Endocrine and Metabolic Responses

A characteristic pattern of change in hormone secretion and the flux of metabolites is seen in a stressed animal. Typically, the sympathoadrenal and HPA systems are stimulated, together with the secretion of glucagon and the pituitary hormones growth hormone (GH), prolactin, and vasopressin. In contrast, there is a suppression of adrenal androgen production, and complex changes take place in the hypothalamic-pituitary-gonadal and hypothalamic-pituitary-thyroid axes that lead to lower concentrations of at least some of their end hormones. As regards metabolites, glycogen and fat are broken down so that the concentrations of glucose, lactate, nonesterified fatty acids (NEFA), glycerol, and ketone bodies increase in the plasma.

In muscle the persistence of stressful stimuli cause net protein breakdown, whereas in the liver the synthesis of a set of plasma proteins known as the acute- phase reactants (APR) is induced.

Any effects of stressors is superimposed on a background of normal age-related changes. In brief, there is little change in plasma adrenocorticotropic hormone (ACTH) or cortisol during aging in humans. The concentration of norepinephrine, which is derived from sympathetic nerve terminals as well as the adrenal medulla, tends to rise whereas that of adrenaline does not.

There is decreased secretion of GH, which takes place mainly at night, but probably not of prolactin or glucagon. Aging is accompanied by a decline in the turnover rate of glucose and a small increase in its concentration, reflecting insulin resistance. There is a tendency for the concentrations of fat metabolites also to increase, but the changes in their turnover rates are more controversial. Protein turnover decreases, as we expect from the decline in lean body mass with age.

Responses to hypoglycemia. The effect of aging on the endocrine responses to hypoglycemia has been of considerable interest because of the possibility of hypoglycemic episodes in treated elderly diabetics. Earlier studies, using a standard insulin tolerance test (0.1 U kg^-1), mostly showed no difference between old and young people, although there are sporadic findings of decreased cortisol, GH, glucagon, and adrenaline responses and an exaggerated rise in norepinephrine in the elderly.

More controlled experiments with the infusion of insulin have confirmed this lack of consistent change in endocrine responses to hypoglycemia, but have shown a tendency for fewer signs of sympathoadrenal activation (e.g., increased heart rate) in the elderly that is not always reflected in diminished responses of plasma catecholamines. There is no unanimity over their sensitivity to the symptoms of neuroglycopenia.

Responses to psychological stress. A variety of procedures have been used to study the effect of age on responses to psychological stress in the laboratory, nearly all of which have shown reduced heart rate responses in the old. A number of authors have measured catecholamines, finding a tendency for the norepinephrine, but not the adrenaline, response to increase with age as well as an increase in baseline values. There have been few studies of HPA responses to psychological stress in the elderly, and these demonstrate no failure to activate the system.

In only one study of the effects of chronic stress do different age groups appear to have been compared directly. Urinary free cortisol excretion has been measured in subjects ages 47-78 whose spouses had life-threatening illnesses. There was a correlation with age only in those who were identified as suffering from severe depressive illness. Others have studied elderly people caring for spouses with Alzheimertype dementia and compared them with spouses not needing to provide care; the caregiving spouses were classified as high or low stress.

Plasma norepinephrine was higher in the high-stress group than the others; there were no differences in plasma adrenaline, cortisol, or the sensitivity or receptor density of lymphocyte ß-adrenergic receptors. Although younger subjects were not studied, the results were contrasted with the ß-adrenergic downregulation observed in young men experiencing the chronic stress of homelessness. A further study reported a tendency for higher insulin (but not glucose) concentrations in those caring for demented spouses.

Response to trauma. The general response. The acute rises in cortisol and norepinephrine after elective surgery tend to be exaggerated in old people, whereas the opposite is seen for pituitary hormones. During the first few hours after moderate to major surgery, plasma cortisol in elderly patients was either similar to or significantly higher than its value in younger ones. There was no difference in plasma ACTH, whereas the GH response was lower in the elderly. However, in men ages 43-77 undergoing colectomy, there was no correlation between plasma cortisol and age. Cortisol and ACTH concentrations also tended to be higher in old than in young patients after cholecystectomy. After a minor procedure (inguinal herniorrhaphy), too, plasma cortisol was higher in the elderly, whereas there was no difference in the response of norepinephrine, adrenaline, or aldosterone.

Over the next few days after injury, plasma cortisol and, perhaps, norepinephrine tend to decline to normal levels more slowly in the elderly than in the young. After moderate to major surgery, cortisol continues to be higher in the old; no corresponding change in ACTH is detectable. There is also no difference in plasma GH during this period. The rate of 17-ketogenic steroid excretion (a crude index of cortisol production rate) was raised for a longer period after surgery in older patients, even though the absolute values were not different between young and old. In apparently the only study of the effect of aging responses very soon after accidental injury, we found no difference in plasma cortisol among patients ages 17-40, 41-65, and 66-92 with a wide range of injury severities.

Hypothalamic-pituitary-adrenal axis activation after hip fracture. We have studied hip-fracture patients, who are often frail and in poor health. Cortisol concentrations remain elevated 2 weeks later (unlike younger people with similar severity injuries). The same was true of the cortisol concentrations after they took dexamethasone overnight (suggesting impaired feedback inhibition by cortisol). These differences persisted when further measurements were made 8 weeks after the injury. Such a prolonged period of raised cortisol concentrations could have deleterious systemic effects.

Most reported cortisol concentrations have been measured in the morning, and comparisons would be invalidated if there were changes in its nychthemeral rhythm. Reliable indices of HPA function can be based on urinary measurements made over a 24-h period. We found that hip-fracture patients had a higher cortisol production rate and urinary freecortisol excretion rate than healthy elderly women.

The latter index, which is particularly useful because it usually reflects the integrated concentration of free, biologically active cortisol in plasma, rose on average approximately threefold. The cortisol production rate rose proportionately less because the metabolic clearance rate of cortisol was lower in the injured patients, so only a modest rise in production rate was necessary to account for the observed increases in plasma cortisol concentration and free-cortisol excretion.

A puzzling aspect of our studies is that hip-fracture patients did not show an increase in plasma ACTH. In fact, when concentrations were measured at the end of a 2-h baseline period leading up to injection of corticotropin releasing hormone (CRH), decreased values were seen compared with the healthy elderly. The total areas under the cortisol- and ACTH-time curves after the injection of CRH reflected this disparity, with that for cortisol being greater and that for ACTH smaller in the patients, although the incremental areas were both decreased.

The correlation between cortisol and ACTH seen in the control subjects was completely lost in the hip-fracture patients, and their apparent sensitization to ACTH was not reflected in the slope of the dose-response curve. Combined with a lack of change in the response to small doses of exogenous ACTH^1-24, this suggests the presence of a stimulus acting independently of ACTH. However, it is very difficult to know what such a stimulus might be.

Myocardial infarction. There have been two comparisons of responses to acute myocardial infarction in elderly and younger patients. One study found no difference in plasma cortisol or ACTH, measured over a period of 12 days after the event, between patients ages 65-81 and 33-55; there was also no difference in the response to a standard ACTH stimulation test on the twelfth day. The other study reported similar peak C-reactive protein (CRP) levels in patients younger than 60 and older than 70.

This study is valuable because it appears to be the only attempt to compare acute-phase protein concentrations between age groups after a standard insult, although it is well known that an acute-phase response occurs in old patients. Comparability is important to establish because CRP levels have been used to distinguish the effects of aging from those of underlying inflammatory disease.