Characteristic of the normal aging process is a decline in physiologic reserve in many body regulatory systems, including those involved in the maintenance of fluid balance. The confluence of normal aging changes, common diseases, and the administration of many classes of drugs can readily lead to clinically evident disturbances of fluid balance, such as water retention or loss and to hyponatremia or hypernatremia with resultant symptomatic consequences. In some individuals, an impaired ability to conserve water may underlie the development of nocturnal urinary frequency as well as urinary incontinence.

The normal regulation of water and electrolyte balance involves the interplay of many homeostatic systems that operate to maintain the composition of fluid and electrolyte compartments within a narrow range. Because of alterations in the normal aging process, these homeostatic systems maybe compromised. The key regulatory components of fluid balance include (1) thirst perception, which governs fluid intake, (2) the kidney, which is governed by hemodynamic forces, and (3) hormonal influences of arginine vasopressin (AVP) or antidiuretic hormone (ADH), atrial natriuretic hormone (ANH), and the renin–angiotensin–aldosterone system, which control renal water and electrolyte excretion. Clinicians who are involved in the care of the elderly recognize that disturbances of water and electrolyte balance are common in this age group, especially when older persons are challenged by disease, drugs, or extrinsic factors such as access to fluids or control of diet composition.

(Table 88-1)

Body Composition

Aging effects on body composition have the potential to contribute to derangements in fluid balance. Normal aging is accompanied by a decrease in lean body mass, an increase in fat mass, and a decrease in total body water. Thus, total body water declines from the approximate values of 60% of body weight in young men and 52% in young women to 54% and 46%, respectively, in individuals older than age 65 years, primarily through a decrease in the intracellular fluid compartment. The decrease in total body water may place the elderly patient at increased risk for dehydration and/or hyponatremia when challenged by fluid loss or decreased fluid intake and at increased risk for fluid overload and hyponatremia when exposed to excessive oral or parenteral fluid intake.

Thirst and Fluid Intake

The ability to ingest a sufficient volume of fluid to meet body needs requires that thirst perception be present, that a suitable source of fluid be available, and that the individual be physically capable of obtaining and consuming the fluid. Under normal conditions, the requirement for daily fluid intake is approximately 30 mL/kg body weight. This requirement is further increased when there is high environmental temperature, fever, increased gastrointestinal, urinary, or respiratory fluid loss. Urine production typically ranges from 1.0 to 2.5 L per 24 hours and is modestly higher in men older than the age of 60 years. Older persons maybe challenged by increased amounts of fluid by either oral or parenteral routes, especially when they are in institutional acute care or long-term care settings and not in control of their fluid intake. In this circumstance, there is risk of volume overload and hyponatremia.

In healthy individuals, fluid intake is largely controlled by thirst sensation, which is regulated by both extracellular fluid volume and blood tonicity. Blood osmolality is the most important factor in the day to day perception of thirst. Thirst is usually stimulated when plasma osmolality rises to values greater than 292 mOsm/kg. Healthy persons older than age 65 years have shown evidence of an age-associated impairment in thirst perception so that when they were subjected to water deprivation sufficient to raise plasma osmolality to greater than 296 mOsm/kg, they exhibited diminished subjective awareness of thirst and consumed significantly less water than young subjects who were similarly water deprived and whose plasma osmolality rose to a lesser level (mean, 290 mOsm/kg). Other studies of elderly patients with cerebrovascular accidents have similarly documented impaired thirst perception in the face of volume depletion and hyperosmolality, both normally being potent stimuli for thirst. Elderly patients with Alzheimer’s disease fail to drink adequately when exposed to water deprivation in spite of the accompanying elevation of blood osmolality to levels above the usual thirst threshold. Further confounding the ability of the elderly person to ingest adequate amounts of fluid is the frequent presence of physical disability (e.g., blindness, arthritis, stroke) and impaired mobility, thus limiting the capacity of the patient to gain access to fluids.

Renal Changes of Normal Aging (Also See Chapter 85)

Structural and Functional Changes

Normal aging is accompanied by changes in renal anatomy and in renal function. Kidney mass undergoes progressive decline from a normal combined weight of approximately 250 to 280 g in young adults to between 180 and 200 g by age 80 to 90 years, with corresponding decrease in length and volume. Histologically, the number of glomeruli decline by 30% to 40% with increasing age and the percent of glomeruli that are hyalinized or sclerotic, increase to 10% to 30%. This process accelerates after the age of 40 years, and the residual glomeruli also undergo changes with age. Thus, there is a decrease in effective filtering surface, and an increase in the number of mesangial cells, a decrease in number of epithelial cells, and thickening of the glomerular basement membrane.

As part of the normal aging process, there are changes in the renal vasculature that lead to obliteration of the arteriolar lumen and loss of the glomerular capillary tuft. These changes take place primarily in the cortical glomeruli. In the juxtamedullary area, glomerular sclerosis may lead to anastomosis between afferent and efferent arterioles with direct shunting of blood between these vessels. Blood flow to the medulla, through the arteria rectae, is maintained in old age.

The anatomic changes of the aging kidney are paralleled by alterations in renal function although a direct relationship between anatomic and functional changes is not firmly established. Renal blood flow declines during the course of normal aging by approximately 10% per decade after young adulthood so that by the age of 90 years, the renal plasma flow is approximately 300 mL/min—a reduction of 50% of the value found at 30 years of age. The decrease in renal perfusion is most extensive in the outer cortex with lesser impairment of inner cortex and minimal effect on the medulla.

Glomerular filtration rate (GFR) remains relatively stable until age 40 years, after which it undergoes decline at an annual rate of approximately 0.8 mL/min/1.73 m2. Because there is much individual variability within the elderly population, longitudinal studies show that not all aging persons will undergo a decline in GFR. The Cockcroft–Gault formula for creatinine clearance has been used to estimate GFR. A more accurate estimate of GFR is now recommended using the abbreviated version of the Modification of Diet in Renal Disease (MDRD) formula, which is based on serum creatinine, age, gender, and race. The calculation can easily be made by use of a downloadable Web-based calculator (). In persons older than the age of 70 years, approximately 26% will have estimated GFR less than 60 mL/min/1.73 m2.

Water Regulatory Capacity

Renal Water Retention

The aging kidney exhibits a modest age-related impairment in the ability to dilute the urine and excrete a water load. The ability to generate free water is dependent on several factors, including adequate delivery of solute to the diluting region (sufficient renal perfusion and GFR), a functional intact distal diluting site (ascending limb of Henle’s loop and the distal tubule), and suppression of ADH in order to escape water reabsorption in the collecting duct. The age-related decline in GFR is the most important factor in the aged kidney’s diluting capacity. The presence of an age-related diluting defect that is independent of changes in GFR remains controversial.

The diluting capacity of the aging kidney has been evaluated in men by determining the urine osmolality and free water clearance response to acute water loading. In young men (mean age 31 years), minimum urine osmolality was 52 mOsm/kg, whereas in middle-aged men (mean age 60 years), minimum urine osmolality was 74 mOsm/kg, and in the older men (mean age 84 years), it was 92 mOsm/kg. The free water clearance was lowest in the older group. However, when these results were expressed as free water clearance per mL of GFR, the values were not different, suggesting that the defect in diluting capacity was a consequence of an age-related reduction in GFR. In a similar study in which healthy elderly subjects aged 63 to 80 years (mean, 72 years) and healthy young subjects aged 21 to 26 years (mean, 22 years) were administered a water load, the peak free water clearance was 5.7 mL/min in the older group and 8.4 mL/min in the younger group. However, when adjustments were made for changes in creatinine clearance, the difference in these indices was not statistically significant.

Other studies suggest that age-related free water clearance defects persist following correction for a lower GFR. A significantly lower free water clearance/creatinine ratio was found in elderly nursing home residents as compared to healthy younger subjects. The maximal urinary dilution (urinary/plasma osmolality) declined from 0.247 in younger subjects to 0.418 in elderly subjects.

In addition to impaired diluting capacity, the age-related decrease in renal plasma flow and GFR can lead to passive reabsorption of fluid, thereby increasing the risk of water overload and hyponatremia. This effect is clinically evident in elderly patients who have congestive heart failure, extracellular volume depletion, and hypoalbuminemia.

Diuretics, especially thiazides, can decrease renal diluting capacity. In the elderly, this effect becomes especially important when it is superimposed on the already diminished diluting capacity of the aged kidney, thus increasing the risk of developing water intoxication by impairing the ability to excrete excess water promptly.

Renal Water Loss

It has been known for many years that there is an age-related change in renal concentrating capacity. In a study of healthy men aged 40 to 101 years who underwent 24 hours of water deprivation, maximum attainable urine specific gravity declined from 1.030 at 40 years to 1.023 at 89 years. Hospitalized men aged 23 to 72 years who underwent 24 hours of dehydration demonstrated a progressive decline in maximum urine osmolality with increasing age. In healthy, active community-dwelling participants in the Baltimore Longitudinal Study of Aging, young subjects responded to 12 hours of water deprivation with a marked decrease in urine flow (1.02±0.10 to 0.49±0.03 mL/min) and a moderate increase in urine osmolality (969±41 to 1109±22 mOsm/kg), whereas elderly subjects were unable to significantly alter urine flow (1.05±0.15 to 1.03±0.13 mL/min) or osmolality (852±64 to 882±49 mOsm/kg). This age-related decline in urine concentrating ability persisted after correction for the age-related decrease in GFR.

The effect of age on renal tubular response to vasopressin has been assessed by measuring the urine-to-plasma inulin concentration ratio in men who ranged in age from 26 to 86 years and who were free of clinically demonstrable cardiovascular and renal disease. The ratio fell from 118 in young men (mean age 35 years) to 77 in the middle-aged group (mean age 55 years) and to 45 in the older men (mean age 73 years). The decreased renal sensitivity to vasopressin with age maybe a result of an age-related increase in vasopressin secretion. Animal studies demonstrate that chronic exposure of the kidney to increased vasopressin results in diminished renal responsiveness to the hormone. Thus, an age-related increase in vasopressin secretion may result in down-regulation of renal AVP receptors and be the basis for decreased renal concentrating capacity in the elderly.

Sodium Regulatory Capacity

Renal Sodium Retention

Several situations may lead to sodium retention and accompanying water overload in the elderly. The previously described age-related decrease in renal blood flow and GFR favors enhanced conservation of sodium. Disease states resulting in secondary hyperaldosteronism, such as congestive heart failure, cirrhosis, or nephrotic syndrome, are common in the elderly. In addition, drugs such as nonsteroidal anti-inflammatory agents, which are frequently used in the elderly, may promote sodium retention.

Renal Sodium Loss

Elderly individuals are more likely to have exaggerated natriuresis after a water load than are younger subjects. Patients with benign hypertension have an excess of sodium excretion in association with increased age. The aged kidney’s response to salt restriction is sluggish so that restriction in sodium intake to 10 mEq/d was followed by a half-time for reduction of urinary sodium excretion of 17.6 hours in young individuals and 30.9 hours in old individuals. These data suggest that the aging kidney is more prone to sodium wasting. Mechanisms underlying this tendency maybe multifactorial and are related to the effects of age on ANH, the renin–angiotensin–aldosterone system, and renal tubular function.

Vasopressin System in Normal Aging (Table 88-2)

Neurohypophyseal System

The magnocellular neurons of the hypothalamus where AVP is synthesized do not appear to undergo age-related degenerative changes. There is no evidence of the cell destruction, neuronal dropout, or loss of dendritic arborization found in other segments of the aged brain. Moreover, neurosecretory material in supraoptic nuclei (SON) and paraventricular nuclei (PVN) in elderly persons does not appear to differ in amount from that in younger subjects.

Morphologic data provide evidence that these nuclei, in fact, become more active with age. In the human hypothalamic neurohypophyseal system of subjects ranging from 10 to 93 years of age, a gradual increase in the size of the SON and PVN was observed after 60 years of age, suggesting that AVP production increases in senescence. Similar changes have been observed in the nuclear size of AVP neurons. Possibly contributing to the maintenance of normal or increased amounts of AVP in the magnocellular neurons is the observation of a 25% reduction in the rate of axonal transport of AVP and its associated neurophysin with advancing age. Thus, it appears that neurosecretory activity of hypothalamic AVP neurons does not decrease but, in fact, remains constant or is elevated with age.

Basal Plasma Vasopressin Levels

There are conflicting data regarding basal concentration of AVP in the blood during normal aging. In young normal individuals, there is a diurnal rhythm of vasopressin secretion, with greatest AVP secretion occurring at night. This rhythm appears to be linked to the wake–sleep cycle rather than to time of day. The sleep-associated peak is absent in the majority of healthy elderly persons. Low AVP levels and the lack of definite diurnal rhythm may, to some extent, explain increased diuresis during the night in some elderly individuals.

Healthy elderly subjects have been found to have basal plasma AVP levels that were significantly lower than in young subjects. In association with the reduced AVP concentration, plasma osmolality was elevated, suggesting that the elderly subjects had a water-losing state similar to partial diabetes insipidus.

Other studies indicate that basal plasma levels of AVP did not differ among young, middle-aged, and elderly healthy individuals who were studied under both supine and ambulatory conditions. Furthermore, there were no differences in plasma osmolality between the groups.

In contrast to the above are reports of elevated basal vasopressin levels in healthy elderly persons as compared with younger individuals. Healthy human subjects aged 20 to 80 years have been observed to exhibit a progressive rise in plasma AVP concentration with age, which become most evident in subjects older than age 60 years. Baseline plasma AVP correlates with serum osmolality in younger adults but not in elderly subjects.

Debate exists regarding a sex-related difference in plasma AVP levels in the elderly population. There are reports of a twofold higher plasma AVP concentration in elderly men as compared to women, while other studies fail to identify a gender effect on basal plasma AVP.

A rise in basal plasma AVP with age cannot be attributed to age-related changes in vasopressin metabolism. No differences were found between young and old subjects in vasopressin half-life, volume of distribution, or clearance. Thus, evidence of increased basal plasma vasopressin most likely reflects age-related changes in central control systems for vasopressin release.

Vasopressin Stimulation

Secretion of AVP normally varies in response to changes in blood tonicity, blood volume, and blood pressure. Hormone release is also affected by other variables such as nausea, pain, emotional stress, a variety of drugs, cigarette smoking, and glucopenia. In recent years, a growing body of information suggests that normal aging affects the way these stimuli act and interact to influence AVP release.

The major physiologic stimulus for vasopressin secretion in humans, plasma osmolality, is regulated by hypothalamic osmoreceptors. Osmoreceptor sensitivity in the elderly population has been assessed by comparing the AVP response to hypertonic saline infusion in healthy elderly persons (age 54–92 years) to the response in younger individuals (age 21–49 years). Hypertonic saline raised plasma osmolality with a consequent increase in plasma AVP in both groups, but the hormone concentrations in the older subjects were almost double those in the younger subjects. Thus, for any given level of osmotic stimulus, there was a greater release of AVP in the elderly, suggesting that aging resulted in osmoreceptor hypersensitivity.

Use of water deprivation as a stimulus for vasopressin secretion has supported the concept of an age-related enhancement in vasopressin secretion. Water deprivation for 24 hours in young healthy individuals (age 20–31 years) and healthy elderly men (age 67–75 years) demonstrated that the older persons responded with higher serum concentrations of AVP than the younger individuals.

The sensitivity of the hypothalamic–neurohypophyseal axis to volume/pressure stimuli has been studied. In response to acute upright posture after overnight water deprivation, older subjects (age 62–80 years) demonstrated the expected changes in pulse and blood pressure but only 8 of 15 older individuals experienced increased plasma vasopressin in contrast to a rise in plasma vasopressin in all young subjects (age 19–31 years). This and other studies suggest the presence of an aged-related impairment of volume-/pressure-mediated vasopressin release.

The ability of intravenous ethanol infusion to inhibit AVP secretion has been evaluated in young (age 21–49 years) and old (age 54–92 years) subjects. Younger subjects demonstrated a sustained inhibition of AVP secretion during the infusion of ethanol, whereas there was a paradoxical response in the older group with initial AVP inhibition followed by breakthrough secretion and rebound to twice basal levels. Not only was ethanol less effective in inhibiting AVP release in the elderly, but it eventually lost its suppressive effect entirely as a result of the introduction of a hyperosmotic stimulus resulting from the ethanol-induced constriction in plasma volume.

Metoclopramide can stimulate vasopressin secretion in humans through cholinergic mechanisms. Intravenous metoclopramide administration to normal elderly subjects aged 65 to 80 years and to normal young subjects aged 16 to 35 years produced significantly higher plasma AVP concentrations in the older group with no significant changes in plasma osmolality, blood pressure, or heart rate. Response of AVP to cigarette smoking and insulin-induced hypoglycemia, as well as to metoclopramide, has been evaluated in male subjects aged 22 to 81 years. The AVP response to metoclopramide and to smoking was significantly higher in the older group as compared to two younger groups. In contrast, the AVP response during the insulin hypoglycemia test was identical in pattern and magnitude in all age groups.

The stimulation studies indicate that, in aging, AVP response to osmotic stimuli is increased because of a hyperresponsive osmoreceptor, whereas AVP response to upright posture is reduced because of impaired baroreceptor function. Input from the baroreceptor to the osmoreceptor is usually inhibitory, so that a defect in this reflex arc would result in lesser dampening and consequent heightening of osmotically stimulated ADH release. When coupled with the many alterations in renal function that occur with aging, these changes can increase the risk of elderly persons for hyponatremia by impairing their ability to excrete excess water promptly.

Age-Related Changes in Atrial Natriuretic Hormone Secretion, Regulation, and Action

ANH is synthesized, stored, and released in the atria of the heart. Through its action on the kidney, ANH produces a pronounced natriuresis and diuresis; through its action on blood vessels, it produces vasodilation and decreases blood pressure in both normal and hypertensive individuals. As an important regulator of sodium excretion, ANH maybe a significant factor in mediating the altered renal sodium handling of age.

In a comparison of young normal men with elderly male nursing home residents, a fivefold increase in mean basal ANH levels and an exaggerated ANH response to the stimulus of saline infusion has been observed in the elderly group. In response to the stimulus of head-out water immersion, ANH levels in healthy old individuals (age 62–73 years), which were twice as high at baseline than in young subjects (age 21–28 years), rose to a greater extent than in the young. Healthy male and female subjects aged 22 to 64 years have been studied to determine the influence of age on circulating levels of ANH, both under basal conditions and after the physiologic stimulation of ANH release by controlled exercise using a bicycle ergometer to increase heart rate to 80% of maximum predicted rate. Subjects older than 50 years of age had higher baseline levels and a greater response to exercise when compared to subjects younger than age 50 years. Thus, increasing age results in increased ANH basal levels and an increased ANH response to both physiologic and pharmacologic stimuli, perhaps as a consequence of age-related decrease in cardiac muscle compliance.

The renal effects of ANH maybe exaggerated in elderly versus young individuals. The natriuretic response to a bolus injection of ANH was higher in older individuals (mean age 52.3 years) as compared with younger subjects (mean age 26 years). No change with age was noted in the blood pressure response to ANH intravenous infusion after correction for higher ANH levels in the elderly.

ANH is known to interact with the renin–angiotensin–aldosterone system. Increases of ANH result in suppression of renal renin secretion, plasma renin activity, plasma angiotensin II, and aldosterone levels, suggesting indirect inhibition of aldosterone secretion by ANH. Minimal increases in ANH within physiologic levels produced by slow-rate ANH infusion can inhibit angiotensin II-induced aldosterone secretion in normal men, suggesting a direct inhibitory effect of ANH on aldosterone release. Thus, ANH may promote renal sodium loss both through inhibition of aldosterone release and through a direct natriuretic action.

ANH maybe an important mediator of age-related renal sodium loss. This effect maybe the consequence of increased basal ANH levels, increased ANH response to stimuli, increased renal sensitivity to ANH, and ANH-induced suppression of adrenal sodium-retaining hormones.

Renin–Angiotensin–Aldosterone