Not Found

Find information on medical topics, symptoms, drugs, procedures, news and more, written for the health care professional.

Central Diabetes Insipidus

(Vasopressin-Sensitive Diabetes Insipidus)

By Ian M. Chapman, MBBS, PhD, Professor of Medicine, Discipline of Medicine, University of Adelaide, Royal Adelaide Hospital

Click here for
Patient Education

Diabetes insipidus (DI) results from a deficiency of vasopressin (ADH) due to a hypothalamic-pituitary disorder (central DI [CDI]) or from resistance of the kidneys to vasopressin (nephrogenic DI [NDI]). Polyuria and polydipsia develop. Diagnosis is by water deprivation test showing failure to maximally concentrate urine; vasopressin levels and response to exogenous vasopressin help distinguish CDI from NDI. Treatment is with desmopressin or lypressin. Nonhormonal treatment includes use of diuretics (mainly thiazides) and vasopressin-releasing drugs, such as chlorpropamide.


The posterior lobe of the pituitary is the primary site of vasopressin storage and release, but vasopressin is synthesized within the hypothalamus. Newly synthesized hormone can still be released into the circulation as long as the hypothalamic nuclei and part of the neurohypophyseal tract are intact. Only about 10% of neurosecretory neurons must remain intact to avoid central diabetes insipidus. The pathology of central diabetes insipidus thus always involves the supraoptic and paraventricular nuclei of the hypothalamus or a major portion of the pituitary stalk.

CDI may be

  • Complete (absence of vasopressin)

  • Partial (insufficient amounts of vasopressin)

CDI may be

  • Primary, in which there is a marked decrease in the hypothalamic nuclei of the neurohypophyseal system

  • Secondary (acquired)


Primary central diabetes insipidus

Genetic abnormalities of the vasopressin gene on chromosome 20 are responsible for autosomal dominant forms of primary CDI, but many cases are idiopathic.

Secondary central diabetes insipidus

CDI may also be secondary (acquired), caused by various lesions, including hypophysectomy, cranial injuries (particularly basal skull fractures), suprasellar and intrasellar tumors (primary or metastatic), Langerhans cell histiocytosis, lymphocytic hypophysitis, granulomas (sarcoidosis or TB), vascular lesions (aneurysm, thrombosis), and infections (encephalitis, meningitis).

Symptoms and Signs

Onset of central diabetes insipidus may be insidious or abrupt, occurring at any age. The only symptoms in primary central diabetes insipidus are polydipsia and polyuria.

In secondary central diabetes insipidus, symptoms and signs of the associated lesions are also present. Enormous quantities of fluid may be ingested, and large volumes (3 to 30 L/day) of very dilute urine (specific gravity usually < 1.005 and osmolality < 200 mOsm/L) are excreted. Nocturia almost always occurs. Dehydration and hypovolemia may develop rapidly if urinary losses are not continuously replaced.

Polyuria may result from


  • Water deprivation test

  • Sometimes vasopressin levels

Central diabetes insipidus must be differentiated from other causes of polyuria, particularly psychogenic polydipsia (see Table: Common Causes of Polyuria) and nephrogenic diabetes insipidus. All tests for CDI (and for NDI) are based on the principle that increasing the plasma osmolality in normal people will lead to decreased excretion of urine with increased urine osmolality.

Common Causes of Polyuria



Vasopressin-sensitive polyuria

Decreased synthesis of vasopressin

Primary diabetes insipidus, hereditary (usually autosomal dominant)

Primary diabetes insipidus, hereditary associated with diabetes mellitus, optic nerve atrophy, nerve deafness, and atonia of bladder and ureters

Acquired (secondary) diabetes insipidus (causes outlined in text)

Decreased release of vasopressin

Psychogenic polydipsia (dipsogenic diabetes insipidus)

Vasopressin-resistant polyuria

Renal resistance to vasopressin

Congenital nephrogenic diabetes insipidus (usually X-linked recessive trait)

Acquired nephrogenic diabetes insipidus: Chronic kidney disease, systemic or metabolic disease (eg, myeloma, amyloidosis, hypercalcemic or hypokalemic nephropathy, sickle cell disease), certain drugs (eg, lithium, demeclocycline)

Osmotic diuresis

Hyperglycemia (in diabetes mellitus)

Poorly resorbed solutes (mannitol, sorbitol, urea)

The water deprivation test is the simplest and most reliable method for diagnosing central diabetes insipidus but should be done only while the patient is under constant supervision. Serious dehydration may result. Additionally, if psychogenic polydipsia is suspected, the patient must be observed to prevent surreptitious drinking. The test is started in the morning by weighing the patient, obtaining venous blood to determine electrolyte concentrations and osmolality, and measuring urinary osmolality. Voided urine is collected hourly, and its specific gravity or, preferably, osmolality is measured. Dehydration is continued until orthostatic hypotension and postural tachycardia appear, 5% of the initial body weight has been lost, or the urinary concentration does not increase >0.001 specific gravity or >30 mOsm/L in sequentially voided specimens. Serum electrolytes and osmolality are again determined. Exogenous vasopressin is then given (5 units of aqueous vasopressin sc, 10 mcg desmopressin [DDAVP] intranasally, or 4 mcg IM or IV). Urine for specific gravity or osmolality measurement is collected one final time 60 min postinjection, and the test is terminated.

A normal response produces maximum urine osmolality after dehydration (often > 1.020 specific gravity or >700 mOsm/L), exceeding the plasma osmolality; osmolality does not increase more than an additional 5% after injection of vasopressin. Patients with CDI are generally unable to concentrate urine to greater than the plasma osmolality but are able to increase their urine osmolality by > 50 to >100% after exogenous vasopressin administration. Patients with partial CDI are often able to concentrate urine to above the plasma osmolality but show a rise in urine osmolality of 15 to 50% after vasopressin administration. Patients with NDI are unable to concentrate urine to greater than the plasma osmolality and show no additional response to vasopressin administration (see Table: Water Deprivation Test Results).

Measurement of circulating vasopressin is the most direct method of diagnosing CDI; levels at the end of the water deprivation test (before the vasopressin injection) are low in CDI and appropriately elevated in NDI. However, vasopressin levels are difficult to measure, and the test is not routinely available. In addition, water deprivation is so accurate that direct measurement of vasopressin is unnecessary. Plasma vasopressin levels are diagnostic after either dehydration or infusion of hypertonic saline.

Pearls & Pitfalls

  • The water deprivation test should be done while the patient is under constant supervision because serious dehydration may occur.

Water Deprivation Test Results



Complete CDI

Partial CDI

Complete NDI

Partial NDI

Psychogenic Polydipsia

Uosm after dehydration (step 1)

Very high (> 700–800 mOsm/kg)

Very low (less than plasma osmolality)

Low (≥ 300 mOsm/kg)

Very low (less than plasma osmolality)

Very low (less than plasma osmolality)

High (500–600 mOsm/kg)

Uosm increase after vasopressin (step 2)

Minimal (< 5%)

50 to > 100%


< 50 mOsm/kg

Up to 45%

No change

Uosm = urine osmolality

Psychogenic polydipsia

Psychogenic polydipsia may present a difficult problem in differential diagnosis. Patients may ingest and excrete up to 6 L of fluid/day and are often emotionally disturbed. Unlike patients with CDI and NDI, they usually do not have nocturia, nor does their thirst wake them at night. Continued ingestion of large volumes of water in this situation can lead to life-threatening hyponatremia.

Patients with acute psychogenic water drinking are able to concentrate their urine during water deprivation. However, because chronic water intake diminishes medullary tonicity in the kidney, patients with long-standing polydipsia are not able to concentrate their urine to maximal levels during water deprivation, a response similar to that of patients with partial CDI. However, unlike CDI, patients with psychogenic polydipsia show no response to exogenous vasopressin after water deprivation. This response resembles NDI, except that basal vasopressin levels are low compared with the elevated levels present in NDI. After prolonged restriction of fluid intake to 2 L/day, normal concentrating ability returns within several weeks.


  • Hormonal drugs, eg, desmopressin

  • Nonhormonal drugs, eg, diuretics

Central diabetes insipidus can be treated with hormone replacement and treatment of any correctable cause. In the absence of appropriate management, permanent renal damage can result.

Restricting salt intake may also help because it reduces urine output by reducing solute load.

Hormonal drugs

Desmopressin, a synthetic analog of vasopressin with minimal vasoconstrictive properties, has prolonged antidiuretic activity, lasting for 12 to 24 h in most patients, and may be administered intranasally, sc, IV, or orally. Desmopressin is the preparation of choice for both adults and children and is available as an intranasal solution in 2 forms. A dropper bottle with a calibrated nasal catheter has the advantage of delivering incremental doses from 5 to 20 mcg but is awkward to use. A spray bottle that delivers 10 mcg of desmopressin in 0.1 mL of fluid is easier to use but delivers a fixed quantity.

For each patient, the duration of action of a given dose must be established, because variation among individuals is great. The duration of action can be established by following timed urine volumes and osmolality. The nightly dose is the lowest dose required to prevent nocturia. The morning and evening doses should be adjusted separately. The usual dosage range in adults is 10 to 40 mcg, with most adults requiring 10 mcg bid. For children age 3 mo to 12 yr, the usual dosage range is 2.5 to 10 mcg bid.

Overdosage can lead to fluid retention and decreased plasma osmolality, possibly resulting in seizures in small children. In such instances, furosemide can be given to induce diuresis. Headache may be a troublesome adverse effect but generally disappears if the dosage is reduced. Infrequently, desmopressin causes a slight increase in BP. Absorption from the nasal mucosa may be erratic, especially when URI or allergic rhinitis occurs. When intranasal delivery of desmopressin is inappropriate, it may be administered sc using about one tenth of the intranasal dose. Desmopressin may be used IV if a rapid effect is necessary (eg, for hypovolemia). With oral desmopressin, dose equivalence with the intranasal formulation is unpredictable, so individual dose titration is needed. The initial dose is 0.1 mg po tid, and the maintenance dose is usually 0.1 to 0.2 mg tid.

Pearls & Pitfalls

  • The duration of action of a given dose of desmopressin varies greatly among individuals and must be established for each patient.

Lypressin (lysine-8- vasopressin), a synthetic agent, is given by nasal spray at doses of 2 to 4 units (7.5 to 15 mcg) q 3 to 8 h but, because of its short duration of action, has been largely replaced by desmopressin.

Aqueous vasopressin 5 to 10 units sc or IM can be given to provide an antidiuretic response that usually lasts 6 h. Thus, this drug has little use in long-term treatment but can be used in the initial therapy of unconscious patients and in patients with central diabetes insipidus who are undergoing surgery. Synthetic vasopressin can also be administered bid to qid as a nasal spray, with the dosage and interval tailored to each patient. Vasopressin tannate in oil 0.3 to 1 mL (1.5 to 5 units) IM may control symptoms for up to 96 h.

Nonhormonal drugs

At least 3 groups of nonhormonal drugs are useful in reducing polyuria:

  • Diuretics, primarily thiazides

  • Vasopressin-releasing drugs (eg, chlorpropamide, carbamazepine, clofibrate)

  • Prostaglandin inhibitors

These drugs have been particularly useful in partial central diabetes insipidus and do not cause the adverse effects of exogenous vasopressin.

Thiazide diuretics paradoxically reduce urine volume in partial and complete CDI (and nephrogenic diabetes insipidus), primarily as a consequence of reducing ECF volume and increasing proximal tubular resorption. Urine volumes may fall by 25 to 50% with 15 to 25 mg/kg of chlorothiazide.

Chlorpropamide, carbamazepine, and clofibrate can reduce or eliminate the need for vasopressin in some patients with partial CDI. None are effective in NDI. Chlorpropamide 3 to 5 mg/kg po once/day or bid causes some release of vasopressin and also potentiates the action of vasopressin on the kidney. Clofibrate 500 to 1000 mg po bid or carbamazepine 100 to 400 mg po bid is recommended for adults only. These drugs may be used synergistically with a diuretic. However, significant hypoglycemia may result from chlorpropamide.

Prostaglandin inhibitors (eg, indomethacin 0.5 to 1.0 mg/kg po tid, although most NSAIDs are effective) are modestly effective. They may reduce urine volume, but generally by no more than 10 to 25%, perhaps by decreasing renal blood flow and GFR. Together with indomethacin, restriction of sodium intake and a thiazide diuretic help further reduce urine volume in NDI.

Key Points

  • Central diabetes insipidus (CDI) is caused by a deficiency of vasopressin, which decreases the kidneys' ability to reabsorb water, resulting in massive polyuria (3 to 30 L/day).

  • The cause may be a primary genetic disorder or various tumors, infiltrative lesions, injuries, or infections that affect the hypothalamic-pituitary system.

  • Diagnosis using a water deprivation test; patients cannot maximally concentrate urine following dehydration but can concentrate urine after receiving exogenous vasopressin.

  • Low vasopressin levels are diagnostic, but vasopressin levels are difficult to measure and the test is not routinely available.

  • Address any treatable causes and give desmopressin, a synthetic analog of vasopressin.

Resources In This Article