Water Deprivation Test

The water deprivation test separates the causes of hypotonic polyuria: central (cranial) diabetes insipidus (deficient antidiuretic hormone, ADH), nephrogenic diabetes insipidus (renal resistance to ADH) and primary polydipsia (excess water intake with ADH suppression). It measures the ability of the kidney to concentrate urine when water is withheld, followed by the response to desmopressin (synthetic ADH).

In diabetes insipidus, the urine stays dilute despite rising plasma osmolality. In primary polydipsia, concentration is normal or nearly normal after deprivation, although medullary washout may reduce it.

Diabetes insipidus is rare (prevalence about 1 in 25 000), and primary polydipsia is more common, especially in psychiatric illness. Polyuria is defined as urine volume over 3 L a day (over 40 to 50 mL/kg), with urine osmolality under 300 mOsm/kg. Direct measurement of plasma copeptin (the C-terminal fragment of pro vasopressin) after hypertonic saline or arginine stimulation has largely replaced the water deprivation test in many centres, with higher accuracy (over 95%), but deprivation remains widely used.

Indications. Polyuria and polydipsia with confirmed dilute urine (urine osmolality under 300 mOsm/kg, 24-hour urine volume over 3 L) with a normal glucose, calcium, potassium and renal function. Suspected diabetes insipidus after pituitary surgery, head injury, or lithium use. Not done if serum sodium is over 145 mmol/L with urine osmolality under 300 mOsm/kg (diabetes insipidus is already confirmed and desmopressin response is tested directly), or in patients with hyponatraemia (primary polydipsia is probable).

Contraindications and cautions. Hypernatraemia, hypovolaemia, hypotension, adrenal insufficiency (hydrocortisone replacement must be given), severe renal impairment, pregnancy (relative), and conditions where dehydration is dangerous. Supervised in hospital by experienced staff. Hypertonic saline testing and copeptin are alternatives.

Preparation. Stop caffeine, alcohol, tobacco and fluid overload the evening before. Light breakfast allowed, no tea or coffee. Patients with primary polydipsia may need gradual fluid restriction over the day before to reduce medullary washout (a prolonged test of 8 to 17 hours can be used). Drugs affecting ADH (carbamazepine, lithium, demeclocycline, diuretics, chlorpropamide) are reviewed. Baseline weight, blood pressure and pulse.

Procedure.

• 08:00: baseline weight, plasma osmolality, sodium, urine osmolality and volume.
• Fluid is withheld (up to 8 hours in adults, or until weight loss reaches 3 to 5%, whichever first). Weight, pulse, blood pressure, urine volume and osmolality are recorded hourly. Plasma osmolality and sodium are measured every 2 hours.
• Stop the deprivation phase when: urine osmolality over 600 to 800 mOsm/kg (adequate concentration), plasma osmolality over 300 mOsm/kg or sodium over 145 mmol/L, weight loss over 3 to 5%, urine osmolality plateaus (rise under 30 mOsm/kg over 2 consecutive hours), or symptoms of dehydration.
• At the end of deprivation, take blood for plasma osmolality, sodium and copeptin (if available), then give desmopressin 2 microg intramuscularly or 10 microg intranasally (or 0.1 to 0.2 mg orally).
• Continue to measure urine osmolality and volume for 2 hours (hourly) after desmopressin. Allow fluid intake of up to 1.5 times urine volume.

Urine osmolality after deprivation and after desmopressin.

Additional measures.

• Plasma ADH or copeptin measured at the end of deprivation: low in central diabetes insipidus, high in nephrogenic diabetes insipidus (copeptin over 21.4 pmol/L after osmotic stimulation excludes central cause, with sensitivity and specificity over 95% for the hypertonic saline test and 90% after arginine).
• Plasma sodium and osmolality rising to over 295 mOsm/kg with urine osmolality under 300 mOsm/kg confirm diabetes insipidus.
• Differentiation of partial forms from primary polydipsia is difficult, and copeptin based testing is preferred.
• Primary polydipsia: low plasma osmolality at baseline (under 280 mOsm/kg), high water intake history, concentrating response slowly recovers with fluid restriction.

Causes. Central diabetes insipidus: idiopathic (autoimmune), pituitary or hypothalamic tumours (craniopharyngioma, germinoma), surgery, trauma, infiltrative disease (Langerhans cell histiocytosis, sarcoidosis), infection, genetic (AVP-NPII), and pregnancy related. Nephrogenic: lithium, hypercalcaemia, hypokalaemia, chronic kidney disease, sickle cell disease, genetic (AVPR2, AQP2), and drugs.

Critical values and findings needing immediate action.

• Plasma sodium over 150 mmol/L, plasma osmolality over 310 mOsm/kg, weight loss over 5%, hypotension or tachycardia during the test: stop deprivation and give fluids.
• Urine osmolality persistently under 300 mOsm/kg with plasma sodium rising: confirmed diabetes insipidus, proceed to desmopressin.
• Confirmed central diabetes insipidus: pituitary MRI, other pituitary hormones tested, and desmopressin started.
• Confirmed primary polydipsia: no desmopressin (risk of severe hyponatraemia), psychiatric review.
• Acute hypernatraemia with confusion: slow correction, not over 10 mmol/L in 24 hours.

Interferences and pitfalls.

• Medullary washout in chronic polyuria (primary polydipsia) lowers maximal urine concentration and creates false suspicion of partial diabetes insipidus.
• Incomplete deprivation produces false results, and the patient is observed for secret drinking.
• Adrenal insufficiency masks diabetes insipidus (cortisol is needed for free water excretion), and appears after replacement.
• Hypokalaemia and hypercalcaemia cause nephrogenic defects and are corrected before testing.
• Pregnancy: increased vasopressinase degrades ADH, causing transient diabetes insipidus, and desmopressin is resistant to vasopressinase.
• Loss of thirst (adipsic diabetes insipidus) after hypothalamic injury is hard to manage.
• Desmopressin carries a risk of hyponatraemia in primary polydipsia, so doses and fluid intake are adjusted carefully.
• The test has diagnostic accuracy of about 70% only, with misclassification in partial forms, and copeptin is more reliable.

Special groups.

• Children: deprivation is shorter and closely monitored, with a weight loss limit of 3 to 5% and hospital supervision.
• Pregnancy: gestational diabetes insipidus is treated with desmopressin.
• Older adults: reduced concentrating ability, and a risk of hypernatraemia and volume depletion.
• Post-pituitary surgery: a triphasic response (diabetes insipidus, SIADH, then recurrent diabetes insipidus) may occur, and daily sodium and fluid balance is monitored.
• Regions: copeptin is not widely available, and the water deprivation test is still the main option.

Follow up tests. Plasma and urine osmolality, sodium, potassium, calcium, glucose, creatinine, 24-hour urine volume, copeptin, hypertonic saline test, pituitary MRI, anterior pituitary function tests, renal ultrasound, genetic testing (AVP, AVPR2, AQP2) in familial disease, and psychiatric assessment in primary polydipsia.