000)=4.930; p=0.035] was found, with a decreased cortisol levels in postexercise for all groups. The effect size (ES) of cortisol reduction was small for patients with MDD (d=0.402) and trivial for patients with AD (d=0.166) and HC group (d=0.090). Conclusions: All participants show a decreased cortisol reactivity to a physical stressor, which can be associated with an impairment in coping with an acute stressor.


INTRODUCTION
L ife expectancy has increased in several countries in the past decades, impacting on the prevalence of some chronic diseases and even neuropsychiatric disorders related to aging such as major depressive disorder (MDD) and Alzheimer's disease (AD) 1,2 . Although there is a genetic component for their development in late life, some modifiable risks are known 3,4 , such as chronic stress 5 and aggravation of the disease courses.
Stress system is regulated mainly by hypothalamic-pituitary-adrenal (HPA) axis resulting in cortisol release in the body, which involves the activation of autonomic nervous system and consequently an increase in blood pressure and heart rate (HR), preparing to a life-threatening situation. Moreover, cortisol crosses the blood-brain barrier, achieving some brain regions to promote negative feedback mechanism. High cortisol levels may cause consequent neuron apoptosis impacting in hippocampus atrophy, also observed in patients with MDD 6 and AD 7 . This mechanism can be aggravated by some other physiological changes related to aging, such as inflammaging, which is a phenomenon with chronic and low-grade inflammation observed in elderly individuals 8 , and wear and tear of chronic exposure to cortisol in the body (allostatic load) 9 .
In this sense, some studies have observed HPA changes in psychiatric disorders, which may result in hypercortisolism or hypocortisolism at rest in older patients with MDD 10,11 and hypercortisolism in patients with AD 12 . Furthermore, chronic stress can also change HPA reactivity, causing blunted cortisol reactivity and increased cortisol levels after exposure to a stressor agent 13 . A previous meta-analysis with an adult sample observed higher cortisol reactivity in MDD individuals after the psychosocial stress task 14 . In this line, previous studies with older healthy adults have shown a higher cortisol response to pharmacological or psychological challenges compared to younger individuals 15 .
Specifically, to patients with AD, it is widely known to cause a hyperactivity of cortisol in basal conditions and a circadian cortisol hypersecretion in individuals with clinically diagnosis 16 . A recent study demonstrated that higher psychosocial stress reactivity may represent a preclinical sign of cognitive impairment 17 . It can explain some difficulty in these individuals to develop coping strategy to lead with stressor event 18 , mainly those with worsened cognitive performance 19 . Regarding patients with AD, it is also expected a difficulty in responding to a stressful stimulus due to failure to elaborate confrontational and problem-solving actions 19 .
Although previous studies have shown higher cortisol response for patients with MDD and AD, these studies have investigated through psychological stressor tests. Perhaps, few studies in the older sample may be explained by the difficulty to introduce stressor tests for this population. In this sense, another type of stressor agent, such as physical exercise (PE), with a lower cognitive demand during the task, may be more efficient, which promotes already known changes in HPA axis 20 . Most studies have observed higher cortisol levels after a single session of exercise in elderly individuals 21 , notwithstanding it is unknown if MDD and AD older adults show the same response for cortisol.
Thus, this study aimed to investigate the cortisol reactivity to a single session of PE in patients with MDD and AD and compare it to healthy control (HC) individuals. We expected an increase in cortisol hormone after acute PE among groups with more pronounced levels for patients with MDD and AD.

Protocol and registration
This research is part of a major study entitled "Efficacy of physical exercise in the Treatment of Major Depression, Alzheimer's Disease and Parkinson's Disease" that evaluates the chronic effect of PE. It was approved by the Ethics Committee for Studies on Humans from the Institute of Psychiatry of Universidade Federal do Rio de Janeiro under registration in CAAE: 24904814.0.0000.5263 and was registered under the Brazilian Registry of Clinical Trials (REBEC) protocol RBR-4M3K2C.

Recruitment and eligibility criteria
We recruited 34 male and female older adults (≥60 years old) with at least 4 years of education attended on Centre for Alzheimer's Disease and Related Disorders in the Institute of Psychiatry at the Universidade Federal do Rio de Janeiro. In this recruitment, three individuals declined to participate in the research, one in MDD and two in HC group. Participants should be diagnosed with MDD and AD according to DSM-IV (Diagnostic and Statistical Manual of Mental Disorders).
All patients were taken medication, and patients with AD were classified into mild and moderate stages according to the Clinical Dementia Rating (CDR). The HC group was recruited through usually familiar patients, neighbors, or colleagues, with age ≥60 years old, at least 4 years of education, no psychiatric disorder, and cognitive decline confirmed by a psychiatric evaluation. The exclusion criteria were severe stage of AD, any other psychiatric disorder, the existence of any musculoskeletal and audiovisual limitation, the use of corticoid medication, and who performed a dental procedure at least 7 days before saliva collection. In all, 10 individuals in HC group, 8 individuals in MDD group, and 13 in AD group were included. Some individuals included in this research were used to workout (walk on treadmill), while others were sedentary. The sample size was calculated using GPower ® software version 3.1, and the estimated sample was about 30 individuals.

Experimental procedure
The experimental procedure was performed in three visits. The first visit took about 2 hours. In this visit, written informed consent, anamnesis (age, education, medication, comorbidities), short-form International Physical Activity Questionnaire (IPAQ) (rating 1-5: sedentary, insufficiently active A and B, active, and very active) involving physical activity over the past 7 days 22 , Global cognitive state assessed by Mini-Mental State Examination (MMSE) 23 , and depressive symptoms by Geriatric Depression Scale (GDS) 24 were assessed.
In the second visit, 1 week after the first visit, individuals were submitted to a cardiopulmonary effort test in a treadmill ergometer (InbraMed Pro ® ) using a ramp protocol to estimate the maximal oxygen consumption (VO 2max ) 25 . The speed/slope increases every 30 seconds and was defined based on the difference between the initial and final speed. The test was configured to be finished in 10 min and, based on its results, the VO-2max was estimated. The electrocardiographic Digital MICROMED ® 12-lead was used to monitor and record blood pressure and electrocardiogram traceability to assure no cardiological disorders. Also, weight and height were recorded to determine the body mass index (BMI).
In the third visit, during the mornings, about 1 month after the first visit, the participants carried out a single bout of acute PE on the treadmill. Individuals underwent a 30-minutes walk on a treadmill, including 5 min of warm-up, 20 minutes of moderate intensity (70%VO 2max ), and 5 min of cool down. Blood pressure was recorded in preintervention and postintervention. The subjective rating perceived exertion (RPE) was measured by Borg's scale 26 , with rating ranging from 6 (no exertion at all) to 20 (maximal exertion). HR was obtained by a monitor, recording prestressor, post-stressor, and every 5 min during the acute PE.
Cortisol assay and collection All participants were instructed to avoid eating and drinking 1 h before the saliva collection. Salivary cortisol samples were collected in two moments, namely, preexercise and immediately postexercise, which were performed between 11 a.m. and 1 p.m. Before collection, participants were asked about sleep hours in the last night and were instructed about how to collect stimulated saliva. The secretion is promoted by mechanical movements in jaw after rinsing some water in the mouth. After collection, falcon tubes were stored and kept frozen (-20°C) until the cortisol assay.
After defrosted and centrifuged tubes at 3000 RPM for 10 min, the ELISA test (Enzyme Linked Immunosorbent Assay), using a previously validated and commercially available kit (IBL ® ) was performed. The cortisol limit of detection was 20.0 μL, and the intra-and inter-assay variability were considered 2.1 and 5.2%, respectively (within a 20.0-800.0 ng/mL dosage range).
A χ 2 test was used to compare gender among groups. Tukey post hoc test was also performed to find differences between them. Furthermore, a repeated-measure ANOVA (F) was used to analyze cortisol levels among groups (HC × MDD × AD) and within moments (preexercise and postexercise). The degree of freedom considered was Greenhouse-Geisser correction. Spearman correlation was performed to investigate correlation between delta cortisol and RPE pre, RPE post, HR pre, HR post, and IPAQ (MET/week). Linear regression was used to investigate the association among variables that showed difference in baseline condition (e.g., gender and hours of sleep) with delta cortisol.
To determine the magnitude of the difference in cortisol levels between pre-and post-values in HC, MDD, and AD groups, the effect size (ES) analyses were used. The results were interpreted according to Cohen's 27 (trivial: d≤0.2; small: d=0.2-0.49; moderate d=0.5-0.79; and large: d≥0.8). Analyses were performed using SPSS ® software version 26.0 (IBM Corp., New York, USA), and figure was performed using GraphPad Prism ® software version 5.01. The p-value <0.05 was the threshold for statistical significance. One-way ANOVA showed significant difference in treadmill slope during exercise (F=4.941; p=0.016; AD=HC<MDD), with significant difference between AD and HC groups (p=0.012) reaching the same exercise intensity stipulated in 70% of VO 2max during stressor test.

Sample characteristics in baseline are presented in
The repeated-measure ANOVA revealed no interaction for cortisol on the moment×group [F(2.000, 28.000)=1.285; p=0.293] and no effect for group (F=0.323; p=0.727). However, there was a significant effect for moment [F(1.000, 28.000)=4.930; p=0.035], with a decreased cortisol levels in postexercise for all groups (Figure 1). Gender as a covariate confirmed our outcomes and showed a trend to reduction in these levels (F=3.949; p=0.057). The comparison of delta cortisol among groups also showed nonsignificant difference (F=1.285; p=0.293). However, a reduction on cortisol levels of 6.5% for HC group, 8.8% for AD group, and 28.76% for MDD group was observed.
ES analysis revealed a small ES for patients with MDD (d=0.402) and trivial for patients with AD (d=0.166) and HC group (d=0.090). As expected, there was a positive correlation between HR pre and RPE pre (r=0.478; p=0.013) and between HR post and RPE post (r=0.478; p=0.016). Besides, in linear regression, nonsignificant results were observed for delta cortisol and gender (R 2 =0.088; p=0.106) and for delta cortisol and hours of sleep (R 2 =0.012; p=0.550).

DISCUSSION
The main objective of this study was to investigate the cortisol reactivity to a single session of PE in MDD, AD, and HC elderly individuals. We hypothesized that cortisol levels would increase after acute PE in all groups, with results more pronounced for patients with MDD and AD. Notwithstanding, a significant cortisol reduction for all groups was observed.
The association between HPA axis dysfunction and MDD has been investigated over the six decades 28 ; since then, both hypocortisolism and hypercortisolism at rest were found, which can impact the acute reactivity to a stressor agent. A previous meta-analysis 15 , with 43 articles in which 39 used pharmacological stressors and 6 used psychological tasks, concluded that aging increased the cortisol response to challenge. It was HC: healthy control; MDD: major depressive disorder; AD: Alzheimer's disease. speculated that the lack of negative feedback resulting in higher cortisol levels was due to an impairment in brain structures, such as the hippocampus and prefrontal cortex generally observed in MDD 29 and AD 30 . However, this explanation does not take into account some specific characteristics of these diseases in older individuals. For example, MDD older individuals are less likely to endorse affective symptoms and more likely to present cognitive changes, somatic symptoms, and loss of interest than are depressed young adults 31 . MDD individuals seem to be more influenced by lifelong stress, which may be related to the exhaustion phase of stress-causing significant changes in the HPA axis 32 . A previous study demonstrated that lower evening cortisol levels may predict poorer course in elderly with MDD 33 . It is important to take into consideration the medication used by this population, because psychoactive medication was generally associated with lower cortisol levels 34 . Other study found a diminished ability to respond to the stress of awakening among depressed older persons 35 . In a longitudinal study, it was observed that a low mean salivary cortisol concentration and a small difference between morning and evening cortisol concentration may be risk factors of depression 36 . Moreover, apathy, fatigue, and exhaustion 37 are very common symptoms in MDD and AD older individuals and are associated with hyporeactivity of HPA axis and consequent decrease in cortisol levels.
There are also other mechanisms related to aging. MDD and AD older individuals present more enhanced susceptibility to inflammaging 8 , which can affect the function of the glucocorticoid receptor (GR) and lead to hypercortisolism 38 and consequently a worse memory 39 . Although a previous meta-analysis has observed hypercortisolism in AD in the mornings 12 , it is possible to hypothesize that individuals with more inflammaging and GR resistance can show reduced cortisol reactivity. It is also expected that AD individuals may encounter difficulties or even be incapable of evaluating the potential threat of a given stress event 40 . In contrast, a recent study showed that higher salivary cortisol reactivity to acute psychosocial stress test in healthy individuals aged 50 years or above was a hallmark of developing cognitive impairment 5 years later 17 , probably because senescence was turned to senility when occur inflammation and disorders come.
In addition to these speculations regarding MDD and AD characteristics, it is important to consider the type of stressor chosen. Physical stressor was less explored and previous studies have investigated adult populations with MDD and found nonsignificant blood cortisol changes after bicycle ergometer till exhaustion 41 and a small increase of blood cortisol after bicycle ergometer till exhaustion 42 . A previous study with salivary sample observed cortisol reduction after bicycle ergometer with moderate intensity, which corroborates our results 43 .
In this study, cortisol reduction showed a small ES for all groups, with the highest effect for MDD compared to the other groups. Perhaps, these individuals had been more influenced by lifelong stress, which could be related to the exhaustion phase of stress-causing significant changes in the HPA axis 44 . Besides, other variables that could influence in results were not evaluated in this study, such as early life experience, chronic fatigue, social support, personality factors, chronic stress, and burnout [45][46][47] . We also have to point out the effect of acute stress using a rolling treadmill, because some individuals were extremely adapted to this type of stressor and others may experience some anxiety or afraid to be evaluated and anticipatory cortisol response must be considered. Meantime, this type of exercise is safe and adequate to use in this population.

Strengths and limitations of the study and clinical relevance
This study has some limitations. First, we cannot generalize for all patients with AD and MDD because of the heterogeneity and individual characteristics (gender, hours of sleep, physical performance, and BMI) of the small sample. Another limitation was that we only took two moments to observe this reactivity, and we know that it is necessary and important to observe the curve of this reactivity with more moments before and especially 15, 30, 45, and 60 min after exercise.
Notwithstanding, it is important to highlight that this is the first study, to the best of our knowledge, to evaluate cortisol reactivity of a physical stressor in MDD and AD individuals. However, there is much to investigate. It is desirable to study the elderly population with neuropsychiatric disorders to better understand how these individuals may cope with stressor agent and also contribute to elucidate the physiopathology of such disorders regarding to the stress, while it is unavoidable to observe the heterogeneity of the subjects considering that it is inherent in this type of study.
Thus, the relationship between acute stress, healthy elderly, and, even more, those with neuropsychiatric disorders is still under debate and more research is required in future. This type of study has a pivotal relevance to clinical practice because their findings could provide adequate support assistance, care, and information to multidisciplinary professionals who follow up patients with AD and MDD. These professionals can offer treatment to prevent or minimize consequence and deterioration on memory and mood in these patients, improving quality of life.
A future study with a larger and more balanced sample needs to be investigated, managing the influence of these variables in the outcome and considering that there is a cortisol variability within individuals of the same group.
Older individuals, regardless of the presence of mental disorder, show a decrease in salivary cortisol after a physical stressor. Nevertheless, this response seems to be more evident in MDD, which can be associated with