Diabetes insipidus

Abstract

Diabetes insipidus (DI) is a disorder characterized by excretion of large amounts of hypotonic urine. Central DI results from a deficiency of the hormone arginine vasopressin (AVP) in the pituitary gland or the hypothalamus, whereas nephrogenic DI results from resistance to AVP in the kidneys. Central and nephrogenic DI are usually acquired, but genetic causes must be evaluated, especially if symptoms occur in early childhood. Central or nephrogenic DI must be differentiated from primary polydipsia, which involves excessive intake of large amounts of water despite normal AVP secretion and action. Primary polydipsia is most common in psychiatric patients and health enthusiasts but the polydipsia in a small subgroup of patients seems to be due to an abnormally low thirst threshold, a condition termed dipsogenic DI. Distinguishing between the different types of DI can be challenging and is done either by a water deprivation test or by hypertonic saline stimulation together with copeptin (or AVP) measurement. Furthermore, a detailed medical history, physical examination and imaging studies are needed to ensure an accurate DI diagnosis. Treatment of DI or primary polydipsia depends on the underlying aetiology and differs in central DI, nephrogenic DI and primary polydipsia.

Full text

Diabetes insipidus (DI) is a form of polyuria–polydipsia
syndrome and is characterized by hypotonic polyuria
(excessive urination; >50 ml/kg body weight/24 h) and
polydipsia (excessive drinking; >3 l/day)1. After exclu-
sion of disorders of osmotic diuresis (such as uncon-
trolled diabetes mellitus), the differential diagnosis
of DI involves distinguishing between primary forms
(ofcentral or renal origin) and secondary forms (result-
ing from primary polydipsia) of polyuria. A third, rare
form of DI termed gestational DI can occur during
pregnancy. Central DI (also known as hypothalamic
or neurogenic DI) results from inadequate secretion
and usually deficient synthesis of arginine vasopressin
(AVP) in the hypothalamic–neurohypophyseal system
in response to osmotic stimulation (FIG.1). Central DI
is most often an acquired disorder that is caused by dis-
ruption of the neurohypophysis (specifically, damage to
the AVP- producing magnocellular neurons), whereas
hereditary forms are less common and are caused by
mutations in AVP 2 (located on the short arm of chromo-
some 20 (20p13)). Nephrogenic DI is the result of an
inadequate response of the kidneys to AVP, either due
to mutations in AVP receptor 2 (AVPR2) or aquaporin2
(AQP2)3 (hereditary nephrogenic DI) or as an adverse
effect of various drugs, most commonly lithium, or due
to electrolyte disorders, such as hypercalcaemia or hypo-
kalaemia (acquired nephrogenic DI). Primary polydipsia
is characterized by excessive fluid intake that leads to
polyuria, despite intact AVP secretion and an appropri-
ate antidiuretic renal response. Gestational DI results
from the enzymatic breakdown of endogenous AVP by
increased placental vasopressinase levels in pregnancy.
Regardless of the aetiology, all four forms of polyuria–
polydipsia syndrome result in a water diuresis due to an
inability to maximally concentrate urine. Distinguishing
between the types of DI is important, astreatment strat-
egies differ and application of the wrong treatment can
be dangerous4. However, DI is often difficult to diagnose
reliably and accurately5, especially in patients with pri-
mary polydipsia or partial, mild forms of central and
nephrogenic DI1,6. In this Primer, we describe the differ-
ent types of DI, their pathophysiology, the methods for
differentiating between them and therapies for optimal
management of each type of DI. We also discuss possi-
bilities for prevention and available data about quality of
life (QOL) in patients with DI.
Epidemiology
Prevalence
DI is a rare disease with a prevalence of ~1 in 25,000 indi-
viduals7 and data on geographical differences in preva-
lence are limited. The disorder can manifest at any age,
and the prevalence is similar among males and females.
The age at presentation depends markedly on the aetio-
logy8,9, with hereditary forms manifesting early in life,
whereas acquired forms manifest after early childhood.
Although precise prevalence data are not available,
acquired forms of DI are much more common than famil-
ial forms. Fewer than 10% of cases of renal and central DI
are hereditary. In a study involving 79 paediatric patients,
Diabetes insipidus
MirjamChrist- Crain1,2*, DanielG.Bichet3,4, WiebkeK.Fenske5, MorrisB.Goldman6,
SorenRittig7, JosephG.Verbalis8 and AlanS.Verkman9,10
Abstract | Diabetes insipidus (DI) is a disorder characterized by excretion of large amounts of
hypotonic urine. Central DI results from a deficiency of the hormone arginine vasopressin (AVP)
inthe pituitary gland or the hypothalamus, whereas nephrogenic DI results from resistance to
AVP in the kidneys. Central and nephrogenic DI are usually acquired, but genetic causes must be
evaluated, especially if symptoms occur in early childhood. Central or nephrogenic DI must
bedifferentiated from primary polydipsia, which involves excessive intake of large amounts of
water despite normal AVP secretion and action. Primary polydipsia is most common in psychiatric
patients and health enthusiasts but the polydipsia in a small subgroup of patients seems to be
dueto an abnormally low thirst threshold, a condition termed dipsogenic DI. Distinguishing
between the different types of DI can be challenging and is done either by a water deprivation
test or by hypertonic saline stimulation together with copeptin (or AVP) measurement.
Furthermore, a detailed medical history , physical examination and imaging studies are needed
toensure an accurate DI diagnosis. Treatment of DI or primary polydipsia depends on the
underlying aetiology and differs in central DI, nephrogenic DI and primary polydipsia.
*e- mail: mirjam.christ- crain@
unibas.ch
https://doi.org/10.1038/
s41572-019-0103-2
1
PRIMeR
NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
0123456789();

DI was found to be familial in only 6% of the patients
and the remaining 94% had various types of acquired DI9.
Central DI is the most common type of DI. In heredi-
tary central DI, the predominant inheritance pattern is
autosomal dominant due to mutations in AVP , although
rarely it can be due to the autosomal recessive disorder
Wolfram syndrome, which is caused by mutations in
WFS1 (encoding wolframin). Only few population-
based data on the incidence of congenital nephrogenic
DI exist, but estimates can be made. X- linked nephro-
genic DI due to loss- of-function mutations in AVPR2
accounts for ~90% of hereditary nephrogenic DI cases
and was reported to occur with a frequency of ~8.8 cases
per million male live births in the general population of
Quebec10, which might be representative of the general
population worldwide3. Autosomal recessive and auto-
somal dominant nephrogenic DI due to loss- of-function
mutations in AQP2 accounts for the remaining ~10%
of hereditary nephrogenic DI cases3,7. The prevalence of
acquired nephrogenic DI in patients receiving long-term
lithium treatment is 30%.
Primary polydipsia is common in patients with
neurodevelopmental disorders (such as autism and
intellectual disability) or psychotic disorders (such as
schizophrenia, schizoaffective disorder, bipolar dis-
order and psychotic depression), particularly chronic
schizophrenia, in which primary polydipsia occurs in
11–20% of patients11. Many of these patients experience
episodesof hyponatraemia, especially during psychotic
relapses, a syndrome often referred to as psychosis inter-
mittent hyponatraemia–polydipsia (PIP) syndrome12.
Many other individuals with non- psychotic Axis I psy-
chiatric disorders also have primary polydipsia, a form
that is often termed compulsive water drinking (CWD;
also known as psychogenic polydipsia)11. These patients
rarely if ever become hyponatraemic in the absence of
other factors (such as treatment with thiazide diuretics).
Outside the psychiatric setting, the prevalence of CWD
is increasing in the general population (and seems to
be more prevalent in women) owing to the increasing
popularity of lifestyle programmes and the current
view that fluid intake is inadequate and that drinking
water is healthy and improves cognition in children and
adults13,14. The extent to which these patients overlap
with those with dipsogenic DI (that is, DI owing to an
abnormally low thirst threshold) is unknown.
DI during pregnancy was first reported in 1942 and
occurs in ~1 in 30,000 pregnancies, with the highest
prevalence in multiparous women. Gestational DI typ-
ically occurs at the end of the second or early third tri-
mester (the peak of placental vasopressinase production)
and is associated with a higher risk of pre- eclampsia15.
Pre- existing asymptomatic partial central DI can become
symptomatic during pregnancy, owing to the inabil-
ity of the pituitary gland to increase AVP secretion in
response to increased degradation of AVPby placental
vasopressinase16. In these patients, symptoms typically
appear early in pregnancy and recur with every preg-
nancy. The severity of polyuria and polydipsia in patients
with pre-existing central DI may increase in pregnancy.
Risk factors
Central DI. Risk factors for central DI include traumatic
and non- traumatic causes of damage to AVP- producing
magnocellular neurons in the hypothalamus. The most
common risk factor for trauma- induced central DI is
surgical resection of tumours in the sellar and suprasellar
area. Although tumours contained within the sella tur-
cica, such as pituitary macroadenoma, rarely cause DI,
surgical resection of these tumours and subsequent
damage to the axons of the AVP- producing magno-
cellular neurons in the pituitary stalk (also known as
the infundibulum) can result in central DI. The inci-
dence of postoperative DI is substantially higher fol-
lowing resection of large suprasellar tumours, such as
craniopharyngioma (10–25% incidence, depending on
the extent of resection17), compared with resection of
sellar- based tumours, such as pituitary microadenoma
or macroadenoma (5–30% incidence of transient DI but
only 1–4% incidence of permanent DI18,19). Resection of
tumours with a more rostral location has a greater risk
of causing adipsic central DI because of damage to
osmoreceptors in the anterior hypothalamus (FIG.2).
Traumatic brain injury can also lead to central DI — in
particular, deceleration injuries that shear the pituitary
stalk at the level of the diaphragma sellae, leading to a
characteristic triphasic response (TABLE1).
Risk factors for non- traumatic central DI include
genetic mutations (TABLE 1), granulomatous diseases
that infiltrate the hypothalamus (such as sarcoidosis
and Langerhans and non- Langerhans cell histiocytosis),
primary brain tumours that invade or compress the
hypothalamus (such as germinoma, meningioma, cranio-
pharyngioma and lymphoma), secondary tumours
in the pituitary gland or pituitary stalk (usually met-
astatic breast or lung tumours) and lymphocytic
infundibulo neurohypophysitis (LIN) that causes auto-
immune destruction of AVP neurons20. Although the
occurrence of LIN is usually unpredictable, some risk
factors have been identified, including states associ-
ated with activation of autoimmune diseases, such as
the post- partum period, personal or family history of
autoimmune disorders, anterior pituitary hypophysi-
tis, chronic inflammation of parasellar structures (for
example, in hypertrophic pachymeningitis and Tolosa–
Hunt syndrome), IgG4-related systemic diseases (such
Author addresses
1Division of Endocrinology, Diabetes and Metabolism, University Hospital Basel, Basel,
Switzerland.
2Department of Clinical Research, University of Basel, Basel, Switzerland.
3Division of Nephrology, Hôpital du Sacré- Cœur de Montréal, Montréal, Québec, Canada.
4Départements de Médecine, Pharmacologie et Physiologie, Faculté de Médecine,
Université de Montréal, Montréal, Québec, Canada.
5Medical Department III, Endocrinology, Nephrology, Rheumatology, University Hospital
of Leipzig, Leipzig, Germany.
6Department of Psychiatry and Behavioral Sciences, Feinberg School of Medicine,
Northwestern University, Chicago, IL, USA.
7Department of Pediatric and Adolescent Medicine, Aarhus University Hospital, Aarhus,
Denmark.
8Division of Endocrinology and Metabolism, Georgetown University Medical Center,
Washington, DC, USA.
9Department of Medicine, University of California, San Francisco, CA, USA.
10Department of Physiology, University of California, San Francisco, CA, USA.
2
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();

as autoimmune pancreatitis and lymphoproliferation)21
and the presence of anti- rabphilin 3A antibodies22,23.
Nephrogenic DI. Apart from genetic risk factors, the
most common risk factors for nephrogenic DI are
lithium treatment, hypercalcaemia and hypokalaemia.
Lithium is one of the most common medications for
treating and preventing relapses of bipolar disorders
and psychotic depression. In a study of 873 patients with
bipolar disorders who were treated with lithium, of the
54% of patients who discontinued lithium treatment,
9% did so because of polyuria, polydipsia or DI, usually
within 5 years. Discontinuation of lithium treatment
because of increasing serum creatinine levels (indicative
of renal dysfunction) occurred after 17 years on average,
although this occurred after 30 years in four patients24.
Primary polydipsia. Patients with primary polydipsia
who have psychotic disorders rarely complain of thirst but
instead provide delusional explanations for their exces-
sive drinking or state that drinking reduces their anxi-
ety or makes them feel better25,26. Patients with primary
polydipsia who have schizophrenia, and the first- degree
relatives of these patients, have a higher incidence of
alcoholism and smoking than other individuals with
schizophrenia27,28. Polydipsic patients with schizophrenia
can often be identified in the community setting because
they have a cup in their hand at all times, drink from toi-
lets and gather around radiators during cold weather26.
They rarely drink at night compared with patients with DI
and their day-time drinking coincides with other stereo-
typies (such as smoking, pacing and mannerisms)29.
Hyponatraemia in patients with PIP must be distin-
guished from that attributable to psychotropic medica-
tions (such as antipsychotic drugs or the anticonvulsant
carbamazepine) or medications for the treatment of
hypertension (such as thiazides) and diabetes (such as
chlorpropamide), which is commonly seen in other
polydipsic patients with and without schizophrenia14,27.
The impaired water excretion induced by antipsychotic
drugs can be distinguished from that in patients with PIP
because it is stable and more marked30.
The prevalence of CWD is higher in women than in
men14. Unlike patients with psychotic disorders, com-
pulsive water drinkers are much more likely to complain
of excessive thirst and seem to be prone to psycho-
somatic disorders31,32. Many compulsive water drinkers
have depression, anxiety, obsessive–compulsive dis order,
anorexia nervosa or alcoholism, and others become
habituated to drinking owing to fear of dehydration or
the belief that water improves health33.
Dipsogenic DI is inevitably associated with increased
thirst and is assumed to involve systemic rather than
psychological factors34. The extent of overlap between
compulsive water drinkers without psychiatric illness
and patients with dipsogenic DI is unclear. Because of
the absence of distinguishing features (noted above),
dipsogenic DI can rarely if ever be definitively distin-
guished from other forms of DI without assessing the
osmotic threshold for thirst35. Primary polydipsia asso-
ciated with organic brain disorders often occurs in con-
junction with polyphagia. Hypothalamic sarcoidosis in
particular is more likely to enhance water intake than to
impair AVP secretion32.
Gestational DI. Gestational DI typically occurs at the
end of the second or early third trimester (the peak of
placental vasopressinase production)36. Vasopressinase
production is proportional to the size of the placenta
and thus the risk of gestational DI is higher for twin and
multiple pregnancies. Transient gestational DI has also
been reported in patients with pre- eclampsia, HELLP
(haemolysis, elevated liver enzymes and low platelet
count) syndrome and acute fatty liver disease (owing to
impaired hepatic degradation of vasopressinase)37.
Mechanisms/pathophysiology
Excessive renal excretion of large volumes of dilute urine
in DI is caused by a decrease in the secretion or action of
AVP (except in primary polydipsia), which can be partial
or almost complete, depending on the underlying cause.
Central DI
Central DI is most often due to various acquired or
hereditary lesions that destroy or damage the neuro-
hypophysis, either by pressure or by infiltration (TABLE1).
Central diabetes
insipidus
Destruction of
neurohypophyseal
neurons or
mutations in AVP
Nephrogenic
diabetes insipidus
Decreased response
to AVP or mutations
in AVPR2 or AQP2
↓ Free water reabsorption
↑ Solute-free water diuresis
↑ AVP degradation
Renal
resistance
to serum AVP
↓ Pituitary
AVP release
Primary
polydipsia
Excessive
fluid intake
Gestational
diabetes insipidus
Excessive activity
of placental
vasopressinase
Fig. 1 | Pathophysiology of DI. Diabetes insipidus (DI) is a form of polyuria–polydipsia
syndrome, which is caused by various acquired or hereditary lesions or disorders. Central
DI results from inadequate production and/or secretion of arginine vasopressin (AVP)
in the hypothalamic–neurohypophyseal system in response to osmotic stimulation.
Acquired central DI is caused by disruption of the neurohypophysis, whereas hereditary
central DI is due to mutations in AVP . Nephrogenic DI is the result of an inadequate
response of the kidneys to AVP, either acquired (as an adverse effect of various drugs or
due to electrolyte disorders) or hereditary (due to mutations in the genes encoding
arginine vasopressin receptor 2 (AVPR2) or the water channel aquaporin 2 (AQP2)).
In primary polydipsia, excessive fluid intake that leads to polyuria occurs, even when AVP
secretion and an appropriate antidiuretic renal response are present. In gestational DI,
increased activity of arginine vasopressinase during pregnancy reduces the levels of AVP,
leading to a presentation similar to that of central DI.
3NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
Primer
0123456789();

The severity of the resulting hypotonic diuresis is
dependent on the extent of the neurohypophyseal dam-
age, resulting in either partial or complete deficiency of
AVP secretion. The levels of oxytocin, the other hor-
mone secreted fromthe posterior pituitary, also seem
to be low in patients with central DI, paralleling an
increased psychopathology that is suggestive of a possi-
ble oxytocin- deficient state. However, larger studies are
required to confirm these findings38.
Acquired central DI. Despite the large number of
lesions that can produce central DI, it is actually more
common for patients with these lesions to not develop
central DI. This apparent inconsistency is the result of
several aspects of neurohypophyseal physiology. First,
AVP synthesis occurs in the hypothalamus and not in
the posterior pituitary gland, which is only the site of
storage and secretion of the AVP- containing neuro-
secretory granules (FIG.2). Thus, lesions in the sella
turcica that damage only the posterior pituitary gland
leave the cell bodies of the AVP- synthesizing magnocel-
lular neurons intact and therefore do not usually cause
central DI; for example, the gradual destruction of the
posterior pituitary gland by slowly enlarging large pitui-
tary macroadenomas damages only the nerve terminals
but not the cell bodies of the AVP- producing neurons,
allowing sufficient time for the site of AVP release to
shift more superiorly to the pituitary stalk and median
eminence. Indeed, the development of DI from a pitui-
tary adenoma is so uncommon that its presence should
lead to consideration of alternative diagnoses, such as
craniopharyngioma or more rapidly enlarging sellar or
suprasellar masses (such as metastatic lesions or acute
haemorrhage), which do not allow sufficient time for a
shift in the site of AVP release.
Second, the AVP synthesis capacity of the neurohypo-
physeal neurons considerably exceeds daily needs for
maintaining water homeostasis. Destruction of 80–90%
of the AVP- synthesizing magnocellular neurons in the
hypothalamus following surgical resection of the pitui-
tary stalk is required to produce polyuria and polydipsia
in dogs39. Thus, even lesions that destroy the cell bodies
of these neurons must produce fairly extensive damage to
cause DI. Necropsy studies of human patients after pitui-
tary stalk resection showed atrophy of the posterior pitu-
itary gland and loss of the magnocellular neurons in the
hypothalamus owing to retrograde degeneration of neu-
rons with axons severed during surgery40. Similar to all
neurons, the probability of retrograde neuronal degen-
eration occurring depends on how close the axotomy is
to the cell body of the magnocellular neuron. Inhumans,
transection of the pituitary stalk at the level of the dia-
phragm sellae (that is, a ‘low stalk transection’) caused
only transient central DI, whereas transection at the level
of the infundibulum (that is, a ‘high stalk transection’)
caused permanent central DI in most patients41.
Neural lobe
Inferior
hypophyseal artery
Autoimmune or
inflammatory destruction
of AVP-producing
neurons or their axons
Axon terminal degeneration:
uncontrolled AVP release
2
Pituitary stalk injury:
axonal shock
1
Rapid enlargement
of anterior pituitary
and compression of
axon terminals of AVP-
producing neurons
Anterior lobe
Axon terminal
Axon
Hypothalamo-
hypophyseal tract
Oxytocin
vesicle
Median eminence
Supraoptic nucleus
Paraventricular nucleus
Vasopressin
vesicle
Retrograde axonal
degradation:
death of cell bodies of
AVP-producing neurons
3
Systemic release of
vasopressin and oxytocin
into the circulation
Fig. 2 | Pathogenetic mechanisms in acquired central DI. In the triphasic response, the first phase of central
diabetesinsipidus (DI) occurs after pituitary stalk damage that severs the connections between the cell bodies (in the
hypothalamus) and axons (in the posterior pituitary gland) of the arginine vasopressin (AVP)-producing magnocellular
neurons, which prevents stimulated secretion of AVP (step 1). This is followed in several days by the second phase,
syndrome of inappropriate antidiuretic hormone secretion (SIADH), which is caused by uncontrolled release of AVP
fromthe degenerating nerve terminals in the posterior pituitary gland into the bloodstream (step 2). After all stored
AVP in the posterior pituitary gland has been released, the third phase of DI occurs if the cell bodies of >80–90% of
theAVP- producing neurons in the hypothalamus undergo retrograde degeneration (step 3). Similar pathogenetic
mechanisms underlie autoimmune and inflammatory aetiologies of central DI that result in destruction of axons and cell
bodies of AVP- producing neurons, and rapid enlargement of anterior pituitary lesions (for example, metastatic lesions and
pituitary apoplexy) that compress the axons of AVP- producing neurons. Adapted with permission from REF.162, Elsevier.
4
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();

Central DI resulting from surgical or traumatic injury
to the neurohypophysis is a unique situation that can
result in several different, well- defined phenotypes.
In some patients, polyuria develops 1–4 days after the
injury but then resolves spontaneously. Less commonly,
DI persists for longer periods and may become per-
manent (this occurs more frequently with suprasellar
lesions, such as craniopharyngioma). Most interest-
ingly, pituitary stalk transection can result in a triphasic
response. The initial DI (first phase) lasts several hours
to several days and is due to axon shock and functional
deficiency of the damaged neurons. The second, anti-
diuretic phase (termed syndrome of inappropriate
antidiuretic hormone secretion (SIADH)) can persist
for 2–14 days and is due to the uncontrolled release of
AVP from the disconnected, degenerating posterior
pituitary gland42. Importantly, in this second phase,
overly aggressive fluid administration does not suppress
AVP secretion and can result in hyponatraemia. In the
third phase, DI recurs after depletion of the AVP from
the degenerating posterior pituitary gland43.
Idiopathic forms of central DI are a large category;
in a study of paediatric patients with central DI, DI was
idiopathic in 54% on initial classification9. However,
longer- term follow up showed a diagnostic accuracy of
96% in those patients initially classified as having idio-
pathic DI44. These patients usually have no history of
previous injury or disease that might have contributed
to their central DI, and pituitary MRI reveals no abnor-
malities other than the absence of the posterior pituitary
bright spot (PBS; see below) and sometimes thickening
of the pituitary stalk. Multiple lines of evidence suggest
that autoimmune destruction of the neurohypophysis
is the most likely cause of the central DI in many of
these patients20,45, including biopsy samples and post-
mortem examination of patients with idiopathic cen-
tral DI demonstrating lymphocytic infiltration of the
pituitary stalk20,45 and studies demonstrating a high
prevalence (67%) of anti- vasopressin cell antibodies
in young patients with non- traumatic central DI46.
Although the presence of anti- vasopressin cell antibod-
ies supports the involvement of an autoimmune process
in many cases of idiopathic DI, these antibodies have
also been detected in DI of other aetiologies, including
Langerhans cell histiocytosis and germinoma, and thus
cannot be considered a reliable marker of autoimmune-
mediated DI42. Furthermore, in a form of infundibulo-
neurohypophysitis that occurs in middle- aged to older
men and is associated with immunoglobulin G4
(IgG4)-related systemic disease47, various organs, espe-
cially the pancreas, are infiltrated with IgG4-secreting
plasma cells, and neurohypophysitis is only one mani-
festation of a multi-organ disease that may affect other
endocrineglands.
Table 1 | Aetiology of polyuria–polydipsia syndromes
Basic defect Acquired causes Hereditary causes
Central DI
Deficiency in
AVP synthesis or
secretion
• Trauma (surgery and deceleration injury)
• Neoplasia (craniopharyngioma, meningioma, germinoma
and metastases)
• Vascular (cerebral or hypothalamic haemorrhage and infarction
or ligation of anterior communicating artery aneurysm)
• Granulomatous (histiocytosis and sarcoidosis)
• Infectious (meningitis, encephalitis and tuberculosis)
• Inflammatory or autoimmune (lymphocytic infundibuloneuro-
hypophysitis and IgG4 neurohypophysitis)
• Drug or toxin exposure
• Osmoreceptor dysfunction (adipsic DI)
• Others (hydrocephalus, ventricular or suprasellar cyst,
andtrauma and degenerative diseases)
• Idiopathic
• Autosomal dominant:
AVP mutations
• Autosomal recessive,
type a and b: AVP mutations
• Autosomal recessive,
type c: WFS1 mutations
• Autosomal recessive,
type d: PCSK1 mutations
• X- linked recessive: gene
unknown
Nephrogenic DI
Reduced renal
sensitivity to
antidiuretic effect
of physiological
AVP levels
• Drug exposure (lithium, demeclocycline, cisplatin, etc.)
• Hypercalcaemia or hypokalaemia
• Infiltrating lesions (sarcoidosis, amyloidosis, multiple
myeloma, etc.)
• Vascular disorders (sickle cell anaemia)
• Mechanical (polycystic kidney disease and urethral obstruction)
• X- linked: AVPR2 mutations
• Autosomal recessive or
dominant: AQP2 mutations
Primary polydipsia
Excessive fluid
intake at a
diminished set
point
• Dipsogenica (idiopathic or similar lesions as with central DI)
• Psychosis intermittent hyponatraemia–polydipsia (PIP) syndrome
• Compulsive water drinking
• Health enthusiasts
NA
Gestational DI
Increased enzymatic
metabolism of
circulating AVP
hormone
Pregnancy NA
aDownward resetting of the thirst threshold. AVP, arginine vasopressin; DI, diabetes insipidus; NA , not applicable.
5NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
Primer
0123456789();

One of the most severe forms of central DI results
from destruction of the osmoreceptors that stimulate
neurohypophyseal secretion of AVP. Studies in animals
indicate that the primary osmoreceptors that control
thirst and the secretion of AVP are located in the ante-
rior hypothalamus in two circumventricular organs, the
subfornical organ (SFO) and the organum vasculosum
of the lamina terminalis (OVLT). Lesions in this region,
termed the AV3V area, cause hyperosmolality through
impaired thirst and impaired osmotically stimulated
secretion of AVP48,49. The SFO has been implicated in
the reciprocal control of appetite for sodium and water50.
Because of the profound thirst deficits found in most
of these patients, this condition is often termed adipsic
central DI51. All reported cases have been due to osmo-
receptor destruction to various extents and are associated
with a range of different brain lesions (TABLE1). Although
many of these lesions are of the same type as those that
cause central DI (FIG.2), they usually occur more rostrally
in the hypothalamus, consistent with the location of the
primary osmoreceptor cells in the anterior hypothala-
mus. Adipsic hypernatraemia without hypothalamic
lesions, accompanied by autoantibodies to the SFO, is a
new, well- described disease in four young patients with
hypernatraemia, no thirst and low vasopressin response
to hypertonicity. Whereas structural abnormalities of the
hypothalamic area are easily visible by MRI in classical
adipsic hypernatraemia, no hypothalamic structural
lesions could be identified in these patients. Specific
circulating antibodies reactive to the mouse SFO and
the sodium channel Nax were present in the serum of
all four patients. Mice injected with an immunoglobulin
fraction of the patient’s serum showed abnormalities in
water and sodium homeostasis, vasopressin release and
diuresis, which resulted in hypernatraemia52.
Hereditary central DI. Familial neurohypophyseal DI
(FNDI; also known as hereditary central DI) comes in
many forms (TABLE1) that are differentiated by the inher-
itance pattern and the underlying genetic lesion, which
include mutations in AVP , WFS1 and PCSK1 (encoding
proprotein convertase subtilisin/kexin type 1)53. PCSK1
is involved in processing numerous hypothalamic and
digestive prohormones and PCSK1 mutations can result
in severe malabsorptive diarrhoea, growth hormone defi-
ciency, central hypothyroidism, central hypogonadism
and central hypocortisolism; ~80% of patients with these
mutations show clinical signs of central DI54. Autosomal
dominant FNDI is caused by >70 different mutations
in different parts of the AVP gene, none of which occur in
the copeptin moiety. Despite the wide range of mutations,
patients with autosomal dominant FNDI have remark-
ably similar presentations. All heterozygous infants are
healthy at birth and show no signs of DI but develop
complete DI at a later age, which seems to vary depend-
ing on the mutation. Later age of onset (up to young
adulthood) occurs in carriers with mutations in the sig-
nal peptide, whereas onset occurs in infancy for carriers
with mutations in the AVP carrier neurophysinII2, which
is produced by cleavage of the AVP preprohormone. The
mechanism underlying the ‘dominant- negative’ effect of
autosomal dominant FNDI mutations is controversial
but most evidence suggests that mutant AVP prohor-
mone is retained in the endoplasmic reticulum (ER)
of magnocellular neurons. Mutant AVP and functional
AVP protein produced from the non- affected allele form
high- molecular-weight complexes that are destined for
ubiquitylation and proteasomal degradation by the ER
quality control pathway ER- associated degradation
(ERAD)55. This model helps explain results obtained in
some mouse models of DI with human proAVP muta-
tions, in which AVP- producing neuronal cell death was
not observed at disease onset and was thus not thought
to have a role in disease initiation56.
Autosomal recessive FNDI is a rare disorder caused
by mutations in AVP (types a and b) or other genes
(types c and d) (TABLE1). The clinical phenotype differs
from that in autosomal dominant FNDI in several ways,
including age of onset, plasma AVP levels during fluid
deprivation, interfamily and intrafamily variation, and
co- morbidity with other symptoms. The most common
form of autosomal recessive FNDI, type c, is due to muta-
tions in WFS1, as a clinical manifestation in Wolfram
syndrome. X- linked recessive FNDI has been reported
in one kindred with a classical FNDIphenotype, and has
been mapped to Xq28, although the responsible gene or
genes have not yet been identified57.
Genetic evaluation of patients with suspected inher-
ited central DI is fairly simple in most patients and
should be considered in those with a positive family
history of DI or with idiopathic forms of DI that appear
at a young age58,59.
Nephrogenic DI
AVP- regulated water permeability is a central com-
ponent of the renal urine- concentrating mechanism.
In diuretic kidneys in which AVP is low or absent,
the collecting duct is water impermeable and dilute
urine is produced. In antidiuretic kidneys, AVP levels
are increased and the collecting duct becomes water-
permeable due to increased levels of the water channel
AQP2 in the apical membrane of principal cells resulting
from exocytosis of AQP2-containing subapical vesicles60.
Urine becomes concentrated as water is transported
osmotically from the lumen of the collecting duct into
the renal interstitium, which is hyperosmolar as a con-
sequence of the renal countercurrent multiplication and
exchange mechanisms. The basolateral cell membrane
is constitutively water- permeable due to expression of
the water channels AQP3 and/or AQP4. AVP binding
to AVPR2 at the basolateral membrane of principal cells
results in cAMP production by adenylyl cyclase 6 (REF.61)
and activation of protein kinase A, which leads to phos-
phorylation of AQP2 and proteins involved in AQP2
exocytosis62,63 (FIG.3a). Although this basic mechanism
is supported by considerable data, questions remain
about mechanisms of apparent AVPR2-dependent but
cAMP- independent AQP2 trafficking, and the role
of non- vasopressin modulators of AQP2 trafficking,
such as prostaglandin E2, nitric oxide and adenosine.
Nephrogenic DI is caused by reduced renal sensitiv-
ity to the antidiuretic effect of physiological levels of
AVP3, owing to an acquired or genetic defect in renal
mechanisms for urine concentration.
6
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();

Acquired nephrogenic DI. Acquired nephrogenic DI is
much more common than hereditary nephrogenic DI,
israrely severe and polyuria and polydipsia are mod-
erate (3–4 l/day). The inability to generate a maximal
urine osmolality results from resistance to the action of
AVP in the collecting tubules or interference with the
countercurrent mechanism secondary to medullary
injury or to decreased sodium chloride reabsorption in
the medullary thick ascending limb of the loop of Henle.
Lithium administration is the most common cause
of acquired nephrogenic DI — 19% of 1,105 unse-
lected patients on chronic lithium therapy had poly-
uria (>3 l/day)64. Lithium treatment causes reduced
AQP2 expression and altered trafficking in the short
term and loss of principal cells in the long term65,66.
Lithium- induced AQP2 downregulation is probably a
consequence of ENaC- mediated influx of lithium into
principal cells, as collecting- duct-specific αENaC defi-
ciency prevents the development of lithium- induced
nephrogenic DI67. Lithium inhibits glycogen synthase
kinase 3 (GSK3) signalling pathways68,69 (FIG.3b). Both
GSK3 isoforms, GSK3α and GSK3β (encoded by GSKA
and GSK3B, respectively), are inhibited by lithium, both
directly and indirectly, by increased phosphorylation of
Ser9 in GSK3β and Ser21 in GSK3α70. Other GSK3 inhib-
itors also reduce AQP2 expression in collecting duct cells
invitro, and ablation of Gsk3a or Gsk3b in mice causes
polyuria70. Inhibition of GSK3β by lithium increases the
expression of cyclooxygenase 2 and the local excretion
of prostaglandin E2 (REF.71), which may counteract vas-
opressin action by causing endocytic retrieval of AQP2
from the plasma membrane. The urinary concentration
of lithium in patients on well- controlled lithium therapy
is sufficient to have this effect.
Other causes of transient acquired nephrogenic DI
include hypercalcaemia, hypercalciuria and obstruc-
tive uropathy. In patients with obstructive uropathy,
the observed suppression of AQP2 expression might be
mediated by hydrostatic pressure72. Autophagic degrada-
tion of AQP2 is implicated in nephrogenic DI induced
by hypokalaemia and hypercalcaemia73,74 (FIG.3b).
Hypokalaemia and hypercalcaemia or hypercalciuria
affect water permeability through regulating autophagic
degradation of water and urea channels (FIG.3).
Hereditary nephrogenic DI. Mutations in AVPR2
(located at Xq28 and encoding a G- protein-coupled
receptor) cause X- linked hereditary nephrogenic DI,
resulting in complete vasopressin insensitivity in affected
males75,76. More than 250 mutations in AVPR2 have been
identified, including missense mutations, nonsense
mutations, deletions and insertions10. Mutations that
a b
AQP2
H2O
H2O
H2O
H2O
ATP
cAMP
Gαs
AVPR2
AQP3
AQP4
EP2 and
EP4
Actin
filament
Microtubule Microtubule
motor
Gi and
endocytic
retrieval
Exocytic
insertion
PDEs
Endocytic
vesicle PKA
Recycling vesicle
Actin
filament
motor
Adenylyl
cyclase
BasolateralLuminal
C2C1
Syntaxin 4
BloodLumen
AQP2
CaSR
Ca2+
UTA1
Autolysosome
AQP2
Autophagosome Phagophore
2K+
3Na
+
Na/K-ATPase
K+
UTA1
K+
Li+Li+
GSK3
P
P
P
P
P
P
Principal cell
AQP2
Fig. 3 | Pathogenetic mechanisms in nephrogenic DI. a | Mechanism of arginine vasopressin (AVP)-stimulated osmotic
water permeability in principal cells of the collecting duct. AVP binding to the G protein- coupled receptor AVP receptor 2
(AVPR2) results in increased production of cAMP by adenylyl cyclase 6, thereby activating protein kinase A and inducing
phosphorylation of target proteins, including the water channel aquaporin 2 (AQP2). These phosphorylations promote
fusion of AQP2-containing vesicles with the apical plasma membrane of the principal cells, and thereby increased AQP2
levels, resulting in increased water uptake from the urine. The basolateral plasma membrane expresses AQP3 and AQP4,
making it constitutively water- permeable. b | Mechanisms of nephrogenic diabetes insipidus (DI) caused by lithium,
hypokalaemia and hypercalcaemia or hypercalciuria. Entry of lithium into the principal cell inhibits glycogen synthase
kinase 3β (GSK3β), reducing expression of AQP2 (which forms tetramers). Hypokalaemia and hypercalcaemia or
hypercalciuria cause autophagy- mediated degradation of monomeric AQP2 and UTA1. Hypercalciuria may activate the
calcium- sensing receptor (CaSR), which increases intracellular Ca2+ levels and enhances basal autophagy by one or more
mechanisms. Autophagy is initiated by the formation of phagophores, which engulf AQP2, UTA1 and other cytoplasmic
proteins (including junctional and cytoskeletal proteins), as well as dysfunctional organelles (such as damaged mitochondria).
Phagophores elongate and close to generate double- membrane autophagosomes, which then fuse with lysosomes to form
single- membrane autolysosomes, thereby delivering cargo for degradation. As a result, decreased abundance of AQP2 and
UTA1 leads to impaired urinary concentrating ability. Part a adapted from REF.3, Springer Nature Limited. Part b adapted with
permission from REF.74, Elsevier.
7NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
Primer
0123456789();

result in the production of a full- length protein are most
common, although these mutant proteins are misfolded
and are targeted for ERAD.
Mutations in AQP2 (located at 12q13 and encoding
the water- selective transporter AQP2) cause heredi-
tary autosomal nephrogenic DI77. Of the ~65 disease-
causing AQP2 mutations identified, most are missense
or nonsense mutations that cause autosomal recessive
nephrogenic DI78. Some missense mutations result
in the production of a mislocalized but functional
water- transporting AQP2, whereas others result in
severely misfolded proteins that are targeted for ERAD.
Interestingly, a few mutations in the carboxyl terminus
of AQP2 result in autosomal dominant nephrogenic
DI, probably because heterotetramers of mutant and
wild- type AQP2 are retained in the Golgi, preventing
their exocytosis79.
Primary polydipsia
Primary polydipsia produces physiological suppression
of AVP secretion by excessive water intake, and is thus
the opposite of the secondary polydipsia that occurs in
response to excessive water loss due to pathologically
diminished AVP activity in other types of DI. The exces-
sive intake in individuals with CWD is attributable to
the sluggish drop in thirst that occurs immediately after
water intake (oropharyngeal regulation) and their dimin-
ished osmotic set point for thirst (but not AVP)80. Actua l
water intake also seems to be greater at any level of thirst.
The extent to which dipsogenic DI is distinct from CWD
is unknown, although patients with dipsogenic DI have
a higher osmotic set point for AVP and a higher plasma
osmolality than those with CWD34. Hypothalamic
sarcoidosis seems to disrupt the normal congruence
of the set points in AVP and desire for water81, and an
analogous mechanism could operate in CWD.
Primary polydipsia occurs most commonly in
patients with chronic schizophrenia. The discussion
below compares these patients and the subset with PIP
to matched non- polydipsic patients with schizophre-
nia. Because the increased water intake in polydipsic
patients seems to be unrelated to thirst25, it is assessed
by asking patients about their ‘desire for water’ (in cups),
which can be subsequently validated by a period of ad
libitumdrinking.
Immediately following water intake, desire for water
drops acutely in polydipsic patients but then rapidly
rebounds, indicating that an external factor overrides
normal oropharyngeal suppression of water intake
(FIG.1). By contrast, plasma AVP exhibits the normal
acute drop and does not abnormally rebound. The
osmotic set point for ‘desire for water’ is diminished
in polydipsic patients, whereas the increase in desire at
higher levels of plasma osmolality is blunted, but only
inthe subset of patients with PIP (reviewed elsewhere82).
The osmotic set point for AVP secretion is also dimin-
ished but only in the PIP subset and is further aggra-
vated, in this subset alone, by acute psychological stress
and psychotic exacerbations to a degree capable of
inducing water intoxication (FIG.4a). Thus, unlike other
patients with primary polydipsia, the patient subset with
PIP fail to appropriately suppress AVP.
These findings in polydipsic psychotic patients seem
attributable to an anterior hippocampus- mediated dis-
ruption of hypothalamic function that may also contrib-
ute to their psychiatric illness83. Polydipsic patients with
and without PIP exhibit discrete deformations on the
surface of the anterior hippocampus, the part of the brain
that is also most consistently implicated in the patho-
physiology of schizophrenia. In polydipsic patients,
these deformations are restricted to the anterior lateral
surface, which projects to the anterior hypothalamus and
normally restrains stress hormone secretagogue release
from parvocellular neurons and AVP release from magno-
cellular neurons during psychological stress84 (FIG.2).
Larger deformations are apparent in polydipsic patients
with PIP than in those without PIP, whereas deforma-
tions in non- polydipsic patients are limited to the oppo-
site (anterior medial) surface. AVP and stress hormone
responses to psychological (but not physical) stress are
greatest in the PIP subset of polydipsic patients, normal
in non- PIP polydipsic patients and blunted in non-
polydipsic patients compared with healthy individuals.
Indeed, the extent of these deformations as well as those
on the medial surface of the amygdala (which is also
heavily implicated in schizophrenia and in hypothalamic
regulation of neuroendocrine and diverse psychologi-
cally driven stress responses) are proportional to the
AVP responses in the three patient groups.
These results support those of other studies show-
ing that anterior hippocampus pathology in polydipsic
patients induces a limbic- based stress diathesis85 that
underlies their water imbalance and features of their
psychiatric illness. In particular, the anterior lateral
hippocampal and medial amygdala deformations are
also proportional to the level of oxytocin that is secreted
from adjacent magnocellular neurons in the anterior
hypothalamus, and normally promotes diverse social
behaviours that are particularly impaired (contributing
to negative psychotic symptoms) in those with polydip-
sia. Impairments in social cognition, and particularly
fear, are proportional to oxytocin levels in patients with
polydipsia and are ameliorated by intranasal oxytocin
administration in those with polydipsia but not in those
without polydipsia83 (FIG.4a).
Familial concordance supports a role for genetic
factors in polydipsia, of which polymorphisms in the
orexin 1 receptor are most compelling86. Why polydipsia
would be a consequence of an anterior hippocampus-
mediated stress diathesis is unclear, although simi-
larities between the polydipsia and other stereotypic
behaviours commonly found in this subset of patients
and schedule- induced polydipsia (SIP) and other stereo-
typic behaviours seen in mammals with hippocampal
lesions may be relevant87. SIP is enhanced by neuroendo-
crine dysfunction88, is associated with other neural and
functional changes commonly seen in individuals with
schizophrenia83,89, is probably an abnormal response to
stress (that is, stress diathesis)90 and, like the polydip-
sia in patients, is preferentially diminished by clozapine
compared with other antipsychotic medications62,91.
Technological advances have helped characterize
how pre- systemic and even pre- ingestion factors that
motivate water intake and enhance AVP secretion92
8
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();

are integrated in the lamina terminalis, along with the
better characterized osmotic, cardiovascular and circa-
dian influences that regulate water balance93,94 (FIG.1).
The anterior hippocampus- mediated stress diathesis
could operate through this pathway and thus the ability
to isolate its disruptive effects on water balance could
reveal how it also disrupts hypothalamically mod-
ulated behaviours and effects that contribute to the
psychoticdisorder95.
Gestational DI
Gestational DI is caused by increased degradation of
AVP by the placental enzyme vasopressinase96,97, which
results in a presentation resembling that of central DI
(FIG.4b). Some patients may be predisposed to gesta-
tional DI as a result of pre- existing, subclinical AVP
deficiency98,99. Vasopressinase is secreted by the kidneys
and the liver100. During pregnancy, placental tropho-
blasts also produce vasopressinase, which is detectable at
10 weeks of gestation. Circulating placental vasopressi-
nase levels increase ~300-fold over the following weeks
(peaking in the third trimester)36, remain high during
labour and delivery and return to undetectable levels
around the second week postpartum. The increased
placental vasopressinase levels lead to increased degra-
dation of AVP101, although AVP levels remain in the nor-
mal range in the majority of pregnant women owing to
increased AVP secretion by the posterior pituitary gland.
Diagnosis, screening and prevention
Clinical manifestations
Polyuria and polydipsia in DI and primary polydipsia
do not necessarily differ in their specific manifesta-
tions, even though the underlying impairment of uri-
nary concentrating mechanisms is different in the two
conditions102. Compared with other forms of DI, patients
with central DI more often describe nocturia and a sud-
den onset of symptoms, as urinary concentration can
often be maintained fairly well until the residual neuro-
nal capacity of the hypothalamus to synthesize AVP falls
below 10–15% of normal capacity, after which urine
output increases dramatically.
• Water excretion diminishes with acute psychosis (1923)
• Polydipsia occurs in 25% of patients with chronic psychosis (1933)
• Acute water intoxication coincides with psychotic exacerbation (1938)
• AVP increases in patients with PIP during psychosis (1975)
• Marked impairments in social function (2007)
• Reset osmostat for AVP in PIP worsens with acute psychosis
• Reset osmostat for desire for water
• Blunted gain of osmostat for desire for water in PIP
• Normal acute drop with drinking but rapid rebound of desire for water
• Impaired social functioning
• Inability to correctly recognize facial emotions greater in PIP
• Enhanced perception of emotional intensity
• Normalization of perceived emotional intensity with intranasal
oxytocin, particularly for fear
• Diminished lateral anterior hippocampus volume greater in PIP
• Diminished medial amygdala volume greater in PIP
• Enhanced resting state connectivity between anterior hippocampus,
medial amygdala and hypothalamus
• Enhanced AVP responses to psychological stress greater in PIP
• Enhanced stress hormone responses to psychological stress greater in PIP
• Reset osmostat for AVP in PIP worsens with psychological stress
• Diminished oxytocin levels greater in PIP
• Diminished hippocampus-mediated glucocorticoid negative
feedback greater in PIP
a b
Labour, delivery
and postpartum
• Vasopressinase levels
remain increased during
labour and delivery and
return to undetectable
levels by the second
week postpartum
Pregnancy
1st trimester
• Vasopressinase produced
in placental trophoblasts,
detectable at 10 weeks
of gestation
2nd trimester
• Increasing production of
placental vasopressinase
3rd trimester
• Maximum levels of
placental vasopressinase;
production proportional
to the size of the placenta,
with highest levels in twin
and multiple pregnancies
Increasing AVP degradation: ↑ Polyuria and polydipsia
Initial observations
(year of observation)
Observations linked
to hypothalamus-
mediated water
imbalance
Observations linked
to psychiatric illness
and oxytocin
Observations linked
to lateral anterior
hippocampal pathology
Observations linked
to AH-mediated
‘stress diathesis’
Fig. 4 | Models of pathogenesis in primary polydipsia in schizophrenia and gestational DI. a | Primary polydipsia.
Findings pertain to schizophrenia patients with primary polydipsia with and without the psychosis intermittent
hyponatraemia–polydipsia syndrome (PIP). The figure highlights the initial unexplained observations suggesting that the
life- threatening water imbalance in patients with PIP was directly linked to their psychotic disorder, as well as subsequent
studies that provide plausible pathophysiological mechanisms arising from disruption of recognized mammalian neural
functions. The structural and functional findings support the view that the more- disrupted neuroendocrine function in
patients with PIP than in non- PIP polydipsic patients is due to more- extensive pathological changes in the anterior lateral
hippocampus, whereas non- polydipsic patients have structural changes on the anterior medial surface which do not
interfere with their normal hippocampus- mediated compensatory neuroendocrine responses to chronic psychological
stress. b | Gestational diabetes insipidus (DI) is caused by increased degradation of arginine vasopressin (AVP) by placental
vasopressinase, which results in a presentation resembling that in central DI. Placental vasopressinase is produced by
placental trophoblasts and is detectable by 10 weeks of gestation. Circulating vasopressinase levels increase about
300-fold over the following weeks, peaking in the third trimester, remain higher during labour and delivery and return
toundetectable levels around the second week postpartum. AH, anterior hippocampus.
9NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
Primer
0123456789();

Patients with DI, especially those with underlying
osmoreceptor defect syndromes, can also show vary-
ing degrees of dehydration and hyperosmolality if renal
water losses cannot be fully compensated for by fluid
intake. The resulting symptoms are due to dehydration
(mostly cardiovascular symptoms, including hypoten-
sion, acute tubular necrosis secondary to renal hypoper-
fusion and hypovolaemic shock)103,104 or hyperosmolality
(mostly neurological symptoms that reflect the extent
of brain dehydration as a result of osmotic water shifts
from the intracellular compartment). Manifestations
may range from non- specific symptoms, such as irri-
tability and cognitive dysfunction, to more severe
manifestations, such as disorientation, reduced level of
consciousness, seizure, coma, focal neurological deficits
and cerebral infarction103,105.
Polyuria in children is defined as excretion of urinary
volumes of >150 ml/kg/day in neonates, >100–110 ml/
kg/day in children ≤2 years of age and >50 ml/kg/day
in older children7. In children, central DI can often be
accompanied by additional signs or symptoms, such
as growth retardation, fatigue, headaches, emesis and
visual field deficits, owing to intracranial neoplasms that
affect CNS structures and other pituitary axes9. A strong
preference for water in children with central DI (which
limits the intake of more caloric liquids or solids) or
associated growth hormone deficiency can slow weight
gain and linear growth.
Differential diagnosis
Once hypotonic polyuria is confirmed, the next step is
to distinguish between central DI, nephrogenic DI and
primary polydipsia, which is crucial because treatment
strategies differ and application of the wrong treatment
can be dangerous4. However, reliable diagnosis is dif-
ficult5, as many available tests are unsatisfactory31 and
often result in false diagnoses, especially in patients with
primary polydipsia or partial, mild forms of central DI1,6.
The indirect water deprivation test was the gold
standard for differential diagnosis of polyuria–polydipsia
syndrome for many years. This test is based on indi-
rect assessment of AVP activity by measurement of the
urine concentration capacity during a prolonged period
of dehydration and again after a subsequent injectionof
an exogenous synthetic AVP analogue, desmopres-
sin106–108. Interpretation of the test results is based on
published recommendations109. If upon thirsting, uri-
nary osmolality remains <300 mOsm/kg and does not
increase by >50% after desmopressin injection, com-
plete nephrogenic DI is diagnosed. Complete central DI
is diagnosed if the urinary osmolality increase is >50%
after desmopressin injection. In partial central DI and
primary polydipsia, urinary concentration increases to
300–800 mOsm/kg, with an increase of >9% (in par-
tial central DI) and <9% (in primary polydipsia) after
desmopressin injection.
However, these published criteria are based on post
hoc data from only 36 patients, who had a wide over-
lap in urinary osmolalities. Furthermore, the diagnos-
tic criteria for this test were derived from a single study
with post- hoc assessment109 and have not been prospec-
tively validated on a larger scale (reviewed elsewhere4).
Consequently, this test has been shown to have consid-
erable diagnostic limitations; overall diagnostic accuracy
is 70%, and accuracy is only 41% in patients with
primary polydipsia6.
Several reasons exist for the disappointing diagnos-
tic outcome of the indirect water deprivation test. First,
chronic polyuria itself can affect renal concentration
capacity, through renal washout110–112 or downregula-
tion of AQP2 expression in the kidneys113, which may
lead to a reduced renal response to osmotic stimula-
tion or exogenous desmopressin113 in different forms
of chronic polyuria109. Second, in patients with AVP
deficiency, urinary concentrations can be higher than
expected114,115, especially in those patients with impaired
glomerular function31,116,117, or can result from a com-
pensatory increase in AVPR2 expression in patients with
chronic central DI118. Finally, patients with acquired
nephrogenic DI are often only partially resistant to
AVP, resulting in a clinical presentation that is similar
to partial central DI.
To overcome these limitations of the indirect water
deprivation test, direct measurement of AVP levels has
been proposed to improve the differential diagnosis of
polyuria–polydipsia syndrome. Indeed, in a 1981 study,
patients with central DI were reported to have AVP lev-
els below a calculated normal range (defining the nor-
mal relationship between plasma osmolality and AVP
levels), whereas AVP levels were above the normal range
in patients with nephrogenic DI and within the normal
range in patients with primary polydipsia119. However,
despite these promising initial results, this method has
failed to enter routine clinical use for various reasons.
First, several technical limitations of the AVP assay
result in a high preanalytical instability of AVP in sam-
ples115,120,121. Second, the accuracy of diagnoses using
commercially available AVP assays has been disappoint-
ing, with correct diagnoses in only 38% of patients with
DI, and particularly poor differentiation between partial
central DI and primary polydipsia4,6. Third, an accurate
definition of the normal physiological range defining the
relationship between plasma AVP levels and osmolal-
ity is still lacking, especially for commercially available
assays119,122,123, which is crucial to identify AVP secre-
tion outside the normal range in patients suspected of
havingDI6.
Copeptin, the C- terminal segment of the AVP pro-
hormone, is an easy- to-measure AVP surrogate that is
very stable ex vivo120. As the serum copeptin level reflects
the osmosensitive circulating AVP concentration, it is a
promising biomarker for differential diagnosis of poly-
uria–polydipsia syndrome. Two studies have shown
that a basal copeptin level of >21.4 pmol/l without prior
thirsting unequivocally identifies nephrogenic DI, ren-
dering a further water deprivation test unnecessary in
these patients6,124. The more difficult differentiation is
between patients with primary polydipsia and those with
central DI, especially mild forms. A study in 144patients
with central DI or primary polydipsia (the largest to
date) directly compared the diagnostic accuracy of a
hypertonic saline infusion and copeptin measurement
with that of the indirect water deprivation test102. An
osmotically stimulated copeptin level of >4.9 pmol/l
10
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();

after infusion of 3% saline (aiming at a sodium level
>150 mmol/l) had an overall diagnostic accuracy of
96.5% (93.2% sensitivity and 100% specificity) in distin-
guishing between patients with primary polydipsia and
those with central DI, compared with only 76% for the
indirect water deprivation test102. The addition of copep-
tin measurement did not improve the diagnostic perfor-
mance of the indirect water deprivation test. By contrast,
the overall diagnostic accuracy of the predefined ratio
of ∆copeptin (0800–1600 h) to plasma sodium 1600 h
in distinguishing between primary polydipsia and cen-
tral DI was lower than that of the water deprivation test
without copeptin, most likely due to the lack of osmotic
stimulus and therefore the lack of a significant increase
in AVP or copeptin by thirsting alone.
Taken together, these data indicate that plasma
copeptin is a promising biomarker for differential
diagnosis of polyuria–polydipsia syndrome and that
hypertonic saline- stimulated copeptin measurement
(using the modified diagnostic workflow in FIG.5)
will probably replace the water deprivation test as the
diagnostic method of choice for hypotonic polyuria.
However, importantly, the hypertonic saline infusion test
requires close monitoring of sodium levels to ascertain
a diagnostically meaningful increase in plasma sodium
within the hyperosmotic range125,126, while preventing a
markedincrease.
Of note, the copeptin test is currently not univer-
sally available, although copeptin assays are commer-
cially available throughout Europe, Australia, India and
Mexico. Registration for commercialization is currently
pending in Taiwan, Korea and Canada. To date, the
copeptin assay has no Clinical Laboratory Amendments
certification in the USA but it is available as a research
use only (RUO) test in two large service laboratories.
Radiological findings
Once the type of DI has been diagnosed, the underlying
pathology must be identified. Gadolinium- enhanced
MRI of the sella and suprasellar regions is used to check
for anatomical disruption of the pituitary or hypotha-
lamic anatomy by, for example, macroadenoma, empty
sella, infiltrative diseases or metastases. Assessment of
the posterior pituitary gland and the pituitary stalk by
unenhanced brain MRI can sometimes be useful for
differential diagnosis of DI. An area of hyperintensity,
referred to as the PBS, is observed in healthy individ-
uals in the posterior part of the sella turcica in sagittal
T1-weighted images127, and is thought to result from the
T1-shortening effects of stored AVP in neurosecretory
Primary polydipsia
Complete or partial
central DI
Nephrogenic DIComplete central DIPartial central DIPrimary polydipsia
Stimulated copeptin
<4.9 pmol/l
(at plasma sodium
>150 mmol/l)
Stimulated copeptin
>4.9 pmol/l
(at plasma sodium
>150 mmol/l)
<50% increase>50% increase<9% increase>9% increase
Mild primary
polydipsia
Urine osmolality
>800 mOsm/kg
Complete or partial
nephrogenic DI
Copeptin
>21.4 pmol/l
Copeptin
<21.4 pmol/l
Urine osmolality
300–800 mOsm/kg
Desmopressin test
Urine osmolality
<300 mOsm/kg
Desmopressin test Hypertonic saline test
Baseline copeptin levelWater deprivation test
Urinary volume <50 ml/kg/24 h
Central or nephrogenic DIPrimary polydipsia
High serum sodium (>147 mmol/l)Low serum sodium (<135 mmol/l)
GU evaluation Urine osmolality <800 mOsm/kg
Normal serum sodium (136–146 mmol/l)
Measure serum sodium, plasma osmolality
Suspected hypotonic polyuria
Confirm the presence of polyuria (>50 ml/kg/24 h)
Fig. 5 | Modified algorithm for differential diagnosis of polyuria–polydipsia syndrome. In a first step, polyuria must
be confirmed, otherwise polyuria–polydipsia syndrome is excluded and genitourinary (GU) evaluation is needed. In case
of polyuria and a urinary osmolality <800 mOsm/kg, serum sodium and plasma osmolality are measured. If these levels
are inthe normal range, further differentiation is done using either a classical water deprivation test or a copeptin- based
algorithm (if copeptin measurement is available). DI, diabetes insipidus. This figure is modified from Figure 1 from Gubbi, S.,
Hannah- Shmouni, F., Koch, C.A. & Verbalis, J.G. in Endotext (eds. Feingold, K.R . etal.). The link to the article on PUBMED
can be found here: https://www.ncbi.nlm.nih.gov/books/NBK537591/ or at Endotext.org.
11NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
Primer
0123456789();

granules in the posterior pituitary gland128. Although
earlier small- scale studies demonstrated the presence of
the PBS in healthy individuals and its absence in patients
with central DI129, subsequent larger studies showed an
age- related absence of a PBS in 52–100% of healthy
individuals130. Conversely, a persistent PBS detected in
some patients with central DI9,131 could be because the
disease is at an early stage or could reflect oxytocin stores
rather than stored AVP. The PBS has been reported to
be absent in some patients with nephrogenic DI and
present inothers132. In a large prospective observational
study in 92patients with polyuria–polydipsia syndrome,
brain MRI revealed the presence of the PBS in only
39% of patients with primary polydipsia but in 70% of
patients with central DI102. Consequently, the presence or
absence of the PBS on MRI is not sufficient to establish
a diagnosis in patients with DI.
A similar caveat applies to imaging- based assessment
of the pituitary stalk: a diameter of >2–3 mm is generally
considered to be pathological133 (for example, in hypo-
physitis, granulomatous disorders, tuberculosis, cranio-
pharyngioma, germinoma or metastasis to the sella or
suprasellar region134) but is not necessarily specific for
idiopathic central DI102,135. However, if scans reveal thick-
ening of the stalk with the absence of the PBS, then a
diligent search for neoplastic or infiltrative lesions of the
hypothalamus or pituitary gland is indicated136.
Diagnostic evaluation of DI during pregnancy is
challenging. Baseline investigation should involve a
complete blood count, liver and kidney values, electro-
lyte levels (including serum calcium) and serum and uri-
nary glucose and osmolality. The water deprivation test
is not recommended during pregnancy due to the high
risk of dehydration and, consequently, utero- placental
insufficiency137. Also, copeptin measurement has never
been prospectively evaluated for DI diagnosis in preg-
nant patients. Importantly, measurement of osmoti-
cally stimulated copeptin levels in pregnancy cannot
be recommended. In view of the generally lower serum
sodium levels during pregnancy, AVP- deficient DI can
be diagnosed in women with an increased sodium level
of >140 mmol/l and inadequately diluted urine (urine
osmolality <300 mOsm/l)138. In all other cases, an over-
night water deprivation test may be considered, provided
the woman is not at risk of dehydration (severe poly-
uria, serum sodium >140 mmol/l or serum osmolality
>290 mOsm/kg). An increase in urine osmolality to
>600 mOsm/l after an overnight water deprivation test
argues against clinically relevant DI, although evidence
for this cut- off is lacking. In patients with urine osmo-
lality <600 mOsm/l, careful evaluation of patient history,
onset of symptoms and presentation is recommended for
further differentiation. Cerebral imaging by MRI dur-
ing pregnancy is only recommended if DI secondary to
haemorrhage, neoplasia or trauma is suspected4.
Diagnosis of DI in children
Once polyuria is established in children, laboratory
measurement of serum osmolality, serum sodium,
urine osmolality, urine specific gravity and potassium,
glucose and calcium is necessary to exclude diabetes
mellitus or nephrogenic DI induced by hypercalcaemia
or hypokalaemia. Concomitant serum osmolality
>300 mOsm/kg and urine osmolality <300 mOsm/kg
is indicative of DI139. By contrast, a patient with urine
osmolality >600 mOsm/kg is unlikely to have DI. If
urine osmolality is intermediate and clinical suspicion of
DI remains, a diagnosis is confirmed using a water dep-
rivation test carried out in a closely monitored medical
setting (not at home). Special caution is needed in neo-
nates, who have a high risk of dehydration. Hypertonic
saline tests in children are not recommended (of note,
the algorithm in FIG.5 is not validated for children).
In children with central DI, hormonal deficiencies
or excess of other pituitary axes must be assessed and
MRI of the sella carried out (see above). In children
with nephrogenic DI, medication history must be
evaluated and electrolyte abnormalities excluded and
a search for underlying acute or chronic renal diseases
conducted. For both central DI140 and nephrogenic DI,
if symptom onset occurs in early childhood, congenital
causes must be evaluated even if most cases are idio-
pathic, especially in the absence of a family history.
In children with primary polydipsia, careful psychiat-
ric evaluation and medication history are important.
Hypothalamic diseases that could lower thirst thresh-
old must be taken into account. A child may also
habitually drink large volumes of water without any
organiccause140.
Prevention
Currently, most forms of DI cannot be prevented. The
incidence of postoperative DI seems to be dependent
mainly on hospital and surgeon case- load, suggesting
that greater experience leads to lower rates of postop-
erative DI141. To date, the prevalence of postsurgical DI
seems to be similar for endoscopic trans- sphenoidal
surgery and microscopic trans- sphenoidal surgery of
large pituitary adenomas142. Perioperative hydrocorti-
sone treatment influences the rate of postoperative DI143.
Administration of hydrocortisone doses lower than the
usual institution’s standard protocols led to almost 50%
lower incidence of DI, possibly owing to suppression of
AVP release by hydrocortisone. Prevention of lithium-
induced nephrogenic DI is an important aspect of the
treatment of affective disorders. In patients receiving
long- term lithium treatment, nephrogenic DI seems
to only be partially reversible after discontinuation of
lithium144. Close monitoring of lithium treatment is
recommended, including annual measurement of the
urinary volume per day to make both the patient and
the physician aware of the development of drug- induced
nephrogenic DI. As gestational DI is rare and there is
no straightforward diagnostic measure for this disorder,
screening in pregnancy is not helpful.
Management
The general goals of treatment for all forms of DI include
correcting pre- existing water deficits and reducing ongo-
ing excessive water loss through urination. Thespe-
cific therapy that is required will depend on the type
of DI and the clinical circumstances. Management of
primary polydipsia entails different challenges and
solutions because therapies are primarily based on
12
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();

behavioural interventions rather than biological and
pharmacological interventions.
Correction of body water deficits
Untreated central and nephrogenic DI often leads to
hyperosmolar dehydration. In a hyperosmolar patient,
the total body water deficit can be estimated using the
following formula:
.∕
Totalbodywater deficit
=06×premorbidweight×
(
1−140
[
Na
])
+
where [Na+] is the serum sodium concentration in
millimoles per litre and weight is in kilograms.
To reduce the risk of brain damage from prolonged
exposure to severe hyperosmolality, in adults, plasma
osmolality should be lowered over the first 24 h of ther-
apy by replacing ~50% of the calculated free water deficit.
Physiologically, neurons increase intracellular osmolality
by increasing the cellular content of organic osmolytes to
protect against excessive osmotic shrinkage during pro-
longed hyperosmolality145. However, once synthesized,
these osmolytes cannot be immediately dissipated, so
correction to a normal plasma osmolality should be
spread over the subsequent 24–72 h to avoid cerebral
oedema from osmotic water shifts into the brain during
treatment146,147. This approach is particularly important
in children, as multiple studies in children have demon-
strated that limiting correction of hypernatraemia to a
maximum rate of 0.5 mmol/l/h prevents symptomatic
cerebral oedema and seizures148,149.
The choice of appropriate fluid replacement is cru-
cial, as treatment of hyperosmolar dehydration with
isotonic saline is dangerous because it can result in
worsened hypernatraemia150. In patients with central
or nephrogenic DI, the urine is essentially pure water.
A child weighing 10 kg has an estimated 7 l of total body
water. Administration of 1 l of isotonic saline (154 mmol
Na+) with excretion of 1 l of hypotonic urine containing
10 mmol Na+ will result in retention of 144 mmol Na+,
and thus will increase serum sodium concentration by
20 mmol/l (144 mmol/7 l). In these patients, isotonic
fluids should only be administered for acute intravas-
cular volume expansion in those with hypovolaemic
shock, which is an exceptionally rare complication, as
extracellular fluid volume is usually preserved with
hyperosmolality. Patients with DI should be treated
with hypotonic fluids, either milk or water consumed
enterally or, if required, 5% dextrose in water adminis-
tered intravenously. The administration of hypotonic
fluids as an intravenous bolus is not recommended;
instead, the infusion rate should be adjusted to exceed
the hourly urine output by an amount necessary to
achieve the desired reduction in the calculated total
body water deficit. The aim is to provide just enough
water to safely normalize serum sodium concentra-
tion at a rate of <0.5 mmol/l/h (<10–12 mmol/l/day)151
or even slower so as to prevent cerebral oedema and,
potentially, death147. As 5% dextrose in water provides
no osmotic load, urine output can decrease substantially,
highlighting the importance of monitoring fluid balance
to avoid rapid swings in serum sodium concentration.
Frequent, careful monitoring of the clinical condition
and biochemistry is crucial for safe treatment and
requires a clinical setting with the necessary experience
in treating complicated electrolyte disorders.
To enable fluid intake to be correctly regulated by
thirst physiology, oral consumption of fluids should
begin as soon as feasible. In most patients with DI,
thirst remains intact and patients will drink sufficient
fluid to maintain a fairly normal fluid balance. Specific
treatments for different types of DI are discussed
separatelybelow.
Central DI
Patients with central DI should be treated to reduce
polyuria and polydipsia to levels that allow maintenance
of a normal lifestyle. As the goal of therapy is improved
symptomatology, the prescribed regimen should be indi-
vidually tailored to individual patients to address their
needs. The safety of the therapeutic regimen and avoid-
ance of detrimental effects of overtreatment are primary
considerations, as in most patients, central DI has a fairly
benign course.
Fluid administration. Patients with central DI will
develop thirst when the plasma osmolality increases by
2–3%1, unless the hypothalamic osmoreceptors are also
affected by the primary lesion that causes adipsic DI.
Consequently, severe hyperosmolality is not a riskin
patients who are alert, ambulatory and able to drink
inresponse to perceived thirst. Although inconveni-
ent and lifestyle- disrupting, polyuria and polydipsia
are not life- threatening. However, hyponatraemia does
not cause specific symptoms initially and can quickly
progress to more symptomatic levels if fluid intake
continues during continuous antidiuresis. Therefore,
treatment of central DI should be designed to minimize
polyuria and polydipsia without causing undue risk of
hyponatraemia as a result of overtreatment.
Pharmacological therapy. Although different agents
have been used in the past (for example, chlorpropa-
mide and pitressin tannate), desmopressin is the current
standard of care for patients with central DI152, owing
to its long half- life, selectivity for AVPR2 and the avail-
ability of multiple preparations. The optimal dose and
dosing intervals should be determined for each patient.
Oral preparations provide greater convenience and are
usually preferred by patients. However, starting with
a nasal spray initially is preferable because of greater
consistency of absorption and physiological effect, after
which the patient can be switched to an oral prepara-
tion. After trying both preparations, patients can then
choose which they prefer for long- term treatment. The
duration of action of individual doses should be ascer-
tained in each patient owing to variability in responses
between patients153. A satisfactory schedule can generally
be determined using modest doses of desmopressin. The
maximum dose of desmopressin required rarely exceeds
0.2 mg orally, 120 µg sublingually or 10 µg (one nasal
spray) two or three times daily. These doses usually
produce plasma desmopressin levels higher than those
required to cause maximum antidiuresis but reduce the
13NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
Primer
0123456789();

need for more frequent treatment154. Once- daily dosing
can sometimes suffice, although this is rare. In some
patients, the effect of intranasal or oral desmopressin
is erratic, due to interference with absorption from the
gastrointestinal tract or nasal mucosa. Administration of
oral desmopressin on an empty stomach155 or intranasal
desmopressin after cleansing of the nostrils can reduce
this variability and prolong the duration of action.
Desmopressin resistance due to antibody production
has not been reported.
Hyponatraemia is the major complication of desmo-
pressin therapy — a 27% incidence of mild hyponatrae-
mia (serum Na+ 131–134 mmol/l) and a 15% incidence
of more severe hyponatraemia (serum Na+ ≤130 mmol/l)
have been reported after long- term follow- up of patients
with chronic central DI156. Hyponatraemia usually occurs
if the patient is continually antidiuretic while continuing
normal fluid intake. Severe hyponatraemia from desmo-
pressin treatment can be avoided by frequent monitoring
of serum electrolyte levels during initiation of therapy.
Patients who develop a low serum sodium concentration
and do not respond to recommended decreases in fluid
intake should be directed to delay a scheduled dose of
desmopressin once or twice weekly until polyuria recurs,
thereby allowing excess retained fluid to be excreted.
Because desmopressin- induced hyponatraemia is usually
chronic (>48 h duration), care must be taken in acutely
treating these patients to avoid osmotic demyelination
syndrome (ODS), a demyelinating disease of motor
neurons that occurs when correction of serum sodium
levels occurs too quickly. Current guidelines recommend
limiting corrections to <12 mmol/l in the first 24 h and
<18 mmol/l in the first 48 h (REF.157). Because cessation of
desmopressin results in a rapid water diuresis (‘aquaresis’)
once the drug is excreted by the kidneys, these patients
can correct their hyponatraemia exceedingly quickly,
putting them at high risk of ODS. Consequently, some
authors recommend continuing to administer desmo-
pressin while correcting the hyponatraemia at a con-
trolled rate using hypertonic (3%) NaCl158. Alternatively,
desmopressin can be re- administered to shut off an ongo-
ing aquaresis once a desired correction of serum sodium
(6–8 mmol/l) has been achieved. Whichever method is
chosen, these patients must be monitored closely to avoid
potentially catastrophic outcomes159.
Central DI occurs frequently after surgery in the
suprasellar region of the hypothalamus160. After con-
firmation of a central DI diagnosis, the best pharma-
cological therapy is desmopressin. However, because
water overload with subsequent brain oedema is a con-
cern after this type of surgery, treatment with oral or
intravenous fluid replacement alone for long periods
oftime sometimes precedes the initiation of desmo-
pressin therapy. If the patient is awake and responds to
thirst, then thirst is a sufficient guide for water replace-
ment. However, fluid balance must be maintained using
intravenous fluids if the patient cannot respond to
thirst because of a decreased level of consciousness or
from damage to the hypothalamic thirst centre. Urine
osmolality and serum sodium concentration should
be checked every 4–6 h during initial therapy and then
daily until stabilization or resolution of the DI. Caution
is warranted regarding the volume of water replacement,
as administration of excess water during continued
administration of AVP or desmopressin can potentially
cause hyponatraemia. Studies in animals suggest that
desmopressin- induced hyponatraemia impairs survival
of AVP- producing neurons after pituitary stalk com-
pression161, suggesting that over- hydration resulting in
decreased stimulation of neurohypophyseal neurons
might increase the probability of permanent DI.
Postoperatively, desmopressin can be administered
parenterally (subcutaneously, intramuscularly or intra-
venously). Intravenous administration is generally pre-
ferred, as it precludes concerns about absorption, does
not have significant pressor activity and the total dura-
tion of action is the same as with the other parenteral
routes. The antidiuretic effect of desmopressin should
occur promptly and typically lasts 6–12 h. Urine osmo-
lality and volume should be monitored to ascertain
whether the dose was effective and serum sodium con-
centration measured frequently (every 4–6 h) to ensure
improvement in hypernatraemia. Allowing a return
of polyuria is advisable before administration of addi-
tional doses of desmopressin because postoperative DI
is often transient, and return of endogenous AVP secre-
tion will become apparent by the absence of return of
the polyuria. In addition, transient postoperative DI is
sometimes part of a triphasic pattern following pituitary
stalk transection (discussed above). Therefore, allowing
recurrence of polyuria before re- dosing with desmo-
pressin will enable earlier detection of a potential sec-
ond phase of inappropriate antidiuresis and reduce the
probability of severe hyponatraemia occurring as a result
of continuing antidiuretic therapy and intravenous fluid
administration when it is no longer required162.
Patients with hypernatraemia due to osmoreceptor
dysfunction (adipsic central DI) should be treated
acutely with the same treatment as any hyperosmolar
patient. The long- term management of osmoreceptor
dysfunction syndromes requires that potentially treat-
able causes are investigated thoroughly, with measures to
prevent dehydration instituted at the same time. Because
hypodipsia cannot be cured, although spontaneous
improvement occurs rarely, education of the patient and
the patient’s family about the importance of regulating
fluid intake according to hydration status is the focus of
management163. This can be accomplished most effica-
ciously by establishing a daily schedule of fluid intake
regardless of the patient’s thirst, which can be adjusted in
response to changes in body weight164. As these patients
will not drink spontaneously, daily fluid intake must be
prescribed. If the patient has polyuria, desmopressin
should also be prescribed, as in any patient with central
DI. The success of fluid prescription should be moni-
tored periodically (weekly at first, later every month,
depending on the stability of the patient) by measur-
ing serum sodium concentration. In addition, periodic
recalculation of the target weight (at which hydration
status and serum sodium concentration are normal)
might be required to account for growth in children or
body fat changes in adults.
Guidelines for pharmacological therapy of central DI
in paediatric patients are not substantially different from
14
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();

those in adults, except that oral or intranasal administra-
tion may be more difficult in very young patients, who
may need to be treated with parenteral (subcutaneous)
desmopressin for a period of time.
Nephrogenic DI
As bypassing a non- functional AVPR2 receptor or
inserting functional water channels into the basolateral
membrane of principal cells of the collecting duct are
not presently feasible, current approaches for treatment
of hereditary nephrogenic DI focus on ameliorating the
symptoms instead of curing the disease. Treatment of
acquired nephrogenic DI should target the underly-
ing cause; for example, relief of urinary obstruction or
amiloride therapy in lithium- associated nephrogenic
DI. If these approaches are not possible, then treat-
ment of acquired nephrogenic DI is the same as that for
hereditary nephrogenic DI.
Diet. Dieticians have a crucial role in managing patients
with nephrogenic DI in the first year of life, when fluid
intake and caloric consumption are coupled. Osmotic
load should be minimized while ensuring recommended
caloric and protein intake to enable normal growth and
development. As 1 g of table salt provides an osmolar
load of ~34 mOsm (17 mOsm Na+ and 17 mOsm Cl−),
the obligatory urine output in a patient with a urine
osmolality of 100 mOsm/kg is increased by 340 ml for
each gram of salt ingested. The dietary osmolar load can
be estimated by multiplying the millimolar amounts
of sodium and potassium by two (to account for the
accompanying anions) and adding the gram amount of
protein multiplied by four.
Pharmacological therapy. For patients on long- term
lithium therapy, amiloride prevents uptake of lithium in
the collecting duct epithelial cells and thus the inhib-
itory effects of intracellular lithium on water trans-
port165. Hydrochlorothiazide has been shown to reduce
urine output in both central and nephrogenic DI166,167.
Thiazides decrease salt reabsorption by inhibiting the
thiazide- sensitive co- transporter SLC12A3 in the distal
tubule. The sodium loss reduces plasma volume, sothat
less water is presented to the collecting duct and lost
in the urine. Also, hydrochlorothiazide administra-
tion reduced urine volume in Slc12a3-knockout mice
with lithium- induced nephrogenic DI, suggesting a
SLC12A3-independent mechanism of thiazide- mediated
reduction in urine output168. Furthermore, inhibition
of carbonic anhydrase by hydrochlorothiazide in the
proximal tubule might reduce proximal sodium uptake
and, via tubulo- glomerular feedback, reduce glomerular
filtration. The carbonic anhydrase inhibitor acetazola-
mide reduces inulin clearance and cortical expression
of sodium/hydrogen exchanger 3 and attenuates the
increased urinary PGE2 levels observed in mice with
lithium- induced nephrogenic DI169, and is effective
in humans with lithium- induced nephrogenic DI170.
Polyuria in these patients is usually moderate (<6 l/day)
and can be decreased by a strict clamping of the plasma
lithium concentration at 0.8 mEq/l, a low- sodium diet
and amiloride administration. Frequent plasma lithium
measurements should be made when instituting a low-
sodium diet and amiloride treatment, as an increase
in lithaemia might be observed with the necessity to
immediately decrease the lithium dosage.
In an animal model of nephrogenic DI, the use of
the NSAID indomethacin reduced water diuresis inde-
pendently of AVP171. A similar effect of prostaglandin
synthesis inhibitors was later reported in patients with
nephrogenic DI172. Since these early studies, prosta-
glandin synthesis inhibitors have become an essential
component of the management of nephrogenic DI, par-
ticularly in the first years of life when management is
most complicated. These drugs can have quite marked
effects when first administered. Indeed, rapid reduction
in plasma sodium levels following the initiation of indo-
methacin and hydrochlorothiazide therapy can induce
hyponatraemic seizures173. Patients with hereditary
partial nephrogenic DI typically carry mutations that
result in partial function of either AVPR2 or AQP2, and
their urine osmolality may increase after desmopressin
treatment174,175.
Hypercalcaemic and hypokalaemic nephrogenic DI
manifestations are usually of mild to moderate sever-
ity and are easily reversed by normalization of plasma
calcium or potassium levels.
Primary polydipsia
Management of primary polydipsia depends on the
psychological profile of the patient and whether there is
concurrent hyponatraemia. Often, no effective treatment
is available, leaving patients at risk of structural urinary
tract abnormalities and other medical complications that
may contribute to a reduced lifespan.
Pharmacological therapy. The most acute danger of
primary polydipsia is episodic water intoxication,
which occurs in psychotic patients with PIP. This sub-
set of polydipsic patients exhibits a reset osmostat for
AVP that drops to levels consistent with water intoxi-
cation during acute psychotic episodes. Symptomatic
hyponatraemia in other patients with polydipsia usu-
ally occurs because of medications that impair renal
diluting capacity. AVPR2 antagonists rapidly normalize
serum sodium concentration in these patients, but they
require close monitoring to avoid dehydration and renal
damage176. Unlike other antipsychotic agents, clozapine
seems to normalize sodium levels in patients with PIP
who are at risk of water intoxication, although none of
the published studies were placebo- controlled and this
drug carries unique, potentially life- threatening risks12.
Clozapine’s effects seem to be attributable to decreased
fluid intake rather than increased fluid excretion177, and
thus may prove effective in normonatraemic psychotic
polydipsic patients as well, assuming that the risks
arejustified.
Many other treatments have been inconclusive or
ineffective in treating polydipsia or impaired water
excretion in psychotic patients, including consuming
electrolyte- containing beverages, reducing the dose or
switching to another antipsychotic agent, and adding
angiotensin inhibitors, α- adrenergic or β- adrenergic
receptor antagonists, or opioid antagonists178. Case
15NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
Primer
0123456789();

reports suggest that acetazolamide can reduce poly-
dipsia, and these observations may warrant further
study179. Unlike psychotic disorders, CWD typically
resolves with successful pharmacological treatment of
the core psychiatric symptoms (such as depression and
anxiety). Desmopressin is generally contraindicated in
patients with primary polydipsia because of increased
risk of hyponatraemia. However, in patients with dipso-
genic DI, the water intake in at least some patients
seems to be driven primarily by a reduced osmotic
threshold for thirst, and desmopressin treatment has
been successful, presumably by reducing plasma osmo-
lality below the set point for thirst. Of particular interest
is a case of central DI in a patient with schizophrenia
whose polydipsia was successfully treated with desmo-
pressin following behavioural therapy of concurrent
primary polydipsia180.
Behavioural therapy. Monitoring diurnal weight gain
can prevent water intoxication in patients with PIP.
Based on pre- determined increases that are predictive
of incipient water intoxication (generally 5–8 kg increase
in body weight) or the presence of prodromal signs, a
brief fluid restriction (1–3 h) is imposed; this procedure
should only be carried out in closely monitored settings
because of the risk of over- correction in patients with
chronic hyponatraemia. Standard behavioural treat-
ments, such as relaxation, response prevention, cognitive
behavioural therapy and token economies, have been
used successfully to supplement medication in psychiat-
ric patients or as stand- alone treatments in those without
a psychiatric disorder, and would most likely be effective
in individuals without an Axis I psychiatric disorder who
have become habitual compulsive water drinkers.
Diet. Ice chips can help reduce polydipsia, perhaps by
acting through recognized thermoregulatory mecha-
nisms and specifically oropharyngeal influences on the
subfornical organ in the lamina terminalis93. In some
patients with primary polydipsia, hard candies (such as
lemon drops) can be used to increase salivary flow and
decrease the dry mouth sensation that leads to increased
fluid ingestion.
Gestational DI
Once diagnosis of gestational DI is confirmed, treatment
with desmopressin is indicated, regardless of whether DI
is permanent or transient152. Although trial data are lack-
ing, no adverse maternal or fetal effects from desmopres-
sin have been reported181. Despite structural similarities
to oxytocin, desmopressin administered intranasally
had no effect on induction of labour15. Desmopressin is
not affected by the placental vasopressinase181. Initiating
treatment with the lowest desmopressin dosage at bed
time to prevent nocturia is recommended. The dose
isthen slowly titrated upwards according to symp-
toms and under regular control of serum sodium levels
(target 133–140 mmol/l)181. The required desmopressin
dose might be somewhat higher than in non- pregnant
patients, as the placental vasopressinase metabolizes any
endogenous AVP. Caution is advisable in patients with
impaired thirst sensation, as they may require a fixed
daily fluid intake to avoid marked changes in plasma
sodium concentration. After delivery, desmopressin can
be stopped within days to weeks in patients with tran-
sient DI or reduced to the pre- pregnancy dose in those
with permanent DI15. Desmopressin can be safely given
during breastfeeding182.
Quality of life
Formal assessment of QOL is limited to two studies in
patients with central DI183,184, although other QOL stud-
ies have included patients with DI who have traumatic
brain injury and concurrent hypopituitarism185. Isolate d
and treated central DI is associated with a fairly normal
QOL in both children184 and adults183, particularly when
oral desmopressin is prescribed. Indeed, in patients with
diminished QOL, the dose or timing of desmopressin
treatment should probably be adjusted.
The Nagasaki Diabetes Insipidus Questionnaire,
which consists of 12 questions, ten of which relate
directly to central DI symptoms (liquid intake and uri-
nation, and limitations of daily life) and two to the effect
of desmopressin and the patient’s satisfaction with des-
mopressin, seems to provide a reliable measure of QOL
in patients with central DI. Studies using this question-
naire revealed that concerns expressed by untreated
patients about polyuria and heightened thirst in public
settings consistently diminish when they receive nasal
desmopressin, and when they are switched to an oral
formulation183. Additional QOL benefits were reported
in a subset of patients who lost weight when switchedto
oral desmopressin, presumably because of reduced
polydipsia- induced weight gain186. Although some
patients noted the inconvenience of not being able to eat
with oral desmopressin, it did not lead to resumption of
nasal treatment. Conversely, patients who were satisfied
with nasal desmopressin showed no further benefit from
switching to oral desmopressin.
The salutary effect of desmopressin on nocturia and
incontinence in some patients with CWD, dipsogenic
DI or partial central DI probably improves their QOL187,
as does the reported ability of clozapine to allow psy-
chotic patients with PIP to be discharged from long- term
inpatient facilities177.
Outlook
Owing to their rarity, DI and primary polydipsia are
fairly neglected disorders, in terms of both their inclu-
sion in the medical education curriculum and research
efforts to improve diagnosis and treatment. The preva-
lence of CWD is increasing in the general population14;
which might be prevented or limited by greater aware-
ness of the tendency of some individuals to take extreme
measures when worried about their physical health.
Although many of the underlying mechanisms lead-
ing to acquired or hereditary DI or primary polydipsia
are known, idiopathic forms of AVP deficiency are a
large pathogenetic category in central DI, which in most
cases results from autoimmune destruction of the neuro-
hypophysis, and better classification and evaluation of
these patients is needed. Furthermore, the increasing
use of immune checkpoint blockade in patients with
cancer has led to increasing incidence of hypophysitis
16
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();

with mainly secondary adrenal insufficiency, second-
ary hypothyroidism and secondary hypogonadism188,
whereas only a few cases of posterior pituitary gland
involvement have been reported189. The expected further
increase in the use of immune checkpoint inhibitors in
the coming years will need to be monitored.
In hereditary central DI, the genes responsible for
autosomal recessive FNDI type b, autosomal recessive
FNDI type d and X- linked recessive FNDI have yet to
be identified57. Furthermore, in hereditary nephrogenic
DI, questions remain about mechanisms of apparent
AVPR2-dependent but cAMP- independent AQP2 traf-
ficking, and the role of non- vasopressin modulators of
AQP2 trafficking, such as prostaglandin E2, nitric oxide
and adenosine.
Primary polydipsia undoubtedly encompasses a num-
ber of distinct disorders arising from learned behaviours,
which induce anticipatory drinking or states of height-
ened arousal that are ameliorated by non- regulatory
water drinking. Technological advances have enabled
dissection of the components of motivational and
affective influences on the homeostatic systems in the
anterior hypothalamus, which regulate drinking behav-
iour and AVP secretion. We expect that the pathophys-
iology of these disorders and their relationship to one
another and to other homeostatic and non- homeostatic
behaviours will be revealed in the comingyears.
For decades, the differential diagnosis of polyuria–
polydipsia syndrome has been based on the indirect
water deprivation test. In 2018, hypertonic saline infu-
sion with copeptin measurement emerged as a new test
with higher diagnostic accuracy, and was proposed as
the new gold standard102. However, alternative tests
should be explored because adverse effects of this test
are quite common and constant surveillance and close
moni toring is needed during the test to prevent a marked
increase in serum sodium levels. A promising alterna-
tive is arginine infusion (followed by serum copeptin
measurement), as it stimulates the release of various
hormones, such as growth hormone and prolactin, from
the anterior pituitary gland190,191 and is a standard test to
evaluate suspected growth hormone deficiency in chil-
dren and adults. Arginine also stimulates the posterior
pituitary gland, with a consequent increase in copeptin
levels (M.C.-C., unpublished data), and the test differ-
entiates between central DI and primary polydipsia with
high diagnostic accuracy192. Further advantages include
a possible lower risk of adverse effects of infusion com-
pared with 3% saline infusion and practicality, even
outside the hospital setting.
Finally, whereas central DI treatment is usually
straightforward, treatment of nephrogenic DI is chal-
lenging. Novel treatments include non- peptide AVPR2
agonists or antagonists that act as molecular chaper-
ones to rescue misfolded AVPR2 receptors (reviewed
elsewhere3). Potential mechanism- based therapies for
nephrogenic DI caused by AVPR2 mutations include
pharmacological chaperones to partially correct cellu-
lar misprocessing of mutant AVPR2 (REF.193), and drugs
to bypass AVPR2, including prostaglandin receptor
inhibitors, β3 adrenoreceptor agonists, secretin recep-
tor agonists, cGMP phosphodiesterase inhibitors and
others194–197. Mechanism- based therapy of nephrogenic
DI caused by AQP2 mutations is challenging because
AQP2-mediated water permeability is required for phys-
iological function of collecting ducts and thus cannot
be bypassed. Chemical and molecular chaperones have
been proposed as potential therapeutic agents for reces-
sive nephrogenic DI caused by mutations that impair
cellular trafficking of AQP2 (REF.198).
Published online xx xx xxxx
1. Robertson, G. L. Diabetes insipidus. Endocrinol.
Metab. Clin. North Am. 24, 549–572 (1995).
2. Babey, M., Kopp, P. & Robertson, G. L. Familial forms of
diabetes insipidus: clinical and molecular characteristics.
Nat. Rev. Endocrinol. 7, 701–714 (2011).
3. Bockenhauer, D. & Bichet, D. G. Pathophysiology,
diagnosis and management of nephrogenic diabetes
insipidus. Nat. Rev. Nephrol. 11, 576–588 (2015).
This article is a detailed review of nephrogenic DI,
including the importance of early genetic testing
and clinical management.
4. Fenske, W. & Allolio, B. Clinical review: current state
and future perspectives in the diagnosis of diabetes
insipidus: a clinical review. J. Clin. Endocrinol. Metab.
97, 3426–3437 (2012).
5. Carter, A. C. & Robbins, J. The use of hypertonic saline
infusions in the differential diagnosis of diabetes
insipidus and psychogenic polydipsia. J. Clin. Endocrinol.
Metab. 7, 753–766 (1947).
6. Fenske, W. etal. Copeptin in the differential diagnosis
of the polydipsia- polyuria syndrome — revisiting the
direct and indirect water deprivation tests. J. Clin.
Endocrinol. Metab. 96, 1506–1515 (2011).
7. Di Iorgi, N. etal. Diabetes insipidus — diagnosis
and management. Horm. Res. Paediatr. 77, 69–84
(2012).
8. Saifan, C. etal. Diabetes insipidus: a challenging
diagnosis with new drug therapies. ISRN Nephrol.
2013, 797620 (2013).
9. Maghnie, M. etal. Central diabetes insipidus in
children and young adults. N. Engl. J. Med. 343,
998–1007 (2000).
10. Arthus, M. F. etal. Report of 33 novel AVPR2
mutations and analysis of 117 families with X- linked
nephrogenic diabetes insipidus. J. Am. Soc. Nephrol.
11, 1044–1054 (2000).
11. Mercier- Guidez, E. & Loas, G. Polydipsia and
water intoxication in 353 psychiatric inpatients:
an epidemiological and psychopathological study.
Eur.Psychiatry 15, 306–311 (2000).
12. Vieweg, W. V. Treatment strategies in the polydipsia-
hyponatremia syndrome. J. Clin. Psychiatry 55,
154–160 (1994).
13. Sailer, C. O. etal. Characteristics and outcomes of
patients with profound hyponatraemia due to primary
polydipsia. Clin. Endocrinol. 87, 492–499 (2017).
14. Benton, D. etal. Executive summary and conclusions
from the European Hydration Institute Expert
Con ference on human hydration, health, and
performance. Nutr. Rev. 73 (Suppl. 2), 148–150 (2015).
15. Ananthakrishnan, S. Diabetes insipidus during
pregnancy. Best Pract. Res. Clin. Endocrinol. Metab.
30, 305–315 (2016).
16. Durr, J. A. & Lindheimer, M. D. Diagnosis and
management of diabetes insipidus during pregnancy.
Endocr. Pract. 2, 353–361 (1996).
17. Clark, A. J. etal. Treatment- related morbidity and
the management of pediatric craniopharyngioma:
a systematic review. J. Neurosurg. Pediatr. 10,
293–301 (2012).
18. Schreckinger, M. etal. Post- operative diabetes
insipidus after endoscopic transsphenoidal surgery.
Pituitary 16, 445–451 (2013).
19. Laws, E. R. Jr etal. A benchmark for preservation
of normal pituitary function after endoscopic
transsphenoidal surgery for pituitary macroadenomas.
World Neurosurg. 91, 371–375 (2016).
20. Imura, H. etal. Lymphocytic infundibuloneuro-
hypophysitis as a cause of central diabetes insipidus.
N. Engl. J. Med. 329, 683–689 (1993).
21. Masaki, Y. etal. Proposal for a new clinical entity,
IgG4-positive multiorgan lymphoproliferative
syndrome: analysis of 64 cases of IgG4-related
disorders. Ann. Rheum. Dis. 68, 1310–1315 (2009).
22. Iwama, S. etal. Rabphilin-3A as a targeted autoantigen
in lymphocytic infundibulo- neurohypophysitis. J. Clin.
Endocrinol. Metab. 100, E946–E954 (2015).
23. Yasuda, Y. etal. Critical role of rabphilin-3A in
the pathophysiology of experimental lymphocytic
neurohypophysitis. J. Pathol. 244, 469–478 (2018).
24. Ohlund, L. etal. Reasons for lithium discontinuation in
men and women with bipolar disorder: a retrospective
cohort study. BMC Psychiatry 18, 37 (2018).
25. Millson, R. C., Koczapski, A. B., Cook, M. I.
&Daszkiewicz, M. A survey of patient attitudes toward
self- induced water intoxication. Can. J. Psychiatry 37,
46–47 (1992).
This article reports the reasons that patients with
schizophrenia and PIP provide for their excess
water intake.
26. May, D. L. Patient perceptions of self- induced water
intoxication. Arch. Psychiatr. Nurs. 9, 295–304 (1995).
27. de Leon, J., Tracy, J., McCann, E. & McGrory, A.
Polydipsia and schizophrenia in a psychiatric hospital:
a replication study. Schizophr. Res. 57, 293–301
(2002).
28. Ahmed, A. G., Heigh, L. M. & Ramachandran, K. V.
Polydipsia, psychosis, and familial psychopathology.
Can. J. Psychiatry 46, 522–527 (2001).
29. Shutty, M. S. Jr., McCulley, K. & Pigott, B. Association
between stereotypic behavior and polydipsia in
chronic schizophrenic patients. J. Behav. Ther. Exp.
Psychiatry 26, 339–343 (1995).
30. Atsariyasing, W. & Goldman, M. B. A systematic
review of the ability of urine concentration to
distinguish antipsychotic- from psychosis- induced
hyponatremia. Psychiatry Res. 217, 129–133
(2014).
17NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
Primer
0123456789();

31. Barlow, E. D. & De Wardener, H. E. Compulsive water
drinking. Q. J. Med. 28, 235–258 (1959).
32. McKenna, K. & Thompson, C. Osmoregulation
in clinical disorders of thirst appreciation.
Clin. Endocrinol. 49, 139–152 (1998).
This review summarizes the osmoregulatory
disruption in desire for water and AVP secretion
inindividuals with primary polydipsia.
33. Sailer, C., Winzeler, B. & Christ- Crain, M. Primary
polydipsia in the medical and psychiatric patient:
characteristics, complications and therapy. Swiss Med.
Wkly 147, w14514 (2017).
34. Robertson, G. L. Dipsogenic diabetes insipidus:
a newly recognized syndrome caused by a selective
defect in the osmoregulation of thirst. Trans. Assoc.
Am. Physicians 100, 241–249 (1987).
35. Robertson, G. L. Diabetes insipidus: differential
diagnosis and management. Best Pract. Res. Clin.
Endocrinol. Metab. 30, 205–218 (2016).
36. Davison, J. M., Sheills, E. A., Philips, P. R.,
Barron, W. M. & Lindheimer, M. D. Metabolic
clearance of vasopressin and an analogue resistant to
vasopressinase in human pregnancy. Am. J. Physiol.
264, F348–F353 (1993).
37. Kennedy, S., Hall, P. M., Seymour, A. E. & Hague, W. M.
Transient diabetes insipidus and acute fatty liver
of pregnancy. Br. J. Obstet. Gynaecol. 101, 387–391
(1994).
38. Aulinas, A. etal. Low plasma oxytocin levels and
increased psychopathology in hypopituitary men with
diabetes insipidus. J. Clin. Endocrinol. Metab. 104,
3181–3191 (2019).
39. Heinbecker, P. & White, H. Hypothalamico-
hypophyseal system and its relation to water balance
in the dog. Am. J. Physiol. 133, 582–593 (1941).
40. Maccubbin, D. A. & Vanburen, J. M. A quantitative
evaluation of hypothalamic degeneration and its
relation to diabetes insipidus following interruption
of the human hypophyseal stalk. Brain 86, 443–464
(1963).
41. Lipsett, M. B., Maclean, J. P., West, C. D., Li, M. C. &
Pearson, O. H. An analysis of the polyuria induced by
hypophysectomy in man. J. Clin. Endocrinol. Metab.
16, 183–195 (1956).
42. Hollinshead, W. H. The interphase of diabetes
insipidus. Mayo Clin. Proc. 39, 92–100 (1964).
43. Verbalis, J. G., Robbins, A. G. & Moses, A. M. in
Diabetes Insipidus in Man (eds Czernichow, P. &
Robinson, A. G.) 247–265 (Karger, Basel, 1984).
44. Di Iorgi, N. etal. Central diabetes insipidus in children
and young adults: etiological diagnosis and long- term
outcome of idiopathic cases. J. Clin. Endocrinol.
Metab. 99, 1264–1272 (2014).
45. Kojima, H. etal. Diabetes insipidus caused by
lymphocytic infundibuloneurohypophysitis.
Arch. Pathol. Lab Med. 113, 1399–1401 (1989).
46. Maghnie, M. etal. Idiopathic central diabetes
insipidus in children and young adults is commonly
associated with vasopressin- cell antibodies and
markers of autoimmunity. Clin. Endocrinol. 65,
470–478 (2006).
47. Shimatsu, A., Oki, Y., Fujisawa, I. & Sano, T. Pituitary
and stalk lesions (infundibulo- hypophysitis) associated
with immunoglobulin G4-related systemic disease:
an emerging clinical entity. Endocr. J. 56, 1033–1041
(2009).
48. Buggy, J. & Jonhson, A. K. Preoptic- hypothalamic
periventricular lesions: thirst deficits and
hypernatremia. Am. J. Physiol. 233, R44–R52
(1977).
49. Thrasher, T. N., Keil, L. C. & Ramsay, D. J. Lesions
of the organum vasculosum of the lamina terminalis
(OVLT) attenuate osmotically- induced drinking and
vasopressin secretion in the dog. Endocrinology 110,
1837–1839 (1982).
50. Matsuda, T. etal. Distinct neural mechanisms for the
control of thirst and salt appetite in the subfornical
organ. Nat. Neurosci. 20, 230–241 (2017).
51. Crowley, R. K., Sherlock, M., Agha, A., Smith, D. &
Thompson, C. J. Clinical insights into adipsic diabetes
insipidus: a large case series. Clin. Endocrinol. 66,
475–482 (2007).
52. Hiyama, T. Y. etal. Adipsic hypernatremia without
hypothalamic lesions accompanied by autoantibodies
to subfornical organ. Brain Pathol. 27, 323–331
(2017).
53. Christensen, J. H. & Rittig, S. in Genetic Diagnosis
of Endocrine Disorders 2nd edn (eds Weiss, R. E. &
Refetoff, S.) 93–101 (Elsevier, 2016).
This book chapter is a comprehensive in- depth
review of the genetic background of hereditary
central DI.
54. Pepin, L. etal. A new case of PCSK1 pathogenic
variant with congenital proprotein convertase 1/3
deficiency and literature review. J. Clin. Endocrinol.
Metab. 104, 985–993 (2019).
55. Shi, G. etal. ER- associated degradation is required
for vasopressin prohormone processing and systemic
water homeostasis. J. Clin. Invest. 127, 3897–3912
(2017).
This study highlights a pathway linking ERAD to
conformational maturation of AVP prohormone
inneuroendocrine cells, indicating the potential
significance of this pathway in the pathogenesis
ofautosomal dominant central DI.
56. Bichet, D. G. & Lussier, Y. Mice deficient for ERAD
machinery component Sel1L develop central diabetes
insipidus. J. Clin. Invest. 127, 3591–3593 (2017).
57. Habiby, R. etal. A novel X- linked form of familial
neurohypophyseal diabetes insipidus [abstract].
J. Invest. Med. 44, 388A (1996).
58. Rutishauser, J., Kopp, P., Gaskill, M. B., Kotlar, T. J.
& Robertson, G. L. Clinical and molecular analysis
of three families with autosomal dominant
neurohypophyseal diabetes insipidus associated with
a novel and recurrent mutations in the vasopressin-
neurophysin II gene. Eur. J. Endocrinol. 146,
649–656 (2002).
59. Perrotta, S. etal. Early- onset central diabetes
insipidus is associated with denovo arginine
vasopressin- neurophysin II or Wolfram syndrome 1
gene mutations. Eur. J. Endocrinol. 172, 461–472
(2015).
60. Nielsen, S. etal. Vasopressin increases water
permeability of kidney collecting duct by inducing
translocation of aquaporin- CD water channels to
plasma membrane. Proc. Natl Acad. Sci. USA 92,
1013–1017 (1995).
61. Rieg, T. etal. Adenylate cyclase 6 determines cAMP
formation and aquaporin-2 phosphorylation and
trafficking in inner medulla. J. Am. Soc. Nephrol. 21,
2059–2068 (2010).
62. Moeller, H. B., Praetorius, J., Rutzler, M. R. &
Fenton, R. A. Phosphorylation of aquaporin-2
regulates its endocytosis and protein- protein
interactions. Proc. Natl Acad. Sci. USA 107,
424–429 (2010).
63. Fushimi, K., Sasaki, S. & Marumo, F. Phosphorylation
of serine 256 is required for cAMP- dependent
regulatory exocytosis of the aquaporin-2 water
channel. J. Biol. Chem. 272, 14800–14804 (1997).
64. Boton, R., Gaviria, M. & Batlle, D. C. Prevalence,
pathogenesis, and treatment of renal dysfunction
associated with chronic lithium therapy. Am. J. Kidney
Dis. 10, 329–345 (1987).
65. Marples, D., Christensen, S., Christensen, E. I.,
Ottosen, P. D. & Nielsen, S. Lithium- induced
downregulation of aquaporin-2 water channel
expression in rat kidney medulla. J. Clin. Invest. 95,
1838–1845 (1995).
66. Christensen, B. M. etal. Changes in cellular
composition of kidney collecting duct cells in rats with
lithium- induced NDI. Am. J. Physiol. Cell Physiol. 286,
C952–C964 (2004).
67. Christensen, B. M. etal. αENaC- mediated lithium
absorption promotes nephrogenic diabetes insipidus.
J. Am. Soc. Nephrol. 22, 253–261 (2011).
68. Grunfeld, J. P. & Rossier, B. C. Lithium nephrotoxicity
revisited. Nat. Rev. Nephrol. 5, 270–276 (2009).
69. Rao, R. Glycogen synthase kinase-3 regulation of
urinary concentrating ability. Curr. Opin. Nephrol.
Hypertens. 21, 541–546 (2012).
70. Quiroz, J. A., Gould, T. D. & Manji, H. K. Molecular
effects of lithium. Mol. Interv. 4, 259–272 (2004).
71. Rao, R. etal. Lithium treatment inhibits renal GSK-3
activity and promotes cyclooxygenase 2-dependent
polyuria. Am. J. Physiol. Renal Physiol. 288,
F642–F649 (2005).
72. Frokiaer, J. etal. Downregulation of aquaporin-2
parallels changes in renal water excretion in unilateral
ureteral obstruction. Am. J. Physiol. 273, F213–F223
(1997).
73. Khositseth, S. etal. Autophagic degradation of
aquaporin-2 is an early event in hypokalemia- induced
nephrogenic diabetes insipidus. Sci. Rep. 5, 18311
(2015).
74. Khositseth, S. etal. Hypercalcemia induces targeted
autophagic degradation of aquaporin-2 at the onset
of nephrogenic diabetes insipidus. Kidney Int. 91,
1070–1087 (2017).
75. van den Ouweland, A. M. etal. Mutations in the
vasopressin type 2 receptor gene (AVPR2) associated
with nephrogenic diabetes insipidus. Nat. Genet. 2,
99–102 (1992).
76. Rosenthal, W. etal. Molecular identification
of the gene responsible for congenital nephrogenic
diabetes insipidus. Nature 359, 233–235
(1992).
77. Noda, Y., Sohara, E., Ohta, E. & Sasaki, S. Aquaporins
in kidney pathophysiology. Nat. Rev. Nephrol. 6,
168–178 (2010).
78. Deen, P. M. etal. Requirement of human renal water
channel aquaporin-2 for vasopressin- dependent
concentration of urine. Science 264, 92–95 (1994).
This paper reports the discovery that loss of
function mutations in AQP2 cause nephrogenic DI.
79. Mulders, S. M. etal. An aquaporin-2 water
channel mutant which causes autosomal dominant
nephrogenic diabetes insipidus is retained in the Golgi
complex. J. Clin. Invest. 102, 57–66 (1998).
80. Thompson, C. J., Edwards, C. R. & Baylis, P. H.
Osmotic and non- osmotic regulation of thirst and
vasopressin secretion in patients with compulsive
water drinking. Clin. Endocrinol. 35, 221–228
(1991).
81. Stuart, C. A., Neelon, F. A. & Lebovitz, H. E.
Disordered control of thirst in hypothalamic- pituitary
sarcoidosis. N. Engl. J. Med. 303, 1078–1082
(1980).
82. Ittasakul, P. & Goldman, M. B in Hyponatremia:
Evaluation and Treatment (ed. Simon, E. E.) 159–173
(Springer, New York, 2013).
83. Goldman, M. B. Brain circuit dysfunction in a distinct
subset of chronic psychotic patients. Schizophr. Res.
157, 204–213 (2014).
This review describes the mechanism of water
imbalance in polydipsic patients with schizophrenia
with and without PIP and its relationship to the
underlying psychotic disorder.
84. Ulrich- Lai, Y. M. & Herman, J. P. Neural regulation of
endocrine and autonomic stress responses. Nat. Rev.
Neurosci. 10, 397–409 (2009).
85. Lodge, D. J. & Grace, A. A. Developmental pathology,
dopamine, stress and schizophrenia. Int. J. Dev.
Neurosci. 29, 207–213 (2011).
86. Fukunaka, Y. etal. The orexin 1 receptor (HCRTR1)
gene as a susceptibility gene contributing to
polydipsia- hyponatremia in schizophrenia.
Neuromolecular Med. 9, 292–297 (2007).
87. Luchins, D. J., Goldman, M. B., Lieb, M. & Hanrahan, P.
Repetitive behaviors in chronically institutionalized
schizophrenic patients. Schizophr. Res. 8, 119–123
(1992).
88. Mittleman, G., Whishaw, I. Q., Jones, G. H., Koch, M.
& Robbins, T. W. Cortical, hippocampal, and striatal
mediation of schedule- induced behaviors. Behav.
Neurosci. 104, 399–409 (1990).
89. Navarro, S. V. etal. Behavioral biomarkers of
schizophrenia in high drinker rats: a potential
endophenotype of compulsive neuropsychiatric
disorders. Schizophr. Bull. 43, 778–787 (2017).
90. Umbricht, D. Polydipsia and hippocampal pathology.
Biol. Psychiatry 36, 709–710 (1994).
91. Didriksen, M., Olsen, G. M. & Christensen, A. V. Effect
of clozapine upon schedule- induced polydipsia (SIP)
resembles neither the actions of dopamine D1 nor D2
blockade. Psychopharmacology 113 , 250–256
(1993).
92. Augustine, V. etal. Hierarchical neural architecture
underlying thirst regulation. Nature 555, 204–209
(2018).
93. Zimmerman, C. A., Leib, D. E. & Knight, Z. A.
Neural circuits underlying thirst and fluid homeostasis.
Nat. Rev. Neurosci. 18, 459–469 (2017).
This review summarizes the evidence for the role
ofindividual populations of neurons in the lamina
terminalis in modulating pre- ingestive, pre- systemic
and homeostatic influences on water intake and
excretion.
94. Leib, D. E. etal. The forebrain thirst circuit drives
drinking through negative reinforcement. Neuron
96, 1272–1281 (2017).
95. Hsu, T. M., McCutcheon, J. E. & Roitman, M. F.
Parallels and overlap: the integration of homeostatic
signals by mesolimbic dopamine neurons. Front.
Psychiatry 9, 410 (2018).
96. Barron, W. M. etal. Transient vasopressin- resistant
diabetes insipidus of pregnancy. N. Engl. J. Med. 310,
442–444 (1984).
97. Durr, J. A., Hoggard, J. G., Hunt, J. M. & Schrier, R. W.
Diabetes insipidus in pregnancy associated with
abnormally high circulating vasopressinase activity.
N. Engl. J. Med. 316, 1070–1074 (1987).
98. Iwasaki, Y. etal. Aggravation of subclinical diabetes
insipidus during pregnancy. N. Engl. J. Med. 324,
522–526 (1991).
18
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();

99. Hashimoto, M. etal. Manifestation of subclinical
diabetes insipidus due to pituitary tumor during
pregnancy. Endocr. J. 43, 577–583 (1996).
100. Czaczkes, J. W., Kleeman, C. R. & Koenig, M.
Physiologic studies of antidiuretic hormone by its
direct measurement in human plasma. J. Clin. Invest.
43, 1625–1640 (1964).
101. Schrier, R. W. Systemic arterial vasodilation, vasopressin,
and vasopressinase in pregnancy. J. Am. Soc. Nephrol.
21, 570–572 (2010).
102. Fenske, W. etal. A copeptin- based approach in the
diagnosis of diabetes insipidus. N. Engl. J. Med. 379,
428–439 (2018).
This study showed that a copeptin- based approach
could be the new gold standard in the diagnosis of
different types of DI.
103. Adrogue, H. J. & Madias, N. E. Hypernatremia.
N.Engl. J. Med. 342, 1493–1499 (2000).
104. Palevsky, P. M., Bhagrath, R. & Greenberg, A.
Hypernatremia in hospitalized patients. Ann. Intern.
Med. 124, 197–203 (1996).
105. Riggs, J. E. Neurologic manifestations of fluid and
electrolyte disturbances. Neurol. Clin. 7, 509–523
(1989).
106. Baumann, G. & Dingman, J. F. Distribution, blood
transport, and degradation of antidiuretic hormone
in man. J. Clin. Invest. 57, 1109–1116 (1976).
107. Robertson, G. L. in Endocrinology and Metabolism
3rd edn (eds Felig, P., Baxter, J. D. & Frohman, L. A.)
385–432 (McGraw- Hill, New York, 1995).
108. Verbalis, J. G. Disorders of body water homeostasis.
Best Pract. Res. Clin. Endocrinol. Metab. 17, 471–503
(2003).
109. Miller, M., Dalakos, T., Moses, A. M., Fellerman, H.
&Streeten, D. H. Recognition of partial defects in
antidiuretic hormone secretion. Ann. Intern. Med. 73,
721–729 (1970).
110 . Van de Heijning, B. J., Koekkoek- van den Herik, I.,
Ivanyi, T. & Van Wimersma Greidanus, T. B. Solid-
phase extraction of plasma vasopressin: evaluation,
validation and application. J. Chromatogr. B Analyt.
Technol. Biomed. Life Sci. 565, 159–171 (1991).
111. Wun, T. Vasopressin and platelets: a concise review.
Platelets 8, 15–22 (1997).
112 . Preibisz, J. J., Sealey, J. E., Laragh, J. H., Cody, R. J.
&Weksler, B. B. Plasma and platelet vasopressin in
essential hypertension and congestive heart failure.
Hypertension 5, I129–138 (1983).
113 . Cadnapaphornchai, M. A. etal. Effect of primary
polydipsia on aquaporin and sodium transporter
abundance. Am. J. Physiol. Renal Physiol. 285,
F965–F971 (2003).
114 . Berliner, R. W. & Davidson, D. G. Production of
hypertonic urine in the absence of pituitary antidiuretic
hormone. J. Clin. Invest. 36, 1416–1427 (1957).
115 . Robertson, G. L., Mahr, E. A., Athar, S. & Sinha, T.
Development and clinical application of a new method
for the radioimmunoassay of arginine vasopressin in
human plasma. J. Clin. Invest. 52, 2340–2352 (1973).
116 . Dies, F., Rangel, S. & Rivera, A. Differential diagnosis
between diabetes insipidus and compulsive polydipsia.
Ann. Intern. Med. 54, 710–725 (1961).
117 . Harrington, A. R. & Valtin, H. Impaired urinary
concentration after vasopressin and its gradual
correction in hypothalamic diabetes insipidus.
J. Clin. Invest. 47, 502–510 (1968).
118 . Block, L. H., Furrer, J., Locher, R. A., Siegenthaler, W.
&Vetter, W. Changes in tissue sensitivity to vasopressin
in hereditary hypothalamic diabetes insipidus.
Klin.Wochenschr. 59, 831–836 (1981).
119 . Zerbe, R. L. & Robertson, G. L. A comparison of
plasma vasopressin measurements with a standard
indirect test in the differential diagnosis of polyuria.
N.Engl. J. Med. 305, 1539–1546 (1981).
120. Morgenthaler, N. G., Struck, J., Alonso, C.
&Bergmann, A. Assay for the measurement of
copeptin, a stable peptide derived from the precursor
of vasopressin. Clin. Chem. 52, 112–119 (2006).
121. Czaczkes, J. W. & Kleeman, C. R. The effect of various
states of hydration and the plasma concentration on
the turnover of antidiuretic hormone in mammals.
J.Clin. Invest. 43, 1649–1658 (1964).
122. Baylis, P. H. Diabetes insipidus. J. R. Coll. Physicians
Lond. 32, 108–111 (1998).
123. Baylis, P. H., Gaskill, M. B. & Robertson, G. L.
Vasopressin secretion in primary polydipsia and
cranial diabetes insipidus. Q. J. Med. 50, 345–358
(1981).
124. Timper, K. etal. Diagnostic accuracy of copeptin in
the differential diagnosis of the polyuria- polydipsia
syndrome: a prospective multicenter study. J. Clin.
Endocrinol. Metab. 100, 2268–2274 (2015).
125. Robertson, G. L., Shelton, R. L. & Athar, S. The
osmoregulation of vasopressin. Kidney Int. 10, 25–37
(1976).
126. Robertson, G. L. The regulation of vasopressin function
in health and disease. Recent Prog. Horm. Res. 33,
333–385 (1976).
127. Fujisawa, I. etal. Posterior lobe of the pituitary in
diabetes insipidus: MR findings. J. Comput. Assist.
Tomogr. 11, 221–225 (1987).
128. Arslan, A., Karaarslan, E. & Dincer, A. High intensity
signal of the posterior pituitary. A study with horizontal
direction of frequency- encoding and fat suppression
MR techniques. Acta Radiol. 40, 142–145 (1999).
129. Moses, A. M., Clayton, B. & Hochhauser, L. Use of
T1-weighted MR imaging to differentiate between
primary polydipsia and central diabetes insipidus.
AJNR Am. J. Neuroradiol. 13, 1273–1277 (1992).
130. Cote, M., Salzman, K. L., Sorour, M. & Couldwell, W. T.
Normal dimensions of the posterior pituitary bright
spot on magnetic resonance imaging. J. Neurosurg.
120, 357–362 (2014).
131. Hannon, M. etal. Anterior hypopituitarism is rare and
autoimmune disease is common in adults with idiopathic
central diabetes insipidus. Clin. Endocrinol. 76,
725–728 (2011).
132. Maghnie, M. etal. Correlation between magnetic
resonance imaging of posterior pituitary and
neurohypophyseal function in children with diabetes
insipidus. J. Clin. Endocrinol. Metab. 74, 795–800
(1992).
133. Bonneville, J. F. MRI of hypophysitis [French].
Ann.Endocrinol. 73, 76–77 (2012).
134. Gubbi, S., Hannah- Shmouni, F., Stratakis, C. A.
&Koch, C. A. Primary hypophysitis and other
autoimmune disorders of the sellar and suprasellar
regions. Rev. Endocr. Metab. Disord. 19, 335–347
(2018).
135. Leger, J., Velasquez, A., Garel, C., Hassan, M.
&Czernichow, P. Thickened pituitary stalk on magnetic
resonance imaging in children with central diabetes
insipidus. J. Clin. Endocrinol. Metab. 84, 1954–1960
(1999).
136. Verbalis, J. G. in Brenner and Rector’s The Kidney
9th edn (eds Maarten, T. etal.) 552–569 (Saunders,
Philadelphia, 2011).
137. Marques, P., Gunawardana, K. & Grossman, A.
Transient diabetes insipidus in pregnancy. Endocrinol.
Diabetes Metab. Case Rep. 2015, 150078 (2015).
138. Lindheimer, M. D. Polyuria and pregnancy: its cause,
its danger. Obstet. Gynecol. 105, 1171–1172 (2005).
139. Dabrowski, E., Kadakia, R. & Zimmerman, D. Diabetes
insipidus in infants and children. Best Pract. Res. Clin.
Endocrinol. Metab. 30, 317–328 (2016).
140. Gubbi, S., Hannah- Shmouni, F., Koch, C. A. &
Verbalis,J. G. in Endotext [Internet] (eds Feingold, K. R.
etal.) (MDText.com, Inc., South Dartmouth (MA),
2000).
141. Barker, J. M. Clinical review: type 1 diabetes- associated
autoimmunity: natural history, genetic associations, and
screening. J. Clin. Endocrinol. Metab. 91, 1210–1217
(2006).
142. Akbari, H. etal. Clinical outcomes of endoscopic
versus microscopic trans- sphenoidal surgery for large
pituitary adenoma. Br. J. Neurosurg. 32, 206–209
(2018).
143. Rajaratnam, S., Seshadri, M. S., Chandy, M. J.
&Rajshekhar, V. Hydrocortisone dose and postoperative
diabetes insipidus in patients undergoing trans-
sphenoidal pituitary surgery: a prospective randomized
controlled study. Br. J. Neurosurg. 17, 437–442
(2003).
144. Garofeanu, C. G. etal. Causes of reversible nephrogenic
diabetes insipidus: a systematic review. Am. J. Kidney
Dis. 45, 626–637 (2005).
145. Gullans, S. R. & Verbalis, J. G. Control of brain volume
during hyperosmolar and hypoosmolar conditions.
Annu. Rev. Med. 44, 289–301 (1993).
146. Verbalis, J. G. Brain volume regulation in response to
changes in osmolality. Neuroscience 168, 862–870
(2010).
147. Sterns, R. H. Disorders of plasma sodium—causes,
consequences, and correction. N. Engl. J. Med. 372,
55–65 (2015).
148. Bruck, E., Abal, G. & Aceto, T. Jr. Pathogenesis and
pathophysiology of hypertonic dehydration with
diarrhea. A clinical study of 59 infants with observations
of respiratory and renal water metabolism. Am. J. Dis.
Child. 115 , 122–144 (1968).
149. Fang, C. etal. Fluid management of hypernatraemic
dehydration to prevent cerebral oedema: a
retrospective case control study of 97 children in
China. J. Paediatr. Child Health 46, 301–303 (2010).
150. Bockenhauer, D. & Bichet, D. G. Nephrogenic diabetes
insipidus. Curr. Opin. Pediatr. 29, 199–205 (2017).
151. Sterns, R. H. Treatment of severe hyponatremia.
Clin.J. Am. Soc. Nephrol. 13, 641–649 (2018).
152. Oiso, Y., Robertson, G. L., Norgaard, J. P. & Juul, K. V.
Clinical review: treatment of neurohypophyseal
diabetes insipidus. J. Clin. Endocrinol. Metab. 98,
3958–3967 (2013).
153. Richardson, D. W. & Robinson, A. G. Desmopressin.
Ann. Intern. Med. 103, 228–239 (1985).
154. Lam, K. S. etal. Pharmacokinetics, pharmacodynamics,
long- term efficacy and safety of oral 1-deamino-
8-D-arginine vasopressin in adult patients with central
diabetes insipidus. Br. J. Clin. Pharmacol. 42,
379–385 (1996).
155. Rittig, S., Jensen, A. R., Jensen, K. T. & Pedersen, E. B.
Effect of food intake on the pharmacokinetics and
antidiuretic activity of oral desmopressin (DDAVP)
in hydrated normal subjects. Clin. Endocrinol. 48,
235–241 (1998).
156. Behan, L. A. etal. Abnormal plasma sodium
concentrations in patients treated with desmopressin
for cranial diabetes insipidus: results of a long- term
retrospective study. Eur. J. Endocrinol. 172, 243–250
(2015).
157. Verbalis, J. G. etal. Diagnosis, evaluation, and treatment
of hyponatremia: expert panel recommendations. Am. J.
Med. 126, S1–S42 (2013).
158. Sood, M. M. etal. Acute kidney injury in critically ill
patients infected with 2009 pandemic influenza
A(H1N1): report from a Canadian Province. Am. J.
Kidney Dis. 55, 848–855 (2010).
159. Achinger, S. G., Arieff, A. I., Kalantar- Zadeh, K.
& Ayus, J. C. Desmopressin acetate (DDAVP)-associated
hyponatremia and brain damage: a case series.
Nephrol. Dial. Transplant. 29, 2310–2315 (2014).
160. Schreckinger, M., Szerlip, N. & Mittal, S. Diabetes
insipidus following resection of pituitary tumors.
Clin.Neurol. Neurosurg. 115, 121–126 (2013).
161. Dohanics, J., Hoffman, G. E. & Verbalis, J. G. Chronic
hyponatremia reduces survival of magnocellular
vasopressin and oxytocin neurons after axonal injury.
J. Neurosci. 16, 2373–2380 (1996).
162. Loh, J. A. & Verbalis, J. G. Disorders of water and
salt metabolism associated with pituitary disease.
Endocrinol. Metab. Clin. North Am. 37, 213–234
(2008).
163. Eisenberg, Y. & Frohman, L. A. Adipsic diabetes
insipidus: a review. Endocr. Pract. 22, 76–83 (2016).
164. Cuesta, M., Hannon, M. J. & Thompson, C. J. Adipsic
diabetes insipidus in adult patients. Pituitary 20,
372–380 (2017).
165. Batlle, D. C., von Riotte, A. B., Gaviria, M. & Grupp, M.
Amelioration of polyuria by amiloride in patients
receiving long- term lithium therapy. N. Engl. J. Med.
312, 408–414 (1985).
166. Crawford, J. D. & Kennedy, G. C. Chlorothiazid in
diabetes insipidus. Nature 183, 891–892 (1959).
167. Earley, L. E. & Orloff, J. The mechanism of
antidiuresisassociated with the administration
ofhydrochlorothiazide to patients with vasopressin-
resistant diabetes insipidus. J. Clin. Invest. 41,
1988–1997 (1962).
168. Sinke, A. P. etal. Hydrochlorothiazide attenuates
lithium- induced nephrogenic diabetes insipidus
independently of the sodium- chloride cotransporter.
Am. J. Physiol. Renal Physiol. 306, F525–F533 (2014).
169. de Groot, T. etal. Acetazolamide attenuates lithium-
induced nephrogenic diabetes insipidus. J. Am. Soc.
Nephrol. 27, 2082–2091 (2015).
170. Gordon, C. E., Vantzelfde, S. & Francis, J. M.
Acetazolamide in lithium- induced nephrogenic
diabetes insipidus. N. Engl. J. Med. 375, 2008–2009
(2016).
171. Rosa, R. M. etal. A study of induced hyponatremia
in the prevention and treatment of sickle- cell crisis.
N.Engl. J. Med. 303, 1138–1143 (1980).
172. Usberti, M. etal. Renal prostaglandin E2 in
nephrogenic diabetes insipidus: effects of inhibition
of prostaglandin synthesis by indomethacin. J. Pediatr.
97, 476–478 (1980).
173. Boussemart, T., Nsota, J., Martin- Coignard, D.
&Champion, G. Nephrogenic diabetes insipidus:
treat with caution. Pediatr. Nephrol. 24, 1761–1763
(2009).
174. Bockenhauer, D. etal. Vasopressin type 2 receptor
V88M mutation: molecular basis of partial and
complete nephrogenic diabetes insipidus. Nephron
Physiol. 114 , 1–10 (2010).
175. Canfield, M. C., Tamarappoo, B. K., Moses, A. M.,
Verkman, A. S. & Holtzman, E. J. Identification
and characterization of aquaporin-2 water channel
19NATURE REVIEWS
|
DIseAse PrIMers
|
Article citation ID: (2019) 5:54
Primer
0123456789();

mutations causing nephrogenic diabetes insipidus with
partial vasopressin response. Hum. Mol. Genet. 6,
1865–1871 (1997).
176. Josiassen, R. C. etal. Double- blind, placebo- controlled,
multicenter trial of a vasopressin V2-receptor antagonist
in patients with schizophrenia and hyponatremia.
Biol.Psychiatry 64, 1097–1100 (2008).
177. Canuso, C. M. & Goldman, M. B. Clozapine restores
water balance in schizophrenic patients with polydipsia-
hyponatremia syndrome. J. Neuropsychiatry Clin.
Neurosci. 11, 86–90 (1999).
178. Goldman, M. & Ittasakul, P. in Schizophrenia: Recent
Advances in Diagnosis and Treatment (eds
Janicak,P.G., Goldman, M., Tandon, R. & Marder, S. R.)
205–224 (Springer, New York, 2014).
179. Ahmed, S. E. & Khan, A. H. Acetazolamide:
treatmentof psychogenic polydipsia. Cureus 9, e1553
(2017).
180. Costanzo, E. S., Antes, L. M. & Christensen, A. J.
Behavioral and medical treatment of chronic polydipsia
in a patient with schizophrenia and diabetes insipidus.
Psychosom. Med. 66, 283–286 (2004).
181. Ray, J. G. DDAVP use during pregnancy: an analysis of
its safety for mother and child. Obstet. Gynecol. Surv.
53, 450–455 (1998).
182. Burrow, G. N., Wassenaar, W., Robertson, G. L.
& Sehl, H. DDAVP treatment of diabetes insipidus
during pregnancy and the post- partum period.
ActaEndocrinol. 97, 23–25 (1981).
183. Nozaki, A. etal. Quality of life in the patients with
central diabetes insipidus assessed by Nagasaki
Diabetes Insipidus Questionnaire. Endocrine 51,
140–147 (2016).
184. Richards, G. E. etal. Natural history of idiopathic
diabetes insipidus. J. Pediatr. 159, 566–570 (2011).
185. Ishii, H. etal. Development and validation of a new
questionnaire assessing quality of life in adults with
hypopituitarism: Adult Hypopituitarism Questionnaire
(AHQ). PLOS ONE 7, e44304 (2012).
186. Ito, A., Nozaki, A., Horie, I., Ando, T. & Kawakami, A.
Relation between change in treatment for central
diabetes insipidus and body weight loss. Minerva
Endocrinol. 44, 85–90 (2019).
187. Juul, K. V., Schroeder, M., Rittig, S. & Norgaard, J. P.
National Surveillance of Central Diabetes Insipidus
(CDI) in Denmark: results from 5 years registration
of 9309 prescriptions of desmopressin to 1285 CDI
patients. J. Clin. Endocrinol. Metab. 99, 2181–2187
(2014).
188. Byun, D. J., Wolchok, J. D., Rosenberg, L. M.
&Girotra, M. Cancer immunotherapy — immune
checkpoint blockade and associated endocrinopathies.
Nat. Rev. Endocrinol. 13, 195–207 (2017).
189. Zhao, C. etal. Anti- PD-L1 treatment induced central
diabetes insipidus. J. Clin. Endocrinol. Metab. 103,
365–369 (2018).
190. Merimee, T. J., Rabinowtitz, D. & Fineberg, S. E.
Arginine- initiated release of human growth
hormone.Factors modifying the response in
normalman. N.Engl. J. Med. 280, 1434–1438
(1969).
191. Nair, N. P. etal. Effect of normal aging on the prolactin
response to graded doses of sulpiride and to arginine.
Prog. Neuropsychopharmacol. Biol. Psychiatry 9,
633–637 (1985).
192. Winzeler, B. etal. Arginine- stimulated copeptin
measurements in the differential diagnosis of diabetes
insipidus: a prospective diagnostic study. Lancet
https://doi.org/10.1016/S0140-6736(19)31255-3
(2019).
193. Morello, J. P. etal. Pharmacological chaperones rescue
cell- surface expression and function of misfolded V2
vasopressin receptor mutants. J. Clin. Invest. 105,
887–895 (2000).
194. Ando, F. & Uchida, S. Activation of AQP2 water
channels without vasopressin: therapeutic strategies
for congenital nephrogenic diabetes insipidus.
Clin. Exp. Nephrol. 22, 501–507 (2018).
195. Milano, S., Carmosino, M., Gerbino, A., Svelto, M.
&Procino, G. Hereditary nephrogenic diabetes
insipidus: pathophysiology and possible treatment.
Anupdate. Int. J. Mol. Sci. 18, E2385 (2017).
196. Feldman, B. J. etal. Nephrogenic syndrome of
inappropriate antidiuresis. N. Engl. J. Med. 352,
1884–1890 (2005).
197. Biebermann, H. etal. A new multi- system disorder
caused by the Gαs mutation p.F376V. J. Clin.
Endocrinol. Metab. 104, 1079–1089 (2019).
198. Tamarappoo, B. K. & Verkman, A. S. Defective
aquaporin-2 trafficking in nephrogenic diabetes
insipidus and correction by chemical chaperones.
J.Clin. Invest. 101, 2257–2267 (1998).
Acknowledgements
The authors thank N. Salvisberg for help with referencing.
Author contributions
Introduction (M.C.-C.); Epidemiology (W.K.F.); Mechanisms/
pathophysiology (D.G.B., M.B.G., S.R., J.G.V. and A.S.V.);
Diagnosis, screening and prevention (M.C.-C and W.K.F.);
Management (D.G.B., M.B.G. and J.G.V.); Quality of life
(M.B.G.); Outlook (M.C.-C., D.G.B., W.K.F., M.B.G., S.R, J.G.V.
and A.S.V.).
Competing interests
M.C.-C. and W.K.F. received speaking honoraria from Thermo
Fisher AG, the manufacturer of the copeptin assay. All the
other authors declare no competing interests.
Peer review information
Nature Reviews Disease Primers thanks H. Arima,
M.Magnie, S. Sasaki and Y. Sugimura for their contribution
to the peer review of this work.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
RELATED LINKS
OMIM: http://www.ncbi.nlm.nih.gov/omim/
Wolfram syndrome: https://omim.org/entry/222300
20
|
Article citation ID: (2019) 5:54 www.nature.com/nrdp
Primer
0123456789();