The American Journal of Medicine (2006) Vol 119 (7A), S36 –S42

Regulation of Arginine Vasopressin in the Syndrome

of Inappropriate Antidiuresis
Gary L. Robertson, MD
Division of Endocrinology, Department of Medicine, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, USA

ABSTRACT

The syndrome of inappropriate antidiuresis (SIAD) is a disorder of sodium and water balance characterized
by hypotonic hyponatremia and impaired water excretion in the absence of renal insufficiency, adrenal
insufficiency, or any recognized stimulus for the antidiuretic hormone arginine vasopressin (AVP).
Hyponatremia is primarily a result of excessive water retention caused by a combination of excessive
intake and inappropriate antidiuresis. It is sometimes aggravated by a sodium deficiency caused by
decreased intake and/or a secondary natriuresis triggered by and largely corrective of the increase in
extracellular volume. Hence, there is neither edema nor signs of hypovolemia. Inappropriate antidiuresis
is usually due to administration or endogenous production of AVP or another vasopressin receptor agonist
such as desmopressin. Endogenous production can be either ectopic (from a tumor) or eutopic (from the
neurohypophysis). The latter apparently is induced by a wide variety of diseases, drugs, or injuries and is
divisible into 3 different types of abnormal AVP release during hypertonic saline infusion: high, erratic
fluctuations unrelated to increases in plasma sodium (type A); a slow constant “leak” that is also unaffected
by increases in plasma sodium (type B); and rapid progressive increases in plasma AVP that correlate
closely with plasma sodium as it rises toward the normal range (type C or “reset osmostat”). In 5% to 10%
of patients, there is no demonstrable abnormality in the osmoregulation of AVP (type D) and the cause of
inappropriate antidiuresis is unclear. In some children it appears to be due to an activating mutation of the
V2 receptor (V2R). In other patients, it may be due to abnormal control of aquaporin-2 water channels in
renal collecting tubules or production of an antidiuretic principle other than AVP. These different types of
osmoregulatory dysfunction underlying SIAD may result in marked differences in clinical presentation or
response to therapy with fluid restriction, hypertonic saline infusion, or vasopressin antagonists. © 2006
Elsevier Inc. All rights reserved.

KEYWORDS: Antidiuretic hormone; Arginine vasopressin; Hyponatremia; Syndrome of inappropriate antidiuresis

The syndrome of inappropriate antidiuresis (SIAD) is a
disorder of sodium and water balance characterized by hy-
potonic hyponatremia and impaired urinary dilution in the
absence of renal disease or any identifiable nonosmotic
stimulus known to release the antidiuretic hormone arginine
vasopressin (AVP).1 It is a diagnosis of exclusion and must
be distinguished from several other types of hyponatremia
because each has a fundamentally different pathogenesis,
Requests for reprints should be addressed to Gary L. Robertson, MD,
Division of Endocrinology, Department of Medicine, Feinberg School of
Medicine, Northwestern University, 300 East Chicago Avenue, Chicago,
Illinois 60611.
E-mail address: robconsult@comcast.net.
presents different abnormalities in sodium and water bal-
ance, and requires different management (Table 1). This
distinction is based on 2 factors: a clinical assessment of
whether the patient’s extracellular fluid (ECF) volume is
increased (type 1), decreased (type 2), or relatively normal
(type 3, or euvolemic); and, if the latter (euvolemic) type is
present whether the patient has a known or unknown cause
for antidiuresis. Thus, to qualify for the diagnosis of SIAD,
the patient should not have a history or signs indicative of
hypervolemia (for example, generalized edema due to se-
vere congestive heart failure, cirrhosis, or nephrosis), hypo-
volemia (for example, hemorrhage, gastroenteritis, diuretic
abuse, or aldosterone deficiency), renal failure, or adrenal

0002-9343/$ -see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjmed.2006.05.006
Robertson AVP Regulation in SIAD
Table 1 Differential diagnosis of hyponatremia
Type 1 (Hypervolemic) Type 2 (Hypovolemic)
History CHF, cirrhosis, nephrosis Bleeding, diuretic, diarrhea
Edema ⫹ ⫺
Blood pressure Low Low
BUN High High
Urine Na Low Low*
PRA High High
Cortisol Normal Normal‡
BUN ⫽ blood (serum) urea nitrogen; CHF ⫽ congestive heart failure; PRA ⫽ plasma renin activity; SIAD ⫽ syndrome of inappropriate antidiuresis.
*Except in cases of diuretic abuse.
†
In late phase.
‡
Except in Addison disease.
§
Except in cases of nausea.
failure. Technically, nausea and vomiting that occurs before
the onset of hyponatremia also excludes the diagnosis of
SIAD because nausea is an extremely potent stimulus for
AVP release and this effect can be prevented or corrected
completely with antiemetic treatment.2 However, because
nausea and vomiting can also result from hyponatremia, it is
sometimes difficult to determine whether they are the cause
or the consequence of SIAD and, if the latter, whether they
are part of a vicious circle that perpetuates excessive AVP
secretion triggered initially by another stimulus. In either
case, treatment with an antiemetic may be effective in ter-
minating SIAD as well as some of the symptoms that can
result from it.
PATHOPHYSIOLOGY
The fundamental cause of hyponatremia in SIAD is an
absolute increase in body water that results from excessive
fluid intake in a patient who cannot dilute his or her urine
sufficiently to mount a compensatory water diuresis. Total
fluid intake must be greater than the sum of urinary and
insensible water losses because AVP by itself does not
increase body water or directly stimulate a significant in-
crease in sodium excretion. AVP merely retards the rate of
urinary water loss enough to prevent complete elimination
of excess intake. This defect in water excretion is the sine
qua non for hyponatremia, because it rarely results from
polydipsia alone even if it is very severe. There also must be
some impairment in urinary dilution and free water excre-
tion. Neither abnormality is sufficient; both are necessary
for hyponatremia to develop.
The rate of water intake required to produce hyponatre-
mia varies inversely with the abnormality in free water
excretion and vice versa. If intake is extremely high, as it is
in some patients with psychogenic polydipsia, even a slight
impairment in urinary dilution (or solute excretion rate) may
suffice to expand body water and decrease serum sodium.3
Conversely, if urine concentration is fixed at a high level
(e.g., urine osmolarity ⬃1000 mOsmol/L), even a subnor-
S37
Type 3 (Euvolemic)
A SIAD
Drug, nausea, hypocortisol Tumor, stroke, infection
⫺ ⫺
Normal/low Normal
Low Low
High† High†
Low Low
Low§ Normal
mal rate of water intake may be enough to produce an
imbalance. The majority of patients with SIAD fall some-
where between these 2 extremes.
The expansion of body water that occurs in SIAD causes
hyponatremia by 2 mechanisms. The first is simple dilution.
This effect is immediate and lowers serum sodium in pro-
portion to the increase in total body water. Thus, for exam-
ple, if body water is expanded 5% (about 2 L in a typical
70-kg patient), serum sodium decreases by about 7 mEq/L
(7 mmol/L). If body water is further expanded, the addi-
tional reduction that results from dilution can be worsened
by a second mechanism, a net decrease in exchangeable
body sodium. This reduction in body sodium serves to
largely counteract the increase in extracellular volume but at
the expense of aggravating the hyponatremia. It is mainly
caused by an increase in urinary sodium excretion triggered
by a homeostatic response to volume expansion.3 The na-
triuresis is probably mediated, at least in part, by suppres-
sion of plasma renin activity and aldosterone secretion,4 and
an increase in plasma atrial natriuretic peptide.5 In some
patients the sodium deficit may also be caused or aggravated
by a reduction in sodium intake, possibly as a result of the
anorexia or nausea and vomiting that is sometimes induced
by acute hyponatremia.
CLINICAL CAUSES
The causes of the excess fluid intake and impaired urinary
water excretion in patients with SIAD are quite varied and
the basic abnormalities underlying most of them are not
well understood. Excessive fluid intake is sometimes due to
inappropriate thirst and may be caused by the same abnor-
mality responsible for the inappropriate secretion of AVP.
In these patients, hyponatremia usually develops spontane-
ously early in the course of the syndrome. In other patients,
however, the excessive intake appears to be psychogenic or
“iatrogenic”—that is, hyponatremia occurs when a care-
giver prescribes too much intravenous or oral fluid in a
misguided effort to prevent dehydration. The inappropriate
S38
Figure 1 Osmoregulation of plasma arginine vasopressin (AVP) in patients with the syndrome of inappropriate antidiuresis is depicted
for types A, B, C, and D. 1 mEq/L ⫽ 1 mmol/L.
antidiuresis can also be iatrogenic in that it results from
administration of AVP, desmopressin (a V2 selective ana-
logue of AVP), large doses of oxytocin, or certain drugs
(such as serotonin-reuptake inhibitors) that stimulate release
of endogenous AVP in some patients. However, it is usually
a result of the release of endogenous AVP produced ectopi-
cally by a tumor or eutrophically by the neurohypophysis in
response to some unknown stimulus induced by a stroke,
infection, or other illness.
PATHOGENESIS OF THE DEFECT IN URINARY
WATER EXCRETION
As shown in Figure 1, 4 types of inappropriate antidiuresis
have been identified.6 In 3 of these types, the osmoregulation
of plasma AVP is clearly abnormal because neither the hor-
mone nor the urine concentration are fully suppressed as they
should be when plasma sodium falls below the normal range.
In type A, the osmoregulatory defect is characterized by mod-
erate to marked increases in plasma AVP that fluctuate widely
with no discernible relation to the increases in plasma osmo-
larity or sodium produced by hypertonic saline infusion. This
pattern appears to be caused by erratic, unregulated release of
AVP and is found in about 30% of patients with SIAD, in-
cluding, but not limited to, patients with malignancy. The large
fluctuations in plasma AVP usually do not result in large
changes in urine concentration, however, because they tend to
The American Journal of Medicine, Vol 119 (7A), July 2006
occur in a range consistently above that required to produce
maximum antidiuresis. Hence, urine osmolarity tends to be
fixed at the highest level possible given the blunting of max-
imum urinary concentrating capacity that also often occurs in
SIAD.7 In type B, plasma AVP is not as high and does not
change appreciably during hypertonic saline infusion until
plasma osmolarity and sodium rise into the normal range. At
that point, the hormone begins to rise appropriately in direct
relation to the strength of the osmotic stimulus. This pattern
suggests a slow, constant “leak” of AVP and could result from
damage to either the neurohypophysis or the inhibitory com-
ponent of the osmoregulatory mechanism. It also occurs in
about 30% of patients, including some with malignancy. In this
type of SIAD, the level of urine concentration also tends to be
fixed but at a lower level than in type A (Figure 2).
In the third osmoregulatory defect, type C (Figure 2),
plasma AVP and urine osmolarity are usually low or max-
imally suppressed in the basal hyponatremic state. How-
ever, during infusion of hypertonic saline, they begin to
increase “inappropriately” in close correlation with plasma
osmolarity and sodium long before the hyponatremia is
corrected. This pattern appears to be due to downward
resetting of the entire osmoregulatory system similar to that
which occurs in the presence of significant hypovolemia or
hypotension.8 However, as noted above, neither hemody-
namic abnormality is present in SIAD, which suggests that
Robertson AVP Regulation in SIAD
the type C defect might be due to aberrant input from the
peripheral mechanisms and/or neural pathways that mediate
the baroreceptor regulation of AVP secretion.6,8 Like types
A and B, the type C defect occurs in about 30% of patients
with SIAD, including some with malignancy. The latter is
surprising since the osmoregulation and production of AVP
normally occur in distinct, highly differentiated hypotha-
lamic cells, and it seems unlikely that a malignancy would
acquire both capacities via dedifferentiation. A more likely
explanation for occurrence of the type C defect in tumor
patients is that some malignancies cause SIAD not by pro-
ducing AVP ectopically, but rather by stimulating eutopic
(neurohypophysial) secretion via alterations in baroreceptor
regulation or other afferent input. Unlike the type A or type
B defects (Figure 2), the level of plasma AVP and antidi-
uresis in patients with the type C defect varies markedly
with changes in hydration and plasma sodium (Figure 2).
The fourth type of inappropriate antidiuresis, type D
(Figure 2), differs from the others in that it cannot be
attributed to a demonstrable defect in the osmoregulation of
AVP (Figure 1). In these patients, plasma AVP is sup-
pressed to undetectable levels and remains so during hyper-
tonic saline infusion until plasma osmolarity and sodium
increase to within the normal range. At that point, plasma
AVP begins to rise normally in close correlation with fur-
ther increases in plasma osmolarity and sodium. Thus, os-
moregulation of the hormone appears to be completely
normal. However, despite the consistent lack of detectable
AVP in plasma under hyponatremic conditions, patients
with the type D defect fail to dilute their urine or mount a
water diuresis even when challenged with an oral water load
(Figure 2). The reason for this failure is not completely
clear. It is not attributable to a significant reduction in
glomerular filtration or an isolated deficiency of cortisol,
either of which is reported to result in urine concentration
even when AVP is deficient.9,10 However, as reported else-
where in this supplement by Gitelman and colleagues,11 it
can sometimes be caused by a germ cell mutation that
constituently activates the AVP V2 receptor (V2R). The
patients in whom this abnormality was found were brothers
who had SIAD since birth. Whether similar mutations ever
cause sporadic type D SIAD in adults remains to be deter-
mined. However, there probably are other causes of urine
concentration in the absence of AVP, because some older
men with the autosomal dominant form of familial pituitary
diabetes insipidus recover the capacity to concentrate their
urine even though they continue to have a severe deficiency
of AVP.12,13 Potential mechanisms include the production
of an antidiuretic compound other than immunoactive AVP
or a postreceptor defect in trafficking of the aquaporin-2
water channels that mediate antidiuresis.14
The type of defect present in SIAD appears to remain
relatively constant over time, at least in patients with the
idiopathic chronic or recurrent form of the syndrome. In 1 of
our patients with chronic SIADH, repeat testing over a
15-year period revealed persistence of the same, identical
S39
type B defect. Another patient with agenesis of the corpus
callosum and recurrent episodes of hypothermia, hypona-
tremia, polyuria, and polydipsia occurring 2 to 3 times a
year for almost 45 years revealed a type C defect (reset
osmostat during acute episodes). The reproducibility of dif-
ferent types of osmoregulatory defects suggests that they
reflect fundamentally different pathogenic mechanisms for
SIAD. However, there does not seem to be any correlation
between the type of osmoregulatory defect and the specific
disease or injury that appears to have caused it. This sug-
gests that diseases most often associated with SIAD (e.g.,
cancer, pneumonia, stroke) can produce the syndrome via
fundamentally different mechanisms.
VARIATIONS IN THE CLINICAL
MANIFESTATIONS OF THE SYNDROME OF
INAPPROPRIATE ANTIDIURESIS
The clinical behavior of SIAD can differ significantly de-
pending on the type of osmoregulatory defect present (Fig-
ure 2). Patients with type A, B, or D always exhibit some
impairment in urinary dilution in the presence of hypona-
tremia; the only clinically significant difference between
them is the magnitude of the diluting defect. However, this
can also vary greatly from patient to patient owing not only
to differences in plasma AVP but probably also to variable
downregulation of the renal concentrating response to the
hormone.7 Urine osmolarity usually does not exceed 600
mOsmol/L in SIADH even when plasma AVP levels are
extremely high. However, even at this submaximum level,
the rate of urine output is not sufficient to permit escape
from the excessive water retention unless insensible loss is
abnormally high and intake is very low. Types A and B also
differ from type D in terms of whether the defect in urinary
dilution is due to inappropriate secretion of AVP, but this
difference does not appear to alter the clinical manifesta-
tions of the syndrome. However, in patients with the type C
defect (reset osmostat), plasma AVP and urine osmolarity
may be maximally suppressed if fluid intake is great enough
to depress plasma osmolarity and sodium to the new os-
motic “set-point.” In this situation, the patient may have a
history of polydipsia as well as signs of a maximum water
diuresis (i.e., urine osmolarity ⬍80 mOsmol/L and a rate of
urine output ⬎10 mL/kg body weight per hour). This di-
uresis limits the decrease in plasma osmolarity and sodium
to the level of the new osmoregulatory set-point no matter
how great the rate of water intake may be. Its presence may
also lead to an erroneous conclusion that the hyponatremia
is due solely to severe polydipsia. However, this miscon-
ception usually can be dispelled simply by monitoring
changes in urine osmolarity (and plasma AVP, if desired) as
the hyponatremia is gradually corrected by fluid restriction
or hypertonic saline infusion. This procedure will almost
always show an inappropriate increase in urine osmolarity
and plasma AVP before serum sodium returns to normal.
The combination of hyponatremia and a maximum water
diuresis also can be observed during spontaneous recovery
S40
Figure 2 Effect of an oral water load on plasma arginine vasopressin (PAVP), urine osmolarity (UOsm) (expressed as milliosmoles per
liter), urine flow (urine volume [UVol] in milliliters per 30 minutes), and plasma (serum) sodium (PNa) in patients with osmoregulatory defect
types A, B, C, and D. 1 mEq/L ⫽ 1 mmol/L.
from acute SIAD. In that case, however, increased urinary
concentration does not begin until the serum sodium returns
to the normal range.
CLINICAL VARIATIONS IN THE RESPONSE TO
THERAPY
The treatment for all forms of endogenous SIAD is re-
striction of fluid intake to less than the rate of urinary
water excretion. If the symptoms of hyponatremia are
severe and a quicker recovery is desired, fluid restriction
can be combined cautiously with either infusion of hy-
pertonic (3%) saline or administration of a V2R antago-
nist. The response to these treatments may vary depend-
ing on the underlying type of antidiuretic defect. In
patients with a type A, B, or D defect, raising plasma
osmolality and sodium by fluid restriction alone or in
combination with hypertonic saline infusion usually does
not significantly alter plasma AVP, urine osmolarity, or
urine flow. In these patients, therefore, it is relatively
easy to predict and control the rate of rise in serum
sodium unless a solute or water diuresis develops due to
a sodium load or spontaneous remission of the underlying
The American Journal of Medicine, Vol 119 (7A), July 2006
defect. The response to treatment with a competitive V2R
antagonist may be less predictable because, as a compet-
itive antagonist, its effect depends much more on changes
in endogenous AVP and it might not work at all in the
relatively small number of patients with the type D de-
fect. In patients with the type C defect (reset osmostat),
the response to therapy is likely to be even less predict-
able. In these patients, any increase in plasma osmolarity/
sodium stimulates a large increase in plasma AVP (Fig-
ure 1). The further reduction in urine output that results
is likely to be small and of little consequence relative to
other influences, such as insensible loss or solute excre-
tion rate, if the therapy is fluid restriction alone or with
hypertonic saline infusion. However, if the treatment is a
competitive V2 antagonist, the rise in plasma AVP that
results from the initial water diuresis may quickly over-
come blockade of the V2R and result in a rapid reduction
in urine output and marked slowing of the rise in serum
sodium even though blood levels of the antagonist may
not yet have declined significantly. This effect would
appear as a shorter duration of action than in patients
with the other types of antidiuretic defect and could be
Robertson AVP Regulation in SIAD
Figure 2
prevented or overcome, if desired, simply by increasing
the dose of the antagonist. Unfortunately, at present it is
not easy to predict the type of antidiuretic defect and
response to expect in a particular patient. Thus the only
solution to the problem remains continuous close moni-
toring of water balance and serum sodium throughout the
treatment period.
SUMMARY
SIAD is a heterogenous disorder characterized by hypo-
natremia and impaired urinary dilution in the absence of
any recognized nonosmotic stimulation of the antidiuretic
hormone, AVP. Hyponatremia results from an increase in
body water caused by a combination of excessive fluid
intake and impaired urinary excretion of solute-free wa-
ter. It may be aggravated by an increase in urinary so-
dium excretion triggered by volume expansion and/or
decrease in sodium intake resulting from hyponatremia-
induced anorexia. The causes of the excessive fluid in-
take and impaired urinary excretion of water can be
exogenous or endogenous. The endogenous forms of in-
appropriate antidiuresis can be subdivided into 4 types
based on the response of plasma AVP to osmotic stimu-
S41
(continued).
lation. Three of four types of antidiuresis indicate differ-
ent abnormalities in the osmoregulation of AVP secre-
tion: erratic, unregulated release (type A); a slow,
constant “leak” (type B); and downward resetting of the
osmostat (type C). The fourth type of antidiuresis (type
D), the osmoregulation of AVP, appears to be normal and
inappropriate antidiuresis probably resulting from a de-
fect in the renal collecting tubules. The type of inappro-
priate antidiuresis in each patient is reproducible with
repeat testing and probably reflects a distinct pathogenic
mechanism. However, it does not seem to correlate with
the associated disease or injury that is thought to cause
SIAD and, apart from ectopic production of AVP by a
tumor, little is known about the basic defect that causes
the various types of inappropriate antidiuresis. The clin-
ical behavior and response to therapy of patients with
various types of SIAD can also differ significantly de-
pending on the type of osmoregulatory defect present.
This is particularly true for patients with type C defect
(reset osmostat) because they may respond less well to
fluid restriction, hypertonic saline infusion, or V2 recep-
tor antagonists because AVP secretion increases as the
hyponatremia is corrected.
S42
References
1. Bartter FC, Schwartz WB. The syndrome of inappropriate secretion of
antidiuretic hormone. Am J Med. 1967;42:790 – 806.
2. Rowe JW, Shelton RL, Helderman JH, Vestal RE, Robertson GL.
Influence of the emetic reflex on vasopressin release in man. Kidney
Int. 1979;16:729 –735.
3. Leaf A, Bartter FC, Santos RF, Wrong O. Evidence in man that urinary
electrolyte loss induced by pitressin is a function of water retention.
J Clin Invest. 1953;32:868 – 878.
4. Fichman MP, Michelakis AM, Horton R. Regulation of aldosterone
secretion in the syndrome of inappropriate antidiuretic hormone se-
cretion (SIADH). J Clin Endocrinol Metab. 1974;39:136 –144.
5. Manoogian C, Pandian M, Ehrlich L, Fisher D, Horton R. Plasma atrial
natriuretic hormone levels in patients with the syndrome of inappro-
priate antidiuretic hormone secretion. J Clin Endocrinol Metab. 1988;
67:571–575.
6. Robertson GL. Antidiuretic hormone: normal and disordered function.
Endocrinol Metab Clin North Am. 2001;30:671– 694.
7. Tian Y, Sandberg K, Murase T, Baker EA, Speth RC, Verbalis JG.
Vasopressin V2 receptor binding is down-regulated during renal es-
cape from vasopressin-induced antidiuresis. Endocrinology. 2000;141:
307–314.
The American Journal of Medicine, Vol 119 (7A), July 2006
8. Robertson GL. Physiology of ADH secretion. Kidney Int Suppl. 1987;
32:S20 –S26.
9. Berliner RW, Bennett CM. Concentration of urine in the mammalian
kidney. Am J Med. 1967;42:777–789.
10. Iwasaki Y, Kondo K, Hasegawa H, Oiso Y. Osmoregulation of plasma
vasopressin in three cases with adrenal insufficiency of diverse etiol-
ogies. Horm Res. 1997;47:38 – 44.
11. Gitelman SE, Feldman BJ, Rosenthal SM. Nephrogenic syndrome of
inappropriate antidiuresis: a novel disorder in water balance in pedi-
atric patients. Am J Med. 2006;119(suppl 7A):S54 –S58.
12. Robertson GL, McLeod JF, Zerbe RL, Baylis PH, Kovacs L, Rittig S.
Vasopressin function inheritable forms of diabetes insipidus. In: Gross
P, Richter D, Robertson GL, eds. Vasopressin: International Vaso-
pressin Conference, May 23–27, 1993, Berlin, Germany. Paris: John
Libbey Eurotext; 1993: 503.
13. Robertson GL. The use of vasopressin assays in vasopressin physiol-
ogy and pathophysiology. Semin Nephrol. 1994;14:368 –383.
14. Nielsen S, Marples D, Frokiaer J, Knepper M, Agre P. The aqua-
porin family of water channels in kidney: an update on physiology
and pathophysiology of aquaporin-2. Kidney Int. 1996;49:1718 –
1723.