Diabetes Insipidus

Also known as:


Continue Reading

Related conditions

Polyuria, polydipsia, nocturia, hypernatremia

1. Description of the problem

What every clinician needs to know

Diabetes insipidus (DI) is characterized by decreased activity of the antidiuretic hormone (ADH) vasopressin (VP) due to partial or complete failure of secretion (central or hypothalamic DI) or renal resistance to its effect (nephrogenic DI). (Vasopressin is a hormone secreted by the posterior pituitary gland. In the kidney, vasopressin causes water reabsorption through the V2 receptor-mediated insertion of aquaporin water channels into the luminal membrane in the collecting duct.) DI is characterized by a decreased ability to generate a concentrated urine, leading to polyuria and secondarily to polydipsia. In the presence of an intact thirst mechanism and free access to water, diabetes insipidius does not result in overt hypernatremia (an increase in the sodium concentration in body fluids to a value above the upper limit of normal); however, if thirst or access to water is impaired, hypernatremia will ensue.

Clinical features

The hallmark of DI is polyuria (urine output >3 L/day). Onset of symptoms may be sudden in hypothalamic DI (HDI) but is usually more insidious in nephrogenic DI (NDI).

Polyuria is defined arbitrarily as a urine volume greater than 3 Liters/day. The degree of polyuria depends on the total solute load, effective circulating volume and severity of DI. Nocturia may be the main symptom in less severe DI and is often the first clue to its diagnosis.

HDI that results from trauma or surgical damage to the hypothalamus or the stalk of the pituitary gland may be transient or permanent or may follow a triphasic course. The transient form is the most common, starting abruptly and resolving over days or weeks. The triphasic response is characterized by an initial polyuric phase from the inhibition of vasopressin release (starting within one day, lasting 4-5 days), followed by a SIADH-like phase due to release of stored ADH from the posterior pituitary (days 6-11), followed by permanent HDI.

HDI may be accompanied by concurrent deficiency of other anterior pituitary hormones such as growth hormone or adrenocorticotropic hormone (ACTH), the latter resulting in adrenal insufficiency. Polyuria may not be manifested until the adrenal insufficiency is treated, since cortisol deficiency increases fluid reabsorption and increases vasopressin release.

Hypernatremia and DI

Hypernatremia (defined as a serum Na >145 mEq/L) may develop as the result of DI infants and in adults with altered mental status or otherwise impaired thirst, or who have restricted access to water. Hypothalamic disorders such as tumors, granulomatous disorders and vascular disease can result in both DI and impaired thirst sensation. In patients with critical illness, DI may become manifest through the development of hypernatremia as the result of sedation, altered mental status from intercurrent critical illness, or as the result of restricted access to fluids.

Symptoms of hypernatremia are mainly neurologic, manifesting as lethargy and weakness that can progress to seizures and coma. Chronic hypernatremia can be asymptomatic.

Key management points in order of priority

– Determine the cause of polyuria (exclude primary polydipsia and osmotic diuresis) by clues from history, physical examination, and laboratory studies, including serum sodium, BUN, glucose and osmolality (Osm) and urine sodium, potassium and osmolality.

– Correct the water deficit if hypernatremia is present and replace ongoing electrolyte-free water losses.

– Conduct a water deprivation test if the patient is polyuric but is not hypernatremic. If the patient is hypernatremic (serum sodium > 145 mEq/L) the response to exogenous vasopressin or dDAVP should be assessed without water deprivation.

-HDI is treated with hormone replacement with vasopressin (short-acting) or with the longer-acting synthetic analogue desmopressin (dDAVP).

– NDI is often resistant to therapy. Treatment should focus on correcting reversible causes (e.g., discontinuation of offending medications) and decreasing urinary solute load by restricting dietary sodium and protein. Thiazide diuretics, which block urinary dilution, may be used as an adjunct to therapy.

– Serum Na, serum Osm, and urine Osm should be monitored closely after start of therapy to ensure correct diagnosis and management.

2. Emergency Management

Emergency situations can occur when unrecognized or new-onset DI causes symptomatic hypernatremia in a patient with altered mental status, impaired thirst mechanism or restricted access to water. Prompt administration of free water (administered via nasogastric tube or as intravenous 5% dextrose in water) is essential to correct hypernatremia. The optimal rate of correction is not known. A reasonable rule is that the initial rate of correction should be approximately 1-2 mEq/L/hr for 3-4 hours or until symptoms resolve, with a goal of correcting approximately half of the free-water deficit over the initial 24 hours of therapy, with full correction achieved in over the ensuing 24 to 48 hours.

Neurologic status should be monitored closely during therapy. Initial neurologic recovery followed by worsening of mental status may suggest the development of cerebral edema from overly rapid correction of hypernatremia. If this develops, water replacement should be temporarily discontinued until for several hours and resumed, at a slower rate, once neurologic status has stabilized.

First-line therapy

The first line of therapy is to correct hypernatremia by replacing the free water deficit.

The free water deficit may be estimated as = TBW * (Plasma Na – 140)/140, where TBW is the estimated total body water, calculated as 0.5* Lean body weight (LBW) for women and 0.6* LBW for men.

Free water replacement should also account for ongoing urinary losses due to polyuria, any GI losses and insensible losses (generally estimated as 1 liter per day in adults in the absence of fever).

3. Diagnosis

The first step in the diagnosis of the patient with polyuria is determining whether the polyuria is due to a solute diuresis or a water diuresis.

-Solute diureses are most commonly due to renal excretion of glucose (usually associated with hyperglycemia), sodium (either physiologic naturesis or renal salt wasting), or urea and are usually characterized by a urine osmolality >250 mOsm/kg.

-In contrast, DI is characterized by a water diuresis, with a urine osmolality <150 mOsm/kg.

The differential diagnosis of polyuria with a water diuresis is primarily between DI and primary (often psychogenic) polydipsia.

The diagnosis of DI is based on demonstration of an inability to concentrate the urine in response to hypertonic stimuli. Differentiation between hypothalamic and nephrogenic DI is generally based on the clinical response to exogneous vasopressin (or the synthetic analogue, desmopressin). This is generally achieved with a water deprivation test (Table I).

Table I.
Urine Osmolality after fasting(mOsm/kg) Increase in Urine Osmolality following exogenous Vasopressin or Desmopressin Possible Diagnoses
<300 >50% increase Complete hypothalamic DI
<50% increase Complete nephrogenic DI
300-800 >10% increase Partial hypothalamic DI
<10% increase Partial nephrogenic DIPrimary polydipsia
>800 Not Applicable Normal

In a water deprivation test, the patient with polyuria is placed on a strict, supervised fast with close monitoring of vital signs, volume status and urine and serum electrolytes.

  • In patients with severe polyuria the test should begin in the early morning with close monitoring for signs of volume depletion; in patients with less severe polyuria the test may begin with an overnight fast.
  • Vital signs should be monitored every 30 to 60 minutes during the test, and if hypotension ensues the test should be immediately terminated.
  • Weight, urine volume and urine osmolality should be monitored hourly.
  • Plasma osmolality should be checked every 4-6 hours.
  • Water deprivation is continued until body weight has decreased by 3% from baseline, plasma osmolality rises to more than 295 mOsm/kg or urine osmolality has reached a plateau (three consecutive hourly values that vary by less than 5%).
  • – Exogenous vasopressin (5-10 units) or desmopressin (4 mcg) should then be administered IV or SC and urine volume and urine osmolality monitored every 30 minutes for an additional 2 hours.

Water deprivation testing should not be performed in patients who are already substantially hypernatremic (serum Na >145 mEq/L) or have a plasma osmolality > 295 mOsm/kg, as they should already have maximal pituitary vasopressin secretion at these levels. Testing in these patients should be limited to administration of exogenous vasopressin or desmopressin.

Measurement of plasma vasopressin levels at baseline and prior to administration of exogenous vasopressin may be useful if the results of the dehydration test are equivocal.

Normal lab values

Serum Na 135-145 mEq/L

Plasma osmolarity 275-290 mOsm/Kg

Maximal VP stimulation occurs when plasma osmolality > 295 mOsm/Kg

UOsm 700-1400 mOsm/Kg in the presence of hypernatremia and normal VP response

UNa < 25 mEq/L if hypernatremia is due to water loss

UNa > 100 mEq/L if hypernatremia and polyuria are due to excessive salt loading

How do I know this is what the patient has?

– History of lithium use, neurologic disease, sarcoidosis, psychiatric illness, no history of uncontrolled diabetes or recent AKI

– Symptoms of severe polyuria (>4-5 L/day), nocturia, thirst. Onset is abrupt in CDI and progressive in NDI or primary polydipsia.

– Labs: Polyuria or hypernatremia with a low urine osmolarity. Plasma Na is high-normal in DI, low-normal in primary polydipsia.

Other possible diagnoses

Polyuria can result from primary polydipsia. Usually serum Na is low-normal and patient has history of psychiatric illness.

Polyuria after neurosurgery may be due to excretion of excessive fluids received during surgery or from hyperosmolar drugs induced to prevent brain edema such as mannitol, hypertonic saline or hyperglycemia from high-dose corticosteroids.

Polyuria developing in the hospital setting may be due to solute diuresis from high-protein enteral feedings, hyperglycemia, mannitol, urea, saline loading, diuretic use, recovery of renal failure or relief of urinary obstruction. An osmotic diuresis is suggested when polyuria is accompanied by a urine with an osmolality similar to or greater than that of serum (urine osmolality >250 mOsm/Kg) and increased urinary solute excretion per day (greater than 900 mOsm/day). Urine Na is elevated (usually well above 100 mEq/L) when polyuria and hypernatremia are due to an elevated sodium load. Measuring urine Na, glucose and urea can identify the major solute that is being excreted.

Confirmatory tests

Water deprivation testing (as described above).

Plasma vasopressin level at the start and end of water deprivation, before exogenous vasopressin or desmopressin is administered, may be useful in equivocal cases. Vasopressin levels are usually less than 1.0 pg/mL in complete CDI, less than 1.5 pg/mL in partial CDI, greater than 2.0 pg/mL in normal, partial NDI & primary polydipsia, >5 pg/mL in complete NDI.

4. Specific Treatment

– First-line therapy for hypothalamic (central) DI is to replace VP with arginine vasopressin or desmopressin (dDAVP). Desmopressin, a 2 amino-acid substitution of vasopressin, is the preferred form of therapy because of longer half-life and a decreased vasoconstrictive effect.

– In partial hypothalamic DI, therapy can also include drugs that potentiate VP effect or increase its secretion such as chlorpropamide (increases VP effect), carbamazepine (increases VP effect), clofibrate (increases VP release).

– In nephrogenic DI, correction of underlying cause is sufficient in certain cases (hypercalcemia, hypokalemia, eliminating certain drugs).

– Induction of mild volume depletion with low-sodium diet and use of thiazide diuretics, in combination with low-protein diet, helps decrease the amount of water diuresis in both hypothalamic DI and nephrogenic DI. This is the major form of therapy for NDI that is not otherwise readily reversible. Addition of amiloride can enhance the effect of thiazides while minimizing K loss. In lithium-induced nephrogenic DI, amiloride presents the added benefit of blocking lithium entry into the collecting tubular cells, but this effect is seen only if lithium use is continued concurrently and the NDI is reversible (UOsm >200-250 mOsm/Kg). Loop diuretics should be avoided as they will result in further impairment of the urine concentrating ability.

– In NDI, use of NSAIDs increases urine concentrating ability by lowering prostaglandin synthesis. That effect is independent of ADH presence, although it does potentiate ADH effect when present. Not all NSAIDs are effective; indomethacin is the classically used one.

Desmopressin (dDAVP, desamino-8-D-arginine vasopressin) 5-20 mcg/dose intranasally once or twice daily; 1-2 mcg (IV or SC) 1 or 2 times per day

Oral desmopressin 0.05 mg qhs with subsequent titration. Usual daily doses: 0.1 mg-0.8 mg PO in divided doses

Aqueous vasopressin 5-10 units SC or IV in acute situations

HCTZ 12.5-25 mg daily or BID

Chlorpropramide 125-250 mg daily or BID

Clofibrate 500 mg PO Q6 hrs

Carbamazepine 100-300 mg BID

Refractory cases of NDI can sometimes be treated with supraphysiologic levels of dDAVP.

5. Disease monitoring, follow-up and disposition

Expected response to treatment

The expected response to treatment is reduction in urine volume and control of nocturia and polyuria. Careful monitoring is essential as hyponatremia can develop with wrong diagnosis or treatment.

When should I suspect I’ve made the wrong diagnosis?

Development of hyponatremia in a patient treated with desmopressin for presumed hypothalamic DI may result from excessively high dose of desmopressin, wrong diagnosis or resolution of DI (for instance, after surgery). Lack of symptom resolution after institution of therapy with desmopressin should prompt reassessment of the diagnosis. Patients with incomplete or partial DI and patients with polydipsia have the same response to the water restriction test.

What follow-up should this patient receive?

Follow-up on symptoms, urine osmolarity, serum Na is essential. Patients diagnosed with DI should carry a medical alert bracelet in case they develop altered mental consciousness.

Patients with hypothalamic DI should be evaluated for concomitant anterior pituitary disease and should have radiologic imaging of the hypothalamus and pituitary gland.


Hypothalamic DI is due to disruption of vasopressin synthesis or secretion. Vasopressin is synthesized in the supraoptic and paraventricular nuclei in the hypothalamus, then released by the posterior pituitary. Hypothalamic DI can result from damage to the osmoreceptors, hypothalamic nuclei or the hypophyseal hypothalamic tract.

  • Idiopathic DI (30% likely autoimmune disease)
  • Trauma or neurosurgery (usually transsphenoidal)
  • Infiltrative diseases (malignancy, sarcoidosis, histiocytosis X)
  • Thrombosis and hemorrhage
  • Hypoxic injury can lead to mild and often subclinical impairment in ADH release
  • Hereditary (rare)
  • Anorexia nervosa can affect ADH release.

Nephrogenic DI (NDI) refers to absent or decreased renal responsiveness to ADH action at the level of the collecting tubules. It can be congenital but is more commonly acquired.

  • Congenital: X-linked recessive (V2 receptor) or autosomal recessive (aquaporin-2 water channels)
  • Electrolyte disturbances: hypercalcemia (Ca>11 mg/dL) and hypokalemia (K<3 mEq/L) impair countercurrent function and the ability to respond to ADH.
  • Drugs (lithium, demeclocycline, amphotericin B, methoxyflurane). Lithium accumulates in the collecting tubule cells and interferes with ADH effect. Demeclocycline, a tetracycline antibiotic, is sometimes used to treat refractory SIADH.
  • Obstructive uropathy
  • An unusual form of NDI is seen in some women during their second trimester where increased levels of vasopressinase synthesized by the placenta lead to polyuria.


NDI occurs in 20-40% of patients on chronic lithium therapy.

The incidence of hypothalamic DI after trauma or neurosurgery depends on the extent of trauma. It can affect 10% of patients undergoing minimally invasive adenomectomy (usually transsphenoidal) and up to 80% of patients undergoing large tumor resections. Most cases of hypothalamic DI are not permanent and most cases of polyuria post-surgery are not due to hypothalamic DI. Study shows serum Na for 5 days after endoscopic pituitary surgery can be used as a predictor of risk for CDI.


Most patients with DI only complain of polyuria, nocturia and polydipsia. They are at risk for hypernatremia with hospitalizations, surgery, altered mental status. Untreated patients with chronic polyuria may develop functional dilation of the bladder, hydroureter and hydronephrosis.

When nephrogenic DI results from lithium therapy, it is often reversible with discontinuation of the drug, although with chronic lithium use, it may become permanent.

Prognosis is generally good once appropriate diagnosis is made, correct treatment is instituted and precautions are taken to avoid restricted access to water. This is especially true for hypothalamic DI, whereas for nephrogenic DI, therapy can be a little more challenging and requires greater patient cooperation.

What’s the evidence?

Adrogue, HJ, Madias, NE. “Hypernatremia”. N Engl J Med. vol. 342. 2000. pp. 1493-9. (Review of hypernatremia, pathophysiology and treatment. Includes discussion of formula used to correct hypernatremia.)

Palevsky, PM. “Hypernatremia”. Semin Nephrol. vol. 18. 1998 Jan. pp. 20-30. (Review of hypernatremia, epidemiology and pathophysiology.)

Rose, BD, Post, TW. Clinical Physiology of Acid-Base and Electrolyte Disorders. 2001. pp. 746-784. (Review of hypernatremia includes diabetes insipidus, pathogenesis and treatment options.)

Sigounas, DG, Sharpless, JL, Cheng, DM, Johnson, TG, Senior, BA. “Predictors and incidence of central diabetes insipidus after endoscopic pituitary surgery”. Neurosurgery. vol. 62. 2008. pp. 71-8. (Shows that minimally invasive surgery is associated with low incidence of DI, and normal serum Na in the first 5 days after surgery.)

Garofeanu, CG, Weir, M, Rosas-Arellano, MP. “Causes of reversible nephrogenic diabetes insipidus: a systematic review”. Am J Kidney Dis. vol. 45. 2005. pp. 626(Systematic review of reversible causes of NDI reported, showing that medications are the most common culprit.)

Sasaki, S. “Nephrogenic diabetes insipidus: update of genetic and clinical aspects”. Nephrol Dial Transplant. vol. 19. 2004. pp. 1351

Hensen, J, Henig, A, Fahlbusch, R. “Prevalence, predictors and patterns of postoperative polyuria and hyponatraemia in the immediate course after transsphenoidal surgery for pituitary adenomas”. Clin Endocrinol (Oxf). vol. 50. 1999. pp. 431

Seckl, JR, Dunger, DB. “Diabetes insipidus. Current treatment recommendations”. Drugs.. vol. 44. 1992. pp. 216-24. (A review of treatment options for central and nephrogenic DI.)

Zerbe, RL, Robertson, GL. “A comparison of plasma vasopressin measurements with a standard indirect test in the differential diagnosis of polyuria”. N Engl J Med.. vol. 305. 1981. pp. 1539-46. (Study that shows the usefulness of vasopressin measurements for accurate diagnosis.)