Evidence that the Antidiuretic Substance in the Plasma of Children with Nephrogenic Diabetes Insipidus is Antidiuretic Hormone

Title: Evidence that the Antidiuretic Substance in the Plasma of Children with Nephrogenic Diabetes Insipidus is Antidiuretic Hormone
Authors: Holliday, Malcolm A.; Burstin, Charles; Harrah, Jean
Publisher: Pediatrics
Date Published: January 01, 1963
Reference Number: 291

The following article is presented with the permission of the publisher for educational use within the NDI community. No part of this article may be reproduced in any way without permission in writing from the publisher.

The polyuria in the "complete form" of nephrogenic diabetes insipidus is presumed to be due to an inability of the kidney to react to either exogenous or endogenous antidiuretic hormone (ADH).1 In several instances, antidiuretic substances have been demonstrated in the plasma2,3 and urine4,5 from children with nephrogenic diabetes insipidus. The methods employed, however, did not permit the identification of the antidiuretic substances.6 In the present study antidiuretic activity was demonstrated in the plasma of children with nephrogenic diabetes insipidus which conformed in many respects to the characteristics of ADH.

FIG. 1. — Graph of results of an assay from study #3. The number beneath each response is the absolute area of depression of urine flow from the control. (a) refers to jugular vein plasma after water deprivation, (b) to femoral vein plasma taken at the same time, and (c) to jugular vein plasma sample obtained after the child drank water. The standards indicated are the amounts delivered in 0.5 ml.
The concentration of antidiuretic activity was determined by the rat assay procedure described by Kalant, Hunter, and Ogilvie,7 with the exception that material for assay was injected into the assay rat via the descending aorta rather than via the jugular vein. The intra-arterial injection increased the sensitivity of the assay so that in some instances as little as 1.5 micropressor units (µU) of posterior pituitary powder° (PPE) could be detected. Several days prior to the assay, a catheter was inserted into the aorta by the technique described by Popovic and Popovic,8 and plastic tubes were tied into the bladder and stomach. Each rat received one or more standard solutions containing from 1.5 to 25 µU PPE in 0.5 ml saline solution to assess its sensitivity. The method of plotting a response is illustrated in Figure 1.

(°The standard used throughout was a Parke Davis dried pituitary powder (Lot #25326) assayed to have 80 pressor units/mg before, and 64 pressor units/mg after, this study. We are indebted to Dr. Frederick Armstrong of Parke Davis and Company for supplying the PPE and for the pressor assays. USP XVI Pituitary Powder prepared as directed exhibited antidiuretic responses in this assay which corresponded to the pituitary powder of Parke Davis when compared in pressor units. A preparation of purified arginine vasopressin assayed to have 295 pressor units/mg supplied by Dr. I. L. Schwartz also gave corresponding antidiuretic responses.)

Blood for assay was collected in a plastic syringe containing heparin and transferred immediately to a plastic centrifuge tube in an ace bath [sic]. The tube was centrifuged at 4°C within 30 minutes and the plasma removed to a second plastic tube which was kept in an ice bath when assayed the same day, or stored at –20°C. Internal jugular vein punctures were performed according to standard pediatric techniques.

The native or diluted plasma was injected in a total volume of 0.5 ml. When possible the response to the plasma was determined both before and after the injection of a standard solution. In some instances the same plasma was assayed in two rats. In most instances the plasma was tested again after incubation with 0.01 M thioglycollate for 1 hour9 (Table I). Ultrafiltrates of plasma were prepared by centrifugation of the plasma suspended in Nojax tubing #8/32 at 18,000 G for 1 hour at 4°C; approximately 0.5 ml of ultrafiltrate was obtained from 1 ml of plasma.



Condition of
Nature of
Plasma Sample
(mOs/kg H20)


. .
. .
no resp.

6-12†; <15§
ThirstedJugular, UF
< 15†
< 3
< 3†

> 6
Thirst & mannitolJugular
<15-30†; <24§
Jugular, UF
Jugular, res.

After vasopressinAntecubital
. .
>30-60†; >24§
Assay of injected
10 µU vasopressin

* Thioglycollate abolished only half the antidiuretic response.
† Responses obtained twice in the same assay animal.
Plasma not tested for thioglycollate inhibition.
§ Results of an assay of the same plasma in a second assay rat.


Four male children with diabetes insipidus since birth, and a family history of this condition, were studied. None of these children exhibited any antidiuretic response to the injection of vasopressin (Pitressin) of demonstrated potency. The children were not given any water overnight and a loss of 5-7% body weight occurred in each instance. In the morning a blood sample was drawn from the internal jugular vein of the child; in addition, blood was taken at the same time from an antecubital vein in one instance, and from the femoral vein in another. From two children, a second blood was drawn from the internal jugular vein in from 30 to 40 minutes after they drank 300 to 400 ml water. Instead of water, one child was given an infusion of 5% mannitol at a rate of 2 ml/min which resulted in an increase in urine flow comparable to that which ensued after the ingestion of water. There was, however, little change in plasma osmolality. Sixty minutes later a second internal jugular vein blood sample was obtained. Urine flow was followed at 20-minute intervals to determine the effect of water ingestion or mannitol infusion in the respective studies. Plasma osmolality was measured in most of the plasma samples assayed.

On [sic] child was given ample fluids orally and in, addition [sic], he was given an intravenous infusion containing 2½% glucose and 2½% fructose at a rate of 2 ml/min. When his urine flow became stable, blood was drawn from the antecubital vein and 100 mU vasopressin was given intravenously. When the child became pale and developed a slight elevation of blood pressure, a second blood was drawn from the opposite antecubital vein. His urine flow and osmolality were determined from samples collected at 20-minute periods.


The antidiuretic activities of the plasmas, expressed as µU per milliliter, are cited in Table I in relation to the site of the venipuncture and to the state of hydration of the child. Because it was not possible to use a standard 2 X 2 assay for each plasma, the responses are expressed as being between responses of two standard concentrations of PPE when possible, or greater than or less than a standard when the standards did not bracket the plasma response.
FIG. 2. — A plot of the urine osmolality and urine flow at 20-minute intervals in study #1. The assay results are noted. The water ingestion was complete at the site of the arrows. The vasopressin (Pitressin) injection had no effect.

In study #1, the internal jugular vein plasma taken during the hydropenic state produced an antidiuretic response in the assay animal that was much higher than the response to the standard (12 µU/ml) tested. The internal jugular vein plasma obtained after the child drank water elicited a response that, on the other hand, was lower than the response to the standard. The rate of urine flow increased and both urine and plasma osmolality decreased. The injection of vasopressin during the diuresis produced no antidiuretic effect (Fig. 2).

In study #2, the internal jugular vein plasma produced an antidiuretic response which was greater than the single standard in two separate tests with the same assay animals. No antidiuretic response was noted when the antecubital vein plasma was injected.

In study #3, the internal jugular vein plasma, obtained when the child was hydropenic, was assayed using two rats. One rat responded to 0.5 ml plasma with two nearly equal responses between 3-6 µU, or 6-12 µU/ml plasma; the second rat was given 0.2 ml plasma and the response was slightly less than that produced by 3 µU, or <15 µU/ml. An ultrafiltrate of this plasma (0.2 ml) induced the responses which were very slightly less than 3 µU or <15 µU/ml on two separate tests in the one assay animal. The femoral vein plasma, obtained at the same time as the first jugular vein plasma, was tested once in the same assay animal as the jugular vein plasma; the response was less than 1.5 µU in 0.5 ml of plasma or <3 µU/ml. The internal jugular vein plasma obtained after the child drank water was tested twice in this assay animal; responses on both tests were distinct but were less than that to 1.5 µU, or <3 µU/ml. The plasma osmolality initially was 340 mOs/l in this child; after he drank water it decreased to 320 mOs/l, but not to normal levels, and plasma antidiuretic activity also decreased.

In study #4, the child was given an infusion of isotonic mannitol; his urine flow increased to that which occurs after the ingestion of water, but the osmolality of the urine did not decrease and the plasma osmolality changed very little. A comparison of the plasma antidiuretic activity before and after the infusion of mannitol was not possible because the initial plasma sample was small, but the post-infusion plasma had a high level of antidiuretic activity. The antidiuretic activity in an ultrafiltrate and in the residue of the plasma sample from the post-infusion blood were in the same range as that of the parent plasma.

In study #5, the antidiuretic activity in the child's plasma after he was given vasopressin was higher than that observed in this child (cf #1) and in the other children after they had been deprived of water. This concentration, however, had no effect on urine flow or urine osmolality although it exerted a pressor and vasoconstrictor action. The plasma obtained prior to the injection of vasopressin exhibited a small antidiuretic response in one test and none in a second test. If the antidiuretic response in the post-injection plasma could be attributed to the injected vasopressin, it would be possible to estimate an approximate volume of distribution for the injected vasopressin. The blood was drawn within 2 minutes of the injection of vasopressin. Allowing the half-life of vasopressin to be 4-11 minutes,10 it is probable that no more than 25% was inactivated. Since the plasma concentration was 30-60 µU/ml 2 minutes post-injection, the zero time concentration would lie between 40-80 µU/ml. This range, divided into the total dose of 100 mU, i.e. 100,000 µU, describes a volume of distribution equal to 1,250-2,500 ml. The plasma volume for this child who weighed 10 kg should be less than 600 ml. Accordingly, the volume of distribution of injected vasopressin in this child was probably greater than the plasma volume.


The antidiuretic activity of the plasma of the four children with nephrogenic diabetes insipidus appears to be due to antidiuretic hormone because (a) the concentration in plasma abruptly decreases when the child replenishes water deficits, (b) the concentration in the internal jugular vein plasma coming from the brain is higher than it is in plasma of the blood draining the extremities, and (c) the antidiuretic activity is destroyed after incubating the plasma with 0.01 M thioglycollate at pH 7.4 for 1 hour. The concentrations observed are in the range of those reported in a few patients where antidiuretic activity presumed to be due to ADH has been demonstrated by suitably specific techniques,11,12 and they are in the range anticipated from calculations based on the half-life of the hormone in the circulation.

It was not possible to obtain comparable degrees of dehydration in other children who had a prompt reduction in urine flow with water deprivation. Three children stimulated to secrete vasopressin by the stress of operation or by injecting 3% salt did develop comparable antidiuretic responses when their jugular vein plasma was tested. We lack sufficient data in these normal subjects to compare with the data reported herein on the ultrafilterability of endogenous ADH, Bocanegra and Lauson,13 however, have shown that endogenous ADH in the blood of normal dogs is ultrafilterable. The probability that the volume of distribution of the injected vasopressin is greater than the plasma volume suggests that the vasopressin is ultrafilterable from plasma.

The increase in urine flow which occurred when the children replenished water deficits raises an interesting point. It can be seen from Figure 2 that the increase is due primarily to an increase in the excretion of water, and that solute excretion increases very little. The increase in water excretion occurred in association with a decrease in plasma antidiuretic activity. However, when vasopressin was injected during a water diuresis it had no effect on urine flow, although the resulting antidiuretic activity of the plasma was higher after the injection of vasopressin than it was after water deprivation. If the antidiuretic activity is due to ADH this would suggest that endogenous ADH is not the mechanism which effects a reduction in urine flow with dehydration. Conversely, its suppression when water is ingested is not the mechanism whereby urine flow increases. These findings support the view that ADH secretion is normal but that the concentrations observed are ineffective in inducing an antidiuretic response in children with complete nephrogenic diabetes insipidus.

The antidiuretic activity in the plasma of four children with nephrogenic diabetes insipidus was measured by a rat assay technique. The evidence presented to indicate that this activity was due to antidiuretic hormone (ADH) was as follows: (a) the activity was higher in jugular vein plasma than in femoral or antecubital vein plasma, (b) it was high when the children were thirsted and decreased when they drank water, (c) it was destroyed when the plasma was incubated with thioglycollate, and (d) it was ultrafilterable, and vasopressin (Pitressin), when injected, was distributed as though it was ultrafilterable.

When the children were given vasopressin, there was no change in urine flow or osmolality, but plasma antidiuretic activity was higher than it was when water deprivation led to a reduction in urine flow and an increase in urine osmolality. The inference of these findings is that ADH is secreted normally in children with nephrogenic diabetes insipidus, it is ultrafilterable, but it is not a factor in modifying urine flow in response to dehydration.


  1. Orloff, J., and Burg, M. B.: Vasopressin resistant diabetes insipidus; in The Metabolic Basis of Inherited Disease; edited by J. B. Stanbury, J. B. Wyngaarden, and D. S. Frederickson. New York, McGraw-Hill, 1960, P. 1274.
  2. MacDonald, W. B.: Congenital pitressin resistant diabetes insipidus of renal origin. PEDIATRICS, 15:298, 1955.
  3. Linneweh, F., Buchborn, E., and Dellbruck, B.: Familiarer renaler diabetes insipidus. Klin. Wschr., 35:22, 1957.
  4. Dancis, J., Birmingham, J. R., and Leslie, S. H.: Congenital diabetes insipidus resistant to treatment with pitressin. Amer. J. Dis. Child., 75:316, 1948.
  5. Luder, J., and Burnett, D.: A congenital renal tubular defect. Arch. Dis. Child., 29:44, 1954.
  6. Lauson, H. D.: The problem of estimating the rate of secretion of antidiuretic hormone in man. Amer. J. Med., 11:135, 1951.
  7. Hunter, J., Kalant, H., and Ogilvie, J. C.: A simple method for the assay of small amounts of antidiuretic hormone. Canad. J. Biochem. Physiol., 37:1215, 1959.
  8. Popovic, V., and Popovic, C.: Permanent cannulation of aorta and vena cava in rats and ground squirrels. J. Appl. Physiol., 15:727, 1960.
  9. van Dyke, H. B., et al.: The isolation of a protein from the pars neuralis of the ox pituitary with constant oxytocic, pressor and diuresis inhibiting properties. J. Pharmacol., 74:190, 1942.
  10. Share, L.: Rate of disappearance of arginine vasopressin from circulating blood in the dog. Amer. J. Physiol., 203:1179, 1962.
  11. Lauson, H. D.: Vasopressin and oxytocin in the plasma of man and other mammals; in Hormones in Human Plasma; edited by H. N. Antoniades. Boston, Mass., Little, Brown and Co., 1960, p. 225.
  12. Bisset, G. W., and Lee, I.: Oxytocic and antidiuretic activity in blood from the conscious subject. Lancet, 2:770, 1957.
  13. Bocanegra, M., and Lauson, H. D.: Ultrafilterability of endogenous antidiuretic hormone from plasma of dogs. Amer. J. Physiol., 200:486, 1961.

Presented in part before the American Federation for Clinical Research, Atlantic City, 1962 (Clin. Res. 10:228, 1962).

Supported in part by USPHS Grant #A-2668 and Health Research and Services Foundation Grant #D-26.

Dr. Burstin's work was done as Student Research Fellow; Dr. Holliday is Career Research Investigator, USPHS.

ADDRESS: (M.A.H.) Children's Hospital of the East Bay, Oakland 9, California.

This translation by the NDI Foundation is to assist the lay reader. To provide a clear, accessible interpretation of the original article, we eliminated or simplified some technical detail and complicated scientific language. We concentrated our translation on those aspects of the article dealing directly with NDI. The NDI Foundation thanks the researchers for their work toward understanding and more effectively treating this disorder.
© Copyright NDI Foundation 2007 (JC)

In 1963, researchers presumed that nephrogenic diabetes insipidus (NDI) was the result of the kidneys' inability to respond to the antidiuretic message of either naturally occurring antidiuretic hormone (ADH) or synthetically modified ADH. At that time, antidiuretic substances had been demonstrated in the urine and plasma of children with NDI, but the substance had not been identified.

Holliday, et al. designed a study which revealed that the antidiuretic substance in the plasma of children with NDI is characteristic of ADH. The researchers studied four male children with NDI. The researchers restricted these subjects from all water intake overnight and drew blood samples from their jugular veins the following morning. In one child they also drew blood from the antecubital vein, and in another from the femoral vein. The researchers also drew blood samples from the children after the children were rehydrated. Two children drank water to rehydrate themselves. Another was given an infusion of 5% of a sugar alcohol called mannitol, and another was given water orally along with an infusion of a glucose, fructose and water mixture.

These blood samples were injected into specially prepared rats. Then the rats' urine flow and the ratio of particles to water in their urine was measured. The measurements revealed that level of antidiuretic activity in the children's blood samples. The results indicated that the antidiuretic activity of the children's plasma was due to ADH because:

  1. the concentration of antidiuretic substance was high when the children were dehydrated and decreased when they were rehydrated;
  2. the antidiuretic activity in the internal jugular vein plasma coming from the brain was higher than it was in the plasma drawn from the antecubital and femoral veins (which drain blood from the extremities);
  3. the antidiuretic activity is destroyed after incubating the plasma in thioglycollate, a sulfur compound.
There was no change in the children's urine flow or osmolality when they were injected with ADH. Their plasma antidiuretic activity was higher when they were dehydrated. These findings led the researchers to infer that children with NDI secrete ADH normally, but their kidneys can't respond to it.