| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Stress and Neuroendocrine Research Center, Exercise and Sport Research Institute, Department of Kinesiology, Arizona State University, Tempe, Arizona 85287-0404
Address all correspondence and requests for reprints to: Kathleen S. Matt, Department of Kinesiology, Arizona State University, Tempe, Arizona 85287-0404. E-mail: kmatt{at}asu.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The appropriate activation of the HPA axis in response to a challenge is important, because both hypo- and hyperactivity can be detrimental (2, 3). What is equally important to the activation of the stress systems is the ability to shut the systems off after the challenge has ended. A more rapid return of elevated ACTH and cortisol levels to basal or nonstimulated levels would be the mark of a dynamic HPA axis, indicating strong negative feedback sensitivity to cortisol. It has been proposed that aging results in a loss of negative feedback sensitivity, resulting in a prolonged recovery after a challenge (4).
Support for this hypothesis has mainly come from animal studies, with results from human studies being equivocal. Some studies have found that older individuals exhibit a greater HPA axis response to a challenge than young individuals (5, 6, 7, 8, 9, 10, 11), whereas others have not found any agerelated differences (12, 13, 14, 15). Many of these earlier studies have focused on the peak response to a challenge, thus giving a limited insight into the recovery process.
The aim of this study was 2-fold: 1) to examine whether aging is associated with a decreased ability to recover from a challenge to the HPA axis, and 2) to examine whether fitness is associated with a greater ability to recover from a challenge to the HPA axis among older individuals.
The existing literature on aging and HPA axis feedback sensitivity comes primarily from studies that used pharmacological challenges, and many of those have shown decreased sensitivity with aging (16, 17, 18). We were interested in using a natural nonpharmacological way to stimulate endogenous increase in ACTH and cortisol and measure the return to baseline values. Acute exercise is a potent stimulus of the pituitary-adrenal axis, especially at high intensity.
We hypothesized that aging would result in a loss of negative feedback sensitivity of the HPA axis. Specifically, we predicted that older women would have a slower recovery of ACTH and cortisol from the exercise bout, compared with young women, as indicated by a less steep slope of recovery and a larger area under the response curve (AURC). We also hypothesized that fitness [as measured by maximum oxygen consumption (VO2 max)] would be associated with a blunting of the age-related loss of negative feedback sensitivity and predicted that fit older women would have a faster recovery of ACTH and cortisol after the exercise bout compared with unfit age-matched women, as indicated by a steeper slope of recovery and a smaller AURC.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Nine young women (19–36 yr) and 22 older women (59–81 yr) were recruited from the community through distribution of fliers, advertisement in the Arizona State University Staff newsletter, a short announcement in a local fitness magazine, and word of mouth. The young women were all eumenorrheic, with a history of normal menstrual cycles, and had not been on oral contraceptives within the previous 6 months. They were tested in the early follicular phase (d 3–7 of the menstrual cycle) to minimize the confounding effects of estrogen on cortisol. They were untrained, as defined by having a VO2 max rated average or below for their respective age group according to the American Heart Association Classification Criteria (19). The older women were postmenopausal and were not taking hormone replacement therapy or any natural estrogen supplements. They were classified as either fit or unfit based on their VO2 max, with unfit having a VO2 max rated as average or below and the fit having a VO2 max rated as good or high for their age group, according to the American Heart Association Classification Criteria (19). All subjects provided a written consent before participating in the study. Additionally, a physician’s release was obtained for the older women before VO2 max testing. Subjects were excluded if they had chronic illness that could affect hormone levels, such as diabetes mellitus, liver or gallbladder disease, coronary heart disease, or renal failure. Also, subjects were excluded if they were taking any medication that could affect the stress response such as ß-blockers, glucocorticoids, diuretics, or other hormonal or hormone-mimetic medications, had smoked in the last year, or had a body mass index greater than 30.
Experimental design
Each subject reported to the laboratory on two separate days. The first visit was to determine the VO2 max and to obtain additional health history. The treadmill test was necessary to verify their aerobic fitness level and to determine the workloads used in the acute exercise trial. The second visit was the acute exercise trial. These trials are described below.
VO2 max. VO2 max was determined with a continuous, inclined treadmill test using the modified Naughton protocol (20) for the older subjects and the modified Åstrand protocol (21) for the younger subjects. The former protocol is a walking protocol, whereas the latter is a running protocol. Although the VO2 max may be slightly higher in an individual when measured during running vs. walking, the different protocols were used to elicit a VO2 max in approximately 7–12 min in each age group, thus minimizing confounding factors such as leg fatigue. Because the VO2 max measure was used to classify an individual as either "fit" or "unfit" based on age-related norms and was not analyzed as a continuous variable, the protocol selection would not be expected to affect the study outcome. Oxygen consumption was measured using indirect calorimetry. All measurements of expiratory gases were determined by standard techniques of open-circuit spirometry using a Parvo Medics True Max metabolic cart (Consentius Technologies, Sandy, UT) as previously described (22). Heart rates were determined from standard 12-lead electrocardiogram recordings (Quinton Instrument Co., Seattle, WA) in the older women and wireless heart rate monitors (Polar Electro Oy, Kempele, Finland) in the young women. At the end of each stage, a rating of perceived exertion was obtained using the Borg scale (23). VO2 max was considered to have been achieved if two of the three of the following criteria were met: 1) a plateau in VO2 with an increase in workload, 2) a respiratory exchange ratio of 1.1 or greater, or 3) a heart rate within 10 beats of the age-predicted maximal heart rate (24). The testing procedure was approved by the Internal Review Board at Arizona State University, and the termination criteria of the American College of Sports Medicine were used (25).
Acute exercise trial.
Subjects reported to the laboratory in the morning (between 0730 and 0800 h) after an overnight fast. An in-dwelling catheter was inserted into an antecubital vein and attached to a normal saline lock. A resting blood sample was drawn after 45 min of rest in a seated semireclined position in a quiet room with the lights dimmed. The subject then moved to the treadmill for a 15-min exercise bout. The initial 5 min served as warm-up, during which the subject exercised at an intensity equivalent to 50% of their VO2 max. The intensity was then increased to 70% VO2 max for 5 min, and the final 5 min was performed at 90% VO2 max. Heart rate was recorded throughout the trial with a Polar heart rate monitor. The proper intensity of the exercise was achieved by monitoring the oxygen consumption via indirect calorimetry during the exercise and adjusting the speed and grade of the treadmill accordingly. Immediately after the 15-min treadmill exercise, the subject momentarily stepped off the treadmill for a blood draw, then resumed walking at a pace of 2.5 mph. for a 5-min cool-down followed by a blood draw. The subject then returned to the recliner and five recovery blood samples were drawn at 30, 40, 55, 70, and 90 min, for a total of eight blood draws. A schematic outline of the test is shown in Fig. 1
.
|
For each blood draw, 5-ml samples were collected into tubes containing EDTA or heparin, centrifuged at 3000 rpm at 4 C for 15 min, and the plasma was saved in three or four aliquots and stored at –80 C until time of analysis. Aprotinin (Sigma-Aldrich, St. Louis, MO) was added as a preservative to each plasma aliquot designated for ACTH (0.27 mg/ml). Plasma ACTH was measured with a dual-antibody RIA, and plasma cortisol was analyzed using a coated-tube RIA, both assayed with commercially available kits (Valeant Pharmaceuticals, Costa Mesa, CA). Intra- and interassay coefficients of variation were less than 3 and 10%, respectively. All samples were analyzed in duplicate.
Data analyses
The responses to the exercise trial were analyzed by repeated-measures ANOVA (group-by-time ANOVA) for heart rate, systolic and diastolic blood pressure, and ACTH and cortisol levels, with a priori contrasts set up between the young and the older women, and the older and the older-fit women, respectively. Additionally, the ACTH and cortisol responses were analyzed as integrated area under the absolute concentration curve (AUC) and the area under the response curve (AUC corrected for baseline value: AURC), calculated by the method of the trapezoidal rule. Slope of recovery was calculated for both ACTH and cortisol by regressing the samples between the mean peak response and the final recovery sample on the respective time points, and using the slope of the line as the outcome variable. As expected, the hormones peaked at different time points; thus the last six time points were used to calculate the slope of recovery for ACTH and the last five time points were used to calculate the slope of recovery for cortisol. Group differences in AURC and the slope of recovery were tested using a one-way ANOVA. Analyses on the effects of aging were also done using an analysis of covariance where fitness was entered as a covariate. A standardized fitness score was used by calculating the z-score for each individual based on the mean of the group to account for normal age-related differences in fitness. All comparisons were considered statistically significant at P < 0.05, and the Huynh-Feldt correction was used when the data did not meet the assumption of sphericity. All data shown are the mean ± SEM. The statistical analyses were conducted using SPSS 11.5 software (SPSS Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. Although the VO2 max was significantly different among all three groups (young > older-unfit > older-fit), the means for the young and the older groups represent "average" levels for the respective age groups, thus indicating that the fitness levels were comparable between these two groups, whereas the older-fit group had a fitness rating of "good," as listed by the American Heart Association Cardiorespiratory Fitness Classification (19). Similar values have been reported by other studies (26, 27).
|
The actual intensities measured during the three stages of the exercise trial were 51, 71, and 89% (±1%) of VO2 max in the group as a whole, with no differences among the three groups. The workload during each stage, calculated as the metabolic equivalent was significantly different among the groups at each intensity, in line with the differences in VO2 max levels (young > older-fit > older-unfit, P < 0.05).
All of the groups had significant increases in heart rate (F3,78 = 567.63; P < 0.001), systolic (F3,68 = 67.54; P < 0.001) and diastolic blood pressure (F6,143 = 11.07; P < 0.001), ACTH (F6,129 = 7.04; P < 0.001), and cortisol (F4,97 = 11.17; P < 0.001) in response to the exercise trial.
Cardiovascular responses
The young women had a significantly greater heart rate response than the older women (F1,14 = 34.25; P < 0.001; Fig. 2), and this did not change after covarying for fitness. The heart rate response was not different between the fit and unfit older women [F1,16 = 0.2; P = not significant (NS); Fig. 2]. There were no group differences in either systolic blood pressure responses (F2,23 = 2.41; P = NS) or diastolic blood pressure responses (F2,23 = 1.00; P = NS) to the exercise bout (data not shown).
|
Baseline and peak ACTH levels were not significantly different between the young and the older women. The slope of recovery for ACTH was significantly steeper in the young women compared with the older women (–0.326 vs. –0.101, respectively; F1,14 = 4.66; P < 0.05; Table 2). These differences remained significant after covarying for fitness (analysis of covariance: F1,13 = 4.82, P < 0.05). However, the groups did not differ in the overall response whether calculated as AUC (F1,14 = 0.48; P = NS, data not shown) or AURC (F1,14 = 0.004; P = NS; Fig. 3 and Table 2). We also calculated the percentage of change for the recovery period (from the end of exercise to end of trial) and found that the young women recovered 109%, whereas the older women exhibited 34% recovery, a difference that did not reach statistical significance (F1,14 = 3.27; P = 0.09).
|
|
) or in the overall response whether calculated as AUC (F1,14 = 1.13; P = NS; data not shown) or AURC (F1,14 = 1.04; P = NS; Fig. 4 and Table 2).
|
Baseline and peak ACTH levels were not significantly different between the older and the older-fit women. The slope of recovery for ACTH was significantly steeper in the older-fit women compared with the older women (–0.346 vs. –0.101, respectively; F1,14 = 5.24; P < 0.05; Table 2). However, the groups did not differ in the overall ACTH response whether calculated as AUC (F1,15 = 0.02; P = NS; data not shown) or AURC (F1,15 = 0.23; P = NS; Fig. 3 and Table 2).
Baseline and peak cortisol levels were not significantly different between the older and the older-fit women. The slope of recovery for cortisol was not significantly different between the fit and the unfit women (F1,15 = 2.62; P = NS; Table 2); however, the older-fit women had a significantly greater overall response calculated as AURC (F1,15 = 4.64; P < 0.05; Fig. 4 and Table 2) or AUC (F1,15 = 7.19; P = 0.017; data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The results indicate that the prediction of prolonged recovery with aging was supported by the ACTH data, as shown by a significantly less steep slope of recovery in the older women compared with the young. When the recovery was calculated as either the percentage of change or the AURC from the end of the treadmill exercise to the end of the measurement period, the young exhibited a 3-fold greater recovery than the older women. Although the difference did not reach statistical significance (P = 0.09), the large effect size (Cohen’s d = 0.77) suggests that the difference was meaningful. However, the cortisol data indicated no differences in the rate of recovery or the AURC and thus did not support the hypothesis of a prolonged cortisol elevation after a challenge with aging.
The recovery from a challenge is an indirect measure of negative feedback sensitivity rather than a direct measure. It is important to consider that the time course of recovery could also be influenced by metabolic clearance rate and withdrawal of drive. Our findings extend the results of pharmacological studies that have found reduced negative feedback sensitivity with aging in both women and men, shown by blunted ACTH suppression in response to exogenous cortisol, either by a bolus injection (18, 28) or a prolonged infusion (17, 29).
The second prediction regarding the effects of fitness was also supported by the results for ACTH where the older-fit women exhibited a significantly steeper slope of recovery than the older-unfit, and looked in fact similar to the young individuals. When the recovery is expressed as the percentage of change, the older-fit women had values that were about 1.6 times greater than the older-unfit women, or about 50% of the recovery in the young women. Similar results have been reported in older men where ACTH levels returned to baseline faster in runners compared with sedentary age-matched controls after 30 min of cycling at 65% VO2 max followed by a low dose of combined CRH, LH, and TSH challenge (30).
The significantly greater cortisol response during recovery in the older-fit women compared with the unfit was an unexpected finding. Exercise at the same relative submaximal intensity has been shown to elicit similar hormonal responses in individuals, regardless of fitness level (31), and the cortisol levels before and immediately after exercise (15 min) were indeed not different between the groups. Although the fit women obviously exercised at a higher absolute intensity than the unfit, it seems an unlikely explanation for the higher levels during the recovery period, because the young women, who had the highest VO2 max and therefore exercised at the highest absolute intensity, had a response that was intermediate to the two older groups. The heart rate and blood pressure data also support that the groups were working at the same relative intensity. A more plausible explanation for the greater cortisol response during the recovery period in the older-fit group is that it is due to greater adrenal capacity, occurring through an adaptation of the physiological machinery to the repeated stimulus of exercise. Trained individuals have shown greater hormonal responses than untrained individuals at maximal or supramaximal intensity, due to greater secretory capacity (32).
This can be regarded as a beneficial adaptation because it increases the chance of survival in the face of acute stress of severe magnitude and/or repeated challenges. Our results are in agreement with Heuser et al. (33), who found that among middle-aged men, endurance athletes had significantly greater cortisol responses to a combined dexamethasone/human CRH challenge than sedentary controls, while exhibiting similar ACTH responses.
It is well recognized that the type of challenge used to activate the HPA axis may influence the response because there are thought to be different pathways of HPA axis activation. Thus, a physical stressor will have a more direct pathway to the paraventricular nucleus, and the response is more regulated by metabolic variables, whereas a psychological stressor will go through the limbic system (34, 35). These results may therefore not necessarily extend to the responses to psychological stress.
We did not include young, fit women in our study because we only expected to see an effect of fitness in individuals with perhaps already compromised responses, i.e. aging. However, the age-related differences were not as impressive as expected, possibly due to the small sample size as well as the excellent health of the women in our sample, a consequence of the strict inclusion /exclusion criteria used in the study. These stringent criteria, while carefully controlling for confounding age-related health variables, probably somewhat minimized the differences. A full factorial design in future studies will give us better means to compare the effects of aging and fitness and how these factors interact.
In summary, this study shows that aging is associated with a slower rate of recovery for ACTH after high-intensity exercise and that higher aerobic fitness among older women can attenuate or prevent these age-related changes in ACTH recovery. Future studies using longitudinal design are needed to explore whether these age-related changes in sedentary older women can be reversed or reduced with an exercise intervention. Additionally, it was found that fitness was associated with higher cortisol levels in the recovery period, suggesting perhaps greater adrenal sensitivity to ACTH after exercise. This adaptation could translate into a greater ability to overcome a physical challenge of severe magnitude through the stimulatory effects of cortisol on fuel mobilization.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AUC, Area under the curve; AURC, area under the response curve; HPA, hypothalamic-pituitary-adrenal; NS, not significant; VO2 max, maximum oxygen consumption.
Received October 1, 2003.
Accepted March 31, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. Aizawa, M. Iemitsu, T. Otsuki, S. Maeda, T. Miyauchi, and N. Mesaki Sex differences in steroidogenesis in skeletal muscle following a single bout of exercise in rats J Appl Physiol, January 1, 2008; 104(1): 67 - 74. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
