Printer Friendly

Water balance in elderly people: is there a deficiency of vasopressin?


The possible change in osmoregulatory ability with age is of importance because the incidence of both hyper- and hypo-natraemia increases in the elderly population |1~. There are currently two lines of thought emerging from the literature on the changes in homoeostatic mechanisms of water balance with increasing age. It is now documented repeatedly that there is a decrease, with age, in the thirst response to osmotic change and it is possible that perception of other stimuli to fluid ingestion is also diminished |2~. There is, however, more controversy regarding changes in osmoregulation of vasopressin (AVP) secretion and the ability to conserve and excrete water appropriately. Early studies by Helderman et al. |3~ suggested that there was an increase in sensitivity of AVP secretion in response to a rise in plasma osmolality. More recent studies have failed to confirm this, and indeed have suggested perhaps the converse may be true |4, 5~.

Studies which have explored the ability of elderly people to excrete a water load have indicated that it is impaired, and that more time is required for full excretion of the load |6~. To explore further the changes in osmoregulatory mechanisms which may occur with age we have compared the biochemical and AVP responses to a short period of fluid deprivation (9 h) and a standard oral water load (WL) in groups of healthy young and elderly volunteers.

Patients and Methods

Ten healthy men aged 18-45 years (mean 28 years) comprised the young group (Y) and 21 men and women volunteers aged 58-74 years (mean 68 years) the elderly group (E). All were non-smokers, none was taking any medication and all gave written informed consent. Exclusions were made on the basis of serious illness (past or present), previous head injury or bacterial meningitis or abnormality of serum electrolytes, liver function tests, blood glucose, full blood count or urinalysis. All subjects abstained from alcohol for 24 h before water-load tests.

Subjects were fasted and deprived of fluid from midnight. A venous cannula was inserted for blood sampling at 09 h 00 and a baseline sample was taken after the subject had rested in the sitting position for a period of 30 min. Subjects emptied their bladders and then drank 20 ml/kg of water over 15-20 min. Blood samples were taken at O, 30, 60, 90, 120, 180 and 240 min after fluid ingestion, and blood pressure was taken at 10 min intervals for the first hour and then at half-hourly intervals over the next 3 hours. Subjects voided urine at O, 60, 90, 120 180 and 240 min after the water load. A visual analogue scale for nausea was recorded before the start of water loading and at standardized intervals throughout the study. Subjects remained seated during the study except when voiding urine.

Blood was drawn into chilled syringes and transferred to cooled, heparinized tubes to be centrifuged immediately at 2000 g at 4|degrees~C. Plasma was separated from the cells, and aliquots were taken for analysis of osmolality (POs) and vasopressin. Plasma for osmolality measurement was stored at 4|degrees~C and measured within 24 h. Plasma for AVP assay was stored at -40|degrees~C for assay within 2 months.

The volume of urine voided at each time point was recorded and aliquots were taken for measurement of osmolality (UOs) and calculation of free water clearance (C|H.sub.2~O) as defined by the function urine volume/time x (1 - UOs/POs).

Osmolality of urine and plasma samples was measured by the depression of freezing point method (Roebling Osmometer). Plasma AVP was estimated by a sensitive and specific radio-immunoassay after extraction from plasma by magnesium silicate absorption |7~. Inter- and intra-assay coefficients of variation were 12.6% and 9.7%, respectively, at 2 pmol/l. The limit of assay detection was 0.2 pmol/l.

Ethical approval was obtained from Newcastle Health Authority Joint Ethics Committee.

The results of each group were compared by two-tailed t tests. Unless otherwise stated, results are shown as mean |+ or -~ SEM.


After 9 h of fluid deprivation, the elderly group had a significantly higher POs and lower UOs than the young. Plasma AVP concentration was, however, significantly lower in the elderly than in the young group despite the higher POs (Table). The relationship between baseline TABULAR DATA OMITTED plasma AVP and UOs is shown for the two groups in Figure 1.

Figure 2 shows the pattern of excretion of the water load (WL) as the calculated range of percentage of load excreted (mean |+ or -~ 2 SD) for the two groups. There was close approximation at most time points, but at 1 h there was a significant difference between the groups at which time the young excreted 12 |+ or -~ 3%, but the elderly subjects 21 |+ or -~ 2%, (p = 0.031). By 90 min there was no longer a significant difference and this persisted throughout the rest of the study. At 4 h both groups had excreted over 90% of the load ingested (Y 91.9 |+ or -~ 4.6%, E 92.8 |+ or -~ 7.6%, p |is greater than~ 0.05).

The pattern of WL excretion showed marked differences between the groups in changes of C|H.sub.2~O and urine osmolality (see below) as demonstrated in Figure 3 (c and d). The elderly subjects had a significantly higher baseline C|H.sub.2~O (E - 0.35|+ or -~ 0.18 ml/min, Y -0.95|+ or -~0.1 ml/min, p = 0.036), consistent with lower urinary concentrating ability. The C|H.sub.2~O of the elderly was significantly greater at 60 min than in the young (E 1.6 |+ or -~ 0.4, Y -0.03 |+ or -~ 0.4 ml/min, p = 0.028) consistent with the higher percentage WL excreted at 60 min by the elderly. After this time, however, the young achieved a significantly greater C|H.sub.2~O. The increased C|H.sub.2~O persisted for longer in the elderly and this group was still excreting free water at 4 h. The young therefore achieved excretion of the water load by a sharp increase in C|H.sub.2~O but short duration, the elderly achieved the same end result by a lesser but more prolonged increase in C|H.sub.2~O.

At all time points, AVP in the elderly was significantly lower than in the young. There also appeared to be a longer suppression of AVP to minimal levels in the elderly than the young.

Figure 3a shows that the initial significant difference in plasma osmolality was no longer apparent after ingestion of the water load and did not recur after the load was excreted. The magnitude of fall in POs after ingestion of water load was, however, significantly different in the two groups (Y 5.1 |+ or -~ 0.7 mOsm/kg, E 7.7 |+ or -~ 0.4 mOsm/kg, p = 0.003). The young returned to a mean POs 1.8 |+ or -~ 0.7 mOsm/kg, lower than that at starting (p = 0.035), but the elderly remained at a POs 3.5 |+ or -~ 0.5 mOsm/kg, lower than baseline (p |is less than~ 0.001). The difference between the two groups was not, however, significantly different.

Figure 3c shows the change in UOs with respect to time after the WL. The young achieved a greater urine dilution than the elderly subjects but this failed to reach significance (Y 72 |+ or -~ 6, E 91 |+ or -~ 8 mOsm/kg, p |is greater than~ 0.05). The elderly, however, maintained a lower urinary osmolality for a longer period than the young and there was a significant difference in UOs at 4 h (Y 392 |+ or -~ 53, E 228 |+ or -~ 23, p = O.002).

There was no significant fall in blood pressure in any of the studies. During the water load, one young subject experienced a significant degree of nausea ( |is greater than~ 10 mm on linear scale). Excluding his data from the analysis did not alter the statistical significance of any of the results. As expected there were no significant differences in the elderly between male and female subject results. Data were therefore combined for the statistical analysis.


There were three remarkable observations from this study. Firstly, there was no effective difference in the ability of healthy elderly or young subjects to excrete an oral water load except in the very early stages when the elderly began to excrete water more rapidly than the young. Secondly, the osmoregulatory mechanisms of achieving this result were quite different between young and elderly subjects, and thirdly, there were substantial differences in baseline biochemistry and AVP between the two groups after a short period of fluid deprivation.

Owing to their lower baseline plasma AVP concentration, the elderly subjects achieved a concentration of AVP sufficiently low to allow urinary dilution more rapidly than the young after ingestion of the water load and consequently diuresis began more rapidly in the elderly subjects. The pattern of water excretion then changed between the two groups. The young subjects had a short but marked increase in C|H.sub.2~O, the elderly a less marked but more prolonged change. The magnitude of maximum C|H.sub.2~O in young and elderly subjects is similar to that found in other studies |6, 8~. This might be solely a consequence of a reduction in GFR with age |6~ or it could be secondary to a reduction in maximal dilution ability with age. The significant difference in group mean value remained, however, after correction of the maximum C|H.sub.2~O of each individual elderly subject for variability in GFR. We also found no significant difference between the young and elderly subjects in minimum urine concentration but there was a trend for the elderly to have a significantly higher urine concentration during maximal free-water clearance. Other studies have found similar results after a water load |8~, and there is evidence that the ageing kidney has a diminished ability to dilute urine owing to an impaired sodium conservation mechanism |9, 10~.

It would seem that the potential problem of incomplete bladder emptying was not a source of error in this study since this would lead to an underestimate of WL excretion by the elderly subjects. The dilution of urine to |is less than~ 100 mOsm/kg would also argue against incomplete urine collection at each time point.

Despite these changes in osmoregulatory mechanisms with age there was no net effect on the ability to excrete the WL. It might be anticipated, however, that a large volume of water or salt-deplete fluids (e.g. 5% dextrose) would lead to a deficit in excretion by elderly subjects compared with young owing to the diminished maximal C|H.sub.2~O. This may be particularly relevant to the occurrence of hyponatraemia in the hospitalized elderly |11, 12~.

After the short period of fluid deprivation, the elderly subjects had a substantial higher baseline POs than the young. Baseline AVP was, however, significantly lower in the elderly group. This is in contrast to other studies which have shown either no difference in basal AVP concentrations in elderly subjects |3, 13, 14~ or higher concentrations |15, 16~. One study has found a lower concentration of AVP in elderly than in young subjects |17~. These differences in findings may be explained in part by differences in the age group studied. The AVP concentrations observed in the elderly group would not be expected to achieve urinary concentration greater than 400 mOsm/kg |18~. We found a mean urinary concentration of over 500 mOsm/kg at baseline in our elderly subjects which suggests that urine was concentrated by a mechanism independent of AVP, or renal AVP receptors were up-regulated.

Low plasma AVP concentration, high normal plasma osmolality and greater than expected urine osmolality would suggest that the elderly subjects were in a similar biochemical and hormonal state to that seen in patients with partial cranial diabetes insipidus (CDI). This could be because of inadequate AVP release owing to decreased stores or a change in release mechanism, or because of an inadequate stimulus to AVP release through dysfunction of the osmoreceptor cells. In partial CDI, despite 'inadequate' plasma concentrations of AVP compared with normals, urinary concentration can occur to a limited extent |19, 20~. This is thought to be because of increased renal tubular sensitivity to AVP, probably as a result of AVP receptor up-regulation |21~. This increased tubular sensitivity to AVP was reported in early studies of water balance in elderly people where infusion of submaximal concentration of vasopressin during sustained maximal diuresis led to a much greater rise in urine osmolality in an elderly than in a young sample |8~. Patients with a mild partial CDI pass 2.5-4 litres of urine per day and maintain, by adequate fluid intake, a high normal plasma osmolality. Elderly people. with reduced thirst stimulus to fluid intake, may therefore have been more dehydrated than the young at the start of the WL and this might have been reflected in the lower maximum C|H.sub.2~O of the elderly group. Despite the putative initial relative dehydration, the elderly subjects excreted a similar percentage of the WL to the young, lending further evidence to the hypothesis that elderly people have an abnormality of the mechanisms of fluid conservation.

This is the first study to our knowledge that has investigated physiological osmoregulation in such a large group of healthy, non-hospitalized elderly subjects. The biochemical and hormonal findings which suggest that partial AVP deficiency develops with age are controversial. Previous research has shown that there may be an increased AVP response to rise in plasma osmolality |2, 3, 22~ and this is the view expressed in many standard texts |23~. Not all studies have confirmed this, however. Li et al. |17~ showed that elderly people had a smaller increase in plasma vasopressin than the young after 14 h of water deprivation and these authors suggested that elderly people may be in a state of incomplete CDI. Mukherjee et al. |24~ showed an inadequate rise in urine osmolality after 54 h of water deprivation in elderly subjects, and suggested that there was a dysfunction of the osmoreceptor leading to inadequate AVP release. Two recent reports have also indicated that there is perhaps a decreased sensitivity of vasopressin release to rise in plasma osmolality in elderly subjects |4, 5~.

Studies on non-osmotic stimulation of vasopressin by orthostatic stimulus have

shown a failure of vasopressin secretion with age |14, 22, 25~. Other studies have shown an age-associated augmentation in AVP release in response to metoclopramide stimulation |13, 22~ but no age effect on AVP release stimulated by hypoglycaemia |13~. The storage of AVP in hypothalamic nuclei and the neurohypophysis has been shown to increase with age |26, 27~. This could suggest an abnormality of the mechanism of AVP secretion although it has previously been interpreted as evidence of an increase in demand in view of the apparent increase in osmotic sensitivity of AVP secretion with age |3~.

Miller |23~ and others |13~ have argued that the apparent changes with age of osmotic regulated AVP secretion are related to an imbalance of afferent inputs to AVP secreting neurons in the hypothalamus secondary to reduction of afferent impulses from arterial baroreceptors. This imbalance leads to altered sensitivity to other stimuli to AVP secretion, possibly mediated by increased endogenous opioid activity. Such a hypothesis is not incompatible with the findings of this study, and further work to explore all facets of AVP secretion in individual elderly subjects rather than population studies may elucidate the complex and probably variable interrelations and pathophysiology.

Many elderly people complain of polyuria, especially at night. This has been usually attributed to a decrease in urinary concentrating ability subsequent to a diminished GFR which is well recognized to occur with age |2, 28-30~ as well as to urological problems such as prostatism. This study has provided evidence that at least part of the diminished urinary concentrating ability may be due to a deficiency of AVP. Miller and Shock |31~ suggested that renal tubular concentrating ability in elderly people was relatively insensitive to AVP administration. This was not confirmed, however, in later studies |8, 24~. This difference in findings remains unexplained. Similar to the argument used to explain the inadequacy of urine concentration after exogenous AVP in both CDI and primary polydipsia |20~, failure of exogenous AVP to increase further maximum urine osmolality in elderly people may be explained by a relatively hypo-osmolal renal medullary interstitium (a consequence of both renal ageing and polyuria itself |30, 32~). It does not necessarily confirm either a renal resistance to AVP nor does it necessarily imply that the senescent loss of concentrating ability cannot be due to inadequate plasma AVP concentrations.

Although it is possible that elderly subjects with abnormalities of osmoregulation and AVP secretion are over-represented in the study group of this research, the apparent high frequency of occurrence them of a deficiency of osmotically stimulated AVP, together with the well documented decrease in thirst perception with age and the difficulty many elderly people may have with the practicalities of toileting, has very important implications for the maintenance of fluid balance and further research in normal elderly subjects is called for.


We wish to thank Lilly Research Centre Ltd (Eli Lilly & Co), Windlesham, Surrey, for supporting this work, and Age Concern and all the volunteers who gave their time so generously.


1. Leaf A. Dehydration in the elderly. New EnglJMed 1984;311:791-2.

2. Phillips PA, Ledingham JGG, Rolls B, Crowe MJ, Wollner L. Reduced thirst perception after water deprivation in healthy elderly man. N EnglJMed 1984;311:7 5 3-9.

3. Helderman JH, Vestal RE, Rowe JW, Tobin JD, Andres R, Robertson GL. The response of arginine vasopressin to intravenous ethanol and hypertonic saline in man: the impact of aging.JGerontol 1978;33:39-47.

4. McLean KA, O'Neill PA, Davies I, Catania J. Changes in the response to a saline load with age. Clin Sci 1991;81:6P.

5. Duggan J, Catania J, O'Neill P, Davies I. Response to dehydration in long term care elderly patients. Clin Sci 1992;82:27P.

6. Crowe MJ, Forsling ML, Rolls BJ, Phillips PA, Ledingham JGG, Smith RF. Altered water excretion in healthy elderly men. Age Ageing 1987;16:285-93.

7. Rooke PA, Baylis PH. A new sensitive radioimmunoassay for arginine vasopressin. J Immunoassay 1982;3:114-31.

8. Lindeman RD, Lee TD, Yiengst MJ, Shock NW. Influences of age, renal disease, hypertension, diuretics and calcium on the antidiuretic responses to suboptimal infusions of vasopressin.JLab Clin Med 1966;68:206-23.

9. Macias Nunez JF, Garcia Iglesias C, Bondia Roman A, et al. Renal handling of sodium in old people: a functional study. Age Ageing 1978;7:178-81.

10. Wienstock Brown WW, Spry L, Davis BB. Alterations in renal homeostasis with aging. In: Davis BB, Wood WG, eds. Homeostatic function and aging. New York: Raven Press, 1985;23-29.

11. Sunderam SG, Mankikar GD. Hyponatraemia in the elderly. Age Ageing 1983;12:77-80.

12. Anderson RJ, Chung HM, Kluge R, Schrier R. Hyponatremia: A prospective analysis of its epidemiology and the pathogenetic role of vasopressin. Ann Intern Med 1985;102:164-8.

13. Chiodera P, Capretti L, Marchesi M, et al. Abnormal arginine vasopressin response to cigarette smoking and metoclopramide (but not to insulin-induced hypoglycaema) in elderly subjects.JGerontol 1991;46:M6-10.

14. Rowe JW, Minaker KL, Sparrow D, Robertson GL. Age-related failure of volume-pressure mediated vasopressin release.JClin Endocrinol Metab 1982;54:661-4.

15. Frolkis VV, Golovchenko SF, Medved VI, Frolkis RA. Vasopressin and cardiovascular system in aging. Gerontology 1982;28:290-302.

16. Kirkland J, Lye M, Goddard C, Vargas E, Davies I. Plasma arginine vasopressin in dehydrated elderly patients. Clin Endocrinol 1984;20:451-6.

17. Li CH, Hsieh SM, Nagai I. The response of plasma arginine vasopressin to 14 h water deprivation in the elderly. Acta Endocrinol 1984;105:314-17.

18. Thompson CJ. Polyuric states in man: water and salt homeostasis in health and disease. Clin Endocrinol Metab 1989;3:473-98.

19. Miller M, Dalakos T, Moses AM, Fellerman H, Streeten DHP. Recognition of partial defects in antidiuretic hormone secretion. Ann Intern Med 1970;73:721-9.

20. Zerbe RL, Robertson GL. A comparison of plasma vasopressin measurements with a standard indirect test in the differential diagnosis of polyuria. N EnglJMed 1981;305:1539-46.

21. Block LH, Furrer J, Locher RA, Siegenthaler W, Vetter W. Changes in tissue sensitivity to vasopressin in hereditary hypothalamic diabetes insipidus. Klin Wochenschr 1981;59:831-6.

22. Bevilacqua M, Norbiato G, Chebat E, et al. Osmotic and nonosmotic control of vasopressin release in the elderly; effect of metaclopramide.JClin Endocrinol Metab 1987;65:1243-7.

23. Miller M. Influence of aging on vasopressin secretion and water regulation. In: Schrier RW, ed. Vasopressin. New York: Raven Press, 1985;249-58.

24. Mukherjee AP, Coni NK, Davison W. Osmoreceptor function among the elderly. Gerontol Clin 1973;15:227-33.

25. Segar WE, Moore WW. The regulation of antidiuretic hormone release in man: effects of change in position and ambient temperature on blood ADH levels.JClin Invest 1968;17:2143-51.

26. Fliers E, Swaab DF, Pool CW, Verwer RWH. The vasopressin and oxytocin neurons in human supraoptic and paraventricular nucleus with aging and in senile dementia. Brain Res 1985;342:45-53.

27. Hoogendijk JE, Fliers E, Swaab DF, Verwer RWH. Activation of vasopressin neurons in the human supraoptic and paraventricular nucleus in senescence and senile dementia.JNeurol Sci 1985;69:291-9.

28. Lindeman RD, Van Buren HC, Raisz LG. Osmolar renal concentrating ability in healthy young men and hospitalised patients without renal disease. N EnglJMed 1960;262:1306-14.

29. Rowe JW. The effect of age on creatinine clearance in man: a cross-sectional and longitudinal study.JGerontol 1976;31:155-63.

30. Rowe JW, Shock NW, De Fronzo RA. The influence of age on the renal response to water deprivation in man. Nephron 1976;17:270-8.

31. Miller JH, Shock NW. Age differences in the renal tubular response to antidiuretic hormone.JGerontol 1953;8:446-51.

32. De Wardener HE, Herxheimer A. The effect of high water intake on the kidney's ability to concentrate urine in man. J Physiol 1957; 139:42-52.
COPYRIGHT 1993 Oxford University Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Faull, Christina M.; Holmes, Catherine; Baylis, Peter H.
Publication:Age and Ageing
Date:Mar 1, 1993
Previous Article:Expressed wish to die and mortality in older people: a community replication.
Next Article:Is there a right age for case finding in elderly people?

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters